Monitoring land subsidence in Yunlin County using Sentinel-1A SAR by SBAS

• Main Authors: Yang, Yi-Jie, 楊苡絜
• Other Authors: Hwang, Cheinway
• Format: Others
• Language: en_US
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/7m2brb

碩士 === 國立交通大學 === 土木工程系所 === 106 === Yunlin County, a coastal county in the Chuoshui River Alluvial Fan (CRAF), has long suffered from severe land subsidence, which poses a threat to the operation of the Taiwan High Speed Railway (THSR). Effective monitoring of Yunlin land subsidence is an important issue of hazard prevention. Interferometric Synthetic Aperture Radar (InSAR) has been widely used for monitoring land subsidence in Taiwan because of its high spatial resolution and wide swath. The orbital errors in InSAR technique cause systematic errors in results, this study estimates the orbital errors by comparing the differences between the InSAR-derived results and the GPSderived results. In this study, we used 35 Sentinel-1A images (22 images along the ascending track and 13 images along the descending track) from April 2016 to April 2017 to determine land subsidence by the method of Small BAseline Subset (SBAS). The orbital errors are then estimated to improve the overall quality of the Sentinel-1A-derived SBAS results. The Root Mean Square (RMS) differences between the Sentinel-1A results and GPS results in the one-year case are 7.6 mm/yr and 33.6 mm/yr using data from the ascending and descending tracks, respectively. And the RMS differences decrease to 6.2 mm/yr and 8.0 mm/yr after moving the effect of the orbital errors. The adjusted Sentinel-1A-derived results are converted to the rates along the vertical direction and compared with the leveling-derived results. This study provides wide-range and high-precision land subsidence rates compatible with those from GPS and leveling (within one mm/yr) in Yunlin.