Long Term Land Subsidence Analysis by Fusing Multi-Sensor Time Series InSAR Results

Land subsidence, as a regional environmental disaster problem, hinders social stability and sustainable development, and endangers people’s lives and property in the long term. As demonstrated by many examples, Time Series InSAR is able to monitor land subsidence using a stack of synthetic aperture radar(SAR) images.However, due to the data missing problem caused by short satellite life and data confidentiality, it is difficult to get the full monitoringresult during long time by using only a single sensor. In recent years,multi-sensor InSAR fusion methods are mostly used for cross validation and comparative analysis. Unfortunately,the datasets used in these methods is repetitive and cohesive in short time, the problem of missing data during long time has not been solved.In this paper, we will exploit three stacks of time-gapped SAR data, acquired from 2004 to 2005 by Envisat ASAR, from 2007 to 2010 by ALOS PALSAR, from 2015 to 2016 by TerraSAR-X/TanDEM-X, to analyze the long term subsidence of Shenyang city. In order to fuse the time-series InSAR results from different sensors, mathematical methods based on nonlinear curve fit will be conducted to find the best-fit model for the deformation time series.We also examine the performance of the nonlinear curve fit by calculating the root mean square error (RMSE) based on the difference between the vertical deformation time series and the best-fit model.