Search

In the last few years, the success of Persistent Scatterers Interferometry (PSI) techniques has largely increased because of the ever enhanced availability of space born radar data. Recently, various authors worked about using PSI techniques in order to better understand slope movements in general and in particular landslides kinematic. PS-time approach represents a new method for the automatic classification of PSI time series based on a conditional sequence of statistical tests. Time series are classified into distinctive predefined target trends, such as uncorrelated, linear, quadratic, bilinear and discontinuous, that describe different styles of ground deformations, so this new approach can be successfully used to improve the radar interpretation of land-slide movements. In this paper we present the results of the application of PS-time to two unstable areas in Northern Apennines of Italy (San Benedetto Val di Sambro, Province of Bologna; Vairo, Province of Parma). Results show that the time series analysis can greatly improve our understanding of the deformation phenomena, and pro-vide useful information in addition to the conventional analysis based on the mean velocity alone.

We present an automated time series analysis of InSAR data for the characterization of ground subsidence induced by longwall mining at Metropolitan mine (New South Wales, Australia). The dataset derives from SqueeSAR™ processing of two Envisat radar data stacks of 44 images acquired from the ascending orbit and 43 from a descending orbit, acquired on a 35-day repeat interval in the period June 2006 to September 2010. Automated time series classification was carried out with PSTime, a specifically designed software that employs a sequence of statistical tests to classify the time series into the six different classes: uncorrelated, linear, quadratic, bilinear, discontinuous without constant velocity and discontinuous with change in velocity. Results highlight a cluster of bilinear trends with acceleration at the front of longwall panel progression and a cluster of bilinear trends with deceleration at the back. Linear trends are found at the centre of the subsidence bowl while outside most trends are uncorrelated. This picture is consistent with the evidence of acceleration in deformation trend when mining is approaching and a deceleration after extraction is completed and an essentially constant while mining takes place just underneath a specific point. Thus, time series analysis proved to be valuable for subsidence dynamics characterization, constraining in space and time the patterns of deformation trend in mining applications.