# Geopera — Full Content Archive

> Geopera is a satellite imagery company providing high-resolution commercial satellite data from multiple providers. We process and deliver orthorectified, pansharpened imagery as Cloud Optimized GeoTIFFs. This file contains the complete text of all published blog posts and case studies.

- Website: https://geopera.com
- Customer Portal: https://portal.geopera.com
- Technical Docs: https://docs.geopera.com
- Lightweight index: https://geopera.com/llms.txt

Total posts: 53
Total case studies: 2

---

## Blog Posts

# Backpropagating Through a Flood

> We calibrated the sediment physics of our Bhote Koshi flood model by gradient descent through the full shallow-water solver, against terrain change measured from stereo satellite imagery. It worked, after three instructive failures.

Published: 2026-09-01 | Author: Darcy Weedman | Reading time: 9 min

Source: https://geopera.com/blog/bhote-koshi-differentiable-calibration

---
<script>
import { PostFigure } from '$lib/components/charts';
</script>

## Summary

- Our 2D flood-and-sediment model of the Bhote Koshi disaster runs on a GPU in PyTorch. That makes it differentiable: every output is a mathematical function of every parameter, with gradients available by backpropagation, straight through roughly eight thousand solver timesteps.
- So instead of hand-tuning the sediment physics, we made six parameters learnable and descended the gradient of the mismatch between simulated terrain change and the terrain change we measured from stereo satellite imagery.
- It reduced the misfit against the measurement by 59 percent in about half an hour on a consumer graphics card, and the calibrated physics came out sensible: finer sediment, lower critical shear, smoother flow than the hand-tuned values.
- Getting there took three failures worth documenting: infinite gradients at dry cells, exploding gradients through shock waves, and a 25-minute compilation from being greedy about kernel fusion.
- The strangest part predates the method. An earlier calibration target turned out to be wrong, and the model's physics had visibly resisted it, putting sediment where the corrected measurement later said it belonged. Sometimes the model is arguing with your data for good reason.

This is the methods post in our Bhote Koshi series; the human context and the measurements live in the [earlier](/blog/bhote-koshi-flood-2026-satellite-analysis) [posts](/blog/bhote-koshi-stereo-elevation-model). It gets more technical than usual, on purpose.

## Why gradients, and why now

Calibrating a flood model has traditionally meant running it many times: pick parameter values, simulate, compare with observations, adjust, repeat. With four or six coupled parameters, the search space grows fast, every sample costs a full simulation, and the tuner's intuition quietly becomes part of the model.

There is another way if your solver happens to be written in an automatic-differentiation framework. Ours is: we built it on PyTorch for GPU speed, back when the goal was just fast ensembles. But a solver written that way is secretly a very deep neural network, one whose "layers" are timesteps of the shallow-water equations with erosion, sediment transport, and deposition riding along. Ask for the derivative of the output misfit with respect to the erosion coefficient, and the framework will chain-rule its way backward through the entire flood.

This is a young but fast-moving corner of the field. Differentiable flood solvers built for exactly this purpose have begun appearing in the literature this year. Our contribution is small and practical: we did it to a real disaster, against a real satellite-derived measurement, and we are writing down everything that broke.

## The setup

Six learnable parameters, in log space so they stay positive: an erosion coefficient, a critical shear stress below which the bed resists, a settling velocity, a speed threshold above which nothing deposits, the injected sediment concentration, and the Manning friction coefficient. The loss compares the model's final bed-elevation change against the stereo-measured change on valley-floor cells, plus a weaker term matching flow heights at surveyed trimline points. One evaluation is a full simulation of the flood's passage through the deposition reach, about eight thousand timesteps at 24-metre resolution.

One structural fix mattered before any gradient did. The original model domain ended at kilometre 42.5 of the corridor, which the stereo measurement revealed to be upstream of where the real deposit ends. The model could never have matched the measurement, because the measurement lay partly outside its world. The calibration domain now runs to kilometre 46.5. It is a good reminder that no optimiser can fix a domain boundary.

## Three failures, in order

**Failure one: infinite gradients at the water's edge.** The wave speed in a shallow-water solver involves the square root of water depth, and the derivative of a square root at exactly zero is infinite. Ninety-five percent of our domain is dry land sitting at exactly zero depth. The forward simulation was fine; the very first backward pass returned NaN for every parameter. The fix is an epsilon inside every square root, and the same medicine for the vector magnitude of velocity, whose gradient is undefined at exactly zero motion. If you take one line from this post into your own differentiable solver, take this one.

**Failure two: exploding gradients through shock waves.** With the poles patched, the descent ran for seven productive iterations and froze. Backpropagating through thousands of steps of nonlinear, shock-forming dynamics is the recurrent-network exploding-gradient problem in fancy dress: tiny perturbations to the flow amplify exponentially, and the gradients overflow. The cure is also borrowed from recurrent networks: truncation. We detach the fast, chaotic flow state every sixty steps, while keeping the slowly accumulating bed-elevation chain, the thing the loss actually measures, differentiable end to end. Gradients through the morphology survive; gradients through the turbulence-adjacent chaos are cut before they can explode. After this change, not a single gradient overflowed in forty iterations.

**Failure three: greed at compile time.** To speed things up we asked the compiler to fuse all sixty timesteps of a segment into one giant kernel graph. Twenty-five minutes later it was still compiling and the GPU had done nothing. Compiling a single timestep instead takes seconds and captures nearly all the win, because the benefit is fusing the forty small operations inside a step, not gluing steps together. With that, plus moving the inflow lookup out of the inner loop, an iteration dropped from 174 seconds to 37.

<PostFigure
src="/blogs/bhotekoshi-diffcal-loss.webp"
alt="Calibration loss by iteration: a first run that froze from exploding gradients, then a restarted run with truncated backpropagation descending to the best loss"
kind="chart"
caption="The whole campaign: the full-adjoint run froze when gradients overflowed at iteration 7; the truncated-adjoint restart descended to the best fit and then wandered a flat basin, which is what running out of parameters looks like."
/>

## What the flood learned

The descent cut the terrain-change misfit by 59 percent, most of it in the first fifteen iterations. The calibrated parameters tell a coherent physical story. Critical shear fell from 250 to 150 pascals and the settling velocity from 0.15 to 0.054 metres per second: the sediment behaves finer than we had assumed, staying in suspension longer and travelling further down-valley before dropping, which is exactly the correction the measured deposit demanded. Manning friction fell from 0.050 to 0.034, echoing a result we found earlier by a completely different route, that this flood moved like water, not like a viscous debris slurry. And total modelled deposition in the reach fell from 21 million cubic metres to about 5, from four times the measured scale to roughly matching it.

<PostFigure
src="/blogs/bhotekoshi-diffcal-before.webp"
alt="Model elevation change before calibration beside the stereo-measured target"
kind="imagery"
caption="Before: the hand-tuned model (left) against the stereo measurement (right). The pattern is right; the amounts are several times too large."
/>

<PostFigure
src="/blogs/bhotekoshi-diffcal-after.webp"
alt="Model elevation change after gradient calibration beside the stereo-measured target"
kind="imagery"
caption="After: the gradient-calibrated model. Amplitudes now sit at the measured scale; the remaining mismatch is spatial detail six scalar parameters cannot express."
/>

Where it stopped is as informative as where it got. After iteration fifteen the loss went flat while parameters drifted: friction pinned against its lower bound, critical shear wandering without improvement. That is what exhausting a parameterisation looks like. Six numbers cannot express that friction differs between bedrock gorge and gravel fan, or that erodibility varies along fifteen kilometres of valley. The obvious next step, spatially varying parameter fields with appropriate regularisation, is exactly what gradient-based calibration is uniquely equipped to handle, and exactly what grid search never could.

## The model that argued with its data

One thread from earlier in this series belongs in the record here. Before the stereo measurement existed, this model was calibrated against a sediment map that later proved wrong, made with a parallax method whose viewing-geometry assumption did not hold. That target put the main deposit five kilometres upstream of where it really is. The remarkable thing, visible only in hindsight: the model never fully complied. Its physics kept carrying sediment through the gorge and dropping it at the valley opening, centred within one bin of where the corrected measurement eventually placed the real deposit. The calibration pulled; the shallow-water equations pushed back.

We do not want to romanticise this. A model can be wrong in ways that also resist your data. But it is a concrete argument for keeping real physics inside learned systems: the structure of the equations encoded knowledge about where sediment can and cannot come to rest, knowledge that survived contact with a corrupted calibration target.

## An honest appendix: the measurement that refused

Not everything we tried this week worked, and one null result deserves recording. Landsat 9 imaged the flood mid-event, and the instrument's spectral bands are captured a fraction of a second apart, a lag that has been used to measure ocean currents and drifting river ice. We attempted to measure the flood's surface velocity from that inter-band displacement. The machinery worked beautifully, down to a detection floor near half a metre per second, and measured the moving floodwater at essentially zero. The reason is structural: on a river a few pixels wide, image correlation locks onto the static banks, not the moving water between them. The technique wants open water or discrete floating objects, and a Himalayan river in a gorge offers neither. We publish the negative so the next person can spend their afternoon on something else.

Code for the simulation lives in the [open reconstruction repository](https://github.com/geo-pera/bhotekoshi-2026-reconstruction). The terrain data this calibration learned from is in the [v1.1 release](https://github.com/geo-pera/bhotekoshi-2026-reconstruction/releases) and mirrored on [Google Drive](https://drive.google.com/drive/folders/1XQt5SYDMHehH3i2lAj69e52Zb2ne7Rqm?usp=sharing).

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# The Flood You Could Hear: Bhote Koshi in the Seismic Record

> A public seismometer 17 km from the river recorded the Bhote Koshi disaster twice: the glacier collapse as a sharp spike, and then three hours of the flood itself, audible through the ground.

Published: 2026-09-01 | Author: Darcy Weedman | Reading time: 6 min

Source: https://geopera.com/blog/bhote-koshi-flood-seismology

---
<script>
import { PostFigure } from '$lib/components/charts';
</script>

## Summary

- A public broadband seismometer at Kakani, 17 to 55 kilometres from the flood corridor, recorded the 26 August disaster twice over.
- First, the glacier collapse itself: a spike ninety times above background, the signal the USGS catalogued as a magnitude 5.2 landslide.
- Then something we had not seen discussed anywhere: after the collapse signal faded, the ground vibration rose again into a broad plateau, twelve times background, and stayed elevated for roughly three hours. That is the flood itself, millions of tonnes of water and rock grinding down the valley, audible through the earth from the next district.
- The rise and fall of that vibration tracks the flood's position from our satellite-and-simulation timeline. Two unrelated instruments, one story.
- The data is public. We give the station, channel, and time window below so anyone can pull the same waveforms.

This post is part of our reconstruction of a disaster in which hundreds of people died, and we have tried to write it with the same care as the rest of the series. Casualty figures reported elsewhere in the series are those of Nepal Police and remain provisional.

## Two signals, one morning

Seismometers are usually described as earthquake instruments, but they are simply very sensitive listeners. Anything that shakes the ground hard enough, for long enough, leaves a record: trains, quarry blasts, storms at sea, and, it turns out, a valley being rearranged.

Station KKN sits on the Kakani ridge northwest of Kathmandu, operated as part of Nepal's open seismic network and freely available through EarthScope. On the morning of 26 August it was between 17 and 55 kilometres from the various reaches of the Bhote Koshi and Trishuli as the flood descended them. We pulled its vertical-channel record for the five hours around the event and filtered it to the 1 to 8 Hz band where debris flows speak.

<PostFigure
src="/blogs/bhotekoshi-kkn-tremor.webp"
alt="Seismic envelope at station KKN: a sharp collapse spike at time zero, then a broad three-hour tremor plateau, with the modelled flood front distance below"
kind="chart"
caption="Top: 1-8 Hz ground-vibration envelope at KKN, 60-second averaging, log scale. Bottom: distance from the station to the modelled flood front. Waveforms: NK.KKN via EarthScope, open data."
/>

The first signal needs little introduction. At 08:37 local time the envelope leaps to ninety times the pre-event background: the glacier collapse, the event that seismic catalogues worldwide logged as a magnitude 5.2 with the unusual annotation "landslide" rather than "earthquake". Our earlier force-inversion work found five distinct pulses inside that spike, consistent with a staged, domino-style failure rather than a single clean break.

The second signal is the one this post exists for. By twenty minutes after the collapse, the spike's coda should be fading toward quiet. Instead the envelope climbs again, reaching a broad plateau around forty to sixty-five minutes after collapse at roughly twelve times background, and then decays slowly over about three hours. There is no aftershock sequence that looks like this. This is flow tremor: the sustained, high-frequency grinding of boulders on bedrock as the flood tore through the gorges, radiating enough seismic energy to be measured tens of kilometres away.

## Reading the tremor against the timeline

The shape of that plateau is not arbitrary. The lower panel shows the distance from KKN to the flood front as our reconstruction has it: the front starts 55 kilometres away in the upper gorge, sweeps down the corridor, and passes its closest approach, about 17 kilometres, around the time it reached the Galchhi reach. The tremor rises as the flood approaches and gains power in the steepest reaches, peaks while the flow is both strong and near, and dies away as the flood attenuates on the lower river even though it remains close.

We want to be careful about what we claim here. With one good near-field station, we cannot triangulate the flood's position from seismic data alone, the way a dense network allowed for the 2021 Chamoli disaster in India. But a simple test holds up better than we expected: scale the vibration by the inverse square of the distance from the station to the modelled front, and the prediction tracks the measured envelope through the whole transit window with a log-space correlation of 0.87. A single moving source at the reconstructed position explains most of what the station heard. What the record provides, then, is an independent clock and a plausibility check on the flood's power. The vibration rises, peaks, and decays exactly when a flood following our reconstructed timeline should make it do so, and that timeline was built entirely from satellites, terrain, and hydraulics, with no seismic input after the initial collapse time.

There is one more feature we cannot yet explain: a short, discrete burst at about three hours and eight minutes after the collapse, well after the main tremor has faded. An aftershock of the collapse zone, a secondary bank failure, a slump into the new channel. We flag it as an open question, and the interferometric radar monitoring now running on the valley may eventually answer it.

## Why this matters beyond one flood

Everything in this post came from a single open station and an afternoon of analysis. Nepal's mountain valleys, like most of the Himalaya, have far more seismometers than river gauges, and the gauges are precisely the instruments a flood destroys first. The Bhote Koshi's gauges went offline at 08:40, three minutes after the collapse. The seismometer at Kakani never stopped recording.

A station that hears a collapse spike followed by a rising tremor plateau is describing a specific, dangerous thing with tens of minutes of lead time for downstream communities. That is not a new idea in the research literature, and operational systems exist for individual instrumented catchments. But the Bhote Koshi record is a clean, public, textbook example of the signature, on an unmonitored river, from an open station. If it helps make the case for listening to mountain rivers through the ground they shake, it will have earned its place in this series.

For anyone who wants the data: station KKN, network NK, channel BHZ, 26 August 2026, 02:30 to 07:30 UTC, via any EarthScope FDSN client. The processing is a bandpass and an envelope, nothing exotic. The rest of the series, including the terrain measurements the timeline rests on, is [here](/blog/bhote-koshi-stereo-elevation-model).

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# The Bhote Koshi Flood in Half-Metre Relief: What the Stereo Data Changed

> Vantor released the raw WorldView-3 stereo strips from the 2026 Nepal flood. The elevation model we built from them rewrote our sediment story, confirmed a model prediction, and counted the buildings the river took.

Published: 2026-09-01 | Author: Darcy Weedman | Reading time: 10 min

Source: https://geopera.com/blog/bhote-koshi-stereo-elevation-model

---
<script>
import { BeforeAfter, PostFigure, Callout } from '$lib/components/charts';
</script>

## Summary

- On 31 August, Vantor added the raw WorldView-3 stereo strips from 27 August to its open data bucket, with camera models attached. From them we built a half-metre elevation model of the flood corridor and differenced it against pre-event terrain.
- The measurement rewrote our sediment story, and we owe readers a correction. The parallax-based deposition map in our first post assumed the two looks view the valley from opposite directions. They do not, and that map is retracted. This post explains the error and what the real measurement says.
- What the flood actually did: it scoured its bed through every confined reach, 2 to 12 metres of floor lowering from the border gorge down, dumped a debris sheet 12 to 18 metres thick and three kilometres long where the valley opens below Syabrubesi, and has already cut a 13 to 21 metre trench back through its own deposit downstream.
- Our morphodynamic model put the deposit in the right place before the data existed to prove it, even though we had calibrated it against the measurement that turned out to be wrong.
- Building by building, where the sky was clear: 27 structures destroyed in the Syabrubesi reach, most of them lost with the riverbank itself rather than to water alone. Timure sits under cloud in the stereo pair, so its count must come from other sources.
- The elevation model, the change rasters with per-pixel uncertainty, the damage census, and a per-kilometre sediment budget are all released under the same open licence as the imagery they came from.

This post continues our [reconstruction of the 26 August flood](/blog/bhote-koshi-flood-2026-satellite-analysis). The human toll of this event is severe and still being counted; casualty figures in the earlier post are reported values from Nepal Police and remain provisional.

## First, the correction

Our first post measured sediment depth with a trick we were fond of: two images of the same valley, both map-corrected using pre-event terrain, disagree about the position of anything the flood raised or lowered, and the size of the disagreement gives you the elevation change. The trick is real and the physics is sound. But it depends entirely on knowing the viewing geometry, and we got that wrong. We assumed the two WorldView-3 strips look at the valley from opposite sides. Reading the actual camera models shows they are same-side, in-track stereo. The along-valley disagreement we converted into a sediment wedge was three times too small to carry the signal we claimed, with the sign flipped, and the map we published was dominated by small registration differences between the vendor's ortho products. The 12 million cubic metre wedge at kilometres 33 to 38 was not sediment. It was noise wearing a costume.

We found this out the honest way: a second, independent measurement refused to agree with the first. When the rigorous elevation model described below came back showing scour where the parallax map showed deposition, we re-derived the viewing geometry from the camera files, re-measured the parallax ourselves under the corrected geometry, and watched the discrepancy resolve to within a metre. Three separate checks converged on the same conclusion. The retraction is posted in the [data repository](https://github.com/geo-pera/bhotekoshi-2026-reconstruction), the invalid map layers are withdrawn, and everything that depended on them has been recomputed from the new measurement.

One lesson worth passing on: the wrong map behaved beautifully on stable ground, which is exactly why we trusted it. Vendor orthos are registered to a reference base, so quiet terrain agreed to half a metre and looked like quality. The failure lived only where the ground had changed, which is the one place a change measurement cannot be checked against itself. Independent methods are not a luxury.

## What the flood actually did to the valley

The stereo strips are a different class of input from the orthos: full-resolution imagery with camera models, which supports true photogrammetry. We refined the cameras until the two looks agreed to a quarter of a pixel, masked the clouds, matched the pair densely at the native 34 centimetres, and co-registered the resulting surface to pre-event terrain on ground the flood never touched. On that stable ground the model agrees with the reference to about a metre and a half, and every number below survived a set of checks designed to kill it: a bias gate on flat ground, exclusion of cells where the pre-event reference itself was interpolated, and a per-pixel uncertainty estimate carried through to every volume.

<PostFigure
src="/blogs/bhotekoshi-dsm-dh.webp"
alt="Map of measured elevation change along the Bhote Koshi valley floor, erosion in red and deposition in blue over a hillshade of the new terrain"
kind="imagery"
caption="Valley-floor elevation change from the 0.5 m stereo model, Syabrubesi reach · 27 Aug 2026 vs pre-event terrain · derived from WorldView-3 © Vantor, open data"
/>

Read the corridor from north to south and the flood tells its own story. Through the border gorge past Rasuwagadhi, the floor came out 8 to 12 metres lower: the flow was digging, not dumping. Down the narrow reach above Syabrubesi it kept digging, 2 to 5 metres in most sections, and it stripped the forested banks so thoroughly that they now sit several metres below the pre-event canopy surface. Then, at kilometre 40.5, the valley opens. Within a kilometre the signal flips from red to blue: a debris sheet three kilometres long and 12 to 18 metres thick in its core, exactly where our velocity measurements had the flow decelerating from 50 metres per second to 11. And below the sheet, where the valley narrows again, the strongest signal in the whole dataset: the river has already incised 13 to 21 metres, partly through its own fresh deposit, partly into the older floor.

The honest numbers on volume are lower bounds, and we will state them as such. Clouds covered most of both stereo looks, so the model sees about 45 percent of the valley floor in the two clear windows. Over the floor it can see, we measure roughly 0.9 million cubic metres of deposition and 3.2 million of erosion, with the thickest part of the deposit partly hidden under cloud. What the measurement rules out is any story in which tens of millions of cubic metres settled in the upper corridor. The bulk of the solid load went further downstream than our retracted map claimed, or is spread thinner than the resolution of a valley-scale budget.

## The model knew before we did

Here is the part we did not expect. Our morphodynamic simulation, the one that models the flood picking up and dropping sediment as it runs, had been calibrated against the retracted parallax map. The calibration tried to pull the model's deposition toward kilometres 33 to 38, because that is where the false wedge sat. The model would not fully go. Its physics kept the sediment in suspension through the confined gorge and dropped it where the flow decelerated at the valley opening, centred near kilometre 40, with its thickest predicted bin at kilometre 42.5. That is, within a bin, where the stereo model now measures the real deposit.

<PostFigure
src="/blogs/bhotekoshi-dh-profile.webp"
alt="Chart comparing measured valley-floor elevation change against the morphodynamic model prediction by distance along the river"
kind="chart"
caption="Median valley-floor elevation change by river kilometre: stereo measurement in blue, morphodynamic model in orange. The model was calibrated before the stereo data existed, against a target that put the deposit 5 km further upstream."
/>

We want to be careful not to over-claim here. The model over-deposits by roughly a factor of two, it smears some mass upstream, and being right about where is easier than being right about how much. A recalibration against the real measurement comes next and should tighten both. But the shape of the result stands: a process model, fed bad calibration data, disagreed with that data in the direction that turned out to be true. When the physics and the measurement argue, it is worth finding out why before betting on either.

## Buildings, one by one

At half-metre resolution, elevation change becomes a damage census. We intersected the model with OpenStreetMap building footprints along the corridor and classified each building by the change under it and around it, with two filters that mattered: a building only counts as buried if the open ground beside it also rose, and only counts near the river, because ten years of tree growth against an older reference produces false positives on hillside villages that no flood reached.

The result, for the reach the stereo can see: 27 buildings destroyed. Twenty-four of them went with the ground they stood on, including a block of about twenty on the Syabrubesi riverfront where three to four metres of terrace eroded away, and two buildings near kilometre 44.5 that dropped with twenty metres of bank into the new trench. Three more were removed from ground that survived. About 1,470 buildings in the same reach read as probably intact, and 1,135 could not be assessed under cloud.

Two limits, stated plainly. Timure, which day-after imagery shows badly damaged, sits almost entirely under the cloud mask, so our census is blind exactly where the damage is known to be severe; treat the counts as a floor, not a total. And OpenStreetMap coverage in the upper valley is sparse, so the census can only count buildings somebody has mapped. The full per-building GeoJSON, classes and caveats included, is in the release.

## Counting boulders from orbit

One more thing a half-metre surface makes possible: a boulder census. We ran a local-relief detector over the debris sheet and it found 4,329 individual clasts larger than a metre across, on 37 hectares of fresh deposit. The 84th-percentile diameter is about 7 metres. The largest block we can resolve is 26 metres across, a piece of mountain the size of an apartment building, carried at least several kilometres and set down on the valley floor.

Boulder sizes are more than a curiosity, because moving a block takes a calculable amount of flow. Applying Costa's empirical competence relation to the largest clasts, bin by bin along the deposit, gives transport velocities of 19 to 26 metres per second through this reach. Our superelevation measurements, made from tilted trimlines kilometres upstream, gave 50 metres per second in the gorge and 11 at Syabrubesi. The boulders slot between them, from an entirely independent line of physics. Two caveats for anyone reusing the census: at half-metre resolution, adjacent boulders merge, so the large tail is more trustworthy than the small end; and near the town, some detections are probably building debris rather than rock.

## Five days later, the river is already editing

The newest scenes in the open data bucket were collected on 1 September, five days after the flood. Cloud cover is heavy, but the Syabrubesi confluence is visible, and the comparison with 27 August shows the next chapter starting: the braided sheet the flood left behind has organised into a defined channel, cutting into the fresh deposit. The 13 to 21 metres of incision our elevation model measured downstream is the same process further along. Valleys do not keep the shape a flood leaves them in.

<BeforeAfter
before={{ src: '/blogs/bhotekoshi-syabru-27aug.webp', alt: 'Syabrubesi confluence on 27 August 2026, one day after the flood, braided debris sheet across the valley floor', width: 1066, height: 1066 }}
after={{ src: '/blogs/bhotekoshi-syabru-01sep.webp', alt: 'Syabrubesi confluence on 1 September 2026, a defined river channel re-established through the debris sheet', width: 1066, height: 1066 }}
beforeLabel="27 Aug"
afterLabel="1 Sep"
caption="The Syabrubesi confluence, one day and six days after the flood · WorldView-3 and Legion © Vantor, open data"
/>

## Take the data

Everything is released under CC BY-NC 4.0, matching the licence on the source imagery, in the [reconstruction repository](https://github.com/geo-pera/bhotekoshi-2026-reconstruction):

- The half-metre elevation models for both clear-sky reaches, in EGM2008 heights, co-registered to the pre-event terrain.
- Elevation-change rasters at 2 metres with a per-pixel uncertainty layer, the artefact masks we applied, and the per-kilometre sediment budget as CSV.
- Corrected deposition and erosion polygons, replacing the retracted layers, the building damage census for both reaches as GeoJSON, and the boulder census as CSV.
- The retraction note itself, in the repository README, with enough detail to check our reasoning.

The full release is also mirrored on [Google Drive](https://drive.google.com/drive/folders/1XQt5SYDMHehH3i2lAj69e52Zb2ne7Rqm?usp=sharing), in the same folder as the original 32 GB analysis-ready archive.

Our thanks again to Vantor for releasing the stereo strips with camera models attached after the community asked, and to Planet, Copernicus, NASA and USGS for the open data underneath everything here. If you find an error in this analysis, tell us. The fastest correction we can make is the one somebody hands us.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Could the Bhote Koshi Collapse Have Been Predicted?

> We tested every satellite method that has ever forecast an ice-rock avalanche against the 26 August 2026 Nepal collapse. The answer matters for how the Himalaya is monitored.

Published: 2026-08-31 | Author: Darcy Weedman | Reading time: 9 min

Source: https://geopera.com/blog/bhote-koshi-collapse-predictability

---
<script>
import { PostFigure, Callout } from '$lib/components/charts';
</script>

## Summary

- After [reconstructing the 26 August flood](/blog/bhote-koshi-flood-2026-satellite-analysis), we asked the question every disaster leaves behind: could anyone have seen it coming?
- We tested all four satellite precursor channels that have ever forecast an ice-rock avalanche: surface motion, radar coherence, elevation change, and thermal preconditioning.
- The mass that killed at least 289 people moved less than two metres in the hundred days before it fell, with no acceleration up to two days before failure.
- Every warning channel returned a null or a saturated signal. This collapse belonged to a class that does not telegraph.
- That is not a counsel of despair. It redirects the investment: away from forecasting individual slopes, toward screening whole ranges and fixing the downstream warning chain, where hours of usable lead time already existed and went unused.

The reconstruction post told the story of what happened. This one is about whether it had to happen by surprise, and we should say up front that we went looking hoping to find a missed warning. A missed warning would mean the next one can be caught. What we found is harder to act on, and more important to say honestly.

## What prediction would take

There is a respectable scientific tradition of forecasting slope collapses, and it rests on one empirical gift: most big failures announce themselves. A rock mass that is about to let go usually creeps first, slowly, then faster, and the speedup follows a pattern regular enough that you can plot the inverse of velocity against time, fit a line, and read off the date where it hits zero. That date is the failure. The method has called mine-wall collapses to within days.

The two most famous Himalayan-adjacent glacier disasters both followed the script. The Aru twin collapses in Tibet in 2016 and the Chamoli disaster in India in 2021 were each preceded by months to years of visible motion, metres upon metres of it, sitting unnoticed in freely available satellite archives. The retrospective papers made an uncomfortable point: the data to raise an alarm existed, and nobody was looking.

So the honest test for 26 August is simple to state. Was this collapse Aru-like, telegraphing its intent to anyone who looked? We had already [located the detachment zone](/blog/bhote-koshi-flood-2026-satellite-analysis) with radar change detection, seismic force direction, and offset tracking, all agreeing on a patch of chronically active ice at about 5,800 metres. That gave us the exact pixels to interrogate.

## The motion test

Between May and the collapse, Sentinel-2 caught the source zone cloud-free on eight days, the last of them on 24 August, two days before failure. Monsoon cloud is merciless over this range, but 5,800 metres pokes above a lot of weather.

We measured the ice mass's motion between every usable pair of those images with sub-pixel image correlation, tied to the surrounding bedrock so that satellite geolocation wobble cancels out. One trap is worth recording for anyone repeating this: comparing images from different orbits produces phantom shifts of several metres over steep terrain, because orthorectification errors depend on viewing angle. Same-orbit pairs only. Once we did it properly, the answer was stark.

<PostFigure
src="/blogs/bhotekoshi-no-creep.webp"
alt="Chart showing measured displacement of the failed ice mass between May and August 2026, all below two metres, far below the tens of metres of creep seen before comparable disasters"
kind="chart"
caption="Displacement of the failure mass across every usable same-orbit image pair, May to 24 August 2026. Every measurement sits at or below 2 m."
/>

Under two metres of total movement in a hundred days. Around two centimetres a day, constant, with no acceleration in any window we could form, including the final one ending 48 hours before collapse. For comparison, the masses at Aru and Chamoli crawled tens of metres before they went. The inverse-velocity method needs a velocity curve that bends upward. This one was a flat line until it simply ended.

We cross-checked with radar. NASA and ISRO's new NISAR satellite, in one of the luckier coincidences of this event, had imaged the source repeatedly through July and August, and its urgent-response products span the collapse itself. The radar told us two things. First, the detachment zone is chronically active: its surface churns fast enough that radar coherence there is saturated at zero in every 12 to 24 day interval we examined, all the way back to early July. You cannot watch for the onset of instability at a site that is permanently unstable at your measurement cadence. Second, radar amplitude tracking on this icefall carries a noise floor near ten metres, so it cannot tighten the optical bound; it simply agrees that nothing dramatic moved.

## The other channels

Elevation: ICESat-2's laser happened to cross the detachment zone on a repeating track, with near-identical passes in January 2024 and December 2025. Sixteen matched footprints, four metres apart on the ground, show 1.3 metres of thinning over two years. That is ordinary glacier ablation, not a mass loading up to fail.

Weather: reanalysis data at the failure elevation shows 2026 was a warm season, 44 days above freezing between mid-June and the collapse, with monsoon rain falling on thawed ice for eight straight weeks. Water working into a fractured ice-rock mass is the standard trigger mechanism, and the seismic record's five-pulse, staged failure fits a mass coming apart wet. But here is the problem with calling that a warning sign:

<PostFigure
src="/blogs/bhotekoshi-thaw-days.webp"
alt="Bar chart of days above freezing at 5,800 metres each summer from 2016 to 2026, showing 2026 tied as the warmest of the decade but with 2024 and 2025 nearly identical"
kind="chart"
caption="Days above freezing at 5,800 m over the source zone, 15 June to 26 August, each year since 2016 (ERA5). The last three summers are almost interchangeable."
/>

The two previous summers were one and two thaw-days short of identical, and nothing fell. Thermal preconditioning flags the era, and it flags a widening set of Himalayan summers every year. It cannot flag the slope or the week.

<Callout type="info" title="The result in one sentence">
Across four independent channels, motion, coherence, elevation and climate, the 26 August collapse produced no detectable precursor that current satellite methods could have converted into a warning: it was a brittle failure of a chronically active ice mass, and it did not telegraph.
</Callout>

## What this means for the Himalaya

It would be easy to read this as fatalism. It is the opposite. A negative result this clean tells you where the effort actually pays, and that is in two places.

The first is screening rather than forecasting. We could not have predicted the day, but the site itself was findable in advance. It sits in radar data as a patch of chronic, violent surface activity, hanging at the top of a steep couloir, above a river, above villages, in a catchment that had already produced a disaster fourteen months earlier. NISAR's open L-band data now makes it possible to map every such chronically active ice mass across the entire Himalaya, seasonally, at modest compute cost. That produces a shortlist: not "this slope will fail on Thursday" but "these two hundred couloirs are the ones that can do this, and these thirty have people under them." Prevention budgets, sensor deployments and land-use decisions can work with a list like that.

The second is the warning chain, and this is where the arithmetic turns brutal. The collapse was detected within minutes; it registered on the global seismic network as its own magnitude 5.2 event. The wave reached the border in 17 minutes, which no siren system can honestly promise to beat. But most of the people who died were far downstream, where the same wave took two to five hours to arrive. Galchhi had 145 minutes. The Chitwan reaches had more. Those hours existed physically. They were lost institutionally, in the gap between a seismometer knowing and a riverbank knowing, a gap widened when the river gauges were destroyed by the very wave they were meant to report. An automated link from seismic detection to downstream sirens is boring, proven technology. On 26 August it was worth more than every satellite in this story combined.

## Data and methods

The displacement series is built from Copernicus Sentinel-2 imagery, the radar analysis from NISAR provisional and urgent-response products (an extraordinary release by the NISAR team, weeks after commissioning), the elevation check from ICESat-2 ATL06, and the climate series from ERA5. All are free and open. Our measurements, the co-registered and analysis-ready intermediate data, and the full reconstruction that this study builds on are in the [project repository](https://github.com/geo-pera/bhotekoshi-2026-reconstruction) and the [data archive](https://geopera.com/blog/bhote-koshi-flood-2026-satellite-analysis) from the first post, under the same open terms. Sub-pixel correlation results are sensitive to orbit geometry and co-registration choices, as described above; we would genuinely welcome independent replication, and the chips to do it with are in the archive.

Casualty figures are those reported by Nepal Police and remain provisional.

## Frequently asked questions

### Could the 2026 Bhote Koshi glacier collapse have been predicted?

Based on our analysis of four independent satellite precursor channels, very likely not with current methods. The failed ice mass moved less than 2 metres in the 100 days before collapse, with no acceleration up to two days before failure, unlike previous forecastable events such as Aru (2016) and Chamoli (2021), which crept tens of metres before failing.

### What usually warns of an ice-rock avalanche?

Most large slope failures accelerate before they fail. Satellite image correlation or radar interferometry can measure that creep, and inverse-velocity analysis can then estimate a failure date. The method works when failures telegraph. The 2026 Bhote Koshi collapse did not.

### Was climate change a factor?

The 2026 melt season was the joint warmest of the past decade at the failure elevation, with eight weeks of sustained thaw and rain before the collapse, which fits the standard trigger mechanism for ice-rock failures. But the two previous summers were nearly as warm without a disaster, so warmth alone could not have identified the timing. Warming raises the base rate of these events across the whole range.

### What would actually reduce deaths from events like this?

Two things our analysis supports: range-wide screening for chronically active ice masses above settlements, now feasible with open NISAR radar data, and automated links from seismic detection to downstream sirens. Most victims on 26 August were hours downstream of the collapse; the detection existed within minutes, the warning never arrived.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Bhote Koshi Flood 2026: A Reconstruction from Open Satellite Data

> How open satellite imagery reconstructed the 26 August 2026 Nepal flood: flow heights, velocities, sediment depths, and an interactive 3D model.

Published: 2026-08-28 | Author: Darcy Weedman | Reading time: 13 min

Source: https://geopera.com/blog/bhote-koshi-flood-2026-satellite-analysis

---
<script>
import { StatStrip, BeforeAfter, PostFigure, Timeline, Callout, FloodViewer3D } from '$lib/components/charts';
</script>

<Callout type="warning" title="Correction, 1 September">
The sediment-depth measurement in this post (the 10 to 18 metre wedge and the ~12 million cubic metre figure) has been retracted: the parallax method used the wrong viewing geometry. The corrected analysis, from a half-metre stereo elevation model, is in [a follow-up post](/blog/bhote-koshi-stereo-elevation-model) — the real deposit sits further downstream, and the full data is now released.
</Callout>

## Summary

- On 26 August 2026 a glacier collapse above the Lende Khola sent roughly 100 million cubic metres of debris-laden water down the Bhote Koshi and Trishuli rivers in Nepal. At least 289 people are confirmed dead as of 28 August, with more than 800 still out of contact, most of them far downstream between Dhading and Chitwan.
- Working only from openly licensed satellite data, we reconstructed the event in about 72 hours: flow heights of 70 to 134 metres in the gorges, velocities of 37 to 52 metres per second, and 10 to 18 metres of new sediment on the valley floor.
- The trigger is in the USGS seismic catalogue: a magnitude 5.2 signal at 08:37 local time, catalogued as a landslide, not an earthquake.
- Day-after WorldView-3 stereo imagery let us measure where the debris settled without anyone setting foot in the valley.
- Everything is downloadable: the source imagery, our co-registered and terrain-corrected rasters, every measurement, the model code, and the interactive 3D model at the top of this page.

**Download the data**: [full archive, 32 GB, on Google Drive](https://drive.google.com/file/d/1eXzkWafRG4_QZoufwqbv3JSJxW51cv0a/view?usp=sharing) · [derived data and measurements, 335 MB](https://github.com/geo-pera/bhotekoshi-2026-reconstruction/releases/download/v1.0/bhotekoshi-2026-flood-data.zip) · [code and methods](https://github.com/geo-pera/bhotekoshi-2026-reconstruction). Licences and checksums are in the Take the Data section below.

This post documents a scientific reconstruction of a disaster in which many people died and many more are still missing. We have tried to write it the way we would want to read it: carefully, with the uncertainty stated plainly, and without turning a tragedy into a promotion. Casualty figures are those reported by Nepal Police and are provisional.

<FloodViewer3D
dataUrl="/interactive/bhotekoshi-event.json"
depthMax={45}
clockStartMin={530}
legend="flow depth"
caption="The simulated flood over real terrain: 8 m shallow-water model, WorldView-2 imagery drape © Vantor. Drag to orbit, scroll to zoom, scrub or play the timeline."
/>

## What Happened on the Morning of 26 August

At 08:37 Nepal time, a piece of glacier roughly 600 metres wide let go from the north flank of the Langtang Himal, at about 5,600 metres. It fell into the headwaters of the Lende Khola, a small river that joins the Bhote Koshi at the Nepal-China border crossing of Rasuwagadhi.

Here is the detail that reframes the whole event. Seismometers on the other side of the planet felt it. The [USGS catalogue](https://earthquake.usgs.gov/) lists a magnitude 5.2 event at 28.271 N, 85.515 E at that exact moment, and the catalogue entry does not say earthquake. It says landslide. Early reporting had a quake triggering the collapse. The seismic record says the collapse WAS the shaking.

Falling ice makes a strange kind of flood. Friction melts it in transit. The melt grabs sediment, the sediment thickens the flow, the flow moves faster than clean water ever could. By the time it reached the border it was a wall. Twenty-three minutes, start to finish, from the seismic signal to the destruction of the customs complex.

<Timeline
title="26 August 2026, Nepal time"
events={[
{ date: '08:37', title: 'Glacier collapse', body: 'M 5.2 landslide-type seismic event, upper Lende Khola. Recorded by USGS.' },
{ date: '~09:00', title: 'Rasuwagadhi border crossing destroyed', body: 'Customs complex, friendship bridge and hydropower works swept away.' },
{ date: '~09:10-09:25', title: 'Timure and Syabrubesi', body: 'Riverside settlements inundated within minutes of the front arriving.' },
{ date: '~10:30', title: 'Peak passes Betrawati', body: '50 km downstream of the source.' },
{ date: '~11:00-12:00', title: 'Gauged reaches', body: 'Reported rises of 9 m in 30 minutes at Galchhi and 7 m at Malekhu.' }
]}
/>

Six hydropower stations gone along 60 kilometres of valley. And a grim piece of context: this same catchment flooded in July 2025, when a lake hiding on the Purepu Glacier drained without warning. Fourteen months, two disasters, one valley.

## What the Satellites Saw

We want to be straight about the limits before showing the pictures, because the limits shaped everything.

It was monsoon season. Cloud cover on the day of the flood ran 62% to 93% across the optical passes, and one 15-kilometre stretch of the corridor, including Rasuwagadhi itself, had exactly zero cloud-free pixels on day one. We checked every scene. Zero. If you see anyone claiming a complete day-of optical picture of this event, they are describing something that does not exist.

What saved the analysis was the day after. Planet's crisis response programme published PlanetScope from the morning of the flood, then SkySat and Pelican collections at 0.8 and 0.55 metres the next day. Vantor's open data programme released four WorldView-3 strips at roughly 30 centimetres, and, in the detail that mattered most, two of them look at the valley from opposite directions. More on why that matters below.

<StatStrip
stats={[
{ value: '30', label: 'satellite scenes analysed' },
{ value: '0.3-3.8 m', label: 'resolution range' },
{ value: '~72 h', label: 'event to full reconstruction' },
{ value: '100%', label: 'openly licensed data' }
]}
caption="Planet Labs PBC and Vantor open data programmes (CC BY-NC 4.0), NASA NSIDC, Copernicus."
/>

The clearest way to grasp the scale is to move a slider across it. Left of the handle is October 2021. Right is the day after the flood.

<BeforeAfter
before={{ src: '/blogs/bhotekoshi-border-pre.webp', alt: 'Rasuwagadhi border crossing in October 2021, WorldView-2 satellite image', width: 1100, height: 1149 }}
after={{ src: '/blogs/bhotekoshi-border-post.webp', alt: 'Rasuwagadhi border crossing on 27 August 2026 after the flood, WorldView-3 satellite image', width: 1100, height: 1149 }}
beforeLabel="Oct 2021"
afterLabel="27 Aug 2026"
revealedName="the day after the flood"
caption="Rasuwagadhi border crossing, Rasuwa District · Oct 2021 vs 27 Aug 2026 · 0.5 m / 0.3 m · © Vantor, open data"
/>

The customs complex, the bridge, the hydropower intake: gone. That grey sheet entering from the upper right is the Lende Khola. It used to be a river maybe 30 metres wide. It is now a debris surface wider than the settlement it destroyed.

<BeforeAfter
before={{ src: '/blogs/bhotekoshi-timure-pre.webp', alt: 'Timure village before the flood, satellite image from 2021', width: 1100, height: 1149 }}
after={{ src: '/blogs/bhotekoshi-timure-post.webp', alt: 'Timure village after the 2026 Bhote Koshi flood', width: 1100, height: 1149 }}
beforeLabel="Oct 2021"
afterLabel="27 Aug 2026"
revealedName="the day after the flood"
caption="Timure, Rasuwa District · Oct 2021 vs 27 Aug 2026 · 0.5 m / 0.3 m · © Vantor, open data"
/>

<BeforeAfter
before={{ src: '/blogs/bhotekoshi-syabru-pre.webp', alt: 'Syabrubesi before the flood, satellite image from 2021', width: 1100, height: 1146 }}
after={{ src: '/blogs/bhotekoshi-syabru-post.webp', alt: 'Syabrubesi after the 2026 Bhote Koshi flood', width: 1100, height: 1146 }}
beforeLabel="Oct 2021"
afterLabel="27 Aug 2026"
revealedName="the day after the flood"
caption="Syabrubesi, Rasuwa District · Oct 2021 vs 27 Aug 2026 · 0.5 m / 0.3 m · © Vantor, open data"
/>

Look at Syabrubesi for a moment. The houses that survived are the ones sitting higher on the slope. That is not a metaphor. It is an elevation contour, drawn in buildings.

## Measuring a Flood Nobody Could Reach

Field teams could not get into most of this corridor for days. So every number in this post had to come from orbit. Three measurements carried the analysis, and each one is a trick worth knowing.

**Flow height, from the bathtub ring.** A debris flow scours vegetation and plasters mud up to its peak level, and that line survives on the valley walls after the water is gone. Hydrologists call it a trimline. We mapped the stripped zone from before-and-after imagery, draped it over a corrected elevation model, and read off the height at 217 individual bank positions. The median through the gorges is about 70 metres. At the border, individual measurements run 40 to 134 metres. One caution we will repeat because it matters: a trimline records the peak, splash and runup included, so treat these as upper bounds on the water itself.

**Speed, from tilted water.** Water going around a bend banks like a motorbike, higher on the outside wall than the inside. The tilt is preserved in the trimlines on the two banks, and a decades-old formula turns that tilt, the bend radius, and the channel width into a velocity. The chain of estimates reads: 37 metres per second in the Lende Khola, 45 to 52 at the border, 50 in the gorge, and then a sudden drop to 11 where the valley opens at Syabrubesi. That collapse in speed is the flood hitting the brakes, and you will see in the next measurement exactly where the braking energy went.

**Sediment depth, from two photos that disagree.** This one is our favourite, and it needs an analogy. Photograph a table from the left, then from the right. If someone slips a book under the tablecloth between your two shots, the bump sits in a different place in each photo. The size of the disagreement tells you the thickness of the book. Two of the WorldView-3 strips view the valley from opposite directions, and both were map-corrected using terrain from BEFORE the flood, so wherever the flood changed the ground, the two images disagree. We matched 20,500 points between them and converted the disagreements into elevation change. The result: a wedge of new valley floor, 10 to 18 metres thick, through the gorge where the flow decelerated, thinning to about 4 metres at Syabrubesi. Roughly 12 million cubic metres of rock and mud, weighed from space.

<PostFigure
src="/blogs/bhotekoshi-deposition.webp"
alt="Map of measured elevation change along the Bhote Koshi showing sediment deposition in brown"
kind="imagery"
caption="Stereo-parallax elevation change, Bhote Koshi corridor · 27 Aug 2026 · derived from WorldView-3 © Vantor, open data"
/>

## What Each Dataset Actually Did

A reconstruction like this is a relay, not a single hero sensor. Every dataset covered a specific blindness in the others.

| Dataset | What it gave the model | What it could not do |
| --- | --- | --- |
| USGS seismic catalogue | The trigger's location, size, and the exact second the clock starts | Nothing about the water |
| PlanetScope (3.8 m, day of) | Flood extent in the cloud gaps, hours after the peak | 62-93% cloud; blind over the border reach |
| SkySat + Pelican (0.8/0.55 m, day 1) | Sharp post-event ground detail; filled the day-one gaps | Still cloudy; no height information |
| WorldView-3 stereo (0.3 m, day 1) | Sediment depths via parallax; trimlines at the border | Its own quality masks flagged clear ground as cloud, so we masked it ourselves |
| WorldView-2 archive (2021) | The before picture every comparison depends on | Predates the July 2025 flood in this valley |
| NASA HMA 8 m DEM | The terrain every height and every simulation stands on | From 2017 imagery; ellipsoidal heights needed datum correction; 12% voids near the channel |
| Copernicus GLO-30 | Filled the DEM voids; cross-checked the datum fix | Too coarse to carry the analysis alone |
| Sentinel-2 | We tried to extend coverage downstream with it | Joint clear-sky with the pre-event pass: 2.1%, none of it on the river. A documented failure |
| OpenStreetMap | Building footprints for impact counts; the Lende Khola's own name | Building heights are guesses |
| Reported gauge readings | The 9 m rise at Galchhi that calibrated the model's volume | Media-relayed figures, not raw agency records |

Two of those rows deserve a sentence each. The Sentinel-2 row is in there because negative results belong in the record: we burned an afternoon proving the monsoon beat us on that axis, and anyone repeating this work deserves to know before they burn theirs. And the WorldView row hides the single most useful accident of the week: those strips were collected as a stereo pair. Without that accident there is no sediment map at all.

## The Model, and Whether to Believe It

We routed the flood down 611 measured river cross-sections in a one-dimensional model, then ran a two-dimensional model of the Syabrubesi gorge on top. One unknown mattered: how much material came down. Two observations that know nothing about each other pin it. The reported 9-metre rise at the Galchhi gauge, 85 kilometres from the source. And the 48-to-70-metre band of trimline heights in the gorge, 35 kilometres upstream of that gauge. One volume satisfies both at once: about 100 million cubic metres, give or take 40%.

Then the model did the thing you hope for and rarely get. We calibrated it before the day-two imagery existed, and it predicted a 90-metre peak stage at Rasuwagadhi, sight unseen. The WorldView strips arrived the next morning. The trimlines at the border measured 40 to 134 metres, maximum 134. The prediction sat inside the observed envelope, made before the observation. That is the difference between a model that explains and a model that anticipates, and it is the main reason we are willing to publish numbers this young.

The same test buried one early theory. A popular explanation held that the avalanche dammed the river, a lake built up, and the dam burst. We simulated it. A sustained dam-and-breach arrives at the border 30 to 60 minutes too late, rises far too gently for the eyewitness accounts of a wall, and peaks at a third of the measured heights. What the data cannot distinguish is a direct debris flow from a blockage that failed within a couple of minutes. Both fit everything we can measure. Honesty requires leaving that open.

<PostFigure
src="/blogs/bhotekoshi-model-validation.webp"
alt="Chart comparing modelled peak stage along the river corridor with measured trimline heights"
kind="imagery"
caption="Modelled peak stage vs 109 measured trimline heights along the corridor, with stage curves at four sites"
/>

<figure>

![Animated simulation of the flood wave travelling down the Bhote Koshi and Trishuli corridor](/blogs/bhotekoshi-corridor-wave.gif)

<figcaption>Simulated flood wave, source to Galchhi, on a Nepal-time clock. Depth exaggerated 20x so a 50 m wave stays visible against 5,000 m of relief.</figcaption>

</figure>

<Callout type="warning" title="An active situation, updated 31 August">
The barrier lake is confirmed. A Sentinel-2 pass on 29 August caught it in a cloud gap: a sediment-laden pond of about 0.06 square kilometres (roughly 240 by 490 metres) at 85.510 E, 28.292 N, about 2.4 km from the collapse source, at the location Chinese authorities reported on 27 August with an area of 0.11 square kilometres. The comparison suggests it may be shrinking, and NASA-ISRO NISAR radar from 28 August independently shows the site as freshly changed surface. At the measured size the impounded volume is of order 0.5 to 1.5 million cubic metres. Routing a breach of that size through our calibrated model, the wave is largely absorbed refilling the scoured Lende channel and stalls about 3 km short of the border; even a flashier release should stay within the one to a few metres our larger scenarios give at Rasuwagadhi. The main danger zone for a breach at the current size is the Lende channel itself, and the lake can grow, so this is a snapshot, not a forecast. That is still lethal for anyone on the riverbed. Follow NDRRMA guidance, not blog posts.
</Callout>

## Take the Data

Everything is packaged and free, and we have done the unglamorous preparation so you do not have to: all rasters are co-registered onto one grid (EPSG:32645), the elevation model is datum-corrected and void-filled, every measurement carries its uncertainty, and the co-registration audit ships as a document beside the data. If you have ever lost a week discovering that your DEM is in ellipsoidal heights while your gauge data is orthometric, you know why we mention this. That week is already spent. It was ours.

- **Full archive (32 GB)**: source imagery, all rasters, measurements, models, documentation. [Download from Google Drive](https://drive.google.com/file/d/1eXzkWafRG4_QZoufwqbv3JSJxW51cv0a/view?usp=sharing). SHA-256 in the manifest alongside.
- **Just the science (335 MB)**: measurements, model outputs, masks, docs. [Download from the repository release](https://github.com/geo-pera/bhotekoshi-2026-reconstruction/releases/download/v1.0/bhotekoshi-2026-flood-data.zip).
- **Code and methods**: the full reconstruction pipeline is public at [github.com/geo-pera/bhotekoshi-2026-reconstruction](https://github.com/geo-pera/bhotekoshi-2026-reconstruction).
- **Source imagery programmes**: [Planet Crisis Response](https://source.coop/planet/disasterdata/nepal-flash-flood-2026-08-26) and Vantor open data (s3://vantor-opendata). A broader guide to [free satellite data sources](/blog/free-sources-of-satellite-data) is maintained separately.

Derived products inherit CC BY-NC 4.0 from the source imagery. If you are working on the response or on research into this event and data processing is your bottleneck, write to us and we will help, without charge. Community groups are already maintaining crisis datasets for this event; they deserve the traffic.

Ten years ago this analysis was a funded research project and a season of work. It took us three days, and the reason is not cleverness. The reason is that Planet published within a day, Vantor released 30-centimetre stereo for free, NASA's terrain was already sitting there, and a seismometer network logged the trigger to the second. Open data did not make the flood less terrible. It made the flood measurable, fast, by anyone. That seems worth building on.

## Frequently Asked Questions

### What caused the 2026 Bhote Koshi flood in Nepal?

A collapse of glacier ice and rock, roughly 600 m wide, from about 5,600 m on the Langtang Himal at 08:37 on 26 August 2026. USGS catalogued the collapse itself as a magnitude 5.2 landslide-type seismic event. It was not triggered by an earthquake.

### How high did the flood reach?

Trimline measurements from satellite imagery show a median flow height near 70 metres through the confined gorges, with individual bank measurements from 40 to 134 metres at the Rasuwagadhi border crossing. These are peak-stage values and include splash and runup.

### Was it a glacial lake outburst flood (GLOF)?

Current evidence says no pre-existing lake drained. The flow's water came from melted avalanche ice, entrained sediment and the river itself. A short-lived blockage that failed within minutes cannot be ruled out; a sustained dam-and-breach sequence is excluded by timing and flow heights.

### Is there satellite imagery of the Nepal flood?

Yes, and most of it is free. Planet's crisis response dataset and Vantor's open data programme both published pre- and post-event imagery, from 3.8 m down to 30 cm, under CC BY-NC 4.0 licences. Direct downloads are in the Take the Data section above.

### How was sediment depth measured without field access?

From the parallax between two WorldView-3 images taken from opposite look angles. Both were orthorectified against pre-event terrain, so surface changes displace the images relative to each other. Dense image correlation converted that displacement into 10 to 18 metres of measured deposition.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# What Is GeoLibre? The Free Cloud-Native GIS, Explained

> What GeoLibre is, what it runs on, and what it can actually do. The free open-source cloud-native GIS explained, including what it does not do yet.

Published: 2026-08-21 | Author: Darcy Weedman | Reading time: 11 min

Source: https://geopera.com/blog/geolibre-explained

---
<script>
import { AskBand, BarChart, Timeline, Callout } from '$lib/components/charts';
</script>

## Summary

- **GeoLibre is a free, MIT-licensed GIS application**, not a library. It's built on Tauri v2, React, MapLibre GL JS, DuckDB-WASM Spatial and deck.gl, and it runs in a browser tab, as a desktop app on Windows, macOS and Linux, as native iOS and Android apps, and inside Jupyter
- It comes from the [opengeos](https://github.com/opengeos/GeoLibre) project, led by Qiusheng Wu at the University of Tennessee, Knoxville. First release 28 May 2026, v1.0.0 thirteen days later, **v2.6.0 by 14 August 2026**
- **1,066 geoprocessing tools run locally through WebAssembly**, including 154 remote sensing tools, so analysis happens on your machine with no server and no upload
- One `.geolibre.json` project file opens in every runtime. It also **imports QGIS `.qgs`/`.qgz` and ArcGIS Pro `.aprx`/`.mapx` projects**, rebuilding layers, groups, styling and the saved view
- It is not finished. Real-time collaboration is labelled MVP by its own authors, several tools are desktop only because they need a local Python process, and the Mac App Store build trades away more still

Search for GeoLibre and one of the first things you'll read is that it's a Python library.

It isn't. GeoLibre ships a Python package, which is a different claim entirely, and the distance between those two sentences turns out to be most of what's interesting about the project. If you're trying to work out what this thing actually is before you spend an afternoon on it, start there.

<AskBand headline="What will you put on the map?" sub="Free Sentinel-2 with spectral indices built in, and commercial imagery from 30 cm when you need the detail." action="Try Pera Portal" href="https://portal.geopera.com" tiles={['/band/fields.webp', '/band/oxbow.webp', '/band/saltponds.webp']} />

## What GeoLibre Actually Is

GeoLibre is a full desktop-class GIS that happens to run in a web browser. Layers panel, style rail, attribute tables, expression builder, print composer, processing toolbox: the furniture a GIS analyst expects, all of it, in a tab.

It's built with Tauri v2, React and TypeScript, with MapLibre GL JS doing the 2D rendering, deck.gl handling 3D and large point data, DuckDB-WASM Spatial running the SQL, and CesiumJS available for a true 3D globe. The Tauri wrapper is why the same codebase installs as a native application on Windows, macOS, Linux, Android and iOS. It's MIT licensed, so free in both senses.

The project moves fast enough that any version number in this post will be stale before long.

<Timeline
title="GeoLibre release history"
events={[
{ date: '28 May 2026', title: 'v0.1.0', body: 'First public release. The repository was created the day before.' },
{ date: '10 June 2026', title: 'v1.0.0', body: 'Thirteen days after the first release.' },
{ date: '11 July 2026', title: 'v2.0.0', body: 'Mobile builds, expanded processing, plugin system.' },
{ date: '14 August 2026', title: 'v2.6.0', body: '27 releases in 78 days. 6,500+ GitHub stars, 640+ forks.' }
]}
/>

## The Python Library Confusion, Cleared Up

This matters because it changes what you'd go and do next.

There is a `geolibre` package on PyPI, and you can `pip install geolibre`. But the package doesn't reimplement GIS operations in Python the way GeoPandas or Rasterio do. It's an anywidget bridge. GeoLibre's own documentation describes it plainly: the widget **"loads the complete GeoLibre app (menus, panels, processing tools) in an iframe"**, and state syncs both ways through a single `.geolibre.json` project.

So data you add from Python appears in the interface, and pans, zooms and layer edits you make in the interface read back into Python. It's the application, embedded, with a scripting handle attached. There's an R package that does the same job for RStudio, Quarto and Shiny, and an MCP server (`geolibre-mcp`) that authors project files headlessly with **"no browser, no running app, and no bundled web build."**

<Callout type="info" title="The short version">

Calling GeoLibre a Python library is roughly like calling QGIS a Python library because PyQGIS exists. The bindings are real and useful. They aren't the thing.

</Callout>

## Five Surfaces, One Project File

The unusual part isn't that GeoLibre runs in several places. Plenty of software does. It's that one `.geolibre.json` file is genuinely portable across all of them: build a map in a notebook, save it, reopen it on a phone.

What you give up varies by surface, and the differences are honest ones rather than marketing tiers.

| Surface           | How you get it                       | WASM toolbox | Notable limits                                              |
| ----------------- | ------------------------------------ | ------------ | ----------------------------------------------------------- |
| Browser           | web.geolibre.app, nothing to install | Yes          | No local file dialogs, MBTiles or filesystem save            |
| Desktop           | Installers, winget, Homebrew Cask    | Yes          | The fullest build, including the Python sidecar              |
| Mac App Store     | Sandboxed macOS build                | Yes          | No Python sidecar, local Jupyter, Earth Engine or plugins    |
| iOS and Android   | App Store, Google Play               | Yes          | Sidecar tools hidden; "the iOS sandbox forbids spawning one" |
| Jupyter, R        | pip, conda, CRAN                     | Yes          | Embeds the app; needs a kernel to host it                    |

The browser build is a progressive web app with a Download Offline Area tool that pre-caches the current view's basemap tiles, and it service-worker caches Pyodide and PGlite so SQL and Python keep working offline after first use. That's a genuinely odd sentence to be able to write about a web page.

## What's Actually Inside It

The processing toolbox is the part most people underestimate. GeoLibre bundles the Whitebox suite compiled to WebAssembly and runs it locally, which means no Python environment, no server round-trip, and nothing uploaded.

<BarChart
title="Whitebox WASM tools by category"
horizontal={true}
showValues={true}
data={[
{ label: 'Vector', value: 313 },
{ label: 'Raster', value: 256 },
{ label: 'Remote sensing', value: 154 },
{ label: 'Hydrology', value: 100 },
{ label: 'Terrain', value: 99 },
{ label: 'LiDAR', value: 65 }
]}
/>

Conversion (49), network (26) and projection (4) bring the total to 1,066. The remote sensing set is the one worth knowing about if you work with imagery: a dedicated spectral index toolbox covering NDVI, GNDVI, NDWI, NDMI, NDBI, NBR, EVI and SAVI, with band layouts preset for Sentinel-2, Landsat 8-9 and NAIP. We've written separately about [what those vegetation indices actually measure](https://geopera.com/blog/remote-sensing-vegetation-indices).

Beyond processing, the feature list runs long: rule-based renderers, a data-driven label engine, virtual fields, an expression builder shared across filters and styling, a SQL workspace running DuckDB Spatial with PGlite and Apache Sedona as alternative engines, story maps, dashboards with cross-filtering widgets, a print composer with atlas generation, field collection that reads an external NMEA GPS receiver over Web Serial or Web Bluetooth, and 18 translation catalogues including right-to-left Arabic and Persian.

There's also a plugin system, with browsers for STAC catalogues, OpenAerialMap, Overture Maps, USGS LiDAR and NASA's Earthdata GIS portal.

## How It Sits Next to QGIS and ArcGIS

Not as a replacement, at least not yet, and the project doesn't really claim otherwise.

QGIS has two decades of plugins, a vast body of documentation, and the accumulated trust that comes from being the thing everyone already knows. Anything unusual you need to do, someone has probably written a QGIS plugin for it. That's very hard to compete with, and GeoLibre isn't trying to on those terms.

What GeoLibre has instead is zero install, a project you can share as a URL, cloud-native formats treated as first-class rather than bolted on, and analysis that runs client-side on a phone. The migration path is unusually considerate too: it imports QGIS `.qgs` and `.qgz` projects, rebuilding layers, nested groups, group visibility, layer order, styling and the saved map view, and reports per-layer why anything was skipped. ArcGIS Pro `.aprx` and `.mapx` projects import as well, read straight from CIM JSON without ArcPy.

Most people who try it will end up running it alongside what they already have. That's the honest answer, and it's fine. If you want the desktop version of the same imagery question, our [QGIS imagery guide](https://geopera.com/blog/qgis-satellite-imagery) covers that canvas.

## What It Doesn't Do

Any explainer that skips this section is selling you something.

**Real-time collaboration is an MVP.** The project labels it that way itself. Per-participant permissions, chat and a session roster exist, and snapshots are portable, but treat it as early.

**Several tools need a local Python process, so they vanish on mobile.** Raster tools, conversion, AI segmentation and the PostgreSQL/Martin tools all depend on the desktop sidecar. The docs are blunt about why on iOS: the sandbox forbids spawning one.

**The Mac App Store build gives up more.** No Python sidecar, no server-backed PostGIS, no local Jupyter server, no Earth Engine sign-in, and no external plugin installs. The in-browser SQL engines survive. If you want everything, take the signed installer rather than the Store version.

**Big local files come with warnings, not magic.** Large vector layers render through client-side vector tiling with a warning before loading very large files, and Apache Iceberg tables load capped by a row limit so something bigger than the browser can hold still opens as a usable subset. Sensible engineering, but it's a ceiling.

**A few features need a remote host.** GDAL export, ONNX object detection, story map HTML export and Pyodide reach outside. There's a `GEOLIBRE_NO_EXTERNAL_CDN=1` build for locked-down deployments that reports this up front instead of failing mid-run. The Cesium 3D globe needs your own Ion token.

## Where Geopera Fits

We don't build GIS software. We sell the pixels that go in it, so our interest here is narrow and worth stating plainly: a cloud-native GIS only pays off if the imagery is written to match.

GeoLibre reads range requests for the tile covering your view rather than the whole scene. That saving evaporates if a provider hands you a flat GeoTIFF, because the entire file has to come down first. Every Geopera order is delivered as a **Cloud Optimized GeoTIFF**, orthorectified, pansharpened and atmospherically corrected at no extra cost, with a STAC 1.0 catalogue and WMTS tiles behind an API key. Paste a URL, and it streams.

We've written the full version of that argument, with the byte arithmetic, in our guide to [adding satellite imagery to GeoLibre](https://geopera.com/blog/geolibre-satellite-imagery). If free data will do the job, our roundup of the [best free sources of satellite data](https://geopera.com/blog/free-sources-of-satellite-data) is the better starting point, and we'd rather you use it than overspend.

An open-source GIS that runs anywhere is worth paying attention to. What you point it at is still your call.

## Frequently Asked Questions

### What is GeoLibre?

GeoLibre is a free, open-source, cloud-native GIS application built on MapLibre GL JS and Tauri v2. It runs in a web browser, as a desktop app on Windows, macOS and Linux, as native iOS and Android apps, and inside Jupyter notebooks.

### Is GeoLibre free?

Yes, and it's MIT licensed. There's no account, no subscription and no paid tier. Data and basemaps you load carry their own separate licences, which is usually where the actual restrictions live rather than in the software.

### Is GeoLibre a Python library?

No. There's a `geolibre` package on PyPI, but it's an anywidget bridge that embeds the full application inside Jupyter and syncs state through a project file. The GIS itself is a Tauri and React application, not a Python library.

### Can GeoLibre open QGIS projects?

Yes. It imports `.qgs` and `.qgz` files, rebuilding layers, nested groups, visibility, layer order, styling and the saved map view, and it reports per layer why anything was skipped. ArcGIS Pro `.aprx` and `.mapx` projects import too.

### Can GeoLibre replace QGIS?

For many jobs, yes, and it's genuinely faster to reach. QGIS still wins on plugin depth and on the unusual workflows people have built over twenty years. Most people will run both rather than choosing, especially while GeoLibre is this young.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# How to Add Satellite Imagery to GeoLibre (2026 Guide)

> Four ways to add satellite imagery to GeoLibre, and why a cloud-native GIS only pays off when your imagery ships as a proper COG.

Published: 2026-08-21 | Author: Darcy Weedman | Reading time: 9 min

Source: https://geopera.com/blog/geolibre-satellite-imagery

---
<script>
import { AskBand, FetchProfile, Callout } from '$lib/components/charts';
</script>

## Summary

- **GeoLibre loads satellite imagery four ways**: a built-in basemap, a remote COG URL, a STAC catalogue search, or a local raster dragged onto the map. The remote-URL path is the one that changes how you work
- New to the tool? Start with [what GeoLibre is](https://geopera.com/blog/geolibre-explained). This post assumes you already have it open
- Its raster reader **"reads range requests for the tile containing the pixel rather than the whole scene."** So a 1.15 GB image can paint from roughly 4.6 MB of traffic
- That saving is not free. It needs a **real Cloud Optimized GeoTIFF**: internally tiled, with an overview pyramid. Point GeoLibre at a flat GeoTIFF and the whole file comes down before you see anything
- **1,000+ geoprocessing tools run locally in WebAssembly**, 154 of them remote sensing. Band math and change detection now happen in the browser tab, with no server and nothing uploaded

A GIS that opens a 1.15 GB satellite scene by pulling about 4.6 MB of it sounds like a rounding error. Or a lie.

It's neither, and the trick behind it decides whether GeoLibre is a genuine upgrade to your workflow or an annoying way to watch a progress bar. If you want to add satellite imagery to GeoLibre and have it feel instant, the app is only half the story. The other half is how your imagery was written to disk, which is a sentence most providers would rather you never think about.

<AskBand headline="How fast should a scene open?" sub="Every Geopera delivery is a Cloud Optimized GeoTIFF, with a STAC 1.0 catalogue and WMTS tiles behind an API key." action="Try Pera Portal" href="https://portal.geopera.com" tiles={['/band/city.webp', '/band/delta.webp', '/band/pivots.webp']} />

## The Thirty-Second Version

GeoLibre is a free, MIT-licensed GIS that runs in a browser tab and installs as a native app on desktop and mobile. Our separate post covers [what GeoLibre is and what's inside it](https://geopera.com/blog/geolibre-explained) properly.

The pitch that matters for imagery work: your data stays on your machine. GeoLibre Web has no server account and no analytics, and files you load are processed client-side. For anyone who has ever had to explain to a client why their pre-tender site imagery got uploaded to someone else's cloud to make a map, that's not a small feature.

## The Four Ways to Get Satellite Imagery onto a GeoLibre Map

Every method comes down to one of these, and they are not interchangeable.

**1. The built-in basemaps.** Open the Layers panel and pick one. OpenFreeMap for streets, EOX Sentinel-2 cloudless for a gap-free global satellite backdrop. Zero effort, zero control. The Sentinel-2 cloudless annual layers are licensed CC-BY-NC-SA 4.0, which means non-commercial, so check that before it ends up under a paid deliverable. (There's also a planet switcher with Moon, Mars and Titan basemaps, which has nothing to do with your job and is worth ten minutes of your afternoon anyway.)

**2. A remote COG URL.** Paste an `https://` link to a Cloud Optimized GeoTIFF and GeoLibre streams it. Nothing lands on your disk and nothing needs converting first. This is the method that makes the app interesting, and most of this post is about it.

**3. A STAC catalogue.** GeoLibre speaks STAC, so you can point it at a catalogue, search by area and date, and pull the matching scenes as layers. Search and load, rather than download-then-load.

**4. A local raster.** Drag a GeoTIFF onto the map. It works exactly as you'd expect, it needs no network, and on the desktop build you also get local MBTiles and filesystem reads. When the imagery is already sitting on your drive, this is still the right answer.

| Method            | You control the date?      | Streams or downloads?      | Bands available     | Best for                          |
| ----------------- | -------------------------- | -------------------------- | ------------------- | --------------------------------- |
| Built-in basemap  | No, it's a fixed mosaic    | Streams tiles              | RGB picture only    | Context behind your vectors       |
| Remote COG URL    | Yes, it's your scene       | Streams byte ranges        | Every band in file  | Big scenes, shared projects       |
| STAC search       | Yes, filter by date        | Streams matched scenes     | Every band in file  | Finding imagery by area and time  |
| Local raster file | Yes, it's your file        | Already downloaded         | Every band in file  | Offline work, archived deliveries |

## The Part the Tutorials Skip: Your File Has to Cooperate

Method 2 is where people get burned, so it's worth understanding what the app is actually doing.

When GeoLibre opens a raster over HTTP, it doesn't ask the server for the file. It asks for specific byte ranges: the header first, to learn how the image is laid out, then only the tiles covering what's on screen at the zoom level you're viewing. GeoLibre's own documentation puts it plainly. It **"reads range requests for the tile containing the pixel rather than the whole scene."**

That behaviour depends entirely on the file being structured for it. A Cloud Optimized GeoTIFF is a normal GeoTIFF with two extra properties: the pixels are arranged in internal tiles instead of long strips, and the file carries a pyramid of pre-computed lower-resolution copies. Zoom out and the reader grabs a small overview. Zoom in and it grabs a handful of full-resolution tiles. Either way it reads a sliver.

Take a modest commercial scene: 12,000 × 12,000 pixels, four bands, 16-bit. At 30 cm that's about a 3.6 km square, a small site. Uncompressed it's 12,000 × 12,000 × 4 × 2 = 1.15 GB.

<FetchProfile
title="Opening that scene over HTTP"
caption="Segment widths show file layout, not byte share. The 1/16 overview is 750 × 750 px, so 750 × 750 × 4 × 2 = 4.5 MB by the same arithmetic, plus a header read. Uncompressed figures, for a zoomed-to-fit first view."
profiles={[
{
label: 'Flat GeoTIFF',
note: 'no internal tiling, no overviews',
readout: '1.15 GB before the first pixel paints',
segments: [
{ label: 'Header', width: 5, fetched: true },
{ label: 'Full-resolution image data', width: 95, fetched: true }
]
},
{
label: 'Cloud Optimized GeoTIFF',
note: 'tiled, with an overview pyramid',
readout: '4.6 MB before the first pixel paints',
segments: [
{ label: 'Header', width: 5, fetched: true },
{ label: '1/16', width: 8, fetched: true },
{ label: '1/4 overview', width: 17, fetched: false },
{ label: 'Full-resolution tiles', width: 70, fetched: false }
]
}
]}
/>

Same URL, same app, and the same pixels on screen at the end. The difference in what crosses the wire is 256-fold, and that number isn't a benchmark result. It's just 16², the scale factor of the overview level the viewer picked.

<Callout type="warning" title="The failure mode is silent">

Nothing errors when you hand a cloud-native GIS a flat GeoTIFF. The map eventually draws. You just sit through a long wait, on every pan, on every machine, forever, and quietly conclude that browser GIS isn't ready. The app was ready. The file wasn't.

</Callout>

This is why "we'll email you a GeoTIFF" is a worse deal in 2026 than it was in 2016, and why the delivery format line on a quote deserves more attention than it usually gets. Plenty of the industry still hands over raw or minimally processed rasters and leaves conversion to you. Converting is possible (GDAL will do it, and GeoLibre's own Conversion toolbox has 49 tools that translate to GeoParquet, PMTiles and COG), but it means downloading the whole 1.15 GB first, which is the cost you were trying to avoid.

## Free Imagery That's Actually Licensed for Work

You can do real work in GeoLibre without spending anything, as long as you're honest about the resolution ceiling.

Sentinel-2, from the European Space Agency's Copernicus programme, is free and open for commercial use with attribution: 13 bands, 10 m in the visible and near-infrared, a five-day revisit. Landsat is US public domain with an archive back to 1972. Both download from sources like the [Copernicus Data Space Ecosystem](https://dataspace.copernicus.eu) and load straight into GeoLibre.

Ten metres per pixel is roughly a tennis court. Enough to watch a crop stress, a dam draw down, a burn scar spread. Not enough to see a vehicle, a fence line, or the edge of a disturbance boundary a regulator will argue about. We've ranked the archives properly in our roundup of the [best free sources of satellite data](https://geopera.com/blog/free-sources-of-satellite-data).

## What the Browser Can Now Do With Those Pixels

The genuinely surprising part of GeoLibre isn't the map. It's **Processing → Whitebox Toolbox**, which exposes 1,000+ geoprocessing tools compiled to WebAssembly and executed locally: 256 raster tools, 154 remote sensing tools, 99 terrain, 65 LiDAR.

Spectral indices, band math, classification and change detection, computed in a browser tab, with no Python environment to install and nothing sent to a server. Five years ago that was a desktop install, a licence key and an afternoon. If you've been putting off learning a scripting workflow just to calculate NDVI over one site, that excuse is gone.

Which brings the resolution question back around. Once analysis is this cheap to run, the constraint stops being your tooling and starts being your pixels. A 10 m index over a paddock tells you something. The same index over a single tree row or a 12-metre easement tells you nothing at all, because the feature is smaller than one pixel.

## Where Geopera Fits

We deliver imagery the way a cloud-native GIS wants to read it, because that's now the difference between a map that opens and a map that sulks.

Every order through [Pera Portal](https://portal.geopera.com) is delivered as a **Cloud Optimized GeoTIFF**, orthorectified, pansharpened and atmospherically corrected at no extra cost. Paste the URL into GeoLibre and it streams. There's also a **STAC 1.0 GeoJSON catalogue** for searching orders programmatically, and **WMTS and XYZ tile endpoints** behind an API key when imagery needs to sit under a dashboard or a design model rather than be copied around.

That processing matters as much as the container. A raster that hasn't been orthorectified can put features tens of metres from their true position, which no amount of clever byte-range reading will fix. Our explainer on [orthorectification](https://geopera.com/blog/orthorectification-explained) covers where that error comes from and why we do it on every order.

You can search archive imagery from 30 cm to 2 m or task a fresh capture across WorldView, Beijing-3, Perascope, SuperView and Wyvern's hyperspectral sensors, and the [pricing is published](https://geopera.com/pricing) so you can cost a job without booking a call. If you're weighing up the market more broadly first, our [guide to buying satellite imagery](https://geopera.com/blog/how-to-buy-satellite-imagery) lists the questions worth asking anyone. Still on the desktop for most things? Our [QGIS imagery guide](https://geopera.com/blog/qgis-satellite-imagery) covers the same ground for that canvas.

GeoLibre made its viewer cloud native in about eleven weeks. The imagery industry has had rather longer.

## Frequently Asked Questions

### How do I add satellite imagery to GeoLibre?

Four ways: pick a built-in basemap from the Layers panel, paste a remote Cloud Optimized GeoTIFF URL to stream a scene, search a STAC catalogue by area and date, or drag a local GeoTIFF onto the map. Remote COG URLs avoid downloading anything.

### Can I use GeoLibre's built-in basemaps for client work?

Check each one. The EOX Sentinel-2 cloudless annual layers are CC-BY-NC-SA 4.0, so non-commercial without a paid licence from EOX. Downloaded Sentinel-2 and Landsat data are both fine commercially, and OpenFreeMap is openly licensed.

### Can GeoLibre open Cloud Optimized GeoTIFFs from a URL?

Yes, and it's the format the app is built around. GeoLibre reads HTTP range requests for the tiles covering your view rather than the whole scene, so a 1.15 GB image can render from a few megabytes of traffic if the file is properly tiled with overviews.

### GeoLibre vs QGIS: which should I use for satellite imagery?

QGIS remains the deeper desktop toolkit with a plugin catalogue built over two decades. GeoLibre wins on instant access, cloud-native streaming and sharing a project as a link. Plenty of people will run both, and imagery delivered as COGs opens in either.

### Do I need commercial imagery, or will free Sentinel-2 do?

Sentinel-2 at 10 m is genuinely free and commercially licensed, and it's enough for crop condition, water extent and broad change. Anything where you must resolve a vehicle, a boundary edge or an individual asset needs sub-metre commercial imagery.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# 10 Best Free Sources of Satellite Data for Australia

> A comprehensive list of the best free satellite data sources in Australia

Published: 2026-08-17 | Author: Darcy Weedman | Reading time: 6 minute

Source: https://geopera.com/blog/free-sources-satellite-data-australia

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- [National Map](https://nationalmap.gov.au/): Best for property owners wanting easy-to-view satellite images without technical knowledge

- [Digital Earth Australia](https://www.ga.gov.au/scientific-topics/dea): Best for researchers and analysts tracking environmental changes over time

- State-Specific Resources:
  - [SIX Maps](https://maps.six.nsw.gov.au/) (NSW)
  - [Queensland Globe](https://qldglobe.information.qld.gov.au/)
  - [Vicmap](https://www.land.vic.gov.au/maps-and-spatial/spatial-data/vicmap-catalogue) (Victoria)
    Best for detailed local planning and property boundary information

- [Google Earth](https://earth.google.com/web): Best for general purpose viewing and basic exploration

- [Pera Portal](https://portal.geopera.com/): Unlimited free Sentinel-2 browsing with 45+ pre-configured spectral indices (NDVI, NDWI, EVI, etc.) viewable in-browser. Advanced band ratio calculations, zero setup required. Plus access to 100+ commercial satellites for premium imagery when needed

- [BoM Satellite Viewer](http://satview.bom.gov.au/): Best for weather monitoring and tracking current atmospheric conditions

- [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/): Best for environmental monitoring and agricultural applications

- [Zoom Earth](https://zoom.earth/): Best for near-current weather and environmental condition tracking

- [NASA Earthdata Search](https://search.earthdata.nasa.gov/): Best for scientific research and complex data analysis

- [Maxar Open Data Program](https://www.maxar.com/open-data): Best for disaster response and emergency management

<AskBand headline="What does your patch of Australia look like this week?" sub="Unlimited free Sentinel-2 with indices, no credit card." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/dunes.webp', '/band/fields.webp', '/band/atoll.webp']} />

## Overview

Looking for satellite images of your house in Australia? Want to view historical satellite imagery of your property for free? You're in the right place.

This guide covers the sources of satellite data available to Australians, whether you want to view imagery directly in your browser or download data for further analysis. (For sources beyond Australia, see our global guide to the [10 free sources of satellite data](/blog/free-sources-of-satellite-data).)

### Is There a Live Satellite Map of Australia?

Not in the way movies suggest: live satellite imagery of your house isn't possible with current technology, and what you see on mapping platforms was usually captured weeks or months ago. The closest thing to a live satellite map of Australia is the Bureau of Meteorology's viewer, which updates every 10 minutes but shows weather-scale pixels, not properties.

### What's Available in Australia

If you're searching for free satellite imagery in Australia, it's important to understand what's available:

- Most free satellite images are updated every few months to years.
- Free imagery generally has lower resolution (you can see buildings, but not detailed features).
- High-resolution imagery with clear details often requires a paid subscription.
- Real-time satellite imagery does exist but is typically limited to weather forecasting or emergency services.

Fortunately, Australia offers several excellent free sources of satellite data. Whether you are a property owner, researcher, or just curious, we'll walk you through the best free and paid options available.

---

## 1. National Map

[National Map](https://nationalmap.gov.au/) is an Australian government platform that allows users to view satellite imagery directly through their browser. It's an excellent starting point if you want to see satellite images of your property without needing to download large datasets.

**Key Features:**

- Simple, Google Maps-style interface for easy navigation.
- Frequently updated satellite views of cities and rural areas.
- Access to a wide range of geographic data, including:
  - Digital elevation models
  - Property boundaries
  - Infrastructure and planning overlays
  - Environmental data

---

## 2. Digital Earth Australia (DEA)

[Digital Earth Australia](https://www.ga.gov.au/scientific-topics/dea) offers detailed satellite imagery and analytical tools, ideal for tracking changes to the environment or specific properties over time.

**Key Features:**

- Free access to time-series satellite data for Australia.
- Historical imagery dating back to 1988.
- Updated every 5 to 16 days, depending on the satellite.
- Products focused on water monitoring, urban development, and vegetation analysis.

---

## 3. State-Specific Resources

Each Australian state provides its own satellite imagery platforms, which often offer higher resolution and more frequent updates than national platforms.

- **NSW Spatial Services ([SIX Maps](https://maps.six.nsw.gov.au/)):** High-resolution imagery and property boundaries for NSW.
- **[Queensland Globe](https://qldglobe.information.qld.gov.au/):** Detailed satellite views and mining overlays for Queensland.
- **Victoria's DataVic ([Vicmap](https://www.land.vic.gov.au/maps-and-spatial/spatial-data/vicmap-catalogue)):** Planning zone maps and regular updates for Victoria.
- **Other state platforms:**
  - Western Australia's Landgate
  - South Australia's Location SA
  - Tasmania's LISTmap
  - Northern Territory's NR Maps

---

## 4. Google Earth

[Google Earth](https://earth.google.com/web) provides high-resolution satellite imagery across the globe, including Australia. However, most images are several months old, and the platform isn't suitable for downloading raw data. Note that [Google Earth Pro desktop is being discontinued](/blog/google-earth-pro-desktop-discontinued); the web version is the one with a future.

---

## 5. Pera Portal

Our own [Pera Portal](https://portal.geopera.com/) has a free tier built on Sentinel-2: unlimited browsing of Australia with 45+ spectral indices computed in the browser. The same interface offers access to premium satellite imagery with higher resolution and more frequent updates than free sources, and you can preview available imagery and order only the specific tiles or captures you need.

---

## 6. BoM Satellite Viewer

The Bureau of Meteorology (BoM) provides near real-time satellite imagery focused on weather patterns through [SatView](http://satview.bom.gov.au/). This is useful for monitoring storm systems and other atmospheric conditions.

**Key Features:**

- Updates every 10 minutes.
- User-friendly interface to zoom in on specific areas.
- Continuous coverage via geostationary satellites.

---

## 7. Copernicus Data Space Ecosystem

The European Union's [Copernicus](https://browser.dataspace.copernicus.eu/) program offers free, high-quality satellite data with frequent updates, imaging every point of Australia at 10 m resolution roughly every 5 days. It's a valuable resource for environmental monitoring, agriculture, and disaster response.

---

## 8. Zoom Earth

[Zoom Earth](https://zoom.earth/) provides frequently updated satellite imagery, focusing on weather patterns and environmental conditions. It's an excellent tool for tracking near-current conditions.

---

## 9. NASA Earthdata Search

[NASA Earthdata Search](https://search.earthdata.nasa.gov/) provides comprehensive datasets for scientific research. While it's not designed for casual users, it offers essential resources for those needing in-depth analysis.

---

## 10. Maxar Open Data Program

[Maxar](https://www.maxar.com/open-data) offers high-resolution satellite imagery during natural disasters in Australia, as it did during the 2019-20 bushfire season and major flood events. This program is critical for emergency response efforts.

**Key Features:**

- Free access to imagery during crises.
- Partnerships with humanitarian organisations.
- High-resolution data for disaster recovery planning.

---

## Tips for Making the Most of Free Satellite Imagery Platforms

**1. Find the Right Platforms.**  
We've highlighted some of the best satellite data providers, but there are many more out there. Starting with the platforms above will save you time and ensure a smoother experience.

**2. Understand Platform Capabilities.**  
Each platform offers unique features. Familiarise yourself with their resolution, data quality, and update frequency to match your project's needs.

**3. Define Your Requirements.**  
Specify what you need (the area of interest, time range, spatial resolution, and spectral bands) and use search filters to narrow down the results.

**4. Use Visualisation Tools.**  
Many platforms provide built-in tools to help you explore and analyse the data. These features can provide deeper insights without needing external software.

**5. Be Selective with Downloads.**  
Rather than downloading full datasets, focus on specific tiles or areas of interest to save time and storage space.

By following these tips, you'll be able to make the most of the available satellite imagery resources, whether you're conducting research, evaluating property, or monitoring environmental changes.

---

## When Free Satellite Data Isn't Enough

Free platforms cover an enormous range of uses, and if you're simply curious about your own property, they're genuinely all you'll ever need. But if you're using imagery professionally, you'll eventually hit the resolution wall:

| Source               | Resolution  | What you can actually see                                  |
| -------------------- | ----------- | ---------------------------------------------------------- |
| Landsat (free)       | 30 m        | Regional land cover: a footy oval is roughly one pixel     |
| Sentinel-2 (free)    | 10 m        | Paddock-scale vegetation patterns; buildings are blurs     |
| Commercial (Geopera) | up to 30 cm | Individual vehicles, fence lines, stockpiles, single trees |

The difference matters the moment you need to **measure rather than look**: tracking earthworks on a mine site, auditing stockpiles, mapping erosion along a specific creek line, monitoring vegetation by the row rather than by the paddock, or proving site conditions on a particular date. Free sensors also can't be tasked: if no satellite happened to capture your site cloud-free last month, there's nothing to download.

Commercial imagery used to mean opaque quotes and weeks of back-and-forth. We publish [transparent per-square-kilometre pricing](/pricing) for both tasking and archive, our guide on [how to buy satellite imagery](/blog/how-to-buy-satellite-imagery) covers the process end to end, and every order arrives analysis-ready: orthorectified, pansharpened, colour-balanced and mosaicked ([here's exactly what that involves](/imagery)).

If free data has taken your project as far as it can go, [explore available imagery through Pera Portal](https://portal.geopera.com/) or [get in touch to discuss your project](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Drone vs Aerial vs Satellite Imagery: Which to Use When

> Drone vs aerial vs satellite imagery compared on resolution, coverage, cost shape and access, with a decision framework for choosing the right platform.

Published: 2026-08-14 | Author: Darcy Weedman | Reading time: 10 min

Source: https://geopera.com/blog/drone-vs-aerial-vs-satellite-imagery

---
<script>
import { AskBand, QuadrantChart } from '$lib/components/charts';
</script>

## Summary

- **Drone, aerial and satellite imagery solve different problems**, and the deciding variables are area size and revisit need, not image quality in the abstract.
- **Drones win below roughly a square kilometre**: 0.5 to 5 cm detail, flown on demand, but every hectare costs pilot time and you need legal access to the site.
- **Crewed aerial survey wins for metro-scale visual base maps**: 5 to 15 cm resolution flown in campaigns, typically a few times a year over cities, with thin coverage outside them.
- **Satellites win on large areas, remote or inaccessible sites, frequent revisit, and history**: 30 cm to 10 m resolution over any coordinates on Earth, with archives reaching back decades.
- **Most serious monitoring programmes end up combining platforms**: satellite for the standing record, drone or aerial where centimetre detail earns its cost.

<AskBand headline="How big is the area, really?" sub="From one paddock to a whole state: archive search is free, with the per-km² price on each result." action="Search the archive" href="https://portal.geopera.com" tiles={['/band/fields.webp', '/band/city.webp', '/band/dunes.webp']} secondary={{ label: 'or talk to us', href: '/contact' }} />

A survey drone in standard mapping mode covers something like five square kilometres in a good day. At that pace, one drone mapping the Shire of East Pilbara in Western Australia, all 372,000 square kilometres of it, would still be flying in the 2220s. A satellite constellation photographs the whole shire before lunch, and did so yesterday as well.

Flip the comparison and it inverts just as hard. Ask that satellite to count rivets on a conveyor gantry and it can't. The drone reads the serial number.

That's the honest shape of the drone vs aerial vs satellite imagery question: there is no best platform, only a best platform for a given area, detail level, and repeat schedule. Buyers who get this wrong usually overpay in one of two directions, either flying drones over areas that satellites cover for a fraction of the price, or buying satellite pixels for a job that needed millimetres. The rest of this post is the decision, worked through properly.

## What Actually Separates the Three Platforms

Altitude, mostly. Everything else follows from it.

A drone flies at 50 to 120 metres, so each pixel covers half a centimetre to a few centimetres of ground. A survey aircraft flies at a few thousand metres and delivers 5 to 15 cm pixels across a whole city in a campaign. An imaging satellite sits several hundred kilometres up: commercial resolution runs from 30 cm at the sharp end to 10 m on free public missions like [Sentinel-2](https://sentiwiki.copernicus.eu/web/s2-mission), and a single pass can image a swath tens of kilometres wide.

| Platform      | Typical resolution | Coverage per day | Ground access needed   | Cloud             | Historical archive       | Cost shape                  |
| ------------- | ------------------ | ---------------- | ---------------------- | ----------------- | ------------------------ | --------------------------- |
| Drone         | 0.5-5 cm           | 1-10 km²         | Yes, plus flight rules | Flies under it    | Starts when you start    | High per km², low per visit |
| Crewed aerial | 5-15 cm            | Hundreds of km²  | No, airspace only      | Mostly under it   | Campaign years only      | Mid; sold via programmes    |
| Satellite     | 30 cm - 10 m       | Entire regions   | None at all            | Blocked (optical) | Back to 1972 via Landsat | Lowest per km² at scale     |

Two rows of that table deserve a closer look, because they decide more procurements than resolution does.

**Access.** A drone needs someone standing near the site with legal permission to fly, under rules like [CASA's in Australia](https://www.casa.gov.au/knowyourdrone/drone-rules). Crewed aircraft need airspace clearance. A satellite needs nothing: not permission, not access, not a person in the country. For land you don't control, sites across a border, offshore assets, or anywhere a travel budget dies, that row settles the argument by itself.

**History.** Imagery you didn't order can't be flown retrospectively. A drone archive begins the day your programme does. Satellite archives already exist: Landsat has imaged the planet since 1972 and Sentinel-2 has covered it at 10 m since 2015, so a dispute about what a site looked like in 2019 can be answered this afternoon. For baselines, legal evidence, and change analysis, the archive is the product.

## When a Drone Wins

Below about a square kilometre, with detail doing the earning, nothing touches a drone. Sub-5 cm pixels resolve individual plants, cracked pavement, corroded fittings, stockpile toes. Photogrammetry from a well-flown drone survey yields elevation models at centimetre accuracy that satellite stereo can't approach. And the drone flies under cloud, on your schedule, the same afternoon you decide you need it.

The costs are real, though, and they're not mainly the hardware. A licensed operator, site access, flight approvals, and processing time all scale with every visit and every site. One quarry, monthly: entirely sensible. Forty quarries across three states, monthly: now you're running an aviation programme, and the spreadsheet stops being funny.

## When Crewed Aerial Wins

Aerial survey occupies the middle deliberately. Cities and growth corridors get flown in organised campaigns at 5 to 10 cm, which is sharp enough to read driveways, roof condition and kerb lines across an entire metro at once. For councils, insurers and utilities that need a consistent visual base map of a city a few times a year, these programmes are mature, well priced, and very good at exactly that.

The limits are the flip side of the model. Coverage concentrates where the subscriber density is: capital cities and large regional centres, flown on the programme's calendar rather than yours. Between campaigns nothing updates. Outside the flown footprint, which for a mining region or agricultural shire can mean everywhere that matters to you, there's simply no data to buy.

## When a Satellite Wins

Satellites hold the rest of the map: the large, the remote, the frequent, and the historical.

<QuadrantChart title="Which platform, for what job" xLabel="Area you need covered" yLabel="How often you need it" schematic={true} caption="Schematic decision space, not a measurement: boundaries move with budget and site access, but the regions hold. The largest region belongs to the platform that needs no access and repeats without mobilising anyone." zones={[{ label: 'Drone', x0: 2, x1: 26, y0: 6, y1: 94 }, { label: 'Crewed aerial', x0: 30, x1: 70, y0: 6, y1: 46 }, { label: 'Satellite', x0: 30, x1: 98, y0: 52, y1: 94 }, { label: 'Satellite', x0: 74, x1: 98, y0: 6, y1: 46 }]} />

Work through the axes and the pattern is mechanical. Once an area passes a few tens of square kilometres, per-km² pricing beats mobilising anything with a pilot. Once revisit passes a few times a year, the platform that's already overhead beats the one that needs scheduling: modern constellations pass over most coordinates daily, and [tasking a fresh capture](/blog/satellite-tasking-explained) is an order form rather than a logistics plan. And the moment a site is inaccessible, the other two platforms leave the board entirely.

Resolution is the trade you're actually making. At 30 cm you can count vehicles, buildings, containers and cleared land; you can't read a serial number or map a crack. The practical question isn't "is satellite imagery sharp enough?" in general, it's whether your measurement survives at 30 or 50 cm. Most monitoring measurements do. Our guide to [choosing the right satellite](/blog/best-satellite-imagery) covers that decision once you're on this side of the map.

Cloud is the other trade: optical satellites can't see through it, drones fly under it. In the wet season over the tropics that matters, and a programme that needs guaranteed capture windows plans around it with tasking windows or accepts the gap.

## The Decision in Practice

Three worked examples, deliberately generic.

**A 40-hectare quarry, monthly compliance photos.** Genuinely contestable. A local drone operator gives you centimetre detail; a 50 cm satellite subscription gives you the record without anyone driving out. If the measurement is "has the disturbed footprint grown", satellite wins on cost and consistency. If it's "is that bench cracking", drone.

**A 300-site infrastructure portfolio, quarterly.** Satellite, and it isn't close. Twelve hundred drone mobilisations a year is a business, not a monitoring programme. One archive-plus-tasking arrangement covers every site in the same quarter at a knowable per-km² price.

**A flood response across three shires.** Satellite first, because it's the only platform that can see the whole event this week, then drones on the specific assets the imagery flags. Which is the general pattern hiding in all three examples: the platforms stack. Satellite carries the standing record; the closer platforms spend their higher cost only where the record says to look.

That stack is the part most platform-versus-platform articles miss. We sell the satellite layer, so weigh our view accordingly, but the strongest monitoring programmes we see run exactly this way: a consistent satellite baseline over everything, processed to be [analysis-ready on arrival](/blog/why-we-process-every-order), with drone or aerial detail commissioned as the baseline demands. Through one account, that baseline spans 100+ satellites with the price per km² published before any order goes in.

## Frequently Asked Questions

### What is the difference between aerial and satellite imagery?

Aerial imagery is captured from crewed aircraft at a few thousand metres and resolves 5-15 cm over campaign areas such as cities. Satellite imagery is captured from orbit at 30 cm to 10 m resolution, covers any location on Earth, and carries archives reaching back decades.

### Is drone imagery better than satellite imagery?

For sites under about a square kilometre where you need sub-5 cm detail and have legal access, yes. For large areas, multiple sites, inaccessible land, or frequent revisit, satellite imagery costs less per km² and needs no one on the ground.

### How accurate is satellite imagery compared to aerial?

Commercial satellite imagery resolves 30-50 cm per pixel against 5-15 cm for aerial survey, so aerial shows finer detail. Positional accuracy after orthorectification is comparable for both: a few metres or better, depending on processing and ground control.

### Is satellite imagery cheaper than aerial photography?

Per square kilometre at scale, yes. Satellite archive imagery is priced per km² of your area, with no mobilisation, flight or crew costs. Aerial campaigns are cost effective mainly within their subscribed metro footprints; outside them, satellite is usually the cheaper source.

### When should you use aerial vs satellite imagery?

Use crewed aerial when you need 5-10 cm detail across a metro area that an existing campaign already covers. Use satellite when the area is large, remote, outside flown footprints, needs frequent revisit, or needs a historical baseline from the archive.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# What Is Mine Rehabilitation? Stages, Standards and Proof

> Mine rehabilitation explained: the stages from decommissioning to relinquishment, Australia's PRCP and MRF rules, and how recovery actually gets proven.

Published: 2026-08-13 | Author: Darcy Weedman | Reading time: 9 min

Source: https://geopera.com/blog/mine-rehabilitation-explained

---
<script>
import { AskBand, LineChart, StatStrip, StepFlow, Timeline } from '$lib/components/charts';
</script>

## Summary

- **Mine rehabilitation is the work of returning mined land to a safe, stable, agreed post-mining land use**: reshaping landforms, replacing growth media, re-establishing vegetation, and then proving the result holds.
- **Australia has roughly 50,000 recorded abandoned mines**, and when researchers surveyed state governments in 2017, agencies could name one mine fully rehabilitated and relinquished in the previous decade.
- **The rules now bind earlier**: Queensland requires a Progressive Rehabilitation and Closure Plan (PRCP) with enforceable milestones, and Western Australia levies every operator annually through the Mining Rehabilitation Fund based on disturbed area.
- **Relinquishment, not revegetation, is the finish line.** A site must demonstrate years of stable, self-sustaining recovery before a government will take the lease back, which makes long-run monitoring evidence the scarcest asset in closure.
- **Satellite time series have become the standard way to build that evidence**: vegetation trajectory from free 10 m archives, landform and erosion detail from sub-metre captures.

Australia has roughly 50,000 recorded abandoned mines. In 2017, researchers at [The Australia Institute](https://australiainstitute.org.au/report/dark-side-of-the-boom/) asked state governments a simple question: name the mines that finished rehabilitation and were handed back. Across the whole country, agencies produced one example from the previous decade.

One.

That gap between "we will rehabilitate this site" and "a government agreed it was done" is what mine rehabilitation actually turns on, and it's why the rules have been tightening in every Australian mining jurisdiction since. If you work anywhere near a mining lease, closure plan, or environmental bond, the mechanics below decide real money.

<AskBand headline="What does recovery look like from orbit?" sub="NDVI over your own site, on free Sentinel-2, in the browser." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/mine.webp', '/band/terraces.webp', '/band/hills.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## What Mine Rehabilitation Actually Involves

Mine rehabilitation is the process of returning land disturbed by mining to a safe, stable condition that supports an agreed post-mining land use. That end use is negotiated up front: grazing country, native ecosystem, forestry, sometimes a pit lake with managed water quality. The obligation sits with the operator, secured by financial provisioning. If the company walks or fails, the state isn't meant to be left holding the bill.

The word "rehabilitation" makes it sound like one activity. It's closer to five, run in sequence over years:

<StepFlow
title="The stages of mine site rehabilitation"
steps={[
{
title: 'Decommissioning',
body: 'Plant, buildings and infrastructure come down. Contaminated equipment and materials leave the site under their own disposal rules.'
},
{
title: 'Landform reconstruction',
body: 'Waste rock dumps and pit edges are reshaped into stable slopes designed to shed water without eroding. This is where most of the earthmoving money goes.'
},
{
title: 'Growth media placement',
body: 'Stockpiled topsoil, or a manufactured substitute, is spread over the reshaped landform. Its depth and chemistry largely decide what can grow.'
},
{
title: 'Revegetation',
body: 'Seeding or planting to the agreed species mix. For native ecosystem targets, this means local provenance seed, not just anything green.'
},
{
title: 'Monitoring and maintenance',
body: 'Years of measurement against completion criteria: vegetation cover, species diversity, erosion, water quality. Failures get repaired and the clock restarts.'
}
]}
/>

The last stage is the one that decides everything, and it's the one that gets the least attention in glossy sustainability reports. A slope that looked stable in year one can gully in year three. A seeded cover that flushed green after autumn rain can thin out over two dry summers. Completion criteria exist precisely because early appearances mislead.

## The Rules: Closure Plans, Levies and Bonds

Every Australian state runs its own scheme, but two reforms in the last decade set the direction, and they took opposite approaches to the same problem.

<Timeline
title="How the regulatory ground shifted"
events={[
{
date: '2013',
title: 'WA launches the Mining Rehabilitation Fund',
body: 'Operators pay an annual levy calculated from their disturbed area. More disturbance, bigger levy. Rehabilitate, and the levy falls.'
},
{
date: 'Nov 2018',
title: 'Queensland passes the Financial Provisioning Act',
body: 'The Mineral and Energy Resources (Financial Provisioning) Act 2018 rebuilds how closure is funded and planned.'
},
{
date: 'Nov 2019',
title: 'PRCPs become mandatory in Queensland',
body: 'Site-specific mines must hold a Progressive Rehabilitation and Closure Plan with legally enforceable milestones.'
},
{
date: '2020-2021',
title: 'Two giants enter closure',
body: 'Argyle diamond mine stops production in late 2020; Ranger uranium mine stops processing in early 2021. Both become decade-scale rehabilitation projects.'
}
]}
/>

**Western Australia priced the problem.** Under the [Mining Rehabilitation Fund](https://www.dmp.wa.gov.au/Environment/Mining-Rehabilitation-Fund-4581.aspx), every operator reports disturbed area annually and pays a levy on it. The fund accumulates to cover abandoned-site cleanup, and the levy structure makes rehabilitation directly reduce next year's bill. What counts as "disturbed" and how it's measured is its own discipline; we covered it in [how disturbed area is measured for MRF reporting](/blog/mining-rehabilitation-fund-disturbed-area).

**Queensland scheduled it.** A [Progressive Rehabilitation and Closure Plan](https://environment.qld.gov.au/management/policy-regulation/mining-rehab-reforms) commits an operator to rehabilitation milestones across the life of the mine, not just at the end, with a Mine Rehabilitation Commissioner watching the system. Miss a milestone and you're out of compliance now, decades before closure.

Both schemes converge on the same operational fact: rehabilitation status is no longer something you describe once in an environmental report. It's something you measure, date, and defend every single year.

## Why Proving It Is the Hard Part

The arithmetic behind that single relinquished mine is uncomfortable.

<StatStrip
stats={[
{ value: '~50,000', label: 'recorded abandoned mines in Australia' },
{ value: '17,000', label: 'of them in Queensland alone' },
{ value: '1', label: 'relinquishment agencies could name, 2007-2017' },
{ value: '2019', label: 'PRCP milestones become law in QLD' }
]}
caption="Abandoned-mine counts per The Australia Institute's Dark Side of the Boom (2017) and the Queensland Government; PRCP timing per the Queensland environment department."
/>

Relinquishment is rare because the burden of proof is heavy, and it should be. To hand a lease back, an operator has to show the land is safe, stable, non-polluting and sustaining its agreed use without intervention. Not this season. Over years, through drought cycles and flood years, with evidence a regulator can audit.

That evidence problem has a specific shape. Ground surveys are precise but sparse: a botanist can tell you everything about a 50 by 50 metre quadrat and nothing about the 4,000 hectares around it. Drone campaigns cover more ground but each flight is a project with mobilisation costs, and flying the same site identically every year for a decade takes discipline few programmes sustain. What the relinquishment case actually needs is the boring thing: the same measurement, made the same way, over the whole site, every year, for a very long time.

| Method        | Coverage per campaign      | Repeat cost | Historical record        | Best at                                  |
| ------------- | -------------------------- | ----------- | ------------------------ | ---------------------------------------- |
| Ground survey | Quadrats and transects     | High        | Only from first visit    | Species ID, soil chemistry, ground truth |
| Drone flight  | Single site                | Medium      | Only from first flight   | Centimetre detail, landform surveys      |
| Satellite     | Whole tenement, every site | Low         | Back to 1972 via Landsat | Trajectory: the same measure, every year |

The third row has a property the other two can't buy at any price: the archive already exists. A mine planning its relinquishment case today can pull Landsat and Sentinel-2 imagery over its rehabilitation areas going back decades, before rehabilitation even started, and show the full recovery curve rather than assert it.

## Watching Recovery from Orbit

The measurement that carries most programmes is a vegetation index time series. NDVI, computed from red and near-infrared bands, tracks how much photosynthesising cover sits on each pixel. Run it across every [Sentinel-2](https://sentiwiki.copernicus.eu/web/s2-mission) capture of a rehabilitated landform, 10 m pixels, revisiting every five days since the second satellite joined in 2017, and you get a curve: seeding, first flush, dry-season dips, and the thing regulators actually care about, whether each year's trough sits higher than the last.

<LineChart
title="What a relinquishment case looks like as a curve"
xLabel="Years since seeding"
yLabel="NDVI"
xTicks={[0, 2, 4, 6, 8]}
refBand={{ from: 0.55, to: 0.75, label: 'undisturbed analogue range' }}
schematic={true}
caption="Schematic, not data from a specific site: seasonal peaks and troughs, a year-5 drought dip, and troughs climbing toward the analogue band. The relinquishment case is this curve, measured."
series={[
{
label: 'Rehabilitated slope',
points: [
{ x: 0, y: 0.08 }, { x: 0.5, y: 0.22 }, { x: 1, y: 0.14 }, { x: 1.5, y: 0.3 },
{ x: 2, y: 0.2 }, { x: 2.5, y: 0.38 }, { x: 3, y: 0.26 }, { x: 3.5, y: 0.44 },
{ x: 4, y: 0.28 }, { x: 4.5, y: 0.4 }, { x: 5, y: 0.24 }, { x: 5.5, y: 0.46 },
{ x: 6, y: 0.34 }, { x: 6.5, y: 0.52 }, { x: 7, y: 0.4 }, { x: 7.5, y: 0.58 },
{ x: 8, y: 0.46 }
]
}
]}
/>

The reference matters more than the raw number. A rehabilitated slope holding 80% of the NDVI of undisturbed vegetation next door through a drought is a strong story. The same value with no reference is just a number. Good monitoring programmes always carry analogue sites in the time series for exactly this reason.

Sub-metre imagery answers the questions the 10 m curve can't. Is that NDVI dip senescence or a gully opening up? Are the contour banks intact after the wet season? Individual erosion features, bare patches, and access track scars all show at 0.5 m, and [change detection between captures](/blog/satellite-change-detection) turns two dates into a map of exactly what moved. Field teams still matter; they're just deployed at the pixels that need them instead of walking the whole landform.

None of this is exotic. It's the same [vegetation monitoring](/blog/vegetation-management-using-satellite-data) toolchain used across agriculture and environmental compliance, pointed at the one industry where a decade of consistent measurements is literally the price of getting your bond back.

Geopera runs this workflow for mining clients from both ends of the archive: free Sentinel-2 and Landsat processed into consistent index time series, and sub-metre captures tasked or pulled from archive over the landforms that need detail, all delivered analysis-ready through one account. The [mining page](/mining) covers the monitoring products; if you're building a rehabilitation evidence base and want the historical record assembled first, [talk to us](/contact).

## Frequently Asked Questions

### What is mine rehabilitation?

Mine rehabilitation is the process of returning land disturbed by mining to a safe, stable condition supporting an agreed post-mining land use, such as native ecosystem or grazing. It covers decommissioning, landform reconstruction, soil replacement, revegetation, and years of monitoring against completion criteria.

### How are mine sites rehabilitated?

In stages: infrastructure is removed, waste rock landforms are reshaped to stable, erosion-resistant slopes, stockpiled topsoil is spread, and the area is seeded or planted to an agreed species mix. The site is then monitored for years, with failures repaired, until completion criteria are met.

### What is a Progressive Rehabilitation and Closure Plan (PRCP)?

A PRCP is a legally binding plan required for site-specific mines in Queensland since November 2019. It schedules rehabilitation milestones across the life of the mine rather than leaving the work to closure, with enforceable consequences for missed milestones.

### Why are so few rehabilitated mines relinquished?

Because relinquishment requires proving the land is safe, stable and self-sustaining over many years, through drought and flood cycles. A 2017 survey by The Australia Institute found state agencies could name only one fully rehabilitated and relinquished mine in the previous decade.

### How is mine rehabilitation monitored?

Through completion criteria covering vegetation cover, species diversity, landform stability and water quality. In practice, programmes combine satellite vegetation-index time series across whole sites, sub-metre imagery for erosion and landform detail, and targeted ground surveys for species-level evidence.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# How to Add Satellite Imagery to QGIS: The 2026 Guide

> Three ways to add satellite imagery to QGIS: XYZ tiles, QuickMapServices and GeoTIFFs, plus the licensing trap most tutorials skip.

Published: 2026-08-01 | Author: Darcy Weedman | Reading time: 9 min

Source: https://geopera.com/blog/qgis-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- There are **three ways to add satellite imagery to QGIS**: an XYZ tile connection, the QuickMapServices plugin, or loading a GeoTIFF file directly. Each suits a different job
- The Google satellite XYZ URL that most tutorials hand out **isn't licensed for professional use**. Google's terms require access through its official APIs, and a scraped tile endpoint isn't one of them
- The Sentinel-2 cloudless basemap from EOX (the pretty one in QuickMapServices and most browser GIS tools) is licensed **CC-BY-NC-SA**: fine for a hobby map, not for client work
- Actual Sentinel-2 data is genuinely open. Download it free from Copernicus, use it commercially, get 13 spectral bands at 10 m resolution
- The moment you need to pick the capture date or see 30 cm detail, you load a commercial GeoTIFF. Delivered analysis-ready, it drags straight onto the QGIS canvas like any other raster

There's a tile URL that gets pasted into QGIS thousands of times a day. It starts with `mt1.google.com`, it ends with a satellite basemap appearing behind your vector layers, and nearly every guide on how to add satellite imagery to QGIS hands it out like a Tim Tam at smoko.

Almost none of them mention that Google never agreed to any of this.

That's the gap in most QGIS satellite imagery tutorials. The mechanics are easy, and we'll cover all of them below. The part that bites people is what you're allowed to do with each imagery source once it's on your canvas. Get that wrong and your beautiful map becomes something you can't legally put in the report it was made for.

<AskBand headline="Are your basemap tiles actually licensed?" sub="Geopera tiles stream into QGIS over licensed WMTS and XYZ endpoints." action="See the API" href="/api" tiles={['/band/city.webp', '/band/terraces.webp', '/band/port.webp']} />

## The Three Ways to Get Satellite Imagery into QGIS

Every method you've ever seen boils down to one of these:

**1. XYZ tile connections.** You give QGIS a URL template, and it streams pre-rendered image tiles from someone's server as you pan and zoom. In the Browser panel, right-click **XYZ Tiles → New Connection**, paste a URL, name it, and drag it onto the canvas. Done in thirty seconds.

**2. The QuickMapServices plugin.** Same streaming idea, but with a curated menu instead of hunting for URLs. Install it from **Plugins → Manage and Install Plugins**, then go to **Web → QuickMapServices → Settings → More services → Get contributed pack** to unlock the full catalogue of several dozen basemaps. This is the answer to "what's the best QGIS satellite imagery plugin" in basically every forum thread since 2016, and fair enough.

**3. Loading a raster file.** A GeoTIFF, a Cloud Optimised GeoTIFF (COG), a JPEG2000 from a Sentinel download. Drag it into the Layers panel. That's it. No plugin required, and no tile server that can vanish out from under a half-finished project.

The first two rent you a view. The third one hands you data.

That distinction sounds pedantic until you try to do real work. A streamed tile is a compressed picture of imagery, served at whatever zoom level you asked for, captured whenever the provider felt like it. A GeoTIFF is the measurement itself: every band, every pixel value, at native resolution, with a capture timestamp you can put in a report.

## The Google Satellite Trick, and Why Pros Quietly Avoid It

The famous URL points QGIS at the tile endpoint that feeds Google Maps. It works. The imagery looks great, because in cities it's often aerial photography at better than 15 cm.

Four problems, in ascending order of how much they'll hurt you:

**You can't pick the date.** Google's basemap is a mosaic stitched from many sources and capture dates. The paddock on your screen might be from March. The neighbouring one, from three years ago. For change detection or anything time-sensitive, that's disqualifying, and we've covered why in our guide to [satellite change detection](https://geopera.com/blog/satellite-change-detection).

**It's a picture, not data.** Compressed RGB with the near-infrared band stripped out, which rules out NDVI and every other vegetation index before you start.

**Positional accuracy is unknown.** The mosaic is optimised to look right, not to meet survey standards. If you're digitising boundaries off it, you're inheriting error nobody quantified for you. Our post on [orthorectification](https://geopera.com/blog/orthorectification-explained) explains where that error comes from.

**The licence.** Google Maps Platform terms require accessing Google's map content through its official APIs and SDKs. A raw tile endpoint scraped into a desktop GIS isn't covered, and the terms explicitly restrict creating derived datasets from Maps content. Tracing features off Google satellite tiles and shipping the result to a client is exactly that.

Will Google send lawyers after your council flood study? Almost certainly not. But "our evidence base violates the imagery provider's terms of service" is a sentence you never want to hear in an audit or a courtroom. Bing and Esri tiles carry their own restrictions too. Always read the terms of whatever streams onto your canvas.

It works right up until it matters that it doesn't.

## Free Satellite Basemaps You Can Actually Use at Work

Good news: the properly licensed free options are better than most people think. You just need to know which layer is a pretty picture and which is open data.

**Sentinel-2 cloudless by EOX** is the gorgeous, gap-free global mosaic you've seen in QuickMapServices and, lately, in the new wave of browser GIS tools (GeoLibre ships it as a default basemap, for one). One caveat: the recent annual layers are licensed **CC-BY-NC-SA 4.0**. Non-commercial. Lovely for a portfolio piece or a community project, not for paid work unless you buy a commercial licence from [EOX](https://s2maps.eu) directly.

**Sentinel-2 itself** is the real prize. The European Space Agency's Copernicus programme publishes it under a genuinely free and open licence: commercial use allowed, attribution requested. Every point on Earth, revisited every five days, 13 spectral bands, 10 m resolution in the visible and near-infrared. Download scenes from the [Copernicus Data Space Ecosystem](https://dataspace.copernicus.eu) and load them as rasters. Now you own the pixels and the licence question evaporates.

**Landsat** goes one better on licensing: US government public domain, no strings at all. 30 m multispectral, 15 m panchromatic, and an archive stretching back to 1972 that makes it the default for long-term change studies.

**OpenAerialMap** offers openly licensed drone and aerial imagery, patchy in coverage but sometimes spectacularly detailed where volunteers have flown.

Ten metres per pixel, for the record, means a Sentinel-2 pixel is roughly a tennis court. You can watch a crop stress out or a dam draw down. You cannot see a car or a fence line. We've ranked all of these and more in our roundup of the [best free sources of satellite data](https://geopera.com/blog/free-sources-of-satellite-data).

## When Streamed Tiles Stop Being Enough

Somewhere along the way, most QGIS users hit the same wall. The project stops being "I need something behind my vectors" and becomes "I need to prove something." Tiles can't do that. Data can.

Here's the honest comparison:

| Option                        | Resolution         | You pick the date?              | Bands               | Licensed for client work? | Cost |
| ----------------------------- | ------------------ | ------------------------------- | ------------------- | ------------------------- | ---- |
| Google satellite (XYZ)        | Varies by location | No, undated mosaic              | RGB picture only    | No, terms don't allow it  | Free |
| Sentinel-2 cloudless (EOX)    | 10 m               | No, annual mosaic               | RGB picture only    | Non-commercial only       | Free |
| Sentinel-2 (downloaded)       | 10 m               | Any archive date, 5-day revisit | 13 spectral bands   | Yes, open licence         | Free |
| Landsat (downloaded)          | 30 m (15 m pan)    | Archive back to 1972            | 11 bands            | Yes, public domain        | Free |
| Commercial archive or tasking | 30-75 cm           | Yes, including new captures     | Multispectral + pan | Yes, explicit licence     | Paid |

The pattern in that table is worth saying out loud: **the free streamed options fail on licensing, and the free downloadable options are properly licensed but capped at 10 m.** The moment your job needs both a clean licence and sub-metre detail (a stockpile volume, a disturbance boundary for a regulator, an asset inspection, a footprint worth digitising), you've arrived at commercial imagery whether you like it or not.

The good news is that commercial data arrives the same way a Sentinel download does. It's just a raster.

## Getting 30 cm Imagery onto Your QGIS Canvas

An ordered scene from a commercial satellite lands as a GeoTIFF or COG. You drag it into QGIS. There is no step three. COGs will even stream straight from cloud storage over a URL, so you get tile-style convenience with actual data underneath.

What matters is the state the file arrives in. Plenty of the industry delivers raw or semi-processed imagery and leaves the orthorectification, pansharpening and atmospheric correction to you, or charges 30-80% extra to do it. That's a day of specialist work before the image is safe to measure from.

We built Geopera to skip that entire stage. Every order through [Pera Portal](https://portal.geopera.com) is delivered **orthorectified, pansharpened and atmospherically corrected at no extra cost**, ready to drop into QGIS and analyse. You can search archive imagery or task a fresh capture across satellites from multiple operators (WorldView at 30 cm, Beijing-3, Perascope, SuperView, plus Wyvern's hyperspectral sensors) and the [pricing is published openly](https://geopera.com/pricing), so you can cost a project without sitting through a sales call.

If you're not sure whether your job needs 30 cm tasking or whether free Sentinel-2 will carry it, [ask us](https://geopera.com/contact). We'll tell you honestly, including when the answer is "keep your money and download Sentinel." And if you're weighing up the whole market first, our [guide to buying satellite imagery](https://geopera.com/blog/how-to-buy-satellite-imagery) walks through the questions worth asking any provider.

Your QGIS canvas doesn't care where the raster came from. Regulators, clients and lawyers very much do.

## Frequently Asked Questions

### How do I add satellite imagery to QGIS?

Three ways: create an XYZ Tiles connection in the Browser panel and paste a tile URL, install the QuickMapServices plugin for a curated basemap menu, or load a GeoTIFF directly by dragging it into the Layers panel. Downloaded rasters give you data; the other two stream pictures.

### How do I add Google satellite imagery to QGIS?

Technically, via an XYZ Tiles connection pointing at Google's tile endpoint. Be aware that Google's terms require access through official APIs, so this method isn't licensed for professional or commercial mapping, and the mosaic has no capture date you can verify or cite.

### What is the best QGIS satellite imagery plugin?

QuickMapServices is the standard choice. Install it, then use Settings → More services → Get contributed pack to unlock dozens of basemaps including Sentinel-2 cloudless. Check each layer's licence before using it in paid work, since several popular ones are non-commercial.

### How do I get high resolution satellite imagery into QGIS?

Order it as a GeoTIFF or Cloud Optimised GeoTIFF from a commercial provider, then drag the file into QGIS like any raster. Commercial sensors such as WorldView-3 or Beijing-3 deliver 30 cm resolution with a capture date you choose, licensed for professional use.

### Can I use free satellite imagery commercially in QGIS?

Downloaded Sentinel-2 and Landsat data, yes: Copernicus data is free and open with attribution, and Landsat is public domain. Streamed basemaps mostly no: Google's tiles sit outside its licence terms and EOX Sentinel-2 cloudless is non-commercial without a paid licence.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Strait of Hormuz Satellite Imagery: Tracking Ships Gone Dark

> What satellite imagery of the Strait of Hormuz shows during the 2026 crisis: dark tankers, jammed GPS, and how to order the data yourself.

Published: 2026-07-21 | Author: Darcy Weedman | Reading time: 10 min

Source: https://geopera.com/blog/strait-of-hormuz-satellite-imagery

---
<script>
	import { ComparisonScale, Callout } from '$lib/components/charts';
</script>

## Summary

- **Iran closed the Strait of Hormuz to most commercial shipping on 28 February 2026.** By midnight on 2 March, not one tanker inside the strait was broadcasting an AIS position
- Maritime intelligence firm Windward counted **167 commercial-size vessels in the strait area on 5 May, with 146 of them running dark**, after dark-ship activity jumped nearly 600% in a fortnight
- GPS jamming corrupted navigation for roughly **470 vessels near Fujairah and Khor Fakkan in a single 24-hour window**, so even ships trying to be honest were reporting wrong positions
- Satellite imagery records ships whether they broadcast or not. A 330 m supertanker is visible in free 10 m Sentinel-2 data and in forensic detail at 30 cm
- Archive imagery of the strait can be ordered through [Pera Portal](https://portal.geopera.com) with published per-km² pricing. New tasking carries operator restrictions that change weekly, so [check with us](https://geopera.com/contact) for the current state

At one minute past midnight on 2 March 2026, the Strait of Hormuz disappeared from every ship-tracking website on the internet.

The ships were still there. Dozens of tankers, anchored or creeping along the Omani side of the channel. What vanished was the signal: for the first time since transponders became mandatory, no tanker in the world's most important oil chokepoint was broadcasting its position.

Since then, satellite imagery of the Strait of Hormuz has become the only reliable way to know what's actually happening there. The tracking apps show an empty sea and the official statements contradict each other, so everyone from oil traders to war insurers now depends on photographs, taken from 500 km up, of ships that would very much prefer not to be photographed.

This post covers what that imagery shows, why the usual tracking tools broke, and, because we get asked this a lot, how an insurer, trader or researcher orders the same data without a defence budget.

## How the World's Biggest Oil Chokepoint Went Dark

Quick recap for anyone who's been living under a rock (a reasonable place to be in 2026).

On 28 February, the United States and Israel opened an air campaign against Iran. Iran answered by closing the Strait of Hormuz, the 39 km-wide channel between Iran and Oman's Musandam Peninsula that carries roughly **20 million barrels of oil every day**. The US Energy Information Administration calls it [the world's most important oil transit chokepoint](https://www.eia.gov/international/analysis/special-topics/World_Oil_Transit_Chokepoints), handling about a quarter of seaborne oil trade and a fifth of global LNG.

The market did what markets do. Brent crude cleared US$100 a barrel by 8 March and peaked at US$126. War-risk insurance premiums for a single transit jumped from 0.125% of hull value to as much as 0.4%, adding around **US$250,000 per voyage** for a large tanker. By late April, some 2,000 ships and 20,000 mariners were stuck waiting on either side of the blockade.

Ceasefires have come and gone since. The latest one frayed in early July, and by mid-month Iran had struck seven commercial vessels in a week. Transits collapsed from about eight supertanker crossings a day in late June to roughly two.

Every number in the paragraph above was either confirmed or discovered by someone looking at satellite imagery.

## AIS Was Never Built for This

Every commercial vessel over 300 gross tonnes carries an AIS transponder, a radio that shouts the ship's identity, position, course and speed to anyone listening. Ship-tracking websites are, at heart, giant AIS receivers. The whole system rests on one assumption: that ships want to be seen.

That assumption is now doing a lot of heavy lifting.

A "dark ship" is simply a vessel that has switched its transponder off, and the Strait of Hormuz is currently the dark shipping capital of the world. [Windward](https://windward.ai/blog/one-month-into-the-ceasefire/) measured a near-600% surge in dark activity between 19 April and 3 May. By 5 May, of 167 commercial-size vessels its analysts could identify across the strait, **146 were operating dark**. Bloomberg's reporting in July found tankers going dark for the Hormuz run itself, then reappearing on AIS once safely into the Gulf of Oman.

It gets worse. Ships that leave their transponders on aren't necessarily telling the truth either, because GPS jamming and GNSS spoofing around the strait have corrupted the position data itself. Roughly 470 vessels near Fujairah and Khor Fakkan reported garbled or impossible positions inside one 24-hour window. Tankers have appeared to teleport inland, or sail in circles over an airport.

<Callout type="info" title="Dark doesn't mean invisible">
Turning off AIS makes a ship disappear from tracking websites, not from the ocean. A very large crude carrier is 330 m long and 60 m wide, sitting on a flat, dark, essentially featureless background. From a satellite's point of view, there are few things on Earth easier to spot.
</Callout>

## What Satellite Imagery of the Strait of Hormuz Actually Shows

Optical satellite imagery works on a different principle from AIS: it doesn't ask, it looks. And what you can pull out of a scene depends almost entirely on resolution.

<figure>

![The same 330 metre supertanker simulated at 10 m, 3 m, 1 m and 0.3 m satellite imagery resolution, from a rough blob to visible deck pipework](/blogs/tanker-resolution-comparison.webp)

<figcaption>A simulated 330 m crude carrier at four ground sample distances. This is an illustration of pixel footprints, not real imagery, but the proportions are accurate.</figcaption>

</figure>

| Resolution | Example source         | A 330 m tanker spans | What you can determine                                                       |
| ---------- | ---------------------- | -------------------- | ---------------------------------------------------------------------------- |
| 10 m       | Sentinel-2 (free)      | ~33 pixels           | A ship is present. Length class, heading, whether it's under way             |
| 0.5–0.75 m | Perascope, SuperView   | ~450–660 pixels      | Vessel type, deck layout, two hulls rafted together for an STS transfer      |
| 0.3–0.5 m  | WorldView-3, Beijing-3 | ~660–1,100 pixels    | Deck pipework, manifolds, whether hoses are connected, individual fast craft |

A few things practitioners have been reading out of Hormuz imagery this month:

**Anchorage counting.** On 19 July, analysts identified 14 tankers holding at Iran's Kharg Island export terminal, 13 of them dark, with a combined capacity in the range of 16-18 million barrels. You can't hide a parked fleet.

**Ship-to-ship transfers.** Two tankers rafted hull-to-hull off the Omani coast are unmistakable at sub-metre resolution. Imagery from 18 July showed exactly one STS pair still working outside the strait, down from several the week before, which told oil analysts the transfer trade was stalling days before any official data could.

**Small craft swarms.** Around 85 high-speed craft were imaged moving through the strait in three groups in mid-July, the largest small-boat presence ever recorded there. Boats that size have no AIS at all. Imagery is the only way they get counted.

One honest caveat, because this is the bit breathless LinkedIn posts get wrong: you will not read a ship's name off its stern from orbit. Identification at 30 cm comes from measuring length and beam, matching deck fittings, and tracking a hull between passes. It's fingerprinting, not name-reading. That's also why [change detection](/blog/satellite-change-detection) matters more than any single image: the story is in what moved between Tuesday and Friday.

<ComparisonScale
title="How much of the Hormuz picture each layer gives you right now"
items={[
{ label: 'AIS tracking alone', position: 2 },
{ label: 'AIS + free Sentinel-2 screening', position: 5 },
{ label: 'Free screening + tasked sub-metre imagery', position: 8 },
]}
leftLabel="Guessing"
rightLabel="Counting hulls"
/>

## The Gulf Is Almost Cloud-Free, Which Changes the Maths

Maritime surveillance talk usually jumps straight to radar satellites, because most of the world's shipping lanes spend half the year under cloud. Radar has real advantages here, and we've written a full explainer on [how SAR satellites see through weather and darkness](/blog/sar-satellite-imagery-explained). ESA's free Sentinel-1 C-band radar covers the Gulf routinely.

But the Strait of Hormuz is a special case. This is one of the least cloudy stretches of ocean on the planet. For most of the year, an optical satellite pass over the strait simply works, which is why so much of the public reporting on this crisis is built on ordinary photographs rather than radar.

Haze and airborne dust are the real problem in the Gulf, and they flatten contrast and shift colour across a scene in a way cloud never gets the chance to. That's an atmospheric correction problem, and it's solvable in processing. We've covered [why atmospheric correction matters](/blog/atmospheric-correction-satellite-imagery) in detail, but the short version: corrected imagery keeps a grey hull separable from a grey sea on a hazy June afternoon, and uncorrected imagery sometimes doesn't.

Revisit is the other half of the equation. Sentinel-2 photographs the strait about every five days at 10 m, free, and it's a genuinely useful screening layer. [Free satellite data sources](/blog/free-sources-of-satellite-data) will get you surprisingly far. When five days isn't enough, or 10 m pixels aren't, you task a high-resolution satellite to capture the area on demand. Our guide to [how satellite tasking works](/blog/satellite-tasking-explained) walks through the mechanics.

## Who's Buying Imagery of a Chokepoint

Navies and intelligence agencies have their own satellites. The interesting shift in 2026 is who else is ordering commercial imagery of Hormuz, because the buyer list looks nothing like it did five years ago.

**Insurers and P&I clubs** are verifying where a vessel actually was before paying a war-risk claim, since the ship's own AIS log is now worth roughly nothing as evidence.

**Commodity traders and analysts** count tankers at Kharg Island and STS pairs off Oman to estimate export volumes days ahead of official figures. When Brent moves double digits on a rumour, one clear image of an anchorage is cheap.

**Charterers and supply chain teams** want independent confirmation that a nominated vessel is where its owner says it is, and that its recent track doesn't include a quiet detour to a sanctioned port.

**Journalists and researchers** have used commercial imagery all year to verify strikes, sinkings and blockade claims that no reporter can reach by boat.

None of these organisations owns a satellite. All of them can order from the same constellations, which is the part of this industry that still feels slightly like science fiction.

## Watching the Strait Without a Navy

A word on what Geopera actually does, since "just buy satellite imagery" has historically been easier said than done.

Through [Pera Portal](https://portal.geopera.com), you can search archive imagery over any area, the Hormuz shipping lanes and anchorages included, or task a new capture from constellations run by multiple operators: WorldView at 30 cm, Beijing-3, Perascope, SuperView, and Wyvern's hyperspectral sensors, which pick up things like oil slicks that broadband colour imagery misses. Multi-operator access matters more in a crisis than any other time. And fair warning: there are real restrictions on Hormuz imagery right now. Some operators have limited tasking over parts of the Gulf, others hold back recent captures from general sale, and the rules shift week to week as the situation does. We work with the operators directly, so we know the current state of play at any given moment. [Reach out](https://geopera.com/contact) before you plan a capture and we'll tell you what can and can't be ordered over the strait this week, rather than letting you find out after you've budgeted for it.

Every order arrives orthorectified, pansharpened and atmospherically corrected at no extra cost, which for maritime work means positions you can measure from and haze you don't have to fight. Most of the industry charges 30-80% extra for that processing or skips it entirely.

Pricing is published on our [pricing page](https://geopera.com/pricing) by resolution tier, per square kilometre, so you can cost a monitoring cadence over an anchorage before speaking to a human. And if you're not sure whether free Sentinel-2 screening covers your use case, [ask us](https://geopera.com/contact). Sometimes the free answer is the right one, and we'll say so.

The strait will reopen eventually, chokepoints always do. What won't reverse is the habit organisations picked up this year of checking what's actually on the water instead of trusting what ships say about themselves.

## Frequently Asked Questions

### Can satellite imagery track ships that have turned off AIS?

Yes. AIS is a self-reported radio broadcast; satellite imagery physically photographs the vessel. A 330 m tanker is detectable in free 10 m Sentinel-2 data and identifiable by length, beam and deck layout in 30-50 cm commercial imagery, transponder or not.

### Is the Strait of Hormuz still closed to shipping in July 2026?

Effectively, yes, for most commercial traffic. After the early-July truce broke down, Iran struck seven vessels in a week and transits fell to roughly two supertankers per day, against a pre-crisis norm of well over a dozen. Conditions change weekly, so verify before planning anything.

### What resolution do you need to identify a ship in satellite imagery?

Detection needs surprisingly little: 10 m data shows presence, heading and length class. Classification (tanker vs bulker, ship-to-ship pairs) needs sub-metre imagery. Fingerprinting a specific hull uses 30-50 cm data and repeat passes. No commercial resolution reads a ship's name.

### How often can satellites capture the Strait of Hormuz?

Sentinel-2 images the strait free about every five days at 10 m. Commercial high-resolution constellations like Perascope, Beijing-3 and WorldView can be tasked for far more frequent captures, and the Gulf's near-permanent clear skies mean optical passes rarely fail.

### Can anyone buy satellite imagery of the Strait of Hormuz?

Mostly, yes. Archive scenes are available to businesses, researchers and individuals through platforms like Pera Portal, priced per square kilometre by resolution tier. New tasking over the strait currently carries operator-specific restrictions that change frequently, so check the current rules with a provider first.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Google Earth Pro Desktop Is Being Discontinued: What Now?

> Google Earth Pro desktop won't be downloadable after 25 June 2027. What the shutdown means, what still works, and what to use instead.

Published: 2026-07-09 | Author: Darcy Weedman | Reading time: 8 min

Source: https://geopera.com/blog/google-earth-pro-desktop-discontinued

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- Google announced on **8 July 2026** that Google Earth Pro desktop will no longer be available for download from **25 June 2027**
- Already-installed copies **keep working after the deadline**, but they'll receive no further updates
- Google's suggested replacement is Google Earth on web and mobile, which users report caps KML imports at around **250,000 vertices** and struggles to display attribute data
- Google Earth Pro was never one tool. It was three: a viewer, a lightweight GIS, and a screenshot machine. Each job has a different replacement
- For work that needs current, accurate, analysis-ready imagery, the replacement isn't another free globe. It's actual satellite data

Google Earth Pro has outlived Google Reader, Google+, Stadia, and close to 300 other products in the Google graveyard. Twenty-one years of dodging the axe. Then, on 8 July 2026, its number came up: **Google Earth Pro desktop is being discontinued**, with new downloads ending on 25 June 2027.

Google didn't announce it with a press release, or even a blog post. The news went out as a support-forum notice.

If "open Google Earth" is step one of half your workflows, and in mining, surveying, agriculture and environmental consulting it usually is, this one's for you. Here's what's actually happening, and what to move to.

<AskBand headline="Losing the desktop app; keeping the workflow?" sub="Pera Portal handles the imagery half in a browser; nothing to install or lose." action="Try the portal" href="https://portal.geopera.com" tiles={['/band/city.webp', '/band/oxbow.webp', '/band/port.webp']} />

## What Google Actually Announced

The facts, stripped of the outrage:

- **From 25 June 2027**, the Google Earth Pro desktop application (Windows, macOS, Linux) will no longer be available for download
- **Existing installations keep working.** Google has confirmed the app will continue to function after the deadline for anyone who already has it
- **No more updates**, ever. No bug fixes, no imagery pipeline improvements, no security patches
- Google recommends migrating saved places and projects to [Google Earth on web or mobile](https://www.google.com/earth/about/versions/)

So this is a slow fade rather than a kill switch. Your installed copy won't stop working on 26 June 2027. It'll just start ageing, and software that never gets patched again tends to age like milk rather than wine.

One wrinkle for Mac users: Google Earth Pro has no native Apple silicon build and runs through Rosetta 2 translation. Apple has already signalled that Rosetta will wind down after macOS 27. Mac users are on a shorter runway than the official date suggests.

Worth remembering how unlikely this product's life was in the first place. Google bought Keyhole in October 2004, launched Google Earth in June 2005, and sold the Pro tier for **US$399 a year** until January 2015, when it became free. An enterprise-grade desktop globe, maintained for over a decade, at a price of zero dollars. The strange part is that it survived this long.

## Why "Just Use the Web Version" Isn't Landing Well

Google's answer to every complaint is the web and mobile version of Google Earth. And to be fair, it has improved a lot. Historical imagery arrived on the web version in 2024, which closed one of the biggest gaps.

But professional users have been blunt about what's missing. [TechRadar's coverage](https://www.techradar.com/computing/the-impact-to-thousands-of-companies-across-industries-will-be-huge-google-earth-pro-for-desktop-is-being-discontinued-and-some-users-are-furious) ran with the quote "the impact to thousands of companies across industries will be huge", and the specifics behind that frustration are real:

- Users report the web version caps imported files at roughly **250,000 vertices**, which rules out the large KML datasets that utilities, councils and exploration teams have built over years
- Attribute data attached to features often doesn't display properly on web
- There's no offline mode, which matters a great deal when your field site has the mobile coverage of the Moon
- Desktop staples like Movie Maker and direct GIS-format import have no web equivalent

None of this makes the web version bad. It's a different product: a viewer. Google Earth Pro desktop grew into something closer to a Swiss Army knife, which is exactly why so many industries wove it into their workflows.

## What People Actually Used Google Earth Pro For

Ask a geologist, a farm agronomist and a town planner what Google Earth Pro is for and you'll get three different answers. That's the point. It did several jobs at once, all free:

**Reconnaissance.** Fly to a site, have a look, get oriented. Nothing beats it for speed.

**The KML lingua franca.** For twenty years, KML files have been how field teams, consultants and clients swap spatial information without anyone needing GIS training. "I'll send you a KML" is a complete sentence in half the industries in Australia.

**The time slider.** The historical imagery slider let anyone scrub back through decades of imagery for free. If you've used it to check what a paddock, pit or coastline looked like in 2005, you know why people are cranky. (We've written a full guide to [finding historical satellite imagery of Australia](https://geopera.com/blog/historical-satellite-images) that goes well beyond the slider.)

**Measurements and screenshots.** Rough areas, rough distances, and a thousand PowerPoint slides with that familiar globe watermark.

Each of those jobs has a replacement. But no single free product replaces all of them, and pretending otherwise is how people end up frustrated.

<figure>

![Scale showing which Google Earth Pro replacement fits which job: Google Earth web for browsing, QGIS with free satellite data for KML and measurement work, commercial satellite imagery for analysis and monitoring](/blogs/google-earth-pro-replacement-scale.webp)

<figcaption>Google Earth Pro covered the first two-thirds of this scale for free. Its replacements split along the same lines.</figcaption>

</figure>

## Similar Programs Like Google Earth Pro: What to Use Instead

The right replacement depends entirely on which of those jobs you're replacing. Here's the honest breakdown:

| Your job                                       | Use instead                                                                | Cost | The catch                                              |
| ---------------------------------------------- | -------------------------------------------------------------------------- | ---- | ------------------------------------------------------ |
| Browsing, presenting, quick looks              | Google Earth (web/mobile)                                                  | Free | Vertex caps, patchy attribute display, no offline mode |
| KML work, site boundaries, layering GIS data   | QGIS                                                                       | Free | A learning curve. Genuine GIS software, not a toy      |
| Free current imagery, vegetation checks        | Copernicus Browser (Sentinel-2)                                            | Free | 10 m resolution. Fields yes, fence lines no            |
| Current high-resolution imagery, real analysis | Commercial satellite imagery via [Pera Portal](https://portal.geopera.com) | Paid | Costs money. That's genuinely the only catch           |

A few notes on that table.

**QGIS** is the closest thing to a true Google Earth Pro desktop successor for the drawing-and-measuring crowd. It's free, open source, runs natively on everything including Apple silicon, opens KML and KMZ directly, and will never be discontinued by a company chasing quarterly focus. The trade-off is that it's real GIS software. Expect a weekend of YouTube tutorials before it feels natural.

**Sentinel-2**, the European Space Agency's free workhorse, photographs every spot on Earth every five days at 10 m resolution. That's dramatically fresher than Google Earth's basemap, which is a mosaic that can be anywhere from months to years old depending on where you're looking. We've pulled together the [best free sources of satellite data](https://geopera.com/blog/free-sources-of-satellite-data) if you want the full menu, Landsat included.

And commercial imagery is for when the free tier stops being enough, which brings us to the part most Google Earth Pro obituaries skip.

## The Jobs Google Earth Pro Was Never Doing Anyway

A lot of organisations were using Google Earth Pro for work it was never built to support.

Google Earth's imagery is a **visual product, not a measurement product**. The basemap is a mosaic stitched from many sources and dates, so you can't control (or sometimes even know) when a given patch was captured. The imagery is compressed RGB, meaning no near-infrared band, no red edge, and therefore no NDVI or any of the vegetation analysis serious monitoring depends on. And positional accuracy varies from place to place, because the imagery is optimised to look right rather than to survey-grade standards. We've covered [why orthorectification matters](https://geopera.com/blog/orthorectification-explained) if you want the gory details.

If you were measuring a stockpile, dating a land clearing event, or putting a Google Earth screenshot in front of a regulator, the desktop app's retirement isn't your real problem. The imagery was.

For a fuller side-by-side, our comparison of [Google Earth vs commercial satellite imagery](https://geopera.com/blog/google-earth-vs-commercial-satellite-imagery) breaks down where the free globe ends and professional data begins.

## If the Deadline Is Your Nudge to Upgrade

We built Geopera for the moment a team realises "having a look" isn't enough anymore.

Through [Pera Portal](https://portal.geopera.com), you can search archive imagery or task a new capture across satellites from multiple operators: WorldView at 30 cm, Beijing-3, Perascope, SuperView, plus Wyvern's hyperspectral sensors. You pick the capture date instead of inheriting whatever mosaic Google served up. Every order is delivered **orthorectified, pansharpened and atmospherically corrected at no extra cost**, where most of the industry either charges 30-80% extra for processing or hands you raw files and wishes you luck.

Pricing is published openly on our [pricing page](https://geopera.com/pricing), so you can work out whether a project stacks up before talking to anyone, and if commercial imagery is new territory, our guide on [how to buy satellite imagery](/blog/how-to-buy-satellite-imagery) walks through the prices, licensing and ordering steps. If you're not sure whether free Sentinel-2 covers your use case or you need 30 cm tasking, [ask us](https://geopera.com/contact). We'll tell you honestly, including when the free option is the right answer.

Google Earth Pro earned its twenty-one years. Pour one out and grab an installer before June 2027. Then put the right tool behind each of the jobs it was doing for you.

## Frequently Asked Questions

### Is Google Earth Pro being discontinued?

The desktop application is. Google announced on 8 July 2026 that Google Earth Pro desktop won't be available for download after 25 June 2027. Google Earth on web and mobile continues, and Google recommends migrating saved projects there.

### Why is Google Earth Pro being discontinued?

Google has not given a detailed public reason. The 8 July 2026 announcement was a support-forum notice recommending migration to Google Earth on web and mobile, which is where development now happens. The desktop app never received a native Apple silicon build, which suggests investment in it stopped some time ago.

### Is Google Earth going away too?

No. Only the Google Earth Pro desktop application is being discontinued. Google Earth on web and mobile continues, and it is Google's recommended destination for saved places and projects. The web version added historical imagery in 2024, though users report a roughly 250,000 vertex cap on KML imports.

### Can I still use Google Earth Pro after June 2027?

Yes, if it's already installed. Google has confirmed existing installations keep working after the download cutoff. The app just won't receive updates, bug fixes or security patches, so expect it to degrade slowly over time.

### What are similar programs like Google Earth Pro?

Google Earth web or mobile covers browsing and presentations. QGIS, which is free and open source, handles KML files, measurement and GIS layers. For current, high-resolution, analysis-ready imagery, commercial platforms like Pera Portal replace what Google Earth never provided.

### Is Google Earth Pro free?

Yes. Google Earth Pro has been free since January 2015, when Google dropped the US$399 annual fee. It stays free to download until 25 June 2027, so grab an installer now if your workflows depend on it.

### What happens to my KML files and saved places?

KML is an open format, so nothing is lost. You can import saved places into Google Earth web (watch the roughly 250,000 vertex limit users report) or open KML and KMZ files directly in QGIS with no size restrictions.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# How Disturbed Area Is Measured for MRF Reporting

> WA's Mining Rehabilitation Fund levy is charged per disturbed hectare. How disturbed area is measured for MRF reporting, and why the number matters.

Published: 2026-06-24 | Author: Darcy Weedman | Reading time: 13 min

Source: https://geopera.com/blog/mining-rehabilitation-fund-disturbed-area

---
<script>
import { AskBand, Callout, ComparisonScale } from '$lib/components/charts';
</script>

## Summary

- **The Mining Rehabilitation Fund (MRF) levy is charged by the hectare.** Each year the levy comes to 1% of a tenement's Rehabilitation Liability Estimate, and that estimate is just your disturbed area in each category multiplied by a per-hectare rate, then added up. Every hectare you report is a line on a recurring bill.
- **You measure and report the disturbed area yourself.** The department doesn't check your figure when you lodge it. It audits a sample afterwards. Overstate the area and you overpay the levy every year; understate it and a false or misleading return carries a penalty of up to $20,000.
- **Unit rates run from about $2,000/ha to $50,000/ha** depending on what the disturbance is. A hundred hectares carried at the wrong rate, or never cleared off the books, quietly moves real money.
- **Measurement error goes both ways.** In a single year's audit sample, WA's mines department clawed back $386,853 in extra levies _and_ refunded $37,667. Operators were both under-reporting and over-reporting their disturbed area.
- **Satellite imagery gives an independent, dated, repeatable measurement of the footprint** that lines up with the annual reporting cycle. The WA Auditor General has recommended exactly this kind of imagery to verify operator-reported ground disturbance.

![Diagram of a Western Australian mine site viewed from above, with disturbance zones (open pit, waste dump, tailings storage, plant area, haul roads, rehabilitation area) outlined as labelled polygons over a faint hectare grid](/blogs/mrf-disturbed-area-hero.webp)

The most expensive number on a West Australian mining tenement isn't the ore grade or the strip ratio. It's a figure in hectares that someone on the environmental team types into an annual return: the disturbed area.

Get it slightly wrong in one direction and the company overpays the Mining Rehabilitation Fund (MRF) levy, year after year, on disturbance that isn't really there. Get it wrong the other way and you've filed a return that understates your footprint, which is a problem the day an auditor pulls your tenement. Same number. Two ways to lose money. And almost nobody outside the compliance team has a clear picture of how that number is actually meant to be measured.

This is the guide that connects the two halves nobody connects: the regulation that prices your disturbance by the hectare, and the measurement method that decides whether the hectare figure is right.

<AskBand headline="What will this year's levy be?" sub="If you'd rather commission the disturbed-area measurement than build it, talk to us." action="Talk to us" href="/contact" tiles={['/band/mine.webp', '/band/hills.webp', '/band/fields.webp']} />

## What the Mining Rehabilitation Fund Actually Charges You For

Start with the money, because the money is the whole reason measurement matters.

The MRF is Western Australia's pooled rehabilitation fund, set up under the Mining Rehabilitation Fund Act 2012 and administered by [DMPE](https://www.wa.gov.au/organisation/department-of-mines-petroleum-and-exploration/about-the-mining-rehabilitation-fund), the Department of Mines, Petroleum and Exploration (the agency that replaced DEMIRS in July 2025). Instead of every tenement holder lodging an individual rehabilitation bond, holders pay a small annual levy into one shared fund. That pool pays to rehabilitate abandoned mine sites once every avenue to recover the cost from the operator has run out. By 30 June 2025 it had grown to roughly **$356 million**, with about **$51.6 million** in levies assessed for the 2024-25 year alone.

Here's the part that runs your bill. The levy on each mining authorisation is **1% of its Rehabilitation Liability Estimate**, or RLE. And the RLE is built from one input: area.

<Callout type="tldr" title="The levy in one line" showIcon={false}>
For each tenement: take your disturbed area in hectares in each category, multiply by that category's per-hectare rate, add it all up to get the Rehabilitation Liability Estimate, then pay <strong>1% of that</strong> as the annual levy. Area is the only number you supply.
</Callout>

<figure>

![Infographic showing the MRF levy calculation flowing left to right: disturbed area in hectares times a per-hectare unit rate gives the Rehabilitation Liability Estimate, multiplied by 1% to give the annual levy](/blogs/mrf-levy-formula.webp)

<figcaption>The MRF levy is a chain of multiplications that all start from one number you provide: the disturbed area in hectares.</figcaption>

</figure>

The per-hectare rates aren't flat. They're set in Schedule 1 of the regulations and scale with how expensive that kind of disturbance is to rehabilitate. Exploration disturbance and land you've already started rehabilitating sit near the bottom at around $2,000/ha. A large or hazardous tailings storage facility sits at the top, around $50,000/ha. Everything else lands somewhere between.

| Disturbance type (example)                            | Indicative unit rate (A$ per ha) |
| ----------------------------------------------------- | -------------------------------- |
| Exploration disturbance, or land under rehabilitation | $2,000                           |
| Salt or halite stockpile (minerals-in-brine)          | $10,000                          |
| Plant site, workshop, fresh-water dam                 | $18,000                          |
| Minerals-in-brine evaporation pond                    | $20,000                          |
| Large waste dump or overburden stockpile              | $30,000                          |
| Large or hazardous tailings storage facility          | $50,000                          |

Those figures are indicative. Your exact category list and rates live in Schedule 1 of the Mining Rehabilitation Fund Regulations 2013, and DMPE's [RLE Calculator](https://ace.dmp.wa.gov.au/ACE/Public/MrfRleCalculator/RleCalculator) will work out the real number for your tenement once you feed it the areas. Which is the catch. The calculator is only ever as good as the hectare figures you type in, and it has no way of knowing whether those figures match what's actually on the ground.

A few mechanics worth pinning down, because they shape how the measurement gets used:

- **You report in hectares, to at least two decimal places.** Acres and square metres aren't accepted. The area is a cumulative snapshot of total disturbance "as at" a date you choose, not just what you disturbed during the year.
- **The reporting period runs 1 July to 30 June**, and returns are lodged through DMPE's online EARS2 system on or before 30 June each year.
- **If your RLE comes to $50,000 or less, the levy is nil**, but you still have to report your disturbance. Plenty of small operators sit under that line, which makes the area figure the thing that keeps them under it or tips them over.

So the regulation is clear about _what_ you owe and dead silent about _how_ to measure the input. That silence is the whole story.

## Why Disturbed Area Is the Most Expensive Number You Report

Most numbers on a compliance form are one-offs. The disturbed-area figure isn't. It's cumulative and it recurs, which means an error in it doesn't cost you once. It costs you every year until someone catches it.

Run the arithmetic. Say 100 hectares is sitting in your return at a $10,000/ha rate when it shouldn't be, maybe it's old mine-plan area that was never actually cleared, or ground that's quietly revegetated and should have been reclassified. That's $1,000,000 of phantom liability bolted onto your RLE. At the 1% levy, you're handing over an extra **$10,000 a year**, indefinitely, for disturbance that isn't there. Nobody sends you a refund for the years you overpaid before you noticed.

Now the other direction. Under-report the area and yes, the levy drops in the short term. But the figures you lodge have to be supported by evidence you can produce on audit, and a return that's false or misleading in a material particular carries a penalty of up to **$20,000**. There's a modified penalty regime too: $4,000 per tenement for missing the deadline. In 2023-24 the department served 171 of those infringement notices in a single reporting round.

This isn't hypothetical. DMPE runs an annual compliance assessment on a sample of returns, and the results show how often the area is simply wrong.

<Callout type="info" title="What one year's MRF audit actually found" showIcon={false}>
Reviewing 1,145 tenements across 43 MRF reports, the department flagged 240 with potential discrepancies and confirmed 226 needed amending. The corrections produced <strong>$386,853 in additional levies</strong> owed and <strong>$37,667 in refunds</strong>. The most common mistakes: roads and tracks left out, mining voids and waste rock dumps put in the wrong category, and some disturbances not reported at all. Errors ran in both directions, which is the point, accurate area protects you from overpaying as much as from underpaying.
</Callout>

That refund figure is the one most operators miss. The instinct is to treat MRF reporting as a cost you minimise by reporting less. The audit data says a chunk of the industry is overpaying, sitting on overstated or stale disturbance that an accurate measurement would clear off the books. Accuracy isn't only a compliance shield. It's a cost-control lever pointing both ways.

## How Disturbed Area Gets Measured Today

Here's the thing the official guidance never spells out. It tells you the area must be accurately measured and that it's your responsibility, then stops. It doesn't tell you how. So in practice, tenement holders fall back on a handful of methods, each with a gap.

**GIS mine-plan polygons.** The cheapest and most common. You already have the planned footprint in your mine's GIS, so you sum the hectares per category and lodge them. The problem is right there in the word _planned_. Designed footprints and real ground don't match. Clearing over-runs or under-runs the plan, infrastructure shifts, and last year's polygons drift out of date. These figures are self-drawn and self-reported, which is exactly why they're the ones an auditor most wants to check against something independent.

**Ground and GPS survey.** Walk it or drive it with RTK GNSS and you get survey-grade boundaries, accurate to a few centimetres. That's the gold standard for a single boundary. It's also slow, expensive, and unsafe along active pit walls, so it gets done in campaigns over small areas, not across a whole tenement every year. Brilliant for validating a spot. Impractical for measuring your entire footprint annually.

**Drones.** An RTK-enabled drone matches survey accuracy and covers a hundred-plus acres in under an hour, far faster than a ground crew. But the swath is small. Covering a large tenement, or a portfolio of scattered ones, means many flights, mobilisation cost, and CASA approvals, and weather and battery life cap how much you get per trip. So it shines on a single rehab cell and bogs down across a whole operation.

**Aerial photography.** High resolution over a wide area in one sortie, but you're commissioning flights, so captures are sporadic and pricey, and consistent annual coverage of every tenement rarely stacks up economically.

Notice the pattern. The methods that are accurate don't scale, and the method that scales, the desktop GIS polygon, isn't an actual measurement of the ground. So how defensible is each one the day someone asks you to prove your number?

<ComparisonScale
title="How defensible is each measurement on audit? (1 = take our word for it, 10 = independent dated evidence)"
leftLabel="Take our word for it"
rightLabel="Independent dated evidence"
items={[
{ label: 'GIS mine-plan polygons (planned, self-drawn)', position: 2, color: '#dc2626' },
{ label: 'Aerial photography (sporadic)', position: 6, color: '#6b7280' },
{ label: 'Ground / GPS survey (accurate but one-off)', position: 7, color: '#6b7280' },
{ label: 'Drone / UAV (accurate, small area)', position: 7, color: '#6b7280' },
{ label: 'Satellite imagery (independent, dated, repeatable)', position: 9, color: '#059669' }
]}
/>

## Measuring the Disturbance Footprint From Satellite Imagery

Satellite imagery answers the question the regulation asks and never explains: how do you produce an accurate, repeatable, defensible disturbance footprint across a whole tenement, on demand, every year?

The method is straightforward in principle. Classify each pixel in a recent capture as disturbed ground or vegetation, trace the boundary of the disturbed area, and compute the hectares inside it per category. Bare-soil indices like the Bare Soil Index pick out exposed ground where the vegetation's been stripped, which is most of what "disturbance" looks like from orbit. Then you compare against earlier captures using [satellite change detection](/blog/satellite-change-detection) to see what's new disturbance and what's land that's started recovering. The mechanics of that comparison, and why getting it right is mostly about consistency between dates, are the same ones we walk through for [satellite imagery across the mining lifecycle](/blog/mining-satellite-imagery).

What makes it fit MRF specifically is the combination of three things the other methods can't all offer at once.

It's **synoptic**. One pass images an entire tenement, or a whole portfolio of them, in a single consistent capture. No mobilisation, no flight planning, no walking the pit edge.

The measurement also **repeats, and it's dated**. Sentinel-2 revisits roughly every five days, and commercial constellations can go daily, so a clean annual snapshot to match the 30 June return is trivially available. Better still, it's measured the same way every cycle, so this year's number is genuinely comparable to last year's instead of being two people's readings of two different mine plans.

And because the imagery comes from an external source with a timestamp, it's **independent** in the way that counts on audit. That's the quality that turns "our GIS says 80 hectares" into "here's dated imagery showing 80 hectares, georeferenced, that you can re-measure yourself."

There's a resolution trade-off to be honest about. At Sentinel-2's free 10 metres, the smallest feature you can reliably map is around half a hectare, so narrow haul tracks, drill pads and thin slivers of clearing slip below the grid. Push to sub-metre commercial imagery and the minimum mapping unit drops to roughly a tenth of a hectare, sharp enough to trace the actual edge of every cleared parcel. For a sprawling operation you often want both: free [Sentinel-2 and Landsat](/blog/free-sources-of-satellite-data) to measure the broad footprint cheaply, and [a tasked sub-metre capture](/blog/satellite-tasking-explained) over the fiddly bits where the boundary has to be exact. Studies of satellite mine-disturbance mapping routinely report classification accuracies above 90%, well inside what a hectare-level return needs.

| Resolution tier                                      | Smallest reliable feature | Cost             | Best use for MRF                                    |
| ---------------------------------------------------- | ------------------------- | ---------------- | --------------------------------------------------- |
| Sentinel-2 (10m, free)                               | ~0.5 ha                   | Free             | Whole-tenement footprint, broad annual change       |
| Sub-metre commercial (Beijing-3, Perascope, SuperView) | ~0.1 ha                 | Tasked / archive | Exact boundaries, tracks, drill pads, small parcels |
| Drone / GPS                                          | Centimetres               | High per ha      | Spot validation, a single rehab cell                |

## Turning Rehabilitation Into a Smaller Levy

Now the lever that actually saves money, and the reason measurement matters on the way down as much as the way up.

The MRF doesn't reward rehabilitation with a discount multiplier. It rewards it by **reclassification**. Land sitting in an active disturbance category at, say, $18,000/ha doesn't get a percentage knocked off as you rehabilitate it. Instead, once the closure earthworks are done, you move that area into the "land under rehabilitation" category at $2,000/ha. And once a DMPE environmental officer signs it off as fully rehabilitated, it drops off your return entirely, at which point it stops costing you anything.

The saving is steep because the gap between the categories is steep. Move 50 hectares from an $18,000/ha active category down to the $2,000/ha rehabilitation rate and you've cut $800,000 off your RLE, which is **$8,000 a year off the levy**. You can report progressive rehabilitation pro-rata too, so finishing earthworks on half of a 10-hectare laydown lets you split it: five hectares disturbed, five under rehabilitation.

<figure>

![Three-panel time series of the same mine site across three years, showing newly disturbed land added in orange and land moved into rehabilitation shown in green, illustrating how the reported disturbed area changes between MRF reporting periods](/blogs/mrf-disturbance-timeseries.webp)

<figcaption>Across reporting years, satellite time-series separates two things the levy treats very differently: newly disturbed hectares (which add liability) and land moved into rehabilitation (which cuts it).</figcaption>

</figure>

But, and this is where a lot of operators leave money on the table, you only get the reclassification once you can **prove** the area qualifies. The rehabilitation criteria are vegetation-based: native cover re-establishing toward what the surrounding landscape looks like. You can't claim that with a stale mine-plan polygon. You need a measured vegetation signal showing recovery, and a delineated boundary around the recovering area, both of which an auditor can interrogate.

That's a measurement problem, and it's one satellite imagery is built for. Tracking [NDVI and other vegetation indices](https://docs.geopera.com) over a rehab cell gives you an objective, dated record of cover returning, which is the evidence that backs the reclassification. We go deeper on the vegetation side of this in our guide to [vegetation management with satellite data](/blog/vegetation-management-using-satellite-data). The same annual capture that measures your active disturbance also measures your recovery, so one consistent dataset drives both halves of the return.

## What an Audit-Ready Disturbance Record Looks Like

Here's the strategic part, and it comes from the regulator's own side of the fence.

The [WA Auditor General's review of compliance with mining environmental conditions](https://audit.wa.gov.au/reports-and-publications/reports/compliance-with-mining-environmental-conditions/) found that operator-reported ground disturbance isn't adequately verified against independent information, and recommended the department check it using independent sources such as publicly available imagery. Read that again from where you're sitting as an operator. The regulator has told itself, in writing, that self-reported area needs independent imagery to verify it. The direction of travel is obvious. The operators who'll sail through the next decade of audits are the ones whose numbers already come with that independent backing attached.

So what does a defensible disturbance record look like in practice? Three components, and satellite imagery supplies all three from the same source:

1. **Dated imagery** of the tenement at or near your assessment date, from an external provider with a verifiable timestamp.
2. **Georeferenced polygons** of the disturbance, classified by type, with the hectares that went into each MRF category.
3. **A consistent, documented method**, so this year's figure was produced the same way as last year's and can be re-run by someone else.

Keep that package and your annual return stops being an assertion and becomes evidence. Records have to be retained for at least two years anyway, and assessments can be reopened for up to two years after lodgement, so having the imagery on file isn't busywork. It's the thing that closes the audit in an afternoon instead of a fortnight.

## Getting MRF-Ready Imagery Without the Headache

The reason most teams don't already measure this way isn't the satellites. It's everything between the raw capture and a number you can defend. Two captures of the same pit a year apart have to line up to the metre before you can trust the area difference between them, which means orthorectifying both so coordinates match and processing them consistently so a brighter, hazier day doesn't masquerade as new disturbance. Skip that and your "change" is half measurement noise. We do that [alignment and correction](/blog/atmospheric-correction-satellite-imagery) on every order, included in the base price rather than billed as an extra, so the imagery arrives analysis-ready and two years' captures are genuinely comparable.

Through one platform you can pull free Sentinel-2 and Landsat for the broad footprint and [task sub-metre captures](/blog/satellite-tasking-explained) over the boundaries that need to be exact, across operators like Beijing-3, Perascope and SuperView, all delivered processed to a consistent standard. That's the difference between imagery you have to caveat and imagery you can staple to a due diligence file.

If you're working out the imagery layer for your next MRF return, start with what's available on [our satellite imagery page](/imagery), or [tell us about your tenements](/contact) and we'll work out which can ride on free Sentinel-2 and which need a tasked capture to pin the boundary. The levy is priced by the hectare. The hectare's worth measuring properly.

## Frequently Asked Questions

**How is the mining rehabilitation fund levy calculated?**

The MRF levy is 1% of a tenement's Rehabilitation Liability Estimate (RLE). The RLE is your disturbed area in each category multiplied by that category's per-hectare unit rate (roughly $2,000 to $50,000/ha), summed across all categories. If the RLE is $50,000 or less, no levy is payable, but you still report.

**What is considered ground disturbance for MRF reporting?**

Ground disturbance is any land cleared or altered by mining activity: open pits and voids, waste dumps, tailings and stockpiles, plant and infrastructure, haul roads and tracks, laydown areas, and exploration disturbance. It's reported cumulatively in hectares per category, including land currently under rehabilitation, until it's signed off as fully rehabilitated.

**Does DMPE check my reported disturbed area?**

Not when you lodge it. DMPE accepts your self-reported figures, then audits a sample each year against the evidence you're required to keep. The WA Auditor General has recommended verifying operator-reported disturbance using independent imagery, so expect that scrutiny to grow, not shrink.

**Can satellite imagery measure disturbed area accurately enough for MRF reporting?**

Yes. Sub-metre satellite imagery resolves disturbance boundaries down to about 0.1 hectare, and mine-disturbance classification studies report accuracies above 90%. Free 10-metre Sentinel-2 handles the broad footprint; tasked sub-metre imagery pins the exact edges. The result is a dated, georeferenced, repeatable area figure.

**How can I reduce my MRF levy?**

Accurately, by clearing overstated or stale disturbance off your return, and by reclassifying rehabilitated land. Moving area from an active category into "land under rehabilitation" drops its rate to about $2,000/ha, and fully rehabilitated land leaves the return entirely. Both require measured evidence of the change to claim and defend.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Satellite-Derived Bathymetry, Put to the Test Against Laser Truth

> Geopera satellite-derived bathymetry from Sentinel-2 and ICESat-2: seabed depth validated to within 38 cm of NASA's laser in clear water and about a metre against airborne LiDAR in turbid water, with honest uncertainty on every pixel.

Published: 2026-06-19 | Author: Darcy Weedman | Reading time: 10 min

Source: https://geopera.com/blog/satellite-derived-bathymetry-validated

---
<script>
import { AskBand, BarChart, Callout, ComparisonScale } from '$lib/components/charts';
</script>

## Summary

- **Geopera now produces satellite-derived bathymetry (SDB)** — seabed depth maps made from satellite imagery, with no boats, no aircraft, and no in-water survey.
- **Over a clear lagoon in the Bahamas**, our depths landed within 38 centimetres of NASA's ICESat-2 laser — 0.38 m across the scene, and sub-half-metre through the shallows. Seabed depth, from orbit, to within a hand's width of a survey-grade laser — accuracy at the front of the field.
- **In turbid, temperate water** — the case that defeats most satellite bathymetry at a few metres — we still hold about a metre across the shallows, validated against a real airborne LiDAR survey at Jervis Bay, NSW.
- **The edge is the processing.** A depth map is only as good as the imagery beneath it, and pulling a clean seabed signal out of [Sentinel-2](/sensors) — through haze, glint and sediment — is exactly the craft Geopera brings to every order. That's a large part of why our numbers come out where they do.
- **Every depth ships with its own honesty:** a calibrated error bound on each pixel, a flag for where the seabed stops being visible, and a clear line between what we measured and what we inferred. It complements a hydrographic survey; it doesn't replace one.

A lagoon in the Bahamas. Water so clear you can read the sand ripples from a plane. We pointed nothing at it — no boat, no survey crew, no aircraft trailing a laser. We just took satellite imagery — refreshed every few days — and, with the same processing we bring to every order, turned it into a map of the seabed.

Then we did the only thing that matters: we checked it against the truth. NASA's ICESat-2 satellite carries a laser precise enough to find the seabed through clear water. We held a stripe of those laser depths back, predicted them from the imagery alone, and then asked the laser how close we got.

The answer was 38 centimetres.

That's the gap across the whole scene between what we computed from orbit and what a survey-grade laser measured directly. For seabed depth pulled from a satellite that was never built for the job, in water nobody surveyed for us, that number is the reason we're writing this post.

![Satellite-derived bathymetry of a clear Bahamas lagoon showing intricate shallow channels and sand banks in turquoise grading to deep blue](/bathy/sdb-bahamas.webp)

And it isn't an average hiding a mess. It's sub-half-metre right through the shallows, where the seabed is brightest and the decisions get made:

<BarChart
title="Geopera satellite depth vs ICESat-2 laser truth, by depth band (Bahamas, clear water). Shorter bars are better — each bar is the error against the laser."
data={[
{ label: '0–5 m', value: 0.26, suffix: ' m', color: '#059669' },
{ label: '5–10 m', value: 0.46, suffix: ' m' },
{ label: '10–16 m', value: 0.65, suffix: ' m' }
]}
barColor="#6b6a63"
/>

From routine satellite imagery, against a laser that costs a fortune to fly. That's not a reconnaissance sketch. That's a depth surface you can plan around.

Here's the catch, and it's the reason this post keeps going. Clear tropical water is the easy mode of bathymetry. Quote a number like 38 cm and stop there, and you've told a true story that happens to be useless to anyone whose coastline isn't a postcard. Most of the world's working coast — the harbours, the dredged channels, the estuaries — is green, murky and full of suspended sediment. So before you lean on that figure, we want to show you what happens when the water fights back.

<AskBand headline="How deep is the water?" sub="SDB is a paid product with a validation report behind it; start with the research." action="Read the research" href="/research" tiles={['/band/atoll.webp', '/band/sandbanks.webp', '/band/bloom.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## Why the seabed is mostly a mystery

Most of the ocean floor has never been measured. Not the deep stuff — the _shallow_ stuff, the first thirty metres, where ports sit, cables come ashore, reefs grow, fish nurseries hide, and storms rearrange the coast overnight. That shallow margin is the part everyone cares about and almost nobody has mapped, because the only ways to map it properly are slow and expensive.

A survey ship dragging sonar gives you charting-grade truth, to any depth, along the lines the ship actually drove. An aircraft flying LiDAR covers more ground, faster, but only out to where the laser can punch through the water, and only over the corridors you paid it to fly. Both are field campaigns. Both cost real money and real time, and both leave you with depth exactly where you looked and nothing where you didn't.

Satellites, meanwhile, pass over those same shallow waters every few days, all over the planet. The whole game has always been turning what a satellite sees into a depth you'd actually bet a decision on. That's the problem we set out to solve, and this is the announcement that we have — with the receipts attached.

## How light tells you the depth

Here's the physics, and it's the same physics that makes a swimming pool look shallow at one end and bottomless at the other.

When sunlight hits the sea, some of it goes _in_. It travels down through the water, bounces off the seabed, and comes back up to the satellite. But the water doesn't let it travel for free. The deeper it goes, the more light gets absorbed on the way down and on the way back — and crucially, different colours get absorbed at different rates. Red light is gone within a few metres. Blue and green reach much deeper before they fade.

So the colour of a patch of shallow sea encodes its depth. A bright sandy bank under two metres of water reads very differently from the same sand under twelve. Read that colour-by-depth signal carefully enough, and you can recover the depth from the imagery alone.

<Callout type="tldr" title="The one-line version">
Light goes into water, deeper water absorbs more of it, and colour encodes how deep it went. Satellite-derived bathymetry reads that signal out of ordinary satellite imagery. The science is old. Making it accurate, honest, and global is the work — and turbid water is where the work shows.
</Callout>

That's the textbook part, and it's why the idea isn't new. The reason satellite bathymetry has spent years stuck with a reputation as a rough reconnaissance trick — fine for a first glance, not for anything you'd stake a survey budget on — is everything _else_ that also changes the colour of a pixel. The haze in the atmosphere. Sun glinting off the surface. The natural brightness of the seabed. And the big one: a cloud of suspended sediment. Get any of those wrong and your "depth map" is really a map of how dirty the water was that morning.

The discipline we already bring to every order — stripping the atmosphere back out of a pixel — matters double here. In a normal image, a bad [atmospheric correction](/blog/atmospheric-correction-satellite-imagery) is just noise. In bathymetry, it's a wrong depth.

We think that reputation is a measurement problem, not a law of nature. So instead of arguing about it, we measured — and we picked an unfair place to do it.

## The hard case: Jervis Bay

Jervis Bay, on the New South Wales coast, is a deliberately mean test for satellite bathymetry. It's temperate, turbid and sediment-rich — the kind of water where the simple methods come apart past a handful of metres. If a technique is going to embarrass itself, this is where.

We mapped it from orbit, then laid our result against a dense airborne LiDAR survey of the same seabed, flown by NSW DCCEEW. First, what we computed from a satellite:

![Geopera satellite-derived bathymetry of Jervis Bay, a coloured seabed depth surface grading from shallow turquoise to deep indigo](/bathy/sdb-jervis-sdb.webp)

Now the reference — what the laser saw, measured directly:

![Airborne LiDAR bathymetry of Jervis Bay used as the independent reference truth, same seabed, same colour scale](/bathy/sdb-jervis-lidar.webp)

And the honest panel, the difference between the two. Green is where we agree; colour creeps in where we drift apart:

![Difference map between Geopera satellite bathymetry and airborne LiDAR at Jervis Bay, mostly neutral green across the shallow bay indicating agreement within about a metre](/bathy/sdb-jervis-diff.webp)

Across the shallow margin — which is most of the bay, and the part that carries the decisions — satellite and laser agree to within about a metre. The colour builds as the water deepens, exactly where the physics says it should, because less light makes it back from down there. So rather than guess in the dark, the product flags where the bottom drops out of sight and leaves it blank. In murky temperate water, at 10-metre resolution, about a metre off a survey-grade laser across the zone that matters is a number we're happy to put our name on.

And here's a detail that quietly works in our favour: that LiDAR survey was flown in **2018**, while our imagery is from **2026**. Some of the colour in the difference map isn't our error at all — it's eight years of a real, moving seabed, with sediment shifting and storms reworking the shallows in between. The true gap between our depths and the seabed as it stands today is, if anything, a little tighter than the picture shows.

<Callout type="info" title="Why we lead you through the murky one">
Clear water flatters every satellite-bathymetry method ever made. Turbid water exposes them. We lead with our best number — 38 cm — and then walk you through the hard case, because the turbid map is the one that tells you whether this works on _your_ coast, not someone's holiday brochure. If a method holds up here, the clear-water result takes care of itself.
</Callout>

## So how accurate is satellite-derived bathymetry, really?

This is the question every buyer asks, and the honest answer is _it depends on your water_ — which is exactly why a single headline number, ours included, is a little dishonest. Clear water gets you toward 38 centimetres. Murky water gets you toward a metre. And once the water is too deep or too cloudy, the seabed isn't visible at all.

So instead of stamping one figure across the whole map, every depth we deliver carries its own. Three things travel with each pixel:

- **Its own error bound.** A calibrated margin on that exact depth — how sure we are right _there_, not one average smeared across the whole scene.
- **A bottom-visibility flag.** It marks where the seabed has stopped being optically visible — where the water got too deep or too murky for light to make the round trip — so we never report a depth we can't actually see.
- **A line between measured and inferred.** Some of the map is pinned to laser truth; the rest is the model joining the dots. We keep the two clearly marked, so you always know which is which.

![Per-pixel vertical uncertainty map of Jervis Bay, green where depth is confident grading to red where it is uncertain](/bathy/sdb-uncertainty.webp)

This is what lets the product be used like a grown-up, not just admired. You can set a safe working depth, decide which patches still need a boat over them before anyone relies on them, and — for navigation — back up how far you'd trust each area. We'll happily tell you where our own map shouldn't be trusted. That's not a weakness in the product; it _is_ the product.

## Why ours is more accurate: the processing underneath

Here's the part that actually separates one satellite-bathymetry map from another, and it isn't the satellite.

The raw seabed signal is faint, and almost everything conspires to drown it: haze in the atmosphere, sun glinting off the surface, suspended sediment in the water column. Most providers reach for the same public imagery we do — what they can't all do is clean it up well enough to leave the real signal standing. That's the craft Geopera has spent years on. The same processing pipeline behind every order — atmospheric correction, deglinting, careful capture selection — hands the depth model a clean picture to work from. Cleaner inputs are most of the distance between a depth you can trust and a confident-looking guess.

The reference we calibrate against is survey-grade laser truth — the same laser that judged our Bahamas map. We hold some of it back, predict it, and check, so the accuracy you see is earned on data the model never trained on.

That's also why this is a product, not a giveaway. The imagery is just the raw material. What you pay for is what our processing and calibration turn it into: a depth surface that lands at the front of the field, with an honest error bar on every pixel.

## Satellite-derived bathymetry vs LiDAR vs sonar

Satellite-derived bathymetry does not replace a survey ship, and anyone who tells you it does is selling you something. Here's how the three honestly relate.

<ComparisonScale
title="Coverage vs absolute accuracy"
leftLabel="Broad & fast"
rightLabel="Exact & local"
items={[
{ label: 'Satellite (SDB)', position: 2 },
{ label: 'Airborne LiDAR', position: 6 },
{ label: 'Multibeam sonar', position: 9 }
]}
/>

Satellite gives you breadth, speed and scale: any sunlit coast on Earth, refreshed every few days, at the lowest cost per square kilometre because there's no field campaign to mount. Airborne LiDAR gives you project-grade accuracy along the lines you commission. Multibeam sonar gives you charting-grade truth along the ship's track, to any depth, at the highest cost.

The strongest coastal programmes run all three as a team — satellite to find where to look and to fill the gaps between surveys, LiDAR and sonar to nail down the depths that carry the most risk. That's the whole reason we validate _against_ LiDAR rather than pretending to beat it. On charting specifically, the international hydrographic community treats satellite-derived bathymetry as a reconnaissance and chart-update source, not a substitute for a proper survey — and we agree. In clear, shallow water our uncertainty is good enough to stand behind a respectable charting confidence tier; in turbid water we'll tell you it's a notch lower. We will never claim a confidence level the data hasn't earned.

## Where this is actually useful

- **Coastal engineering and dredging** — scope a site, plan a campaign, and watch sediment move between surveys.
- **Offshore energy and cable routing** — screen nearshore approaches and landing corridors at country scale before you mobilise a single asset.
- **Reef, habitat and marine science** — map shallow seabed structure across whole atolls, and do it again next season.
- **Nautical chart reconnaissance** — flag what's changed and point the survey boat where it'll matter first.
- **Disaster and storm response** — re-map a coastline that moved overnight in days, not the months a survey would take.

## Why we're different

There are good people in this market. Here, plainly, is where we draw the line in a different place.

- **We show our working.** An independent LiDAR comparison in turbid water — not a cherry-picked clear-water number. It's the first thing on our [product page](/satellite-derived-bathymetry), not a footnote.
- **We lead with honesty, not just the flattering number.** The error bound, the visibility flag and the measured-versus-inferred line are the product, not the small print.
- **We earned the right to mention turbid water.** The case most providers quietly cap out on is the one we built to survive.
- **Our processing is the moat.** The same pipeline behind every Geopera order pulls a cleaner seabed signal out of the imagery — which is most of why our accuracy lands at the front of the field.
- **We're upfront about what we can and can't see**, and about price, the same way we [approach every product](/why-geopera).

The thread running through all of that is the same one that runs through [everything we process](/blog/atmospheric-correction-satellite-imagery): the inputs can be standard, even off the shelf — the advantage is the processing that cleans them up, the calibration against laser truth, and the honesty about the edges. That's what you're paying for, and it's what puts the numbers where they are.

## Try it on your coastline

If you've got a stretch of coast you need depth for — before you commission a survey, between surveys, or somewhere no survey has ever run — we can turn around a validated satellite-derived bathymetry map, with an error bound on every pixel, for any sunlit coast on Earth.

[See the product and the full LiDAR validation →](/satellite-derived-bathymetry)

Or [tell us where you need depth](/contact) and we'll map it.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# EUDR Compliance: How Satellite Imagery Provides Proof

> EUDR compliance means proving commodities did not drive deforestation after 2020. How satellite imagery provides geolocated proof before the 2026 deadline.

Published: 2026-06-16 | Author: Darcy Weedman | Reading time: 10 min

Source: https://geopera.com/blog/eudr-compliance-satellite-imagery

---
<script>
import { AskBand, Callout } from '$lib/components/charts';
</script>

## Summary

- **EUDR compliance means proving that the land behind your commodities was not deforested after 31 December 2020.** The EU Deforestation Regulation covers cattle, cocoa, coffee, oil palm, rubber, soya and wood, plus products derived from them.
- **The deadline moved.** After a one-year delay agreed in December 2025, the rules apply from **30 December 2026** for large and medium companies and **30 June 2027** for micro and small businesses.
- **Geolocation is the part everyone underestimates.** Every plot of land a commodity came from has to be pinned to coordinates: polygons for plots above 4 hectares, point coordinates for anything smaller.
- **Satellite imagery turns those coordinates into evidence.** Once you know exactly where a plot is, you compare its forest cover at the 2020 baseline against now. If the trees are still there, that plot is deforestation-free, and you can show it.
- **It applies to you even outside the EU.** Any business placing covered goods on the EU market, or exporting them into it, carries the obligation, which is why "does EUDR apply to the UK" has a simple answer: if you sell into the EU, yes.

![Diagram of EUDR compliance: a geolocated farm plot with its forest cover checked between 2020 and 2026, marked as deforestation-free and geolocated](/blogs/eudr-compliance-hero.webp)

A shipment of cocoa arrives at Rotterdam. Before it clears, someone has to be able to show that the specific plots of land where those beans grew were not carved out of forest after the end of 2020. Not the region. Not the co-operative. The actual plots. With coordinates.

That, in one sentence, is what the EU Deforestation Regulation did to global supply chains. It shifted the burden of proof. For decades "sustainably sourced" was a claim you made. Under EUDR it's a claim you have to evidence, plot by plot, and satellite imagery is the only practical way to do it at the scale of a real supply chain. Here's how EUDR compliance actually works, and where the imagery fits.

<AskBand headline="Can you evidence the supply chain?" sub="Sentinel-2 and Landsat, the archives EUDR verification leans on, are searchable free." action="Search the archives" href="https://portal.geopera.com" tiles={['/band/delta.webp', '/band/fields.webp', '/band/hills.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## What the EU Deforestation Regulation Actually Requires

Strip away the jargon and EUDR asks three things of every covered product before it can be sold in or exported from the EU.

**It has to be deforestation-free.** The commodity must come from land that was not subject to deforestation after 31 December 2020. That date is the line in the sand. Clear forest on 1 January 2021 to plant coffee, and that coffee is non-compliant, permanently. The regulation uses the FAO definition of forest, so the assessment is about genuine forest loss, not a farmer trimming a hedgerow.

**It has to be legal.** Production must comply with the relevant laws of the country of origin: land rights, environmental rules, labour law, anti-corruption. Deforestation-free but illegally produced still fails.

**It has to be traceable to coordinates.** This is the operational heart of it. Companies file a **Due Diligence Statement (DDS)** that includes the geolocation of every plot where the commodity was produced. No coordinates, no compliant statement, no sale.

<Callout type="info" title="The seven commodities, and what comes with them" showIcon={false}>
EUDR covers cattle, cocoa, coffee, oil palm, rubber, soya and wood. The catch most people miss is "derived products." Leather and beef trace back to cattle. Chocolate traces to cocoa. Tyres trace to rubber. Furniture, paper, charcoal and printed packaging trace to wood. If your product contains any of the seven, somewhere up the chain you inherit the obligation to prove where the raw material grew.
</Callout>

## When EUDR Applies: The 2026 and 2027 Deadlines

EUDR has been one of the most rescheduled pieces of EU legislation in recent memory, so the dates are worth getting right.

The regulation was originally meant to bite at the end of 2025. Weeks before that, in December 2025, the [European Parliament and Council agreed another one-year postponement](https://www.consilium.europa.eu/en/press/press-releases/2025/12/04/eu-deforestation-law-council-and-parliament-reach-a-deal-on-targeted-revision/), alongside a set of simplifications. So the timeline that actually matters now is this:

| Company size                           | EUDR applies from |
| -------------------------------------- | ----------------- |
| Large and medium operators and traders | 30 December 2026  |
| Micro and small enterprises            | 30 June 2027      |

The delays have lulled some businesses into treating EUDR as a someday problem. That's a mistake. The deforestation-free cutoff is still 31 December 2020, which means the evidence you'll need in December 2026 is about land use that already happened. You can't go back and re-photograph 2021. Either you have a clean imagery record for each plot or you're reconstructing one, and reconstruction gets harder the longer you leave it.

<Callout type="info" title="Don't confuse the deadline with the cutoff" showIcon={false}>
Two different dates do two different jobs. The <strong>cutoff date</strong> (31 December 2020) is what your land has to be measured against: no deforestation after it. The <strong>application date</strong> (30 December 2026 for larger firms) is simply when you have to start filing compliant due diligence statements. Pushing the application date back does nothing to the cutoff.
</Callout>

## Why Geolocation Is the Hard Part

Ask anyone mid-way through an EUDR programme what's eating their time and it won't be the satellite analysis. It's getting coordinates for every plot in the first place.

The regulation wants **polygons for any plot larger than 4 hectares** and **point coordinates for plots at or below that size**. For a vertically integrated timber operation with a handful of large concessions, that's manageable. For a chocolate company sourcing from tens of thousands of smallholder cocoa farms across West Africa, each often under a hectare, it's an enormous data-collection exercise involving field teams, GPS apps and a lot of messy spreadsheets.

And the coordinates are only useful if they're accurate. A plot boundary that's offset by 50 metres will sit partly over the neighbour's land, and when you run the deforestation check you'll either flag clearing that wasn't yours or miss clearing that was. Garbage coordinates produce garbage compliance evidence. The geolocation step and the imagery step are joined at the hip: precise boundaries make the satellite analysis trustworthy, and the satellite analysis is what makes the boundaries worth collecting.

## How Satellite Imagery Proves a Plot Is Deforestation-Free

Once a plot has coordinates, satellite imagery answers the only question EUDR really cares about: was this land forest after the cutoff, and did that forest get cleared? You answer it by comparing the plot's forest cover at the 2020 baseline against recent imagery. That comparison is [satellite change detection](/blog/satellite-change-detection), and it's the engine under every EUDR monitoring platform on the market.

<figure>

![Four-step EUDR compliance workflow diagram: geolocate the plot, pull baseline imagery since 2020, check for deforestation, produce a due diligence statement](/blogs/eudr-compliance-workflow.webp)

<figcaption>EUDR in four moves: pin the plot, pull its imagery history back to the baseline, check for forest loss, then file the due diligence statement that proof supports.</figcaption>

</figure>

The baseline is where free archives earn their keep. **Sentinel-2 has imaged the whole planet at 10-metre resolution every few days since 2015**, which means a clean, consistent record of nearly every plot on Earth as it looked at the end of 2020 already exists, for free. Landsat stretches the record back to the 1970s if you need deeper context. We list where to pull both in our guide to [free sources of satellite data](/blog/free-sources-of-satellite-data). For a large plantation, 10-metre Sentinel-2 is often enough to settle the question on its own.

Smallholder plots are where it gets harder. At 10 metres, a sub-hectare cocoa farm is only a handful of pixels, and you can't reliably tell a cleared corner from image noise. That's when you [task a high-resolution capture](/blog/satellite-tasking-explained), sub-metre imagery from satellites like Beijing-3 or Perascope, to resolve change at the level of an individual field. The other recurring headache is cloud. The commodities EUDR targets grow mostly in the wet tropics, exactly where persistent cloud cover thins out your usable optical imagery, so a deep archive and frequent revisit matter as much as raw sharpness.

| Plot situation                         | Practical imagery approach                                                       |
| -------------------------------------- | -------------------------------------------------------------------------------- |
| Large plantation or concession (>4 ha) | Free Sentinel-2 (10m) baseline vs recent, change detection on the polygon        |
| Smallholder plot (under 4 ha)          | High-resolution tasking (sub-metre) to resolve plot-level clearing               |
| Persistent cloud (humid tropics)       | Deep archive + frequent revisit to find clear-sky scenes near the dates you need |
| Deep historical context                | Landsat archive to extend the record before 2015                                 |

None of this is exotic. It's the same baseline-versus-now comparison used for [monitoring mines](/blog/mining-satellite-imagery) or urban growth, pointed at a different question and tied to a regulatory date.

## Getting EUDR-Ready Imagery Without the Headache

The imagery side of EUDR stalls in the same place every time: the free baseline is sitting there, but turning it into trustworthy, plot-level evidence means orthorectifying it so coordinates line up, atmospherically correcting it so two dates are actually comparable, and pairing free archive with sharp tasked imagery for the small plots. Skip any of that and your "deforestation" flags are as likely to be sun angle or misalignment as real forest loss.

That processing is what we do on every order, included in the base price rather than charged as an extra. Through one platform you can pull the Sentinel-2 and Landsat baseline for free, [task sub-metre captures](/blog/satellite-tasking-explained) over the plots that need them across operators like Beijing-3, Perascope and SuperView, and get all of it delivered already aligned and corrected, ready to drop against your plot boundaries. That's the difference between imagery you can defend in a due diligence statement and imagery you have to caveat.

If you're scoping an EUDR programme and working out the imagery layer, start with what's available on [our satellite imagery page](/imagery), or [tell us about your supply chain](/contact) and we'll help you work out which plots can ride on free Sentinel-2 and which need tasking. The deadline is fixed. The 2020 baseline isn't getting any fresher.

## Frequently Asked Questions

**What is EUDR compliance?**

EUDR compliance means demonstrating that commodities sold in or exported from the EU, cattle, cocoa, coffee, oil palm, rubber, soya and wood, plus derived products, come from land that was not deforested after 31 December 2020 and was produced legally. Proof is filed through a due diligence statement that includes the geolocation of every plot.

**How do you comply with EUDR?**

Collect the geolocation coordinates of every plot a commodity came from, assess each plot for deforestation after 31 December 2020 using satellite imagery, address any risk you find, and submit a due diligence statement. Polygons are required for plots over 4 hectares and point coordinates for smaller ones.

**When is the EUDR deadline?**

After a one-year delay agreed in December 2025, EUDR applies from 30 December 2026 for large and medium companies and 30 June 2027 for micro and small enterprises. The deforestation-free cutoff date stays fixed at 31 December 2020 regardless of the application date.

**Does EUDR apply to the UK?**

Yes, if you sell into the EU. EUDR applies to any operator or trader placing covered goods on the EU market or exporting them from it, wherever they are based. UK businesses supplying EU customers must comply. The UK is also developing its own separate forest risk commodity rules under the Environment Act.

**What resolution of satellite imagery does EUDR need?**

There's no mandated resolution. Free 10-metre Sentinel-2 is usually sufficient for larger plots and plantations. Smallholder plots under a hectare often need sub-metre tasked imagery to resolve clearing reliably, because at 10 metres a small plot is only a few pixels.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Satellite Change Detection: How to Track Change Over Time

> Satellite change detection compares imagery over time to map what changed. How it works, the data consistency it needs, and which satellites to use.

Published: 2026-06-16 | Author: Darcy Weedman | Reading time: 11 min

Source: https://geopera.com/blog/satellite-change-detection

---
<script>
import { AskBand, Callout, ComparisonScale } from '$lib/components/charts';
</script>

## Summary

- **Satellite change detection compares two or more images of the same place, captured at different times, to map what actually changed on the ground**: new clearings, growing mines, urban sprawl, flooded land, harvested fields.
- **Consistency matters more than resolution.** The same patch of ground imaged at a different sun angle, different season, or through different haze can look "changed" when nothing happened. Good change detection is mostly about removing those false alarms.
- **Free archives carry the historical baseline.** Sentinel-2 (10m, since 2015) and Landsat (30m, since 1972) give you decades of consistent, pre-corrected imagery, the backbone of most time-series work.
- **High-resolution tasking catches what 10m misses.** For change you need to see at object level (a single new building, an encroaching pit wall, an illegal track), sub-metre satellites like Beijing-3 (0.5m), Perascope and SuperView are the tools.
- **Regulation is now driving demand.** The EU Deforestation Regulation (EUDR) effectively forces companies to prove a plot of land wasn't cleared after December 2020, and satellite change detection is how that proof gets made.

![Diagram illustrating satellite change detection: a stylised map of the same area in 2019 and 2026, showing deforestation, a new road and an open-pit mine, with genuinely changed areas outlined in green](/blogs/satellite-change-detection-hero.webp)

Play a quick game. Look at the two panels above. Same patch of ground, two dates years apart. Now find everything that's different.

You spotted the mine, obviously. Probably the cleared land where forest used to be. Did you catch the new road slicing across? The way the river thinned out? The town that doubled in size?

That game, spot the difference but for the whole planet, at scale, without a human squinting at every frame, is **satellite change detection**. It's one of the most useful things you can do with satellite imagery, and also one of the easiest to get wrong. Most of the difficulty isn't finding change. It's telling real change apart from the imagery just looking a bit different because the light moved.

<AskBand headline="What changed since last year?" sub="Free Sentinel-2 reaches back to 2015; run a two-date comparison over your site." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/oxbow.webp', '/band/city.webp', '/band/mine.webp']} />

## What Satellite Change Detection Actually Is

**Change detection in remote sensing is the process of comparing imagery of the same location from two or more dates to identify and measure what has changed between them.** That's the whole idea. You take an image from Date 1, an image from Date 2, and you ask the data a simple question: where are these two different, and by how much?

The output is usually a _change map_: a layer that's blank where nothing moved and lit up where something did. Forest that became bare earth. Empty lots that became warehouses. A reservoir that dropped two metres. The map turns "stuff happened somewhere out there" into "here, exactly here, this much, between these dates."

People reach for it whenever the question is about _change over time_ rather than a single snapshot. How fast is this city spreading. Is that mine staying inside its permit boundary. How much forest did we lose this quarter. Did the flood reach the town or stop short. None of those can be answered by one image. They all fall out of comparing two.

<Callout type="tldr" title="The one-line version" showIcon={false}>
A single satellite image tells you what a place looks like. Change detection tells you what a place is <em>doing</em>, by comparing how it looked then with how it looks now.
</Callout>

## How Change Detection Works, Step by Step

Under the hood there are dozens of algorithms, ranging from "subtract one image from the other" to deep-learning models with names that sound like they belong in a particle physics paper. But the workflow underneath almost all of them is the same four moves.

<figure>

![Three-panel diagram showing satellite change detection: Image A from Date 1, Image B from Date 2, and the resulting Change Map where only genuinely changed areas glow green](/blogs/satellite-change-detection-difference-map.webp)

<figcaption>The core of change detection: line up two dates of the same place, compare them pixel for pixel, and keep only what genuinely changed.</figcaption>

</figure>

**1. Pick two dates of the same place.** Sounds trivial. It isn't, and we'll get to why in a minute. The two images need to cover the same ground, ideally from the same sensor, ideally captured at a similar time of year.

**2. Line them up exactly.** This is called co-registration. If the two images are even one pixel out of alignment, the algorithm "sees" the edges of every road, field and rooftop as change, because the road in Image A sits where the grass in Image B is. Sub-pixel registration is non-negotiable. Get this wrong and your change map is just a tracing of every boundary in the scene.

**3. Compare.** Here's where the methods split. The simplest is _image differencing_: subtract the pixel values of Date 1 from Date 2 and look at what's left. Bigger difference, bigger change. A step up is _spectral index differencing_. Instead of raw brightness, you compare an index like [NDVI](https://docs.geopera.com), which tracks vegetation, so you're measuring "how much did the greenness change" rather than "how much did the brightness change." Then there's _post-classification comparison_, where you classify each image into land-cover types (forest, water, built-up) and compare the labels. And at the top end, deep-learning models trained to recognise change directly. This is where AI and machine learning have genuinely moved the field forward.

**4. Keep the real change, drop the noise.** The raw comparison flags far more "change" than actually happened. Thresholding, filtering, and a bit of human judgement separate the genuine signal (a new clearing) from the junk, like a slightly brighter field because the sun was lower that day.

<Callout type="info" title="Where AI and ML actually help" showIcon={false}>
Deep-learning change detection is the real deal. Models like Siamese networks and transformers are genuinely good at the hard part: telling a harvested field apart from a clear-felled forest, or a shadow apart from a new building, which the older subtract-and-threshold methods constantly confused. They learn what _kind_ of change happened, not just that pixels moved. The one thing they can't do is fix bad inputs. A model trained on consistent, well-aligned imagery is powerful; feed it two mismatched dates and it'll confidently label the mismatch as change. The data quality underneath still decides the ceiling.
</Callout>

## The Thing Nobody Tells You: Consistency Beats Resolution

Here's the part that trips up almost everyone who tries change detection for the first time.

You'd assume the sharpest possible imagery gives the best results. It doesn't, necessarily. Two 30cm images that were captured under wildly different conditions will produce a worse change map than two 10m images that match each other. **For change detection, consistency between your two dates matters more than the resolution of either one.**

Think about photographing the same room on two different days to see if anything moved. If you shoot one photo at noon and one at dusk, half the room "changes": the shadows shifted, the colours warmed, the corners went dark. Nothing actually moved. You created phantom change purely by changing the lighting. Satellites have the same problem, except the lighting is the sun angle, the haze is the atmosphere, and the seasons swap the vegetation in and out underneath you.

So good change detection obsesses over matching the two dates:

- **Same sensor.** Two images from the same satellite are directly comparable. Mixing a 50cm image from one provider with a 30cm image from another means matching their spectral bands, their calibration, and their geometry first.
- **Similar time of day and sun angle.** Sun-synchronous satellites help here by crossing the equator at roughly the same local time each pass, so the lighting stays comparable.
- **Near-anniversary dates.** Comparing June to June beats comparing June to December. A field looks completely different across seasons even when the land use never changed. Imaging close to the same date each year cancels most of that out.
- **Atmospheric correction on both.** This is the big one. Raw imagery includes whatever haze, dust and water vapour sat between the satellite and the ground that day. Two clear days can still produce surface brightness that differs by tens of percent, purely from the atmosphere. Convert both images to surface reflectance first and that phantom change disappears. We went deep on this in [our guide to atmospheric correction](/blog/atmospheric-correction-satellite-imagery). It's the single most common reason time-series analysis goes sideways.

<ComparisonScale
title="What actually drives change-detection accuracy (1 = barely matters, 10 = make or break)"
leftLabel="Barely matters"
rightLabel="Make or break"
items={[
{ label: 'Image co-registration', position: 10, color: '#059669' },
{ label: 'Atmospheric correction', position: 9, color: '#059669' },
{ label: 'Near-anniversary dates', position: 8, color: '#059669' },
{ label: 'Same sensor / consistent calibration', position: 8, color: '#059669' },
{ label: 'Raw spatial resolution', position: 5, color: '#6b7280' },
{ label: 'Number of spectral bands', position: 6, color: '#6b7280' }
]}
/>

Get the consistency right and even modest imagery gives you a clean change map. Get it wrong and the best imagery money can buy lights up like a fireworks display over change that never happened.

## What You're Actually Detecting

Change detection isn't one use case. It's a method that bends to whatever you point it at. A few of the big ones, because the specifics matter:

**Deforestation monitoring.** This is the fastest-growing reason people ask us about change detection right now, largely because of the EU Deforestation Regulation, which pushes companies to prove the land their commodities came from wasn't cleared after a fixed baseline date. Underneath the paperwork it's pure change detection: compare the baseline against now, plot by plot. We walk through the regulation itself, the 2026 deadlines and exactly what counts as proof in our guide to [EUDR compliance and satellite imagery](/blog/eudr-compliance-satellite-imagery).

**Mining site monitoring.** Open-pit mines change shape constantly, and _mining site change detection_ answers questions regulators and operators both care about: is the disturbance footprint staying inside the permitted boundary, how fast are tailings dams growing, is there illegal artisanal mining creeping in at the edges. We dig into this in our piece on [satellite imagery for mining](/blog/mining-satellite-imagery).

**Urban growth.** Cities expand at the fringes, building by building. _Urban change detection_ tracks new construction, informal settlements, infrastructure rollout and land-use shifts, which matters to planning authorities, infrastructure operators, and anyone valuing land near a growing edge.

**Agriculture and vegetation.** Harvest timing, crop rotation, land clearing, irrigation expansion, the slow creep of drought stress across a region. Most of this runs on index-based change ([NDVI](https://docs.geopera.com) and friends) rather than raw imagery, and ties into broader [vegetation management with satellite data](/blog/vegetation-management-using-satellite-data).

**Disaster response.** Before-and-after is the purest form of change detection. Flood extent, fire scars, building collapse after an earthquake, landslide runout. The "after" image gets compared against the most recent clean "before" image, and the difference is the damage map that crews work from.

## Which Satellites to Use for Change Detection

The right satellite depends entirely on what kind of change you're chasing and how often you need to see it. There's no single best answer. There's a baseline layer and a detail layer, and most serious monitoring uses both.

The **free archives** are where almost everyone starts, and for good reason. They're consistent, pre-corrected, and they go back decades:

- **Landsat**: 30m resolution, but an archive stretching to 1972. Nothing else gives you a fifty-year baseline. If your question is "what did this place look like in 1990," Landsat is often the only answer.
- **Sentinel-2**: 10m resolution, free, with a revisit of about five days and consistent processing since 2015. This is the workhorse of operational change monitoring. Plenty of detail for land cover, agriculture, deforestation and water, and frequent enough to actually track change as it happens rather than after the fact.

We keep a running list of where to pull these in our guide to [free sources of satellite data](/blog/free-sources-of-satellite-data).

Where the free archives run out is _object-level_ change. At 10m, a pixel is the size of a generous living room: fine for "the forest shrank," useless for "someone built a shed and cut a new access track." For that you need **high-resolution commercial imagery**, and because change detection needs two matched dates, you usually want to [task a fresh capture](/blog/satellite-tasking-explained) to pair against an archive image or an earlier task.

| Source                | Resolution | Revisit  | Archive depth     | Best for change detection                                                   |
| --------------------- | ---------- | -------- | ----------------- | --------------------------------------------------------------------------- |
| Landsat               | 30m        | ~16 days | 1972 → now        | Long-baseline land-cover trends, decade-scale comparisons                   |
| Sentinel-2            | 10m        | ~5 days  | 2015 → now        | Operational monitoring: deforestation, agriculture, water, broad land cover |
| Beijing-3 (21AT)      | 0.3–0.5m   | Tasking  | Archive + tasking | Object-level urban, mining boundary, infrastructure change                  |
| Perascope (Geopera)   | 0.5–0.75m  | Tasking  | Archive + tasking | High-frequency high-res monitoring, large-area tasking                      |
| SuperView (SpaceWill) | 0.5m       | Tasking  | Archive + tasking | Detailed site monitoring, encroachment detection                            |

A common pattern: run Sentinel-2 continuously to flag _where_ something is changing across a wide area, then task a sub-metre satellite to zoom in on the flagged spots and see _what_ actually changed. Cheap wide net, sharp targeted follow-up. You don't pay for 50cm imagery across a whole province. You pay for it over the three places that lit up.

<Callout type="info" title="A note on resolution and revisit" showIcon={false}>
There's always a trade-off between how sharp the imagery is and how often you can get it. Free Sentinel-2 hands you a new image every few days for nothing, but at 10m. Sub-metre tasking shows you a single new building, but you order each capture and wait for a clear pass. Match the satellite to the question, not the other way round.
</Callout>

## Getting Change-Ready Imagery Without the Headache

Here's where most change detection projects stall. The method is well understood. The free data is right there. And then you spend three weeks fighting co-registration, hunting down matching dates, and discovering your two "clear" images were corrected differently and disagree by 20%.

That alignment-and-consistency work is exactly what we handle. Every order through Geopera is orthorectified, atmospherically corrected, and processed to a consistent standard before it reaches you, included in the base price, not bolted on as an extra. That means two images you order for a comparison actually _match_, instead of needing a fortnight of cleanup before you can trust the difference between them. Most providers ship raw or semi-processed data and leave that part to you.

Through one platform you can pull archive imagery for your baseline and [task new captures](/blog/satellite-tasking-explained) for the current date, across multiple satellite operators (Beijing-3, Perascope, SuperView and others) without juggling separate accounts, formats and processing pipelines. Free Sentinel-2 and Landsat for the wide-area baseline, commercial sub-metre for the detail, all delivered analysis-ready.

If you're building any kind of monitoring workflow (EUDR plots, mine boundaries, urban growth, vegetation trends), start with the imagery and pricing on [our satellite imagery page](/imagery), or [tell us what you're trying to monitor](/contact) and we'll work out the right baseline-plus-tasking mix for it. You order, we make the dates match.

## Frequently Asked Questions

**What is change detection in remote sensing?**

Change detection is the process of comparing satellite or aerial imagery of the same location from two or more dates to identify and measure what changed on the ground between them. The output is typically a change map that flags areas of genuine change, used for monitoring deforestation, urban growth, mining and disasters.

**What satellites are best for change detection?**

It depends on the scale. Sentinel-2 (10m, free, ~5-day revisit) and Landsat (30m, archive to 1972) are the standard choices for wide-area and historical monitoring. For object-level change, sub-metre satellites like Beijing-3, Perascope and SuperView are needed, usually tasked to match an archive baseline.

**How much imagery do you need for change detection?**

A minimum of two images of the same area from different dates. For reliable monitoring you want more: a time series of consistent, near-anniversary captures, so you can separate genuine change from seasonal variation and one-off atmospheric noise rather than reacting to a single misleading pair.

**Why does change detection produce false positives?**

Most false change comes from inconsistency between the two dates: misaligned images, different sun angles, seasonal vegetation differences, or uncorrected atmospheric haze. Co-registering the images and converting both to atmospherically corrected surface reflectance removes the majority of these phantom changes before analysis.

**How often do you need new imagery for change monitoring?**

It depends on how fast the thing you're tracking moves. Slow processes like urban sprawl or forest regrowth work fine on annual or seasonal captures. Active sites, like a working mine or a flood event, need weekly or even daily revisit. Match the capture frequency to the rate of change, not the other way round.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# As 10 Melhores Fontes Gratuitas de Imagens de Satélite do Brasil

> Uma lista completa das melhores fontes gratuitas de imagens e dados de satélite do Brasil

Published: 2026-06-11 | Author: Darcy Weedman | Reading time: 6 minutos

Source: https://geopera.com/blog/imagens-de-satelite-gratuitas-brasil

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Resumo

- [Pera Portal](https://portal.geopera.com/): Navegação gratuita e ilimitada em imagens Sentinel-2 com mais de 45 índices espectrais pré-configurados (NDVI, NDWI, EVI etc.) visualizáveis direto no navegador. Cálculos personalizados de razão de bandas, sem nenhuma instalação. Cobertura global, incluindo todo o território brasileiro — além de acesso a mais de 100 satélites comerciais quando você precisar de mais resolução

- [INPE (Instituto Nacional de Pesquisas Espaciais)](http://www.dgi.inpe.br/catalogo/): A melhor fonte de dados de satélite do Brasil, incluindo imagens CBERS e Amazonia-1

- [MapBiomas](https://mapbiomas.org/): O melhor mapeamento histórico de uso e cobertura da terra no Brasil, com atualizações anuais desde 1985

- [TerraClass Amazônia](http://www.inpe.br/cra/projetos_pesquisas/terraclass.php): O melhor para monitoramento da Amazônia e acompanhamento do desmatamento

- [Google Earth](https://earth.google.com/web): O mais fácil para visualização geral — por exemplo, do seu próprio imóvel

- [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/): O melhor para monitoramento ambiental e agrícola, com atualizações frequentes dos satélites Sentinel

- [NASA FIRMS (Informações sobre Queimadas)](https://firms.modaps.eosdis.nasa.gov/): O melhor para detecção de focos de incêndio em tempo quase real em todo o território brasileiro

- [Global Forest Watch](https://www.globalforestwatch.org/): O melhor para monitoramento florestal, alertas de desmatamento e acompanhamento ambiental

- [IBGE Geociências](https://www.ibge.gov.br/geociencias/): A melhor fonte de dados geoespaciais oficiais do governo brasileiro

- [Zoom Earth](https://zoom.earth/): O melhor para acompanhamento de clima e condições ambientais quase em tempo real no Brasil

<AskBand headline="Como está a sua região do Brasil esta semana?" sub="Sentinel-2 gratuito com índices no navegador, sem cartão de crédito." action="Abrir o nível gratuito" href="https://portal.geopera.com" tiles={['/band/delta.webp', '/band/fields.webp', '/band/bloom.webp']} />

## Visão geral

Procurando imagens de satélite da sua casa no Brasil? Quer ver imagens de satélite históricas do seu imóvel de graça? Você está no lugar certo.

Este guia apresenta as principais fontes de dados de satélite disponíveis no Brasil — seja para visualizar imagens direto no navegador ou para baixar dados e fazer suas próprias análises.

### Imagens de satélite em tempo real: a verdade

Muita gente procura por "imagem de satélite ao vivo" da própria casa, mas a realidade é esta: ao contrário do que os filmes sugerem, imagens de satélite ao vivo não são possíveis com a tecnologia atual. O que você vê nas plataformas de mapas normalmente foi capturado semanas ou até meses atrás.

### O que está disponível no Brasil

Se você está buscando imagens de satélite gratuitas do Brasil, é importante entender o que existe:

- A maioria das imagens gratuitas é atualizada a cada alguns meses ou anos.
- Imagens gratuitas geralmente têm resolução mais baixa (dá para ver edifícios, mas não detalhes).
- Imagens de alta resolução, com detalhes nítidos, normalmente são pagas.
- Imagens em tempo real existem, mas se limitam basicamente à previsão do tempo e a serviços de emergência.

Felizmente, o Brasil tem uma das melhores infraestruturas de dados de satélite do mundo, graças ao INPE e a diversas outras plataformas governamentais e de pesquisa. Seja você proprietário, pesquisador, produtor rural ou apenas curioso, vamos mostrar as melhores opções gratuitas (e pagas) disponíveis.

---

## 1. INPE — Instituto Nacional de Pesquisas Espaciais

O [Catálogo de Imagens do INPE](http://www.dgi.inpe.br/catalogo/) é a principal plataforma brasileira de acesso gratuito a imagens de satélite. O INPE opera os satélites do próprio país e oferece cobertura completa do território nacional.

**Principais recursos:**

- Acesso gratuito às imagens do CBERS (Satélite Sino-Brasileiro de Recursos Terrestres).
- Dados do Amazonia-1 — o primeiro satélite de observação da Terra 100% projetado e construído no Brasil.
- Dados Landsat processados especificamente para o território brasileiro.
- Acervo histórico de várias décadas.
- Busca avançada por localização, data e sensor.
- Ideal para monitoramento agrícola, pesquisa ambiental e gestão de recursos naturais.

---

## 2. MapBiomas

O [MapBiomas](https://mapbiomas.org/) é uma iniciativa brasileira extraordinária que produz mapas anuais de uso e cobertura da terra para todo o Brasil, de 1985 até hoje.

**Principais recursos:**

- Classificação completa de uso da terra para todos os biomas brasileiros (Amazônia, Cerrado, Caatinga, Mata Atlântica, Pantanal e Pampa).
- Mapas anuais mostrando desmatamento, expansão agrícola e urbanização.
- Plataforma interativa para visualizar mudanças ao longo do tempo.
- Download gratuito de mapas de classificação e estatísticas.
- Essencial para monitoramento ambiental, planejamento agrícola e análise de políticas públicas.
- Matrizes de transição mostrando as mudanças de uso da terra entre anos.

---

## 3. Pera Portal

Nosso [Pera Portal](https://portal.geopera.com/) oferece acesso totalmente gratuito e ilimitado a imagens Sentinel-2, com recursos de análise que superam a maioria das plataformas gratuitas.

**Principais recursos:**

- **Navegação ilimitada e gratuita em imagens Sentinel-2**, sem cadastro obrigatório nem limite de downloads.
- **Mais de 45 índices espectrais pré-configurados**, visualizáveis na hora, direto no navegador:
  - NDVI (saúde da vegetação e monitoramento de lavouras)
  - NDWI (detecção de água, essencial para as bacias hidrográficas brasileiras)
  - EVI (índice de vegetação otimizado para florestas tropicais)
  - SAVI (ajustado ao solo, para as regiões agrícolas do país)
  - MSAVI, NDBI e muitos outros
- **Cálculos personalizados de razão de bandas** — crie seus próprios índices para análises específicas.
- **Nenhuma instalação necessária** — visualize imagens processadas imediatamente, sem baixar nada.
- **Cobertura global**, incluindo todo o Brasil, com atualização a cada 5 dias.
- **Aplicações típicas no Brasil**:
  - Monitoramento agrícola no Cerrado, em Mato Grosso e nos estados do Sul
  - Acompanhamento da saúde da floresta amazônica
  - Monitoramento da conservação da Mata Atlântica
  - Avaliação de lavouras de cana-de-açúcar e soja
  - Monitoramento das áreas úmidas do Pantanal
  - Gestão de recursos hídricos para o planejamento hidrelétrico
- **Acesso a imagens comerciais** — quando precisar de mais detalhe, navegue por mais de 100 satélites comerciais com resolução submétrica.

Ao contrário de plataformas que exigem conhecimento técnico ou downloads pesados, o Pera Portal torna a análise avançada de imagens de satélite acessível a todos — do fazendeiro monitorando a lavoura no interior ao pesquisador acompanhando padrões de desmatamento na Amazônia.

---

## 4. TerraClass Amazônia

O [TerraClass Amazônia](http://www.inpe.br/cra/projetos_pesquisas/terraclass.php) é a plataforma especializada do INPE para monitorar o uso da terra nas áreas desmatadas da Amazônia brasileira.

**Principais recursos:**

- Classificação detalhada do uso da terra em áreas já desmatadas da Amazônia.
- Séries multianuais mostrando conversão agrícola e regeneração.
- Acesso gratuito a mapas de classificação e relatórios.
- Fundamental para entender a dinâmica da terra na Amazônia além do simples mapa floresta/não-floresta.
- Diferencia pastagem, agricultura, vegetação secundária e outros usos.

---

## 5. Google Earth

O [Google Earth](https://earth.google.com/web) oferece imagens de satélite de alta resolução de todo o Brasil, com cobertura tanto das áreas urbanas quanto das regiões mais remotas da Amazônia.

**Principais recursos:**

- Interface simples e intuitiva, acessível de qualquer navegador.
- Linha do tempo com imagens históricas — especialmente útil para acompanhar o desmatamento.
- Visualização 3D do terreno, ideal para regiões montanhosas.
- Cobertura de áreas remotas, incluindo o interior da Amazônia e o Pantanal.
- Não serve para baixar dados brutos nem para análises técnicas.

---

## 6. Copernicus Data Space Ecosystem

O programa [Copernicus](https://browser.dataspace.copernicus.eu/) da União Europeia oferece dados gratuitos e de alta qualidade dos satélites Sentinel, com cobertura global que inclui todo o Brasil.

**Principais recursos:**

- Imagens ópticas Sentinel-2 atualizadas a cada 5 dias, com 10 metros de resolução.
- Dados de radar Sentinel-1 para monitoramento em qualquer condição de tempo (essencial na Amazônia, onde a nebulosidade é constante).
- Acesso gratuito, sem limite de downloads nem barreiras de cadastro.
- Ideal para o monitoramento agrícola no Sul do país e no Cerrado.
- Particularmente valioso na Amazônia, onde as nuvens frequentemente inviabilizam imagens ópticas.
- Visualização e download na nuvem.

---

## 7. NASA FIRMS — Sistema de Informações sobre Queimadas

O [NASA FIRMS](https://firms.modaps.eosdis.nasa.gov/) fornece detecção de focos de incêndio em tempo quase real — fundamental para monitorar queimadas no território brasileiro, especialmente na Amazônia e no Cerrado.

**Principais recursos:**

- Detecção de focos ativos atualizada a cada 3 horas.
- Dados históricos de queimadas desde 2000.
- Alertas por e-mail para focos detectados nas suas áreas de interesse.
- Essencial na estação seca para acompanhar queimadas agrícolas e incêndios florestais.
- API gratuita para integração com sistemas de monitoramento.
- Ferramenta crítica para fiscalização ambiental e manejo do fogo.

---

## 8. Global Forest Watch

O [Global Forest Watch](https://www.globalforestwatch.org/) oferece ferramentas completas de monitoramento florestal, com ampla cobertura do Brasil e recursos específicos para a Amazônia.

**Principais recursos:**

- Alertas de desmatamento em tempo quase real para as florestas brasileiras.
- Dados históricos de perda de cobertura arbórea desde 2000.
- Mapas interativos mostrando padrões de desmatamento em todos os biomas.
- Integração com terras indígenas e unidades de conservação.
- Alertas gratuitos por e-mail para as áreas que você monitora.
- Essencial para organizações de conservação e monitoramento ambiental.

---

## 9. IBGE Geociências

O portal de [Geociências do IBGE](https://www.ibge.gov.br/geociencias/) disponibiliza dados geoespaciais oficiais do governo brasileiro, incluindo produtos derivados de satélite.

**Principais recursos:**

- Dados territoriais oficiais do governo brasileiro.
- Mapas de vegetação e classificação de biomas.
- Modelos digitais de elevação para todo o país.
- Dados de uso e cobertura da terra integrados às informações censitárias.
- Download gratuito de bases oficiais.
- Essencial para planejamento e pesquisa.

---

## 10. Zoom Earth

O [Zoom Earth](https://zoom.earth/) mostra imagens de satélite atualizadas com frequência, com foco em padrões climáticos e condições ambientais em todo o Brasil.

**Principais recursos:**

- Acompanhamento quase em tempo real do clima e de tempestades.
- Atualização várias vezes ao dia com as passagens mais recentes.
- Excelente para acompanhar os sistemas climáticos da Amazônia e os padrões de chuva.
- Animações em time-lapse mostrando a evolução do tempo.
- Visualização básica sem necessidade de cadastro.
- Útil para planejamento agrícola e monitoramento climático.

---

## Dicas para aproveitar ao máximo as plataformas gratuitas

**1. Escolha as plataformas certas.**
As plataformas do INPE foram feitas para as necessidades brasileiras e devem ser sua primeira opção para dados locais.

**2. Conheça os recursos de cada plataforma.**
Para monitorar a Amazônia, considere os dados de radar do Sentinel-1, que atravessam a nebulosidade constante. Para a agricultura no Sul e no Cerrado, os dados ópticos costumam funcionar bem.

**3. Defina bem o que você precisa.**
Especifique a área de interesse, o período, a resolução espacial e as bandas espectrais — e use os filtros de busca. Com um território do tamanho do Brasil, mirar com precisão é essencial.

**4. Use as ferramentas de visualização.**
Muitas plataformas têm ferramentas próprias de análise. Os índices espectrais do Pera Portal, por exemplo, revelam a saúde das lavouras, das florestas e dos recursos hídricos sem precisar de software externo nem conhecimento técnico.

**5. Baixe só o necessário.**
Em vez de baixar conjuntos de dados completos, foque nas cenas ou áreas de interesse. Para muitos casos de uso, plataformas em nuvem como o Pera Portal eliminam a necessidade de downloads.

Com essas dicas, você vai aproveitar ao máximo as fontes de imagens de satélite disponíveis para o Brasil — seja para pesquisa, gestão agrícola, monitoramento ambiental ou acompanhamento do desmatamento.

---

## Quando os dados gratuitos deixam de ser suficientes

As plataformas gratuitas cobrem uma enorme variedade de usos — e se você só quer ver o próprio imóvel, elas sempre serão suficientes. Mas quem usa imagens profissionalmente acaba esbarrando, cedo ou tarde, no limite da resolução:

| Fonte                 | Resolução | O que dá para ver de verdade                                                         |
| --------------------- | --------- | ------------------------------------------------------------------------------------ |
| Landsat (gratuito)    | 30 m      | Cobertura da terra em escala regional — um campo de futebol é mais ou menos um pixel |
| Sentinel-2 (gratuito) | 10 m      | Padrões de vegetação na escala do talhão; edifícios ficam borrados                   |
| Comercial (Geopera)   | até 30 cm | Veículos individuais, cercas, maquinário, árvores isoladas                           |

A diferença importa no momento em que você precisa **medir, e não apenas olhar**: comprovar o uso da terra no nível do imóvel para o CAR, monitorar a lavoura linha por linha em vez de município por município, documentar a supressão de vegetação ao longo de uma divisa específica, ou provar as condições de uma área em uma data exata. Satélites gratuitos também não aceitam encomendas de captura — e sobre a Amazônia e a fronteira agrícola, a nebulosidade persistente faz com que o acervo muitas vezes simplesmente não tenha uma cena utilizável da janela que você precisa.

Imagem comercial costumava significar orçamentos opacos e semanas de idas e vindas. Nós publicamos [preços transparentes por quilômetro quadrado](/pricing), tanto para novas capturas quanto para o acervo — e cada pedido chega pronto para análise: ortorretificado, com fusão pancromática, balanceamento de cores e mosaico ([veja exatamente o que isso envolve](/imagery)).

Se os dados gratuitos já levaram o seu projeto até onde podiam, [explore as imagens disponíveis no Pera Portal](https://portal.geopera.com/) ou [fale com a gente sobre o seu projeto](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Die 10 besten kostenlosen Quellen für Satellitenbilder in Deutschland

> Eine umfassende Übersicht der besten kostenlosen Quellen für Satellitenbilder und Satellitendaten in Deutschland

Published: 2026-06-11 | Author: Darcy Weedman | Reading time: 6 Minuten

Source: https://geopera.com/blog/kostenlose-satellitenbilder-deutschland

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Zusammenfassung

- [Pera Portal](https://portal.geopera.com/): Unbegrenztes, kostenloses Durchsuchen von Sentinel-2-Daten mit über 45 vorkonfigurierten Spektralindizes (NDVI, NDWI, EVI u. v. m.) direkt im Browser. Eigene Bandkombinationen, keine Installation nötig. Globale Abdeckung inklusive ganz Deutschlands — plus Zugang zu über 100 kommerziellen Satelliten, wenn Sie mehr Auflösung brauchen

- [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/): Die umfassendste Quelle für europäische Satellitendaten mit vollständiger Deutschland-Abdeckung

- [BKG (Bundesamt für Kartographie und Geodäsie)](https://gdz.bkg.bund.de/): Die beste Quelle für amtliche Geodaten und topographische Karten des Bundes

- [Geoportal.de](https://www.geoportal.de/): Zentraler Zugang zu Geodaten von Bund, Ländern und Kommunen

- [Google Earth](https://earth.google.com/web): Am einfachsten für die allgemeine Betrachtung — etwa des eigenen Grundstücks

- [Geoportale der Bundesländer](https://www.geoportal.de/): Hochauflösende Luftbilder und Geodaten aus Bayern, NRW, Baden-Württemberg und weiteren Ländern

- [ESA Earth Online](https://earth.esa.int/): Zugang zu den Satellitenmissionen und Forschungsdaten der Europäischen Weltraumorganisation

- [NASA Earthdata Search](https://search.earthdata.nasa.gov/): Ideal für wissenschaftliche Forschung und globale Satellitendaten-Archive

- [Maxar Open Data Program](https://www.maxar.com/open-data): Kostenlose hochauflösende Aufnahmen bei Hochwasser- und Katastrophenlagen

- [Zoom Earth](https://zoom.earth/): Nahezu aktuelle Wetter- und Umweltbeobachtung über Deutschland

<AskBand headline="Wie sieht Ihr Stück Deutschland diese Woche aus?" sub="Kostenloses Sentinel-2 mit Indizes im Browser, ohne Kreditkarte." action="Kostenlos ansehen" href="https://portal.geopera.com" tiles={['/band/fields.webp', '/band/city.webp', '/band/oxbow.webp']} />

## Überblick

Sie suchen ein Satellitenbild von Ihrem Haus in Deutschland? Oder möchten historische Satellitenaufnahmen Ihres Grundstücks kostenlos ansehen? Dann sind Sie hier richtig.

Dieser Leitfaden zeigt Ihnen die wichtigsten Quellen für Satellitendaten in Deutschland — egal, ob Sie Bilder direkt im Browser betrachten oder Daten für eigene Analysen herunterladen möchten.

### Gibt es Live-Satellitenbilder?

Nein. Echte Live-Satellitenbilder vom eigenen Haus sind mit heutiger Technik nicht möglich, anders als im Kino. Was Kartendienste zeigen, wurde in der Regel vor Wochen oder Monaten aufgenommen. Am nächsten kommen Wettersatelliten, deren Bilder im Minutentakt eintreffen, aber nur kilometergroße Pixel zeigen, und Sentinel-2, das jeden Ort etwa alle fünf Tage neu aufnimmt.

### Was in Deutschland verfügbar ist

Wenn Sie kostenlose Satellitenbilder von Deutschland suchen, sollten Sie Folgendes wissen:

- Die meisten kostenlosen Satellitenbilder werden nur alle paar Monate bis Jahre aktualisiert.
- Kostenlose Aufnahmen haben meist eine geringere Auflösung — Gebäude sind erkennbar, Details nicht.
- Hochauflösende Bilder mit klaren Details sind in der Regel kostenpflichtig.
- Echtzeit-Satellitendaten existieren, sind aber im Wesentlichen auf Wettervorhersage und Katastrophenschutz beschränkt.

Glücklicherweise ist Deutschland hervorragend versorgt: über das Copernicus-Programm der EU, über Bundesbehörden und über die Geoportale der Länder. Ob Eigentümer, Forscherin oder einfach neugierig — hier sind die besten kostenlosen (und kostenpflichtigen) Optionen.

---

## 1. Copernicus Data Space Ecosystem

Das [Copernicus-Programm](https://browser.dataspace.copernicus.eu/) der Europäischen Union ist die erste Adresse für kostenlose, hochwertige Satellitendaten in Europa — mit vollständiger Abdeckung Deutschlands.

**Wichtigste Merkmale:**

- Optische Sentinel-2-Aufnahmen alle 5 Tage mit 10 Metern Auflösung.
- Sentinel-1-Radardaten für wetterunabhängige Beobachtung.
- Vollständiges Archiv zurück bis 2014 (Sentinel-1) bzw. 2015 (Sentinel-2).
- Kostenloser Zugang ohne Download-Limits.
- Visualisierung und Download direkt in der Cloud.
- Ideal für Umweltmonitoring, Land- und Forstwirtschaft sowie Stadtentwicklung.

---

## 2. BKG — Bundesamt für Kartographie und Geodäsie

Das [Geodatenzentrum des BKG](https://gdz.bkg.bund.de/) stellt amtliche Geodaten des Bundes bereit, darunter Satellitenbildprodukte und topographische Karten.

**Wichtigste Merkmale:**

- Amtliche Geobasisdaten für ganz Deutschland.
- Digitale Geländemodelle und Höhendaten.
- Topographische Karten und Verwaltungsgrenzen.
- Eingebunden in europäische Geodaten-Initiativen.
- Viele Datensätze frei und ohne Registrierung zugänglich.
- Verbindliche Grundlage für Planung und Forschung.

---

## 3. Pera Portal

Unser [Pera Portal](https://portal.geopera.com/) bietet komplett kostenlosen, unbegrenzten Zugriff auf Sentinel-2-Satellitenbilder — mit Analysefunktionen, die über die meisten kostenlosen Plattformen hinausgehen.

**Wichtigste Merkmale:**

- **Unbegrenztes kostenloses Sentinel-2-Browsing** ohne Registrierungspflicht oder Download-Limits.
- **Über 45 vorkonfigurierte Spektralindizes**, sofort im Browser sichtbar:
  - NDVI (Vegetationszustand und Pflanzenmonitoring)
  - NDWI (Gewässererkennung für Flüsse und Seen)
  - EVI (erweiterter Vegetationsindex für die Landwirtschaft)
  - SAVI (bodenkorrigiert für Agrarregionen)
  - MSAVI, NDBI und viele mehr
- **Eigene Bandkombinationen** — erstellen Sie individuelle Indizes für spezielle Analysen.
- **Keine Einrichtung nötig** — aufbereitete Bilder sofort ansehen, ohne Download oder Software-Installation.
- **Globale Abdeckung**, inklusive ganz Deutschlands, alle 5 Tage aktualisiert.
- **Typische Anwendungen in Deutschland**:
  - Landwirtschaftliches Monitoring in Bayern, Niedersachsen und anderen Agrarregionen
  - Waldzustandserfassung in Schwarzwald, Bayerischem Wald und Harz
  - Beobachtung von Rhein, Donau und Elbe
  - Verfolgung der Stadtentwicklung in wachsenden Ballungsräumen
  - Standortbewertung für erneuerbare Energien (z. B. Solarparks)
  - Küstenmonitoring an Nord- und Ostsee
- **Zugang zu kommerziellen Aufnahmen** — bei Bedarf auf über 100 kommerzielle Satelliten mit Auflösungen unter einem Meter erweitern.

Anders als Plattformen, die technisches Vorwissen oder große Downloads voraussetzen, macht Pera Portal die Auswertung von Satellitenbildern für alle zugänglich — vom Landwirt bis zur Forschungsgruppe.

---

## 4. Geoportal.de

[Geoportal.de](https://www.geoportal.de/) ist der zentrale Zugangspunkt zu Geodaten von Bund, Ländern und Kommunen.

**Wichtigste Merkmale:**

- Einheitlicher Zugriff auf Daten aller 16 Bundesländer.
- Bündelt Satelliten- und Luftbilddaten von Bundes- und Landesebene.
- Standardisierte Suche über viele Datenanbieter hinweg.
- Verlinkt auf die Geoportale der Länder für detaillierte lokale Daten.
- Kostenloser Zugang zu harmonisierten Datensätzen.
- Konform mit der EU-INSPIRE-Richtlinie.

---

## 5. Google Earth

[Google Earth](https://earth.google.com/web) bietet hochauflösende Satellitenbilder für ganz Deutschland und ist die einfachste Plattform für alle, die ihr Haus aus dem All sehen möchten.

**Wichtigste Merkmale:**

- Einfache, intuitive Bedienung direkt im Browser.
- Zeitschieberegler für historische Aufnahmen.
- 3D-Gebäudemodelle für Berlin, München, Hamburg, Frankfurt und weitere Städte.
- Abdeckung aller Regionen von der Nordseeküste bis zu den Alpen.
- Nicht geeignet für den Download von Rohdaten oder fachliche Analysen.

---

## 6. Geoportale der Bundesländer

Die Bundesländer betreiben eigene Geoportale mit hochauflösenden Luftbildern und Geodaten — oft aktueller als die Angebote des Bundes:

### Bayern — BayernAtlas

Der [BayernAtlas](https://geoportal.bayern.de/bayernatlas/) bietet hochauflösende Luftbilder und Satellitendaten für ganz Bayern.

### Nordrhein-Westfalen — GEOportal.NRW

Das [GEOportal.NRW](https://www.geoportal.nrw/) stellt umfassende Geodaten für das bevölkerungsreichste Bundesland bereit.

### Baden-Württemberg — Geoportal BW

Das [Geoportal BW](https://www.geoportal-bw.de/) liefert detaillierte Luftbilder und Geodaten für Baden-Württemberg.

### Weitere Landesportale

- **Berlin**: [FIS-Broker Berlin](https://fbinter.stadt-berlin.de/)
- **Sachsen**: [GeoSN Geodatenportal](https://geoportal.sachsen.de/)
- **Hessen**: [Geoportal Hessen](https://geoportal.hessen.de/)
- **Niedersachsen**: [LGLN Geodatenportal](https://www.geodaten.niedersachsen.de/)

---

## 7. ESA Earth Online

[Earth Online](https://earth.esa.int/) der Europäischen Weltraumorganisation bietet Zugang zu den ESA-Satellitenmissionen und Erdbeobachtungsdaten.

**Wichtigste Merkmale:**

- Zugriff auf das vollständige Missionsarchiv der ESA.
- Sentinel-Daten mit zusätzlichen Verarbeitungsstufen und Produkten.
- Historische Daten älterer ESA-Missionen.
- Datensätze und Analysewerkzeuge in Forschungsqualität.
- Kostenlose Registrierung für den Archivzugang.
- Ideal für Wissenschaft und anspruchsvolle Anwendungen.

---

## 8. NASA Earthdata Search

[NASA Earthdata Search](https://search.earthdata.nasa.gov/) bietet umfassende Satellitendatensätze mit globaler Abdeckung — Deutschland eingeschlossen.

**Wichtigste Merkmale:**

- Vollständiges Landsat-Archiv zurück bis 1972 für Langzeitanalysen.
- MODIS-Daten für großräumiges Umweltmonitoring.
- Kostenlose Registrierung, unbegrenzte Downloads.
- Detaillierte Suchfilter nach Ort, Datum und Wolkenbedeckung.
- Unverzichtbar für Forschung und Klimastudien.

---

## 9. Maxar Open Data Program

[Maxar](https://www.maxar.com/open-data) stellt bei Naturkatastrophen und Notlagen in Deutschland hochauflösende kommerzielle Satellitenbilder kostenlos bereit.

**Wichtigste Merkmale:**

- Aufnahmen mit Auflösungen unter einem Meter bei Hochwasserlagen (etwa an Rhein, Elbe und Donau).
- Veröffentlichung innerhalb von 24–48 Stunden nach dem Ereignis.
- Entscheidend für die Einsatzkräfte bei Extremwetterereignissen.
- Vorher-Nachher-Aufnahmen für Schadensbewertung und Wiederaufbau.
- Partnerschaften mit europäischen Katastrophenschutzbehörden.

---

## 10. Zoom Earth

[Zoom Earth](https://zoom.earth/) zeigt häufig aktualisierte Satellitenbilder mit Schwerpunkt auf Wetterlagen und Umweltbedingungen über Deutschland.

**Wichtigste Merkmale:**

- Nahezu aktuelle Wetter- und Sturmverfolgung.
- Mehrmals täglich aktualisiert mit den neuesten Überflügen.
- Hervorragend zur Verfolgung von Wettersystemen von Nordsee und Atlantik.
- Zeitraffer-Animationen der Wetterentwicklung.
- Grundfunktionen ohne Registrierung nutzbar.

---

## Tipps für den Umgang mit kostenlosen Satellitenbild-Plattformen

**1. Die richtige Plattform wählen.**
Für deutschlandspezifische Daten beginnen Sie am besten bei Copernicus und den Geoportalen der Länder.

**2. Die Möglichkeiten der Plattformen kennen.**
Jede Plattform hat eigene Stärken. Prüfen Sie Auflösung, Datenqualität und Aktualisierungsrhythmus im Hinblick auf Ihr Projekt — Deutschlands Geodaten-Infrastruktur bietet ausgezeichnete kostenlose Optionen.

**3. Anforderungen klar definieren.**
Legen Sie fest, was Sie brauchen — Untersuchungsgebiet, Zeitraum, räumliche Auflösung, Spektralbänder — und nutzen Sie die Suchfilter gezielt.

**4. Visualisierungswerkzeuge nutzen.**
Viele Plattformen bringen eigene Analysewerkzeuge mit. Die Spektralindizes im Pera Portal etwa liefern Erkenntnisse zu Pflanzenzustand, Waldgesundheit und Gewässern — ganz ohne zusätzliche Software oder Fachkenntnisse.

**5. Gezielt herunterladen.**
Laden Sie statt kompletter Datensätze nur die Kacheln oder Gebiete herunter, die Sie wirklich brauchen. Für viele Anwendungsfälle machen Cloud-Plattformen wie Pera Portal Downloads ganz überflüssig.

Mit diesen Tipps holen Sie das Maximum aus den verfügbaren Satellitenbild-Ressourcen für Deutschland heraus — ob für Forschung, Immobilienbewertung, landwirtschaftliches Monitoring oder Umweltbeobachtung.

---

## Wenn kostenlose Satellitendaten nicht mehr ausreichen

Kostenlose Plattformen decken enorm viele Anwendungsfälle ab — wer nur das eigene Grundstück ansehen möchte, braucht nie etwas anderes. Wer Satellitenbilder jedoch beruflich nutzt, stößt früher oder später an die Auflösungsgrenze:

| Quelle                 | Auflösung | Was tatsächlich erkennbar ist                                 |
| ---------------------- | --------- | ------------------------------------------------------------- |
| Landsat (kostenlos)    | 30 m      | Regionale Landbedeckung — ein Fußballplatz ist etwa ein Pixel |
| Sentinel-2 (kostenlos) | 10 m      | Vegetationsmuster auf Feldebene; Gebäude bleiben unscharf     |
| Kommerziell (Geopera)  | bis 30 cm | Einzelne Fahrzeuge, Zäune, Solarmodule, einzelne Bäume        |

Der Unterschied zählt, sobald Sie **messen statt nur betrachten** wollen: Baufortschritt oder Industrieflächen Woche für Woche dokumentieren, Solar- und Windanlagen prüfen, Veränderungen auf Flurstücksebene statt auf Landkreisebene verfolgen — oder den Zustand eines Standorts zu einem bestimmten Stichtag belegen. Kostenlose Satelliten lassen sich zudem nicht beauftragen: Gab es im relevanten Zeitfenster keine wolkenfreie Aufnahme Ihres Gebiets, existiert schlicht kein Bild.

Kommerzielle Satellitenbilder bedeuteten früher intransparente Angebote und wochenlange Abstimmung. Wir veröffentlichen [transparente Preise pro Quadratkilometer](/pricing) — für Neuaufnahmen wie für Archivdaten — und jede Bestellung kommt analysefertig an: orthorektifiziert, pansharpened, farblich angeglichen und mosaikiert ([hier zeigen wir Schritt für Schritt, was dahintersteckt](/imagery)).

Wenn kostenlose Daten Ihr Projekt so weit gebracht haben, wie sie können: [Erkunden Sie verfügbare Aufnahmen im Pera Portal](https://portal.geopera.com/) oder [sprechen Sie mit uns über Ihr Projekt](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# What is Atmospheric Correction in Satellite Imagery?

> Atmospheric correction turns raw satellite imagery into surface reflectance, the data you need for accurate NDVI, change detection, and analysis.

Published: 2026-05-26 | Author: Darcy Weedman | Reading time: 13 min

Source: https://geopera.com/blog/atmospheric-correction-satellite-imagery

---
<script>
import { AskBand, BarChart, Callout, ComparisonScale, SpectrumBar } from '$lib/components/charts';
</script>

## Summary

- **Atmospheric correction removes the optical effects of the atmosphere from satellite imagery**, converting top-of-atmosphere (TOA) reflectance into surface reflectance, the actual brightness of the ground.
- **Raw satellite imagery includes everything between the satellite and the ground**: aerosols, water vapour, dust, ozone, smoke. That's not what you want to measure.
- **Without correction, the same patch of ground can produce wildly different NDVI values** from week to week, purely because the atmosphere changed. That kills time-series analysis.
- **Sentinel-2 Level-2A and Landsat Collection 2 Level-2 ship pre-corrected** using Sen2Cor and LaSRC respectively. Most commercial high-resolution imagery does not.
- **Different methods exist for different needs**: Sen2Cor and ACOLITE are free, FLAASH and ATCOR are paid, and dark object subtraction (DOS) is the rough-and-ready option when nothing better is available.

A vegetation monitoring project images the same patch of crops twice, three weeks apart. Cloud-free both times. Same satellite, same sensor, same bands. The NDVI numbers swing by 30%.

The crops didn't change that much. The atmosphere did.

This is the part of satellite imagery nobody warns you about. Every photo you've ever seen taken from space is, in a small but important way, lying to you. The satellite isn't really looking at the ground. It's looking at the top of the atmosphere, with the ground sort of fuzzily visible through several kilometres of air, water vapour, dust, and whatever forest-fire smoke happened to be drifting through that morning.

Atmospheric correction is how we strip that atmosphere out. Without it, you're not measuring what's on the ground. You're measuring what's on the ground plus everything between you and the ground. For some applications that's fine. For most serious ones, it's a disaster.

<AskBand headline="What is the sky doing to your pixels?" sub="Free Sentinel-2 arrives with ESA's correction applied; compare over your own area." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/bloom.webp', '/band/atoll.webp', '/band/fields.webp']} />

## The Atmosphere Is Lying to Your Satellite

Here's the basic problem. Sunlight starts at the sun, travels 150 million kilometres through empty space, and then hits Earth's atmosphere. That last bit, maybe 100 kilometres of stuff, is where everything gets messy.

The atmosphere scatters some of the light sideways before it ever reaches the ground. (That's why the sky is blue. Blue wavelengths scatter more than red.) It absorbs some of it entirely, especially in specific water-vapour and ozone bands. The light that does make it through hits the surface, bounces back up, and then has to make the whole trip in reverse, getting scattered and absorbed all over again before it reaches the satellite.

![Diagram showing how sunlight scatters and gets absorbed as it passes through the atmosphere down to the surface and back up to a satellite, illustrating why atmospheric correction is needed for satellite imagery](/blogs/atmospheric-correction-light-path.webp)

So when a satellite "sees" a pixel of imagery, it's actually seeing a mixture of:

1. Light that reflected cleanly off the ground (the bit you want)
2. Light that scattered off aerosols in the air and never touched the ground at all (called "path radiance", the bit you don't want)
3. Light from the ground that got attenuated on its way back up (less than there should be)
4. Light absorbed by water vapour, ozone, oxygen, CO2, and methane in specific narrow bands

The raw measurement, the number the sensor records before any processing, is called **top-of-atmosphere (TOA) reflectance**. It's perfectly accurate. It's also not what you want.

What you want is **surface reflectance**: how bright the actual ground is, independent of whatever atmospheric conditions happened to exist that day. That number is comparable across dates, across sensors, across locations. TOA isn't.

<Callout type="tldr" title="The two-word version">
TOA reflectance = ground + atmosphere. Surface reflectance = ground only. Atmospheric correction is the maths that gets you from the first to the second.
</Callout>

## What Atmospheric Correction Actually Does

The correction has to undo physical effects, so it works by modelling those effects and reversing them. The classic approach uses a radiative transfer model. Basically a computational simulation of how light moves through the atmosphere given certain conditions.

Feed the model with:

- **Aerosol optical depth** (how dusty or hazy the air is)
- **Water vapour content** (how much moisture sits between the ground and the satellite)
- **Ozone column** (mostly relevant for UV bands)
- **Sun-target-sensor geometry** (the angles involved)
- **Surface elevation** (more atmosphere above sea-level surfaces than above mountains)

The model then tells you: given these conditions, this is how much atmospheric distortion exists in each pixel's measurement. Subtract that out, and what's left is surface reflectance.

The hard part is getting the atmospheric parameters. Some come from external sources, like ozone columns from NOAA and water vapour from MODIS retrievals or from the satellite's own bands. Aerosol optical depth is the trickiest one to nail down, and it's also the parameter that matters most. Half the engineering of atmospheric correction is "how do we estimate aerosols well enough."

## TOA vs Surface Reflectance, Visually

Here's the same scene in both forms. Same satellite, same date, same area. The only difference is whether atmospheric correction has been applied.

![Side-by-side comparison of the same satellite image showing top-of-atmosphere reflectance with a hazy blue cast on the left and atmospherically corrected surface reflectance with vivid true colours on the right](/blogs/atmospheric-correction-before-after.webp)

On the left, that milky blue cast is path radiance. Light bouncing off aerosols in the air, never actually touching the ground. It makes water look brighter than it is. It softens contrast. It pushes everything toward the same washed-out tone.

On the right, after correction, dark features stay dark. The water is properly deep blue. The forest is properly dark green. The brown fields are properly brown. This is what you'd see if you somehow stripped away the atmosphere and looked straight at the ground.

The visual difference is obvious. The analytical difference is far more important. The NDVI you calculate from the right image actually reflects vegetation health. The NDVI from the left image reflects vegetation health plus whatever the atmosphere was doing that morning, and the atmosphere changes constantly.

## The Methods People Actually Use

A handful of correction methods dominate the field. They differ in physical rigour, computational cost, and how badly they fall over when atmospheric inputs are missing.

| Method       | What it is                                                 | Free? | Best for                                       |
| ------------ | ---------------------------------------------------------- | ----- | ---------------------------------------------- |
| **Sen2Cor**  | ESA's official Sentinel-2 processor (Level-1C to Level-2A) | Yes   | Default Sentinel-2 correction                  |
| **LaSRC**    | USGS Landsat Surface Reflectance Code                      | Yes   | Default Landsat correction                     |
| **ACOLITE**  | RBINS open-source tool, especially good over water         | Yes   | Coastal, inland water, lakes                   |
| **FLAASH**   | MODTRAN-based, sold with ENVI                              | No    | Commercial multispectral and hyperspectral     |
| **ATCOR**    | Closed-source, sold with ERDAS or standalone               | No    | Production environments with mixed sensors     |
| **6S / 6SV** | Academic radiative transfer model, basis of many others    | Yes   | Research, custom pipelines                     |
| **DOS**      | Dark Object Subtraction, empirical not physics-based       | Yes   | Quick-and-dirty when nothing else is available |
| **iCOR**     | VITO's correction tool, works for Sentinel-2 and Landsat   | Yes   | Multi-sensor consistency                       |
| **Polymer**  | Aerosol-tolerant correction for ocean colour               | Yes   | Open ocean, turbid water                       |

A few patterns are worth pointing out.

**Sen2Cor and LaSRC are the default for the free data sources.** If you download Sentinel-2 Level-2A or Landsat Collection 2 Level-2 from [ESA or USGS](/blog/free-sources-of-satellite-data), the correction has already been applied. You don't have to do anything. This is one of the underrated wins of the free data programmes: billions of square kilometres of atmospherically corrected imagery, no licence cost, no processing required.

**Commercial high-resolution imagery usually does not ship corrected.** WorldView, Pléiades, Beijing-3, Perascope, SuperView. When you buy these products as standard, you typically get either raw TOA imagery or basic radiometrically calibrated imagery with the atmosphere still very much present. Atmospheric correction is an additional step you (or your provider) have to do.

**ACOLITE and Polymer were built specifically because the standard methods fall apart over water.** Surface reflectance over deep water is tiny. Most of what the satellite sees is path radiance from the atmosphere, not bottom reflectance or even surface reflectance. Ocean colour applications need correction methods that are aerosol-tolerant in a way Sen2Cor wasn't designed for.

**Dark Object Subtraction is the duct-tape solution.** It assumes that the darkest pixel in your image (a deep shadow, a clear lake) should have effectively zero reflectance in the visible bands, so whatever brightness it does have must be atmospheric. Subtract that value from every pixel and call it done. It's empirical, ignores wavelength-specific scattering, and produces results that range from "surprisingly OK" to "definitely wrong" depending on the scene. It works with zero auxiliary data, which is sometimes all you have.

<Callout type="info" title="What level number am I looking at?">
ESA Level-2A and USGS Level-2 both mean "atmospherically corrected surface reflectance." Level-1C (Sentinel-2) and Level-1 (Landsat) mean TOA reflectance, uncorrected. If a product file ends in "_SR" or "_L2A" you're working with corrected data.
</Callout>

## What Wavelengths Get Hit Hardest

Not all bands suffer equally. Atmospheric effects are strongly wavelength-dependent, and that determines how much correction matters for any given application.

<SpectrumBar
title="How Much the Atmosphere Distorts Each Spectral Region"
segments={[
{ label: 'Blue', sublabel: '~450 nm', width: 3, color: '#1d4ed8' },
{ label: 'Green', sublabel: '~550 nm', width: 2.5, color: '#059669' },
{ label: 'Red', sublabel: '~650 nm', width: 2, color: '#dc2626' },
{ label: 'NIR', sublabel: '~850 nm', width: 1.5, color: '#7c2d12' },
{ label: 'SWIR1', sublabel: '~1600 nm', width: 1, color: '#6b7280' },
{ label: 'SWIR2', sublabel: '~2200 nm', width: 0.8, color: '#374151' },
]}
leftLabel="Most distortion"
rightLabel="Least distortion"
/>

Blue light scatters off everything. That's why the sky is blue, why distant mountains look hazy, and why the blue band is the most affected by aerosols. Green and red scatter less. Near-infrared barely scatters at all, which is part of why NIR is so useful for vegetation analysis. Shortwave infrared barely interacts with the atmosphere at clean-sky conditions but gets absorbed strongly by water vapour in specific bands.

The practical implication: anything that uses the blue band (true-colour visualisation, water turbidity, ocean colour) needs proper correction. Anything that primarily uses NIR (most vegetation indices) is more forgiving but still benefits.

[NDVI](/blog/remote-sensing-vegetation-indices) uses red and NIR. Without correction, the red value is too high because atmospheric path radiance adds brightness to it, which compresses the difference between red and NIR, which makes vegetation look less vigorous than it actually is. Over time, as atmospheric conditions vary, your NDVI baseline drifts. Suddenly a healthy field looks like it's declining. It isn't.

## When You Need Correction (and When You Can Skip It)

This is the part most guides skip. Atmospheric correction matters more for some applications than others. Knowing which is which saves time and money.

**You absolutely need atmospheric correction for:**

- **Time-series analysis** of any kind. Comparing images from different dates only works if the atmosphere has been removed. Otherwise you're partly measuring weather.
- **Quantitative vegetation indices** (NDVI, EVI, NDWI, NDRE). Even small TOA errors compound into meaningful index drift. Geopera's docs site catalogues [over 350 spectral indices](https://docs.geopera.com), and nearly every one assumes you're working with surface reflectance.
- **Change detection** across multiple acquisitions. Same logic. Without correction, atmospheric variation will look like real change.
- **Multi-sensor analysis.** Combining Sentinel-2 with Landsat with WorldView without correction is impossible. Each sensor has slightly different atmospheric responses; correction normalises them.
- **Quantitative water applications.** Bathymetry, water quality, algal bloom detection. The signal from water is tiny. The atmosphere dominates the raw measurement.
- **Anything you'll feed to a machine learning model** that you want to generalise across times and places.

**You can skip it (mostly) for:**

- **One-off visualisation** where you just need a pretty picture for a slide deck. TOA looks fine enough for a single date.
- **Visual interpretation** by an analyst comparing features in a single scene. Object detection, asset counting, parking lot censuses. The atmosphere doesn't really change the answer.
- **Cases where you'll do an empirical relative calibration** anyway, like normalising to known invariant features within the scene.

<Callout type="warning" title="The expensive trap">
Buying high-resolution commercial imagery without checking whether it's atmospherically corrected. Most isn't. You'll get a beautiful 30cm WorldView image, throw it into your NDVI pipeline, and get garbage results. Not because the imagery is bad, but because nobody corrected for the air between the satellite and the field.
</Callout>

## The Honest Limitations

Atmospheric correction is physics-based modelling, which means it's only as good as its inputs. A few things can go wrong.

**Bad aerosol estimates.** If the model thinks there's clean air over your scene but actually there's smoke from a forest fire, the correction will be wrong. The error is sometimes larger than the atmospheric effect itself.

**Adjacency effects.** Light that scattered off bright neighbouring pixels (a white roof, a snowfield, a cloud edge) bleeds into the measurement of dark pixels. Standard atmospheric correction doesn't fully handle this.

**Thin cirrus clouds.** These are nearly invisible in standard cloud masks but still affect reflectance. Some processors use a specific cirrus band (Sentinel-2 has one at 1380nm) to detect and partially correct for this.

**Cloud shadows.** A pixel in cloud shadow gets less direct sunlight than a pixel in full sun. Atmospheric correction handles direct illumination but not shadow geometry.

**Bright over-corrections in shallow water.** Some methods, tuned for land, can produce nonsense negative reflectance over clear shallow water where the actual signal is mostly bottom reflectance.

Knowing these failure modes matters. A correction product isn't the same as ground truth. It's an estimate, with error bars, and those error bars get bigger when conditions are unusual.

## Where This Fits With What We Do

At Geopera, atmospheric correction is part of [the standard processing](/blog/why-we-process-every-order) we apply to every commercial imagery order, alongside [orthorectification](/blog/orthorectification-explained) and [pansharpening](/blog/pansharpening-satellite-imagery-explained). It's not an extra. It's not a paid upgrade. It happens before the file ever reaches you.

The reason is straightforward. If you bought satellite imagery to do something with it (calculate vegetation indices, detect change, monitor a coastline, feed a model) then uncorrected TOA data isn't fit for purpose. Delivering it raw and letting you "handle the correction yourself" is the polite version of delivering broken product.

Most of the [free data sources](/blog/free-sources-of-satellite-data) already come corrected (Sentinel-2 L2A, Landsat C2 L2). For everything else, including [Maxar/Vantor WorldView](/blog/vantor-lanteris-maxar-rebrand), 21AT's Beijing-3, our own Perascope, SpaceWill's SuperView, and Wyvern hyperspectral, we run correction ourselves before delivery. You get surface reflectance, not raw TOA, with no extra step on your end.

This is why our [pricing](/pricing) looks different from other providers'. Most charge a base rate for raw imagery and then 30-80% extra for processing. We charge one number that already includes the work that turns raw imagery into something analysis-ready.

If you'd rather see it work than read about it, you can [order processed imagery through the Pera Portal](https://portal.geopera.com), or [talk to our team](/contact) about what processing your project actually needs.

## Frequently Asked Questions

**What is atmospheric correction in remote sensing?**

Atmospheric correction is the process of removing the optical effects of Earth's atmosphere (scattering and absorption by aerosols, water vapour, ozone, and gases) from satellite imagery. The output is surface reflectance, which represents the actual brightness of the ground rather than the brightness recorded at the top of the atmosphere.

**Do I need atmospheric correction for NDVI?**

For single-date qualitative visualisation, no. For any quantitative or time-series NDVI analysis, yes. Without correction, NDVI values from the same field on different dates can swing by 20-40% purely from atmospheric variability, which makes trend analysis unreliable.

**Is Sentinel-2 Level-2A already atmospherically corrected?**

Yes. Sentinel-2 Level-2A products are corrected using ESA's Sen2Cor processor and represent surface reflectance. Level-1C is the uncorrected top-of-atmosphere version. Most users should download L2A unless they need to apply a different correction method.

**What's the difference between TOA reflectance and surface reflectance?**

TOA (top-of-atmosphere) reflectance is what the satellite sensor measures directly, which includes both the ground signal and atmospheric effects. Surface reflectance is what's left after atmospheric correction removes those effects, representing the actual brightness of the ground. Surface reflectance is comparable across dates and sensors; TOA reflectance is not.

**Which atmospheric correction software is best?**

It depends on the use case. For Sentinel-2 over land, Sen2Cor is the default and works well. For Sentinel-2 over coastal and inland water, ACOLITE generally outperforms Sen2Cor. For Landsat, USGS Collection 2 Level-2 (LaSRC) is the standard. For commercial high-resolution imagery, FLAASH and ATCOR are industry standards but require paid licences. 6S underlies several of these and is the choice for research and custom pipelines.

**Can dark object subtraction replace proper atmospheric correction?**

Sometimes. DOS is fast, requires no auxiliary data, and produces acceptable results for visual interpretation. It struggles with wavelength-dependent scattering, doesn't handle absorption properly, and falls over when no genuinely dark pixels exist in the scene. For quantitative work, prefer a physics-based correction method when one is available.

---

Need analysis-ready satellite imagery without the processing headache? Geopera applies atmospheric correction (along with orthorectification, pansharpening, and spectral index calculation) to every order, included in the base price. [Order through the Pera Portal](https://portal.geopera.com) or [talk to our team](/contact) about what your project needs.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Multispectral vs Hyperspectral Satellite Imagery Explained

> Multispectral vs hyperspectral satellite imagery: what changes when a sensor captures 200 bands instead of 4? Resolution trade-offs, costs, and which to choose.

Published: 2026-03-24 | Author: Darcy Weedman | Reading time: 12 min

Source: https://geopera.com/blog/multispectral-vs-hyperspectral-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- **The multispectral vs hyperspectral question** comes down to breadth versus depth. Multispectral captures 4-16 broad bands and remains the workhorse of commercial Earth observation, powering vegetation indices, land-use classification, and change detection at resolutions down to 0.3 metres.
- **Hyperspectral sensors capture 20-200+ narrow, contiguous bands**, enabling identification of specific materials, minerals, and vegetation species that multispectral can't distinguish.
- **The core trade-off is spatial resolution vs spectral detail.** Commercial multispectral delivers 0.3-3m pixels. Most hyperspectral satellites sit at 5-30m.
- **Commercial hyperspectral satellite data became widely available in 2024-2025**, with Wyvern's Dragonette constellation (5.3m, 23-32 bands) joining research missions like PRISMA, EnMAP, and NASA's EMIT.
- **Most projects need multispectral.** Hyperspectral becomes worth the investment when you need to identify _what_ something is, not just _where_ it is.

Here's something odd about the colour green.

To your eyes, a healthy wheat field and a paddock full of invasive weeds look almost identical. Same shade. Same vibe. A standard multispectral satellite can confirm they're both vegetation, picking up chlorophyll absorption in the red and near-infrared bands without breaking a sweat. But it can't tell you which is the crop and which is the weed.

A hyperspectral sensor can.

That distinction is worth serious money to the right buyer. Wheat from weeds, hematite from goethite, healthy coral from bleached coral. It all comes down to one thing: how many slices of the electromagnetic spectrum your sensor captures. The _multispectral vs hyperspectral_ question gets asked constantly, and the honest answer is "you probably need multispectral." But "probably" isn't good enough when you're spending thousands on satellite data, so let's actually break down what these terms mean, where the trade-offs bite, and when hyperspectral genuinely justifies the extra cost.

<AskBand headline="How many bands does the job need?" sub="The catalogue lists band counts so you can compare like for like." action="Compare band counts" href="/sensors" tiles={['/band/bloom.webp', '/band/saltponds.webp', '/band/fields.webp']} />

## The Imaging Spectrum: Panchromatic to Hyperspectral

Before we compare multispectral and hyperspectral, it helps to see where they sit on the broader spectrum of [optical satellite imaging](/blog/understanding-optical-satellite-imagery).

<figure>

![Diagram showing the progression from panchromatic (1 band) to multispectral (4-16 bands) to hyperspectral (20-200+ bands) satellite imaging](/blogs/panchromatic-multispectral-hyperspectral-spectrum.webp)

<figcaption>The three tiers of optical satellite imaging — panchromatic captures one wide band for maximum spatial resolution, multispectral captures several targeted bands, and hyperspectral captures hundreds of contiguous narrow bands.</figcaption>

</figure>

**Panchromatic** sensors capture a single, wide band spanning most of the visible spectrum. One band. Maximum spatial resolution. Black and white. [WorldView-3](/sensors/worldview-3) achieves 0.31m resolution in panchromatic mode. You can count cars in a parking lot, but you can't tell a red one from a blue one.

**Multispectral** sensors split the spectrum into 4-16 discrete bands, typically blue, green, red, near-infrared (NIR), and sometimes short-wave infrared (SWIR). Each band is 40-200nm wide. This is what most commercial satellites do. Sentinel-2 captures 13 bands at 10-60m resolution. WorldView-3 captures 16 bands (including 8 SWIR) at 1.24m multispectral resolution. Planet's SuperDove fleet captures 8 bands at 3m. These sensors sample the spectrum at specific, strategically chosen wavelengths.

**Hyperspectral** sensors take a fundamentally different approach. Instead of sampling a few broad chunks of the spectrum, they capture dozens to hundreds of narrow, contiguous bands, each just 5-20nm wide. No gaps between them. Think of it as the difference between sampling a song at five points versus recording the entire track from start to finish.

## How Multispectral Imaging Actually Works

A multispectral sensor is selective by design. It captures specific wavelength ranges that scientists have determined are useful for particular types of analysis. The near-infrared band is included because healthy vegetation reflects NIR strongly. That's where [NDVI and other vegetation indices](/blog/remote-sensing-vegetation-indices) come from.

The gaps between bands aren't a limitation. They're a deliberate engineering choice. By capturing fewer, wider bands, multispectral sensors collect more light per band, which means higher spatial resolution and a better signal-to-noise ratio. For most applications, that's the right trade-off.

What multispectral does well:

- **Vegetation health monitoring** using NDVI, EVI, and SAVI indices calculated from red and NIR bands. This is the backbone of precision agriculture and forestry monitoring globally.
- **Land cover classification** to distinguish water, urban areas, vegetation, and bare soil. Four or five bands is plenty.
- **Change detection** by comparing imagery from different dates to spot construction, deforestation, flooding, or urban expansion.
- **Water quality assessment** using blue and green bands to reveal turbidity, algal blooms, and sediment concentration.

What multispectral _can't_ do: tell you which species of vegetation is stressed, or which specific mineral is in that exposed rock face, or which chemical contaminated that waterway. All green vegetation looks roughly the same in 4-8 bands. You know it's there. You just don't know what it is.

## How Hyperspectral Imaging Works

Hyperspectral sensors capture the full spectral curve of every pixel. Where a multispectral sensor gives you five or ten data points along the electromagnetic spectrum, a hyperspectral sensor gives you two hundred.

Why does that matter?

Because every material on Earth has a unique spectral signature. A specific pattern of how it absorbs and reflects different wavelengths based on its molecular composition. Chlorophyll absorbs red light and reflects NIR. Iron oxide absorbs around 900nm. Calcite has a distinctive absorption feature near 2,300nm.

With enough spectral detail, you can identify these absorption features and match them against known spectral libraries. That's how a geologist can distinguish hematite (Fe₂O₃) from goethite (FeOOH) from 600 kilometres above the Earth. They look the same colour to the naked eye but have completely different spectral curves between 800-1000nm.

<figure>

![Comparison of multispectral vs hyperspectral band coverage across the electromagnetic spectrum](/blogs/multispectral-vs-hyperspectral-bands.webp)

<figcaption>Multispectral sensors capture a handful of broad bands with gaps between them. Hyperspectral sensors capture hundreds of narrow, contiguous bands, revealing the full spectral signature of every material in the scene.</figcaption>

</figure>

What hyperspectral does that multispectral can't:

- **Mineral identification** to distinguish individual mineral species in exposed rock and soil. Copper carbonate, gypsum, kaolinite, alunite — each has unique absorption features only visible in narrow bands.
- **Vegetation species discrimination** that can tell eucalyptus from acacia from pine. Not just "it's green" but "it's _this specific_ green."
- **Soil composition mapping** to identify organic carbon content, moisture levels, clay mineralogy, and contamination.
- **Water constituent analysis** that detects specific phytoplankton species, dissolved organic matter, and individual pollutant types.
- **Gas emission detection** for methane, CO₂, and other greenhouse gases with spectral absorption signatures that hyperspectral sensors can pick up.

## Multispectral vs Hyperspectral: The Key Differences

Here's where the comparison gets concrete.

| Feature                                | Multispectral                       | Hyperspectral                         |
| -------------------------------------- | ----------------------------------- | ------------------------------------- |
| **Spectral bands**                     | 4-16 bands                          | 20-200+ bands                         |
| **Band width**                         | 40-200nm (broad)                    | 5-20nm (narrow)                       |
| **Band arrangement**                   | Discrete, with gaps                 | Contiguous, no gaps                   |
| **Best commercial spatial resolution** | 0.31m pan / 1.24m MS (WorldView-3)  | 5.3m (Wyvern Dragonette)              |
| **Typical spatial resolution**         | 0.5-10m                             | 5-30m                                 |
| **Data volume per scene**              | Moderate (4-16 layers)              | Very large (100-200+ layers)          |
| **Processing complexity**              | Standard GIS tools                  | Specialist spectral analysis software |
| **Primary strength**                   | Where things are                    | What things are                       |
| **Approximate cost**                   | $2-55/km² (archive)                 | $10-80/km² (varies by provider)       |
| **Commercial availability**            | Abundant (dozens of constellations) | Limited but growing fast              |

The spatial resolution gap is the thing that trips people up. If you need to count individual trees, multispectral at 0.5m beats hyperspectral at 30m every time. But if you need to know which trees are _Eucalyptus camaldulensis_ and which are _Eucalyptus globulus_, the spectral detail wins, even at lower spatial resolution.

**Data volume is the other big consideration.** A single hyperspectral scene might be 10-50x larger than the equivalent multispectral image. More storage, more processing time, more specialised software. Standard GIS packages handle multispectral data out of the box. Hyperspectral analysis typically requires tools like ENVI, or custom Python workflows using libraries like `spectral` or `hyperspy`.

## When to Use Which: A Decision Framework

Don't default to hyperspectral because it sounds fancier. It isn't always better. It's different.

**Use multispectral when you need:**

- High spatial resolution (sub-metre)
- Frequent revisit for time-series monitoring
- Standard index calculations (NDVI, NDWI, NDMI)
- Land cover classification and change detection
- Cost-effective coverage of large areas
- Quick turnaround and simple processing

**Use hyperspectral when you need to:**

- Identify specific minerals in mining exploration or environmental assessment
- Discriminate between similar plant species for biodiversity mapping or invasive species detection
- Map soil properties like organic carbon, clay content, or contamination levels
- Detect subtle crop stress _before_ it shows up in standard NDVI
- Monitor specific water pollutants or algal species
- Quantify atmospheric gas concentrations

**Consider using the two together** when the project demands it. A mining company might use multispectral imagery at 0.5m to map the site layout and track physical changes monthly, then task a hyperspectral collection once or twice a year to assess mineral composition in tailings and monitor vegetation species during rehabilitation. That layered approach gets you spatial precision and spectral depth without blowing the budget on either.

## Commercial Hyperspectral Satellites in 2026

For a long time, hyperspectral satellite data meant waiting months for access to research missions with restrictive licences and 30m resolution at best. That picture has changed.

**Currently operational:**

- **Wyvern Dragonette** (Dragonette-001 to -004): 5.3m GSD, 23-32 bands in the VNIR range (400-1000nm). The highest-resolution commercial hyperspectral constellation flying today. Wyvern, a Canadian company founded in 2018, has Dragonette-005 and -006 scheduled for launch later in 2026. [Geopera is a Wyvern distribution partner](/blog/geopera-wyvern-partner), offering Dragonette data through our [imagery platform](/imagery).
- **PRISMA** (ASI, Italy): 30m resolution, 239 bands covering 400-2505nm across VNIR and SWIR. Operational since 2019. Free for research use through the [Italian Space Agency](https://www.asi.it/en/earth-science/prisma/).
- **EnMAP** (DLR, Germany): 30m resolution, 242 bands covering 420-2450nm. Launched April 2022. Data available through DLR's EOWEB GeoPortal.
- **EMIT** (NASA/JPL): 60m resolution, 285 bands covering 380-2500nm. Mounted on the International Space Station since July 2022. Originally designed for mineral dust source mapping, but all data is publicly accessible through [NASA Earthdata](https://www.earthdata.nasa.gov/).
- **DESIS** (DLR/Teledyne): 30m resolution, 235 bands covering 400-1000nm (VNIR only). Also mounted on the ISS.

**Compared to multispectral availability**, hyperspectral options are still limited. Dozens of multispectral constellations orbit Earth right now. Sentinel-2, Landsat 9, WorldView-3, Beijing-3, Perascope, Planet SuperDove. They cover the planet daily at resolutions from 0.3m to 10m. Hyperspectral doesn't have that density yet. Wyvern's constellation expansion and new commercial entrants will change the equation over the next two to four years, but today, getting cloud-free hyperspectral data over a specific area still requires more planning and patience than pulling multispectral from an archive.

## Hyperspectral vs Multispectral Processing: What You're Signing Up For

Something the typical comparison article never tells you: the processing burden is wildly different between hyperspectral and multispectral data.

**Multispectral processing** is well-understood territory. Orthorectification, atmospheric correction, pansharpening, index calculation. These are standard workflows that any competent imagery provider handles as part of delivery. When you [order satellite imagery](/blog/how-to-buy-satellite-imagery) that's multispectral, you can reasonably expect analysis-ready data that loads straight into QGIS or ArcGIS.

**Hyperspectral processing** is a different animal. Beyond the standard geometric and atmospheric corrections, you'll typically need:

- **Noise reduction** across hundreds of bands (atmospheric water vapour absorption creates noisy bands that need masking or interpolation)
- **Dimensionality reduction** using Principal Component Analysis (PCA) or Minimum Noise Fraction (MNF) transforms to identify the most information-rich band combinations
- **Spectral unmixing**, because most hyperspectral pixels at 30m contain a mix of materials. Unmixing algorithms decompose each pixel into its constituent materials and their proportions.
- **Spectral library matching** to compare pixel signatures against reference databases (like the USGS Spectral Library) and identify materials

This isn't a few clicks in QGIS. It's specialist work that requires someone comfortable with ENVI, Python spectral analysis libraries, or similar tools. Without that expertise, the data just sits on a hard drive doing nothing.

**The practical takeaway:** factor processing capability into your decision. The [cost of satellite imagery](/blog/satellite-imagery-cost-guide) isn't just the acquisition price. It's acquisition plus the expertise to turn 200 bands of raw data into actionable information. If you don't have that in-house, find a provider who handles it.

## How We Handle This at Geopera

We work across the full optical spectrum. Our platform connects you to [multispectral sensors](/sensors) from Vantor (formerly Maxar), 21AT's Beijing-3, our own Perascope, SpaceWill's SuperView, and Sentinel-2, plus Wyvern's [Dragonette constellation](/sensors/dragonette-001) for hyperspectral. We also maintain [over 350 spectral indices](https://docs.geopera.com) on our documentation site, calculated and documented with code samples.

Every order goes through our processing pipeline: orthorectification, atmospheric correction, pansharpening where applicable. That processing is included, not an add-on. Whether you need a 0.3m multispectral scene for infrastructure monitoring or a 5.3m hyperspectral capture for mineral exploration, you get analysis-ready data.

Not sure which type fits your project? See how we deliver both multispectral and hyperspectral data, processed and analysis-ready, through our [satellite imagery platform](/imagery).

## Frequently Asked Questions

### What is the difference between multispectral and hyperspectral satellite imagery?

**Multispectral sensors capture 4-16 broad, non-contiguous spectral bands** (40-200nm wide each), optimised for general-purpose Earth observation like vegetation monitoring and land classification. **Hyperspectral sensors capture 20-200+ narrow, contiguous bands** (5-20nm wide each), enabling identification of specific materials through their unique spectral absorption signatures. In practical terms, multispectral tells you _where_ things are; hyperspectral tells you _what_ they are.

### How does hyperspectral imaging work?

Hyperspectral sensors record reflected sunlight across hundreds of narrow, adjacent wavelength bands, producing a complete spectral curve for every pixel. Each material on Earth absorbs and reflects light differently at specific wavelengths due to its molecular composition. Analysts compare these per-pixel spectral curves against known reference libraries to identify the chemical composition and physical properties of surface materials from orbit.

### Is hyperspectral satellite imagery commercially available in 2026?

Yes. Commercial hyperspectral data is available from Wyvern's Dragonette constellation at 5.3m resolution with 23-32 VNIR bands, accessible through providers including Geopera. Free research-grade data is available from PRISMA (30m, 239 bands), EnMAP (30m, 242 bands), and NASA's EMIT (60m, 285 bands). Commercial availability remains more limited than multispectral but is expanding as Wyvern and others grow their constellations.

### When should I choose hyperspectral over multispectral?

Choose hyperspectral when your analysis requires material identification rather than spatial mapping. That means mineral exploration, vegetation species discrimination, soil composition analysis, water quality constituent detection, or atmospheric gas monitoring. For change detection, land cover classification, general vegetation health monitoring, or any application requiring sub-metre resolution, multispectral is the better and more cost-effective choice.

### What spatial resolution do hyperspectral satellites achieve?

The highest-resolution commercial hyperspectral satellite in 2026 is Wyvern's Dragonette at **5.3m ground sampling distance (GSD)**. Research missions like PRISMA and EnMAP operate at 30m, and NASA's EMIT at 60m. For comparison, leading multispectral satellites achieve 0.31m (WorldView-3 panchromatic) to 3m (Planet SuperDove). The resolution gap exists because capturing more spectral bands means each detector element collects less light per band, reducing signal strength at finer spatial scales.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# What is Pansharpening in Satellite Imagery?

> Pansharpening fuses high-res panchromatic and lower-res multispectral satellite bands into one sharp colour image. How it works, when to use it, and the trade-offs.

Published: 2026-03-24 | Author: Darcy Weedman | Reading time: 10 min

Source: https://geopera.com/blog/pansharpening-satellite-imagery-explained

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- **Pansharpening is an image fusion technique** that combines a high-resolution panchromatic (black-and-white) band with lower-resolution multispectral (colour) bands to produce a single sharp, full-colour image.
- **The spatial resolution improvement is typically 4x.** A satellite like [WorldView-3](/sensors/worldview-3) captures 0.31m panchromatic and 1.24m multispectral. Pansharpening gives you 0.31m colour imagery.
- **The trade-off is spectral fidelity.** Pansharpened images look better, but the fusion process alters the original spectral values. Running NDVI or other quantitative analyses on pansharpened data can produce unreliable results.
- **Most satellite imagery providers charge extra for pansharpening** as a processing add-on, typically 30-80% on top of the base imagery price. Geopera includes it in every order at no additional cost.
- **You should almost always request pansharpened imagery** for visual interpretation, mapping, and feature extraction. Skip it only when you need the raw spectral values for quantitative analysis.

Why do satellites still take black-and-white photos? It's 2026. We've got phones that shoot 200-megapixel colour photos. Surely a billion-dollar satellite can manage colour at full resolution.

It can't. And the reason is physics, not budget.

A satellite sensor has to make a choice. It can collect all visible light into one wide band and produce a sharp, detailed greyscale image. Or it can split that light into separate colour channels and get a colour image, but at lower resolution because each channel gets less light. You can't have full spatial detail and full colour at the same time. Not from a single sensor pass.

**Pansharpening** is the workaround. It takes the sharp black-and-white image (the **panchromatic** band) and the blurry colour image (the **multispectral** bands) and fuses them into one product: a sharp colour image at the panchromatic resolution. If you've ever looked at commercial satellite imagery and thought "that's impressively detailed for a colour photo taken from 600km up," you were probably looking at a pansharpened image.

<AskBand headline="Would you trust it at full zoom?" sub="Judge pansharpened output over your own area; sample applications are open." action="Apply for a sample" href="/trust" tiles={['/band/city.webp', '/band/port.webp', '/band/pivots.webp']} />

## What is a Panchromatic Image?

A **panchromatic image** is a single-band greyscale image that captures light across a wide range of the visible spectrum, roughly 450-900nm. "Pan" means all, "chromatic" means colour. The sensor collects all visible wavelengths into one bucket instead of separating them into red, green, and blue.

The advantage is simple: more photons per pixel. Because the sensor isn't splitting light into separate channels, each detector element receives more energy, which means you can make the pixels smaller while still getting a clean signal. That's why panchromatic bands consistently deliver the highest spatial resolution on any given satellite.

On [WorldView-3](/sensors/worldview-3), the panchromatic band captures at 0.31m resolution. The multispectral bands on the same satellite? 1.24m. Four times coarser. Same satellite, same orbit, same moment in time. The only difference is how the sensor divides the incoming light.

This 4:1 ratio isn't arbitrary. It shows up across nearly every commercial optical satellite:

| Satellite             | Panchromatic Resolution | Multispectral Resolution | Ratio |
| --------------------- | ----------------------- | ------------------------ | ----- |
| WorldView-3 (Vantor)  | 0.31m                   | 1.24m                    | 4:1   |
| Beijing-3 (21AT)      | 0.5m                    | 2.0m                     | 4:1   |
| SuperView (SpaceWill) | 0.5m                    | 2.0m                     | 4:1   |
| Perascope (Geopera)   | 0.75m                   | 3.0m                     | 4:1   |
| WorldView-2 (Vantor)  | 0.46m                   | 1.84m                    | 4:1   |
| KOMPSAT-3A (KARI)     | 0.55m                   | 2.2m                     | 4:1   |
| Pleiades Neo (Airbus) | 0.3m                    | 1.2m                     | 4:1   |

The 4:1 ratio is baked into the sensor hardware. Each panchromatic detector element is one quarter the size of each multispectral detector element, collecting the same total amount of light because it doesn't need to split it by wavelength.

## How Pansharpening Works

The concept is simple. You have two images of the same area captured at the same time:

1. **Panchromatic**: sharp spatial detail, no colour information
2. **Multispectral**: colour information, fuzzy spatial detail

Pansharpening extracts the spatial detail from the panchromatic band and injects it into the multispectral bands, producing colour imagery at the panchromatic resolution.

<figure>

![Pansharpening process showing panchromatic plus multispectral equals sharp colour pansharpened satellite image](/blogs/pansharpening-before-after-diagram.webp)

<figcaption>Pansharpening combines the spatial detail of a panchromatic band with the colour information from multispectral bands. The result is a colour image at the higher panchromatic resolution.</figcaption>

</figure>

The actual fusion happens through one of several algorithms. The most common ones in commercial satellite processing:

**IHS (Intensity-Hue-Saturation)** transforms the multispectral bands from RGB colour space into IHS colour space, replaces the Intensity component with the panchromatic band, then transforms back. Fast and produces visually sharp results, but tends to distort colours.

**Brovey Transform** normalises each multispectral band by the sum of all bands, then multiplies by the panchromatic band. Computationally simple. Good colour preservation for natural colour composites, but again shifts spectral values.

**Gram-Schmidt** simulates a lower-resolution panchromatic band from the multispectral data, then uses the difference between the simulated and real panchromatic bands to sharpen the multispectral image. Generally produces the best balance of spatial sharpness and colour fidelity, as [NASA's Earth Observatory has documented](https://earthobservatory.nasa.gov/blogs/earthmatters/2017/06/13/how-to-pan-sharpen-landsat-imagery/). This is the default in most commercial processing pipelines, including ours at Geopera.

**HPF (High-Pass Filter)** extracts high-frequency spatial detail from the panchromatic band using a filter, then adds that detail to each upsampled multispectral band. Simple and effective, with reasonable spectral preservation.

No algorithm is perfect. They all involve a compromise between spatial sharpness and spectral accuracy. But for the majority of commercial use cases, the visual improvement massively outweighs the spectral trade-off.

## The Spectral Integrity Trade-off

This is the part that matters if you're doing anything quantitative with satellite data.

**Pansharpening changes pixel values.** The fusion process alters the original digital numbers recorded by the multispectral sensor. The degree of alteration depends on the algorithm, the scene content, and how well the panchromatic band's spectral range overlaps with the multispectral bands.

What does that mean in practice? If you calculate [NDVI from pansharpened imagery](/blog/remote-sensing-vegetation-indices), the values won't exactly match NDVI calculated from the original multispectral bands. For some algorithms, the difference is small. For others, it can be significant enough to misclassify vegetation health categories.

**Rule of thumb:**

- **Visual interpretation, mapping, digitising features, presentations** = use pansharpened. The improved spatial detail makes everything easier to see and interpret.
- **Quantitative spectral analysis (NDVI, NDWI, classification, change detection)** = use the original multispectral bands. Don't sacrifice spectral accuracy for pixels you don't need.
- **Mixed workflows** = request both. Use the pansharpened product for visual reference and feature identification, then run your spectral calculations on the native multispectral data.

Most commercial providers deliver pansharpened imagery as the default product. If you need the raw multispectral bands for analysis, you'll typically need to specify that when ordering. At Geopera, we can deliver either or both, depending on your workflow.

## When Pansharpening Adds Real Value

Not every project needs pansharpened imagery. But most do. Here's when the 4x resolution boost makes a genuine difference.

**Urban mapping and infrastructure monitoring.** At 1.2m multispectral resolution, you can see buildings but not their outlines. At 0.3m pansharpened, you can trace individual building footprints, identify road markings, and spot vehicles. The difference between "there's a structure" and "that's a two-storey commercial building with rooftop solar panels."

**Defence and security applications.** Object identification demands the highest possible spatial resolution combined with colour. Pansharpened WorldView-3 imagery at 0.31m is the standard for intelligence-grade analysis.

**Insurance and damage assessment.** After natural disasters, adjusters need colour imagery sharp enough to assess roof damage, identify debris fields, and estimate structural impacts. Pansharpened imagery delivers that level of detail.

**Mining site monitoring.** Tracking stockpile volumes, road conditions, and equipment locations requires colour imagery at sub-metre resolution. Running [NDVI for rehabilitation monitoring](/blog/remote-sensing-vegetation-indices) should use the native multispectral bands, but pansharpened products handle everything else.

**Agriculture at field scale.** Identifying individual pivot irrigators, checking fence lines, mapping waterways. For crop health analysis, stick with native multispectral. For everything else, pansharpened wins.

## When to Skip Pansharpening

There are valid reasons to work with the native multispectral bands instead.

**Spectral index calculations.** If your primary deliverable is an NDVI map, an NDWI analysis, or any index that relies on precise spectral values, use the original [multispectral data](/blog/multispectral-vs-hyperspectral-satellite-imagery). The spectral distortion from pansharpening, even with Gram-Schmidt, introduces noise into your results.

**Land cover classification.** Supervised and unsupervised classification algorithms train on spectral signatures. Altered signatures from pansharpening can reduce classification accuracy, sometimes by several percentage points. Train and classify on native multispectral, then overlay results on pansharpened imagery for presentation if needed.

**Time-series analysis.** If you're comparing imagery across multiple dates to detect change, consistency matters more than resolution. Using pansharpened imagery from different dates, potentially processed with different algorithms or parameters, introduces variables that have nothing to do with actual ground change.

**Large-area mapping at coarse resolution.** If you're mapping land cover across a 10,000 km² region using Sentinel-2 at 10m resolution, pansharpening won't help. Sentinel-2 doesn't even have a panchromatic band.

## What This Means When You're Buying Satellite Imagery

Here's the thing most buyers don't realise until after they've placed an order: **pansharpening is usually a paid add-on.**

Most satellite imagery providers sell you the raw or semi-processed data, then charge extra for processing steps like [orthorectification, atmospheric correction, and pansharpening](/blog/satellite-imagery-cost-guide). These fees can add 30-80% to the base imagery cost. You order a 50 km² area at $12/km², expecting to pay $600, and the invoice comes back at $900-$1,080 because processing was extra.

We think that's backwards. At Geopera, pansharpening is included in every order as part of our standard processing pipeline. So is orthorectification and atmospheric correction. You get analysis-ready, pansharpened colour imagery at the full panchromatic resolution without paying more for it. That's how [buying satellite imagery](/blog/how-to-buy-satellite-imagery) should work.

If you need the native multispectral bands alongside the pansharpened product, we deliver those too. Same order, no surcharge. See how our [satellite imagery processing pipeline](/imagery) handles pansharpening, orthorectification, and atmospheric correction as standard on every order.

## Frequently Asked Questions

### What is pansharpening in satellite imagery?

**Pansharpening is an image fusion technique** that merges a high-resolution panchromatic (greyscale) satellite band with lower-resolution multispectral (colour) bands to create a single colour image at the higher panchromatic resolution. The process typically improves spatial resolution by 4x while preserving the colour information from the multispectral bands.

### What is a panchromatic image?

**A panchromatic image is a single-band greyscale image** that captures all visible wavelengths (roughly 450-900nm) into one channel. Because the sensor collects more total light per pixel compared to splitting it into colour channels, panchromatic bands achieve the highest spatial resolution available from any given satellite. For example, WorldView-3 captures 0.31m panchromatic vs 1.24m multispectral.

### Does pansharpening affect spectral accuracy?

Yes. **All pansharpening algorithms alter the original spectral values** to some degree. For visual interpretation and mapping, this is acceptable. For quantitative analysis like NDVI calculation, land cover classification, or change detection, you should use the original multispectral bands to preserve spectral fidelity. The Gram-Schmidt method generally produces the least spectral distortion.

### Which pansharpening algorithm is best?

**Gram-Schmidt spectral sharpening** is considered the best general-purpose algorithm, offering the strongest balance between spatial sharpness and spectral preservation. IHS fusion produces sharp results but distorts colours more. Brovey Transform works well for visual products. The optimal choice depends on whether your priority is visual quality or spectral accuracy.

### Can you pansharpen Sentinel-2 data?

**Sentinel-2 does not have a panchromatic band**, so traditional pansharpening isn't possible. However, Sentinel-2's 10m visible bands (B2, B3, B4, B8) can be used to sharpen the 20m bands (red edge, SWIR) using similar fusion techniques, as described in the [ESA Sentinel-2 technical guide](https://sentinels.copernicus.eu/web/sentinel/user-guides/sentinel-2-msi). Some researchers also apply super-resolution methods to enhance Sentinel-2 spatial detail, though these aren't true pansharpening.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# What is SAR Satellite Imagery? How Radar Sees What Optical Can't

> SAR satellites use radar to image Earth through clouds, at night, in any weather. Learn how synthetic aperture radar works and which SAR satellites to use.

Published: 2026-03-17 | Author: Darcy Weedman | Reading time: 14 min

Source: https://geopera.com/blog/sar-satellite-imagery-explained

---
<script>
	import { BarChart, ComparisonScale, SpectrumBar, Callout } from '$lib/components/charts';
</script>

## Summary

- **Synthetic aperture radar (SAR) satellites image Earth using radar pulses, not sunlight.** They see through clouds, work at night, and don't care about weather. Optical satellites can't do any of that.
- **Two free SAR data sources now exist**: ESA's Sentinel-1C (C-band, 5-40m) and NASA/ISRO's NISAR (L-band + S-band, 5-10m), which started releasing data in early 2026.
- **Commercial SAR resolution has hit 16cm** from ICEYE and Umbra. That rivals the best optical satellites.
- **SAR captures completely different information than optical** — surface roughness, moisture, ground deformation — and works best when combined with optical data.
- **The wavelength matters more than you'd think.** X-band sees the surface. L-band sees through entire forest canopies. P-band goes into the soil.

Last month, Planet Labs told its customers they'd have to wait 14 days for optical satellite imagery over the Middle East. Fourteen days. The US had just launched strikes on Iranian targets, and apparently nobody wanted commercial satellite photos showing the aftermath circulating on Twitter.

SAR imagery from other providers? Available as normal. No restrictions.

That one detail tells you almost everything about the difference between optical and synthetic aperture radar satellite imagery. Optical photos are intuitive — anyone can look at them and understand what they're seeing, which is exactly what makes them politically sensitive. SAR images look like grainy TV static to untrained eyes, even though they contain a staggering amount of information.

We deal with [optical imagery](/blog/understanding-optical-satellite-imagery) every day at Geopera. We process thousands of orders from providers like [Maxar/Vantor](/blog/vantor-lanteris-maxar-rebrand) and 21AT, plus our own Perascope constellation. But the question we keep getting is: "What about radar?" So let's talk about it.

## The Embarrassingly Simple Idea Behind SAR

Here's the concept: shout at a wall and listen to the echo.

That's it. That's synthetic aperture radar. A satellite fires microwave pulses at the ground and records what bounces back. Different surfaces bounce radar differently — buildings reflect it straight back (loud echo), water reflects it away like a mirror (silence), trees scatter it in every direction (messy echo).

The "synthetic aperture" part is the trick that makes it actually useful. To get a sharp radar image from space, you'd need an antenna hundreds of metres long. Nobody's launching that. Instead, SAR exploits the fact that the satellite is moving. As it flies along its orbit, it fires pulses and records echoes from thousands of slightly different positions. Then software stitches all those returns together to simulate the effect of one enormous antenna.

Think of it like dragging a microphone across a room while someone claps — you hear the clap from many positions, and your brain combines them to pinpoint exactly where the sound came from. SAR does this with radar at 7km/s from 500km up.

![How SAR satellites transmit radar pulses through clouds to image different surface types on Earth](/blogs/sar-satellite-explained.webp)

The result: detailed imagery from an antenna that's actually only a few metres wide, taken through clouds, at midnight, during a monsoon. Optical satellites are sitting this one out completely.

<Callout type="fun-fact" title="Why SAR images look weird">
SAR images are grayscale. No colour. No pretty RGB composites. You can't calculate [NDVI or any of the 350+ spectral indices](https://docs.geopera.com) that make optical imagery so powerful for vegetation analysis. What you get instead is information about surface texture, moisture, and structure — things that are invisible in a photograph.
</Callout>

## Not All Radar is Created Equal

Here's where it gets fun. SAR satellites don't all use the same type of radar. The wavelength they transmit determines what they can see and — this is the critical part — how deep they can look.

<SpectrumBar
title="SAR Frequency Bands — What Each Wavelength Penetrates"
segments={[
{ label: 'X-band', sublabel: '~3 cm', width: 1, color: '#dc2626' },
{ label: 'C-band', sublabel: '~5.6 cm', width: 1.5, color: '#ea580c' },
{ label: 'S-band', sublabel: '~9.4 cm', width: 2, color: '#ca8a04' },
{ label: 'L-band', sublabel: '~23 cm', width: 3, color: '#059669' },
{ label: 'P-band', sublabel: '~69 cm', width: 4, color: '#1d4ed8' },
]}
leftLabel="Surface only"
rightLabel="Penetrates deep into soil & vegetation"
/>

Short wavelengths (X-band, ~3cm) bounce off the very top of whatever they hit. Leaves, rooftops, the surface of water. Great for seeing fine detail, terrible for seeing beneath anything.

Long wavelengths (L-band, ~23cm) pass right through leaves, branches, and light vegetation to interact with trunks and the ground below. Imagine the difference between throwing a tennis ball at a chain-link fence versus a basketball — the basketball goes through gaps the tennis ball can't.

P-band (~69cm) goes even further, penetrating into dry soil and dense vegetation. ESA's upcoming [BIOMASS satellite](/blog/esa-biomass-explained) uses P-band specifically because it can measure forest biomass by interacting with the largest branches and trunks that contain most of the carbon.

Why does this matter practically?

If you're monitoring deforestation in the Amazon — where cloud cover blocks optical satellites for half the year — you need L-band SAR that cuts through both the clouds AND the canopy. C-band won't get through dense tropical forest. X-band won't even get through the leaves.

But if you're tracking ship movements in open water? X-band gives you the sharpest, crispest images at the highest resolution. Horses for courses.

## The SAR Satellites You Can Actually Use in 2026

Five years ago, your SAR options were basically Sentinel-1 or writing a very large cheque to a government agency. That's changed dramatically.

### The Free Stuff

**Sentinel-1C** (ESA) launched December 2024 after Sentinel-1B died unexpectedly in 2022 (just... stopped working, like a printer on a Monday morning, except this one cost €300 million). C-band, 5-40m resolution, 12-day revisit, and the [entire archive is free](/blog/free-sources-of-satellite-data) through the Copernicus Data Space.

**NISAR** (NASA/ISRO) is the big one. Launched July 2025 with both L-band and S-band instruments. It maps every landmass on Earth every 12 days at 5-10m resolution and generates 85 terabytes of data per day. Over 100,000 data products hit the [Alaska Satellite Facility](https://asf.alaska.edu/) in February 2026, with fully calibrated global data coming mid-2026. All free. All open access. L-band SAR data at this scale has simply never existed before.

### The Commercial Stuff

<BarChart
title="Best Resolution by Commercial SAR Provider (cm)"
data={[
{ label: 'ICEYE', value: 16, suffix: ' cm' },
{ label: 'Umbra', value: 16, suffix: ' cm' },
{ label: 'Synspective StriX', value: 25, suffix: ' cm' },
{ label: 'Capella Space', value: 31, suffix: ' cm' },
{ label: 'TerraSAR-X', value: 100, suffix: ' cm', color: '#6b7280' },
{ label: 'RADARSAT', value: 300, suffix: ' cm', color: '#6b7280' },
]}
maxValue={350}
barColor="#059669"
/>

A few things jump out from that chart.

ICEYE (Finland) and Umbra (US) both deliver **16cm resolution** from commercial constellations. Each satellite weighs under 100kg. That's comparable to what Maxar achieves with optical at 15cm — except the radar version works through clouds and at night.

Synspective (Japan) is launching its 8th StriX satellite this month and building toward a 30-satellite constellation. Capella Space offers 31cm, still sub-metre, still impressive. Both are X-band.

The older systems — TerraSAR-X (Airbus, launched 2007) and RADARSAT (Canadian Space Agency, 2019) — are lower resolution but cover enormous swaths. Different tools for different jobs.

In Germany, you might search for "SAR-Satelliten" and land on different results than English-language markets. The European SAR ecosystem runs deep: Airbus with TerraSAR-X, ESA with Sentinel-1 and BIOMASS, and a growing constellation of startups like constellr and SatVu in the thermal-adjacent space.

<Callout type="info" title="Quick resolution gut-check">
At 16cm resolution, you can identify the make of a car. At 3m resolution, you can tell there's a car. At 40m resolution (free Sentinel-1 data), you can tell there's a parking lot. Different projects need different levels of detail — and different budgets.
</Callout>

## SAR vs Optical: A Practical Decision Guide

Most guides present this as a contest. SAR vs optical, pick a side. That completely misses the point. They capture fundamentally different information about Earth's surface. The question isn't "which is better" — it's "what do I actually need to see?"

<ComparisonScale
title="Where SAR and Optical Each Excel (1 = weak, 10 = strong)"
items={[
{ label: 'SAR — clouds', position: 10, color: '#059669' },
{ label: 'Optical — clouds', position: 1, color: '#dc2626' },
{ label: 'SAR — colour info', position: 1, color: '#059669' },
{ label: 'Optical — colour info', position: 10, color: '#dc2626' },
{ label: 'SAR — night imaging', position: 10, color: '#059669' },
{ label: 'Optical — night imaging', position: 1, color: '#dc2626' },
{ label: 'SAR — deformation', position: 10, color: '#059669' },
{ label: 'Optical — deformation', position: 1, color: '#dc2626' },
{ label: 'SAR — ease of use', position: 3, color: '#059669' },
{ label: 'Optical — ease of use', position: 9, color: '#dc2626' },
]}
min={1}
max={10}
leftLabel="Weak"
rightLabel="Strong"
/>

See the pattern? SAR dominates anything weather-related, time-related, or deformation-related. Optical dominates anything colour-related, spectral-related, or interpretability-related.

Neither replaces the other. They're doing different things entirely.

**Reach for SAR when:**

- Cloud cover makes optical unreliable (tropical regions, monsoon seasons, northern Europe, the UK on any given Tuesday)
- You need imagery regardless of daylight
- You're measuring ground subsidence or structural deformation with InSAR (millimetre precision, which is honestly absurd from space)
- You're finding ships, oil spills, or sea ice
- You need to see beneath forest canopy

**Reach for optical when:**

- You need colour or spectral analysis (vegetation health, water quality, mineral ID)
- Non-technical people need to understand the imagery (boardroom presentations, insurance reports, planning applications)
- You're calculating vegetation indices like NDVI or EVI
- You need the absolute finest resolution (Maxar at 15cm still edges out commercial SAR)
- Budget is tight (optical archive imagery is generally cheaper per square kilometre)

**Use both when the stakes are high.** For mining companies monitoring tailings dams, combining optical spectral analysis with SAR-based InSAR deformation tracking provides something neither technology delivers alone. For [agricultural monitoring](/agriculture) in cloudy regions, SAR fills the gaps between clear-sky optical acquisitions. For disaster response, you can't afford to wait for clear skies.

## What SAR Can't Do (The Honest Part)

We'd be doing you a disservice if we didn't mention the limitations. SAR is powerful, but it's not magic.

**Interpretation is hard.** SAR images look nothing like photographs. Bright doesn't mean "light-coloured" — it means "rough surface that reflected radar back at the satellite." Water appears jet black. Buildings glow. Trees are speckly grey. It takes training to read SAR imagery, and most people outside the remote sensing world find it confusing at first.

**Speckle noise is real.** Every SAR image has a granular salt-and-pepper texture that makes fine details harder to see. It's inherent to how coherent radar works — not a processing error. Filtering reduces it but softens detail.

**No spectral information.** You can't calculate NDVI from SAR data. You can't tell the difference between a red roof and a blue roof. If colour matters, you need optical.

**Geometric distortions near mountains.** Tall features "lean" toward the satellite in SAR images (called layover). Mountains cast radar shadows. Slopes facing the satellite get compressed (foreshortening). Processing handles some of this, but steep terrain is always harder.

**It won't see through solid rock.** Despite what movies suggest, SAR doesn't see underground in any meaningful way. L-band penetrates a metre or two into dry sand. P-band reaches a few metres into dry soil. But wet ground, rock, concrete, metal? Nothing gets through.

## Where This Fits With What We Do

We process [optical satellite imagery](/blog/why-we-process-every-order) at Geopera — [orthorectification](/blog/orthorectification-explained), pansharpening, atmospheric correction, spectral index calculation. That's our core. Every order gets the full treatment at no extra cost, because we think unprocessed imagery isn't worth delivering.

The SAR question comes up more and more. And honestly, the answer is usually "you need both." Optical gives you the spectral richness. SAR gives you the all-weather reliability. Together they get you something closer to continuous situational awareness than either manages alone.

If you're figuring out what mix of optical and SAR fits your project — or you just need properly processed satellite imagery from [any of the major providers](/blog/how-to-buy-satellite-imagery) — [talk to our team](/contact). We'll help you work it out without overselling you something you don't need.

Or poke around the [Pera Portal](https://portal.geopera.com) yourself.

## Frequently Asked Questions

**What is synthetic aperture radar used for?**

SAR is used for flood mapping through clouds, ground deformation monitoring in mining and infrastructure (via InSAR, with millimetre precision), crop monitoring during cloudy seasons, deforestation detection beneath forest canopy, maritime surveillance, and defence. The key advantage is that SAR works in any weather and at any time of day.

**How does synthetic aperture radar work?**

A SAR satellite transmits microwave pulses at Earth and records the reflected energy. As it moves along its orbit, it collects returns from thousands of positions, then computationally combines them to simulate a much larger antenna — the "synthetic aperture." This creates high-resolution images from a physically small antenna, through clouds and darkness.

**Can synthetic aperture radar see underground?**

In very limited cases. L-band SAR (23cm wavelength) penetrates 1-2 metres into dry sand and sees through forest canopy to the ground. P-band (69cm) reaches a few metres into dry soil. But SAR can't see through wet ground, rock, concrete, or metal. The "underground vision" capability depends entirely on the surface material and how dry it is.

**How deep can synthetic aperture radar penetrate?**

It depends on the band and the material. X-band (~3cm) stays at the surface. C-band (~5.6cm) pushes through light vegetation. L-band (~23cm) gets through forest canopy and 1-2m into dry sand. P-band (~69cm) can reach several metres into dry soil. Water kills penetration at every frequency — even L-band barely gets past wet ground.

**Is SAR satellite imagery free?**

Some of it. ESA's Sentinel-1C provides free C-band SAR globally at 5-40m resolution. NASA/ISRO's NISAR started releasing free L-band and S-band data at 5-10m in 2026. Umbra offers select free imagery through its open data programme. For sub-metre commercial SAR from ICEYE, Capella, Umbra, or Synspective, you'll need to purchase — pricing depends on resolution, coverage, and licensing.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# How Much Does Satellite Imagery Cost in 2026?

> Real satellite imagery prices per km2 by resolution tier. Archive vs tasking costs, hidden processing fees, and what you actually end up paying.

Published: 2026-03-17 | Author: Darcy Weedman | Reading time: 11 min

Source: https://geopera.com/blog/satellite-imagery-cost-guide

---
<script>
import { AskBand, BarChart, Callout, ComparisonScale } from '$lib/components/charts';
</script>

## Summary

- **Commercial satellite imagery ranges from ~$2/km2 to $55+/km2** depending on resolution, whether you're buying archive or tasking new captures, and your licensing needs.
- **Archive imagery costs 40-60% less than new tasking.** If someone already photographed your area, you save substantially.
- **The brochure price isn't what you'll actually pay.** Processing fees add 30-80% at most providers. Minimum order areas mean you might buy 25km2 when you only need 2km2. Multi-user licensing adds another 30%.
- **Free satellite imagery exists** at lower resolutions: Sentinel-2 at 10m (ESA), Landsat at 30m (USGS), and now NISAR at 5-10m for [SAR applications](/blog/sar-satellite-imagery-explained).
- **Some providers include processing in their price. Most don't.** This single factor can swing your total cost by thousands of dollars.

Try finding satellite imagery pricing online. Go ahead. Pick any major provider's website and look for actual numbers.

You'll find "Contact Sales." Or "Request a Demo." Or "Get a Quote." Maybe a "Starting at..." followed by no actual detail.

This is an industry that sells data products at scale, and almost nobody publishes what they charge. They want you on a sales call before you know whether you're looking at $5/km2 or $50/km2. For a technology business in 2026, the pricing opacity is genuinely wild.

We publish [our pricing](/pricing). Openly. With actual numbers. Because we think you should know what things cost before committing to a conversation. So here's everything we know about what satellite imagery costs across the industry — not just from us.

<AskBand headline="What would your area cost?" sub="The rate card is published per km²; no quote round-trip required." action="See the rate card" href="/pricing" tiles={['/band/pivots.webp', '/band/port.webp', '/band/mine.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## Resolution Is the Biggest Cost Driver

Two factors determine roughly 80% of your satellite imagery cost. Resolution is the first, and it's the big one.

The relationship isn't linear. Going from 3m to 1m resolution roughly doubles the price. Going from 1m to 50cm doubles it again. And the jump from 50cm to 30cm? Nearly doubles once more.

<BarChart
title="Typical Archive Price by Resolution (USD per km2)"
data={[
{ label: '30cm', value: 22, suffix: '-55', color: '#dc2626' },
{ label: '40-50cm', value: 8, suffix: '-18' },
{ label: '1m', value: 4, suffix: '-10', color: '#059669' },
{ label: '1.5m', value: 2, suffix: '-6', color: '#059669' },
{ label: '3-5m', value: 1, suffix: '-5', color: '#059669' },
{ label: '10m+', value: 0, suffix: ' Free', color: '#6b7280' },
]}
maxValue={60}
barColor="#2563eb"
/>

At 30cm, you can identify the make of a car in a parking lot. At 3m, you can tell there's a parking lot. At 10m, you can tell there's a commercial area. Different projects need different levels of detail, and paying for 30cm when 1m would do is like hiring a helicopter when a drone would get the job done.

The common mistake? Defaulting to the highest resolution available "just in case." We see this constantly. Someone orders 30cm imagery for a vegetation monitoring project where 50cm or even Sentinel-2's free 10m data would give them everything they need.

## Archive vs Tasking: The 50-100% Price Gap

The second big factor. And it catches more people off guard than it should.

**Archive imagery** already exists. At some point in the past, a satellite flew over your area and captured an image. You're buying access to that existing data. Delivery is often same-day.

**Tasking** means you're asking a satellite operator to point their camera at your specific location on a future date. You're essentially booking a photoshoot from space.

The price difference is dramatic:

|                       | Archive       | Tasking       |
| --------------------- | ------------- | ------------- |
| Typical cost (50cm)   | $8-18/km2     | $15-35/km2    |
| Typical cost (30cm)   | $17-26/km2    | $25-55/km2    |
| Minimum order area    | 25 km2        | 50-100 km2    |
| Delivery time         | Hours to days | Days to weeks |
| You control the date? | No            | Yes           |

And within tasking, there's often a further split between "flexible" and "assured." Flexible means the operator will try to capture your area but makes no guarantee — if clouds roll in or a higher-priority customer needs the slot, you wait. Assured means they commit to delivering, and it can cost 3-4x more than flexible.

<Callout type="tldr" title="Quick rule of thumb">
If archive imagery exists for your area with acceptable cloud cover and recency, always check archive first. You'll typically pay half the price, with smaller minimums, and get it the same day.
</Callout>

## The Hidden Costs Nobody Warns You About

Here's where the published per-km2 rates get misleading. The price you see on a rate card or quote is often just the starting point.

### Processing fees: the 30-80% surprise

Raw satellite imagery needs [orthorectification](/blog/orthorectification-explained), pansharpening, atmospheric correction, and potentially mosaicking before you can actually use it for analysis. Most providers charge separately for this. The processing markup typically adds 30% to 80% on top of the base imagery price.

So that $15/km2 quote for 50cm imagery? After processing, you're looking at $20-27/km2. On a 100km2 order, that's an extra $500-$1,200 nobody mentioned during the sales call.

<Callout type="warning" title="This is the single biggest hidden cost in the industry">
Always ask: "Does that price include orthorectification, pansharpening, and atmospheric correction?" If the answer is no, budget an additional 30-80% on top. Or use a provider that includes processing — like <a href="/pricing">we do</a>.
</Callout>

### Minimum order areas: paying for what you don't need

You can't buy 1km2 of high-resolution imagery from most providers. Minimum order areas are typically 25km2 for archive and 50-100km2 for tasking. Some go as high as 500km2 for medium-resolution tasking.

What that means in practice: if you need imagery of a 5km2 mine site, you're paying for 25km2 at minimum. At $15/km2, that's $375 instead of $75. A 5x overpurchase that nobody flags at quote stage.

### Licensing: the 30% you didn't budget for

The base price assumes a single user accessing the data for internal purposes. Need two people to look at it? That's usually a 30% uplift. Government use? Additional uplift. Want to include imagery in a published report? Different licensing tier entirely.

Most buyers don't think about licensing until the contract shows up. By then, the budget's already approved at the "base" price.

### Cloud cover specs: paying more for clear skies

Standard pricing usually assumes 10-15% cloud cover tolerance. Want guaranteed less than 5% cloud cover? Expect a 25% uplift. Cloud-free? Even more. If you're working in tropical regions where clear skies are rare, these uplifts add up fast.

## A Real-World Cost Breakdown

Let's walk through a real example. You need 30cm resolution, pansharpened imagery of a 50km2 mining concession. Archive from the last 6 months, less than 10% cloud cover.

**Scenario A: Traditional provider with separate processing**

- Base imagery: 50km2 x $22/km2 = $1,100
- Processing (ortho + pansharp + atmos): ~50% uplift = $550
- Standard licensing (single user): included
- **Total: ~$1,650**

**Scenario B: Provider with lower-res but cheaper option (50cm)**

- Base imagery: 50km2 x $8/km2 = $400
- Processing: ~50% uplift = $200
- **Total: ~$600**

**Scenario C: Provider that includes processing (like Geopera)**

- All-in price: 50km2 x per-km2 rate from our [pricing page](/pricing)
- Processing: included. $0 extra.
- **Total: Just the imagery cost. No surprises.**

The gap between Scenario A and Scenario B is nearly 3x, for imagery that's genuinely good enough for most monitoring applications. And the gap between any traditional provider and one that includes processing can save hundreds to thousands depending on order size.

Whether the jump from 50cm to 30cm justifies $1,000+ depends entirely on your use case. For [mining rehabilitation monitoring](/blog/mining-satellite-imagery), 50cm is usually plenty. For counting individual structures in an urban damage assessment, you probably need 30cm.

## When Free Imagery Is Good Enough

Not every project needs commercial data. We say this even though we sell the stuff, because recommending expensive imagery for a project that Sentinel-2 could handle would be dishonest.

| Source             | Resolution | Revisit | Cost | Best For                                                              |
| ------------------ | ---------- | ------- | ---- | --------------------------------------------------------------------- |
| Sentinel-2 (ESA)   | 10m        | 5 days  | Free | Agriculture, land use, [vegetation indices](https://docs.geopera.com) |
| Landsat 8/9 (USGS) | 30m        | 16 days | Free | Long-term change detection, environmental                             |
| Sentinel-1 (ESA)   | 5-40m      | 12 days | Free | [All-weather SAR monitoring](/blog/sar-satellite-imagery-explained)   |
| NISAR (NASA/ISRO)  | 5-10m      | 12 days | Free | Ground deformation, forestry (new in 2026)                            |

We've written [free satellite data guides by country](/blog/free-sources-of-satellite-data) covering where to access all of this.

The honest take: if you're monitoring crop health across a large agricultural area, Sentinel-2 at 10m is excellent and costs nothing. If you need broad environmental change detection over decades, Landsat's 40+ year archive is unbeatable. But if you need to identify individual trees, assess building-level damage, or read the text on a shipping container, you need commercial resolution.

## How to Not Overpay

After processing thousands of imagery orders, here's what we've learned:

**Check archive first.** Always. The 50-100% price premium for tasking only makes sense when you genuinely need a specific capture date. For most projects, archive imagery from the last few months works perfectly.

**Match resolution to your actual need.** The most common waste we see is people buying 30cm imagery for projects where 50cm or even 1m would deliver identical analytical outcomes. Ask yourself: "What's the smallest thing I need to identify?" and pick the resolution that just covers it.

**Ask about processing before you commit.** If a provider quotes base imagery cost without mentioning processing, get the all-in number. Then compare all-in prices across providers, not headline rates.

**Watch minimum order areas.** If you need 5km2, find a provider with minimums that don't force you to buy 100km2. Some modern platforms let you order exactly the area you need.

**Consider non-Western satellite providers.** Several Asian constellations now offer 50cm archive at $6-8/km2, undercutting traditional Western providers by 40-50%. The quality has improved enormously in recent years, and at Geopera we offer imagery from multiple providers through our [Pera Portal](https://portal.geopera.com) so you can compare options.

For transparent, all-in pricing with processing included, check our [pricing page](/pricing). For available imagery over your specific area, browse the [Pera Portal](https://portal.geopera.com). And if you'd rather just tell someone what you need, [talk to our team](/contact).

## Frequently Asked Questions

**How much does satellite imagery cost per square kilometre?**

Commercial satellite imagery ranges from about $2/km2 (1.5m resolution, archive) to $55+/km2 (30cm stereo, new tasking). The most commonly purchased tier — 50cm archive — typically falls between $8-18/km2 depending on provider. Free options exist at 10m resolution (Sentinel-2) and 30m (Landsat).

**Why do most satellite imagery providers hide their pricing?**

The industry traditionally uses enterprise sales models where pricing is negotiated per deal. Providers argue that pricing "depends on requirements," but the real reason is that opacity enables higher margins. A few modern platforms now publish transparent pricing. We're one of them — our rates are on our [pricing page](/pricing).

**Does satellite imagery processing cost extra?**

At most traditional providers, yes. Orthorectification, pansharpening, and atmospheric correction typically add 30-80% to the base imagery cost. Some providers include all processing at no extra charge — at Geopera, every order ships fully processed and analysis-ready. Always ask whether the quoted price includes processing.

**What is the minimum order for satellite imagery?**

Minimum order areas vary. Most providers require 25km2 for archive orders and 50-100km2 for tasking. Some medium-resolution products have 500km2 minimums. A few modern platforms have reduced or eliminated minimums, letting you order exactly the coverage you need without paying for excess area.

**Is free satellite imagery good enough for commercial projects?**

For many applications, yes. ESA's Sentinel-2 provides free 10m resolution imagery with 5-day revisit, which is sufficient for regional agriculture monitoring, land use classification, and broad environmental tracking. USGS Landsat offers free 30m data going back to 1972. For projects requiring sub-metre detail — infrastructure inspection, urban mapping, property-level analysis — commercial imagery is necessary.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# That Time Everyone Thought the Three Gorges Dam Was Collapsing Due to an Orthorectification Error

> In 2019, satellite imagery showing distortions in China's Three Gorges Dam sparked global panic—but the dam was fine. The orthorectification wasn't.

Published: 2025-11-19 | Author: Darcy Weedman | Reading time: 15 min

Source: https://geopera.com/blog/three-gorges-dam-orthorectification-error

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

The Three Gorges Dam is not deforming. The wavy structure in the viral 2019 Google Earth images was a processing artifact: orthorectification with elevation data that didn't account for the 185-metre structure. Monitoring data released by the dam's operator in July 2019 put horizontal crest movement at 1.4 to 26.7 mm, elastic and within design limits.

In 2019, satellite images showing severe geometric distortion in China's Three Gorges Dam went viral, sparking fears of structural failure and calls for mass evacuation. But the dam's structural integrity was sound—the distortion existed only in the imagery processing. Inadequate orthorectification using insufficient elevation data for the 185-meter artificial structure created geometric artifacts that resembled structural deformation. This incident demonstrated that when monitoring critical infrastructure, the quality of image preprocessing directly affects whether you're analyzing reality or processing artifacts. The lesson is clear: professional applications require professional-grade orthorectification with proper validation.

At Geopera, we deliver satellite imagery processed to professional standards—with comprehensive orthorectification, validated accuracy metrics, and complete documentation included at no extra cost—so you can trust that what you're seeing represents reality, not processing artifacts.

<AskBand headline="Would your imagery survive this scrutiny?" sub="Geometric QC runs on every order before delivery; the pipeline is documented." action="See how Legato works" href="/legato" tiles={['/band/caldera.webp', '/band/city.webp', '/band/hills.webp']} />

## When Bad Imagery Goes Viral

In July 2019, something unusual started appearing in social media feeds across China. Satellite images from Google Earth showed the Three Gorges Dam—one of the world's largest engineering projects—with visible geometric distortion along its structure. Instead of the expected straight line spanning the Yangtze River, the images showed what appeared to be undulations and warping.

To understand why this was alarming: the Three Gorges Dam is over 2 kilometers wide, 185 meters tall, and holds back a reservoir extending 600 kilometers. Hundreds of millions of people live downstream. The dam's scale is such that NASA detected measurable effects on Earth's rotation when the reservoir was filled.

So when images suggested the structure might be deforming, the reaction was swift and severe.

Posts proliferated across Weibo, Twitter, and WeChat. Some claimed imminent structural failure. Screenshots circulated allegedly showing warnings from Huang Xiaokun, a senior engineer at the China Academy of Building Research, calling for immediate evacuations of the Yichang area. International media picked up the story. Suddenly "concerns about the Three Gorges Dam's structural integrity" were making headlines worldwide.

Chinese authorities scrambled to respond. They released official statements, wheeled out engineers to explain the dam's monitoring systems, and published alternative high-resolution imagery from the Gaofen-6 satellite showing the structure perfectly straight.

The dam was fine. It had always been fine.

But the imagery processing? That was another story.

## The Technical Reality: What Orthorectification Does (and What Happens When It Fails)

Here's what most people don't realize about satellite imagery: what satellites capture isn't immediately usable for accurate measurement or analysis. Raw satellite data contains geometric distortions that need to be corrected before the imagery can be trusted for professional applications.

This correction process is called [orthorectification](/blog/orthorectification-explained), and it's fundamental to turning satellite acquisitions into analysis-ready data.

### How Orthorectification Works

When a satellite captures imagery, several factors introduce geometric distortions:

- **Sensor geometry**: The satellite's position, attitude, and viewing angle at acquisition
- **Terrain relief**: Elevation variations cause displacement, especially in oblique imagery
- **Earth curvature**: The spherical nature of Earth affects pixel positioning
- **Platform motion**: The satellite's movement during image capture

Think of it this way: if you photograph a tall building from an angle, the building appears to lean. The top of the building is displaced from its base in the image, even though the building itself is vertical. This is relief displacement, and it happens with every feature that has elevation when photographed from an oblique angle.

Now scale that up to entire landscapes with mountains, valleys, and yes—large artificial structures like dams.

![Animation showing the orthorectification process](/blogs/OrthoAnimation.gif)

Orthorectification corrects these distortions by using:

- **Rational Polynomial Coefficients (RPCs)**: Metadata describing the mathematical relationship between image coordinates and ground coordinates
- **Digital Elevation Models (DEMs)**: Detailed terrain height information
- **Ground control points (GCPs)**: Known positions used for validation in high-accuracy applications

When done properly, you get an orthophoto where every pixel represents its true ground position. Features appear where they actually are, measurements are reliable, and you can overlay the imagery with other geospatial data without misalignment.

When done improperly—or when the input data is inadequate—you get the Three Gorges Dam incident.

## The Perfect Storm: Why the Dam Processing Failed

The Three Gorges Dam presented a particularly challenging case for automated orthorectification:

**The elevation problem**: The dam is a 185-meter tall artificial structure creating a massive elevation change. Standard [DEMs](/blog/understanding-dems-dsms-dtms-guide), which typically derive from natural terrain data, often don't include accurate elevation information for recently constructed large-scale infrastructure. Without accurate elevation data for the dam itself, the orthorectification algorithms were essentially trying to correct for terrain that their model said was much flatter than reality.

**The stitching problem**: Google Earth creates its global imagery by mosaicking together satellite and aerial photographs acquired at different times, from different angles, under different atmospheric conditions. Each individual image requires orthorectification before being integrated into the mosaic. When the elevation model is inadequate, each image gets corrected slightly differently, creating visible distortions when they're stitched together.

**The algorithm problem**: Consumer mapping platforms optimize for global coverage and processing speed rather than precision for every specific feature. Their algorithms work well for most use cases—natural terrain, low-relief areas, standard structures. But edge cases like 185-meter dams spanning major rivers can break the assumptions these algorithms rely on.

The result? Geometric artifacts that made a perfectly stable structure look deformed.

## Why Smart People Believed Bad Data

Here's the uncomfortable part: the imagery looked convincing. People weren't being irrational when they panicked.

The distortions were visible, consistent across multiple viewpoints in Google Earth, and appeared in what most people consider an authoritative source. When you zoom in on Google Earth and see imagery that looks photographic, it's natural to trust what you're seeing.

This is the hidden risk of democratized satellite imagery access. Platforms like Google Earth have made satellite imagery available to billions of people—an incredible achievement. But most users don't understand the processing that happens between satellite acquisition and the image they see on screen. They don't know to look for processing artifacts. They don't know that certain features (like very tall structures) are more prone to orthorectification errors.

And when the stakes are high enough—when you're looking at infrastructure protecting hundreds of millions of people—misinterpreting geometric artifacts as structural deformation becomes genuinely frightening.

The Chinese government faced an interesting challenge: they needed to prove that the imagery was wrong, not the dam. That's a hard sell when people can see the "evidence" with their own eyes.

## The Broader Implications for Remote Sensing Applications

The Three Gorges Dam incident isn't just an interesting case study—it's a warning about the real-world consequences of inadequate image processing across professional applications.

### When Orthorectification Quality Matters

**Infrastructure monitoring**: Accurate geometric positioning is essential for detecting actual deformation, measuring clearances, and planning maintenance. Geometric artifacts that mimic structural movement can trigger false alarms or, worse, mask real problems.

**Mining operations**: Pit wall stability analysis depends on accurate 3D positioning over time. Orthorectification errors of several meters can make stable walls appear to be moving or obscure dangerous actual movement.

**Precision agriculture**: Field boundaries, equipment guidance, and yield mapping all require sub-meter accuracy. Poor orthorectification means GPS-guided equipment doesn't match the imagery, leading to inefficient operations.

**Environmental monitoring**: Tracking coastal erosion, glacier retreat, or deforestation requires comparing imagery over time. If images from different dates have different geometric errors, you're measuring processing differences rather than actual change.

**Disaster response**: After earthquakes or floods, response teams use satellite imagery to assess damage and prioritize resources. Geometric distortions could show damage where none exists or miss critical areas.

**Legal and regulatory applications**: Property boundaries, environmental compliance, and planning approvals often depend on satellite imagery. Geometric inaccuracies can have legal consequences.

The pattern is clear: when decisions have consequences—safety, financial, legal, environmental—the quality of orthorectification becomes non-negotiable.

### Professional Standards vs. Consumer Platforms

Consumer mapping platforms like Google Earth serve a different purpose than professional remote sensing applications. They prioritize:

- Global coverage over local precision
- Processing speed over validation
- Visual consistency over geometric accuracy
- Accessibility over technical documentation

This approach works well for their use case: providing general reference imagery to billions of users. But it creates problems when people use consumer imagery for professional applications without understanding its limitations.

Professional orthorectification, by contrast, requires:

**High-resolution elevation data**: Standard DEMs might be 30-90 meter posting. Professional applications often require 1-meter or better resolution, particularly around complex terrain or artificial structures. For the Three Gorges Dam, proper orthorectification would need elevation data that accurately represents the dam's 185-meter height and geometric complexity.

**Appropriate algorithms**: Different terrain types, structure heights, and sensor geometries require different correction approaches. Professional workflows adapt the orthorectification method to the specific characteristics of each scene.

**Rigorous validation**: Processed imagery should be verified against ground control points to quantify accuracy. Professional applications typically target sub-pixel geometric accuracy and document achieved RMSE (Root Mean Square Error).

**Proper documentation**: Professional imagery delivery includes metadata specifying DEM source and resolution, processing methods, geometric accuracy assessment, and coordinate system information.

**Computational resources**: Processing high-resolution imagery with detailed elevation models and sophisticated algorithms requires significant infrastructure that consumer platforms don't apply to every pixel of global coverage.

## What Professional Orthorectification Looks Like

Let's be specific about what separates adequate processing from professional-grade orthorectification.

### The Input Requirements

**Imagery quality**: Starting with high-quality raw imagery with complete metadata, including RPCs, sensor calibration information, and acquisition parameters.

**Elevation models**: Using DEMs with resolution appropriate to the application. For infrastructure monitoring, this often means 1-meter or better resolution. For the Three Gorges Dam, this would require elevation data that accurately captures the dam's geometry, possibly derived from stereo imagery or lidar specifically acquired for the structure.

**Ground control**: When absolute positional accuracy is critical, incorporating GCPs with known coordinates to refine the geometric model. Professional applications might use 10-20 well-distributed GCPs depending on area size and accuracy requirements.

### The Processing Standards

Professional orthorectification workflows include:

- Sensor-specific correction algorithms that account for each satellite's unique optical characteristics
- Adaptive resampling methods that preserve radiometric fidelity while correcting geometry
- Automated quality assessment checking for residual distortions
- Validation against independent reference data
- Documentation of achieved geometric accuracy

### The Validation Process

This is where professional processing diverges most from consumer platforms. Professional imagery undergoes systematic validation:

**Checkpoints**: Independent ground control points not used in the processing are used to assess accuracy

**Accuracy metrics**: RMSE, CE90 (Circular Error 90%), or other standard metrics quantify geometric precision

**Visual inspection**: Experienced analysts review the imagery for artifacts, edge effects, and alignment issues

**Cross-validation**: When possible, comparing with other data sources to verify consistency

The result is imagery where you know the accuracy, understand the limitations, and can trust measurements derived from it.

## The Geopera Approach: Making Professional Processing Standard

At Geopera, our philosophy is straightforward: satellite imagery should be delivered ready for professional use, with the processing quality that critical applications demand.

Every image from our platform undergoes comprehensive orthorectification using:

- High-resolution elevation models appropriate for each scene's characteristics
- Advanced algorithms that adapt to terrain complexity and feature types
- Rigorous quality control ensuring sub-pixel geometric accuracy
- Complete validation with documented accuracy metrics

This isn't positioned as a premium service or optional add-on—it's our standard. Because we believe inadequate orthorectification isn't just a technical shortcoming; it's a fundamental failure to deliver usable data.

For applications monitoring infrastructure, analyzing environmental change, planning operations, or supporting critical decisions, you need to know that what you're seeing represents reality, not processing artifacts.

We handle the complexity—elevation data acquisition, algorithm selection, computational processing, quality validation—so you can focus on analysis and insights rather than questioning whether your imagery is geometrically reliable.

Learn more about [why we process every image](/blog/why-we-process-every-order) to professional standards.

## What to Look for in Satellite Imagery Providers

If you're sourcing satellite imagery for professional applications, here are the questions you should ask:

**What DEM resolution is used for orthorectification?** If the answer is vague or the provider doesn't specify, that's a red flag. Professional processing requires elevation data appropriate to the application.

**What geometric accuracy is achieved?** Look for specific metrics—RMSE in meters or CE90 values. "High accuracy" or "professional quality" without numbers means nothing.

**How is accuracy validated?** There should be independent checkpoints and documented validation procedures.

**Is processing documentation included?** Professional delivery should include metadata specifying processing methods, elevation data sources, coordinate systems, and accuracy assessment.

**Can you provide custom processing?** Different applications have different requirements. A provider should be able to discuss trade-offs between processing approaches and adapt to your specific needs.

If a provider can't answer these questions, or if orthorectification is treated as an optional upgrade, you're likely getting consumer-grade processing inadequate for professional applications.

## The Takeaway

The Three Gorges Dam incident happened because imagery processing designed for general consumer use was interpreted as if it had professional-grade geometric accuracy. The dam wasn't failing—the orthorectification was inadequate for that specific application.

This matters beyond China and beyond large dams. Every day, decisions get made based on satellite imagery. Some of those decisions affect safety, investments, environmental protection, or legal compliance. When the imagery processing is inadequate, those decisions get made on faulty information.

The solution isn't to distrust all satellite imagery. The solution is to understand that not all imagery is created equal, and to demand professional-grade processing for professional applications.

Consumer platforms are excellent for what they're designed to do. But when accuracy matters—when you're measuring, monitoring, or analyzing features where precision affects outcomes—you need imagery processed to professional standards, with validation to prove it.

Nobody needs their own Three Gorges Dam moment. Whether you're monitoring infrastructure, managing resources, or conducting environmental analysis, you deserve imagery that shows reality, not artifacts.

Ready to work with properly processed satellite imagery? [Explore the Pera Portal](https://portal.geopera.com) to see what professional-grade orthorectification delivers. Every image comes with complete processing and documented accuracy—because that's what professional applications require.

Or [contact our team](/contact) to discuss your specific requirements. We're here to ensure your satellite imagery supports confident decision-making, not confusion about whether what you're seeing is real.

## Frequently Asked Questions

**Q: Is the Three Gorges Dam bending?**

No. The apparent bending in the viral 2019 satellite images existed only in the image processing, not the structure. The dam's operator responded with monitoring data showing horizontal crest movement of 1.4 to 26.7 mm, described as elastic deformation within design limits, and imagery from the Gaofen-6 satellite showing the dam straight. All large dams move at millimetre scale with water level and temperature; that movement is designed for.

**Q: How common are orthorectification errors in consumer satellite imagery platforms?**

Geometric artifacts are relatively common in consumer platforms, particularly in areas with extreme relief, tall artificial structures, or at the edges of image tiles where different acquisitions are mosaicked together. Most users don't notice because they're using the imagery for general reference rather than precision measurements. The Three Gorges Dam case was unusual because the stakes were high enough to generate panic when people noticed the distortions.

**Q: Can orthorectification errors be corrected after imagery is delivered?**

In some cases, yes—if you have access to the original raw imagery, proper elevation data, and the technical expertise. However, this requires essentially reprocessing the imagery from scratch, which is why it's far more efficient to ensure proper orthorectification during initial processing. Trying to "fix" poorly orthorectified imagery is technically challenging and often produces suboptimal results.

**Q: What accuracy should I expect from professionally orthorectified imagery?**

This depends on several factors including sensor resolution, DEM quality, and whether ground control points are used. For high-resolution satellite imagery (0.3-1m resolution) with proper orthorectification, you should expect horizontal accuracy of 2-3 meters RMSE without GCPs, or sub-meter accuracy when GCPs are incorporated. Your provider should document the achieved accuracy for your specific imagery.

**Q: Is Google Earth imagery suitable for any professional applications?**

Google Earth imagery can be appropriate for reconnaissance, general planning, visual reference, and applications where approximate positioning is sufficient. However, for applications requiring accurate measurements, change detection, or integration with other geospatial datasets, professionally orthorectified imagery with documented accuracy is necessary. Always match the imagery quality to your accuracy requirements.

**Q: How much does professional orthorectification typically add to imagery costs?**

With traditional providers, orthorectification can add 30-80% to base imagery costs, and high-accuracy processing with GCPs can cost even more. At Geopera, we include professional-grade orthorectification as standard with every order at no additional cost, because we believe analysis-ready imagery is the only type worth delivering.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# 10 Best Free Sources of Satellite Data for Brazil

> A comprehensive list of the best free satellite data sources in Brazil

Published: 2025-10-29 | Author: Darcy Weedman | Reading time: 6 minute

Source: https://geopera.com/blog/free-sources-satellite-data-brazil

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- [Pera Portal](https://portal.geopera.com/): Unlimited free Sentinel-2 browsing with 45+ pre-configured spectral indices (NDVI, NDWI, EVI, etc.) viewable in-browser. Advanced band ratio calculations, zero setup required. Global coverage including comprehensive Brazilian imagery. Plus access to 100+ commercial satellites for premium imagery when needed

- [INPE (National Institute for Space Research)](http://www.dgi.inpe.br/catalogo/): Best comprehensive source for Brazilian satellite data including CBERS and Amazonia-1 imagery

- [MapBiomas](https://mapbiomas.org/): Best for historical land use and land cover mapping across Brazil with annual updates since 1985

- [TerraClass Amazonia](http://www.inpe.br/cra/projetos_pesquisas/terraclass.php): Best for Amazon rainforest monitoring and deforestation tracking

- [Google Earth](https://earth.google.com/web): Best for general purpose viewing and basic exploration of your property

- [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/): Best for environmental monitoring and agricultural applications with frequent Sentinel updates

- [NASA FIRMS (Fire Information)](https://firms.modaps.eosdis.nasa.gov/): Best for real-time fire detection and monitoring across Brazilian territories

- [Global Forest Watch](https://www.globalforestwatch.org/): Best for forest monitoring, deforestation alerts, and environmental tracking

- [IBGE Geosciences](https://www.ibge.gov.br/geociencias/): Best for official Brazilian government geospatial data and statistical mapping

- [Zoom Earth](https://zoom.earth/): Best for near-current weather and environmental condition tracking across Brazil

<AskBand headline="What does your part of Brazil look like this week?" sub="Free Sentinel-2 with indices in the browser, no credit card." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/delta.webp', '/band/fields.webp', '/band/bloom.webp']} />

## Overview

Looking for satellite images of your house in Brazil? Want to view historical satellite imagery of your property for free? You're in the right place.

This guide will help you explore the various sources of satellite data available in Brazil, whether you want to view imagery directly in your browser or download data for further analysis.

### Understanding Real-Time Satellite Imagery

Many people search for "live satellite view" or "real-time satellite images" of their house, but here's the truth — despite what movies might suggest, live satellite imagery isn't possible with current technology. What you see on mapping platforms is usually imagery captured weeks or even months ago.

### What's Available in Brazil

If you're searching for free satellite imagery in Brazil, it's important to understand what's available:

- Most free satellite images are updated every few months to years.
- Free imagery generally has lower resolution (you can see buildings, but not detailed features).
- High-resolution imagery with clear details often requires a paid subscription.
- Real-time satellite imagery does exist but is typically limited to weather forecasting or emergency services.

Fortunately, Brazil has world-leading satellite data infrastructure through INPE and several other government and research platforms. Whether you are a property owner, researcher, farmer, or just curious, we'll walk you through the best free and paid options available.

---

## 1. INPE - National Institute for Space Research

[INPE's Image Catalog](http://www.dgi.inpe.br/catalogo/) is Brazil's premier platform for accessing free satellite imagery. INPE operates Brazil's own satellites and provides comprehensive coverage of Brazilian territory.

**Key Features:**

- Free access to CBERS (China-Brazil Earth Resources Satellite) imagery.
- Amazonia-1 satellite data — Brazil's first 100% nationally-built Earth observation satellite.
- Landsat data specifically processed for Brazilian territories.
- Historical archives dating back decades.
- Advanced search by location, date, and satellite sensor.
- Ideal for agricultural monitoring, environmental research, and natural resource management across Brazil.

---

## 2. MapBiomas

[MapBiomas](https://mapbiomas.org/) is an extraordinary Brazilian initiative providing annual land use and land cover maps for all of Brazil from 1985 to present.

**Key Features:**

- Complete land use classification for all Brazilian biomes (Amazon, Cerrado, Caatinga, Atlantic Forest, Pantanal, Pampa).
- Annual maps showing deforestation, agricultural expansion, and urbanization.
- Interactive platform for visualizing changes over time.
- Free downloads of classification maps and statistics.
- Essential for environmental monitoring, agricultural planning, and policy analysis.
- Transition matrices showing land use changes between years.

---

## 3. Pera Portal

Our [Pera Portal](https://portal.geopera.com/) provides completely free, unlimited access to Sentinel-2 satellite imagery with advanced analytical capabilities that surpass most free platforms.

**Key Features:**

- **Unlimited free Sentinel-2 browsing** with no registration walls or download limits.
- **45+ pre-configured spectral indices** viewable instantly in your browser:
  - NDVI (vegetation health and crop monitoring)
  - NDWI (water detection crucial for Brazilian watersheds)
  - EVI (enhanced vegetation index optimized for tropical forests)
  - SAVI (soil-adjusted for Brazilian agricultural regions)
  - MSAVI, NDBI, and many more
- **Custom band ratio calculations** — create your own indices for specialized analysis.
- **Zero setup required** — view processed imagery immediately without downloading or installing software.
- **Global coverage** including comprehensive Brazilian imagery updated every 5 days.
- **Ideal for Brazilian applications**:
  - Agricultural monitoring across Cerrado, Mato Grosso, and southern states
  - Amazon rainforest health tracking
  - Atlantic Forest conservation monitoring
  - Sugarcane and soybean crop assessment
  - Pantanal wetland monitoring
  - Water resource management for hydroelectric planning
- **Premium commercial imagery access** — upgrade to browse 100+ commercial satellites with sub-meter resolution when you need more detail.

Unlike other platforms that require technical knowledge or large downloads, Pera Portal makes advanced satellite imagery analysis accessible to everyone — from fazendeiros monitoring crop health in the interior to researchers tracking Amazon deforestation patterns.

---

## 4. TerraClass Amazonia

[TerraClass Amazonia](http://www.inpe.br/cra/projetos_pesquisas/terraclass.php) is INPE's specialized platform for monitoring land use in deforested areas of the Brazilian Amazon.

**Key Features:**

- Detailed classification of land use in previously deforested Amazon areas.
- Multi-year datasets showing agricultural conversion and regeneration.
- Free access to classification maps and reports.
- Critical for understanding Amazon land dynamics beyond simple forest/non-forest mapping.
- Distinguishes between pasture, agriculture, secondary vegetation, and other uses.

---

## 5. Google Earth

[Google Earth](https://earth.google.com/web) provides high-resolution satellite imagery across Brazil, including comprehensive coverage of both urban and remote Amazonian regions.

**Key Features:**

- Simple, intuitive interface accessible from any web browser.
- Historical imagery slider to view changes over time — particularly valuable for tracking deforestation.
- 3D terrain visualization perfect for mountainous regions.
- Coverage of remote areas including deep Amazon and Pantanal.
- Not suitable for downloading raw satellite data or technical analysis.

---

## 6. Copernicus Data Space Ecosystem

The European Union's [Copernicus](https://browser.dataspace.copernicus.eu/) program offers free, high-quality Sentinel satellite data with global coverage including all of Brazil.

**Key Features:**

- Sentinel-2 optical imagery updated every 5 days with 10-meter resolution.
- Sentinel-1 radar data for all-weather monitoring (crucial for frequently cloudy Amazon conditions).
- Free access without download limits or registration barriers.
- Ideal for agricultural monitoring in southern Brazil and Cerrado.
- Particularly valuable for monitoring Amazon where cloud cover often obscures optical imagery.
- Cloud-based visualization and download options.

---

## 7. NASA FIRMS - Fire Information for Resource Management System

[NASA FIRMS](https://firms.modaps.eosdis.nasa.gov/) provides near real-time active fire detection crucial for monitoring fires across Brazilian territories, especially in the Amazon and Cerrado.

**Key Features:**

- Active fire detection updated every 3 hours.
- Historical fire data dating back to 2000.
- Email alerts for fire detection in areas of interest.
- Essential during dry season for tracking agricultural burning and wildfires.
- Free API access for integration into monitoring systems.
- Critical tool for environmental enforcement and fire management.

---

## 8. Global Forest Watch

[Global Forest Watch](https://www.globalforestwatch.org/) provides comprehensive forest monitoring tools with extensive Brazil coverage and Amazon-specific features.

**Key Features:**

- Near real-time deforestation alerts for Brazilian forests.
- Historical tree cover loss data since 2000.
- Interactive maps showing deforestation patterns across all Brazilian biomes.
- Integration with indigenous territories and protected areas.
- Free email alerts for deforestation in areas you monitor.
- Essential for conservation organizations and environmental monitoring.

---

## 9. IBGE Geosciences

The Brazilian Institute of Geography and Statistics [IBGE Geosciences](https://www.ibge.gov.br/geociencias/) provides official government geospatial data including satellite-derived products.

**Key Features:**

- Official Brazilian government territorial data.
- Vegetation maps and biome classifications.
- Digital elevation models for all of Brazil.
- Land use and land cover data integrated with census information.
- Free downloads of authoritative datasets.
- Essential for official planning and research applications.

---

## 10. Zoom Earth

[Zoom Earth](https://zoom.earth/) provides frequently updated satellite imagery, focusing on weather patterns and environmental conditions across Brazil.

**Key Features:**

- Near-current weather and storm tracking.
- Updated multiple times daily with latest satellite passes.
- Excellent for tracking Amazon weather systems and rainfall patterns.
- Time-lapse animations showing weather development.
- No registration required for basic viewing.
- Useful for agricultural planning and weather monitoring.

---

## Tips for Making the Most of Free Satellite Imagery Platforms

**1. Find the Right Platforms.**
We've highlighted some of the best satellite data providers for Brazilian applications. INPE's platforms are specifically designed for Brazilian needs and should be your first choice for local data.

**2. Understand Platform Capabilities.**
Each platform offers unique features. For Amazon monitoring, consider radar-based Sentinel-1 data which penetrates frequent cloud cover. For agriculture in southern Brazil and Cerrado, optical data often works well.

**3. Define Your Requirements.**
Specify what you need — the area of interest, time range, spatial resolution, and spectral bands — and use search filters to narrow down the results. Brazil's vast territory means precise targeting is essential.

**4. Use Visualisation Tools.**
Many platforms provide built-in tools to help you explore and analyse the data. Pera Portal's spectral index capabilities, for example, can provide insights into crop health, forest conditions, and water resources without needing external software or technical expertise.

**5. Be Selective with Downloads.**
Rather than downloading full datasets, focus on specific tiles or areas of interest to save time and storage space. For many use cases, cloud-based platforms like Pera Portal eliminate the need for downloads entirely.

By following these tips, you'll be able to make the most of the available satellite imagery resources in Brazil, whether you're conducting research, managing agricultural operations, monitoring environmental changes, or tracking deforestation across this ecologically critical country.

---

## When Free Satellite Data Isn't Enough

Free platforms cover an enormous range of uses — and if you're simply curious about your own property, they're genuinely all you'll ever need. But if you're using imagery professionally, you'll eventually hit the resolution wall:

| Source               | Resolution  | What you can actually see                                 |
| -------------------- | ----------- | --------------------------------------------------------- |
| Landsat (free)       | 30 m        | Regional land cover — a soccer pitch is roughly one pixel |
| Sentinel-2 (free)    | 10 m        | Field-scale vegetation patterns; buildings are blurs      |
| Commercial (Geopera) | up to 30 cm | Individual vehicles, fence lines, machinery, single trees |

The difference matters the moment you need to **measure rather than look**: verifying land use at the property level for CAR compliance, monitoring crops by the row rather than by the municipality, documenting clearing along a specific boundary, or proving conditions on a specific date. Free sensors also can't be tasked — and over the Amazon and the agricultural frontier, persistent cloud means the archive often holds no usable capture for the window you need.

Commercial imagery used to mean opaque quotes and weeks of back-and-forth. We publish [transparent per-square-kilometre pricing](/pricing) for both tasking and archive, and every order arrives analysis-ready — orthorectified, pansharpened, colour-balanced and mosaicked ([here's exactly what that involves](/imagery)).

If free data has taken your project as far as it can go, [explore available imagery through Pera Portal](https://portal.geopera.com/) or [get in touch to discuss your project](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# 10 Best Free Sources of Satellite Data for Germany

> A comprehensive list of the best free satellite data sources in Germany

Published: 2025-10-28 | Author: Darcy Weedman | Reading time: 6 minute

Source: https://geopera.com/blog/free-sources-satellite-data-germany

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- [Pera Portal](https://portal.geopera.com/): Unlimited free Sentinel-2 browsing with 45+ pre-configured spectral indices (NDVI, NDWI, EVI, etc.) viewable in-browser. Advanced band ratio calculations, zero setup required. Global coverage including comprehensive German imagery. Plus access to 100+ commercial satellites for premium imagery when needed

- [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/): Best comprehensive source for European satellite data with extensive Germany coverage

- [BKG (Federal Agency for Cartography and Geodesy)](https://gdz.bkg.bund.de/): Best for official German government geospatial data and topographic maps

- [GeoPortal.de](https://www.geoportal.de/): Best for accessing German state and federal geospatial data from multiple agencies

- [Google Earth](https://earth.google.com/web): Best for general purpose viewing and basic exploration of your property

- [State-Level Geoportals](https://www.geoportal.de/): Best for high-resolution state-specific imagery from Bavaria, North Rhine-Westphalia, Baden-Württemberg, etc.

- [ESA Earth Online](https://earth.esa.int/): Best for accessing European Space Agency satellite missions and research data

- [NASA Earthdata Search](https://search.earthdata.nasa.gov/): Best for scientific research and comprehensive global satellite data

- [Maxar Open Data Program](https://www.maxar.com/open-data): Best for disaster response imagery during floods and emergencies

- [Zoom Earth](https://zoom.earth/): Best for near-current weather and environmental condition tracking across Germany

<AskBand headline="What does your part of Germany look like this week?" sub="Free Sentinel-2 with indices in the browser, no credit card." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/fields.webp', '/band/city.webp', '/band/oxbow.webp']} />

## Overview

Looking for satellite images of your house in Germany? Want to view historical satellite imagery of your property for free? You're in the right place.

This guide will help you explore the various sources of satellite data available in Germany, whether you want to view imagery directly in your browser or download data for further analysis.

### Understanding Real-Time Satellite Imagery

Many people search for "live satellite view" or "real-time satellite images" of their house, but here's the truth — despite what movies might suggest, live satellite imagery isn't possible with current technology. What you see on mapping platforms is usually imagery captured weeks or even months ago.

### What's Available in Germany

If you're searching for free satellite imagery in Germany, it's important to understand what's available:

- Most free satellite images are updated every few months to years.
- Free imagery generally has lower resolution (you can see buildings, but not detailed features).
- High-resolution imagery with clear details often requires a paid subscription.
- Real-time satellite imagery does exist but is typically limited to weather forecasting or emergency services.

Fortunately, Germany has excellent access to satellite data through the EU's Copernicus program, German federal agencies, and state-level geoportals. Whether you are a property owner, researcher, or just curious, we'll walk you through the best free and paid options available.

---

## 1. Copernicus Data Space Ecosystem

The European Union's [Copernicus](https://browser.dataspace.copernicus.eu/) program is the premier platform for accessing free, high-quality satellite data across Europe including comprehensive Germany coverage.

**Key Features:**

- Sentinel-2 optical imagery updated every 5 days with 10-meter resolution.
- Sentinel-1 radar data for all-weather monitoring.
- Complete historical archive dating back to 2014 (Sentinel-1) and 2015 (Sentinel-2).
- Free access without download limits or registration barriers.
- Cloud-based visualization and download options.
- Ideal for environmental monitoring, agriculture, forestry, and urban development tracking.
- Pan-European coverage with consistent data quality.

---

## 2. BKG - Federal Agency for Cartography and Geodesy

[BKG's Geodata Centre](https://gdz.bkg.bund.de/) provides official German government geospatial data including satellite imagery and topographic products.

**Key Features:**

- Official federal geospatial data for all of Germany.
- Digital elevation models and terrain data.
- Topographic maps and administrative boundaries.
- Integration with European geodata initiatives.
- Free access to many datasets without registration.
- Authoritative data for official planning and research applications.

---

## 3. Pera Portal

Our [Pera Portal](https://portal.geopera.com/) provides completely free, unlimited access to Sentinel-2 satellite imagery with advanced analytical capabilities that surpass most free platforms.

**Key Features:**

- **Unlimited free Sentinel-2 browsing** with no registration walls or download limits.
- **45+ pre-configured spectral indices** viewable instantly in your browser:
  - NDVI (vegetation health and crop monitoring)
  - NDWI (water detection for German rivers and lakes)
  - EVI (enhanced vegetation index for German agriculture)
  - SAVI (soil-adjusted for agricultural regions)
  - MSAVI, NDBI, and many more
- **Custom band ratio calculations** — create your own indices for specialized analysis.
- **Zero setup required** — view processed imagery immediately without downloading or installing software.
- **Global coverage** including comprehensive German imagery updated every 5 days.
- **Ideal for German applications**:
  - Agricultural monitoring across Bavaria, Lower Saxony, and other agricultural states
  - Forest health assessment in Black Forest, Bavarian Forest, and Harz regions
  - Rhine, Danube, and Elbe river monitoring
  - Urban development tracking in growing cities
  - Renewable energy site assessment (solar farm monitoring, etc.)
  - Coastal monitoring along North Sea and Baltic Sea coasts
- **Premium commercial imagery access** — upgrade to browse 100+ commercial satellites with sub-meter resolution when you need more detail.

Unlike other platforms that require technical knowledge or large downloads, Pera Portal makes advanced satellite imagery analysis accessible to everyone — from farmers monitoring precision agriculture to researchers tracking land use changes across German landscapes.

---

## 4. GeoPortal.de

[GeoPortal.de](https://www.geoportal.de/) is Germany's central access point for geospatial data from federal, state, and local government agencies.

**Key Features:**

- Unified access to data from all 16 German states (Bundesländer).
- Integration of federal and state-level satellite imagery.
- Standardized search across multiple data providers.
- Links to state-specific geoportals for detailed local data.
- Free access to harmonized datasets across Germany.
- Compliance with EU INSPIRE directive for data sharing.

---

## 5. Google Earth

[Google Earth](https://earth.google.com/web) provides high-resolution satellite imagery across Germany. It's the easiest platform for casual users who want to view satellite images of their house.

**Key Features:**

- Simple, intuitive interface accessible from any web browser.
- Historical imagery slider to view changes over time.
- 3D building models for major German cities including Berlin, Munich, Hamburg, and Frankfurt.
- Coverage of all regions from North Sea coast to Bavarian Alps.
- Not suitable for downloading raw satellite data or technical analysis.

---

## 6. State-Level Geoportals

German states (Bundesländer) provide their own high-resolution imagery and geospatial data, often with more frequent updates than federal sources:

### Bavaria - BayernAtlas

[BayernAtlas](https://geoportal.bayern.de/bayernatlas/) provides high-resolution aerial imagery and satellite data for Bavaria.

### North Rhine-Westphalia - GEOportal.NRW

[GEOportal.NRW](https://www.geoportal.nrw/) offers comprehensive geospatial data for Germany's most populous state.

### Baden-Württemberg - Geoportal BW

[Geoportal BW](https://www.geoportal-bw.de/) provides detailed imagery and spatial data for Baden-Württemberg.

### Other State Geoportals

- **Berlin**: [FIS-Broker Berlin](https://fbinter.stadt-berlin.de/)
- **Saxony**: [GeoSN Geodatenportal](https://geoportal.sachsen.de/)
- **Hesse**: [Geoportal Hessen](https://geoportal.hessen.de/)
- **Lower Saxony**: [LGLN Geodatenportal](https://www.geodaten.niedersachsen.de/)

---

## 7. ESA Earth Online

The European Space Agency's [Earth Online](https://earth.esa.int/) provides access to ESA satellite missions and Earth observation data.

**Key Features:**

- Access to ESA's complete satellite mission archive.
- Sentinel data with additional processing and products.
- Historical data from older ESA missions.
- Research-grade datasets and analysis tools.
- Free registration for access to comprehensive archives.
- Ideal for scientific research and advanced applications.

---

## 8. NASA Earthdata Search

[NASA Earthdata Search](https://search.earthdata.nasa.gov/) provides comprehensive satellite datasets with global coverage including Germany.

**Key Features:**

- Complete Landsat archive dating back to 1972 for long-term German landscape analysis.
- MODIS data for large-scale environmental monitoring.
- Free registration and unlimited downloads.
- Advanced search filters by location, date, and cloud cover.
- Essential for scientific research and climate studies.

---

## 9. Maxar Open Data Program

[Maxar](https://www.maxar.com/open-data) offers high-resolution commercial satellite imagery for free during natural disasters and emergencies in Germany.

**Key Features:**

- Sub-meter resolution imagery during floods (particularly along Rhine, Elbe, and Danube).
- Released within 24-48 hours of disaster events.
- Critical for emergency response during severe weather events.
- Before-and-after imagery for disaster assessment and recovery planning.
- Partnerships with European emergency management agencies.

---

## 10. Zoom Earth

[Zoom Earth](https://zoom.earth/) provides frequently updated satellite imagery, focusing on weather patterns and environmental conditions across Germany.

**Key Features:**

- Near-current weather and storm tracking.
- Updated multiple times daily with latest satellite passes.
- Excellent for tracking weather systems from the North Sea and Atlantic.
- Time-lapse animations showing weather development.
- No registration required for basic viewing.
- Useful for agricultural planning and weather monitoring.

---

## Tips for Making the Most of Free Satellite Imagery Platforms

**1. Find the Right Platforms.**
We've highlighted some of the best satellite data providers for German applications. Start with Copernicus and state-level geoportals for the best German-specific data.

**2. Understand Platform Capabilities.**
Each platform offers unique features. Familiarise yourself with their resolution, data quality, and update frequency to match your project's needs. Germany's well-developed data infrastructure provides excellent free options.

**3. Define Your Requirements.**
Specify what you need — the area of interest, time range, spatial resolution, and spectral bands — and use search filters to narrow down the results. German territories have excellent coverage across all platforms.

**4. Use Visualisation Tools.**
Many platforms provide built-in tools to help you explore and analyse the data. Pera Portal's spectral index capabilities, for example, can provide insights into crop health, forest conditions, and water resources without needing external software or technical expertise.

**5. Be Selective with Downloads.**
Rather than downloading full datasets, focus on specific tiles or areas of interest to save time and storage space. For many use cases, cloud-based platforms like Pera Portal eliminate the need for downloads entirely.

By following these tips, you'll be able to make the most of the available satellite imagery resources in Germany, whether you're conducting research, evaluating property, monitoring agricultural land, tracking environmental changes, or assessing infrastructure development across this technologically advanced nation.

---

## When Free Satellite Data Isn't Enough

Free platforms cover an enormous range of uses — and if you're simply curious about your own property, they're genuinely all you'll ever need. But if you're using imagery professionally, you'll eventually hit the resolution wall:

| Source               | Resolution  | What you can actually see                                    |
| -------------------- | ----------- | ------------------------------------------------------------ |
| Landsat (free)       | 30 m        | Regional land cover — a football pitch is roughly one pixel  |
| Sentinel-2 (free)    | 10 m        | Field-scale vegetation patterns; buildings are blurs         |
| Commercial (Geopera) | up to 30 cm | Individual vehicles, fence lines, solar panels, single trees |

The difference matters the moment you need to **measure rather than look**: monitoring construction or industrial sites week-by-week, auditing solar and wind assets, tracking change at the parcel level rather than the Landkreis, or proving site conditions on a specific date. Free sensors also can't be tasked — if no satellite happened to capture your site cloud-free in the window you care about, there's nothing to download.

Commercial imagery used to mean opaque quotes and weeks of back-and-forth. We publish [transparent per-square-kilometre pricing](/pricing) for both tasking and archive, and every order arrives analysis-ready — orthorectified, pansharpened, colour-balanced and mosaicked ([here's exactly what that involves](/imagery)).

If free data has taken your project as far as it can go, [explore available imagery through Pera Portal](https://portal.geopera.com/) or [get in touch to discuss your project](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# 10 Best Free Sources of Satellite Data for the UK

> A comprehensive list of the best free satellite data sources in the United Kingdom

Published: 2025-10-27 | Author: Darcy Weedman | Reading time: 6 minute

Source: https://geopera.com/blog/free-sources-satellite-data-uk

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- [Pera Portal](https://portal.geopera.com/): Unlimited free Sentinel-2 browsing with 45+ pre-configured spectral indices (NDVI, NDWI, EVI, etc.) viewable in-browser. Advanced band ratio calculations, zero setup required. Global coverage including comprehensive UK imagery. Plus access to 100+ commercial satellites for premium imagery when needed

- [Ordnance Survey Open Data](https://www.ordnancesurvey.co.uk/products/os-open-data): Best for detailed mapping products, aerial imagery, and property boundaries across the UK

- [UK Centre for Ecology & Hydrology Countryside Survey](https://www.ceh.ac.uk/data): Best for environmental and land cover data specific to British landscapes

- [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/): Best for environmental monitoring and agricultural applications with frequent Sentinel updates across Europe

- [Google Earth](https://earth.google.com/web): Best for general purpose viewing and basic exploration of your property

- [DEFRA Data Services Platform](https://environment.data.gov.uk/): Best for UK environmental data, land use, and agricultural monitoring

- [Scotland's Spatial Data](https://spatialdata.gov.scot/): Best for Scottish-specific satellite imagery and geospatial data

- [NASA Earthdata Search](https://search.earthdata.nasa.gov/): Best for scientific research and comprehensive global satellite data

- [Maxar Open Data Program](https://www.maxar.com/open-data): Best for disaster response imagery during floods and emergencies

- [Zoom Earth](https://zoom.earth/): Best for near-current weather and environmental condition tracking across the UK

<AskBand headline="What does your patch of the UK look like this week?" sub="Free Sentinel-2 with indices in the browser, no credit card." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/fields.webp', '/band/port.webp', '/band/hills.webp']} />

## Overview

Looking for satellite images of your house in the United Kingdom? Want to view historical satellite imagery of your property for free? You're in the right place.

This guide will help you explore the various sources of satellite data available in the UK, whether you want to view imagery directly in your browser or download data for further analysis.

### Understanding Real-Time Satellite Imagery

Many people search for "live satellite view" or "real-time satellite images" of their house, but here's the truth — despite what movies might suggest, live satellite imagery isn't possible with current technology. What you see on mapping platforms is usually imagery captured weeks or even months ago.

### What's Available in the UK

If you're searching for free satellite imagery in the UK, it's important to understand what's available:

- Most free satellite images are updated every few months to years.
- Free imagery generally has lower resolution (you can see buildings, but not detailed features).
- High-resolution imagery with clear details often requires a paid subscription.
- Real-time satellite imagery does exist but is typically limited to weather forecasting or emergency services.

Fortunately, the United Kingdom offers several excellent free sources of satellite data through government agencies, the EU's Copernicus program, and research institutions. Whether you are a property owner, researcher, or just curious, we'll walk you through the best free and paid options available.

---

## 1. Ordnance Survey Open Data

[Ordnance Survey Open Data](https://www.ordnancesurvey.co.uk/products/os-open-data) is the UK's national mapping agency, providing free access to geospatial data including aerial imagery and detailed mapping products.

**Key Features:**

- High-quality mapping products covering all of Great Britain.
- OS Open Data includes topographic maps, road networks, and terrain models.
- Aerial imagery available through various OS products.
- Regular updates reflecting changes in the British landscape.
- Free downloads of comprehensive datasets without registration barriers.

---

## 2. UK Centre for Ecology & Hydrology Data

The [UK Centre for Ecology & Hydrology](https://www.ceh.ac.uk/data) provides environmental monitoring data including satellite-derived land cover maps and vegetation indices.

**Key Features:**

- UK-specific land cover maps derived from satellite data.
- Long-term environmental monitoring datasets dating back decades.
- Countryside Survey data tracking changes in British landscapes.
- Free access to research-quality datasets.
- Ideal for environmental research, conservation, and land management.

---

## 3. Pera Portal

Our [Pera Portal](https://portal.geopera.com/) provides completely free, unlimited access to Sentinel-2 satellite imagery with advanced analytical capabilities that surpass most free platforms.

**Key Features:**

- **Unlimited free Sentinel-2 browsing** with no registration walls or download limits.
- **45+ pre-configured spectral indices** viewable instantly in your browser:
  - NDVI (vegetation health and crop monitoring)
  - NDWI (water detection and flood monitoring)
  - EVI (enhanced vegetation index for British agriculture)
  - SAVI, MSAVI, NDBI, and many more
- **Custom band ratio calculations** — create your own indices for specialized analysis.
- **Zero setup required** — view processed imagery immediately without downloading or installing software.
- **Global coverage** including comprehensive UK imagery updated every 5 days.
- **Ideal for UK applications**:
  - Agricultural monitoring across England, Scotland, Wales, and Northern Ireland
  - Flood risk assessment and water resource management
  - Urban development tracking in growing cities
  - Coastal erosion monitoring
  - Forestry and conservation management
- **Premium commercial imagery access** — upgrade to browse 100+ commercial satellites with sub-meter resolution when you need more detail.

Unlike other platforms that require technical knowledge or large downloads, Pera Portal makes advanced satellite imagery analysis accessible to everyone — from farmers monitoring crop health across the Home Counties to researchers tracking peatland restoration in Scotland.

---

## 4. Copernicus Data Space Ecosystem

The European Union's [Copernicus](https://browser.dataspace.copernicus.eu/) program offers free, high-quality Sentinel satellite data with comprehensive coverage of the United Kingdom and Europe.

**Key Features:**

- Sentinel-2 optical imagery updated every 5 days with 10-meter resolution.
- Sentinel-1 radar data for all-weather monitoring (crucial for frequently cloudy UK conditions).
- Free access without download limits or registration barriers.
- Ideal for agricultural monitoring, forestry, and environmental research.
- Cloud-based visualization and download options.
- Perfect for tracking changes across British landscapes.

---

## 5. Google Earth

[Google Earth](https://earth.google.com/web) provides high-resolution satellite imagery across the United Kingdom. It's the easiest platform for casual users who want to view satellite images of their house.

**Key Features:**

- Simple, intuitive interface accessible from any web browser.
- Historical imagery slider to view changes over time.
- 3D building models for major UK cities including London, Manchester, and Edinburgh.
- Coverage of all regions including remote Scottish Highlands and islands.
- Not suitable for downloading raw satellite data or technical analysis.

---

## 6. DEFRA Data Services Platform

The Department for Environment, Food & Rural Affairs [Data Services Platform](https://environment.data.gov.uk/) provides environmental and agricultural data including satellite-derived products.

**Key Features:**

- UK environmental data including land use and agricultural statistics.
- Integration with EU Copernicus data for British territories.
- Flood risk mapping and water quality monitoring.
- Free access to government environmental datasets.
- Useful for planning, environmental assessment, and agricultural applications.

---

## 7. Scotland's Spatial Data

[Scotland's Spatial Data](https://spatialdata.gov.scot/) provides comprehensive geospatial data specific to Scotland, including satellite imagery and aerial photography.

**Key Features:**

- High-resolution aerial photography covering all of Scotland.
- Satellite imagery and derived products for Scottish territories.
- Land cover maps and environmental monitoring data.
- Integration with Ordnance Survey data for seamless coverage.
- Free access to Scottish government spatial data.

---

## 8. NASA Earthdata Search

[NASA Earthdata Search](https://search.earthdata.nasa.gov/) provides comprehensive satellite datasets with global coverage including the United Kingdom.

**Key Features:**

- Complete Landsat archive dating back to 1972 for long-term UK landscape analysis.
- MODIS data for large-scale environmental monitoring.
- Free registration and unlimited downloads.
- Advanced search filters by location, date, and cloud cover.
- Essential for scientific research and climate studies.

---

## 9. Maxar Open Data Program

[Maxar](https://www.maxar.com/open-data) offers high-resolution commercial satellite imagery for free during natural disasters and emergencies in the UK.

**Key Features:**

- Sub-meter resolution imagery during floods and severe weather events.
- Released within 24-48 hours of disaster events.
- Critical for emergency response, particularly during widespread flooding.
- Before-and-after imagery for disaster assessment and recovery planning.
- Partnerships with UK emergency management agencies.

---

## 10. Zoom Earth

[Zoom Earth](https://zoom.earth/) provides frequently updated satellite imagery, focusing on weather patterns and environmental conditions across the UK.

**Key Features:**

- Near-current weather and storm tracking.
- Updated multiple times daily with latest satellite passes.
- Excellent for tracking Atlantic storms, wind patterns, and weather systems affecting the UK.
- Time-lapse animations showing weather development.
- No registration required for basic viewing.

---

## Tips for Making the Most of Free Satellite Imagery Platforms

**1. Find the Right Platforms.**
We've highlighted some of the best satellite data providers for UK applications. Starting with the platforms above will save you time and ensure access to UK-specific datasets from trusted government sources.

**2. Understand Platform Capabilities.**
Each platform offers unique features. Familiarise yourself with their resolution, data quality, and update frequency to match your project's needs. For the frequently cloudy UK climate, consider radar-based Sentinel-1 data which penetrates clouds.

**3. Define Your Requirements.**
Specify what you need — the area of interest, time range, spatial resolution, and spectral bands — and use search filters to narrow down the results. British territories have excellent coverage, so precise targeting is important.

**4. Use Visualisation Tools.**
Many platforms provide built-in tools to help you explore and analyse the data. Pera Portal's spectral index capabilities, for example, can provide insights into crop health, water resources, and land use without needing external software or technical expertise.

**5. Be Selective with Downloads.**
Rather than downloading full datasets, focus on specific tiles or areas of interest to save time and storage space. For many use cases, cloud-based platforms like Pera Portal eliminate the need for downloads entirely.

By following these tips, you'll be able to make the most of the available satellite imagery resources in the UK, whether you're conducting research, evaluating property, monitoring agricultural land, tracking environmental changes, or assessing flood risks.

---

## When Free Satellite Data Isn't Enough

Free platforms cover an enormous range of uses — and if you're simply curious about your own property, they're genuinely all you'll ever need. But if you're using imagery professionally, you'll eventually hit the resolution wall:

| Source               | Resolution  | What you can actually see                                         |
| -------------------- | ----------- | ----------------------------------------------------------------- |
| Landsat (free)       | 30 m        | Regional land cover — a football pitch is roughly one pixel       |
| Sentinel-2 (free)    | 10 m        | Field-scale vegetation patterns; buildings are blurs              |
| Commercial (Geopera) | up to 30 cm | Individual vehicles, fence lines, garden boundaries, single trees |

The difference matters the moment you need to **measure rather than look**: documenting flood extent on a specific date, monitoring construction progress site-by-site, resolving a boundary question, or tracking change at the scale of a single property rather than a parish. Free sensors also can't be tasked — and with the UK's cloud cover, the archive may simply not hold a usable capture of your site for the window you need.

Commercial imagery used to mean opaque quotes and weeks of back-and-forth. We publish [transparent per-square-kilometre pricing](/pricing) for both tasking and archive, and every order arrives analysis-ready — orthorectified, pansharpened, colour-balanced and mosaicked ([here's exactly what that involves](/imagery)).

If free data has taken your project as far as it can go, [explore available imagery through Pera Portal](https://portal.geopera.com/) or [get in touch to discuss your project](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# 10 Best Free Sources of Satellite Data for Canada

> A comprehensive list of the best free satellite data sources in Canada

Published: 2025-10-26 | Author: Darcy Weedman | Reading time: 6 minute

Source: https://geopera.com/blog/free-sources-satellite-data-canada

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- [Pera Portal](https://portal.geopera.com/): Unlimited free Sentinel-2 browsing with 45+ pre-configured spectral indices (NDVI, NDWI, EVI, etc.) viewable in-browser. Advanced band ratio calculations, zero setup required. Global coverage including comprehensive Canadian imagery. Plus access to 100+ commercial satellites for premium imagery when needed

- [Natural Resources Canada Open Data](https://open.canada.ca/en/open-data): Best comprehensive source for Canadian government satellite and geospatial data

- [GeoGratis](https://geogratis.gc.ca/): Best for free downloads of topographic maps, satellite imagery, and aerial photography across Canada

- [Canadian Wildland Fire Information System](https://cwfis.cfs.nrcan.gc.ca/maps/fm): Best for wildfire monitoring and near real-time active fire detection across Canada

- [Google Earth](https://earth.google.com/web): Best for general purpose viewing and basic exploration of your property

- Provincial Resources:
  - [GeoBC](https://www2.gov.bc.ca/gov/content/data/geographic-data-services) (British Columbia)
  - [AltaLIS](https://altalis.com/) (Alberta)
  - [Ontario GeoHub](https://geohub.lio.gov.on.ca/)
    Best for high-resolution provincial imagery and local planning data

- [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/): Best for environmental monitoring and agricultural applications with frequent Sentinel updates

- [NASA Earthdata Search](https://search.earthdata.nasa.gov/): Best for scientific research and comprehensive North American satellite data

- [Maxar Open Data Program](https://www.maxar.com/open-data): Best for disaster response imagery during floods, wildfires, and emergencies

- [Zoom Earth](https://zoom.earth/): Best for near-current weather and environmental condition tracking across Canada

<AskBand headline="What does your part of Canada look like this week?" sub="Free Sentinel-2 with indices in the browser, no credit card." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/glacier.webp', '/band/fields.webp', '/band/delta.webp']} />

## Overview

Looking for satellite images of your house in Canada? Want to view historical satellite imagery of your property for free? You're in the right place.

This guide will help you explore the various sources of satellite data available to Canadians, whether you want to view imagery directly in your browser or download data for further analysis.

### Understanding Real-Time Satellite Imagery

Many people search for "live satellite view" or "real-time satellite images" of their house, but here's the truth — despite what movies might suggest, live satellite imagery isn't possible with current technology. What you see on mapping platforms is usually imagery captured weeks or even months ago.

### What's Available in Canada

If you're searching for free satellite imagery in Canada, it's important to understand what's available:

- Most free satellite images are updated every few months to years.
- Free imagery generally has lower resolution (you can see buildings, but not detailed features).
- High-resolution imagery with clear details often requires a paid subscription.
- Real-time satellite imagery does exist but is typically limited to weather forecasting or emergency services.

Fortunately, Canada offers several excellent free sources of satellite data through federal and provincial government platforms. Whether you are a property owner, researcher, or just curious, we'll walk you through the best free and paid options available.

---

## 1. Natural Resources Canada Open Data

[Natural Resources Canada Open Data](https://open.canada.ca/en/open-data) is the primary gateway to Canadian government satellite imagery and geospatial datasets. It provides comprehensive access to Earth observation data collected over Canada.

**Key Features:**

- Free access to RADARSAT-2 and RCM (RADARSAT Constellation Mission) satellite imagery.
- Landsat data specifically processed for Canadian territories.
- Historical archives dating back decades.
- Advanced search by location, date, and satellite sensor.
- Ideal for researchers, environmental monitoring, and natural resource management.

---

## 2. GeoGratis

[GeoGratis](https://geogratis.gc.ca/) is Natural Resources Canada's free data distribution service offering satellite imagery, aerial photography, and topographic maps.

**Key Features:**

- Free downloads of Landsat imagery covering all of Canada.
- High-resolution aerial photography for many populated regions.
- Digital elevation models and terrain data.
- No registration required for many datasets.
- Comprehensive coverage of northern and remote Canadian territories.

---

## 3. Pera Portal

Our [Pera Portal](https://portal.geopera.com/) provides completely free, unlimited access to Sentinel-2 satellite imagery with advanced analytical capabilities that surpass most free platforms.

**Key Features:**

- **Unlimited free Sentinel-2 browsing** with no registration walls or download limits.
- **45+ pre-configured spectral indices** viewable instantly in your browser:
  - NDVI (vegetation health and forestry monitoring)
  - NDWI (water detection and lake monitoring)
  - EVI (enhanced vegetation index for Canadian agriculture)
  - SAVI, MSAVI, NDBI, and many more
- **Custom band ratio calculations** — create your own indices for specialized analysis.
- **Zero setup required** — view processed imagery immediately without downloading or installing software.
- **Global coverage** including comprehensive Canadian imagery updated every 5 days.
- **Ideal for Canadian applications**:
  - Agricultural monitoring across prairie provinces
  - Forestry management in BC and Quebec
  - Mining exploration and monitoring
  - Northern territory environmental tracking
- **Premium commercial imagery access** — upgrade to browse 100+ commercial satellites with sub-meter resolution when you need more detail.

Unlike other platforms that require technical knowledge or large downloads, Pera Portal makes advanced satellite imagery analysis accessible to everyone — from prairie farmers monitoring crop health to researchers tracking permafrost changes in the Arctic.

---

## 4. Canadian Wildland Fire Information System (CWFIS)

The [Canadian Wildland Fire Information System](https://cwfis.cfs.nrcan.gc.ca/maps/fm) provides near real-time satellite-based fire detection and monitoring across Canada.

**Key Features:**

- Active fire detection updated multiple times daily.
- Historical fire perimeter mapping.
- Satellite-based smoke and air quality monitoring.
- Essential for wildfire season monitoring, especially in BC, Alberta, and northern territories.
- Free access to fire danger ratings and fire weather data.

---

## 5. Google Earth

[Google Earth](https://earth.google.com/web) provides high-resolution satellite imagery across Canada, including remote northern regions. It's the easiest platform for casual users who want to view satellite images of their house.

**Key Features:**

- Simple, intuitive interface accessible from any web browser.
- Historical imagery slider to view changes over time.
- 3D terrain visualization perfect for mountainous regions like the Rockies.
- Coverage of remote northern communities and territories.
- Not suitable for downloading raw satellite data or technical analysis.

---

## 6. Provincial GIS Platforms

Canadian provinces provide their own high-resolution imagery and geospatial data, often with more frequent updates than federal sources:

### British Columbia - GeoBC

[GeoBC](https://www2.gov.bc.ca/gov/content/data/geographic-data-services) offers comprehensive geospatial data for BC including satellite imagery, aerial photography, and forestry data.

### Alberta - AltaLIS

[AltaLIS](https://altalis.com/) provides satellite imagery and geospatial data for Alberta, with emphasis on agricultural and oil & gas applications.

### Ontario - GeoHub

[Ontario GeoHub](https://geohub.lio.gov.on.ca/) offers satellite imagery, land use data, and property boundaries for Ontario.

### Other Provincial Resources

- **Quebec**: [Données Québec](https://www.donneesquebec.ca/)
- **Saskatchewan**: [ISC GeoWarehouse](https://www.isc.ca/Pages/default.aspx)
- **Manitoba**: [Manitoba Land Initiative](https://mli2.gov.mb.ca/)

---

## 7. Copernicus Data Space Ecosystem

The European Union's [Copernicus](https://browser.dataspace.copernicus.eu/) program offers free, high-quality Sentinel satellite data with global coverage including all of Canada.

**Key Features:**

- Sentinel-2 optical imagery updated every 5 days with 10-meter resolution.
- Sentinel-1 radar data for all-weather monitoring (crucial for cloudy Canadian conditions).
- Free access without download limits or registration barriers.
- Ideal for agricultural monitoring, forestry, and environmental research.
- Particularly valuable for monitoring Canada's northern regions.

---

## 8. NASA Earthdata Search

[NASA Earthdata Search](https://search.earthdata.nasa.gov/) provides comprehensive satellite datasets with excellent coverage of North America including Canada.

**Key Features:**

- Complete Landsat archive dating back to 1972.
- MODIS data for large-scale environmental monitoring.
- Free registration and unlimited downloads.
- Advanced search filters by location, date, and cloud cover.
- Essential for scientific research and climate studies.

---

## 9. Maxar Open Data Program

[Maxar](https://www.maxar.com/open-data) offers high-resolution commercial satellite imagery for free during natural disasters and emergencies in Canada.

**Key Features:**

- Sub-meter resolution imagery during floods, wildfires, and other disasters.
- Released within 24-48 hours of disaster events.
- Critical for emergency response in remote northern communities.
- Before-and-after imagery for disaster recovery planning.
- Partnerships with Canadian emergency management agencies.

---

## 10. Zoom Earth

[Zoom Earth](https://zoom.earth/) provides frequently updated satellite imagery, focusing on weather patterns and environmental conditions across Canada.

**Key Features:**

- Near-current weather and storm tracking.
- Updated multiple times daily with latest satellite passes.
- Excellent for tracking severe weather, winter storms, and atmospheric rivers.
- Time-lapse animations showing weather system development.
- No registration required for basic viewing.

---

## Tips for Making the Most of Free Satellite Imagery Platforms

**1. Find the Right Platforms.**
We've highlighted some of the best satellite data providers for Canadian applications. Starting with the platforms above will save you time and ensure access to Canada-specific datasets.

**2. Understand Platform Capabilities.**
Each platform offers unique features. Familiarise yourself with their resolution, data quality, and update frequency to match your project's needs. Consider weather patterns — radar-based Sentinel-1 data is often more useful than optical imagery in cloudy regions.

**3. Define Your Requirements.**
Specify what you need — the area of interest, time range, spatial resolution, and spectral bands — and use search filters to narrow down the results. Canadian territories are vast, so precise targeting saves time.

**4. Use Visualisation Tools.**
Many platforms provide built-in tools to help you explore and analyse the data. Pera Portal's spectral index capabilities, for example, can provide insights into vegetation health, water resources, and land use without needing external software or technical expertise.

**5. Be Selective with Downloads.**
Rather than downloading full datasets, focus on specific tiles or areas of interest to save time and storage space. For many use cases, cloud-based platforms like Pera Portal eliminate the need for downloads entirely.

By following these tips, you'll be able to make the most of the available satellite imagery resources in Canada, whether you're conducting research, evaluating property, monitoring agricultural land, or tracking environmental changes across this vast country.

---

## When Free Satellite Data Isn't Enough

Free platforms cover an enormous range of uses — and if you're simply curious about your own property, they're genuinely all you'll ever need. But if you're using imagery professionally, you'll eventually hit the resolution wall:

| Source               | Resolution  | What you can actually see                                     |
| -------------------- | ----------- | ------------------------------------------------------------- |
| Landsat (free)       | 30 m        | Regional land cover — a hockey rink is well under one pixel   |
| Sentinel-2 (free)    | 10 m        | Field-scale vegetation patterns; buildings are blurs          |
| Commercial (Geopera) | up to 30 cm | Individual vehicles, cut-block edges, equipment, single trees |

The difference matters the moment you need to **measure rather than look**: auditing harvest blocks against a forestry plan, tracking earthworks and stockpiles on a mine site, monitoring remote assets you can't economically fly, or proving site conditions on a specific date. Free sensors also can't be tasked — with short northern capture seasons, if the archive doesn't hold a cloud-free pass of your site, there's nothing to download.

Commercial imagery used to mean opaque quotes and weeks of back-and-forth. We publish [transparent per-square-kilometre pricing](/pricing) for both tasking and archive, and every order arrives analysis-ready — orthorectified, pansharpened, colour-balanced and mosaicked ([here's exactly what that involves](/imagery)).

If free data has taken your project as far as it can go, [explore available imagery through Pera Portal](https://portal.geopera.com/) or [get in touch to discuss your project](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# 10 Best Free Sources of Satellite Data for the USA

> A comprehensive list of the best free satellite data sources in the United States

Published: 2025-10-25 | Author: Darcy Weedman | Reading time: 6 minute

Source: https://geopera.com/blog/free-sources-satellite-data-usa

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- [Pera Portal](https://portal.geopera.com/): Unlimited free Sentinel-2 browsing with 45+ pre-configured spectral indices (NDVI, NDWI, EVI, etc.) viewable in-browser. Advanced band ratio calculations, zero setup required. Global coverage including comprehensive USA imagery. Plus access to 100+ commercial satellites for premium imagery when needed

- [USGS EarthExplorer](https://earthexplorer.usgs.gov/): Best comprehensive archive for Landsat, aerial imagery, and historical satellite data across the USA

- [NASA Worldview](https://worldview.earthdata.nasa.gov/): Best for near real-time satellite imagery and tracking environmental changes like wildfires and floods

- [NOAA Data Access Viewer](https://coast.noaa.gov/dataviewer/): Best for coastal zones, ocean monitoring, and weather-related satellite data

- [Google Earth](https://earth.google.com/web): Best for general purpose viewing and basic exploration of your property

- [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/): Best for environmental monitoring and agricultural applications with frequent Sentinel updates

- [Zoom Earth](https://zoom.earth/): Best for near-current weather and environmental condition tracking across the USA

- [Maxar Open Data Program](https://www.maxar.com/open-data): Best for disaster response imagery during hurricanes, wildfires, and other emergencies

- [State & Local GIS Portals](https://www.census.gov/geographies/mapping-files.html): Best for high-resolution local imagery and property boundary information

- [USDA CropScape](https://nassgeodata.gmu.edu/CropScape/): Best for agricultural land use data and crop classification across the USA

<AskBand headline="What does your patch of the US look like this week?" sub="Free Sentinel-2 with indices in the browser, no credit card." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/pivots.webp', '/band/city.webp', '/band/delta.webp']} />

## Overview

Looking for satellite images of your house in the United States? Want to view historical satellite imagery of your property for free? You're in the right place.

This guide will help you explore the various sources of satellite data available in the USA, whether you want to view imagery directly in your browser or download data for further analysis.

### Understanding Real-Time Satellite Imagery

Many people search for "live satellite view" or "real-time satellite images" of their house, but here's the truth — despite what movies might suggest, live satellite imagery isn't possible with current technology. What you see on mapping platforms is usually imagery captured weeks or even months ago.

### What's Available in the USA

If you're searching for free satellite imagery in the USA, it's important to understand what's available:

- Most free satellite images are updated every few months to years.
- Free imagery generally has lower resolution (you can see buildings, but not detailed features).
- High-resolution imagery with clear details often requires a paid subscription.
- Real-time satellite imagery does exist but is typically limited to weather forecasting or emergency services.

Fortunately, the United States offers several excellent free sources of satellite data, including world-leading NASA and USGS platforms. Whether you are a property owner, researcher, or just curious, we'll walk you through the best free and paid options available.

---

## 1. USGS EarthExplorer

[USGS EarthExplorer](https://earthexplorer.usgs.gov/) is the premier platform for accessing free satellite imagery in the United States. Operated by the U.S. Geological Survey, it provides comprehensive access to Landsat data, aerial imagery, and historical archives.

**Key Features:**

- Complete Landsat archive dating back to 1972 — over 50 years of historical imagery.
- Free download of full-resolution satellite data.
- Advanced search filters by location, date, cloud cover, and satellite sensor.
- Access to ASTER, MODIS, and aerial photography collections.
- Ideal for researchers, environmental monitoring, and land use analysis.

---

## 2. NASA Worldview

[NASA Worldview](https://worldview.earthdata.nasa.gov/) offers near real-time satellite imagery from NASA's Earth Observing System. It's exceptionally useful for tracking current environmental events like wildfires, hurricanes, and air quality.

**Key Features:**

- Near real-time imagery updated within hours of satellite passes.
- Interactive visualization of atmospheric conditions, fires, and weather patterns.
- Time-slider to view historical changes over days, months, or years.
- Multiple satellite sensors including MODIS, VIIRS, and Landsat.
- No registration required for basic browsing and visualization.

---

## 3. Pera Portal

Our [Pera Portal](https://portal.geopera.com/) provides completely free, unlimited access to Sentinel-2 satellite imagery with advanced analytical capabilities that surpass most free platforms.

**Key Features:**

- **Unlimited free Sentinel-2 browsing** with no registration walls or download limits.
- **45+ pre-configured spectral indices** viewable instantly in your browser:
  - NDVI (vegetation health)
  - NDWI (water detection)
  - EVI (enhanced vegetation index)
  - SAVI, MSAVI, NDBI, and many more
- **Custom band ratio calculations** — create your own indices for specialized analysis.
- **Zero setup required** — view processed imagery immediately without downloading or installing software.
- **Global coverage** including comprehensive USA imagery updated every 5 days.
- **Premium commercial imagery access** — upgrade to browse 100+ commercial satellites with sub-meter resolution when you need more detail.

Unlike other platforms that require technical knowledge or large downloads, Pera Portal makes advanced satellite imagery analysis accessible to everyone — from farmers monitoring crop health to researchers tracking urban development.

---

## 4. NOAA Data Access Viewer

The [NOAA Data Access Viewer](https://coast.noaa.gov/dataviewer/) is the go-to platform for coastal and oceanic satellite imagery, operated by the National Oceanic and Atmospheric Administration.

**Key Features:**

- Comprehensive coverage of U.S. coastal zones and Great Lakes.
- LiDAR elevation data and bathymetric (underwater) mapping.
- Storm surge modeling and coastal flood risk data.
- Frequent updates for weather monitoring and marine applications.
- Free download of high-quality datasets for coastal research.

---

## 5. Google Earth

[Google Earth](https://earth.google.com/web) provides high-resolution satellite imagery across the United States. It's the easiest platform for casual users who want to view satellite images of their house without technical complexity.

**Key Features:**

- Simple, intuitive interface accessible from any web browser.
- Historical imagery slider to view changes over time (dating back to the 1980s in some areas).
- 3D terrain visualization and building models.
- Integration with Street View for ground-level perspectives.
- Not suitable for downloading raw satellite data or technical analysis.

---

## 6. Copernicus Data Space Ecosystem

The European Union's [Copernicus](https://browser.dataspace.copernicus.eu/) program offers free, high-quality Sentinel satellite data with global coverage including the entire United States.

**Key Features:**

- Sentinel-2 imagery updated every 5 days with 10-meter resolution.
- Sentinel-1 radar data for all-weather monitoring.
- Free access without download limits or registration barriers.
- Ideal for agricultural monitoring, forestry, and environmental research.
- Cloud-based visualization and download options.

---

## 7. Zoom Earth

[Zoom Earth](https://zoom.earth/) provides frequently updated satellite imagery, focusing on weather patterns and environmental conditions across the USA.

**Key Features:**

- Near-current weather and storm tracking.
- Updated multiple times daily with latest satellite passes.
- Simple interface for tracking hurricanes, wildfires, and severe weather.
- Time-lapse animations showing storm development.
- No registration required for basic viewing.

---

## 8. Maxar Open Data Program

[Maxar](https://www.maxar.com/open-data) offers high-resolution commercial satellite imagery for free during natural disasters and emergencies in the United States.

**Key Features:**

- Sub-meter resolution imagery during hurricanes, wildfires, and floods.
- Released within 24-48 hours of disaster events.
- Critical for emergency response and damage assessment.
- Partnerships with FEMA and humanitarian organizations.
- Before-and-after imagery for disaster recovery planning.

---

## 9. State & Local GIS Portals

Many U.S. states and counties provide their own high-resolution aerial imagery and satellite data through dedicated GIS portals:

- **California**: [Cal-Atlas](https://atlas.ca.gov/)
- **Texas**: [Texas Natural Resources Information System](https://tnris.org/)
- **Florida**: [Florida Geographic Data Library](https://www.fgdl.org/)
- **New York**: [NYS GIS Clearinghouse](https://gis.ny.gov/)
- **Colorado**: [Colorado Information Marketplace](https://data.colorado.gov/)

Many counties also offer aerial imagery through their assessor or planning department websites — often with higher resolution and more frequent updates than national platforms.

---

## 10. USDA CropScape

[USDA CropScape](https://nassgeodata.gmu.edu/CropScape/) is a specialized platform from the U.S. Department of Agriculture providing satellite-derived crop data and agricultural land use information.

**Key Features:**

- Annual Cropland Data Layer (CDL) covering all agricultural regions of the USA.
- Detailed crop type classification (corn, soybeans, wheat, cotton, etc.).
- Historical data dating back to 1997 for tracking agricultural trends.
- Free downloads of raster and vector data for analysis.
- Interactive web interface for visualizing crop patterns.
- Essential for farmers, agricultural researchers, and land use planners.
- Updated annually with the latest growing season data.

---

## Tips for Making the Most of Free Satellite Imagery Platforms

**1. Find the Right Platforms.**
We've highlighted some of the best satellite data providers, but there are many more out there. Starting with the platforms above will save you time and ensure a smoother experience.

**2. Understand Platform Capabilities.**
Each platform offers unique features. Familiarise yourself with their resolution, data quality, and update frequency to match your project's needs.

**3. Define Your Requirements.**
Specify what you need — the area of interest, time range, spatial resolution, and spectral bands — and use search filters to narrow down the results.

**4. Use Visualisation Tools.**
Many platforms provide built-in tools to help you explore and analyse the data. Pera Portal's spectral index capabilities, for example, can provide deeper insights without needing external software or technical expertise.

**5. Be Selective with Downloads.**
Rather than downloading full datasets, focus on specific tiles or areas of interest to save time and storage space. For many use cases, cloud-based platforms like Pera Portal eliminate the need for downloads entirely.

By following these tips, you'll be able to make the most of the available satellite imagery resources in the USA, whether you're conducting research, evaluating property, or monitoring environmental changes.

---

## When Free Satellite Data Isn't Enough

Free platforms cover an enormous range of uses — and if you're simply curious about your own property, they're genuinely all you'll ever need. But if you're using imagery professionally, you'll eventually hit the resolution wall:

| Source               | Resolution  | What you can actually see                                   |
| -------------------- | ----------- | ----------------------------------------------------------- |
| Landsat (free)       | 30 m        | Regional land cover — a football field is roughly one pixel |
| Sentinel-2 (free)    | 10 m        | Field-scale vegetation patterns; buildings are blurs        |
| Commercial (Geopera) | up to 30 cm | Individual vehicles, fence lines, equipment, single trees   |

The difference matters the moment you need to **measure rather than look**: documenting construction progress for a draw inspection, assessing storm damage parcel-by-parcel for claims, monitoring crops by the row rather than by the county, or proving site conditions on a specific date. Free sensors also can't be tasked — if no satellite happened to capture your site cloud-free in the window you care about, there's nothing to download.

Commercial imagery used to mean opaque quotes and weeks of back-and-forth. We publish [transparent per-square-kilometre pricing](/pricing) for both tasking and archive, and every order arrives analysis-ready — orthorectified, pansharpened, color-balanced and mosaicked ([here's exactly what that involves](/imagery)).

If free data has taken your project as far as it can go, [explore available imagery through Pera Portal](https://portal.geopera.com/) or [get in touch to discuss your project](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Understanding DEMs, DSMs, and DTMs: A Complete Guide to Digital Elevation Data

> Learn the key differences between DEMs, DSMs, and DTMs, how they're created, their applications, and how to choose the right elevation data for your project.

Published: 2025-10-23 | Author: Darcy Weedman | Reading time: 14 min

Source: https://geopera.com/blog/understanding-dems-dsms-dtms-guide

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- **What DEMs are:** Raster grids representing Earth's surface elevation, where each pixel contains an elevation value normalized to a datum (typically mean sea level)
- **Three types:** DEM is an umbrella term encompassing Digital Surface Models (DSMs) that include all objects, and Digital Terrain Models (DTMs) that show only bare-earth
- **Creation methods:** InSAR (radar interferometry), stereo photogrammetry (multiple optical angles), LiDAR (laser scanning), contour digitization, and ground surveys
- **Key applications:** Flood modeling, urban planning, infrastructure design, mining volumetrics, hydrology analysis, environmental monitoring, and disaster response
- **Accuracy factors:** Spatial resolution, vertical resolution, acquisition method, temporal currency, and processing algorithms all affect DEM quality
- **Data sources:** Free global datasets (SRTM, ASTER, ALOS) serve many applications, while commercial high-resolution options deliver sub-meter accuracy for critical projects
- **Need high-quality DEMs?** [Contact Geopera](https://geopera.com/contact) to access stereoscopic satellite imagery for custom DEM generation or discuss which elevation data source best fits your project requirements

Our planet is full of peaks, valleys, natural habitats, and human-made structures. Digital elevation data brings these highs, lows, and features into sharp focus.

By visualizing landscapes as elevation data, you can estimate areas vulnerable to sea-level rise, detect vegetation encroachment, identify optimal infrastructure routes, and solve problems ranging from flood risk assessment to mine planning.

There are multiple ways to model elevation, and the terminology can be confusing. The term Digital Elevation Model (DEM) is often used as a blanket term, but it encompasses distinct product types—Digital Surface Models (DSMs) and Digital Terrain Models (DTMs)—each serving different purposes.

Understanding these differences is crucial for selecting the right elevation data for your application.

<AskBand headline="Surface or terrain: which model do you need?" sub="When free elevation is not enough, tri-stereo capture is priced per AOI." action="See elevation products" href="/imagery" tiles={['/band/terraces.webp', '/band/hills.webp', '/band/caldera.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## What Are Digital Elevation Models?

A Digital Elevation Model, also known as a DEM, is a raster GIS layer representing the Earth's surface as a pixel grid. Each pixel in this grid contains an elevation value that indicates the height of that location above a reference point.

These elevation values are normalized to a vertical datum—a standardized reference for zero elevation, typically based on mean sea level. This consistent reference point allows elevation data from different sources and times to be compared and analyzed together.

The term DEM is commonly used as an umbrella term encompassing both Digital Surface Models and Digital Terrain Models, which we'll explore in detail below. In practice, when working with coarser-resolution global datasets that can't distinguish individual buildings or trees, the generic term DEM is most appropriate.

DEMs are stored in various file formats including GeoTIFF (.tif), the USGS DEM format (.dem), binary float files (.flt), and text-based ASCII grids (.asc). The GeoTIFF format has become the de facto standard due to its wide software support and ability to embed georeferencing information directly in the file.

## How Are DEMs Created?

To capture elevation data, a single standard optical image isn't sufficient. You need a way to capture depth. Here are the primary methods used to create digital elevation models:

### SAR Interferometry (InSAR)

This method uses synthetic aperture radar (SAR) to collect multiple radar images of the same area from different angles, building a 3D representation through phase difference analysis.

SAR sensors emit microwave pulses and measure the reflected signal. By comparing the phase differences between two radar images captured from slightly different positions, interferometric algorithms can calculate the distance from the sensor to each point on the ground—revealing elevation.

The key advantage of InSAR is its ability to work in all weather conditions, day or night, since radar penetrates clouds. The Shuttle Radar Topography Mission (SRTM) used this technique during its 11-day mission in 2000 to map most of Earth's landmass.

### Stereo Photogrammetry

Similar to InSAR but using optical sensors, stereo photogrammetry captures multiple images of the same area from different angles to extract 3D information. This can be accomplished using satellite, aerial, or drone-based platforms.

![Stereoscopic satellite imagery showing dual capture angles](/blogs/stereo-imagery.webp)
_Stereoscopic collection captures the same location from two different viewing angles, creating the geometric conditions necessary to extract precise elevation data._

Modern high-resolution satellites like WorldView-2 and WorldView-3 excel at stereo DEM generation through in-track stereo collection, where both images are captured during a single orbital pass within 45-90 seconds of each other. This temporal consistency—where ground conditions remain nearly identical between captures—improves the reliability of photogrammetric matching and the resulting elevation accuracy.

Learn more about how [stereoscopic satellite imagery](/blog/stereoscopic-satellite-imagery) creates accurate 3D elevation models from paired 2D images.

### LiDAR (Light Detection and Ranging)

LiDAR sensors actively fire laser pulses and measure the reflected light to determine the elevation of Earth's surface. Instead of producing a raster grid directly, LiDAR typically creates a dense point cloud—millions of precisely positioned elevation measurements.

These point clouds can be converted into raster grids for easier processing and integration with other GIS data. LiDAR offers the highest spatial and vertical accuracy of any elevation capture method, with centimeter-level precision achievable under ideal conditions.

The trade-off is cost. While LiDAR is cost-effective for small critical areas, it becomes prohibitively expensive for large-scale regional mapping compared to satellite-based methods.

### Digitizing Contour Lines

By using an existing topographic map with contour lines, you can extract elevation data using GIS software. This legacy method interpolates a continuous surface from the discrete contour elevations shown on the map.

While less accurate than modern sensing techniques, contour digitization remains useful for historical analysis and areas where no other elevation data exists.

### Ground Surveying

Traditional surveying with theodolites and differential GPS involves assessing known XYZ positions and systematically measuring neighboring areas. This method requires skilled labor, substantial time investment, and very careful inputs to maintain accuracy.

Ground surveying delivers excellent accuracy but is only practical for small, critical areas where the investment is justified—such as engineering-grade surveys for major infrastructure projects.

## What Are Digital Surface Models (DSMs)?

A Digital Surface Model captures the surface along with all natural and human-made structures—vegetation, buildings, power lines, and any other objects elevated above the ground.

In short: DSMs represent the ground AND everything on it.

![Comparison between Digital Terrain Models (DTM) and Digital Surface Models (DSM)](/dtm-vs-dsm.webp)
_Figure: Comparison between DTM and DSM. DTMs represent the bare earth surface without buildings and vegetation, while DSMs include all objects on the terrain. Source: heliguy.com_

DSMs capture the "first return" surface—the top of whatever is present when looking straight down from above. In a forest, this would be the tree canopy. In a city, this would be building rooftops. Only in open areas does the DSM show the actual ground surface.

### Common Applications of DSMs

Because DSMs represent the surface and all of its above-ground features, they're particularly valuable for:

**Urban planning and 3D city modeling:** DSMs provide the foundation for creating accurate 3D representations of cities, essential for visualization, planning, and analysis of urban development.

**Runway approach zone analysis in aviation:** Detecting the safest landing routes for aircraft requires knowing the height of all obstacles—buildings, trees, towers—in the approach corridor. DSMs excel at this application.

**Building height extraction:** By comparing a DSM with a bare-earth DTM of the same area, you can calculate precise building heights across entire urban areas.

**Vegetation canopy height mapping:** The difference between DSM and DTM reveals canopy height, critical for forestry management and biomass estimation.

**Disaster management and damage assessment:** Post-disaster DSMs can be compared with pre-event data to identify collapsed buildings, vegetation loss, and infrastructure damage.

**Telecommunications line-of-sight planning:** Planning microwave links and radio networks requires knowing exactly what obstacles exist between transmission points—a job perfectly suited to DSMs.

**Solar panel placement optimization:** DSMs combined with sun angle calculations identify roof areas with optimal solar exposure while accounting for shadowing from nearby buildings and trees.

## What Are Digital Terrain Models (DTMs)?

Digital Terrain Models represent the bare-earth surface without natural or man-made features. All above-ground objects—trees, buildings, vehicles, power lines—are removed through processing, revealing only the underlying terrain.

The distinction between DTMs and DSMs becomes most evident in heavily developed urban areas with high-rise buildings. Places like Manhattan or Hong Kong Island have dramatically different surface and terrain elevations—the DSM captures skyscraper tops hundreds of meters above the DTM's ground surface.

### How DTMs Differ from DSMs

A critical point: DTMs can be derived from DSMs, but not the other way around.

Creating a DTM from a DSM requires sophisticated classification algorithms that identify and remove above-ground features while preserving the ground surface. This is considerably more complex than simply capturing the surface, which is why DTM production is more computationally intensive and often more expensive than DSM generation.

### The nDSM Concept

The combination of DSM and DTM creates height information for above-ground features, known as a normalized Digital Surface Model (nDSM). The math is simple and intuitive:

**DSM - DTM = nDSM**

This normalized model shows the height of objects above the ground. A tree might appear as 15 meters in the nDSM, while a building shows its actual structural height rather than its absolute elevation.

nDSMs are particularly useful for:

- **Forestry applications:** Calculating tree heights across large forest areas
- **Urban planning:** Extracting building heights for zoning compliance and 3D modeling
- **Vegetation management:** Identifying power line encroachment risks
- **Change detection:** Tracking growth of vegetation or construction of new structures

### Common Applications of DTMs

DTMs shine in applications where bare-earth elevation is essential:

**Hydrology and watershed analysis:** Water flows across the ground surface, not over tree canopies or building roofs. Accurate flow direction and accumulation modeling requires bare-earth DTMs.

**Infrastructure routing:** Planning roads, railways, pipelines, and power transmission lines requires knowing the actual ground elevation to calculate cut-and-fill volumes, assess slope stability, and optimize routes.

**Flood modeling:** Predicting flood inundation extents demands bare-earth elevation data. A DSM would incorrectly suggest that buildings block water flow, when in reality water flows around and through urban areas at ground level.

**Geomorphology studies:** Understanding landforms, erosion patterns, and geological processes requires seeing the actual terrain without the visual clutter of surface features.

**Archaeological feature detection:** Subtle terrain variations that indicate buried structures or historical earthworks are only visible in bare-earth DTMs, particularly those derived from LiDAR that penetrates vegetation.

**Terrain-based military planning:** Ground movement analysis, line-of-sight calculations for ground observers, and trafficability assessment all require accurate bare-earth terrain models.

## DEM vs DSM vs DTM: Key Differences

Understanding the distinctions between these elevation model types is essential for selecting the right product for your application:

| Feature                   | DEM                              | DSM                   | DTM                     |
| ------------------------- | -------------------------------- | --------------------- | ----------------------- |
| **Definition**            | Umbrella term for elevation data | Surface + all objects | Bare-earth only         |
| **Includes vegetation**   | Depends on product type          | Yes                   | No                      |
| **Includes buildings**    | Depends on product type          | Yes                   | No                      |
| **Best for hydrology**    | Only if it's a DTM               | No                    | Yes                     |
| **Best for urban 3D**     | Only if it's a DSM               | Yes                   | No                      |
| **Derivation**            | -                                | Direct from sensors   | Requires classification |
| **Processing complexity** | -                                | Lower                 | Higher                  |

### Quick Recap

- **DEM:** Generic term that could refer to either a DSM or DTM. Often used for coarse-resolution data where the distinction doesn't matter.
- **DSM:** Everything visible from above—the canopy model showing tree tops, building roofs, and any other elevated features.
- **DTM:** Ground surface only, with all features removed through classification processing.

In practice, low-resolution global datasets (30m pixel size) don't meaningfully distinguish between DSM and DTM because they can't resolve individual trees and buildings anyway. The distinction becomes critical when working with high-resolution data where above-ground features are clearly visible.

## DEM Quality and Accuracy

DEM accuracy is most commonly estimated by calculating the Root Mean Square Error (RMSE) of elevation—comparing DEM-derived heights against reference points with known elevations measured through precise ground surveys.

However, there's much more to DEM quality than just vertical accuracy. Horizontal positional accuracy matters too, particularly when overlaying DEMs with other spatial data like imagery or vector features.

### Factors Affecting DEM Quality

The quality of any digital elevation model depends on multiple interrelated factors:

#### 1. Acquisition Method

Different elevation capture methods produce vastly different accuracy levels:

- **LiDAR:** Best vertical accuracy, with 5-15cm precision achievable under ideal conditions. The active laser approach and dense point cloud output make this the gold standard for high-accuracy applications.

- **Stereo photogrammetry:** Good accuracy, typically 30cm to 1m vertical precision with high-resolution satellites like WorldView-3. Accuracy depends heavily on image resolution, convergence angle between stereo pairs, and ground control points.

- **InSAR:** Moderate accuracy, ranging from several meters to tens of meters depending on radar wavelength, baseline distance, and processing techniques. Excellent for large-area mapping where meter-level accuracy is sufficient.

- **Ground survey:** Excellent accuracy but spatially limited. Practical only for small critical areas requiring engineering-grade precision.

#### 2. Spatial Resolution

Spatial resolution is determined by the distance between sample points. This can be relatively uniform in stereo imagery, somewhat uniform in radar and LiDAR datasets, or highly variable in DEMs created from manual surveying methods.

Coarser resolution means small terrain features can't be captured. A 30m resolution DEM will completely miss a 5m wide gully, while a 1m resolution model will capture it clearly.

#### 3. Vertical Resolution

The vertical resolution of elevation data defines the possible height difference between the modeled elevation and the actual ground-truthed elevation of the surface.

This is often the most critical specification for applications like infrastructure design, where even small elevation errors can affect drainage patterns, cut-and-fill calculations, and slope stability assessments.

As a general hierarchy: LiDAR produces the best vertical resolution, followed by high-resolution stereo photogrammetry, then InSAR and lower-resolution optical stereo methods.

#### 4. Temporal Resolution

One consideration often overlooked: How recently was the elevation data acquired?

This temporal resolution becomes particularly relevant for change detection applications or when using DSMs to study temporally variable features like vegetation growth or new construction. A 2000-vintage SRTM DEM won't reflect the massive urban development that's occurred in many cities over the past 25 years.

#### 5. Error Sources

Multiple factors introduce errors during DEM acquisition and processing:

- **Atmospheric and ionospheric interference:** Affects radar and optical satellite observations, introducing phase delays (SAR) or image distortions (optical)
- **Cloud cover:** Completely blocks optical methods, creating data voids in stereo-derived DEMs
- **Vegetation penetration:** LiDAR often penetrates vegetation to reach the ground, while photogrammetry sees only the canopy top
- **Co-registration errors:** Misalignment between stereo image pairs degrades elevation extraction accuracy
- **Processing artifacts:** Interpolation errors, edge effects, and algorithm limitations introduce systematic biases

### Vertical Errors in DEMs

Vertical errors typically manifest as two distinct feature types:

**Sinks (also called depressions or pits):** Areas surrounded by higher elevation values, representing internal drainage that may or may not be real. Some sinks are natural features, particularly in glacial or karst landscapes, but many result from DEM imperfections.

**Peaks (also called spikes):** Areas surrounded by cells of lower elevation. These are usually natural features and are less problematic for analysis than sinks.

The prevalence of sinks increases with coarser resolution. Often 1% of cells in a 30-meter resolution DEM consist of artificial sinks, increasing to 5% or more in three arc-second (approximately 90m) datasets.

Another common error type is **striping artifacts**—systematic patterns resulting from sampling errors during DEM creation. These are most noticeable in flat areas when elevation is stored as integer values rather than floating-point numbers.

### Creating Depressionless DEMs

When faced with sinks in DEMs, it's often necessary to remove or fill them to create a depressionless DEM. This is particularly critical for hydrological applications.

A DEM free of artificial sinks serves as the proper input for flow direction processing. Artificial sinks trap simulated water flow, leading to erroneous flow-direction rasters and incorrect watershed delineation.

Most GIS applications include tools to:

- Identify sinks and their spatial distribution
- Fill sinks to create continuous flow surfaces
- Calculate sink depth to distinguish real depressions from artifacts

This pre-processing step is essential before conducting any hydrological analysis with elevation data.

## Common DEM Applications

DEMs serve critical functions across nearly every industry that works with spatial data. The vertical dimension unlocks analyses impossible with 2D imagery alone.

### 1. Hydrology and Flood Modeling

Water flows downhill—a simple fact that makes elevation data fundamental to all hydrological analysis.

DEMs enable:

- **Flow direction and accumulation analysis:** Determining the path water takes across a landscape and which areas contribute flow to specific points
- **Watershed delineation:** Automatically extracting drainage basin boundaries based on terrain
- **Flood inundation mapping:** Modeling which areas would flood under various water level scenarios, critical for sea-level rise planning and storm surge prediction
- **Stream network extraction:** Deriving channel networks from flow accumulation patterns

Accurate flood models require high-resolution, bare-earth DTMs. Research has shown that current global DEMs often fail to capture topographic details in floodplains, leading to significant errors in flood extent predictions. For critical flood risk assessment, commercial high-resolution DEMs or LiDAR-derived DTMs deliver the accuracy needed.

As our climate changes and sea levels rise, elevation-based coastal vulnerability assessment has become increasingly important for planning adaptation strategies.

### 2. Infrastructure Planning

Elevation data is fundamental to infrastructure route optimization and design.

DEMs support:

- **Road, rail, and pipeline route optimization:** Finding paths that minimize cut-and-fill requirements while avoiding excessive slopes
- **Cut-and-fill volume calculations:** Precisely quantifying earthwork requirements for construction planning and cost estimation
- **Line-of-sight analysis:** Determining visibility for telecommunications towers, ensuring clear signal paths
- **Slope stability assessment:** Identifying high-risk terrain where landslides or erosion could threaten infrastructure

Avoiding high-slope areas reduces both construction costs and environmental impacts. The ability to analyze thousands of potential route alternatives using elevation data leads to more efficient, sustainable infrastructure.

Explore how Geopera supports [infrastructure](/infrastructure) applications with high-resolution satellite imagery and elevation data.

### 3. Mining Operations

The mining industry relies heavily on accurate, current elevation data for operational planning and regulatory compliance.

DEMs enable:

- **Volumetric calculations for stockpiles and excavations:** Accurately measuring ore, waste rock, and product volumes for inventory management and revenue calculation
- **Pit progression monitoring:** Tracking how excavations evolve over time, comparing planned vs. actual mining progress
- **Haul road design:** Optimizing road grades and routes to minimize fuel consumption and equipment wear
- **Rehabilitation monitoring:** Demonstrating compliance with environmental obligations by tracking restoration of disturbed areas

The temporal component is critical in mining. Time-series DEMs created from regular stereo satellite collections track volume changes with precision, providing the data needed for operational decisions and regulatory reporting.

Learn more about satellite imagery applications for [mining](/mining) operations.

### 4. Agriculture

Topography influences soil properties, water movement, and microclimate—making elevation data valuable for precision agriculture.

DEMs support:

- **Topography-informed management zones:** Dividing fields based on slope, aspect, and elevation to apply variable-rate inputs
- **Drainage planning:** Identifying low spots where water accumulates, informing tile drainage system design
- **Erosion risk assessment:** Calculating slope length and steepness factors for erosion prediction models
- **Precision irrigation design:** Optimizing sprinkler placement and flow rates based on topography
- **Field-scale elevation mapping:** Informing precision planting decisions, particularly for crops sensitive to waterlogging

Understanding field-scale topographic variation helps farmers optimize inputs, reduce costs, and minimize environmental impacts.

Discover how Geopera supports [agriculture](/agriculture) with satellite-derived insights.

### 5. Urban Planning

Cities exist in three dimensions, and elevation data brings that third dimension into planning workflows.

DEMs enable:

- **3D city modeling:** Creating realistic urban visualizations for planning, public consultation, and design review
- **Building height extraction:** Deriving building heights by subtracting DTM from DSM
- **Viewshed analysis:** Determining what's visible from specific locations, critical for visual impact studies of new developments
- **Solar potential mapping:** Identifying roof areas with optimal sun exposure while accounting for shadowing
- **Green infrastructure design:** Planning drainage solutions, green roofs, and urban forests with proper understanding of topography

### 6. Environmental Monitoring

Environmental applications span from disaster assessment to ecosystem management.

DEMs support:

- **Landslide risk assessment and detection:** Identifying unstable slopes and measuring terrain deformation after mass movement events
- **Coastal erosion measurement:** Quantifying beach and cliff elevation changes over time
- **Forest canopy height modeling:** Estimating biomass by calculating tree heights from DSM minus DTM
- **Habitat mapping:** Many species distributions correlate with topographic features like slope, aspect, and elevation

Explore Geopera's capabilities for [environmental monitoring](/environmental) applications.

### 7. Disaster Response

After earthquakes, landslides, volcanic eruptions, or major storms, elevation change detection reveals the extent and severity of impacts.

DEMs enable:

- **Earthquake deformation mapping:** Measuring ground surface changes caused by tectonic movement
- **Landslide extent and volume calculation:** Quantifying material movement for response planning
- **Volcanic topographic change tracking:** Monitoring lava dome growth or crater formation
- **Infrastructure damage assessment:** Identifying collapsed buildings, damaged roads, and altered drainage patterns

Rapid DEM generation from satellite stereo imagery collected immediately after disasters provides critical information for response efforts.

### 8. Archaeology

Bare-earth LiDAR DEMs have revolutionized archaeology by revealing subtle terrain features invisible under vegetation.

Applications include:

- **Buried structure detection:** Ancient building foundations, defensive earthworks, and agricultural terraces create subtle elevation variations visible in high-resolution DTMs
- **Landscape reconstruction:** Understanding historical environments and settlement patterns
- **Site prospection:** Identifying promising areas for investigation before expensive excavation

Famous examples include LiDAR revealing extensive Mayan cities beneath Central American jungle canopy—settlements that were completely invisible in conventional imagery.

### 9. Geologic Studies

Elevation patterns reveal Earth's geological structure and active processes.

DEMs support:

- **Tectonic feature identification:** Fault scarps, rift valleys, and volcanic features are clearly visible in elevation data
- **Fault mapping:** Identifying and characterizing active faults for seismic hazard assessment
- **Geomorphology analysis:** Understanding how landscapes evolve through erosion, deposition, and tectonic processes

A compelling example: Digital elevation models of the East African Rift System clearly show the thermal bulges and grabens where the African continent is slowly splitting apart, eventually forming a new ocean.

### 10. Orthorectification

DEMs are required to remove terrain-induced distortions from satellite and aerial imagery, converting raw imagery into map-accurate products.

Without elevation data to account for terrain relief, imagery of mountainous areas shows geometric distortions where ridges appear to lean and valleys appear compressed. Orthorectification uses DEMs to model these distortions and remove them, producing imagery where every pixel is in its correct geographic position.

This is critical for:

- Creating seamless image mosaics
- Accurate distance and area measurements
- Change detection between images from different dates
- Overlaying imagery with other GIS data layers

## Where to Find Elevation Data

There are plenty of places to find digital elevation models. From free satellite data to LiDAR sources, here's how to find the elevation data you need.

### Free Global Datasets

#### 1. SRTM (Shuttle Radar Topography Mission)

During its 11-day mission in February 2000, the Space Shuttle Endeavour orbited Earth 16 times per day, systematically mapping the planet's topography using interferometric synthetic aperture radar.

**Coverage:** Approximately 80% of Earth's landmass (between 60°N and 56°S latitude)

**Resolution:** 1 arc-second (approximately 30 meters) globally; originally 90m outside the United States but 30m data now released worldwide

**Method:** InSAR with two radar antennas separated by a 60-meter mast

**Vertical accuracy:** Less than 16m absolute vertical error for most areas

**Access:** Free download from USGS Earth Explorer

**Pros:** Free, global, consistent methodology, well-documented accuracy, widely used as a baseline

**Cons:** Data from 2000 (now 25 years old), moderate resolution, data voids in challenging terrain like steep mountains, doesn't reflect recent landscape changes

SRTM remains one of the most widely used global elevation datasets due to its consistent quality and complete documentation, despite its age.

#### 2. ASTER GDEM (Global Digital Elevation Model)

The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) is a joint operation by NASA and Japan's Ministry of Economy, Trade, and Industry (METI). The ASTER Global Digital Elevation Model was generated from stereo pair images collected by the ASTER instrument aboard the Terra satellite.

**Coverage:** Approximately 80% of Earth (83°N to 83°S)

**Resolution:** 1 arc-second (approximately 30 meters)

**Method:** Stereo optical imagery from nadir and backward-looking telescopes

**Versions:** GDEM Version 2 (2011) and Version 3 (2019) with improved artifact correction and void filling

**Access:** Free download from NASA Earthdata and USGS Earth Explorer

**Pros:** Free, newer than SRTM in some areas, performs well in rugged mountainous terrain

**Cons:** Artifacts in persistently cloudy regions, variable accuracy due to cloud contamination during source image collection, can be less accurate than SRTM in flat areas

ASTER GDEM Version 3 represents significant improvements over earlier versions, with better handling of water bodies and reduced artifacts.

#### 3. JAXA ALOS World 3D

ALOS World 3D is a global DSM dataset generated by the Japan Aerospace Exploration Agency (JAXA) from images collected by the Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) aboard the Advanced Land Observing Satellite (ALOS).

**Coverage:** Global land area

**Resolution:** 30 meters (1 arc-second) freely available; 5-meter version available for some areas through commercial licensing

**Method:** PRISM stereo optical imagery

**Acquisition period:** 2006-2011

**Access:** Free registration required through JAXA's Earth Observation Research Center portal

**Pros:** Good accuracy, more recent than SRTM, excellent void filling, better handling of steep terrain than ASTER

**Cons:** Requires registration, still approximately 15 years old, doesn't reflect very recent landscape changes

The ALOS World 3D dataset has become increasingly popular as an alternative to SRTM and ASTER for applications requiring global coverage.

#### 4. Regional LiDAR Sources

Many national, state, and local governments have invested in LiDAR collection and make the resulting elevation data freely available.

**Examples include:**

- USGS 3D Elevation Program (3DEP) covering the United States
- Environment Agency LiDAR data for England
- Australian state government programs (ELVIS portal provides unified access)
- Various European national mapping agencies

**Characteristics:** Highest accuracy of freely available options (often sub-meter vertical accuracy), typically bare-earth DTMs and surface DSMs both available, limited to specific geographic areas with government-funded collection

**Access:** Usually through government geospatial data portals

If you're working in a developed country, it's worth checking whether your local or regional government provides free LiDAR-derived DEMs. The accuracy is typically far superior to global satellite datasets.

### Commercial High-Resolution Options

Free global datasets serve many applications well, but some projects require higher accuracy, more recent data, or specific product characteristics.

#### When Free Data Isn't Enough

Consider commercial elevation data when you need:

- **Current data for change detection:** Comparing recent elevations against historical baselines to measure terrain changes
- **Sub-meter vertical accuracy:** Engineering-grade precision for critical infrastructure or regulatory compliance
- **Specific coverage without data voids:** Ensuring complete coverage of your area of interest
- **Defined product type (DSM vs DTM):** Many free datasets don't clearly distinguish between surface and terrain models

#### Satellite-Derived DEMs

High-resolution satellites like WorldView-2 and WorldView-3 can collect stereoscopic imagery that, when processed photogrammetrically, produces DEMs with sub-meter vertical accuracy.

**Characteristics:**

- 50cm to 1m vertical accuracy achievable with proper ground control
- Custom tasking enables collection of current stereo pairs
- Large-area coverage possible (thousands of square kilometers)
- Both DSM and DTM products can be generated with appropriate processing

Geopera provides access to [stereoscopic satellite imagery](/blog/stereoscopic-satellite-imagery) from the WorldView constellation, including existing archive data and custom tasking for new collections. Stereo pairs can be delivered for in-house DEM extraction or processed into elevation products by our partners.

#### Aerial LiDAR

For the highest accuracy requirements, aerial LiDAR collection remains the gold standard.

**Characteristics:**

- Centimeter-level vertical accuracy (5-15cm typical)
- Ultra-dense point clouds (10+ points per square meter)
- Simultaneous DTM and DSM generation
- Excellent vegetation penetration for ground extraction

**Trade-offs:**

- Expensive for large areas (costs scale with area)
- Requires specialized aircraft and equipment
- Weather-dependent collection
- Longer lead times for mobilization

Aerial LiDAR makes economic sense for critical infrastructure projects, detailed urban mapping, or smaller areas requiring engineering-grade accuracy.

### Getting Help Choosing the Right Data

Not sure which elevation data source fits your project requirements?

Contact Geopera to discuss your specific needs. Our team can help you determine whether free global datasets, commercial satellite-derived DEMs, or custom stereo tasking best meets your accuracy, spatial coverage, temporal requirements, and budget constraints.

We can provide:

- Assessment of existing elevation data coverage and quality for your area of interest
- Access to archive stereoscopic satellite imagery for photogrammetric DEM extraction
- Coordination of custom stereo collection missions
- Connections to processing partners for DEM generation services

## Working with DEM Data

Once you've acquired elevation data, you'll need appropriate tools to open, visualize, analyze, and extract insights from it.

### File Formats

DEMs are distributed in various file formats, each with different characteristics:

**GeoTIFF (.tif):** The most common format, widely supported across GIS platforms. Embeds georeferencing information within the file, simplifying data management. Can store elevation as integer or floating-point values.

**USGS DEM (.dem):** Legacy format from the U.S. Geological Survey. Less common now but still encountered in older datasets.

**Float (.flt):** Binary raster format storing elevation as floating-point values, often paired with a header file (.hdr) containing georeferencing information.

**ASCII Grid (.asc):** Text-based format where elevation values are stored as readable numbers in a grid structure. Human-readable but inefficient for large datasets.

**Point Clouds (.las, .laz):** LiDAR's native format stores irregular point measurements rather than regular grids. LAZ is a compressed version of LAS. These require conversion to raster DEMs for most GIS analysis workflows.

### Software Tools

You'll need Geographic Information System (GIS) software or specialized applications to work with elevation data, as DEMs aren't directly viewable in standard image viewers or web browsers.

**QGIS:** Free and open-source GIS platform with comprehensive DEM analysis capabilities. Includes terrain visualization, contour generation, slope/aspect analysis, and hydrological tools. The active development community ensures good documentation and regular updates.

**ArcGIS:** Industry-standard commercial GIS platform with extensive elevation analysis tools through the Spatial Analyst extension. Offers advanced algorithms for terrain processing, watershed delineation, and visibility analysis.

**GRASS GIS:** Free and open-source platform with over 350 raster and terrain manipulation tools. Particularly strong for advanced topographic analysis and hydrological modeling.

**Python libraries:** For programmatic DEM processing, libraries like Rasterio, GDAL/OGR, and WhiteboxTools provide powerful capabilities for automation and custom analysis workflows.

**CloudCompare:** Specialized tool for point cloud processing, essential when working with LiDAR data before conversion to raster DEMs.

### Pre-Processing Requirements

Elevation data straight from the source often requires cleaning before analysis:

**Common issues to address:**

- Numerous data voids or no-data areas
- Ill-defined coastlines where elevation doesn't transition properly to sea level
- Water bodies that should be flat but show elevation variation
- Artificial sinks requiring filling for hydrological analysis
- Coordinate system mismatches requiring re-projection

Most DEM analysis workflows begin with identifying and correcting these issues to ensure reliable results.

### Essential DEM Analysis Operations

Standard elevation analysis operations include:

**Slope analysis:** Calculate terrain gradient (steepness) for each cell, typically expressed in degrees or percent. Critical for identifying landslide hazards, determining trafficability, and planning infrastructure.

**Aspect analysis:** Determine the compass direction that slopes face. Important for understanding solar exposure, prevailing wind effects, and ecosystem characteristics that vary with orientation.

**Hillshade generation:** Create 3D-like visualization by simulating shadows cast by terrain under specified sun angle. Essential for visual interpretation and communication.

**Contour generation:** Extract elevation isolines at specified intervals, converting raster elevation into vector contour lines for traditional map display.

**Viewshed analysis:** Calculate which areas are visible from specified observation points, accounting for terrain blocking the line of sight.

**Cut/fill calculations:** Estimate earthwork volumes by comparing existing elevation against proposed design surfaces.

These operations form the foundation for more complex terrain analysis workflows.

## Advanced Applications

Beyond basic terrain analysis, elevation data enables sophisticated applications across multiple domains.

### Machine Learning and DEMs

Recent advances in machine learning have opened new possibilities for elevation data processing and analysis.

**Image inpainting to fill data voids:** Deep learning models trained on complete elevation data can intelligently fill gaps where clouds, water, or sensor limitations created voids in the original dataset. These algorithms learn terrain patterns and generate realistic elevation values for missing areas.

**Automated feature extraction:** Machine learning models can automatically identify buildings, roads, vegetation, and other features in high-resolution DEMs, accelerating mapping workflows.

**Terrain classification:** Neural networks classify terrain into categories like ridges, valleys, plains, and slopes based on elevation and derived parameters.

**Landslide susceptibility modeling:** Machine learning algorithms combine elevation, slope, geology, soil, and precipitation data to predict landslide-prone areas with greater accuracy than traditional statistical approaches.

These techniques are particularly valuable in data-scarce regions where filling elevation gaps enables flood modeling, infrastructure planning, and hazard assessment that would otherwise be impossible.

### Multi-Temporal Analysis

Collecting elevation data at multiple points in time creates time-series DEMs that reveal how terrain changes.

**Applications include:**

**Mine pit evolution tracking:** Regular DEM collection over mining operations tracks excavation progress, calculates volumes extracted, and verifies compliance with mining plans. Monthly or quarterly time series provide operational intelligence and regulatory documentation.

**Coastal erosion rate calculation:** Comparing DEMs from different years quantifies beach elevation changes, dune migration, and cliff retreat rates. Essential for coastal management and climate adaptation planning.

**Glacier mass balance:** Annual DEMs of glaciers reveal ice thickness changes, providing direct measurements of mass loss or gain. Critical data for understanding climate change impacts on water resources.

**Urban growth monitoring:** Time-series DSMs show the expansion of built areas, tracking new construction and urban sprawl patterns.

**Landslide deformation:** Frequent DEM collection over unstable slopes detects subtle terrain movement before catastrophic failure, enabling early warning systems.

The temporal component transforms elevation data from static terrain representation into a dynamic monitoring tool.

### DEM Derivatives

Beyond elevation itself, numerous derivative products extract additional terrain information:

**Terrain Ruggedness Index (TRI):** Quantifies landscape roughness by measuring elevation variance within a moving window. Used for habitat modeling (many species prefer specific ruggedness levels) and trafficability assessment.

**Topographic Position Index (TPI):** Classifies terrain by comparing each cell's elevation to the average of surrounding cells. Identifies ridges (positive values), valleys (negative values), and flat areas (near zero).

**Terrain Wetness Index:** Combines slope and flow accumulation to predict soil moisture patterns. Valuable for ecological modeling and precision agriculture.

**Stream Power Index:** Estimates erosive power of flowing water based on slope and contributing area. Helps predict where erosion and deposition occur.

These derivatives extract information embedded in elevation patterns, enabling applications from soil mapping to wildlife habitat assessment.

### Fusion with Other Data

DEMs become even more powerful when combined with other datasets:

**Combining DEMs with multispectral imagery:** Elevation context enhances image classification. For example, vegetation indices combined with elevation improve forest type mapping, as species distributions often correlate with altitude.

**Integrating with hydrological models:** Rainfall-runoff models require elevation data to route water across landscapes. The DEM provides the terrain framework for simulating how storms translate into streamflow.

**3D building models:** Combining building footprint vectors with DSM-minus-DTM height data creates 3D city models without expensive LiDAR collection.

**Augmented reality applications:** Mobile apps overlay digital information on real-world views by combining device position/orientation with elevation data to understand the terrain context.

The integration possibilities are nearly limitless, as elevation provides fundamental spatial context for countless applications.

## Conclusion

Digital elevation models—whether broad-coverage DEMs, feature-inclusive DSMs, or bare-earth DTMs—provide the vertical dimension essential for spatial analysis across countless applications.

Understanding the differences between these products, how they're created, and their accuracy characteristics enables you to select the right elevation data for your needs. Free global datasets like SRTM, ASTER, and ALOS World 3D serve many applications well, providing consistent coverage for regional analysis, preliminary studies, and projects where meter-level accuracy suffices.

When your application demands higher precision, more recent data, or specific product characteristics, commercial high-resolution products deliver the performance required. Satellite-derived DEMs from stereo photogrammetry offer sub-meter vertical accuracy with flexible coverage areas. Aerial LiDAR provides centimeter-level precision for critical infrastructure and engineering projects.

The choice between DSM and DTM depends entirely on your application. Hydrological modeling demands bare-earth DTMs, as water flows across the ground surface rather than over building roofs. Urban 3D visualization requires feature-rich DSMs to represent cities accurately. Many projects benefit from both product types—using the difference between them to extract object heights through normalized DSMs.

Whether you're modeling flood risk to inform climate adaptation, planning infrastructure routes to minimize environmental impact, calculating mine volumes for operational management, or analyzing terrain for any other purpose, elevation data transforms 2D maps into 3D intelligence.

The vertical dimension reveals patterns invisible in traditional imagery: subtle terrain features indicating buried archaeological sites, slope characteristics controlling erosion and vegetation, watershed boundaries determining water flow paths, and elevation changes documenting landscape evolution.

Ready to explore elevation data options for your project? [Contact Geopera](https://geopera.com/contact) to discuss which approach best fits your accuracy requirements, spatial coverage needs, temporal constraints, and budget. Our team can help you access archive stereo imagery, coordinate custom satellite collections, or connect you with processing partners for DEM extraction services.

## Key Takeaways

- DEM is a broad umbrella term: DSMs capture the surface including all objects, while DTMs represent only bare-earth terrain
- DEMs are created using multiple methods: InSAR (all-weather radar), stereo photogrammetry (optical imagery from multiple angles), LiDAR (laser scanning with highest accuracy), contour digitization (legacy method), or ground surveys (limited scope, high precision)
- Quality depends on multiple factors: spatial resolution (pixel size), vertical resolution (height precision), acquisition method, temporal currency, and processing algorithms all affect DEM fitness for purpose
- DSMs include all features above ground: buildings, vegetation, power lines, and other structures. Best for urban 3D modeling, aviation obstacle assessment, and canopy height mapping
- DTMs show only bare earth: all features removed through classification. Essential for hydrology, flood modeling, infrastructure routing, and geomorphology
- Subtracting DTM from DSM yields object heights: nDSM = DSM - DTM provides vegetation and building heights above ground
- Free global datasets work for many applications: SRTM (30m, year 2000), ASTER GDEM (30m, cloud artifacts), and ALOS World 3D (30m, 2006-2011) provide baseline coverage
- High-accuracy needs require commercial data: Satellite stereo imagery produces sub-meter DEMs; aerial LiDAR achieves centimeter-level precision
- Applications span every geospatial domain: flood modeling, infrastructure planning, mining volumetrics, precision agriculture, urban planning, environmental monitoring, disaster response, archaeology, and geology
- Contact Geopera to determine the best elevation data source for your specific project requirements and constraints

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Understanding Stereoscopic Satellite Imagery: From Capture to DEM

> Learn how stereoscopic satellite imagery creates accurate 3D elevation models from two 2D images captured from different angles.

Published: 2025-10-14 | Author: Darcy Weedman | Reading time: 7 min

Source: https://geopera.com/blog/stereoscopic-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- **What It Is:** Stereoscopic imagery captures the same location from two different angles, creating a stereo pair that contains 3D information
- **How It Works:** Satellites capture two images 45-90 seconds apart during the same orbital pass, creating geometric disparity that reveals elevation
- **The Process:** Photogrammetric algorithms match pixels between images, calculate disparity, and generate precise elevation models (DEMs)
- **The Products:** Strip DEMs provide timestamped elevation data for change detection; mosaic DEMs cover large areas for regional analysis
- **Best Sensors:** WorldView-1, WorldView-2, and WorldView-3 produce the highest quality DEMs, with WorldView Legion adding new capabilities
- **Access at Geopera:** Browse existing stereo imagery in our [Pera Portal](https://portal.geopera.com) or [contact us](https://geopera.com/contact) for custom new collections

A single satellite image shows you terrain in two dimensions. But when that same location is captured from two different angles, something remarkable happens—the images contain enough information to calculate precise elevation data across the entire scene.

This is stereoscopic satellite imagery, and it's the foundation for creating accurate Digital Elevation Models (DEMs) that reveal the three-dimensional structure of Earth's surface.

<AskBand headline="Could you measure the height of that?" sub="Stereo-capable sensors are flagged in the catalogue; quotes show the price up front." action="Browse stereo sensors" href="/sensors" tiles={['/band/caldera.webp', '/band/terraces.webp', '/band/city.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## Understanding Monoscopic vs Stereoscopic Collection

Traditional satellite imagery is monoscopic—a single image captured from one viewing angle. While valuable for many applications, monoscopic imagery provides no direct information about elevation or terrain relief.

Stereoscopic collection changes this by acquiring two images of the same location from different angles. These paired images—called a stereo pair—create the geometric conditions necessary to extract elevation data through photogrammetric processing.

The key difference: stereoscopic imagery doesn't just show what's on the ground. It shows how high or low that ground is.

## How Satellites Capture Stereo Pairs

Modern high-resolution satellites like Vantor's WorldView constellation (formerly Maxar) have sophisticated pointing capabilities that enable stereoscopic collection through two primary methods.

### In-Track Stereo Collection

The most common approach is in-track stereo, where both images are captured during a single orbital pass.

Here's how it works:

1. The satellite captures the first image while passing over the target location
2. It continues along its orbit for 45-90 seconds
3. The satellite rotates its camera to point back at the same location
4. It captures the second image from a different viewing angle

Because both captures occur within about 90 seconds, very little changes on the ground between shots. This temporal consistency is crucial—vehicles, shadows, and other transient features remain nearly identical, which improves the reliability of the photogrammetric matching process.

### Cross-Track Stereo Collection

An alternative method is cross-track stereo, where the satellite completes at least one full orbit between captures. This creates a stereo pair from different orbital paths, resulting in a larger baseline (the distance between viewing positions) and potentially greater geometric separation.

However, the time gap between captures—typically 90 minutes or more—means ground conditions may have changed. Moving vehicles, shifting shadows, and even changing atmospheric conditions can complicate the matching process and reduce accuracy.

For most applications, in-track stereo delivers superior results.

### The WorldView Constellation

All satellites in Vantor's WorldView constellation support stereoscopic collection, including:

- WorldView-1 (panchromatic only)
- WorldView-2 (8-band multispectral)
- WorldView-3 (16-band SWIR + multispectral)
- WorldView-4 (decommissioned 2019)
- Six WorldView Legion satellites (launched 2023-2024)

While all these sensors can collect stereo imagery, WorldView-1, WorldView-2, and WorldView-3 produce the most accurate DEMs due to their refined sensor models and extensive calibration data. The WorldView Legion constellation also delivers excellent stereo capabilities with improved revisit times and global coverage.

## From Two 2D Images to 3D Elevation Data

The transformation from stereo pair to elevation model relies on photogrammetry—the science of making measurements from photographs. The process involves several computational steps that convert geometric disparity between images into precise elevation values.

### Step 1: Image Matching

Photogrammetric algorithms first identify corresponding features between the two images. Unlike traditional feature matching that focuses on distinctive points, DEM extraction attempts to match every single pixel with its corresponding pixel in the second image.

This pixel-by-pixel correspondence is where the geometric magic happens. Because the two images were captured from different angles, any given ground feature will appear in slightly different positions in each image. This positional difference—the disparity—is directly related to that feature's elevation.

### Step 2: Calculating Disparity

Once corresponding pixels are identified, the algorithm calculates the precise positional difference between them. A tall building, for instance, will show greater disparity than flat ground because its height causes it to appear in noticeably different locations between the two views.

The relationship between disparity and elevation depends on the collection geometry—specifically, the convergence angle (how different the two viewing angles are) and the satellite's altitude at the time of capture.

### Step 3: Converting Disparity to Coordinates

The satellite's position during each capture is precisely known, recorded by onboard GPS to centimetre-level accuracy. Using this positional data, the collection geometry, and the measured disparity for each pixel, the algorithms calculate the exact distance from that pixel to the satellite.

With the satellite position known and the distance calculated, basic triangulation provides the pixel's precise horizontal (x, y) and vertical (z) coordinates in geographic space.

### Step 4: Generating the DEM

This process repeats across the entire image extent, producing a dense point cloud—millions of precisely positioned elevation measurements. The algorithm then interpolates a continuous surface through this point cloud, divides it into a regular grid, and writes the result as a raster DEM.

The output is a stereo-derived elevation model where each pixel value represents the elevation at that location.

## Strip DEMs vs Mosaic DEMs

Stereo-derived elevation models come in two primary forms, each suited to different applications.

### Strip DEMs: Timestamped Elevation Data

A DEM strip is generated from a single stereo pair (or a sequence of stereo pairs collected back-to-back). These strips typically measure 15-17 km wide and 15-100 km long—matching the coverage area of the source imagery.

The defining characteristic of strip DEMs is their timestamp. Because the source images were captured within a narrow time window, the DEM represents the terrain's elevation at that specific moment in time.

This makes strip DEMs ideal for change detection applications:

- **Mining operations:** Calculate volumetric changes in stockpiles, pits, and waste dumps over time
- **Landslide monitoring:** Detect terrain deformation and mass movement
- **Coastal erosion:** Measure beach and dune elevation changes
- **Infrastructure construction:** Track earthwork progress on large projects

By collecting multiple stereo pairs over the same area at different times, you can create a time series of elevation models that reveal how the terrain has changed.

### Mosaic DEMs: Regional Elevation Coverage

Mosaic DEMs are created by seamlessly blending multiple strip DEMs to cover larger geographic areas—regions, states, or even entire continents.

To avoid visible discontinuities at strip boundaries, the source DEMs are co-registered (aligned precisely to one another) and their edges are blended. The result looks like a single, continuous elevation model.

However, this comes with a trade-off: mosaic DEMs no longer represent a single moment in time. The source strips may have been collected over months or even years, meaning different parts of the mosaic reflect terrain conditions from different dates.

Mosaic DEMs excel at applications requiring broad spatial coverage:

- **Regional planning:** Infrastructure routing, watershed analysis, flood modeling
- **Agricultural drainage design:** Topographic mapping for irrigation planning
- **Baseline terrain mapping:** Creating reference elevation datasets for remote areas
- **Geologic mapping:** Understanding large-scale landforms and structures

| Feature                | Strip DEMs                   | Mosaic DEMs                  |
| ---------------------- | ---------------------------- | ---------------------------- |
| **Coverage**           | 15-100 km strips             | Regional to continental      |
| **Timestamp**          | Single capture time          | Multi-year composite         |
| **Best For**           | Change detection, monitoring | Large-area mapping, planning |
| **Typical Resolution** | 50cm - 1m per pixel          | 50cm - 1m per pixel          |
| **Vertical Accuracy**  | Sub-meter (with WorldView)   | Sub-meter (with WorldView)   |

## Applications Across Industries

Stereoscopic imagery and the DEMs derived from it serve critical functions across multiple sectors.

### Mining

- Volumetric calculations for ore stockpiles and excavations
- Mine pit progression tracking for planning and reporting
- Haul road design and slope stability analysis
- Environmental rehabilitation monitoring

Learn more about satellite imagery for [mining operations](/mining).

### Infrastructure

- Terrain analysis for road, rail, and pipeline routing
- Cut-and-fill calculations for construction planning
- Flood risk modeling along infrastructure corridors
- As-built verification for earthworks projects

Explore applications for [infrastructure planning](/infrastructure).

### Agriculture

- Precision agriculture with topography-informed management zones
- Drainage planning and erosion monitoring
- Terrain-based irrigation system design
- Field-scale elevation mapping for precision planting

See how satellite imagery supports [agriculture](/agriculture).

### Environmental Monitoring

- Landslide risk assessment and post-event deformation mapping
- Coastal change detection and erosion measurement
- Forest canopy height modeling for biomass estimation
- Watershed and stream network delineation

Discover [environmental monitoring](/environmental) capabilities.

### Disaster Response

- Before-and-after elevation change detection following earthquakes
- Landslide and debris flow mapping
- Flood extent and impact modeling
- Infrastructure damage assessment

## Accessing Stereoscopic Imagery with Geopera

Geopera provides access to stereoscopic satellite imagery from the WorldView constellation, including both existing archive data and new custom collections.

### Browse the Archive

Our [Pera Portal](https://portal.geopera.com) includes extensive coverage of stereo imagery captured by WorldView satellites worldwide. You can:

- Search for existing stereo pairs over your area of interest
- Preview coverage footprints and acquisition dates
- Filter by sensor, collection date, and quality parameters
- Download imagery directly or stream it for visualization

Many projects can be completed using existing archive imagery, often available for immediate delivery at lower cost than new tasking orders.

### Order Custom Stereo Collections

When archive coverage isn't available or you need current data, Geopera can facilitate tasking orders for new stereoscopic captures. Our team handles:

- Collection planning and feasibility analysis
- Sensor selection based on your DEM accuracy requirements
- Coordination with satellite operators
- Processing and delivery of orthorectified stereo pairs

### DEM Extraction Services

If you need the final DEM product rather than raw stereo imagery, Geopera can connect you with photogrammetric processing partners or provide guidance on recommended tools and workflows for DEM extraction.

Not sure which approach is right for your project? [Contact our team](https://geopera.com/contact) to discuss your specific requirements. We can help you determine whether archive imagery, new tasking, or alternative elevation data sources best fit your needs and budget.

## The Bottom Line

Stereoscopic satellite imagery transforms the way we understand terrain. By capturing the same location from two angles, satellites create the geometric foundation for precise elevation modeling—turning 2D images into 3D intelligence.

For applications that demand accurate elevation data—from mining volumetrics to infrastructure planning to environmental monitoring—stereo-derived DEMs provide a flexible, cost-effective alternative to LiDAR and traditional surveying methods.

Whether you're tracking elevation changes over time with strip DEMs or mapping large regions with mosaics, stereoscopic imagery delivers the vertical dimension your project requires.

Ready to explore stereo imagery for your project? [Browse our archive](https://portal.geopera.com) or [contact us](https://geopera.com/contact) to discuss custom collection options.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Maxar Rebrand: Vantor and Lanteris Replace Maxar Intelligence and Space Systems

> Maxar is now Vantor (satellite imagery) and Lanteris (space systems). Learn what this 2025 rebrand means for WorldView satellite users and the industry

Published: 2025-10-02 | Author: Darcy Weedman | Reading time: 8 min

Source: https://geopera.com/blog/vantor-lanteris-maxar-rebrand

---
**Update October 2025**: Maxar Technologies has officially rebranded. **Maxar Intelligence is now called Vantor** and **Maxar Space Systems is now Lanteris Space Systems**.

After eight years as one of the most recognisable names in commercial Earth observation, Maxar Technologies has officially ceased to exist. The company has split into two distinct entities: **Vantor** (formerly Maxar Intelligence, owner of WorldView satellites) and **Lanteris Space Systems** (formerly Maxar Space Systems). This restructuring represents the largest shake-up in the commercial satellite imagery sector since DigitalGlobe's acquisition in 2017.

The split, orchestrated by private equity firm Advent International following their $6.4 billion acquisition of Maxar in 2023, marks a strategic pivot from vertical integration to specialisation. For organisations relying on satellite imagery for critical operations, this transformation raises important questions about service continuity, pricing stability, and the future direction of Earth observation technology.

## The Strategic Split: Focus Over Integration

Maxar's journey from formation to dissolution tells a story of changing market dynamics in the space industry. Created in 2017 through the merger of DigitalGlobe and MDA, Maxar attempted to build a vertically integrated space company combining satellite imagery with spacecraft manufacturing capabilities.

The decision to split reflects a fundamental shift in how the industry views value creation. Rather than maintaining a single entity spanning from satellite construction to data delivery, Advent International recognised that each business unit could achieve greater focus and efficiency operating independently.

**Vantor** inherits the crown jewels of the former DigitalGlobe empire: the WorldView satellite constellation, decades of archived imagery, and established relationships with government and commercial customers. The company maintains operational control over WorldView-1, WorldView-2, WorldView-3, and GeoEye-1 satellites, collectively providing sub-metre resolution imagery coverage globally.

**Lanteris Space Systems** takes ownership of the spacecraft manufacturing heritage from SSL and MDA, focusing on power systems, propulsion technology, and satellite bus construction for defense and commercial applications. Their portfolio spans from geostationary communications satellites to deep space exploration vehicles.

## What is Vantor? Understanding Maxar Intelligence's New Identity

Under CEO Dan Smoot's leadership, Vantor positions itself as more than a satellite imagery provider. The company's new identity emphasises "end-to-end solutions" that combine multiple data sources into actionable intelligence products.

The centrepiece of Vantor's strategy is **TensorGlobe**, an AI-powered platform designed to create a comprehensive digital twin of Earth. This system integrates satellite imagery with ground sensor data, aircraft reconnaissance, and other intelligence sources to deliver what Smoot calls "a unified intelligence picture."

This shift from selling raw imagery to providing processed intelligence reflects broader industry trends. Machine learning algorithms now automatically detect changes, classify objects, and predict patterns across vast archives of satellite data. For Vantor's customers, this means receiving answers rather than images—a fundamental change in the value proposition.

The emphasis on artificial intelligence isn't merely marketing rhetoric. Vantor has invested heavily in computer vision capabilities, automated feature extraction, and predictive analytics. Their systems can now identify specific vehicle types, assess crop health variations, and detect infrastructure changes with minimal human intervention.

## Market Implications for Imagery Users

For organisations currently using Maxar imagery or considering satellite data for their operations, the rebrand introduces both opportunities and uncertainties.

### Service Continuity Considerations

While Vantor assures customers of seamless transition, any major corporate restructuring creates potential disruption points. Existing contracts remain valid, but renewal terms, pricing structures, and service level agreements may evolve as Vantor establishes its independent market position.

The private equity ownership model typically emphasises efficiency and profitability, potentially affecting how Vantor prioritises different market segments. Government contracts, which historically provided stable revenue for Maxar, will likely remain central to Vantor's strategy. Commercial customers, particularly those in mining, agriculture, and environmental monitoring sectors, should prepare for potential changes in service offerings and pricing models.

### Competitive Dynamics Shift

The fragmentation of Maxar creates opportunities for alternative providers. Companies like [Planet Labs](https://geopera.com/blog/best-satellite-imagery), with their large constellation of small satellites, and emerging players like [BlackSky](https://geopera.com/satellite-sensors) offer different approaches to Earth observation.

For Australian organisations, this market evolution highlights the value of platform-agnostic solutions. Rather than committing to a single satellite provider, many businesses now prefer access to multiple imagery sources through unified platforms. This approach ensures continuity regardless of individual provider changes while enabling selection of the most appropriate data for each specific application.

### Technology Integration Accelerates

Vantor's focus on AI-driven intelligence products signals an industry-wide shift toward processed data delivery. Raw satellite imagery, while still essential for certain applications, increasingly serves as input for sophisticated analysis pipelines rather than the final product.

This evolution benefits end-users by reducing the technical expertise required for satellite data utilisation. However, it also raises questions about data sovereignty, processing transparency, and the ability to perform independent analysis when needed.

## The Australian Perspective: Stability Through Diversity

For Australian businesses relying on satellite imagery for operations, the Maxar split underscores the importance of vendor diversity and platform flexibility. The mining sector, which depends on regular monitoring of vast operational areas, cannot afford disruption in imagery supply. Similarly, agricultural enterprises using satellite data for [precision farming](https://geopera.com/industries/agriculture) require consistent, reliable access throughout growing seasons.

This market reality drives demand for integrated platforms that aggregate multiple satellite sources. Rather than navigating relationships with individual providers like Vantor, many Australian organisations prefer working through local partners who manage the complexity of multi-source data acquisition and processing.

The ability to access WorldView imagery alongside data from other providers—including [Wyvern's hyperspectral sensors](https://geopera.com/blog/geopera-wyvern-partner), [21AT's Beijing-3 constellation](https://geopera.com/blog/geopera-21at-partner), and free Sentinel-2 data—provides operational flexibility and cost optimisation opportunities.

## Technical Capabilities and Future Direction

Vantor's technical roadmap emphasises three key areas that will shape the future of commercial Earth observation:

### Edge Computing and Real-Time Analysis

Moving processing closer to data acquisition points reduces latency and enables near real-time intelligence delivery. Vantor's investment in edge computing infrastructure allows preliminary analysis aboard satellites or at ground stations before full resolution data reaches central processing facilities.

### Multi-Modal Sensor Fusion

Beyond traditional optical imagery, Vantor integrates synthetic aperture radar (SAR), thermal sensors, and atmospheric monitoring data. This multi-modal approach enables continuous monitoring regardless of weather conditions or time of day—critical for applications like [disaster response and emergency management](https://geopera.com/environmental).

### Automated Change Detection

Machine learning models trained on Vantor's vast imagery archive can automatically identify and classify changes across time series data. This capability transforms reactive monitoring into proactive intelligence, alerting users to significant changes before they become critical issues.

## Navigating the New Landscape

The transformation of Maxar into Vantor and Lanteris represents more than corporate rebranding—it signals fundamental shifts in how the satellite imagery industry creates and delivers value. For imagery users, this evolution presents both challenges and opportunities.

Organisations should evaluate their satellite imagery strategies considering several factors:

**Vendor Relationships**: Direct relationships with providers like Vantor offer certain advantages but create dependency risks. Platform-based approaches provide flexibility at potentially higher costs.

**Technical Requirements**: Determine whether your applications require raw imagery for custom processing or if pre-processed intelligence products meet your needs.

**Budget Predictability**: Private equity ownership often drives pricing optimisation. Consider locking in long-term agreements or maintaining alternative supply options.

**Geographic Coverage**: Ensure your chosen approach provides adequate coverage for all operational areas, particularly for Australian and Asia-Pacific regions where certain satellites may have limited tasking priorities.

## Looking Forward: The Evolution Continues

The satellite imagery industry stands at an inflection point. Traditional providers like Vantor compete with new entrants offering novel approaches—from large constellations of small satellites to specialised sensors for specific applications. Meanwhile, the democratisation of space technology continues driving down launch costs and enabling new business models.

For Australian businesses, this dynamic environment creates unprecedented opportunities to leverage satellite technology for competitive advantage. Whether monitoring [mining operations](https://geopera.com/mining), optimising agricultural productivity, or managing environmental compliance, access to diverse, high-quality satellite data has never been more critical.

The key to success lies not in betting on any single provider or technology but in maintaining flexibility to adapt as the market evolves. Platforms that aggregate multiple data sources, provide consistent processing capabilities, and offer transparent pricing models will become increasingly valuable as the industry continues its transformation.

## Frequently Asked Questions About the Maxar Rebrand

### What is Maxar's new name?

Maxar no longer exists as a single company. **Maxar Intelligence is now called Vantor**, and **Maxar Space Systems is now called Lanteris Space Systems**. The rebrand was announced in October 2025.

### Who owns the WorldView satellites now?

**Vantor** (formerly Maxar Intelligence) owns and operates the WorldView satellite constellation, including WorldView-1, WorldView-2, WorldView-3, and GeoEye-1. These satellites continue to provide the same high-resolution imagery services.

### Why did Maxar change its name to Vantor?

Private equity firm Advent International, which acquired Maxar in 2023, split the company to allow each business unit to focus on its core competencies. Vantor focuses exclusively on Earth observation and satellite imagery, while Lanteris handles spacecraft manufacturing.

### Is Vantor the same as DigitalGlobe?

Vantor is the successor to both DigitalGlobe and Maxar Intelligence. The company inherited DigitalGlobe's satellite constellation and imagery archives when DigitalGlobe merged into Maxar in 2017, and now continues that legacy under the Vantor brand.

### Will my Maxar imagery contracts transfer to Vantor?

Yes, existing Maxar Intelligence contracts and services continue under Vantor. However, renewal terms and pricing may change as the company establishes its new market position.

### What's the difference between Vantor and Lanteris?

**Vantor** provides satellite imagery and geospatial intelligence services, operating Earth observation satellites. **Lanteris Space Systems** manufactures spacecraft, satellite components, and space infrastructure for defense and commercial customers.

---

**Ready to explore your satellite imagery options?** Geopera provides seamless access to WorldView imagery from Vantor alongside 20+ other satellite constellations through our unified Pera Portal platform. Our processing pipeline ensures consistent, analysis-ready data delivery regardless of source, giving you the flexibility to choose the best imagery for each application while maintaining operational continuity.

[Explore available imagery through our Pera Portal →](https://portal.geopera.com) or [contact our team](https://geopera.com/contact) to discuss your specific requirements.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Commercial Optical Satellite Imagery: Bands, Resolution and Ordering

> How optical satellite imagery works, what each spectral band shows, which resolution tier suits which job, and how to order commercial optical imagery.

Published: 2025-09-24 | Author: Darcy Weedman | Reading time: 12 min

Source: https://geopera.com/blog/understanding-optical-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

Optical satellite imagery is captured by passive sensors that record reflected sunlight, in the visible range and in wavelengths just outside it. Commercial optical imagery is sold from archive or by tasking a new capture, priced by area, at resolutions from 0.3 m to 10 m. What separates one product from another is spatial resolution, which bands the sensor records, and how often it revisits.

- Resolution sets what you can identify: 0.3 to 1 m resolves vehicles and building features, 10 to 30 m resolves land cover patterns.
- Bands set what you can measure. Red Edge detects plant stress before it is visible; SWIR separates alteration minerals; coastal blue penetrates water.
- The hard limits are cloud and daylight. Optical sensors see neither through cloud nor at night. Radar does both.
- Commercial optical imagery is ordered by drawing an area of interest and choosing archive or tasking. Prices are per km², not per scene.

Optical satellites are the workhorse of Earth observation, and most imagery anyone has ever looked at came from one. This covers how they work, what each band is for, and how to buy the output.

<AskBand headline="What do the bands show over your site?" sub="Compute indices on free Sentinel-2 in the browser." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/bloom.webp', '/band/fields.webp', '/band/atoll.webp']} />

## What is Optical Satellite Imagery?

Optical satellites are passive sensors that capture reflected sunlight from Earth's surface, much like a camera in space. Unlike radar satellites that emit their own signals, optical sensors depend on the sun's illumination, making them energy-efficient and cost-effective for many applications.

These satellites operate mostly in the visible and near-visible parts of the electromagnetic spectrum, capturing wavelengths our eyes can see plus some they cannot. The result is imagery that is intuitive to read and still carries data layers useful for analysis.

### Advantages

Optical systems cost less to operate than active sensors like radar, because they do not have to generate the energy they measure. The output is natural colour, so it needs no interpretation training to look at. Resolution reaches 30 cm per pixel commercially, and the extra spectral bands support analysis well beyond what the visible range alone would allow.

### Limitations

Cloud blocks the view entirely, which in some regions rules out optical imagery for months at a time. Daylight dependency means no night capture. Atmospheric haze degrades quality, and seasonal changes in sun angle alter shadow length and illumination, which matters when comparing captures taken at different times of year.

For monitoring that must not be interrupted by weather or darkness, pair optical with radar. For most commercial work, optical is the cheaper and more directly interpretable option.

## Understanding Spatial Resolution: The Clarity Factor

Spatial resolution determines how much detail you can see in satellite imagery. It is measured as Ground Sample Distance (GSD), the real-world size each pixel represents.

### Resolution Categories

**Very High Resolution (0.3-1m)**: Individual objects like cars, trees, and building features are distinguishable. Essential for detailed infrastructure monitoring and precision agriculture.

**High Resolution (1-5m)**: Buildings, roads, and field boundaries are clear. Ideal for urban planning and agricultural field management.

**Medium Resolution (5-30m)**: Land cover patterns and large infrastructure visible. Perfect for regional monitoring and environmental studies.

**Low Resolution (>30m)**: Continental and global patterns observable. Used for climate studies and large-scale environmental monitoring.

### Geopera's Available Resolutions

Through our [Pera Portal](https://portal.geopera.com), you can access:

- **[Beijing-3](/sensors/bj3n)**: 0.5m panchromatic, 2m multispectral
- **[Perascope Constellation](/sensors/jl1-75cm)**: 0.75m panchromatic, 3m multispectral
- **[SuperView](/sensors/superview-1)**: 0.5m panchromatic, 2m multispectral
- **[MAXAR WorldView](/sensors/worldview-3)**: 0.3m panchromatic, 1.2m multispectral
- **Sentinel-2**: 10m multispectral (free data)

Remember: higher resolution means smaller coverage area per image. Choose based on your project's balance of detail versus coverage needs.

## The Electromagnetic Spectrum and Spectral Bands

Optical satellites record more than we can see. They detect wavelengths invisible to the human eye, which is where most of the analytical value sits.

### Understanding Spectral Bands

Each spectral band captures a specific range of electromagnetic wavelengths, and different materials reflect these wavelengths uniquely. This "spectral signature" allows us to identify and analyze surface features beyond what's visible in regular photographs.

### Essential Spectral Bands

**Coastal/Deep Blue (400-450nm)**
Penetrates water to reveal bathymetry and detect phytoplankton. Critical for coastal monitoring and water quality assessment.

**Blue (450-510nm)**
Sensitive to atmospheric conditions and water bodies. Used for discriminating soil from vegetation and detecting industrial pollution.

**Green (510-580nm)**
Peak reflectance for healthy vegetation. Essential for assessing plant vigor and differentiating plant species.

**Red (630-690nm)**
Absorbed by chlorophyll in healthy plants. Combined with NIR for vegetation indices like NDVI.

**Red Edge (690-730nm)**
Captures the transition zone where vegetation reflectance dramatically increases. Highly sensitive to plant stress and chlorophyll content, often detecting problems before visible symptoms appear.

**Near-Infrared/NIR (760-900nm)**
Strongly reflected by healthy vegetation. The foundation for most vegetation health assessments and moisture detection.

**Panchromatic (450-800nm)**
Combines all visible wavelengths into high-resolution grayscale. Provides the sharpest detail but without color information.

### Multispectral vs Panchromatic: Making the Right Choice

**Multispectral imagery** provides several bands of spectral information at moderate resolution. It is what you need for spectral indices or false-colour composites.

**Panchromatic imagery** trades colour for spatial detail, which suits identifying small features or precise boundaries.

**Pansharpening** combines both: fusing high-resolution panchromatic with multispectral bands creates sharp, full-color imagery. At Geopera, we offer pansharpened products that deliver both detail and spectral richness. [Learn more about our processing capabilities](/why-we-process-every-order).

## Real-World Applications

Where optical imagery earns its cost, by sector.

### Agriculture & Crop Monitoring

Modern farming relies on satellite-derived vegetation indices to optimize operations:

- **Crop health assessment** using NDVI identifies stress zones before visible symptoms
- **Irrigation management** through soil moisture analysis reduces water waste
- **Yield prediction** helps with harvest planning and market positioning
- **Pest and disease detection** enables targeted treatment instead of blanket applications

[Explore our agriculture solutions](/agriculture) to see how Australian farmers are increasing yields while reducing inputs.

### Mining & Resource Management

Mining uses optical imagery across the whole project lifecycle, covered in more depth in the [mineral exploration guide](/blog/mining-satellite-imagery):

- **Exploration** using multispectral analysis to identify mineral signatures
- **Site monitoring** tracks excavation progress and equipment deployment
- **Environmental compliance** documents vegetation clearing and rehabilitation
- **Infrastructure planning** optimizes haul road design and processing facility placement

Our [mining industry solutions](/mining) help companies maintain operational efficiency while meeting environmental obligations.

### Environmental Monitoring

Conservation efforts depend on consistent, accurate Earth observation:

- **Deforestation tracking** with change detection algorithms
- **Water quality monitoring** using spectral signatures of algae and sediments
- **Coastal erosion measurement** through time-series analysis
- **Fire scar mapping** for post-disaster assessment and recovery planning

[View our environmental monitoring capabilities](/environmental) to understand ecosystem changes at any scale.

### Urban Planning & Infrastructure

Cities use optical imagery for smarter development:

- **Urban sprawl analysis** guides sustainable growth policies
- **Green space monitoring** ensures adequate urban vegetation
- **Construction progress tracking** keeps projects on schedule
- **Traffic pattern analysis** from high-frequency imaging

[Discover infrastructure monitoring solutions](/infrastructure) that help build better cities.

### Forestry & Vegetation Management

Forest managers protect valuable resources with satellite insights:

- **Tree species classification** using multispectral signatures
- **Disease and pest outbreak detection** through red edge analysis
- **Biomass estimation** for carbon credit calculations
- **Fire risk assessment** combining vegetation indices with weather data

Our [forestry solutions](/forestry) support sustainable forest management across Australia's diverse ecosystems.

## Commercial Optical Satellite Imagery

Not all optical imagery is commercial. It is worth being clear about the split, because it determines what you can get and what it costs.

**Public programmes** give data away. Sentinel-2 delivers 10 m multispectral globally, Landsat 30 m back to 1972. Free, unlimited, and adequate for regional work, land cover mapping and long time series. If 10 m answers your question, stop here and use it.

**Commercial optical imagery** is what you buy when it does not. Operators including Maxar, 21AT and SpaceWill sell captures at 0.3 m to 2 m, either from archive or by tasking a new acquisition, alongside Geopera's own Perascope constellation. You are paying for resolution, for the ability to specify a date, and for the licensing and provenance that lets the imagery support a formal claim.

The practical dividing line is the object you need to identify. Vehicles, individual trees, building features and equipment need commercial resolution. Paddocks, forest blocks, urban extent and water bodies do not.

Commercial optical imagery is priced by area rather than by scene, which matters more than it sounds: it means ordering several dates over the same footprint for change detection does not multiply the bill the way per-scene pricing does. Current rates by resolution tier are in the [satellite imagery cost guide](/blog/satellite-imagery-cost-guide), and the [buying guide](/blog/how-to-buy-satellite-imagery) covers minimum order areas, licensing and the process end to end.

## How to Access Optical Satellite Imagery with Geopera

Through [Pera Portal](https://portal.geopera.com), the process runs from discovery to delivery in one place: draw or upload an area of interest, see what exists over it, compare options across operators, and see the price per km² before ordering.

### Archive vs. Tasking

**Archive Imagery**: Previously captured data available for immediate download. Cost-effective for historical analysis or when recent imagery exists. Most areas have archive coverage within the past 12 months.

**Tasking Orders**: Request new captures over your area of interest. Essential for time-critical projects or specific acquisition parameters. Typical delivery within 7-14 days, weather permitting.

### Our Satellite Partners

We hold supplier agreements across several optical operators, plus our own exclusive constellation, so the sensor gets chosen for the job rather than because it is the only one on the shelf:

- **Perascope**: Geopera's own exclusive constellation, frequent revisit
- **[21AT](/blog/geopera-21at-partner)**: Beijing-3, 0.5 m resolution
- **[MAXAR](/blog/geopera-maxar-partner)**: 30 cm resolution
- **[Wyvern](/blog/geopera-wyvern-partner)**: hyperspectral, 23 to 32 bands
- **Sentinel Hub**: free Sentinel-2 data at 10 m

### Processing Options

All Geopera imagery comes [fully processed](/blog/why-we-process-every-order) and analysis-ready:

- **Orthorectification** removes geometric distortions
- **Atmospheric correction** ensures accurate spectral values
- **Pansharpening** combines detail with color
- **Custom band combinations** for specific applications
- **Vegetation indices** calculated and delivered

### Getting Started

1. **Define your requirements**: Area size, resolution needs, temporal requirements
2. **Search available data**: Use our catalog to find archive imagery or plan a tasking
3. **Select processing level**: Choose from our standard or custom processing options
4. **Place your order**: Simple checkout with transparent pricing
5. **Access your data**: Download via portal or integrate through our API

[Start exploring available imagery](https://portal.geopera.com) or [contact our team](/contact) for guidance on your specific requirements.

## Choosing the Right Optical Imagery

Selecting optimal imagery involves balancing multiple factors:

### Resolution Requirements

**Infrastructure projects**: 0.3-1m for detailed engineering
**Agricultural monitoring**: 1-5m for field-level analysis
**Regional studies**: 10-30m for cost-effective coverage
**Continental mapping**: 30m+ for large-scale patterns

### Temporal Considerations

**Change detection**: Consistent acquisition dates and sun angles
**Crop monitoring**: Weekly to monthly during growing season
**Construction tracking**: Monthly or quarterly updates
**Emergency response**: Daily tasking if available

### Spectral Requirements

**Visual interpretation**: RGB natural color
**Vegetation analysis**: Red, NIR, and Red Edge bands
**Water studies**: Blue and coastal bands
**Mineral exploration**: SWIR and thermal bands

### Budget Optimization

- Start with free Sentinel-2 data for regional assessments
- Use medium resolution for initial surveys
- Reserve high resolution for detailed areas of interest
- Consider archive imagery before requesting new tasking

## Advanced Techniques and Analysis

Modern optical satellite analysis extends far beyond visual interpretation:

### Spectral Indices

Calculate specialized indices from band combinations:

- **NDVI** (Normalized Difference Vegetation Index): Vegetation health
- **NDWI** (Normalized Difference Water Index): Water body detection
- **NDSI** (Normalized Difference Soil Index): Bare soil mapping
- **Custom indices**: Tailored to specific materials or conditions

[Explore our vegetation indices documentation](https://docs.geopera.com/en/docs/spectral-indices) for calculation methods and applications.

### Time Series Analysis

Track changes over time with multi-temporal imagery:

- Seasonal vegetation patterns
- Urban growth trajectories
- Erosion and deposition rates
- Post-disaster recovery monitoring

### Machine Learning Integration

Apply AI to extract insights at scale:

- Automated feature detection
- Land cover classification
- Change detection algorithms
- Predictive modeling

## The Future of Optical Satellite Technology

The optical satellite industry is evolving rapidly, with innovations that will transform Earth observation:

### Emerging Capabilities

**Hyperspectral imaging** with hundreds of narrow bands for material identification

**Video from space** capturing dynamic events in real-time

**30cm resolution** becoming commercially available

**Daily global coverage** through large constellations

**On-board AI processing** for immediate insight delivery

### Geopera's Technology Roadmap

We're continuously expanding our capabilities:

- Adding new satellite partnerships for better coverage
- Developing advanced processing algorithms
- Integrating AI-powered analysis tools
- Streamlining data delivery through API enhancements

Stay updated with our latest capabilities by [contacting us](/contact).

## Common Questions About Optical Imagery

### How does cloud cover affect imagery?

Clouds obstruct the satellite's view of the ground. We provide cloud coverage percentages for all archive imagery and can schedule multiple tasking attempts to capture cloud-free data.

### What's the difference between optical and radar satellites?

Optical satellites use reflected sunlight (passive), while radar satellites emit their own signals (active). Optical provides clearer, more intuitive imagery but can't penetrate clouds. Radar works in all weather but requires more processing for interpretation.

### How current is archive imagery?

Most populated areas have archive imagery within 6-12 months. High-interest areas may have weekly or monthly coverage. Search our catalog for specific availability over your area of interest.

### Can I request specific acquisition angles?

Yes, for tasking orders you can specify off-nadir angles, sun elevation requirements, and other parameters. This is particularly useful for avoiding shadows in mountainous terrain or capturing building facades.

## Starting an Optical Imagery Project

The useful first step is almost always the cheap one. Run your question against free Sentinel-2 at 10 m and see how far it gets you. A surprising number of projects finish there, and the ones that do not will have told you exactly which constraint forced the upgrade, which makes the commercial order much easier to specify.

When you do need commercial optical imagery, search the archive over your area of interest at [portal.geopera.com](https://portal.geopera.com) and see what exists and what it costs before committing to anything.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Orthorectification: Turning Raw Satellite Data Into Map-Ready Imagery

> What orthorectification is, how RPCs and a DEM correct terrain and sensor-angle distortion, how to run it in GDAL, and where to order orthorectified imagery.

Published: 2025-09-17 | Author: Darcy Weedman | Reading time: 7 min

Source: https://geopera.com/blog/orthorectification-explained

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

Orthorectification is the geometric correction that moves every pixel in a satellite image to its true ground position. It removes displacement caused by terrain relief, sensor view angle and the curvature of the Earth, using the image's rational polynomial coefficients together with a digital elevation model. The corrected output is an orthophoto: uniform scale throughout, and measurable like a map.

- Raw imagery is not a map. Objects sit away from their real positions, by metres on flat ground and by considerably more in steep terrain.
- The correction needs two inputs beyond the image: RPCs, which are supplied in the image metadata, and a DEM, which you have to source.
- You can run it yourself with GDAL's `gdalwarp`. The command is one line. Choosing the right DEM and validating the result is the actual work.
- Every image Geopera delivers arrives already orthorectified, so the question usually does not come up.

When satellites orbit hundreds of kilometres above Earth, they capture a lot of detail. What comes straight off the sensor is not a map. Raw satellite images carry distortions from terrain, sensor tilt, and the curvature of the planet itself.

Orthorectification removes those distortions. It aligns each pixel to its true ground position and turns the image into something you can measure against.

<AskBand headline="What does corrected look like over your site?" sub="The free Sentinel-2 tier is pre-corrected; see the end state before buying anything." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/terraces.webp', '/band/caldera.webp', '/band/city.webp']} />

## Why Orthorectification Matters

A satellite is not always looking straight down. The sensor has a view angle, the ground has hills and valleys, and the planet curves away underneath. Each of those shifts objects away from where they actually are.

Without correction:

- Features do not appear in their real locations.
- Distance and area measurements are unreliable.
- Overlaying other geospatial data produces visible misalignment.

The size of the error scales with relief and with how far off nadir the sensor was looking. On flat ground at near-nadir it can be small enough to ignore. On a steep slope at a high view angle it is large enough to put a feature on the wrong side of a boundary.

For [mining](/mining), [infrastructure](/infrastructure), [agriculture](/agriculture) and [environmental monitoring](/environmental), that matters. Engineers cannot measure, planners cannot design against it with confidence, and analysts risk drawing conclusions from imagery that disagrees with the ground.

Orthorectification produces an orthophoto, where every pixel sits in its proper place on Earth.

## How Orthorectification Works

![Animation showing the orthorectification process](/blogs/OrthoAnimation.gif)

Three inputs go in:

- **Raw image data**, the original satellite capture.
- **RPCs (Rational Polynomial Coefficients)**, metadata that describes the relationship between image pixels and ground coordinates.
- **A DEM (Digital Elevation Model)**, which supplies terrain height.

They come together in three steps. The RPCs give a mathematical model linking each pixel to a location on Earth. The DEM then adjusts those positions for elevation, so that slopes, valleys and ridges land where they belong rather than where a flat-earth assumption would put them. Finally the imagery is resampled onto a consistent grid in a chosen map projection.

The result is an image whose spatial relationships hold up: features align with maps, site plans and other geospatial layers.

## Doing It Yourself with GDAL

The open-source GDAL library will do this. `gdalwarp` applies the image's RPCs and a DEM:

```bash
gdalwarp -rpc -to RPC_DEM=dem.tif raw_image.tif ortho_image.tif
```

`-rpc` applies the Rational Polynomial Coefficients from the image metadata. `-to RPC_DEM=dem.tif` names the terrain model used for elevation correction. The output file is the orthorectified product.

That command is the easy part. The work is in everything around it: picking a DEM whose resolution and vintage suit the terrain, choosing a resampling method that does not smear detail you need, validating the result against known ground positions, and running it over datasets large enough that memory and I/O start to matter. A coarse or outdated DEM will happily produce a confidently wrong orthophoto.

If you need elevation data to try this, the [free sources of DEM data](/blog/free-sources-of-dem-data) guide covers where to get it.

## Ordering Orthorectified Imagery

If you would rather not run the pipeline, orthorectification is something you can buy already done.

Every image Geopera delivers is orthorectified before it reaches you, using our own elevation models or your supplied DEM and ground control where you have them. It arrives projection-ready for GIS, CAD and analytics platforms.

Orthorectification is one stage in a longer chain. Ingest, co-registration, ground control and ortho, pansharpening, atmospheric correction, cloud and shadow masking, seamline optimisation, harmonisation and QC all run as one automated flow, with nothing manual between stages and nothing hand-finished. That is the [Legato process](/legato), and it is why imagery from different sensors and different dates arrives in a state where you can compare it directly. Delivery is under 24 hours.

You can search the archive and see prices per area at [portal.geopera.com](https://portal.geopera.com). For what the resolution tiers cost, see the [satellite imagery cost guide](/blog/satellite-imagery-cost-guide).

## Common Questions

### What is the difference between georeferencing and orthorectification?

Georeferencing assigns map coordinates to an image. Orthorectification goes further and removes the geometric distortion first, using a terrain model, so that scale is uniform across the whole frame. A georeferenced image can still be wrong in the hills. An orthorectified one has had that displacement taken out.

### What is an orthophoto?

An orthophoto is the output of orthorectification: an aerial or satellite image corrected so that it has the geometric properties of a map. You can measure distance and area off it directly.

### Do I need orthorectified imagery?

If you are measuring anything, overlaying imagery on other spatial data, or comparing captures from different dates or sensors, yes. If you only need a visual impression of a location, raw imagery may be adequate. Anything that ends up in front of a regulator, an engineer or a court should be orthorectified.

### Does orthorectification fix everything?

No. It corrects geometry, not radiometry. Haze, illumination differences between dates and colour mismatches between sensors survive it untouched, which is what atmospheric correction and harmonisation deal with later in the chain. It also has a known limit in dense urban areas: a standard terrain model corrects the ground, so the bases of tall buildings land correctly while their tops still lean.

## The Takeaway

Raw satellite data is not reliable for precision work until it has been orthorectified. The process makes images line up with the real world, which is what makes measurement and change detection defensible.

You can do it yourself with GDAL. If you would rather receive imagery that is already corrected, aligned and map-ready, Geopera delivers it that way by default.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Why We Process Every Image

> Discover how comprehensive satellite image processing transforms raw data into reliable, analysis-ready imagery that delivers accurate results for your critical projects.

Published: 2025-07-21 | Author: Darcy Weedman | Reading time: 7 min

Source: https://geopera.com/blog/why-we-process-every-order

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

**We believe you shouldn't have to choose between affordability and excellence.** That simple principle drives everything we do.

Every Geopera order includes complete processing that transforms raw satellite data into immediately usable, analysis-ready imagery—at no extra cost. While processing is often treated as an additional service, **we've made it our standard** because we believe this is how satellite imagery should always be delivered.

Your time is valuable. Your projects matter. **You deserve imagery that works perfectly from the moment you download it.**

<AskBand headline="What does analysis-ready actually cost?" sub="The processing this post argues for is in the published rates, not quoted separately." action="See the rates" href="/pricing" tiles={['/band/delta.webp', '/band/terraces.webp', '/band/city.webp']} />

## Why Satellite Images Need Processing

Here's what many people don't realise: satellite images straight from space require processing before they're ready for professional work. Think of them like digital photos taken from a moving aeroplane—they're distorted, the colours are affected by atmospheric conditions, and they don't align properly with maps or GPS coordinates.

**Processing transforms raw satellite data into something you can actually use.** It corrects distortions, fixes colours, aligns images geographically, and combines different types of data to create the clear, accurate imagery you need for analysis.

## Our Approach: Making Excellence Accessible

We've designed our business around a simple belief: **excellent satellite imagery should be accessible to everyone.**

Instead of treating processing as an expensive add-on, we've built it into our core service. This means you get:

**✓ Professional-grade processing included with every order**  
**✓ Transparent pricing with no hidden processing fees**  
**✓ Imagery that's immediately ready for analysis**  
**✓ Consistent quality regardless of project size**

**The result:** You can focus on your analysis and insights instead of wrestling with data preparation.

## Making Your Life Easier: What Processing Actually Does

Let's translate the technical jargon into what actually matters for your work:

**Without processing, satellite imagery is like receiving:**

- A puzzle with pieces that don't fit together properly
- Photos where the colours change randomly between shots
- Maps where nothing lines up with your GPS coordinates
- Images so pixelated you can't see important details

**With our processing, you get:**

- Images that align perfectly with your maps and GPS data
- Consistent, natural colours that show the world as it actually looks
- Crystal-clear detail that reveals everything you need to see
- Seamless coverage across large areas without visible boundaries
- Instant compatibility with any mapping software you already use

**The bottom line: You can start analysing immediately instead of spending weeks trying to fix broken data.**

## How We Make It Work: Seven Steps That Save You Time and Money

Every image from [Geopera's platform](https://portal.geopera.com) goes through our comprehensive processing pipeline. Here's what happens behind the scenes to make your life easier:

### 1. We Start With the Best

We work directly with satellite operators to get the highest quality raw data available. Think of this as buying ingredients from the farm instead of the grocery store—fresher, better, and with full knowledge of where it came from.

### 2. We Put Everything in the Right Place

Raw satellite images are geographically confused—like a map where the roads don't match reality. Our [location correction process](/imagery) fixes this completely, so when you click on a point in our imagery, **your GPS will take you to that exact spot**.

No more images that don't line up with your existing maps or field locations.

### 3. We Make Details Crystal Clear

Satellites capture two types of images: super-detailed black and white photos, and lower-detail colour photos. We intelligently combine these to give you **both perfect detail AND full colour**—something that would normally require expensive specialist software.

**You get the clarity you need without the complexity you don't want.**

### 4. We Fix the Colours

Earth's atmosphere acts like a dirty filter, making satellite images look hazy or strangely coloured. We automatically remove these atmospheric effects, so your imagery shows **true, natural colours** that match what you'd see standing on the ground.

### 5. We Stitch Large Areas Seamlessly

Need coverage of a large area? We automatically combine multiple satellite images into one seamless view, eliminating colour differences and visible boundaries. **No more patchwork imagery that looks like a quilt.**

### 6. We Check Everything Twice

Before any image reaches you, our automated systems verify that everything is perfect: colours are consistent, locations are accurate, and quality meets professional standards. **If it's not perfect, you don't see it.**

### 7. We Deliver It Ready to Use

Your imagery arrives in standard formats that work with any mapping software, complete with all the technical information you need. **Download, open, and start working immediately.**

**No processing delays. No additional software required. No technical headaches.**

## Excellence as Standard

**We believe you shouldn't have to choose between affordability and excellence.** That's why comprehensive processing isn't a premium add-on—it's our standard for every order, from the smallest research project to the largest enterprise deployment.

**We've designed our approach around what should be possible:** high-quality satellite imagery that's accessible, transparent, and immediately usable. When everyone on your team gets the same perfectly processed imagery, you eliminate confusion, reduce costs, and achieve better results.

## Your Time Is Worth More Than Processing Data

Here's what our approach eliminates from your workflow:

- Weeks spent learning complex processing software
- Budget uncertainty from processing fees and specialist tools
- Technical hurdles that delay project progress
- Inconsistent imagery quality across different orders
- Time lost troubleshooting data compatibility issues

**Instead, you get:** Imagery that works perfectly from day one, letting you focus on what actually matters—your analysis, your insights, your results.

## The Simplicity You Deserve

Whether you're monitoring [agricultural health](/agriculture), tracking [infrastructure changes](/infrastructure), or conducting [environmental assessments](/environmental), complex technology should make your work simpler, not harder.

**We handle the complexity so you don't have to.** Every technical challenge, every processing step, every quality check happens behind the scenes. You get professional-grade satellite imagery that just works—like it should have from the beginning.

## Start With Perfect Data Today

The best projects start with perfect data. **Why settle for anything less?**

**[Browse our Pera Portal](https://portal.geopera.com)** to instantly see what satellite imagery is available for your area. Every image comes with complete processing included—no surprises, no extra costs, no technical headaches.

Ready to experience satellite imagery that **works perfectly from the moment you download it**? [Contact our team](/contact) to see how we can make your next project easier, faster, and more successful.

## FAQs About Satellite Image Processing

**Q: What's the difference between raw and processed satellite imagery?**  
A: Raw satellite imagery contains geometric distortions, atmospheric effects, and radiometric inconsistencies that make accurate analysis difficult. Processed imagery corrects these limitations to provide reliable, analysis-ready data.

**Q: Why do some providers charge extra for processing?**  
A: Processing requires sophisticated algorithms, computational resources, and quality control systems. While some providers treat it as an optional service, we include comprehensive processing with every order because unprocessed imagery isn't suitable for professional applications.

**Q: How does processing affect imagery accuracy?**  
A: Proper processing significantly improves accuracy by correcting geometric distortions, atmospheric effects, and radiometric inconsistencies. Our processing typically achieves sub-pixel geometric accuracy and consistent radiometric quality across all delivery formats.

**Q: Can I specify processing requirements for my project?**  
A: Our standard processing pipeline meets the requirements for most professional applications. For specialised needs, our team can discuss custom processing options that align with your specific analytical requirements.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Instant NDVI in Browser, Capture Notifications, and More Updates to the Pera Portal

> Discover the latest Pera Portal enhancements: Sentinel-2 integration, 28 browser-based indices, multi-provider samples, comparison tools, and smart alerts for satellite imagery analysis.

Published: 2025-07-14 | Author: Darcy Weedman | Reading time: 7 min

Source: https://geopera.com/blog/pera-portal-major-updates-2025

---
## New Pera Portal Features for 2025

Satellite imagery analysis workflows have increasingly moved to browser-based platforms, reducing the need for specialised desktop software and local processing power. We've released five new features for the [Pera Portal](https://portal.geopera.com) that expand analysis capabilities while maintaining the platform's focus on accessibility and ease of use.

These updates include Sentinel-2 data integration, browser-based spectral index calculations, multi-provider data samples, comparison tools, and automated alert functionality for new imagery acquisitions.

## 1. Sentinel-2 Data Integration

The European Space Agency's Sentinel-2 mission provides free, high-quality satellite imagery with global coverage every five days. This dataset is now integrated into the Pera Portal, allowing users to search and visualise Sentinel-2 data alongside commercial imagery within the same interface.

**What makes this integration special:**

- **Search and visualise** Sentinel-2 data directly alongside commercial imagery
- **10-meter resolution** multispectral data perfect for large-scale monitoring projects
- **5-day revisit cycle** enabling detailed change detection and temporal analysis
- **13 spectral bands** including crucial vegetation red-edge channels
- **Seamless data fusion** with commercial providers for comprehensive analysis

This integration is useful for environmental monitoring and agricultural applications where Sentinel-2's 5-day revisit frequency and red-edge spectral bands support vegetation health tracking and crop development monitoring. The dataset provides baseline coverage for environmental impact assessment and compliance monitoring, while commercial imagery can be used for higher resolution analysis of specific areas.

## 2. Browser-Based Spectral Index Calculations

The portal now includes 25+ pre-configured visualizations and spectral indices for Sentinel-2 data, covering RGB composites and specialised analysis indices. These are calculated in the browser without requiring data downloads or additional software installations.

**RGB Composite Visualizations:**

- Natural Colour (Red, Green, Blue)
- Colour Infrared (NIR, Red, Green)
- Agriculture (SWIR1, NIR, Blue)
- Forestry Coverage (Red, NIR, Blue)
- False Colour Urban (SWIR2, SWIR1, Red)
- Land/Water (NIR, SWIR1, Red)
- Healthy Vegetation (NIR, SWIR1, Blue)

**Vegetation Indices:**

- NDVI (Normalized Difference Vegetation Index)
- NDRE (Normalized Difference Red Edge)
- EVI (Enhanced Vegetation Index)
- SAVI (Soil-Adjusted Vegetation Index)
- MSAVI2 (Modified Soil-Adjusted Vegetation Index v2)
- ARVI (Atmospherically Resistant Vegetation Index)
- Red Edge NDVI
- Chlorophyll Red Edge Index
- SIPI (Structure Insensitive Pigment Index)

**Water and Environmental Indices:**

- NDWI (Normalized Difference Water Index)
- NDWI with SWIR
- NDSI (Normalized Difference Snow Index)
- NBR (Normalized Burn Ratio)

**Specialised Analysis:**

- Deforestation Index
- Fire Detection Index
- Scene Classification Layer

Previously, calculating these indices required downloading imagery and setting up local processing environments. The browser-based approach enables faster iteration and immediate visual feedback when comparing different indices. Users can select an area of interest, choose an index, and view results without data downloads.

For agricultural applications, this allows rapid comparison of different indices to identify the most suitable approach for specific crops and growth stages. Environmental monitoring workflows benefit from the ability to quickly switch between vegetation health, water stress, and soil condition indicators.

For a deeper understanding of how these indices work and when to use each one, check out our comprehensive guide to [remote sensing vegetation indices](../blog/remote-sensing-vegetation-indices).

## 3. Multi-Provider Sample Data

The portal now includes sample datasets from four commercial providers: Wyvern, 21AT, CGST, and Spacewill. These samples allow users to evaluate different sensor capabilities and data quality before making procurement decisions.

**What's available:**

- **Pre-processed sample datasets** covering diverse landscapes and applications
- **Applied spectral indices** showing real-world analysis examples
- **Quality comparison tools** to evaluate resolution, spectral characteristics, and processing quality
- **Use case demonstrations** tailored to specific industries

This feature supports project planning and vendor selection. Users can compare hyperspectral capabilities for mineral identification, evaluate different sensors for crop monitoring, and assess temporal resolution and spectral bands for specific monitoring requirements.

The samples include diverse geographic regions and applications:

- **Agricultural sites** showing crop health monitoring capabilities
- **Mining areas** demonstrating geological and environmental applications
- **Forest regions** highlighting deforestation and biodiversity monitoring
- **Urban areas** showcasing infrastructure and development tracking
- **Water bodies** illustrating flood monitoring and water quality assessment

Each sample comes with metadata about acquisition parameters, processing methods, and suggested applications, giving you the information needed to make informed decisions about data procurement for your projects.

## 4. Comparison Tools

The portal includes new tools for side-by-side imagery comparison, supporting temporal analysis, multi-sensor evaluation, and processing method assessment.

**The Comparison Slider:** Perfect for temporal analysis, this tool enables precise before-and-after comparisons. The implementation allows pixel-perfect alignment for change detection applications, making it ideal for:

- **Environmental impact assessment** (pre and post-mining, construction, or natural disasters)
- **Agricultural monitoring** (seasonal crop development, harvest timing, irrigation effectiveness)
- **Urban development tracking** (infrastructure growth, land use changes)
- **Disaster response** (flood extent mapping, fire damage assessment, recovery monitoring)

**Mirror View:** The split-screen comparison tool enables simultaneous viewing of different datasets, spectral bands, or processing methods. This is particularly powerful for:

- **Multi-sensor fusion** (comparing commercial high-resolution with Sentinel-2)
- **Spectral analysis** (side-by-side viewing of different vegetation indices)
- **Quality assessment** (comparing different processing methods or data providers)
- **Workflow optimization** (testing different analysis approaches simultaneously)

The tools maintain spatial accuracy across different zoom levels and projections, supporting professional applications requiring precise alignment.

## 5. Automated Imagery Alerts

The portal now includes an alert system that notifies users when new imagery becomes available over specified areas of interest. This supports time-sensitive applications requiring rapid access to updated imagery.

**How the system works:**

- **Define areas of interest** using our intuitive drawing tools or by uploading shapefiles
- **Select commercial providers** to monitor (Maxar, Planet, Airbus, and more)
- **Set notification preferences** (email, portal notifications, or both)
- **Customise alert frequency** (immediate, daily digest, or weekly summary)

**Real-world applications:**

- **Disaster response preparation:** Get immediate alerts when post-event imagery becomes available
- **Construction project monitoring:** Track progress with regular imagery updates over project sites
- **Environmental compliance:** Ensure rapid detection of changes in sensitive areas
- **Security applications:** Monitor critical infrastructure or border regions
- **Agricultural management:** Time field visits and management decisions with fresh imagery

The system integrates with existing data partnerships, providing notifications for imagery from multiple commercial providers through a single interface. Users can configure immediate email notifications for high-priority applications or daily/weekly digest formats for routine monitoring.

## Accessing the Updated Portal

These features are now available in the Pera Portal for environmental monitoring, agricultural analysis, mining exploration, and infrastructure planning applications.

**Getting started:**

1. **Existing users** can access new features by logging into the [Pera Portal](https://portal.geopera.com)
2. **New users** can request access through our [contact page](../contact)
3. **Enterprise teams** can [contact our technical team](../contact) for custom integrations or training

The updated portal provides expanded analysis capabilities while maintaining browser-based accessibility for satellite imagery workflows.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# How to Buy Satellite Imagery in 2026: Prices and Process

> Buying satellite imagery in 2026: archive vs tasking, real per-km² prices ($2-55+), minimum order areas, processing fees, and how to avoid overpaying.

Published: 2025-07-10 | Author: Darcy Weedman | Reading time: 8 min

Source: https://geopera.com/blog/how-to-buy-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- Buying satellite imagery comes down to four decisions: the area of interest, archive or new tasking, the resolution tier, and the channel you buy through
- Commercial prices run from about $2/km² for 1.5 m archive to $55+/km² for 30 cm tasking. The most commonly bought tier, 50 cm archive, is typically $8-18/km²
- Processing is the classic hidden cost. Orthorectification, pansharpening and atmospheric correction add 30-80% at providers that bill them separately
- Minimum order areas (commonly 25 km² for archive, 50-100 km² for tasking) often matter more than the per-km² rate on small sites
- SAR imagery is bought through the same channels as optical; tasking matters more there because cloud cover never blocks a capture

To buy satellite imagery, you define an area of interest, choose between archive imagery and tasking a new capture, pick a resolution tier, and order through a satellite operator, a reseller, or an online platform. Budget $2-55+ USD per km² depending on resolution and freshness, and confirm whether processing is included before you compare quotes.

The rest of this guide walks through each decision, with the prices, minimums and licensing terms you should expect in 2026.

<AskBand headline="What would your order actually cost?" sub="Archive search is free, and the price is on the result before anything goes in a cart." action="Search the archive" href="https://portal.geopera.com" tiles={['/band/port.webp', '/band/city.webp', '/band/fields.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## Step 1: Work Out What You Actually Need

Three specifications drive everything else: what you need to see, how often, and in which bands.

**Spatial resolution.** At 30 cm ([WorldView-3](/sensors/worldview-3) territory) you can count vehicles and trace fence lines. At 50 cm ([Perascope](/sensors/jl1-50cm), SuperView) you can map buildings, roads and site layouts with confidence. At 1-2 m you are monitoring paddocks, pits and coastlines rather than individual assets. And for regional vegetation or land-use work, free [Sentinel-2 at 10 m](/blog/free-sources-of-satellite-data) may already cover you at no cost.

**Temporal needs.** A single capture, a seasonal series, or standing revisit? Monitoring programmes are priced and planned differently from one-off orders. Season matters too: vegetation analysis wants growing-season captures, while infrastructure audits often prefer leaf-off conditions for ground visibility.

**Spectral bands.** RGB is enough for visual interpretation. Vegetation health needs near-infrared or [red edge](/blog/red-edge-remote-sensing) bands. Mineral mapping wants shortwave infrared or hyperspectral. If your deliverable includes NDVI or any vegetation index, an RGB-only product cannot produce it.

Then draw your area of interest as a polygon, not a place name. Commercial imagery is priced per km² of that polygon, so a tight boundary is money saved.

## Archive or Tasking?

This is the biggest fork in the process, and it is mostly a question of whether the imagery you need already exists.

**Archive imagery has already been captured.** You search a catalogue, check the capture date and cloud cover of actual scenes, and buy exactly what you can see. It is the cheaper path: across the tiers in our [cost guide](/blog/satellite-imagery-cost-guide), archive typically costs 40-60% less than tasking the same resolution. It is also the faster and lower-risk path, because there is no weather to wait on and no uncertainty about what you will receive.

**Tasking is a new capture to your specification.** You define the area, the capture window, and sometimes the look angle, and a satellite is scheduled to collect it. You pay more for that control, and optical tasking carries weather risk: a cloudy window can push delivery out. Our guide to [satellite tasking](/blog/satellite-tasking-explained) covers the mechanics in detail.

The rule of thumb: always check the archive first. High-resolution satellites have been imaging continuously for years, and many of the "we need new imagery" requests we see are satisfied by a capture from the last few months at half the price.

## Where to Buy: Operators, Resellers and Platforms

**Directly from satellite operators.** Buying from the organisation that flies the satellite gives you access to their full archive and tasking calendar. It suits programmes with steady, high volume in one constellation. Expect framework negotiations, minimum commitments, and a separate agreement per operator.

**Through a reseller.** One agreement covers catalogues from many operators, which matters more than it sounds: no single constellation is best at every resolution, revisit rate and price point. A multi-vendor seat means you get the right sensor for each job rather than the sensor one operator needs to sell.

**Through an online platform.** The newest channel. You browse available coverage over your area, see the price, and order without a sales cycle. [Pera Portal](https://portal.geopera.com) is ours: archive search and tasking across suppliers including WorldView at 30 cm, Beijing-3, Perascope, SuperView and Wyvern's hyperspectral sensors, with the per-km² price shown on the result before you commit to anything.

We sell through the reseller and platform channels, so weigh our view accordingly. What we would tell any buyer regardless: get the all-in price from every channel you compare, because the differences between quotes are usually in the processing and licensing lines, not the headline rate.

## What Satellite Imagery Costs in 2026

Typical commercial rates in USD per km², consistent with the fuller breakdown in our [cost guide](/blog/satellite-imagery-cost-guide):

| Resolution tier | Archive      | Tasking            |
| --------------- | ------------ | ------------------ |
| 30 cm           | $17-26/km²   | $25-55/km²         |
| 50 cm           | $8-18/km²    | $15-35/km²         |
| 1-2 m           | from ~$2/km² | often archive-only |

Four line items decide what you actually pay:

1. **Processing fees.** Orthorectification, pansharpening and atmospheric correction add 30-80% at providers that bill them separately. A $15/km² quote can become $20-27/km² once the imagery is actually usable.
2. **Minimum order areas.** Commonly 25 km² for archive and 50-100 km² for tasking. If your site is 5 km², a 25 km² minimum means paying five times the area you need.
3. **Licensing.** A single-user licence is the default quote. Multi-user or organisation-wide licensing typically adds around 30%.
4. **Priority fees.** Rush tasking and tight capture windows carry premiums.

Our own rate card is published on the [pricing page](/pricing): archive from AUD $3/km², tasking from AUD $14/km², with the full processing chain included in the rate. We publish it precisely so you can do this comparison before talking to anyone.

## Buying SAR Imagery

Synthetic aperture radar is bought through the same channels as optical, with three differences worth knowing before you request a quote.

First, tasking is more dependable: SAR images through cloud and at night, so a capture window is a schedule, not a hope. That makes SAR the default for monitoring programmes in persistently cloudy regions. Second, you specify an imaging mode rather than just a resolution: spotlight modes trade footprint for fine detail, strip modes cover more ground at moderate resolution, and the mode drives the price more than in optical procurement. Third, the product needs more interpretation expertise than an optical image; our [guide to SAR imagery](/blog/sar-satellite-imagery-explained) covers what the data does and doesn't show.

The buying advice is unchanged: compare all-in quotes, and be explicit about the mode and polarisation you need.

## What "Analysis-Ready" Actually Means

A raw satellite capture is not ready to measure from. Between the downlink and your GIS sit five processing steps: orthorectification (correcting geometry against terrain so positions are true), pansharpening (merging the sharp panchromatic band with the colour bands), atmospheric correction (converting to surface reflectance so scenes are comparable), colour balancing, and mosaicking (joining scenes without visible seams).

Providers handle this differently. Some deliver raw files and leave the processing to you. Some bill each step as a line item, which is where the 30-80% surcharge comes from. Some include the chain in the base rate. At Geopera every order runs the full chain at no extra cost, which is [why we process every order](/blog/why-we-process-every-order): the per-km² rate on the rate card is the entire price of a finished, analysis-ready product.

Whichever provider you choose, ask which processing level the quote covers. "Imagery" can mean anything from raw radiance to a finished mosaic, and the difference is real money and real accuracy.

## The Buying Process, Step by Step

1. **Check the archive over your polygon.** Coverage, capture dates, cloud cover. This step is free on platforms with open catalogues and tells you immediately whether tasking is even needed.
2. **Get all-in quotes.** Base rate, processing, licensing, minimums, delivery. Compare the totals rather than the headline rates.
3. **Check the licence.** Who in your organisation may use the imagery, whether it can appear in public deliverables, and whether derived products (maps, analysis outputs) are unrestricted. Regulators and clients usually need derived-product rights, which most licences grant.
4. **Place the order with explicit thresholds.** For tasking, set the capture window and a cloud-cover ceiling (under 10% is a common specification) so an unusable capture is not deliverable.
5. **Receive and verify.** Delivery is typically GeoTIFF into your GIS or cloud storage. Archive orders at Geopera are often delivered the same day; tasking delivers once the capture window and weather allow. Check the imagery against your AOI boundary and the stated processing level before signing off.

## Mistakes That Cost Buyers Money

**Comparing headline rates instead of all-in prices.** The cheapest base rate with separately billed processing regularly totals more than a higher rate with processing included.

**Ignoring minimum order areas on small sites.** For a 5 km² site, the provider's minimum matters more than its per-km² price.

**Over-buying resolution.** Paying 30 cm prices for field-scale vegetation monitoring that free 10 m [Sentinel-2 data](/blog/free-sources-of-satellite-data) serves is the most common overspend we see.

**Treating basemap screenshots as measurement products.** Web-map imagery is a mosaic of unknown dates without survey-grade positioning. If the output faces a regulator, buy imagery with a capture date and [proper orthorectification](/blog/orthorectification-explained); our [Google Earth vs commercial imagery](/blog/google-earth-vs-commercial-satellite-imagery) comparison covers where the line sits.

**Ordering tasking without thresholds.** No cloud ceiling in the order means a 60%-cloud capture can count as delivered.

Our rate card is on the [pricing page](/pricing), and the archive is browsable in [Pera Portal](https://portal.geopera.com) with prices on the results. If you would rather test the product than read about it, you can [apply for a processed sample over your own AOI](/trust), or [tell us what you're weighing up](/contact) and we'll answer plainly, including when free data is the right call.

## Frequently Asked Questions

### Where can I buy satellite imagery?

From three channels: satellite operators directly (suits high-volume programmes on one constellation), resellers that aggregate multiple operators under one agreement, or online platforms where you browse coverage and order with the price shown upfront. For most buyers the practical question is which channel gives an all-in price fastest.

### Can anyone buy satellite imagery?

Yes. Commercial satellite imagery is sold to businesses, governments and individuals; there is no special accreditation to purchase. Licensing terms govern how you may use and share it, and captures over a small number of regions are occasionally restricted, but ordinary commercial purchases involve neither issue.

### How much does it cost to buy satellite imagery?

From about $2/km² for 1-2 m archive imagery to $55+/km² for 30 cm tasking, in USD. The most commonly purchased tier, 50 cm archive, typically runs $8-18/km². Processing adds 30-80% where billed separately. Our [cost guide](/blog/satellite-imagery-cost-guide) breaks down every tier with worked examples.

### Can I buy satellite imagery of any location?

Nearly. Archive coverage varies by location, with populated and economically active areas imaged most often, and tasking can capture almost anywhere on Earth on request. Captures over certain regions are occasionally restricted for regulatory or security reasons; a provider will flag this at quote stage.

### How quickly is satellite imagery delivered?

Archive orders are the fast path: at Geopera they are often delivered the same day, fully processed. Tasking depends on the capture window, satellite schedules and, for optical sensors, weather; delivery follows once a valid capture is made. Ask for the expected window with the quote.

### Can I buy historical satellite imagery?

Yes. Commercial high-resolution archives reach back to the 2000s, and the free Landsat record extends to 1972 at 30 m resolution. Historical archive imagery is bought the same way as recent archive. Our guide to [historical satellite images](/blog/historical-satellite-images) covers the sources and their limits.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# 21AT Data Now Available Through Geopera

> Geopera announces partnership with 21AT, bringing high-quality, affordable satellite imagery to Australian organisations through the Pera Portal

Published: 2025-05-12 | Author: Darcy Weedman | Reading time: 5 min

Source: https://geopera.com/blog/geopera-21at-partner

---
## Geopera Partners with 21AT to Deliver Premium Satellite Imagery at Competitive Prices

We're excited to announce that Geopera has partnered with 21AT, adding their high-quality satellite imagery to our comprehensive data offerings for the Australian market. This strategic collaboration allows our customers to access 21AT's extensive satellite constellation data directly through the Pera Portal, providing exceptional imagery at industry-leading prices.

- [Log in to Pera Portal](https://portal.geopera.com) to view available 21AT data coverage across Australia
- [Contact our Australian team](/contact) for information about specifications, pricing, and availability for your region

## Premium Quality, Exceptional Value

The partnership brings 21AT's impressive satellite fleet to Australian users with unprecedented accessibility and affordability:

- BJ3N satellite imagery at just $40/sq km for 30cm tasking (AUD)
- Archive imagery available for only $25/sq km (AUD)
- Rapid delivery through Geopera's efficient processing pipeline
- Seamless integration with existing workflows in the Pera Portal

These price points represent exceptional value in the high-resolution satellite imagery market while maintaining the premium quality that professional applications demand.

## 21AT's Comprehensive Satellite Constellation

21AT (Twenty First Century Aerospace Technology) has established itself as a leading provider of Earth observation satellite data with an impressive and growing constellation:

### BJ3N Series

The flagship BJ3N satellite delivers stunning 30cm resolution imagery, providing the detail necessary for precise analysis across numerous applications. This level of resolution, previously available only at premium price points, is now accessible at a fraction of traditional costs through our partnership.

### BJ3A Fleet

The constellation includes five BJ3A satellites capturing 50cm resolution imagery, offering an excellent balance between detail and coverage for regional-scale projects where sub-meter resolution is essential.

### Triple Sat Constellation

Complementing the high-resolution offerings, 21AT's three 80cm Triple Sat satellites provide reliable imagery for broader coverage requirements and affordable archive applications.

## Applications: High-Resolution Solutions for Australian Industries

The combination of exceptional resolution and affordable pricing makes 21AT data particularly valuable across multiple Australian sectors:

### Urban Planning and Infrastructure

- Detailed mapping of urban environments at 30cm resolution
- Infrastructure monitoring and development planning
- Property boundary assessment and cadastral mapping updates
- Transportation network analysis and planning

### Agriculture and Land Management

- Precise crop monitoring at field and plant level
- Detailed farm infrastructure mapping and planning
- Property boundary verification and land use classification
- Irrigation system assessment and optimization

### Environmental Monitoring

- Vegetation health assessment and change detection
- Coastal erosion monitoring and management
- Habitat mapping for conservation planning
- Environmental compliance verification

### Mining and Resources

- Detailed site monitoring and planning
- Environmental compliance documentation
- Infrastructure development tracking
- Operation optimization and safety monitoring

The exceptional quality-to-price ratio of 21AT data provides Australian organisations with cost-effective solutions for projects requiring high-resolution imagery without premium pricing constraints.

## Geopera + 21AT: Streamlined Access to Premium Data

Through this partnership, Geopera customers across Australia can now access 21AT's high-resolution satellite data through our Pera Portal platform with several advantages:

- Integration with Geopera's processing pipeline for efficient delivery
- Compatibility with existing geospatial workflows
- Streamlined data access through the intuitive Pera Portal interface
- Local technical support from our Australian-based team
- Significant cost savings compared to traditional high-resolution data sources

This collaboration provides Australian organisations across all sectors with practical, affordable tools to enhance their geospatial capabilities while addressing budget constraints that have traditionally limited access to the highest quality satellite imagery.

## Access 21AT Data Through Geopera Today

For Australian organisations ready to explore how high-resolution, affordable satellite imagery can transform their operations:

1. [Log in to Pera Portal](https://portal.geopera.com) to view available 21AT data coverage across Australia
2. [Contact our Australian team](/contact) for information about specifications, pricing, and availability for your region
3. Experience the perfect balance of quality and affordability with 21AT data

21AT's complete satellite imagery catalog for Australian territories is now available on the Pera Portal.

## Learn More About 21AT

Learn more about 21AT and their satellite constellation at https://www.21at.sg/

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# ESA's Biomass Mission Explained: Peering Inside the World's Forests

> Explore ESA's groundbreaking Biomass mission, using novel P-band radar technology to measure forest carbon storage and unlock secrets of the global carbon cycle.

Published: 2025-05-02 | Author: Darcy Weedman | Reading time: 8 min

Source: https://geopera.com/blog/esa-biomass-explained

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary: ESA's Biomass Mission

ESA's Biomass mission, launched on April 29, 2025, is a pioneering Earth Explorer satellite designed to provide unprecedented insights into the world's forests. Key aspects include:

- **Unique Technology:** It carries the first-ever spaceborne **P-band synthetic aperture radar (SAR)**.
- **Deep Penetration:** The long wavelength (P-band) radar can see through dense forest canopies to measure the woody biomass (trunks, branches) where most carbon is stored.
- **Carbon Cycle Focus:** The primary goal is to reduce uncertainties in how much carbon is stored in forests and how this changes over time, improving our understanding of the global carbon cycle and its role in climate change.
- **Global Mapping:** The mission will create detailed global maps of forest biomass and height.
- **Advanced Techniques:** It uses both **tomographic** (3D imaging) and **interferometric** phases to gather comprehensive data.
- **Beyond Forests:** P-band radar also offers unique capabilities for measuring glacier/ice sheet velocity, mapping sub-surface geology in arid regions, and determining topography beneath dense vegetation.

<AskBand headline="What pairs with the radar?" sub="Optical context for missions like Biomass: Sentinel-2 and Landsat, searchable free." action="Search the archives" href="https://portal.geopera.com" tiles={['/band/glacier.webp', '/band/delta.webp', '/band/bloom.webp']} />

## ESA's Biomass Mission: A New Eye on Global Forests

Forests cover nearly a third of our planet's land surface and are indispensable players in the Earth's climate system. Through photosynthesis, they draw vast amounts of carbon dioxide (CO₂) from the atmosphere, storing carbon primarily in their woody biomass – the trunks, branches, and stems. This natural process is critical for regulating atmospheric CO₂ levels and maintaining climate stability.

However, accurately quantifying the total carbon stored in global forests and tracking its changes over time remains a significant challenge. Factors like deforestation, land-use change, forest degradation, and regrowth, compounded by climate change impacts, create large uncertainties in our understanding of the forest carbon cycle.

To address these critical knowledge gaps, the European Space Agency (ESA) developed the **Biomass mission**. As one of ESA's innovative Earth Explorer missions, Biomass employs groundbreaking technology to deliver crucial data on forest biomass, height, and carbon stocks, enhancing our grasp of forests' role in the complex global carbon cycle.

## The Carbon Cycle Challenge

Human activities, mainly burning fossil fuels and changing land use, have increased atmospheric CO₂ levels by 50% compared to pre-industrial times, significantly impacting our climate. While reducing emissions is paramount, a precise understanding of the natural carbon cycle – how carbon moves between the atmosphere, oceans, and land – is vital for accurate climate modelling and effective mitigation strategies.

Currently, the terrestrial components of the carbon cycle, especially emissions from land-use change and the carbon uptake by land ecosystems, are the least certain parts of the equation. The Biomass mission directly targets this uncertainty by providing accurate, consistent, global measurements of forest biomass, a key indicator of stored carbon on land.

![Global Carbon Cycle Diagram - illustrating stocks and fluxes, highlighting uncertainty in land components.](/blogs/global-carbon-cycle-diagram.webp)
_The global carbon cycle, showing average annual carbon stocks (GtC) and fluxes (GtC per year) for 2014–2023. Terrestrial components show significant uncertainty. (Adapted from Global Carbon Project)_

## What Makes the Biomass Mission Unique? P-Band Radar

Measuring the mass of forests from space is incredibly complex. Traditional optical sensors see the top of the canopy, while shorter-wavelength radar interacts mainly with leaves and smaller branches. The **Biomass mission's breakthrough is its use of P-band synthetic aperture radar (SAR)** – the first of its kind in space.

**Why P-Band?**

- **Long Wavelength:** P-band radar uses a long wavelength (around 70 cm / 435 MHz) that can penetrate deep into the forest canopy, reaching the main trunks and larger branches where the majority (around 75%) of a tree's biomass is stored.
- **All-Weather Capability:** Like other radar systems, P-band can acquire imagery day and night, regardless of cloud cover or weather conditions, ensuring consistent data collection.
- **Structural Information:** By analysing how the P-band signal interacts with and reflects off different parts of the forest structure, scientists can estimate not just biomass but also forest height.

To effectively focus this long-wavelength radar signal onto Earth's surface, the Biomass satellite is equipped with a large **12-metre diameter deployable reflector antenna**.

![Diagram showing P-band radar penetrating forest canopy compared to shorter wavelengths.](/blogs/p-band-penetration.webp)
_Illustration showing how the long wavelength of P-band radar penetrates the forest canopy to interact with trunks and branches, unlike shorter wavelengths._

## How Biomass Measures Forests: Tomography and Interferometry

To achieve its ambitious goals over its planned 5.5-year lifespan, the Biomass mission operates in two distinct phases:

1.  **Tomographic Phase (First ~18 months):**
    - This phase uses techniques similar to a medical CT scan. Multiple images are taken from slightly different orbital positions over the same area.
    - Combining these images allows scientists to create a 3D map of the forest structure, revealing information about the vertical distribution of biomass.
    - This results in a single, detailed global map with a vertical resolution of 15-20 metres and a spatial resolution of 200 metres. Biomass is the first mission to systematically apply this technique from space for forest mapping.

2.  **Interferometric Phase (Remaining ~4 years):**
    - During this phase, the satellite repeatedly observes the same areas. By comparing the phase of the radar signals from different passes (interferometry), scientists can estimate forest height and track changes in above-ground biomass over time.
    - This phase aims to produce around five global maps, allowing for the quantification of carbon fluxes – the crucial variable needed to understand how forest carbon storage is changing and impacting the climate.

## Beyond the Forest: Expanding Earth Observation Frontiers

While forests are the primary focus, the unique capabilities of P-band radar open up exciting possibilities for other areas of Earth observation:

- **Ice Sheets and Glaciers:** P-band can penetrate deeper into ice than shorter wavelengths, making it less susceptible to surface melt or snowfall distortions. This allows for more accurate measurements of glacier and ice-sheet velocities, especially in challenging regions like Antarctica, which Biomass will map using P-band for the first time.
- **Sub-surface Geology:** In arid environments, P-band radar can penetrate dry sand up to several metres deep. This enables the mapping of hidden geological features like ancient riverbeds or palaeolakes, offering insights into past climates and potential fossil water resources.
- **Topography Beneath Vegetation:** By penetrating dense forest canopies, Biomass can reveal the true ground elevation, helping to correct biases present in digital elevation models (DEMs) created using shorter-wavelength radar or optical methods.

## Mission Status and Launch

The ESA Biomass satellite embarked on its mission on **April 29, 2025**. It was launched aboard a **Vega-C rocket** from ESA’s Spaceport in Kourou, French Guiana. This mission represents a significant collaborative effort involving hundreds of people across Europe, Canada, and the USA, with Airbus UK as the prime contractor and Airbus DE leading the development of the P-band radar instrument.

## Conclusion: A Vital Tool for Climate Science

ESA's Biomass mission stands as a testament to innovation in Earth observation. By deploying the first-ever spaceborne P-band radar, it promises to dramatically reduce uncertainties surrounding the role of forests in the global carbon cycle. The data gathered will be invaluable for climate scientists, policymakers, and forest managers worldwide, providing a clearer picture of forest carbon stocks and fluxes.

Beyond its primary objective, Biomass pushes the boundaries of radar remote sensing, offering new ways to study ice dynamics, uncover hidden geological features, and accurately map terrain under dense vegetation. This pioneering mission will undoubtedly enhance our understanding of Earth's complex systems and provide critical information for tackling the challenges of climate change.

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Hyperspectral Arrives: Geopera Partners with Wyvern

> Geopera is excited to announce our partnership with Wyvern, bringing high-resolution hyperspectral satellite imagery to Australian organisations through our platform

Published: 2025-04-22 | Author: Darcy Weedman | Reading time: 6 min

Source: https://geopera.com/blog/geopera-wyvern-partner

---
## Geopera Partners with Wyvern to Bring Hyperspectral Imagery to Australian Users

We're pleased to announce that Geopera has partnered with Wyvern, adding high-resolution hyperspectral satellite imagery to our data offerings for the Australian market. This collaboration enables our customers to access Wyvern's Dragonette constellation data directly through the Pera Portal, providing new analytical capabilities for various applications specific to Australian environments and industries.

For those unfamiliar with hyperspectral imagery and its applications, let's explore what this technology offers:

## Understanding Hyperspectral Imagery

When we refer to "spectral," we're discussing portions of the electromagnetic spectrum—the range of energy wavelengths that includes everything from gamma rays to radio waves. While human vision is limited to the visible light spectrum (approximately 380-750 nanometers), some satellite sensors can detect a much broader spectral range.

Traditional multispectral satellites capture 4-10 colours, or spectral bands. These spectral bands sample some portion of th overall spectral range. Hyperspectral imagery significantly expands this capability by capturing dozens of narrow, contiguous spectral bands across the spectral range, greatly increasing the amount of spectral information taken of the scene.

The value of hyperspectral data lies in its ability to detect distinctive spectral signatures. Each material on Earth has a unique way of reflecting or absorbing different wavelengths based on its molecular composition. With the improved spectral sampling of hyperspectral imagery giving a better represntation of these distinct spectral signatures, imagery can be used to distinguish between similar materials, identify specific vegetation types, detect particular minerals, or monitor environmental conditions with greater precision than conventional multispectral imagery allows.

## Wyvern's Dragonette Constellation

Founded in 2018 in Edmonton, Canada, Wyvern specialises in commercial hyperspectral imaging. Their Dragonette constellation currently includes four satellites (Dragonette-001 to -004) delivering 5.3 m ground sampling distance (GSD) imagery over 23-31 spectral bands in the visible near infrared (VNIR) range.

Wyvern plans to expand the constellation with Dragonette-005, and 006 beginning later this year, which will increase coverage and revisit frequency.

The Dragonette constellation provides 5.3m resolution hyperspectral imagery, offering higher resolution than previously available commercial hyperspectral satellite options. This combination of spatial detail and spectral range makes it suitable for a variety of applications across different industries.

![Aster Satellite Imager over a Minesite](/blogs/wyvern-wa-image.webp)_Dragonette-001 image of Purnululu National Park, Western Australia taken on February 2nd, 2025 01:33:30 UTC (PCA). ©2025 Wyvern Incorporated. All Rights Reserved._

## Applications: Hyperspectral Solutions for Local Challenges

Australia faces unique environmental and agricultural challenges that hyperspectral imagery is particularly well-suited to address:

### Mining Rehabilitation

With Australia's extensive mining operations across Western Australia, Queensland, and New South Wales, environmental rehabilitation monitoring is a critical concern. Hyperspectral data can help:

- Monitor mine site rehabilitation progress with precision
- Detect acid mine drainage issues specific to Australian geological conditions
- Assess native vegetation regrowth on remediated sites
- Ensure compliance with Australia's strict environmental regulations

### Agriculture and Drought Monitoring

Australia's agricultural sector regularly faces drought conditions and requires precise water management. Hyperspectral imagery provides:

- Early stress detection in key Australian crops like wheat, cotton, and sugar cane
- Improved water use efficiency through detailed crop health monitoring
- Better discrimination between similar crop varieties in Australian farming systems
- Support for precision agriculture practices in Australia's variable climate conditions

### Bushfire Management

Following devastating bushfire seasons, Australian land managers need better tools for prevention and recovery assessment. Wyvern's Dragonette constellation offers:

- Enhanced fuel load mapping in eucalypt forests and other Australian vegetation types
- Improved detection of vegetation dryness before it becomes visibly apparent
- Post-fire recovery monitoring of native Australian ecosystems
- Data to support bushfire planning specifically tailored to Australian conditions

The addition of Wyvern's hyperspectral capabilities to Geopera's platform provides Australian organizations with locally-relevant tools to address these national priorities.

## Geopera + Wyvern: Expanding Access to Hyperspectral Data for Australian Users

Through this partnership, Geopera customers across Australia can now access Wyvern's hyperspectral data through our Pera Portal platform with several advantages:

- Integration with Geopera's processing pipeline for efficient delivery
- Compatibility with existing geospatial workflows
- Streamlined data access through the Pera Portal
- Local technical support from our Australian-based team

This partnership provides Australian organisations in agriculture, environmental monitoring, forestry, mining, and related fields with practical tools to enhance their geospatial analysis capabilities while addressing regional challenges.

## Access Wyvern Data Through Geopera

For Australian organisations ready to explore hyperspectral imaging capabilities:

1. [Log in to Pera Portal](https://portal.geopera.com) to view available Wyvern data coverage across Australia
2. [Contact our Australian team](/contact) for information about data specifications and regional availability

Wyvern's hyperspectral data for Australian territories is now available on the Pera Portal.

## Learn More about Wyvern

Learn more about Wyvern at https://wyvern.space/

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Satellite Tasking Explained: How It Works in 2026

> How satellite tasking works in 2026, when to use it over archive imagery, what it costs per km², and how to plan a capture that actually succeeds.

Published: 2025-04-06 | Author: Darcy Weedman | Reading time: 8 minute

Source: https://geopera.com/blog/satellite-tasking-explained

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

Satellite tasking lets you request brand-new, custom imagery of your specific location instead of relying on archive imagery that might be weeks, months, or even years old. Here's what you need to know:

- **What It Is:** Satellite tasking is the process of requesting a satellite to capture fresh imagery of your specific Area of Interest (AOI), giving you up-to-date data when you need it
- **When To Use It:** Ideal for monitoring changing situations like construction projects, natural disasters, environmental changes, or any scenario where recent imagery (within days, not months) is critical
- **Available Options:** Choose from optical imagery (30cm-2m resolution with [WorldView Legion](/sensors/worldview-legion), [SuperView Neo](/sensors/superview-30cm), [Perascope](/sensors/jl1-50cm)) or radar imagery that can penetrate clouds and capture imagery regardless of weather or daylight conditions
- **Tasking Timeline:** From request to delivery, expect 3-14 days depending on satellite availability, weather conditions, and priority level (rush options available for time-sensitive applications)
- **Budget Considerations:** Tasking runs AUD $14-55 per km² depending on resolution (the full rate card is on our [pricing page](/pricing)), while archive imagery of the same resolution typically costs 30-70% less
- **Expert Planning:** Improve acquisition odds by considering cloud cover, revisit frequencies, incidence angles, and collection windows when placing your tasking request through [our tasking service](/tasking)
- **Processing Options:** Raw imagery delivery or value-added services available including orthorectification, atmospheric correction, and advanced analytics to make your imagery immediately actionable

<AskBand headline="What would a fresh capture cost?" sub="Current price bands per resolution tier are published." action="See tasking rates" href="/pricing" tiles={['/band/city.webp', '/band/dunes.webp', '/band/saltponds.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## Introduction

In the past, getting your hands on satellite imagery was a bureaucratic nightmare. Even when commercial satellite imagery became available in the 1990s, the process was painfully slow and complex. By the time you finally received your imagery, it was often outdated and no longer useful for your needs.

Today the process is direct. Instead of being limited to whatever archive imagery might be available, you can request that satellites capture brand-new images of your exact area of interest, often delivered within days.

This guide will walk you through everything you need to know about satellite tasking: how it works, when you should use it, what options are available, and how to ensure you get the best possible results for your specific needs.

## What Is Satellite Tasking?

Satellite tasking is the process of requesting a satellite operator to direct their satellite to capture fresh imagery of your specified Area of Interest (AOI). Unlike archive imagery, which consists of previously captured images stored in a database, tasked imagery is collected specifically for you based on your unique requirements.

### How Satellite Tasking Evolved

The ability for commercial entities to task satellites is relatively recent in the history of remote sensing:

- **Pre-1990s:** Satellite imagery was exclusively controlled by governments and military agencies
- **1992:** Russia began selling satellite imagery to commercial customers
- **1994:** The US allowed private companies to build and launch commercial imaging satellites
- **1999:** IKONOS-2 became the first high-resolution satellite available for commercial tasking
- **2010s:** Web platforms began streamlining the satellite tasking process
- **Today:** Modern platforms like [Geopera's Pera Portal](https://portal.geopera.com) enable direct, user-friendly satellite tasking with delivery times measured in days, not months

### Why Task a Satellite?

There are several compelling reasons to task a satellite rather than relying on archive imagery:

1. **Currency:** You need the most up-to-date view of a location
2. **Specificity:** You require imagery with exact parameters (resolution, angle, time of day)
3. **Availability:** No suitable archive imagery exists for your area of interest
4. **Custom Collection:** You need repeated captures over time to monitor changes
5. **Quality Control:** You want to ensure cloud-free, high-quality imagery

## Types of Satellite Tasking

When tasking a satellite, you'll need to choose between two main types of imagery: optical and radar. Each has distinct advantages for different applications.

### Optical Satellite Tasking

Optical satellites capture imagery similar to traditional cameras but with capabilities far beyond what the human eye can see.

**Key Advantages:**

- High-resolution imagery (as sharp as 30cm with [WorldView Legion](/sensors/worldview-legion))
- Natural-looking images that are easy to interpret
- Multispectral capabilities including visible and invisible bands (RGB, NIR, SWIR)
- Ideal for visual assessment, mapping, and monitoring visible changes

**Limitations:**

- Cannot penetrate clouds (success depends on clear weather)
- Requires daylight for collection
- May require multiple collection attempts in cloudy regions

**Best For:**

- Urban monitoring and planning
- Infrastructure inspection
- Agricultural assessment
- Land use classification
- Environmental change detection

**Available Sensors Through Geopera:**

- [WorldView Legion](/sensors/worldview-legion): 30cm resolution, 6-satellite constellation
- [SuperView Neo](/sensors/superview-30cm): 30cm resolution, 4-satellite constellation
- [Perascope 50cm](/sensors/jl1-50cm): 50cm resolution, frequent revisits
- [WorldView-3](/sensors/worldview-3): 30cm resolution with 16 spectral bands

### Radar Satellite Tasking

Radar satellites emit their own energy pulses and measure the return signal, creating images based on the reflected data.

**Key Advantages:**

- All-weather imaging capability (can penetrate clouds, rain, and smoke)
- Day and night operation (doesn't require sunlight)
- Guaranteed acquisition on first attempt
- Provides unique data on structure, elevation, and surface characteristics

**Limitations:**

- Lower resolution than optical imagery
- More complex to interpret visually
- Typically more expensive than optical imagery

**Best For:**

- Monitoring during monsoon seasons or in frequently cloudy regions
- Emergency response during storms or poor visibility conditions
- Infrastructure monitoring regardless of weather conditions
- Surface deformation and subsidence detection
- Maritime surveillance and ship detection

**Available Through Geopera:**

- SAR (Synthetic Aperture Radar) options available upon request

## How to Task a Satellite

The satellite tasking process has been dramatically simplified through Geopera's portal, but understanding each step will help ensure you get exactly the imagery you need.

### 1. Define Your Area of Interest (AOI)

Your first step is to precisely define the geographical area you want captured:

- **Web Interface:** Draw a polygon directly on the map in [Pera Portal](https://portal.geopera.com)
- **Upload Option:** Upload a KML, GeoJSON, or Shapefile of your AOI
- **API Integration:** Define your AOI programmatically through Geopera's API

**Best Practices for AOI Definition:**

- Keep your AOI as compact as possible to minimise costs
- Ensure your polygon has no self-intersections or topology errors
- Limit complexity to under 1,000 vertices for optimal processing
- Include a small buffer around critical features to ensure complete coverage

### 2. Specify Your Requirements

After defining your AOI, you'll need to specify your collection parameters:

- **Satellite Selection:** Choose specific satellites or let Geopera recommend the best option
- **Resolution:** Select the spatial resolution needed (30cm, 50cm, 75cm, 1m, 2m, etc.)
- **Collection Window:** Set the time period during which imagery should be captured
- **Cloud Cover:** Specify maximum acceptable cloud coverage percentage (typically 15-20% or less)
- **Off-Nadir Angle:** Define acceptable viewing angles (affects image quality)
- **Spectral Bands:** Specify which bands you need (RGB, NIR, SWIR, etc.)
- **Priority Level:** Set standard or rush priority (affects price and timeline)

### 3. Review Feasibility and Pricing

After submitting your requirements, Geopera's team will:

- Assess the feasibility of your request based on satellite availability and conditions
- Calculate the probability of successful acquisition within your timeframe
- Provide a detailed quote based on your specifications
- Suggest alternatives if your original requirements present challenges

**Typical Pricing Factors:**

- AOI size (larger areas cost more)
- Resolution (higher resolution commands premium pricing)
- Priority level (rush tasking costs more than standard)
- Specific satellites requested (some sensors have premium pricing)
- Special processing requirements

### 4. Confirm and Track Your Order

Once you've approved the quote:

- Your tasking request enters the collection queue
- You'll receive updates on planned acquisition dates
- You can track the status of your order through the Pera Portal dashboard
- You'll be notified when the satellite successfully captures your imagery

### 5. Receive and Utilise Your Imagery

Upon successful collection:

- Your imagery undergoes quality control and processing
- The data is delivered through Pera Portal in your preferred format
- Additional processing options are available (orthorectification, analytics, etc.)
- Cloud storage and API access options are available for enterprise users

## Maximising Satellite Tasking Success

To ensure the highest probability of successful acquisition, consider these expert tips:

### Understand Revisit Rates

Different satellites have different revisit capabilities, which affects how quickly they can capture your AOI:

- **Daily Revisit:** [WorldView Legion](/sensors/worldview-legion) (constellation of 6 satellites)
- **2-3 Day Revisit:** [SuperView Neo](/sensors/superview-30cm) (constellation of 4 satellites)
- **Multiple Daily Passes:** [Perascope 75cm](/sensors/jl1-75cm) (constellation of 50+ satellites)

### Plan Around Weather Patterns

For optical imagery, cloud cover is the primary challenge:

- **Cloud Cover Assessment:** Geopera provides historical cloud cover analysis to choose optimal collection windows
- **Seasonal Planning:** Time your tasking requests during typically clear seasons for your region
- **Extended Collection Windows:** For challenging areas, set longer collection windows to increase success probability
- **Radar Alternative:** Consider radar imagery for consistently cloudy regions

### Consider Collection Angles

The satellite's angle relative to your AOI affects image quality:

- **Near-Nadir (0-20°):** Provides the least distortion but may have limited availability
- **Off-Nadir (20-30°):** Good balance of quality and availability
- **Extreme Off-Nadir (>30°):** May have significant distortion but enables quicker acquisition

### Set Realistic Timeframes

Understanding typical timelines helps set proper expectations:

- **Rush Tasking:** 3-7 days from order to delivery (weather permitting)
- **Standard Tasking:** 7-14 days from order to delivery
- **Complex AOIs or Specific Requirements:** May require additional time

## Common Applications for Satellite Tasking

Satellite tasking is particularly valuable for these applications:

### Construction and Development Monitoring

Track progress, verify timelines, and document site conditions with regular updates:

- **Initial Site Assessment:** Capture pre-construction conditions
- **Progress Monitoring:** Task satellites bi-weekly or monthly to document development
- **Compliance Verification:** Document adherence to permits and regulations
- **Dispute Resolution:** Maintain a visual timeline of project evolution

### Disaster Response and Assessment

Get current imagery immediately following natural disasters:

- **Damage Assessment:** Quantify affected areas after floods, fires, or storms
- **Response Planning:** Guide emergency teams with current ground conditions
- **Insurance Documentation:** Provide evidence for claims processing
- **Recovery Monitoring:** Track reconstruction efforts over time

### Agricultural Management

Monitor crop conditions throughout the growing season:

- **Pre-Planting Assessment:** Evaluate field conditions before planting
- **Growth Monitoring:** Track crop development at critical stages
- **Stress Detection:** Identify irrigation issues or disease outbreaks early
- **Harvest Planning:** Optimise timing based on current crop conditions

### Environmental Monitoring

Document changes to sensitive ecosystems:

- **Deforestation Tracking:** Monitor forest clearing activities
- **Pollution Events:** Document spills or contamination
- **Habitat Changes:** Track wildlife habitat modifications
- **Climate Impact Assessment:** Document effects of climate change on specific regions

## Satellite Tasking vs. Archive Imagery

Understanding when to use each option will help optimise your imagery budget:

### When to Use Archive Imagery

- **Historical Analysis:** When you need to understand past conditions
- **Budget Constraints:** Archive imagery typically costs 30-70% less than tasked imagery
- **Immediate Needs:** When you need imagery right away (instant delivery)
- **Low-Change Areas:** For regions where conditions change slowly
- **Initial Assessment:** For preliminary analysis before deciding on tasking

### When to Choose Satellite Tasking

- **Current Conditions:** When you need to see exactly what's happening now
- **Specific Parameters:** When you need precise angles, times, or conditions
- **Monitoring Programs:** For consistent, scheduled imaging of the same area
- **Recent Changes:** Following construction, natural events, or other modifications
- **Quality Requirements:** When you need guaranteed cloud-free, high-quality imagery

## Processing Options for Tasked Imagery

Once your imagery is captured, Geopera offers various processing levels to suit your needs:

### Basic Processing

- **Radiometric Correction:** Adjusting for sensor characteristics
- **Geometric Correction:** Basic alignment and distortion removal
- **Band Composites:** Standard RGB, Color-Infrared, or custom band combinations

### Advanced Processing

- **Orthorectification:** Precise geometric correction using elevation models
- **Atmospheric Correction:** Removing atmospheric effects for true reflectance values
- **Pansharpening:** Enhancing resolution by combining panchromatic and multispectral data
- **Mosaicking:** Seamlessly combining multiple images

### Value-Added Analytics

- **Change Detection:** Automated comparison with previous imagery
- **Feature Extraction:** Identifying specific objects or infrastructure
- **Vegetation Indices:** NDVI, EVI, and other indicators of plant health
- **Classification:** Land use/land cover mapping
- **3D Products:** Digital Surface Models and elevation products

## Conclusion

Satellite tasking has transformed how organisations access current, high-quality imagery. By understanding the process, options, and best practices outlined in this guide, you can make informed decisions about when and how to use this powerful capability.

Whether you're monitoring construction, responding to disasters, tracking environmental changes, or managing agricultural operations, tasked imagery gives you a capture on your terms rather than whatever the archive happens to hold.

Ready to task a satellite for your specific area of interest? Our [satellite tasking service](/tasking) walks through scoping a capture with the price per km² on screen, or [contact our team](/contact) for guidance on feasibility and timing.

## FAQs About Satellite Tasking

**Q: How much does satellite tasking typically cost?**  
A: Geopera's tasking rates run from AUD $14 per km² (75cm resolution) to AUD $40-55 per km² (30cm), with the full rate card published on our [pricing page](/pricing). The final price depends on AOI size, resolution, urgency, and processing needs; archive imagery of the same resolution typically costs 30-70% less.

**Q: How long does the satellite tasking process take?**  
A: Standard tasking typically takes 7-14 days from order to delivery, while rush tasking can be completed in 3-7 days, weather permitting. The timeline depends on satellite availability, weather conditions, and your specified parameters.

**Q: What if clouds obstruct my area of interest during collection?**  
A: For optical imagery, Geopera will continue collection attempts until acceptable imagery is obtained or your collection window closes. Our feasibility analysis includes cloud cover prediction to maximise success. Alternatively, radar imagery can be used for cloud-penetrating capabilities.

**Q: Can I task satellites for any location on Earth?**  
A: While commercial satellites can image most of the Earth, some restrictions apply based on international regulations and security considerations. Geopera will advise you if your AOI falls under any restrictions. See our [supported countries](/supported-countries) page for more details.

**Q: What's the difference between satellite tasking and satellite data subscriptions?**  
A: Tasking is a one-time request for new imagery of a specific area, while subscriptions provide regular, scheduled imagery of your areas of interest over time. For consistent monitoring needs, subscriptions typically offer better value. Learn more about our [subscription options](/enterprise).

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Geopera Joins Maxar Intellgence's Partner Ecosystem

> Geopera, a leading provider of commercial geospatial data solutions, today announced its strategic partnership with Maxar Intelligence

Published: 2025-04-03 | Author: Darcy Weedman | Reading time: 4 min

Source: https://geopera.com/blog/geopera-maxar-partner

---
**Brisbane, Queensland, Australia - April 3, 2025** - Geopera, a leading provider of commercial geospatial data solutions, today announced its strategic partnership with Maxar Intelligence, a global leader in secure, precise geospatial intelligence. Through this partnership, Geopera will distribute Maxar’s industry-leading products and solutions across Australia, focusing specifically on the environmental, resources, agricultural, and energy sectors.

Maxar uses the power of geospatial data and technology to help customers get mission-critical insights, faster. The company owns and operates the most advanced Earth imaging constellation of satellites, which collect the highest resolution satellite imagery commercially available. Maxar builds cutting-edge products and solutions to enable customers to access reliable answers for actionable intelligence. Geopera’s customers can now learn more about the power of Maxar’s innovative geospatial solutions directly through the Pera Portal—Geopera’s proprietary data platform.

Geopera's unique processing pipeline will further enhance Maxar's data, significantly reducing the time required for delivery while maintaining exceptional data quality. This integration empowers customers to rapidly access precise, analysis-ready data, enabling faster, smarter decision-making.

## About Geopera

Geopera is a premier provider of analysis-ready geospatial data. We specialise in the processing of raw satellite imagery into large scale mosaic datasets delivered to the specific requirements of our users. Our comprehensive Pera Portal platform enables organisations transparent and affordable access to the world's highest quality satellite data, delivering everything needed for geospatial projects of any scale.

## About Maxar Intelligence

Maxar Intelligence is a leading provider of secure, precise geospatial intelligence. Operating the most advanced commercial Earth observation constellation on orbit, we use the power of very high-resolution satellite imagery and software technology to deliver mission success on Earth and in space. Our secure, AI-powered products and services deliver ground truth in near real-time to keep nations safe, improve navigation, protect our planet, speed up disaster response and more. For more information, visit www.maxar.com.

## Contact

Explore Maxar's solutions on the [Pera Portal](https://portal.geopera.com) today, or [Contact Us](/contact) to get started.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Which Satellite Imagery Should You Buy? Matching Sensors to Applications

> Which satellite sensor fits which job: resolution, spectral bands, revisit rate and swath compared across eight common applications, and how to order each.

Published: 2025-04-01 | Author: Darcy Weedman | Reading time: 9 minute

Source: https://geopera.com/blog/best-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

There is no single right answer. The sensor that suits your project depends on how small the objects are that you need to see, which wavelengths carry your signal, how often you need a fresh capture, and how much ground you have to cover. Those four constraints usually pick the sensor for you.

- Finest detail: [WorldView-3](/sensors/worldview-3), 30 cm native, with super-resolution enhancement to 15 cm.
- Mineral exploration: [WorldView-3 SWIR](/sensors/worldview-3), 8 SWIR bands in the 1.0 to 2.5 μm range, which separate clay, sulfate and iron-oxide signatures.
- Fine spectral discrimination: Wyvern Dragonette, 23 to 32 bands at 18 to 35 nm spectral resolution.
- Cost per km²: [Perascope 50 cm](/sensors/jl1-50cm), where 50 cm is enough detail.
- Frequent revisit: [WorldView Legion](/sensors/worldview-legion) at daily, [SuperView Neo](/sensors/superview-30cm) every 2 days, Perascope with 60+ satellites.
- Agriculture: [WorldView-2](/sensors/worldview-2), 8 bands including Red Edge, which responds to chlorophyll before stress is visible in RGB.
- Large-area mapping: [Perascope 75 cm](/sensors/jl1-75cm), 150 km swath.

<AskBand headline="Which sensor fits the job?" sub="Compare them side by side; archive search is free, priced per km² on each result." action="Compare sensors" href="https://portal.geopera.com" tiles={['/band/city.webp', '/band/atoll.webp', '/band/pivots.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## Start With the Constraint, Not the Sensor

Most people approach this backwards. They ask which satellite is best, get a list of specifications, and are no closer to a decision.

The faster route is to work out which constraint actually binds on your project. Usually only one does.

If you need to identify individual vehicles, resolution binds and everything else is negotiable. If you are mapping alteration mineralogy, spectral coverage binds and 30 cm panchromatic detail is beside the point. If you are tracking construction fortnightly, revisit binds. If you are mapping a state, swath width and price per km² bind together, and 30 cm imagery will bankrupt the project long before it finishes.

The sections below are organised by constraint rather than by satellite.

## Finest Detail

[WorldView-3](/sensors/worldview-3) resolves more than anything else commercially available.

- 30 cm native resolution
- Super-resolution enhancement to 15 cm
- Geolocation accuracy of 5 m CE90 without ground control points

That level of detail separates individual vehicles, small infrastructure elements and larger equipment, which 50 cm imagery renders as ambiguous blobs. It suits defence and intelligence work, infrastructure inspection, detailed urban mapping, construction monitoring on high-value projects, and archaeology.

It also costs the most per km², so it rewards a tightly drawn area of interest. Buying 30 cm over a whole lease when you only need it over the plant is a common and expensive mistake.

## Mineral Exploration

For geology, the shortwave infrared bands on [WorldView-3](/sensors/worldview-3) do the work that panchromatic resolution cannot.

- 8 SWIR bands covering roughly 1.0 to 2.5 μm
- Detects mineralogy invisible to RGB sensors
- Penetrates thin cloud, smoke and haze

Those bands pick up the diagnostic absorption features of clay minerals, sulfates and carbonates, which is how hydrothermal alteration gets mapped from orbit. Alteration is an indicator rather than proof of a deposit, but it narrows a large licence area to a ranked target list before any ground team is mobilised.

The [mineral exploration guide](/blog/mining-satellite-imagery) covers the method in more depth, including which mineral groups separate cleanly and what the technique cannot do. The [WorldView-3 alteration mapping case study](/case-studies/wv3-mineral-exploration) has the band ratio formulations and classification workflow.

## Fine Spectral Discrimination

Where multispectral sensors sample the spectrum at 4 to 16 points, hyperspectral instruments sample it continuously. The Wyvern Dragonette series sits in this category.

- Dragonette-001: 23 bands, 500 to 800 nm
- Dragonette-002 and 003: 32 bands, 445 to 880 nm
- Spectral resolution of 18 to 35 nm

Continuous coverage detects differences that fall between the bands of a multispectral sensor: early disease indicators in crops, specific pollutants in water, material identification. It suits research, water quality analysis and pollution monitoring more than routine mapping.

## Cost per Square Kilometre

[Perascope 50 cm](/sensors/jl1-50cm) is the volume option.

- 50 cm resolution
- 4 multispectral bands
- Large constellation, so archive availability is good

For a great many jobs, 50 cm is sufficient: buildings, infrastructure change, land use classification, field-level agriculture. The saving over 30 cm often pays for more frequent monitoring or a larger footprint, which is usually worth more to a project than the extra detail would have been. Current per-km² rates for each tier are in the [satellite imagery cost guide](/blog/satellite-imagery-cost-guide).

## Frequent Revisit

When you need the same location repeatedly, constellation size matters more than any individual satellite's specifications.

**[WorldView Legion](/sensors/worldview-legion)** has 6 satellites at 30 cm with daily revisit, weighted toward high-demand areas.

**[SuperView Neo](/sensors/superview-30cm)** has 4 satellites at 30 cm, revisiting every 2 days at mid-latitudes.

**[Perascope](/sensors/jl1-50cm)** runs 60+ satellites at 50 cm and 75 cm, giving multiple passes per day in many places.

For monitoring, consistency between captures usually beats maximum resolution. A regular series at a stable view angle supports change detection far better than occasional sharper images taken from wherever the satellite happened to be. Making captures from different sensors comparable is a processing problem, which is covered below.

## Agriculture and Vegetation

[WorldView-2](/sensors/worldview-2) carries 8 multispectral bands, and the one that matters for crops is Red Edge.

- Red Edge band, sensitive to chlorophyll content
- 50 cm resolution, enough for field-level assessment
- Coastal Blue band, which improves atmospheric correction
- Yellow band, which helps discriminate vegetation types

Red Edge responds to plant stress before it is visible in RGB or standard 4-band imagery, which is the difference between intervening and documenting. The [red edge guide](/blog/red-edge-remote-sensing) covers the physics and the index formulations.

## Urban Planning and Infrastructure

[SuperView Neo](/sensors/superview-30cm) balances detail against update frequency for development monitoring.

- 30 cm resolution
- 4 bands, RGB plus NIR
- 4-satellite constellation
- Wide swath for efficient area coverage

Rapidly developing urban areas need both detail and currency, and SuperView Neo is usually the cheaper way to get both than tasking a single high-resolution satellite repeatedly.

## Disaster Response

[WorldView Legion](/sensors/worldview-legion) responds fastest.

- 6 satellites, so most locations can be reached quickly
- 30 cm for damage assessment
- Daily revisit to track an evolving situation
- Enough imaging capacity to cover a large affected area

In an emergency, time to first capture dominates every other consideration. A constellation reaches an arbitrary location sooner than any single satellite can.

## Large-Area Mapping

[Perascope 75 cm](/sensors/jl1-75cm) covers ground faster than anything else in the catalogue.

- 150 km swath, roughly six times wider than most high-resolution satellites
- 50+ satellites
- 75 cm resolution

Wide swath plus a large constellation means regional and national mapping finishes in days rather than months. It suits land cover assessment, forestry, watershed management and national infrastructure inventory.

## Satellite, Aerial or Drone?

Sometimes the answer is not a satellite at all. Aerial survey resolves finer and drone survey finer still, but both need aircraft, crew, airspace approval and a weather window, and neither scales to a whole region cheaply. Satellite wins on area, on repeat coverage, and on being able to buy a capture over somewhere you cannot easily fly. The [drone versus aerial versus satellite comparison](/blog/drone-vs-aerial-vs-satellite-imagery) works through where each one stops making sense.

## Choosing, in Order

1. **Resolution.** What is the smallest object you must identify? That sets the floor.
2. **Spectral bands.** Is your signal in the visible range, or does it need Red Edge, SWIR or continuous hyperspectral coverage?
3. **Revisit.** One capture, or a time series? A series constrains which constellations qualify.
4. **Area.** A single site, or a region? Swath and per-km² price start to dominate above a few hundred km².
5. **Archive or tasking.** Does an existing capture answer the question, or do you need a new one on a specified date?
6. **Processing.** Are you receiving something analysis-ready, or budgeting time to correct it yourself?

## Ordering and Delivery

You can search every sensor above in one place at [Pera Portal](https://portal.geopera.com/): draw or upload an area of interest, compare what is available across providers, see the price per km² on each result, and order only the footprint you need.

Delivery is where multi-sensor projects tend to come apart. Captures from different satellites disagree with each other, in band positions, view geometry and radiometry, so a time series assembled from several of them can show change that is an artefact of the instrument.

Every order runs through the same chain before it reaches you: ingest, co-registration, ground control and ortho, pansharpening, atmospheric correction, cloud and shadow masking, seamline optimisation, harmonisation and QC, with nothing manual between stages. That is the [Legato process](/legato). It means sensors that disagree are made to behave like one instrument, and imagery arrives analysis-ready in under 24 hours.

For ongoing monitoring, [talk to us](/contact) about setting up a recurring capture programme rather than ordering ad hoc.

## Common Questions

**How recent is commercial satellite imagery?**
The freshest archive imagery is typically hours to days old, depending on the satellite, your location and cloud cover. If you need a specific date, task a new capture rather than searching the archive.

**Can satellite imagery see through clouds?**
Optical imagery cannot. SWIR bands penetrate thin cloud, haze and smoke, but not solid cover. For that you need radar, which images through cloud and at night.

**What is the practical difference between 30 cm and 50 cm?**
30 cm delivers roughly 2.8 times more pixels over the same ground. In practice it is the difference between counting vehicles and knowing that something is parked there.

**How often can I get imagery of the same location?**
Most locations can be imaged daily to weekly. Mid-latitudes do better than equatorial or polar regions, and revisit depends on the constellation rather than the individual satellite.

**Is Google Earth imagery the same as commercial satellite imagery?**
No. Google Earth mixes commercial satellite imagery, aerial photography and other sources, often months or years old and processed to look good rather than to measure accurately. Commercial imagery is current, radiometrically consistent and carries the metadata technical work requires. The [Google Earth comparison](/blog/google-earth-vs-commercial-satellite-imagery) covers the differences.

**Can I order SuperView imagery directly?**
Yes. SuperView Neo is in the archive alongside every other sensor listed here, searchable and priced per km² at [portal.geopera.com](https://portal.geopera.com), with specifications on the [SuperView Neo sensor page](/sensors/superview-30cm).

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Satellite Imagery for Mineral Exploration and the Mining Lifecycle

> How SWIR satellite imagery maps alteration minerals for exploration, measures disturbance for MRF and PRCP reporting, and tracks rehabilitation vegetation.

Published: 2025-03-20 | Author: Darcy Weedman | Reading time: 9 min

Source: https://geopera.com/blog/mining-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

Satellite imagery is used in mineral exploration to map hydrothermal alteration minerals from orbit, before any ground team is deployed. Sensors that record shortwave infrared can distinguish clays, sulfates, iron oxides and carbonates by their absorption features, which narrows a large licence area down to a short list of targets. The same imagery later measures disturbance for regulators and tracks vegetation through rehabilitation.

- Exploration: SWIR bands identify alteration assemblages; panchromatic imagery maps the faults and fractures that control where mineralising fluids went.
- Operations: repeat captures measure disturbed area for Western Australia's Mining Rehabilitation Fund and Queensland's Progressive Rehabilitation and Closure Plan.
- Closure: multispectral indices quantify vegetation cover and plant health across rehabilitated ground.
- The limit worth knowing: satellite spectral mapping finds surface alteration. It does not find ore, and it cannot see through cover.

<AskBand headline="What is happening across the tenement?" sub="Archive over any tenement is searchable free, with per-km² prices on the results." action="Search the archive" href="https://portal.geopera.com" tiles={['/band/mine.webp', '/band/terraces.webp', '/band/dunes.webp']}
secondary={{ label: 'or talk to us', href: '/contact' }}
/>

## Satellite Imagery for Mineral Exploration

Hydrothermal alteration leaves a mineralogical halo around many deposits, and those minerals have diagnostic absorption features in the shortwave infrared. A sensor with enough SWIR bands can tell them apart from orbit.

[WorldView-3](https://geopera.com/sensors/worldview-3) is the instrument most exploration teams reach for. Its SWIR sensor records 8 bands between 1195 and 2365 nm at 3.7 m resolution, which is fine enough to separate individual alteration minerals rather than lumping them into a general "clay" class. In a copper exploration project run with Arc GeoTech, that band set resolved four groups:

- Phyllosilicates: kaolinite, illite, smectite, muscovite, chlorite
- Sulfates: alunite, jarosite, gypsum
- Iron minerals: hematite, goethite, pyrite
- Carbonates: calcite, dolomite

Separating those matters because the assemblage carries information the presence of alteration alone does not. Alunite and kaolinite together point somewhere different from chlorite and calcite. The full method, including the band ratio formulations and the Spectral Angle Mapper classification against reference spectra, is written up in the [WorldView-3 alteration mapping case study](/case-studies/wv3-mineral-exploration).

ASTER is the other name that comes up, and it deserves a caveat. Its 30 m SWIR data underpins a large part of the published exploration literature, but the SWIR subsystem has been unusable since 2008. Work using ASTER SWIR today is working from the pre-2008 archive, which is still valuable for regional reconnaissance and completely unhelpful if you need current conditions.

![ASTER SWIR composite over Morenci copper mine, Arizona](/aster-arizona-morenci-mine.webp)

### Structure, Not Just Spectra

Alteration tells you where fluids reacted. Structure tells you how they got there.

High-resolution panchromatic imagery, at 30 cm, resolves lineaments that are invisible at moderate resolution: faults, fractures, and the intersections between them. In the Arc GeoTech work, lineament analysis revealed a high-density fault network with primary NW-SE and NNW-SSE trends, and the mapped alteration zones correlated strongly with those structures.

That correlation is the useful part. An alteration anomaly sitting on a structural intersection is a better target than an equally strong anomaly sitting on nothing, and you can only make that call if you have mapped both.

### What Spectral Mapping Cannot Do

Worth stating plainly, because the exploration literature is full of enthusiasm and thin on limits.

Satellite spectral mapping detects surface mineralogy. If the prospective geology is under transported cover, soil or dense vegetation, the sensor sees the cover. It also detects alteration, which is an indicator, not ore. Plenty of alteration systems are barren. What the imagery buys you is a ranked target list and a reason to put the drill rig in one valley rather than another, which is worth a great deal when ground access is expensive and the licence area is large.

## Satellite Imagery for Operating Mine Sites

Once a site is operating, the question changes from "where should we look" to "what is the disturbed area this quarter, and can we prove it".

Repeat satellite capture answers that. Western Australia's Mining Rehabilitation Fund levy is calculated on disturbance, and Queensland's Progressive Rehabilitation and Closure Plan requires demonstrated progress against a schedule. Both need an area measurement somebody else can check. Imagery with a timestamp and a documented processing chain gives you that in a form a regulator will accept.

Operationally, the same captures let teams:

- Measure disturbed area for regulatory reporting
- Track progressive rehabilitation against the approved plan
- Monitor impacts beyond the immediate lease boundary
- Document compliance with approved mining plans
- Watch infrastructure progress: haul roads, waste rock dump expansion, tailings storage facility construction

![High Resolution Satellite Image of a Mine Site](/minesite.webp)

The comparison that matters is against the alternative. Ground survey is more accurate over a small area and becomes expensive over a large one. Aerial survey is sharper and needs an aircraft, a window of good weather, and scheduling. Satellite capture covers the whole lease in one pass and can be ordered against a specific date. For a quarterly disturbance figure across thousands of hectares, that trade usually favours satellite. For a volumetric on a single stockpile, it usually does not.

## Satellite Imagery for Rehabilitation and Closure

Rehabilitation reporting is where satellite imagery earns its place most clearly, because the reporting period runs for years and the evidence has to be consistent across all of it.

Multispectral captures let environmental teams:

- Quantify vegetation cover across rehabilitated areas
- Assess plant health using [vegetation indices](https://geopera.com/blog/remote-sensing-vegetation-indices)
- Track establishment of target vegetation communities against the closure criteria
- Flag areas where establishment is failing, early enough to intervene
- Detect invasive species spread, which tends to show up as a spectral mismatch against the intended community

The before-and-after record is the deliverable for MRF and PRCP reporting. It is also the thing that is impossible to reconstruct retrospectively: if nobody captured the pre-disturbance baseline, no amount of later imagery will produce it. Archive imagery can sometimes fill that gap, which is one of the more common reasons mining clients go looking for [old satellite images](/blog/historical-satellite-images).

## Making Captures Comparable Across Years

A rehabilitation record spanning eight years will draw on several sensors, because constellations change. Comparing them is not automatic. Different sensors have different band positions, different view geometries and different radiometric behaviour, so a vegetation index calculated naively across two of them will show a change that is an artefact of the instrument rather than anything on the ground.

Every Geopera order runs through the same chain before delivery: ingest, co-registration, ground control and ortho, pansharpening, atmospheric correction, cloud and shadow masking, seamline optimisation, harmonisation and QC, with nothing manual between stages. That is the [Legato process](/legato), and its point here is narrow but important: sensors that disagree are made to behave like one instrument, so a time series assembled from several of them is measuring the site rather than the satellite.

## Getting Started

Search the archive over your tenement at [portal.geopera.com](https://portal.geopera.com) and see what exists and what it costs before committing to anything. Pricing is per area rather than per capture, which matters for monitoring work where you want several dates over the same footprint. The [satellite imagery cost guide](/blog/satellite-imagery-cost-guide) covers what each resolution tier costs, and the [mining page](/mining) covers how we work with exploration and environmental teams.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Vegetation Management in Australia Using Satellite Data

> How satellite imagery is transforming vegetation monitoring and management across Australias diverse landscapes with focus on high-resolution options.

Published: 2025-03-10 | Author: Darcy Weedman | Reading time: 7 min

Source: https://geopera.com/blog/vegetation-management-using-satellite-data

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

**TL;DR: Satellite imagery has revolutionized vegetation management across Australia, offering cost-effective monitoring for properties of any size. While free satellite data works for many applications, high-resolution commercial imagery delivers superior results for compliance verification, carbon projects, and precision agriculture. Geopera provides access to both, including cost-effective 50cm imagery and specialized SWIR data.**

Australia's vast landscapes present unique challenges in vegetation management. Traditional ground surveys are expensive and time-consuming, often missing critical changes. Satellite monitoring has emerged as the game-changer, providing frequent, detailed views of vegetation at any scale. This blog explores why satellite data is transforming how we monitor Australia's vegetation and when higher-resolution commercial options deliver additional value.

<AskBand headline="How is your vegetation tracking this season?" sub="NDVI on free Sentinel-2 is a browser tab away." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/fields.webp', '/band/hills.webp', '/band/bloom.webp']} />

## The Challenge of Vegetation Management in Australia

Managing vegetation across Australia's diverse ecosystems is no small feat. From dense tropical forests in Queensland to arid rangelands in Western Australia, each environment presents unique monitoring challenges. Traditional approaches like on-ground surveys and aerial photography are not only costly but also provide just a snapshot in time.

Environmental regulators, land managers, agricultural enterprises, and mining companies all face similar questions: Where is vegetation being lost or gained? How healthy is the existing vegetation? Are compliance requirements being met? Answering these questions efficiently and accurately requires a more comprehensive approach.

Enter satellite-based monitoring—a method that combines broad coverage with frequent revisits to provide unprecedented insight into vegetation dynamics. While there are [free satellite data sources](/blog/free-sources-of-satellite-data) available, commercial high-resolution imagery is essential for any serious vegetation management project. Understanding the limitations of free data helps organizations avoid costly missteps and achieve reliable results.

## The Power of Satellite-Based Vegetation Monitoring

### 1. Near Real-Time Clearing Detections

**Satellite advantage:** Detect clearing activities within days instead of months.

A Queensland mining company saved $350,000 in potential fines when satellite imagery alerted them to contractors clearing beyond approved boundaries. The early detection allowed immediate intervention before regulators became involved.

**Resolution matters:** Free Sentinel-2 imagery (10m resolution) can only detect massive clearings, while commercial high-resolution imagery (50cm) reveals individual trees and small disturbances—critical for legal compliance and early intervention on projects of any meaningful size.

The Queensland Government's Statewide Landcover and Trees Study (SLATS) demonstrates the power of consistent satellite monitoring. After implementing satellite-based enforcement, regulated forest clearing decreased by 64% over four years—proof that when landholders know satellite "eyes in the sky" are watching, compliance improves dramatically.

### 2. Precision Vegetation Health Assessment

**Satellite advantage:** Identify stressed vegetation before visible symptoms appear.

A Western Australian cattle station increased weight gains by 12% after implementing satellite-based pasture rotation. By monitoring vegetation health via NDVI, they optimized grazing patterns and prevented overutilization.

**Resolution matters:** Free satellite data provides only the most basic broad-scale health trends, while commercial imagery with additional spectral bands (including Short-Wave Infrared) delivers the detail needed for real-world applications, detecting subtle moisture differences and early disease indicators across agricultural operations and rehabilitation projects of any significant scale.

For agricultural enterprises, this translates to early problem detection. Stressed vegetation appears in satellite data up to two weeks before visible symptoms develop. This early warning system allows for targeted intervention—whether it's adjusting irrigation, treating disease outbreaks, or rotating grazing—before productivity losses occur.

Our [Red Edge remote sensing](/blog/red-edge-remote-sensing) capabilities take this a step further by providing enhanced sensitivity to chlorophyll content, giving users even earlier detection of plant stress.

### 3. Superior Carbon Project Verification

**Satellite advantage:** Verify carbon sequestration remotely without expensive field campaigns.

A carbon farming project in Queensland secured premium pricing (15% above market rates) by providing investors with detailed satellite evidence of vegetation growth over time.

**Resolution matters:** Small-scale projects or those requiring tree-level verification benefit from commercial high-resolution imagery that can demonstrate exact changes in canopy cover and tree counts.

Carbon project developers must provide credible evidence of vegetation increases compared to historical baselines. With decades of [historical satellite imagery](/blog/historical-satellite-images) available, developers can establish these baselines and then verify ongoing improvements. While some projects might use free satellite data for initial scoping, any serious carbon project requires commercial imagery for reliable verification that will stand up to investor scrutiny and regulatory requirements.

### 4. Invasive Species and Fire Risk Management

**Satellite advantage:** Identify emerging threats across vast areas.

A regional council reduced herbicide use by 40% by targeting specific areas where satellite imagery revealed early-stage weed infestations, rather than treating entire regions uniformly.

**Resolution matters:** Broad invasive patterns are barely visible in free imagery and typically only after significant spread has occurred. Early detection of invasion fronts—when intervention is most cost-effective—absolutely requires the detail that only commercial imagery provides.

Fire management agencies are increasingly using satellite data to assess fuel loads and prioritize controlled burning activities. By combining [vegetation indices](/blog/remote-sensing-vegetation-indices) with seasonal data, they can identify areas where dry biomass presents the highest risk, directing limited resources to where they'll have the greatest impact.

## When to Use Free vs. Commercial Imagery

### Free Satellite Data Has Limited Applications:

- Initial exploratory mapping of very large regions (>10,000 hectares)
- Basic educational and research purposes with severe budget constraints
- Simple background context for more detailed analysis
- Non-critical vegetation trend monitoring where precision isn't essential

Free data from satellites like Sentinel-2 (10m resolution) or Landsat (30m resolution) can provide a basic overview, but serious projects typically require the detail and reliability of commercial imagery.

### High-Resolution Commercial Imagery Adds Value When:

- Legal compliance requires detailed evidence
- Monitoring small properties or features
- Detecting early clearing or disturbance
- Carbon projects need tree-level verification
- Precision agriculture applications
- Early invasive species detection

Commercial satellites like [WorldView-3](/sensors/worldview-3) (30cm resolution), [JL1-50cm](/sensors/jl1-50cm), and others provide extraordinary detail that can make the difference in critical applications. For instance, environmental consultants preparing legal evidence often require commercial imagery that can clearly show individual trees or small areas of impact.

## Real-World ROI from High-Resolution Satellite Monitoring

The investment in commercial satellite imagery typically delivers returns far exceeding costs:

- **Compliance assurance:** A single avoided fine can cover years of monitoring costs
- **Field survey reduction:** One agricultural client replaced $75,000 in annual aerial surveys with $25,000 in satellite monitoring while increasing frequency
- **Carbon premium:** Projects with detailed verification consistently command 10-15% higher prices
- **Precision intervention:** Targeted remediation based on detailed imagery reduced treatment costs by 30-40% versus grid-based approaches

For [mining companies](/mining) in particular, consistent satellite monitoring provides powerful evidence of regulatory compliance. One Queensland operation estimated that their quarterly satellite monitoring program cost less than 5% of what a single environmental violation would have cost in fines, legal fees, and remediation—not counting reputational damage.

## Understanding Satellite Vegetation Analysis Techniques

Satellite vegetation monitoring uses several key techniques to extract meaningful insights:

### 1. Multi-Spectral Analysis

Satellites see beyond the visible spectrum, capturing light in bands invisible to the human eye. Sensors can detect near-infrared, short-wave infrared, and other wavelengths that reveal vegetation health, moisture content, and species composition.

This capability allows analysts to differentiate between healthy and stressed vegetation, identify specific plant communities, and even detect early signs of disease or nutrient deficiencies.

### 2. Time Series Analysis

The real power of satellite monitoring comes from consistent observation over time. By comparing images from different dates, analysts can detect:

- **Gradual trends:** Long-term vegetation decline or improvement
- **Seasonal patterns:** Normal growth cycles vs. abnormal changes
- **Sudden changes:** Clearing, fire impacts, or flood damage

These temporal patterns often reveal issues that might be missed in a single snapshot, making satellite monitoring superior to infrequent field visits or aerial surveys.

### 3. Machine Learning Classification

Advanced machine learning algorithms can classify vegetation types, estimate biomass, and detect changes automatically across vast areas. These techniques are becoming increasingly sophisticated, allowing for automated alerting when unexpected vegetation changes occur.

## The Geopera Advantage

Geopera provides the most comprehensive satellite access in Australia, with unique benefits:

- **Complete coverage:** Access to both free public satellites and the latest commercial sensors
- **Cost-effective high-resolution:** Australia's most competitive pricing on 50cm imagery
- **Specialized capabilities:** Advanced options including [WorldView-3 SWIR](/sensors/worldview-3) for superior vegetation analysis
- **Analysis-ready data:** Receive processed, ready-to-use imagery without technical complexity
- **Flexible solutions:** Scale from free options to premium imagery based on your specific needs

Our clients across [environmental](/environmental), [agricultural](/agriculture), [mining](/mining), [energy](/energy), and [government](/government) sectors benefit from Geopera's unique approach: providing the right satellite data for each specific need, rather than a one-size-fits-all solution.

## Industry Applications

### Mining and Resources

Mining operations must comply with strict vegetation management requirements, including buffer zones, rehabilitation targets, and clearing limits. Satellite monitoring provides auditable evidence of compliance while identifying any issues early.

One Queensland mine used high-resolution monitoring to demonstrate successful rehabilitation progress, securing early release of environmental bonds worth over $5 million. The detailed imagery provided regulators with confidence that revegetation was meeting or exceeding requirements.

### Agriculture and Pastoral

From broad-acre cropping to intensive horticulture, agricultural enterprises use satellite data to optimize production and demonstrate sustainable practices.

A Northern Territory cattle station implemented satellite-based pasture monitoring across 180,000 hectares, allowing precise rotation of cattle based on actual vegetation conditions rather than fixed schedules. The result was a 15% increase in carrying capacity while improving land condition—a win-win for profitability and sustainability.

### Environmental Management

Conservation organizations and environmental consultants use satellite monitoring to track ecosystem health, detect threats, and prioritize management actions.

One project in the Kimberley initially tried free imagery but quickly upgraded to commercial high-resolution data when they realized the limitations. Using proper commercial imagery, they monitored the impact of a controlled burning program across a 63,000 sq km area. The detailed data revealed a 60% reduction in late-season wildfires and corresponding improvements in vegetation diversity—powerful evidence that secured continued program funding.

## Take Action: Elevate Your Vegetation Management

Satellite monitoring has transformed vegetation management from reactive to proactive, saving time and money while improving environmental outcomes. Whether you're managing a small property or millions of hectares, there's a satellite solution that fits your needs and budget.

Ready to see the difference? [Contact Geopera](/contact) today to discuss how satellite imagery can transform your approach to vegetation management. Our team will help you determine whether free public data meets your needs or if commercial imagery would deliver superior results for your specific application.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Old Satellite Images: Where to Find Historical Imagery Since 1972

> Where to find old satellite images, free and commercial. Archives run back to Landsat-1 in 1972, with sources, resolutions and coverage for each.

Published: 2025-03-03 | Author: Darcy Weedman | Reading time: 10 min

Source: https://geopera.com/blog/historical-satellite-images

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

- Free optical archives start in 1972 with Landsat-1, at 30 m resolution. That is the deepest continuous record of the Earth's surface available to anyone.
- Aerial photography goes back further than satellites do. National archives in several countries hold film surveys from the early twentieth century.
- Commercial archives are shallower, starting around 1999, but they resolve at 0.3 to 2 m instead of 30 m.
- Which one you need depends on the question. Landsat answers "how has this landscape changed since the seventies". It cannot tell you what was standing on a single lot in 2005.

<AskBand headline="What did your site look like in 2005?" sub="Archive captures over any coordinates, with dates and prices shown on the results." action="Search the archive" href="https://portal.geopera.com" tiles={['/band/oxbow.webp', '/band/city.webp', '/band/hills.webp']} />

## Where to Find Old Satellite Images

Old satellite images come from three kinds of archive. Government programmes such as Landsat and Sentinel give away moderate-resolution imagery going back to 1972, free and without registration limits. National mapping agencies hold scanned aerial film that predates satellites entirely. Commercial operators sell high-resolution archive captures from roughly 1999 onward, priced by area.

The rest of this post covers what each archive holds and where the boundaries sit.

## A Timeline of Old Satellite Images

The Landsat programme has been returning imagery for more than fifty years, which makes it the longest continuous record of the Earth's surface anyone has. The oldest images date to 1972, when Landsat-1 launched.

Eight more Landsat satellites have followed, operated by USGS and NASA, each revisiting a given location every 16 days. Landsat-7 launched in 1999 and Landsat-8 in 2013. Landsat-9 joined them in 2021. Landsat 1 through 5 have finished their missions, but their data stayed in the public archive, which is why the record is continuous rather than a set of disconnected snapshots.

Other programmes filled in around it. Terra and Aqua carry NASA's MODIS instruments, launched in 1999 and 2002. ESA's Sentinel-2 has been returning 10 m optical imagery since 2015, and Sentinel-1 adds radar that works through cloud and at night. CBERS, a joint Chinese and Brazilian programme, covers South America in more detail than the global missions manage.

> ## Why Are Old Satellite Images Black and White?
>
> Most satellite imagery starts as a set of separate greyscale band measurements, one per wavelength the sensor records. Colour appears when those bands are combined. Natural colour imagery maps the red, green and blue bands to what the eye expects. False colour imagery assigns visible colours to bands such as near infrared, which is why healthy vegetation shows up bright red in a lot of older published imagery.

## Free Sources of Old Satellite Images

Several public archives cost nothing to search and nothing to download.

**USGS EarthExplorer** is the main one. It holds the full <a href="https://earthexplorer.usgs.gov/" target="_blank">Landsat record back to 1972</a>, plus declassified reconnaissance imagery and decades of aerial photography. Registration is free.

**Copernicus Data Space Ecosystem** carries <a href="https://browser.dataspace.copernicus.eu/" target="_blank">Sentinel-1 and Sentinel-2 data</a> from 2014 onward at 10 m. Shallower than Landsat, but sharper and more frequent.

**Google Earth** has a <a href="https://developers.google.com/maps/documentation/earth/historical-imagery" target="_blank">historical imagery layer</a> with a time slider. It is the fastest way to eyeball how a location changed, though you cannot export the underlying pixels for analysis.

**Historic Aerials** covers the United States with <a href="https://www.historicaerials.com/viewer" target="_blank">aerial photography going back decades</a>, viewable by year.

**National mapping agencies** hold the oldest material of all, because aerial survey predates satellites. Geoscience Australia publishes a <a href="https://www.ga.gov.au/scientific-topics/national-location-information/historical-aerial-photography" target="_blank">collection of historical aerial imagery</a>, and Queensland's <a href="https://qimagery.information.qld.gov.au/" target="_blank">QImagery</a> serves scanned state film. Most countries run an equivalent. It is worth checking the national agency before assuming imagery does not exist.

The limit on all of these is resolution. Landsat pixels are 30 m across, so a pixel covers roughly a tennis court. That answers questions about land cover, fire scars, water extent and urban sprawl. It will not tell you how many vehicles were parked at a site, or whether a particular roof had been replaced.

## Commercial Archive Imagery

Commercial archives trade historical depth for detail. They start around 1999 rather than 1972, and they resolve at 0.3 to 2 m per pixel rather than 30 m.

Geopera's [Pera Portal](https://portal.geopera.com) sells from that archive. You draw or upload an area of interest, set a date range, and see what exists over it along with the price before you commit. Pricing is by area rather than by capture, so pulling several dates over the same footprint, which is what change detection actually needs, does not multiply the cost the way per-scene pricing does.

The archive draws on several operators, including the Maxar WorldView constellation, the SuperView series, CGST at 50 cm and 75 cm, and SIIS KOMPSAT. Every order runs through the same processing chain before delivery, so captures from different sensors and different years arrive in a state where you can compare them directly. For what the tiers actually cost, see the [satellite imagery cost guide](/blog/satellite-imagery-cost-guide).

## What People Use Old Satellite Images For

**Environmental monitoring and disaster assessment.** Fire scar mapping, flood extent reconstruction, coastal erosion measured across decades, and drought cycles. Insurers and public agencies use before-and-after pairs to establish what conditions were like prior to an event.

**Urban development and planning.** Tracking how a city expanded, when infrastructure went in, and where natural land was converted. Planning disputes often turn on what was on a site at a particular date, and an archive capture with a timestamp settles the question.

**Mining and resource management.** Documenting site conditions over time, establishing pre-disturbance baselines for rehabilitation, and keeping a dated record for regulators. Rehabilitation reporting in particular needs a credible picture of what the ground looked like before work started, which is difficult to reconstruct once the record is gone.

**Due diligence and baselines.** Property and land assessment before acquisition, environmental impact studies that need a documented starting condition, and carbon stock assessment, where the accuracy of the baseline drives the accuracy of everything calculated from it.

## Choosing Between Free and Commercial

Start free. If Landsat or Sentinel answers the question, there is no reason to pay, and the historical depth is better than anything commercial.

Move to commercial archive when the question needs detail that 30 m cannot resolve, when you need a specific date rather than whatever the free satellite happened to catch, or when the imagery has to support a claim in front of a regulator, an insurer or a court, and the provenance has to hold up.

Most projects end up using both: free imagery to establish the long arc, commercial captures at the few dates that matter.

You can search the commercial archive at [portal.geopera.com](https://portal.geopera.com), or email sales@geopera.com with the area and dates you are after.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Red Edge in Remote Sensing: Applications and Advantages

> What the red edge band is (680-750 nm), why it detects plant stress 1-2 weeks before visible symptoms, and where red edge indices beat NDVI.

Published: 2025-03-03 | Author: Darcy Weedman | Reading time: 9 min

Source: https://geopera.com/blog/red-edge-remote-sensing

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary: Red Edge Remote Sensing

Red edge refers to the 680-750nm wavelength region between red and near-infrared light where vegetation reflectance changes dramatically. This spectral region is valuable because:

- It's highly sensitive to chlorophyll content, making it excellent for detecting plant stress and health
- It can detect vegetation problems 1-2 weeks earlier than traditional methods (before visible symptoms appear)
- It overcomes limitations of NDVI, which saturates in dense vegetation
- It's especially useful for precision agriculture (nutrient management, yield prediction) and forestry (early stress detection)
- It improves vegetation species classification accuracy across diverse landscapes

Modern satellites like [WorldView-3](/sensors/worldview-3) & Legion include dedicated red edge bands, making this technology widely accessible for agriculture, forestry, and environmental monitoring applications.

<AskBand headline="Which sensors carry red edge?" sub="Band specs are in the catalogue; archive search over your area is free." action="Browse the catalogue" href="/sensors" tiles={['/band/fields.webp', '/band/bloom.webp', '/band/pivots.webp']} />

## Understanding Red Edge Remote Sensing: Applications and Advantages

Remote sensing techniques capture the reflectance of surfaces across different portions of the electromagnetic spectrum. For vegetated areas, this reflectance varies based on plant species, health, and moisture content. By analyzing how this reflectance relates to vegetation properties, we can extract information about plant characteristics from remotely sensed data. **Red edge remote sensing**, in particular, gives earlier and more sensitive readings of vegetation health and structure than the standard bands.

## What Is the Red Edge Band?

The **red edge band** is the specific portion of the electromagnetic spectrum that lies between the red and near-infrared bands without overlapping either. This spectral region typically spans wavelengths from 680 to 750 nanometers, though the exact range can vary depending on vegetation species and conditions.

What makes the **red edge** so distinctive is the sharp increase in reflectance of green vegetation that occurs at this transition from red to near-infrared wavelengths. This dramatic change creates a "edge" in spectral reflectance curves that gives this region its name.

![Approximate reflectance spectrum of green plants showing the red edge band](/red-edge-spectrum.webp)
_Approximate reflectance spectrum of green plants showing the distinct red edge transition. (Source: Ai, S.; Zheng, H.; Yu, J. Materials 2020, 13, 1540)_

## What Causes the Shift in Vegetation Reflectance in the Red Edge?

The **red edge phenomenon** occurs due to two key physiological factors:

1. **Chlorophyll absorption**: The leaves of healthy vegetation contain chlorophyll, which strongly absorbs red light for photosynthesis
2. **Leaf structure reflectance**: The internal cellular structure of leaves scatters near-infrared light, causing strong reflection

This transition from strong absorption of red light to high reflectance of near-infrared light results in the steep change in reflectance that defines the **red edge bands** in vegetation remote sensing.

## Factors Affecting Red Edge Position

Several biophysical factors influence the shape and position of the **red edge**:

- Chlorophyll concentration
- Internal leaf structure
- Canopy architecture
- Leaf area index (LAI)
- Biomass volume
- Water content

These factors cause the **red edge position** to shift in measurable ways. Variations closer to the red wavelengths primarily relate to chlorophyll content, while variations nearer to the near-infrared wavelengths connect to leaf structural properties.

### Chlorophyll and Water Content Effects

Increases in chlorophyll and/or water content in leaves cause the **red edge** to shift toward longer wavelengths. Conversely, decreases in chlorophyll and water content—often resulting from stress or senescence—lead to shifts toward shorter wavelengths. This relationship makes **red edge remote sensing** particularly valuable for detecting plant stress before visual symptoms appear.

## Red Edge vs. Traditional Red and Near-Infrared Bands

What makes the **red edge band** different from standard red and near-infrared bands used in conventional vegetation monitoring?

While strong chlorophyll absorption causes low red reflectance and internal leaf scattering creates strong near-infrared reflectance, the **red edge**—positioned between these regions—captures nuanced variations in both chlorophyll content and leaf structure simultaneously. This provides more comprehensive information about vegetation status.

The **red edge reflectance** demonstrates significantly higher sensitivity to chlorophyll a and b concentrations and subtle changes in overall chlorophyll content. This enhanced sensitivity enables earlier detection of plant stress conditions before they become visible to the human eye.

## Red Edge-Based Vegetation Indices

Vegetation indices combine two or more spectral bands into a single value to enhance sensitivity to specific vegetation traits. Indices that incorporate **red edge bands** have proven to significantly improve the detection of certain vegetation attributes from remote sensing data.

These **red edge indices** address important limitations of traditional vegetation indices like NDVI that use only red and near-infrared bands. In dense vegetation conditions, NDVI reaches a saturation threshold where it becomes insensitive to further changes in chlorophyll content. Consequently, NDVI becomes less effective at estimating important characteristics like biomass, leaf area index, and nutrient status in dense canopies.

**Red edge-based vegetation indices** maintain sensitivity across a wider range of vegetation densities, making them superior for many advanced monitoring applications. The most widely used is NDRE, covered with its formula and value ranges in our [guide to vegetation indices](/blog/remote-sensing-vegetation-indices).

![Examples of red edge spectral indices derived from Sentinel-2 imagery](/red-edge-indices.webp)
_Examples of red edge spectral indices derived from Sentinel-2 imagery. (Source: Sun, L.; Chen, J.; Guo, S.; et al. Remote Sens. 2020, 12, 158)_

## Applications of Red Edge Remote Sensing

The **red edge** spectral region provides valuable insights for numerous applications across agriculture, forestry, water quality assessment, and land use/land cover mapping. Let's explore some key applications:

### Monitoring Forest Health and Disturbance

Early detection of forest stress and disturbance is crucial for timely intervention and protection. Stressors—including water deficiency, waterlogging, nutrient depletion, and disease—disrupt vegetation growth, functioning, and productivity. Meanwhile, disturbances like windstorms, wildfires, and disease outbreaks can dramatically alter affected areas.

Since vegetation stress cannot be directly determined visually, an indicator is needed. Chlorophyll content serves as an effective proxy for plant health and stress levels. Stressed plants typically contain lower chlorophyll concentrations than unstressed plants, resulting in higher reflectance in the red region and lower reflectance in the near-infrared region.

**Red edge remote sensing** excels at detecting these subtle chlorophyll changes. Studies have shown that satellite-based vegetation indices incorporating **red edge bands** can detect stressed forest stands 13 to 16 days earlier than indices using only red, green, blue, and near-infrared bands. This earlier detection window is possible because the **red edge region** responds more sensitively to changes in chlorophyll a and b than traditional red and green spectral regions.

For forest disturbance classification, **red edge**-based vegetation indices achieve better results at classifying the magnitude of forest disturbance in complex environments like spruce-dominated mountainous areas compared to traditional indices like NDVI.

![Mean spectral reflectance of poplars under different stressors](/poplar-stress-reflectance.webp)
_Mean spectral reflectance of poplars grouped by stress type, extracted from UAV-hyperspectral imagery. (Source: Zhou, Q.; Kuang, J.; Yu, L.; et al. Remote Sens. 2024, 16, 3751)_

### Precision Agriculture Applications

Precision agriculture aims to optimize yields while efficiently using inputs and reducing environmental impacts. Monitoring crop conditions is essential for early issue detection and implementing site-specific management solutions. **Red edge remote sensing** contributes significantly to these efforts in several ways:

1. **Crop health detection**: The sensitivity of **red edge bands** to chlorophyll content makes them excellent indicators of overall crop health.

2. **Nutrient status assessment**: Since chlorophyll concentration correlates with nitrogen supply, **red edge bands** and derived indices help understand spatial variability in crop nutrient status, informing variable-rate fertilizer application strategies.

3. **Growth monitoring and yield prediction**: **Red edge indices** are less affected by canopy density than traditional indices, providing more accurate estimates of key growth parameters like leaf area index (LAI) and chlorophyll content.

Studies have identified the **red edge region** among the most crucial variables for estimating chlorophyll content in crops like maize across various growth stages. This information helps farmers assess crop responses to management practices and environmental conditions, enabling field-specific optimization.

![Spatial distribution of chlorophyll content in a maize field](/maize-chlorophyll-map.webp)
_Spatial distribution of chlorophyll content across a maize field during vegetative growth stages. (Source: Brewer, K.; Clulow, A.; Sibanda, M.; et al. Remote Sens. 2022, 14, 518)_

When using **red edge bands** and derived indices to inform management decisions, it's important to consider the multiple factors affecting plant growth and health, including soil conditions, topography, hydrology, and other environmental variables. Knowledge of the primary growth-limiting factors in specific fields helps link observed reflectance variability to appropriate management inputs.

### Classifying Vegetation Species

Vegetation classification is essential for environmental monitoring, decision-making, and resource planning. The **red edge wavelengths** offer unique advantages for distinguishing between different vegetation types due to their sensitivity to species-specific attributes like leaf and canopy structure, chlorophyll content, and leaf area index.

The subtle differences in how various plant species reflect electromagnetic radiation in the **red edge region** can be leveraged for accurate classification. The shift in **red edge position** often serves as a distinctive signature for differentiating vegetation species.

**Red edge remote sensing** has proven valuable for classification across diverse contexts and landscapes:

- **Crop type mapping**: Sentinel-2 **red edge bands** significantly improve distinction between different crop classes
- **Wetland classification**: **Red edge** and near-infrared bands increase separability between wetland vegetation types
- **Forest-dominated landscapes**: In boreal landscapes, a quarter of the most important classification variables belong to the **red edge band**
- **Grass species discrimination**: Sentinel-2 **red edge band 5** has been identified as the most influential for distinguishing between different grass species

Vegetation classification accuracy can be further enhanced by considering phenology—the seasonal changes in plant characteristics that affect reflectance patterns. Understanding species-specific relationships between **red edge reflectance** and biophysical attributes like chlorophyll content and canopy structure can inform optimal timing and methods for species separation.

## The Red Edge Advantage

The **red edge region** offers several distinct advantages over traditional spectral bands for vegetation monitoring:

- **Higher sensitivity** to vegetation characteristics like chlorophyll content and leaf structure
- **More nuanced insights** into vegetation status, including health, type, structure, and nutrient conditions
- **Earlier stress detection** before visible symptoms appear, enabling timely interventions in agriculture and forestry
- **Enhanced discrimination** between vegetation types, improving mapping accuracy in heterogeneous landscapes
- **Broader application range** across different vegetation densities where traditional indices saturate

**Red edge remote sensing** can be implemented across multiple platforms and scales. Satellite-based measurements enable large-scale, long-term monitoring of vegetation dynamics at regional scales. Hyperspectral imagery provides detailed insights at local scales, while ground-based measurements offer high-resolution data to complement and validate satellite and airborne observations.

## Conclusion

**Red edge remote sensing** represents a significant advancement in our ability to monitor and understand vegetation. By capturing the unique spectral transition between red and near-infrared wavelengths, the **red edge band** provides more sensitive and nuanced information about plant health, structure, and composition than traditional spectral regions alone.

From precision agriculture to forest health monitoring and species classification, **red edge**-based approaches consistently demonstrate superior performance for critical vegetation assessment tasks. As remote sensing technologies continue to evolve, the **red edge advantage** will likely play an increasingly important role in environmental monitoring, resource management, and sustainable development practices worldwide.

Whether you're a researcher, environmental manager, agricultural consultant, or forest steward, incorporating **red edge bands** into your remote sensing toolkit can provide deeper insights and earlier detection capabilities for more effective vegetation management.

## Frequently Asked Questions

### What is the red edge band?

The red edge band is the region of the electromagnetic spectrum between red and near-infrared light, spanning roughly 680 to 750 nanometres, where the reflectance of green vegetation rises sharply. That steep rise, driven by chlorophyll absorption on one side and leaf-structure scattering on the other, is the "edge" the name refers to.

### Why is the red edge useful for detecting plant stress?

Because its position and shape respond directly to chlorophyll content. As stress reduces chlorophyll, the red edge shifts toward shorter wavelengths before any change is visible to the eye. Studies on forest stands measured detection 13 to 16 days earlier with red edge indices than with indices built only from red, green, blue and near-infrared bands.

### Which satellites carry red edge bands?

Sentinel-2 carries three red edge bands (bands 5, 6 and 7) at 20 m resolution for free, and commercial sensors including [WorldView-3](/sensors/worldview-3) and WorldView Legion carry dedicated red edge bands at higher resolution. Availability of the band, not the index, is the constraint: NDRE cannot be computed from RGB+NIR imagery.

### What is the difference between NDVI and red edge indices?

NDVI uses only red and near-infrared bands, and saturates over dense canopy. Red edge indices like NDRE stay sensitive in dense vegetation and respond earlier to chlorophyll change, at the cost of requiring a sensor that carries the band. The [vegetation indices guide](/blog/remote-sensing-vegetation-indices) compares the formulas side by side.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# The 5 Best Free Sources of Elevation Data in Australia

> A comprehensive guide to finding and using free Digital Elevation Model (DEM) data sources in Australia for environmental studies, engineering projects, and research.

Published: 2024-12-07 | Author: Darcy Weedman | Reading time: 7 min

Source: https://geopera.com/blog/free-sources-of-dem-data

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

Looking to analyze terrain for environmental studies, engineering projects, or research? This guide will walk you through the best free Digital Elevation Model (DEM) sources available to Australians.

## Summary

- [ELVIS - Elevation Information System](https://elevation.fsdf.org.au/): Best for comprehensive national elevation data access and direct downloads

- [Digital Earth Australia](https://www.ga.gov.au/scientific-topics/dea): Best for time-series analysis and environmental monitoring

- State-Specific Resources:
  - [QSpatial](https://qldspatial.information.qld.gov.au/) (Queensland)
  - [NSW Spatial Services](https://spatial.nsw.gov.au/)
  - [VicMap](https://www.land.vic.gov.au/) (Victoria)
    Best for high-resolution local and regional elevation data

- [SRTM Data](https://earthexplorer.usgs.gov/): Best for consistent 30m resolution coverage across Australia

- [Copernicus DEM](https://spacedata.copernicus.eu/): Best for recent, high-accuracy elevation data with global coverage

<AskBand headline="How high is that, exactly?" sub="When free elevation runs out of resolution, tri-stereo capture is what we quote." action="See elevation products" href="/imagery" tiles={['/band/caldera.webp', '/band/terraces.webp', '/band/hills.webp']} />

## Understanding Digital Elevation Data

A Digital Elevation Model represents the Earth's terrain as a three-dimensional digital grid known as a raster GIS layer. Each cell in the grid stores precise elevation information. These heights are measured against what's known as a vertical datum—a standardized reference point for zero elevation, typically based on mean sea level. These models are fundamental tools for understanding terrain and topography.

Many techniques are used to create digital elevation models, each with its own strengths and applications. Drone photogrammetry excels at capturing detailed data for small project areas. Aerial LiDAR provides precise elevation measurements and can penetrate vegetation, making it ideal for both Digital Surface Models (DSMs) that include buildings and vegetation, and Digital Terrain Models (DTMs) that represent bare earth. Satellite-derived elevation data offers consistent coverage over vast areas, perfect for regional or continental-scale analysis.

## Types of Digital Elevation Models

### Digital Surface Models (DSM)

DSMs capture the Earth's surface including all objects on it - buildings, vegetation, and other human-made structures. They're particularly valuable for urban planning, aviation studies, and vegetation management. These models represent the first surface that would be hit by a raindrop, whether that's a tree canopy, building roof, or the ground.

![Comparison between Digital Terrain Models (DTM) and Digital Surface Models (DSM)](/dtm-vs-dsm.webp)
_Figure 1: Comparison between DTM and DSM. DTM represents the bare earth surface without buildings and vegetation, while DSM includes all objects on the terrain._

### Digital Terrain Models (DTM)

DTMs represent the bare earth surface with all vegetation and built structures removed. This makes them ideal for flood modeling, watershed analysis, and infrastructure planning. They show the ground level that would be exposed if all buildings and vegetation were removed.

### Digital Elevation Models (DEM)

DEM is often used as an umbrella term encompassing both DSMs and DTMs. At larger scales where the distinction between surface features and bare earth becomes less significant (such as with 30m resolution data), the term DEM is commonly used. The choice between these models depends on your project's requirements - from high-precision local surveys to broad regional studies.

## Common Applications

### Terrain Analysis and Resource Assessment

DEMs help analysts identify geological features like faults, folds, and lineaments in the terrain. By analyzing surface expressions and drainage patterns, field teams can better understand an area's structural geology and plan their investigation programs more efficiently.

### Site Planning and Development

When developing new projects in complex terrain, DEMs are crucial for optimizing site layouts, calculating earthwork volumes, and planning access roads. The elevation data helps engineers determine the most efficient construction approaches while ensuring proper drainage and minimal environmental impact.

### Slope Stability Monitoring

Large infrastructure projects use repeated DEM surveys to track subtle changes in ground movement and detect potential stability issues before they occur. This ongoing monitoring is essential for maintaining site safety and preventing costly disruptions that could result from ground movement or landslides.

## What's Available in Australia

If you're searching for free digital elevation maps in Australia, it's important to understand what's available.

Australia benefits from excellent access to open-source elevation data, with multiple options available to suit different project needs. Nation-wide coverage is readily available through 30-meter resolution DEMs, providing a solid foundation for regional analysis.

Higher resolution models, typically ranging from 1 to 5 meters, are concentrated around urban centers and coastal regions where detailed terrain information is crucial. Several state governments have taken the initiative to release comprehensive statewide elevation models, further expanding access to quality terrain data for specific regions.

## Detailed Source Overview

### 1. ELVIS - Elevation Information System

ELVIS is Australia's national elevation data portal, managed by Geoscience Australia. It's the perfect starting point for any terrain analysis project.

Key Features:

- User-friendly web interface for data discovery
- Multiple resolution options (1m to 30m)
- Direct download capabilities
- Extensive metadata and quality information
- Regular updates for priority areas

### 2. Digital Earth Australia

Digital Earth Australia is a government platform that transforms raw satellite and elevation data into ready-to-use information. It provides powerful tools for understanding Australia's changing landscape.

Key Features:

- Time-series elevation data
- Advanced analysis tools
- Integration with satellite imagery
- Focus on environmental monitoring
- Regular data updates

### 3. State-Specific Resources

#### QSpatial (Queensland)

Queensland's open data portal offers comprehensive elevation datasets through QSpatial, providing high-quality DEMs particularly for coastal and urban regions.

Key Features:

- 5m resolution LiDAR-derived DEMs
- Excellent coastal coverage
- Regular updates for development areas
- Integration with other spatial datasets

#### NSW Spatial Services

The NSW Spatial Services portal delivers authoritative elevation data for New South Wales, with particular emphasis on urban and development areas.

Key Features:

- High-resolution elevation data
- Comprehensive urban coverage
- Advanced terrain modeling
- Detailed metadata

#### VicMap (Victoria)

VicMap provides Victoria's foundational elevation dataset, offering consistent coverage across the entire state through their spatial data infrastructure.

Key Features:

- State-wide 10m resolution DEM
- Derived from various data sources including LiDAR
- Regular updates and quality improvements
- Seamless coverage across Victoria

### 4. Global Digital Elevation Models

#### SRTM (Shuttle Radar Topography Mission)

The SRTM dataset, collected by NASA in 2000, remains one of the most widely used global elevation datasets due to its consistent quality and complete coverage.

Key Features:

- 30m resolution globally
- Well-validated accuracy
- Complete coverage of Australia
- Suitable for regional analysis

#### AW3D30

ALOS World 3D 30m (AW3D30) is JAXA's global elevation dataset, collected between 2006-2011, offering improved accuracy over SRTM particularly in areas of steep terrain.

Key Features:

- 30m resolution global DEM
- Updated more recently than SRTM
- Better handling of void areas
- Improved vertical accuracy

#### Copernicus DEM

The Copernicus DEM is a recent addition to global elevation datasets, providing high-quality terrain data through the European Union's Earth observation program.

Key Features:

- Global coverage at 30m resolution (GLO-30)
- Higher resolution (10m) available over Europe only
- Based on TerraSAR-X and TanDEM-X data (2011-2015)
- High absolute vertical accuracy (2-4m)
- Freely available through the Copernicus Data Space Ecosystem

## Conclusion

Australia offers excellent free Digital Elevation Model resources suitable for various applications. Whether you're conducting environmental research, planning infrastructure, or analyzing watersheds, these platforms provide the elevation data you need. Start with the platform that best matches your technical requirements and project scope.

Need satellite imagery to complement your elevation data? Check out our guide to the [10 free sources of satellite data](https://geopera.com/blog/free-sources-of-satellite-data) worth knowing in 2026.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Google Earth vs Commercial Satellite Imagery: Understanding Your Options

> Comprehensive comparison of Google Earth and commercial satellite imagery platforms, exploring key differences in update frequency, accuracy, and analytical capabilities

Published: 2024-11-22 | Author: Darcy Weedman | Reading time: 5 min

Source: https://geopera.com/blog/google-earth-vs-commercial-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

When it comes to satellite imagery, professionals need to understand the key differences between free platforms like Google Earth and commercial solutions like our [Pera Portal](https://portal.geopera.com/). While Google Earth offers excellent visualization for general use, commercial platforms provide superior accuracy, more frequent updates, and advanced analytical capabilities essential for professional applications. This guide explores these differences to help you make an informed choice for your projects.

For many professionals and enthusiasts, Google Earth serves as their first introduction to satellite imagery. While this platform has revolutionized how we view our planet and made satellite imagery accessible to billions, understanding its limitations is crucial for making informed decisions about geospatial data needs.

<AskBand headline="Ready for real bands and real dates?" sub="Free Sentinel-2 is the gentlest step up: multispectral, dated, still free." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/city.webp', '/band/fields.webp', '/band/atoll.webp']} />

## Understanding Update Frequencies

One of Google Earth's primary limitations becomes apparent when examining its update cycle. While many users hope to access real-time or live satellite views, the reality is more complex. Google Earth typically refreshes imagery every 6-12 months, with urban areas seeing more frequent updates while rural regions might go years without new imagery.

The update patterns reveal significant differences between platforms:

- Google Earth: Updates every 6-12 months with no guaranteed schedule
- Commercial Platforms: Access to imagery within 30 days and ability to request new captures

This difference in update frequency becomes critical for time-sensitive projects where current ground conditions matter. Construction monitoring, environmental assessments, and urban development tracking all require recent imagery that free platforms simply cannot provide.

## Image Quality and Resolution

Resolution and image quality play crucial roles in satellite imagery applications. Google Earth provides variable resolution depending on location, typically ranging from 15m to 15cm, with a focus on visual appeal over technical accuracy. The platform compresses imagery for streaming, which can impact analysis capabilities.

Commercial platforms take a different approach by offering consistent resolution across coverage areas, with options down to 30cm resolution. More importantly, they provide access to raw data, enabling custom processing and advanced analysis that's impossible with compressed imagery.

## Geometric Accuracy for Professional Use

For professional applications, geometric accuracy becomes a critical factor. While Google Earth prioritizes visual appeal, commercial platforms focus on technical precision through orthorectified imagery aligned to ground control points. This precision makes commercial imagery essential for:

- Infrastructure planning and construction
- Legal documentation and compliance
- Scientific research and analysis
- Environmental impact assessments
- Urban development monitoring

The improved accuracy translates directly to reduced field verification costs and more reliable measurements for professional projects.

## Advanced Analytics

The most significant difference lies in analytical capabilities. As covered in our [guide to vegetation indices](../blog/remote-sensing-vegetation-indices), commercial platforms enable sophisticated analysis techniques for vegetation monitoring, spectral analysis, and change detection.

### Vegetation Monitoring

Commercial platforms excel at vegetation analysis through various indices like NDVI, EVI, and SAVI. These tools help professionals monitor crop health, forest coverage, and environmental changes with precision that's impossible with basic visualization platforms. The ability to process raw spectral data allows for customized analysis tailored to specific vegetation types and conditions.

### Change Detection

Professional platforms enable precise temporal analysis for tracking urban development, environmental impacts, and disaster response. Users can quantify surface changes and generate detailed reports for stakeholders. This capability proves invaluable for projects requiring historical comparison or ongoing monitoring.

## Data Integration and Access

Google Earth provides a web-based interface with limited export options and basic measurement tools. Commercial platforms offer significantly more flexibility through direct data downloads, GIS software integration, and API access for automation. This integration capability allows organizations to incorporate satellite imagery into their existing workflows and systems.

The ability to batch process imagery and automate analysis saves significant time for large-scale projects. Professional support services ensure that technical questions get answered quickly, minimizing project delays.

## Cost Considerations

While Google Earth's free access is attractive, professional projects require reliable, accurate, and current data. The investment in commercial imagery often provides significant returns through reduced field verification needs, more accurate planning, and better decision-making capabilities.

Organizations typically find that commercial imagery pays for itself through:

- Reduced time spent on field visits
- More accurate project planning
- Fewer measurement errors
- Better risk assessment
- Improved client communication

## Real-World Applications

Commercial satellite imagery proves invaluable across multiple sectors. Urban planners use it for development monitoring and infrastructure planning. Agricultural professionals rely on it for crop health assessment and irrigation management. Environmental scientists track changes in forest cover and coastal erosion. Emergency responders use it for disaster assessment and response planning.

Each of these applications requires the precision, timeliness, and analytical capabilities that only commercial platforms provide. The ability to process raw data and perform custom analysis allows organizations to extract exactly the information they need for their specific use case.

## Choosing the Right Platform

Your choice between Google Earth and commercial imagery should depend on your specific needs. Google Earth works well for general reference imagery, basic visualization, and cases where exact timing isn't crucial. However, professional work requiring accurate measurements, current data, or advanced analysis demands commercial imagery.

Learn more about utilizing different data sources in our [guide to free satellite data sources](../blog/free-sources-of-satellite-data).

## Experience the Professional Advantage

Ready to move beyond Google Earth's limitations? Visit our [Pera Portal](https://portal.geopera.com) to explore our extensive collection of high-resolution satellite imagery. Our platform offers superior resolution, current imagery, advanced analytics capabilities, and professional support to transform your projects.

Need help choosing the right imagery for your specific requirements? [Get in touch](../quote) with our team of geospatial experts who can guide you through the selection process.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Understanding Remote Sensing Vegetation Indices

> A comprehensive guide to NDVI, EVI, SAVI and other vegetation indices used in satellite imagery analysis for monitoring plant health and agricultural productivity

Published: 2024-11-01 | Author: Darcy Weedman | Reading time: 4 min

Source: https://geopera.com/blog/remote-sensing-vegetation-indices

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

A vegetation index is a single number computed from two or more spectral bands that summarises plant condition in each pixel. The five most used are NDVI, EVI, SAVI, NDWI and NDRE, and all exploit the same physics: healthy vegetation reflects near-infrared light strongly and absorbs red. This guide gives the formula, value ranges and best use for each.

<AskBand headline="What is the NDVI of your paddock today?" sub="Many of these indices run in the browser on free Sentinel-2." action="Compute an index" href="https://portal.geopera.com" tiles={['/band/fields.webp', '/band/pivots.webp', '/band/bloom.webp']} />

## Getting Started with Vegetation Indices

Now that you've learned how to access satellite data from our [guide to free satellite data sources](../blog/free-sources-of-satellite-data), you might be wondering: what can you actually do with this data?
One of the most powerful applications is vegetation analysis using spectral indices.
Most high-resolution satellite imagery comes with four fundamental spectral bands: Red, Green, Blue (RGB), and Near-Infrared (NIR). While the RGB bands show us what we naturally see, the NIR band reveals crucial information about vegetation health that's invisible to the human eye. By combining these bands mathematically, we create vegetation indices - powerful tools that help us understand plant health, biomass, and growth patterns.
Here are the five most commonly used vegetation indices in remote sensing, each offering unique insights into vegetation characteristics.

## 1. Normalized Difference Vegetation Index (NDVI)

Formula: `NDVI = (NIR - Red) / (NIR + Red)`

NDVI is mainly used for crop health monitoring, biomass estimation, drought assessment, and long-term vegetation studies. It provides a value ranging from `-1 to +1`, where healthy vegetation typically ranges from `0.2 to 0.8`, and bare soil sits around `0.1`. The higher the NDVI value, the healthier and denser the vegetation. This makes NDVI particularly effective for tracking vegetation changes over time and comparing plant health across different areas.

Key Values:

- Dense, healthy vegetation: `0.6 to 0.9`
- Moderate vegetation: `0.2 to 0.5`
- Sparse vegetation: `0.1 to 0.2`
- Bare soil: `0 to 0.1`
- Water bodies: `-0.25 to 0`

Key Applications:

- Agricultural monitoring
- Crop yield prediction
- Environmental management
- Carbon sequestration estimation

## 2. Enhanced Vegetation Index (EVI)

Formula: `EVI = G × ((NIR - Red) / (NIR + C1 × Red - C2 × Blue + L))`

The coefficients adopted in the MODIS-EVI algorithm are: `L=1`, `C1 = 6`, `C2 = 7.5`, and `G = 2.5`.
EVI produces values that better represent vegetation health in areas where NDVI might overly saturate. The values typically range from `-1 to +1`, with healthy vegetation showing values between `0.2 to 0.8`.
Unlike NDVI, EVI remains sensitive to changes in dense canopy areas, making it particularly valuable for monitoring rainforests and other areas of high biomass. The addition of the blue band also helps correct for atmospheric interference and soil background noise.

Key Values:

- Dense, healthy vegetation: `0.4 to 0.9`
- Moderate vegetation: `0.2 to 0.4`
- Sparse vegetation: `0.1 to 0.2`
- Bare soil: `< 0.1`
- Water: `< 0`

Key Applications:

- Rainforest monitoring
- Areas with dense vegetation
- Canopy structure studies

## 3. Soil Adjusted Vegetation Index (SAVI)

Formula: `SAVI = ((NIR - Red) / (NIR + Red + L)) × (1 + L)`

where L is the soil brightness correction factor (typically `0.5`)
SAVI generates values similar to NDVI but with better accuracy in areas where soil is visible through the vegetation. It produces values from `-1 to +1`, with healthy vegetation typically ranging from `0.2 to 0.8`. The key difference is its ability to minimize soil brightness influences, making it especially useful in arid regions or areas with sparse vegetation where soil background effects can significantly impact other vegetation indices.

Key Values:

- Dense vegetation: `> 0.7`
- Moderate vegetation: `0.4 to 0.7`
- Sparse vegetation: `0.2 to 0.4`
- Exposed soil: `< 0.2`

Key Applications:

- Arid region monitoring
- Early crop growth stage monitoring

## 4. Normalized Difference Water Index (NDWI)

Formula: `NDWI = (Green - NIR) / (Green + NIR)`

NDWI produces values that indicate vegetation water content and water stress. Values range from `-1 to +1`, where positive values generally indicate healthy, well-watered vegetation, and negative values suggest water stress. Water bodies typically show high positive values (`>0.3`), while dry vegetation and soil show negative values. This makes NDWI particularly effective for monitoring drought conditions and irrigation needs.

Key Values:

- Water bodies: `> 0.3`
- Wet vegetation: `0.1 to 0.3`
- Dry vegetation: `-0.1 to 0.1`
- Soil: `< -0.1`

Key Applications:

- Water stress monitoring
- Irrigation planning
- Fire risk assessment
- Wetland mapping

## 5. Normalized Difference Red Edge (NDRE)

Formula: `NDRE = (NIR - RedEdge) / (NIR + RedEdge)`

NDRE produces values that indicate chlorophyll content and nitrogen status in vegetation. Values typically range from `-1 to +1`, with healthy vegetation showing values between `0.2 to 0.5`. This index is particularly sensitive to subtle changes in plant health and can detect stress before it becomes visible to the naked eye or shows up in NDVI analysis, which makes it valuable for precision agriculture where early detection matters. It depends on a sensor carrying a dedicated band between red and near-infrared; our [red edge explainer](/blog/red-edge-remote-sensing) covers why that narrow band is so sensitive.

Key Values:

- Healthy vegetation: `0.2 to 0.5`
- Stressed vegetation: `0.1 to 0.2`
- Very stressed/senescent: `< 0.1`
- Non-vegetation: `< 0`

Key Applications:

- Precision agriculture
- Crop yield optimization
- Crop management
  - Early stress detection
  - Disease monitoring
  - Growth stage assessment

## Choosing the Right Index

The choice of vegetation index depends on several factors:

- Your specific application and goals
- Environmental conditions of your study area
- Vegetation density and types
- Available spectral bands in your imagery
- Required accuracy and sensitivity
  Often, using multiple indices in combination provides the most comprehensive understanding of vegetation conditions.

## Frequently Asked Questions

### What is a good NDVI value?

Dense, healthy vegetation reads 0.6 to 0.9, moderate vegetation 0.2 to 0.5, sparse vegetation 0.1 to 0.2, and bare soil around 0 to 0.1. Water is negative. "Good" depends on the crop and season: a mid-season cereal paddock at 0.3 signals a problem, while the same value in early growth is normal.

### What is the difference between NDVI and EVI?

Both measure vegetation vigour from red and near-infrared reflectance, but EVI adds the blue band and correction coefficients. That keeps it sensitive in dense canopy where NDVI saturates, and reduces atmospheric and soil background effects. Use NDVI for general monitoring and comparability; use EVI over rainforest and other high-biomass areas.

### Which vegetation index detects crop stress earliest?

NDRE, because it uses the red edge band, where reflectance responds to chlorophyll and nitrogen changes before the red and near-infrared bands move. NDRE can flag stress days before it appears in NDVI or to the eye.

### Which satellites provide the bands these indices need?

Every index here except NDRE needs only red, green, blue and near-infrared, which free Sentinel-2 provides at 10 m. NDRE additionally needs a red edge band, which Sentinel-2 carries at 20 m and several commercial sensors carry at higher resolution.

## Need Help With Your Vegetation Monitoring Project?

Whether you're monitoring crop health, tracking deforestation, or quantifying carbon sequestration, choosing the right satellite imagery and vegetation indices is crucial for your project's success.
While free satellite data sources are great for many applications, some projects require higher resolution or more frequent monitoring.
We can help you find the right satellite data for your specific needs. Explore available imagery through our [Pera Portal](https://portal.geopera.com) or [get in touch](../quote) to discuss your project specific requirements.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# 10 Free Sources of Satellite Data in 2026

> Ten free sources of satellite data in 2026: Sentinel-2, Landsat, NASA viewers and disaster imagery, plus country guides for the USA, UK, Canada, Germany and Brazil.

Published: 2024-10-24 | Author: Darcy Weedman | Reading time: 7 minute

Source: https://geopera.com/blog/free-sources-of-satellite-data

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

Free satellite data comes from a handful of open programs: Europe's Copernicus (Sentinel-2 at 10 m, roughly 5-day revisit), the USGS Landsat archive (30 m, back to 1972), NASA's open data services, and browser viewers built on top of them. The ten sources below cover viewing, downloading and analysis, with no payment required for any of them.

- [Pera Portal free tier](https://portal.geopera.com/): Unlimited free Sentinel-2 browsing with 45+ pre-configured spectral indices (NDVI, NDWI, EVI, etc.) viewable in-browser, no setup and no credit card
- [Copernicus Browser](https://browser.dataspace.copernicus.eu/): The source of truth for Sentinel data, for anyone who wants to search and download the rasters
- [USGS EarthExplorer](https://earthexplorer.usgs.gov/): The Landsat archive back to 1972, the longest continuous satellite record there is
- [NASA Earthdata Search](https://search.earthdata.nasa.gov/): Hundreds of science datasets in one search, for research-grade work
- [NASA Worldview](https://worldview.earthdata.nasa.gov/): Whole-Earth imagery from MODIS and VIIRS, often hours old
- [Google Earth](https://earth.google.com/web): The easiest general-purpose viewer, with a mosaic that can be months to years old
- [Zoom Earth](https://zoom.earth/): Near-current weather and environmental viewing
- [Maxar Open Data Program](https://www.maxar.com/open-data): Free high-resolution imagery released during major disasters
- Weather satellite viewers ([BoM SatView](http://satview.bom.gov.au/), [NOAA GOES viewer](https://www.star.nesdis.noaa.gov/GOES/)): Geostationary imagery updating every few minutes
- National open-data programs ([Digital Earth Australia](https://www.ga.gov.au/scientific-topics/dea), [National Map](https://nationalmap.gov.au/) and their counterparts): Country-level archives and property-scale viewers

Looking for sources specific to your country? We keep dedicated guides for [Australia](/blog/free-sources-satellite-data-australia), the [USA](/blog/free-sources-satellite-data-usa), the [UK](/blog/free-sources-satellite-data-uk), [Canada](/blog/free-sources-satellite-data-canada), [Germany](/blog/free-sources-satellite-data-germany) (also [auf Deutsch](/blog/kostenlose-satellitenbilder-deutschland)) and [Brazil](/blog/free-sources-satellite-data-brazil).

<AskBand headline="What does your site look like this week?" sub="Unlimited free Sentinel-2 with indices, no credit card." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/dunes.webp', '/band/fields.webp', '/band/atoll.webp']} />

## Overview

Whether you want to look at your own property, monitor vegetation across a region, or pull rasters into a GIS, there is a free source that covers it. This guide walks through the ten worth knowing in 2026, what each is best at, and where each one stops being enough.

### Understanding Real-Time Satellite Imagery

Many people search for "live satellite view" or "real-time satellite images" of their house, but despite what movies suggest, live satellite imagery isn't available with current technology. What you see on mapping platforms is usually imagery captured weeks or months ago. The closest things to live are the weather satellites (minutes old, but kilometre-scale pixels) and NASA's Worldview (hours old, 250 m to 1 km pixels).

### What Free Data Can and Can't Do

- Most free imagery basemaps are updated every few months to years
- Free downloadable data tops out at 10 m resolution (Sentinel-2): you can see fields and buildings as shapes, not details
- High-resolution imagery with clear detail (individual vehicles, fence lines) is commercial
- You cannot task a free satellite: if nothing captured your site cloud-free recently, there is nothing to download

With that calibration done, here are the sources.

---

## 1. Pera Portal Free Tier

Our own [Pera Portal](https://portal.geopera.com/) has a free tier built on Sentinel-2: unlimited browsing with 45+ pre-configured spectral indices (NDVI, NDWI, EVI and the rest) computed in the browser, so you can check vegetation or water condition without downloading anything or setting up software.

**Key Features:**

- Unlimited free Sentinel-2 viewing, no credit card
- 45+ spectral indices ready to apply, plus custom band ratios
- The same interface then searches 100+ commercial satellites when a project needs sharper detail

---

## 2. Copernicus Browser

The European Union's [Copernicus Data Space Ecosystem](https://browser.dataspace.copernicus.eu/) is the definitive source for Sentinel data. Sentinel-2 images every point on Earth at 10 m resolution roughly every 5 days, and the browser searches, previews and downloads the full rasters.

**Key Features:**

- Free registration, full-resolution downloads
- Sentinel-1 radar and Sentinel-3 ocean/land products alongside Sentinel-2
- The standard choice when you need the actual data in a GIS

---

## 3. USGS EarthExplorer

[EarthExplorer](https://earthexplorer.usgs.gov/) is the front door to the Landsat archive: 30 m resolution reaching back to 1972, which makes it the longest continuous satellite record of the Earth's surface. If the question is "what did this place look like decades ago", this is the answer.

---

## 4. NASA Earthdata Search

[NASA Earthdata Search](https://search.earthdata.nasa.gov/) indexes hundreds of datasets across NASA's missions. It is built for research rather than casual browsing, and it is where you go for atmospheric, ocean and land products beyond plain imagery.

---

## 5. NASA Worldview

[Worldview](https://worldview.earthdata.nasa.gov/) shows the whole planet from MODIS and VIIRS, usually within hours of the overpass. The pixels are coarse (250 m at best), but for fires, smoke, dust and floods at regional scale it is the fastest free view there is.

---

## 6. Google Earth

[Google Earth](https://earth.google.com/web) remains the easiest way to look at high-resolution imagery anywhere on the globe. The catch: the basemap is a mosaic whose capture dates you can't control, often months to years old, and the imagery can't be downloaded as data. With [Google Earth Pro desktop being discontinued](/blog/google-earth-pro-desktop-discontinued), the web version is the one with a future.

---

## 7. Zoom Earth

[Zoom Earth](https://zoom.earth/) provides frequently updated imagery focused on weather and environmental conditions, and is an easy way to track storms, fires and other near-current events.

---

## 8. Maxar Open Data Program

[Maxar](https://www.maxar.com/open-data) releases high-resolution imagery for free during major natural disasters worldwide, supporting emergency response and humanitarian work. Outside disaster events the archive stays closed, but during one it is often the sharpest free imagery available.

---

## 9. Weather Satellite Viewers

Geostationary weather satellites image continuously, and their agencies publish free viewers: Australia's Bureau of Meteorology runs [SatView](http://satview.bom.gov.au/) and NOAA runs the [GOES image viewer](https://www.star.nesdis.noaa.gov/GOES/) for the Americas. Updates land every few minutes, which makes these the only genuinely near-live free sources, at kilometre-scale resolution suited to weather rather than property viewing.

---

## 10. National Open-Data Programs

Most countries run their own satellite and aerial open-data platforms, and they often beat the global sources for local detail. Australia has [Digital Earth Australia](https://www.ga.gov.au/scientific-topics/dea) (analysis-ready time series back to 1988) and [National Map](https://nationalmap.gov.au/) (browser viewing with property boundaries), plus state platforms like NSW's SIX Maps and Queensland Globe; our guide to [free satellite data for Australia](/blog/free-sources-satellite-data-australia) covers them all. Other countries' equivalents are covered in the dedicated guides: [USA](/blog/free-sources-satellite-data-usa), [UK](/blog/free-sources-satellite-data-uk), [Canada](/blog/free-sources-satellite-data-canada), [Germany](/blog/free-sources-satellite-data-germany) and [Brazil](/blog/free-sources-satellite-data-brazil).

---

## Tips for Making the Most of Free Satellite Imagery

**1. Match the platform to the question.** Viewers (Google Earth, Zoom Earth, National Map) answer "what does it look like". Data platforms (Copernicus Browser, EarthExplorer, Earthdata) answer "what can I measure".

**2. Check the capture date, always.** The single most common mistake with free imagery is treating a years-old basemap tile as current.

**3. Define your requirements before searching.** Area of interest, time range, resolution and bands. Search filters do the rest.

**4. Use the built-in analysis tools.** Indices and time-series tools inside the platforms often answer the question without any downloads.

**5. Download selectively.** Pull the tiles and dates you need rather than whole datasets; the archives aren't going anywhere.

---

## When Free Satellite Data Isn't Enough

Free platforms cover an enormous range of uses, and if you're simply curious about your own property, they're genuinely all you'll ever need. But if you're using imagery professionally, you'll eventually hit the resolution wall:

| Source               | Resolution  | What you can actually see                                  |
| -------------------- | ----------- | ---------------------------------------------------------- |
| Landsat (free)       | 30 m        | Regional land cover: a sports field is roughly one pixel   |
| Sentinel-2 (free)    | 10 m        | Field-scale vegetation patterns; buildings are blurs       |
| Commercial (Geopera) | up to 30 cm | Individual vehicles, fence lines, stockpiles, single trees |

The difference matters the moment you need to **measure rather than look**: tracking earthworks on a mine site, auditing stockpiles, mapping erosion along a specific creek line, monitoring vegetation by the row rather than by the field, or proving site conditions on a particular date. Free sensors also can't be tasked: if no satellite happened to capture your site cloud-free last month, there's nothing to download.

Commercial imagery used to mean opaque quotes and weeks of back-and-forth. We publish [transparent per-square-kilometre pricing](/pricing) for both tasking and archive, our guide on [how to buy satellite imagery](/blog/how-to-buy-satellite-imagery) covers the process end to end, and every order arrives analysis-ready: orthorectified, pansharpened, colour-balanced and mosaicked ([here's exactly what that involves](/imagery)).

If free data has taken your project as far as it can go, [explore available imagery through Pera Portal](https://portal.geopera.com/) or [get in touch to discuss your project](/contact).

---

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# What is Satellite Imagery?

> What satellite imagery is, how it is captured, resolutions from 30 cm to 30 m and what each shows, archives back to 1972, and what it costs per km².

Published: 2024-10-10 | Author: Darcy Weedman | Reading time: 7 min

Source: https://geopera.com/blog/what-is-satellite-imagery

---
<script>
import { AskBand } from '$lib/components/charts';
</script>

## Summary

Satellite imagery is imagery of the Earth's surface captured by sensors on orbiting satellites. Commercial systems resolve 30 cm to 10 m per pixel, revisit the same location daily to every few days, and hold archives reaching back decades: the free Landsat record starts in 1972 at 30 m, and commercial archive imagery is priced at roughly $2-55+ USD per km².

That is the short answer. The rest of this guide covers how the images are captured, what each resolution actually shows, the main imagery types, what industries do with it, and where the free data ends and the commercial data begins.

<AskBand headline="What does home look like from 786 km up?" sub="Free Sentinel-2, no card, no install: build intuition on real data." action="Open the free tier" href="https://portal.geopera.com" tiles={['/band/city.webp', '/band/atoll.webp', '/band/dunes.webp']} />

## How Satellite Imagery Is Captured

A satellite image is not a photograph in the everyday sense. It is measured radiation, turned into a picture through a chain with four stages:

1. **Acquisition.** Sensors on the satellite record energy from the Earth's surface: reflected sunlight for optical sensors, emitted heat for thermal, or the echo of the satellite's own radar pulse for SAR.
2. **Downlink.** The data is transmitted to ground stations on the satellite's next pass, or relayed in near real time.
3. **Processing.** Raw data is corrected for sensor characteristics, atmosphere, and terrain-induced geometric distortion. Done fully, this chain is orthorectification, pansharpening, atmospheric correction, colour balancing and mosaicking; we've written up [why that processing matters](/blog/why-we-process-every-order) and what skipping it costs in accuracy.
4. **Analysis.** The processed image goes into a GIS or analysis pipeline, where it becomes measurements: areas, volumes, vegetation indices, change maps.

The Landsat program, run by NASA and the USGS, has operated this loop continuously since 1972, which is why a 50-year time series of any place on Earth exists at 30 m resolution. Modern commercial constellations run it at far higher resolution and cadence: Maxar's WorldView Legion satellites resolve 30 cm, and Geopera's Perascope constellation delivers high-cadence sub-metre revisit through our exclusive access rights.

## Resolution: What You Can Actually See

Resolution is the single spec that matters most, and the differences are concrete:

| Resolution  | Example source          | What a pixel shows                                       |
| ----------- | ----------------------- | -------------------------------------------------------- |
| 30 cm       | WorldView-3, Legion     | Individual vehicles, fence lines, road markings          |
| 50 cm       | Perascope, SuperView    | Buildings, site layouts, stockpiles                      |
| 1-2 m       | Various archive sensors | Land parcels, large structures, clearings                |
| 10 m (free) | Sentinel-2              | Field-scale vegetation patterns; buildings become blurs  |
| 30 m (free) | Landsat                 | Regional land cover; a sports field is roughly one pixel |

Two other specs complete the picture. **Revisit rate** is how often a satellite can image the same spot: Sentinel-2 covers everywhere roughly every 5 days, and the large commercial constellations manage daily or better. **Spectral bands** determine what can be measured rather than merely seen: near-infrared and [red edge](/blog/red-edge-remote-sensing) bands drive vegetation analysis, and shortwave infrared supports mineral mapping.

## The Main Types of Satellite Imagery

**Optical** imagery records reflected sunlight, in the visible bands and beyond. It is the most intuitive type to interpret and the workhorse for mapping and monitoring, with the limitation that clouds block it and it needs daylight.

**Radar (SAR)** imagery is built from the satellite's own microwave pulses, so it works through cloud and at night. It reads as texture and structure rather than colour, and supports measurements optical cannot make, such as millimetre-scale ground deformation. Our [SAR explainer](/blog/sar-satellite-imagery-explained) covers how to read it.

**Multispectral and hyperspectral** imagery captures many narrow bands across the spectrum, from a handful (multispectral) to hundreds (hyperspectral), revealing properties like crop stress or mineralogy that RGB cannot. The [difference between the two](/blog/multispectral-vs-hyperspectral-satellite-imagery) is mostly a trade between spectral detail and coverage.

**Thermal** sensors measure emitted heat, used for energy audits, volcanic monitoring and water temperature. **Stereo** collection captures the same area from two angles on one pass, which is how [elevation models are built from satellites](/blog/stereoscopic-satellite-imagery).

## What Industries Do With It

The common thread is measurement over large or hard-to-reach areas, repeated on a schedule no ground survey can match.

**[Agriculture](/agriculture).** Vegetation indices computed from multispectral bands flag crop stress before it is visible from the road; growers use them to target irrigation, fertiliser and inspection.

**[Mining](/mining).** Exploration teams screen geology before fieldwork; operating mines monitor disturbance footprints, stockpiles and rehabilitation progress against regulatory commitments.

**[Environmental monitoring](/environmental).** Deforestation, coastal change, water quality and carbon-project baselines, tracked as time series across decades of archive.

**[Infrastructure](/infrastructure) and construction.** Progress documentation with capture dates that stand up to a regulator or a dispute, plus corridor monitoring for pipelines and transmission lines.

**[Energy](/energy), insurance and government.** Site selection for renewables, catastrophe damage assessment after storms and floods (radar working through the weather), and mapping programs at national scale.

The pattern across all of them: the imagery is the raw material, and the value arrives when it is processed well enough to measure from, which is why processing quality separates providers more than sensor lists do.

## Free vs Commercial Satellite Imagery

A great deal of satellite imagery is free. Sentinel-2 (10 m, every ~5 days) and Landsat (30 m, since 1972) cover the whole planet, and our guide to the [10 free sources of satellite data](/blog/free-sources-of-satellite-data) covers where to get them. Free data genuinely answers regional-scale questions: land cover, broad vegetation trends, water extent.

Commercial imagery starts where pixels need to be smaller than a building. It runs roughly $2-55+ USD per km² depending on resolution and whether you buy archive or task a new capture; the full numbers are in our [satellite imagery cost guide](/blog/satellite-imagery-cost-guide), and the process end to end is in [how to buy satellite imagery](/blog/how-to-buy-satellite-imagery).

## Frequently Asked Questions

### What is satellite imagery used for?

Measurement at scale: monitoring crops, mines, forests, coastlines and infrastructure; mapping; damage assessment after disasters; and documenting site conditions on a known date. Any task that needs a repeated, consistent view of a large or remote area is a satellite imagery task.

### How is satellite imagery different from aerial imagery?

Aerial imagery is captured from aircraft at a few thousand metres and resolves 5-15 cm over campaign areas. Satellite imagery is captured from orbit at 30 cm to 30 m, covers anywhere on Earth without mobilising anything, and carries decades of archive. Our [drone vs aerial vs satellite comparison](/blog/drone-vs-aerial-vs-satellite-imagery) maps which platform fits which job.

### How often is satellite imagery updated?

It depends on the source. Free Sentinel-2 images everywhere roughly every 5 days; large commercial constellations revisit daily or better; and web basemaps like Google Earth are mosaics that may be months to years old. For a guaranteed fresh capture, imagery is [tasked](/blog/satellite-tasking-explained) rather than pulled from archive.

### How accurate is satellite imagery?

Resolution and positional accuracy are separate things. Commercial sensors resolve 30-50 cm per pixel, and after proper [orthorectification](/blog/orthorectification-explained) positions are accurate to a few metres or better. Imagery without that correction can be displaced by tens of metres in terrain, which is why processing level matters when measurements face scrutiny.

### Is satellite imagery free?

At 10-30 m resolution, yes: Sentinel-2 and Landsat are free and open. Sub-metre imagery is commercial, priced per km² of the area ordered, from about $2/km² for coarse archive to $55+/km² for 30 cm tasking.

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

## Case Studies

# Using WorldView-3 SWIR to Enable Individual Alteration Mineral Mapping

> How 16-Band SWIR Imagery Enables Precise Hydrothermal Alteration Mapping for Mineral Exploration

Published: 2025-05-26 | Author: Geopera Team | Reading time: 3 min

Source: https://geopera.com/case-studies/wv3-mineral-exploration

---
## The Challenge: Remote Mineral Exploration

Traditional mineral exploration requires expensive field surveys to identify promising areas. Hydrothermal alteration minerals - the key indicators of potential mineralisation - create unique spectral signatures in the shortwave infrared (SWIR) range that can be detected remotely, vastly improving how we approach early-stage exploration.

## WorldView-3: Game-Changing SWIR Technology

WorldView-3's advanced 8-band SWIR sensor (1195-2365 nm at 3.7m resolution) captures diagnostic absorption features of critical alteration minerals:

- **Phyllosilicates**: Kaolinite, illite, smectite, muscovite, chlorite
- **Sulfates**: Alunite, jarosite, gypsum
- **Iron minerals**: Hematite, goethite, pyrite
- **Carbonates**: Calcite, dolomite

## Breakthrough Results

Working with a geologically complex area featuring chloritized granodiorite, metamorphosed volcanics, and known mineralisation zones, Arc GeoTech achieved:

**Precise Mineral Mapping**

- Successfully identified and mapped distinct alteration mineral assemblages
- Detected clay-rich zones with kaolinite, illite, smectite, and muscovite
- Mapped iron oxide concentrations and sulfate mineral occurrences
- Distinguished between different alteration types and intensities

**Structural Discovery**

- Revealed high-density fault networks with primary NW-SE and NNW-SSE trends
- Demonstrated strong correlation between alteration zones and structural lineaments
- Identified structural controls on mineralisation pathways

## Advanced Analysis Methods

Arc GeoTech's multi-technique approach combined:

- **Band ratio analysis** targeting specific mineral signatures
- **Spectral Angle Mapper (SAM)** classification with reference spectra
- **Multi-band composites** optimised for alteration discrimination
- **High-resolution lineament analysis** using 30cm panchromatic data

---

## Get the Complete Technical Analysis

The comprehensive report delivers:

- **Step-by-step processing workflows** with detailed parameters
- **Complete mineral distribution maps** with classification confidence
- **Structural analysis** including rose diagrams and lineament density maps
- **Band ratio formulations** for different mineral targets
- **Visual spectral comparisons** across all composite combinations
- **Geological targeting recommendations** for follow-up exploration

**[Download the full technical case study →](#download)**

---

_Case study prepared by Arc GeoTech utilising Geopera data_

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# Scaling Carbon Stock Assessment by Leveraging High-Resolution Satellite Imagery

> How Commercial Satellite Data Delivers 28% Better Vegetation Detection at a Fraction of the Cost

Published: 2025-03-28 | Author: Geopera Team | Reading time: 5 min

Source: https://geopera.com/case-studies/scaling-carbon-stock-assessment

---
## The Carbon Measurement Challenge

As carbon stock assessment scales from local projects to regional territories, organizations face mounting pressure to:

- **Measure accurately** across increasingly vast areas
- **Reduce costs** while maintaining data quality
- **Deliver timely assessments** to meet regulatory requirements
- **Detect smaller vegetation features** critical for precise carbon calculations

## Key Findings from Our Comparison Study

Our analysis compared high-resolution commercial satellite imagery (Perascope) with freely available Sentinel-2 data:

- **28% increase in vegetation detection** using Perascope's 50cm resolution
- **Individual tree crown identification** versus only detecting larger vegetation clusters
- **Significantly improved boundary accuracy** for more precise carbon calculations
- **Daily revisit capability** compared to Sentinel-2's 5-day revisit frequency

## Cost-Effective Solution

At just **$0.12 AUD per hectare**, Perascope imagery offers:

- ✅ **5-10x cost savings** compared to traditional aerial or drone surveys
- ✅ **Superior resolution** to free satellite options (50cm vs 10m)
- ✅ **Scalability** across vast assessment territories
- ✅ **No practical deployment limitations** unlike drone surveys

---

## Download the Complete Case Study

The full report includes:

- Detailed methodology of our comparative analysis
- Visual examples showing detection differences
- Cost-benefit analysis across multiple project scales
- Implementation guidelines for integrating high-resolution data into your carbon assessment workflow
- Complete comparative specifications of remote sensing options

**[Access the full case study now →](#download)**

---

_Prepared by Geopera | Helping organizations make better environmental decisions with precision satellite data_

---

*This article was published by [Geopera](https://geopera.com). Every order includes the full processing chain, from raw to orthorectified to surface reflectance to pansharpened mosaics, delivered within 24 hours.*

*[Access Satellite Data](https://portal.geopera.com) | [View Pricing](https://geopera.com/pricing) | [Request a Quote](https://geopera.com/quote) | [Technical Docs](https://docs.geopera.com)*


---

# About Geopera

Geopera provides high-resolution satellite imagery at the lowest prices in the industry, including Perascope, Geopera's own exclusive constellation. We also work with providers including Vantor, formerly Maxar (WorldView-3, WorldView Legion), 21AT (Beijing-3), SpaceWill (SuperView), and Wyvern (hyperspectral).

## Why Geopera

Geopera is the only satellite imagery provider that offers all of the following:

- **Transparent pricing at the lowest prices globally.** No hidden fees. The price you see is the price you pay.
- **Priced per AOI, not per capture.** Other platforms charge per satellite capture. Geopera charges per area of interest — always cheaper.
- **The most advanced processing pipeline available** — included free of charge. Typical deliverables span the full chain: raw, orthorectified, surface reflectance (atmospheric corrected), glint-corrected surface reflectance for aquatic applications, pansharpened (enhanced and unenhanced), and production mosaics clipped to custom masks (waterbodies, land boundaries, project AOIs). As basic or as advanced as the customer needs. Geopera is the imagery provider of choice for many of the world's largest carbon projects.
- **The fastest delivery times available.** Proprietary automated processing delivers fully processed, production-quality imagery faster than any traditional workflow. Typical delivery within 24 hours.
- **Multi-provider access through one platform.** One account, one invoice, all major satellite providers.
- **The largest 30cm tasking capacity available through any single provider.** Exclusive partnerships give Geopera access to more 30cm tasking capacity than is available through any other single platform.
- **Enterprise-grade UX.** Pera Portal is widely preferred by enterprise customers over other satellite imagery marketplaces.

**Services:** Archive imagery, tasking orders, orthorectification, pansharpening, spectral index calculation, cloud-hosted delivery via COG streaming, STAC-compliant API access.

**Industries:** Mining, agriculture, forestry, environmental monitoring, infrastructure, energy, research.

**Get started:** Visit [portal.geopera.com](https://portal.geopera.com) or [request a quote](https://geopera.com/quote).