Monitoring the subsidence in the plain of south Thessaly with SAR inteferometry

Περίληψη

Over-exploitation of groundwater has led to extensive land subsidence in the region of South Thessaly, causing considerable damages in public and private property. The above phenomenon, was a result of the extensive cultivation due to the increase of water demand in South Thessaly Plain. Satellite Interferometric Synthetic Aperture Radar (InSAR) is an ideal technique for measuring the spatial extent and magnitude of surface deformation associated with fluid extraction. It is often less expensive than obtaining sparse point measurements from labor-intensive spirit-leveling and Global Positioning System (GPS) surveys and can provide millions of data points in a region about 10,000 square kilometers. It is also able to detect changes with very high level accuracy, even in the scale of 1mm. Scenes from two different datasets and two different satellites were utilised in order to investigate the study area. 24 Advanced SAR (ASAR) scenes from Envisat satellite in Descending orbit and 14 scenes from ALOSPALSAR satellite in both single and dual polarization and Ascending orbit as well, covering a total time span of 7 years (2003-2010) for ASAR/Envisat ones and 4 years (2007-2011) for ALOS-PALSAR were used, respectively. The software we used for processing the imagery was GAMMA Interferometric Software. The method that was primarily selected to reveal the deformation patterns in the area and the evolution of it in the temporal dimension was PSI (Persistent Scatterers Interferometry) provided by IPTA(Interferometric Point Target Analysis) package on GAMMA s/w. Due to the high and rapid rates of deformation in the area, IPTA failed to deliver sufficient and accurate results. Since PSI technique was ruled out, a hybrid INSAR methodology using elements of conventional DINSAR, short baseline interferometry approaches and PSI was implemented. With this technique, starting from the multireference stack of unwrapped phases from conventional DINSAR, we derived a single reference time series using Singular Value Decomposition (SVD) to obtain the leastsquares solution for the phase timeseries. It is the most suitable and useful method to evaluate vertical land surface changes, despite the fact that decorrelation posed as a significant limitation. It is an advanced DInSAR technique which provide information not only on the spatial distribution of deformation, but also the temporal evolution of ground deformation over time, in pixel basis allowing us to produce time series like PSI. This technique also helps us overcome several other problems caused by baseline limitations and also when an area has not cover from wide number of images. Thus, it allows us to obtain interferograms and organise them on consecutive time intervals. The most important condition that has to be fulfilled in order to form interferograms and generally extract information from SAR systems is coherence between images. In general, interferometry is successful and accurate only where higher values of coherence are present. Since Thessaly’s plain surface is covered by croplands spatial decoralation occurred over the largest part of therefore only urban areas are correlated. Another reason that justifies the implementation of this particular technique is that the data was also temporally decorelated since the imagery acquired is relatively small covering a large time span, as already mentioned. With these two datasets we achieved to reveal the deformation patterns caused by subsidence in the broader area of the plain and in particular in the villages that located there. The results will be validated with tectonic, geohydrologic and meteo data.