Detection and monitoring of ground deformation in urban areas with advanced multi-interferogram techniques

Differential Synthetic Aperture Radar Interferometry (DInSAR) has been successfully used in recent years to detect and monitor ground deformation of large areas using space-borne radar imaging. DInSAR is based on the comparison of SAR images acquired at different times and is characterized by wide area coverage, high spatial resolution and centimetric accuracy. Its intrinsic limits (i.e. temporal and geometric decorrelations) can be overcome by PSInSAR (Permanent Scatterers InSAR) technique, developed at Politecnico di Milano and improved by its spin-off, TeleRilevamento Europa. The algorithm employed belongs to multi-interferogram techniques and takes advantage of long temporal series of SAR data to identify coherent targets (PS), about which is possible to detect millimetric displacements along the sensor-target direction, over time.
This work illustrates an application of PSInSAR for detecting and monitoring ground deformations in the town of Naro (Sicily, Italy) located at the edge of the Apennine-Maghrebian thrust belt, constituted in this area from the Gela Nappe, mostly made by Plio-Pleistocene and Miocene sediments.
The activities are carried out within the framework of the SAR.net project, promoted by the Italian National Department of Civil Protection (DPC). The PS technique was used on Naro urban area to study the spatial extension, the temporal evolution and the causes of the event occurred on February 4, 2005 which consisted in the opening of a major fracture within the urban area (Vanelle St.). At the beginning, the origin of the event was unknown. The most probable hypothesis were two: tectonic causes or a landslide.
The analysis was based on the integration of both remote sensing data and traditional in-situ investigations, in order to obtain a thorough knowledge of the event. The availability of historical datasets of SAR acquisitions has made it possible to reconstruct the history of deformations starting from 1992. Images from 1992 to 2000 (ERS1/2) and from 2003 to 2007 (RADARSAT) were used and more than 1500 and 2500 PS were respectively identified.
The results highlight the stability of the town between 1992 and 2000 and the development of remarkable deformations between 2003 and 2007. The data allows us to detect ground deformations ranging from 2 to 6 mm/year in the historic and the south-east areas of the town and to identify the buildings which moved during the event of 2005 and their deformation rate. The affected area is larger than the one detected by in-situ observations and it includes most of the town.
The integration of radar data and in-situ monitoring highlights that the event was related to tectonic causes and did not happen suddenly but it was preceded by deformation precursors starting from September 2004. After the event PSInSAR shows a decreasing in deformation rate in almost the whole area.
The results confirm PSInSAR capability for detecting and monitoring ground deformations at local scale.