Monitoring land surface deformation using persistent scatterers interferometric synthetic aperture radar technique

Abstract

Land subsidence is one of the major hazards occurring globally due to several reasons including natural and human activities. The effect of land subsidence depends on the extent and severity. The consequences of this hazard can be seen in many forms including damaged of infrastructures and loss of human lives. Although land subsidence is a global problem, but it is very common in urban and sub urban areas especially in rapidly developing countries. This problem needs to be monitored effectively. Several techniques such as land surveying, aerial photogrammetry and Global Positioning System (GPS) can be used to monitor or detect the subsidence effectively but these techniques are mostly expensive and time consuming especially for large area. In recent decades, Interferometric Synthetic Aperture Radar (InSAR) technique has been used widely for the monitoring of land subsidence successfully although this technique has several limitations due to temporal decorrelation, atmospheric effects and so on. However, the uncertainties related to InSAR technique have been reduced significantly with the recent Persistent Scatterers Interferometric Synthetic Aperture Radar (PSInSAR) technique which utilized a stack of interferograms generated from several radar images to estimate deformation by finding a bunch of stable points. This study investigates the surface deformation focusing on Kuala Lumpur, a rapidly growing city and Selangor using PSInSAR technique with a set of ALOS PALSAR images from 2007 to 2011. The research methodology consists of several steps of image processing that incudes i) generation of Differential Interferometric Synthetic Aperture Radar (DInSAR), ii) selection of Persistent Scatterers (PS) points, iii) removal of noise, iv) optimization of PS point selection, and v) generation of time series deformation map. However, special consideration was given to optimize the PS selection process using two master images. Results indicate a complete variation of mean line-of-sight (LOS) velocities over the study area. Stable areas (mean LOS=1.1 mm/year) were mostly found in the urban center of Kuala Lumpur, while medium rate of LOS (from 20 mm/year to 30 mm/year) was observed in the south west area in Kuala Langat and Sepang districts. The infrastructures in Kuala Lumpur are mostly stable except in Kuala Lumpur International Airport (KLIA) where a significant subsidence was detected (28.7 mm/year). Meanwhile, other parts of the study area such as Hulu Langat, Petaling Jaya and Klang districts show a very low and non-continuous movement (LOS < 20 mm/year), although comparatively higher subsidence rate (28 mm/year) was detected in the mining area. As conclusion, PSInSAR technique has a potential to monitor subsidence in urban and sub urban areas, but optimization of PS selection processing is necessary in order to reduce the noise and get better estimation accuracy.