The Influence of External Digital Elevation Models on PS-InSAR and SBAS Results: Implications for the Analysis of Deformation Signals Caused by Slow Moving Landslides in the Northern Apennines (Italy)

Advanced interferometric synthetic aperture radar (InSAR) postprocessing, like persistent scatterer InSAR (PS-InSAR), offers the possibility to investigate slow moving landslides, where standard interferometry is problematic. These advanced algorithms involve the analysis of a series of SAR acquisitions in both time and space. One input that requires particular attention for landslide applications is the external digital elevation model (DEM) that is used to correct the interferograms for the topographic phase term. When multiple elevation data sets are available for a given study area, it is difficult to decide which one should be used. In this paper, we test the sensitivity of PS-InSAR/Small Baseline Subset (SBAS) results to different DEMs. The study area is located in the Northern Apennines of Italy, where chaotic clay shales and fine-grained flysch host slow-moving earth flows and ancient rock slides.
$C$
-band (Envisat) and
$X$
-band (Cosmo-SkyMed) data are processed with different DEMs. We describe a simple framework to statistically analyze the influence of these models on the final PS-InSAR/SBAS results. We find that individual interferograms do not vary much depending on the DEM, while the results from PS-InSAR and SBAS analysis do vary. This is likely caused by the way the DEM error is estimated. We find also that the quality of the DEM is more important than the resolution and that
$X$
-band InSAR data are more sensitive to the choice of the DEM than
$C$
-band. The significance of the results is discussed with reference to two landslide areas.