Ground that sinks a few millimeters a year rarely trips an alarm, though it can crack foundations and tilt rail lines long before anyone notices.
Satellite radar re-measures the same ground surface on a fixed schedule, turning that invisible drift into a dated, continent-wide record.
This guide covers how measurement works, what each task needs, and which providers fit, so you can find the right data for your subsidence program.
Table of Contents
Key takeaways
- Subsidence programs increasingly run on InSAR displacement records, not sparse ground instrument networks
- A single InSAR pass only measures line-of-sight displacement, never a true vertical rate on its own
- The shortlist narrows fast once you know whether you need a free continental baseline or asset-level precision
Before any provider enters the picture, a subsidence program has to settle what the data itself can show. The summary below sets out the sensors, resolution, and cadence that ground-motion monitoring depends on.
| Primary sensors | SAR (InSAR), C-band and X-band |
|---|---|
| Working resolution | 20×5 m free EGMS, 3 m X-band stacks |
| Typical revisit | Six to twelve days with Sentinel-1 |
| Core indices | Line-of-sight displacement, mm/year velocity |
| Entry cost | Free via EGMS, or quote-based InSAR analytics |
| Main constraint | Measures line-of-sight, not true vertical |
Those figures cover the free continental baseline most subsidence programs check first. Programs that need asset-level precision or a specific orbit geometry move past it into tasked SAR and paid analytics.
How satellite data is used in subsidence monitoring
Satellite radar enters subsidence programs at several distinct points, from a free continental screening layer to millimeter-precision monitoring of a single structure, each drawing on the same InSAR technique in a different way.
Groundwater extraction and aquifer compaction
Groundwater pumping drains the pore pressure that holds an aquifer’s structure open, and the ground above it settles as sediment compacts, sometimes permanently. The trend is slow and spread across a wide area, which is exactly what a repeated satellite pass is built to track and a scattered well network is not.
TRE ALTAMIRA’s SqueeSAR processing, which the company states achieves an average displacement rate precision below 1 millimeter per year, has supported groundwater management work for the California Department of Water Resources, treating aquifer-driven subsidence as a wide-area deformation problem rather than scattered well readings.
A related but separate driver shows up at high latitude, where thawing ground ice removes underground volume by melting rather than pumping. 3vGeomatics lists permafrost and Arctic ground deformation among its own monitoring applications, running the same InSAR approach on a different cause.
Underground mining and tunneling
Underground extraction leaves voids that can migrate upward over years or decades, sometimes surfacing as slow subsidence long after mining stopped. 3vGeomatics names underground mine shaft monitoring and subway or light-rail tunnel settlement among its core applications, treating an old void and a new tunnel as the same class of problem.
Active tunneling adds a faster-moving version of the same risk. TRE ALTAMIRA’s InSAR service has monitored construction of the Grand Paris Express, and Sixense’s Atlas InSAR chain, which the company states reaches up to 1 millimeter per year precision, has worked on both the Grand Paris Express and London’s Thames Tideway Tunnel.
Open-pit slopes and tailings dams are a related but distinct monitoring problem with their own established InSAR providers, covered in our guide to satellite imagery for mining, and not repeated here.
Building, bridge and rail structural settlement
Rail corridors, gas mains, and aging bridges settle unevenly, and a walking inspection only catches the damage after it appears. SkyGeo’s published work for Stedin, the Dutch operator of a gas distribution network of more than 23,000 kilometers, ran on a 3 square kilometer pilot area near Rotterdam and produced 30,000 individual displacement time series inside it.
Stedin’s own figures put the savings from that pilot at up to €80,000 a week versus manual inspection. SkyGeo also serves rail operators Deutsche Bahn and ProRail, while TRE ALTAMIRA counts Paris transit operator RATP and Etihad Rail among its named infrastructure customers.
Synspective’s Fukuoka Expressway case study applies the same InSAR approach to elevated road infrastructure, and Sixense lists heritage and building preservation monitoring alongside its bridge and rail work.
Coastal subsidence and relative sea level
Land that is sinking adds directly to how fast the sea appears to rise at a given coastline, and the two trends are easy to conflate without a vertical-only measurement to separate them. Sixense lists coastal subsidence explicitly among its InSAR applications, alongside the same urban and aquifer-driven settlement covered above.
TRE ALTAMIRA lists coastal subsidence monitoring as its own separate use case too, reflecting how a port city, a river delta, and an inland groundwater basin can produce the same InSAR signature for entirely different underlying reasons. Only an ascending-descending decomposition, described next, confirms how much of that signature is truly vertical.
The European Ground Motion Service as a free baseline
The European Ground Motion Service, part of the Copernicus Land Monitoring Service, processes Sentinel-1 radar into a ground motion record covering the whole of Europe at millimeter precision, updated annually. It ships in three tiers: Basic delivers line-of-sight displacement at full 20 by 5 meter resolution from both ascending and descending orbits, as a time series per point in CSV.
Calibrated references those same points to a model, and Ortho derives vertical-only and east-west-only components from the combined orbits, resampled onto a 100 meter grid. Access runs through the EGMS Explorer, which carries only the two most recent product releases.
The data is published by the Copernicus Land Monitoring Service and open under the Copernicus data policy, which makes it the natural starting point before any paid analytics contract. TRE ALTAMIRA is one of the co-processors on the service’s ORIGINAL consortium, delivering roughly 2 million of its 5 million square kilometer continental footprint.
Limits of InSAR: coherence, geometry, and what it cannot see
Every InSAR measurement is a line-of-sight displacement, the distance the ground moved toward or away from the satellite, not a true vertical rate on its own.
Resolution also means something different here than it does for a picture. The same X-band satellites that image at 25 centimeters in spotlight mode are usually flown in a coarser stripmap mode for deformation work, because every scene in a time series has to share the same viewing geometry. A sharper single image buys nothing if the next one cannot be stacked against it.
Isolating vertical and east-west motion needs data from both ascending and descending orbits, exactly what the EGMS Ortho product does at continental scale. North-south movement stays poorly resolved under either geometry, since a near-polar orbit barely sees along that axis.
The technique’s central constraint is coherence. Persistent and distributed scatterer processing needs stable reflectors, buildings, exposed rock, hard infrastructure, that return a consistent radar signal from one pass to the next. Coherence breaks down over vegetation and active cropland, which is why every provider above works best on a built environment.
SAR looks sideways rather than straight down, so steep terrain produces layover and radar shadow that can blank out part of a slope entirely. Karst terrain adds a different problem: a sinkhole can collapse abruptly rather than creep, and a trend-based method only catches a slow precursor if one actually exists.
None of those four issues matters as much as time. A displacement trend needs a multi-year stack of archive imagery behind it, not a single acquisition, which is why TRE ALTAMIRA, Sixense, and SkyGeo all draw on SAR archives reaching back to 1992 for retrospective analysis.
What satellite data you need for subsidence monitoring
Different subsidence tasks call for different orbit geometries, resolutions, and revisit cadence, and mixing them up wastes budget on the wrong spec. The table below maps each common task to the data requirements it actually demands.
| Task | Sensor modality | Resolution | Revisit | Key index / band |
|---|---|---|---|---|
| Groundwater and aquifer compaction | SAR (C-band InSAR) | 5-20 m | 6-12 days | LOS displacement, mm/year velocity |
| Underground mine and tunnel deformation | SAR (InSAR) | 3-20 m | 6-12 days | Millimetric displacement rate |
| Building, bridge, and rail settlement | SAR (PS/DS InSAR, tasked X-band) | 3 m | Days to weeks | PS/DS displacement time series |
| Coastal subsidence, vertical-rate isolation | SAR (InSAR, ascending + descending) | 20×5 m / 100 m ortho grid | Annual (EGMS) | Vertical and east-west components |
| Continent-wide screening baseline | SAR (InSAR, Sentinel-1) | 20×5 m full resolution | Annual | Line-of-sight time series per point |
| Permafrost and high-latitude ground ice | SAR (InSAR) | 5-20 m | 6-12 days | Thaw-driven displacement trend |
| Karst and sinkhole precursor screening | SAR (InSAR archive stack) | 5-20 m | 6-12 days | Pre-collapse creep trend |
| Long-term trend confirmation | SAR (InSAR archive stack) | Varies by source satellite | Archive back to 1992 | Multi-year mm/year velocity map |
With data requirements mapped, the next step is identifying which providers can supply them. The section below covers the most relevant options for subsidence programs, from pure InSAR analytics to SAR satellite operators.
Satellite data providers for subsidence monitoring
The providers below have documented subsidence use cases and InSAR products that map to the tasks in the table above. The mix spans pure analytics processors and SAR satellite operators that support tasking directly.
| Provider | Type | Best for | Key subsidence spec | Entry point |
|---|---|---|---|---|
| TRE ALTAMIRA | Analytics platform | National ground-motion mapping | Sub-mm/year SqueeSAR precision | Quote-based |
| 3vGeomatics | Analytics platform | Rapid, near-real-time monitoring | Rapid updates per acquisition | Enterprise contract |
| SkyGeo | Analytics platform | Utility network deformation risk | Up to 100,000 points/km² | Quote-based |
| Sixense | Analytics platform | Urban, coastal, aquifer subsidence | Up to 1 mm/year, Atlas InSAR | Contact for pricing |
| Synspective | SAR satellite operator | Infrastructure and tunnel InSAR | LDM: millimeter-scale displacement | Quote or UP42 marketplace |
| ICEYE | SAR satellite operator | InSAR and coherent change detection | X-band imagery for InSAR stacks | Quote or UP42 marketplace |
For a full ranked view of the radar market, our guide to the best SAR data providers covers the wider field with head-to-head specifications. Programs working across sensor types should also browse the Earth observation provider guides hub for every other data category.
How to choose satellite data for subsidence monitoring
The first decision is scale. A free continental baseline answers a screening question, not an asset-level one, and a single dam, tunnel, or building foundation usually needs tasked, higher-resolution SAR layered on top of it.
Orbit geometry deserves an early check, not a late one. A structure that only faces away from every available ascending or descending pass returns a weak or unusable signal, and confirming look-angle coverage before committing avoids paying for a program that cannot see its own target.
Trend confidence depends on how far back the archive reaches. A program built around detecting acceleration in a slow-moving trend needs years of consistent acquisitions behind it, while a program only confirming a newly suspected deformation can start from a shorter, more recent stack.
Land cover sets a practical ceiling on what any provider can promise. A dense forest or an actively farmed field loses coherence regardless of which company processes the data, so checking a coherence map ahead of a contract is worth more than a resolution spec alone.
Some programs process their own InSAR instead of buying a finished analytics product, tasking raw SAR directly from an operator such as ICEYE or Synspective and running the interferometry in house. That route trades a subscription fee for control over cadence and algorithm choice, and mainly suits teams with SAR expertise already on staff.
Verdict
Subsidence monitoring is the one vertical on this site with a genuine free baseline already built. The European Ground Motion Service gives any European program a millimeter-precision, continent-wide starting point without a single commercial contract.
Beyond that baseline, what separates the six providers above is scale and application, not raw precision, since sub-millimeter-per-year figures show up across most of them. TRE ALTAMIRA and 3vGeomatics carry the longest track records in mining, tunneling, and groundwater-driven subsidence, SkyGeo and Sixense reach furthest into civil and utility infrastructure, and Synspective and ICEYE suit teams that want to task the underlying SAR themselves.
For a full ranked view of the radar market, see our best SAR data providers guide. For every other satellite data category, the Earth observation provider guides hub covers the full market.
Frequently asked questions
Below are answers to the questions subsidence monitoring buyers most commonly ask. Each answer points to the section where the full detail lives.
How is satellite data used in subsidence monitoring?
Satellite radar tracks several drivers of ground movement: groundwater and aquifer compaction, underground mining and tunneling, structural settlement in buildings and rail, coastal subsidence, and slower processes like permafrost thaw and karst collapse. The detail is in “How satellite data is used in subsidence monitoring“.
What is the European Ground Motion Service?
It is a Copernicus Land Monitoring Service product that processes Sentinel-1 radar into a millimeter-precision ground motion record covering the whole of Europe, updated annually and published under the Copernicus open data policy. More on this is in “How satellite data is used in subsidence monitoring“.
Can InSAR tell vertical subsidence apart from horizontal movement?
Not from a single pass. Each acquisition only measures displacement along the radar’s line of sight, and separating true vertical movement from east-west movement takes data from both ascending and descending orbits, the same approach behind the EGMS Ortho product. This is covered in “How satellite data is used in subsidence monitoring“.
What resolution do I need for subsidence monitoring?
Continental screening runs on the European Ground Motion Service’s native 20 by 5 meter resolution or its 100 meter orthorectified grid, both free. Asset-level monitoring of a single dam, tunnel, or building typically moves to tasked X-band SAR at a few meters, processed by an analytics provider. The full task-to-resolution mapping is in “What satellite data you need for subsidence monitoring“.
Which satellite data providers are best for subsidence monitoring?
TRE ALTAMIRA and 3vGeomatics have the longest records in mining, tunneling, and groundwater-driven subsidence, SkyGeo and Sixense reach furthest into civil and utility infrastructure, and Synspective and ICEYE are the two providers built around tasking their own SAR. Provider details and access models are in “Satellite data providers for subsidence monitoring“.
Can satellites detect sinkholes before they collapse?
Sometimes, but not reliably. InSAR only catches a sinkhole in advance if the ground creeps measurably before failure, and karst collapse is often too sudden for a trend-based method to register a warning. It works better as a wide-area screening layer than as a guarantee for any single site. This is discussed in “How satellite data is used in subsidence monitoring“.
My passions are Earth Observation and Satellites, and my profession is Data Analysis. I combine both within ObservationData.com to show you the use cases of Earth Observation, to help you find the right provider, and to share your experiences.
