Interferometric synthetic aperture radar (InSAR) compares the phase of two or more coregistered complex SAR observations of the same area. The phase difference records a change in the radar signal’s travel path. With known acquisition geometry, this permits the construction of digital elevation models or the measurement of surface displacement between acquisitions.1

Measurement geometry

An interferogram wraps phase into repeating cycles, often displayed as coloured fringes. One full cycle corresponds to a path-length change related to the radar wavelength. The measurement is projected along the sensor’s line of sight, so it is not automatically vertical displacement. A single look direction cannot recover a complete three-dimensional motion vector. Combining ascending and descending observations or other geodetic measurements can help separate directional components, subject to their geometry.

Topographic InSAR uses the spatial separation between observations to infer elevation. Differential InSAR instead models and removes topographic phase, commonly with an external digital elevation model, so that change between acquisition dates can be studied.2 ESA describes the method as capable of measuring centimetre-scale deformation under suitable conditions.3 This scale is neither a universal accuracy nor a guarantee for each pixel.

Processing and uncertainty

A practical workflow uses precise orbit data, selects a compatible image pair or time series, coregisters the complex pixels, forms an interferogram and estimates coherence. Sentinel-1 TOPS processing also requires careful handling of bursts. The chain can then include topographic-phase removal, filtering, phase unwrapping, terrain correction and geocoding.2 Each stage affects the result and should be recorded with the input product versions.

Residual phase is not deformation alone. It can contain orbit error, delay caused by atmospheric water vapour, digital-elevation-model error, thermal noise and processing artefacts. Temporal decorrelation occurs when scatterers change between acquisitions; geometric decorrelation occurs when the viewing geometries are too different. Vegetation, snow, water and disturbed ground can therefore lose coherence. Phase unwrapping may fail across low-coherence areas or where phase changes steeply. Time-series methods add observations and can separate some temporal signals, but they do not remove every bias.

Acquisition cadence is another constraint. Sentinel-1’s two-satellite design provides a nominal six-day repeat at the equator.4 The interval between usable interferometric observations at a particular site also depends on acquisition planning, compatible orbit and viewing geometry, mission health and surface coherence.

UK research and application

The UK Centre for the Observation and Modelling of Earthquakes, Volcanoes and Tectonics uses InSAR to map topography and surface change associated with earthquakes and volcanoes.1 The British Geological Survey is developing processing that can automatically download and update satellite data, classify ground motion and identify anomalous behaviour.5 These programmes establish UK research and processing capability. They should not be described as a universal operational warning service: alerting depends on acquisition latency, processing, interpretation and integration with other observations.

A defensible deformation product should state the wavelength, orbit direction, acquisition dates, perpendicular and temporal baselines, reference area, coherence threshold, atmospheric treatment, digital elevation model and uncertainty method. Without that provenance, a smooth displacement map can conceal unresolved ambiguity.

References

Footnotes

  1. NERC Centre for the Observation and Modelling of Earthquakes, Volcanoes and Tectonics, InSAR. ↩ ↩2

  2. European Space Agency, Sentinel-1 TOPS interferometry tutorial. ↩ ↩2

  3. European Space Agency, InSAR Principles: Guidelines for SAR Interferometry Processing and Interpretation. ↩

  4. European Space Agency, Sentinel-1 facts and figures. ↩

  5. British Geological Survey, InSAR research. ↩