Optical satellites record visible light reflected by the Earth’s surface. A radar works differently: it emits its own electromagnetic wave and measures the echo reflected back toward the satellite.

This capability allows observations day and night, and through many atmospheric conditions.

The Sentinel-1 satellites of the European Copernicus programme repeatedly acquire radar images of the Earth with strong temporal regularity.

Each radar pixel contains amplitude information, but also phase information related to the distance travelled by the radar wave between the satellite and the ground.

InSAR stands for “Interferometric Synthetic Aperture Radar”. The principle consists in comparing several radar acquisitions taken over the same area at different dates.

If the ground slightly moves between two acquisitions, the travel distance of the radar wave also changes. This produces a measurable phase difference.

These phase differences can then be converted into extremely small displacements, sometimes at millimetric scale.

But the observed signal never corresponds only to ground motion. It also contains atmospheric, geometric, orbital and instrumental effects.

A radar satellite mainly measures displacement along its Line Of Sight (LOS).

This means that an observed displacement can combine several components: vertical, horizontal or inclined motion.

Vertical subsidence, horizontal sliding or slope-related deformation can sometimes produce similar radar signatures.

Geophysical interpretation therefore strongly depends on topography, geology and acquisition geometry.

InSAR time series always contain noise.

The atmosphere slightly modifies radar wave propagation. Water vapour, meteorological variability and ionospheric disturbances can generate important spatial artefacts.

Surface changes also play a major role: vegetation, agriculture, soil moisture or urbanisation affect radar coherence over time.

InSAR processing therefore attempts to separate actual deformation from atmospheric and instrumental noise. This step is fundamental.

A single InSAR image has limited meaning. The real value comes from time series.

Observing signal evolution over several years makes it possible to detect long-term trends, seasonal cycles or responses to hydrological and climatic events.

In the context of clay shrink–swell, InSAR time series can reveal seasonal deformation dynamics related to soil moisture variations.

Temporal analysis therefore becomes more important than the static map itself.

An InSAR signal never automatically provides the cause of deformation.

Similar deformation patterns may result from very different processes: clay shrink–swell, groundwater pumping, embankment consolidation, mining activity, thermal effects, tectonic motion or slope instability.

InSAR must therefore be interpreted together with geology, hydrology, field observations, geotechnics and land use.

The satellite measures a dynamic. Understanding its origin requires broader territorial analysis.

The main strength of InSAR lies in its ability to document slow, diffuse and otherwise difficult-to-detect phenomena.

Ground motion is not always spectacular. Many processes evolve progressively over several years, with small but repeated amplitudes.

These slow dynamics can nevertheless produce significant impacts on buildings, infrastructure and territories.

Satellite observation makes it possible to place local damage within a broader spatial and temporal context.

InSAR maps sometimes give an impression of absolute precision. Yet each pixel results from complex processing, methodological choices and multiple sources of uncertainty.

A colour alone is never sufficient evidence.

Understanding ground motion therefore requires connecting geology, hydrology, radar geometry, time series and territorial context.

The objective is not only to produce a map. The objective is to better understand ground dynamics through time.

These articles extend this reading: understanding clay soils, comparing the 2026 RGA map update, then crossing hydrological and CatNat signals at territorial scale.

Understanding clay soils and shrink–swell

2026 clay shrink–swell map: geology has not changed, risk has

When physical drought and insurance signals tell different stories