|
PSInSAR method is relatively new method and its fields of application are still in progress. In our work we tried to evaluate the presence of systematic errors in PSInSAR data, information on which could be used as important information when estimating actual insitu surface deformations as a consequence of present geological processes. Comparison between two sets of surface deformation time series on two completely isolated areas represents one possible approach in determination of error integrated in the PSInSAR data. Possible correlations of PS variations between separate locations could indicate possible algorithmic (systematic) errors that occur at processing of original data.
The applicability of the Permanent Scatterers Synthetic Aperture Radar Interferometry (PSInSAR) technique for detecting and monitoring ground displacements and over periods of time was tested in the two Slovenian regions (NW part of Slovenia and central part of Slovenia), which could be representative of similar tectonically and seismic contexts . The study area, which extends for almost 2700 km2, is characterized by a complex geological and structural setting and the presence of different sedimentary formations. In the NW part of Slovenia 16304 permanent scatterers were detected and only 10 % of them satisfied our conditions while more than 25.000 PS were identified in Ljubljana basin and about 14.000 PS were useful. PS data were derived by Synthetic Aperture Radar (SAR) images acquired from April (June) 1992 to December 2000 by the European Remote Sensing satellites (ERS) in both cases.
Considering the fact that time series data were not acquired at the same time the analysis was focused on the comparison of temporary closest two consecutive acquisitions and the relative difference that occurred between the two acquisitions. First analyses indications on some resemblance (correlation) of two series in years 1996 and 1999, which have been observed and some analyses are still in progress. During these two periods there are similar changes in measurements notable in both areas, where a portion of PS is "subsided" and a portion is "uplifted". During the rest of the observation period (1992 - 2000) indications of similarity are scarce. Since the seismic activity recorded for the whole observation period was carefully studied, the tectonic influence on atypical displacements is eliminated. Results obviously indicate some sort of systematic error PSInSAR data.
|
