Generation of deformation maps at low resolution using differential interferometric SAR data

In this paper, an advanced technique for the generation of deformation maps using SAR data is presented. The input data is a set of low resolution Differential Interferograms (multi-looked data) and their associated coherence images. An important advantage of this algorithm is that it can work with a reduced number of SAR images with a diversity of spatial baselines. The algorithm takes advantage of those pixels presenting a good coherence level in the whole set of interferograms, avoiding the rest affected by temporal decorrelation. All pixels accomplishing the selection criteria are related using a Delaunay triangulation. The subsidence velocity map over the scene is obtained adjusting an interferometric phase model, which also considers the error on the DEM used to remove the topography from the interferogram set, to the pixel phase increments. If the density of quality pixels is high enough over the scene, an interpolation of the areas with no information can be performed to obtain a complete deformation map of the zone. Otherwise, the information can be presented only on those zones with enough pixel density. This algorithm has been tested with ERS data from an area of Catalonia (Spain) and validated with precise levelling measurements.