Validation and intercomparison of Persistent Scatterers Interferometry: PSIC4 project results

Abstract

This article presents the main results of the Persistent Scatterer Interferometry Codes Cross Comparison and Certification for long term differential interferometry (PSIC4) project. The project was based on the validation of the PSI (Persistent Scatterer Interferometry) data with respect to levelling data on a subsiding mining area near Gardanne, in the South of France. Eight PSI participant teams processed the SAR data without any a priori information, as a blind test. Intercomparison of the different teams' results was then carried out in order to assess any similarities and discrepancies. The subsidence velocity intercomparison results obtained from the PSI data showed a standard deviation between 0.6 and 1.9 mm/year between the teams. The velocity validation against rates measured on the ground showed a standard deviation between 5 and 7 mm/year. A comparison of the PSI time series and levelling time series shows that if the displacement is larger than about 2 cm in between two consecutive SAR-images, PS-InSAR starts to seriously deviate from the levelling time series. Non-linear deformation rates up to several cm/year appear to be the main reason for these reduced performances, as no prior information was used to adjust the processing parameters. Under such testing conditions and without good ground-truth information, the phase-unwrapping errors for this type of work are a major issue. This point illustrates the importance of having ground truth information and a strong interaction with the end-user of the data, in order to properly understand the type and speed of the deformation that is to be measured, and thus determine the applicability of the technique.