Extracting damages caused by the 2008 Ms 8.0 Wenchuan earthquake from SAR remote sensing data

摘要

Bad weather conditions usually limit the acquisition of optical remote sensing images, while all day and all weather synthetic aperture radar (SAR) shows the ability of providing timely remote sensing data for emergency response and rescue works after earthquake. Because SAR is sensitive to the surface changes caused by earthquake, the modified electromagnetic behaviour by geological disasters and the collapse of buildings can be recorded in SAR images as backscattering intensity changes. Absolute radiometric calibration was performed to SAR products to derive backscattering coefficient sigma nought from image digital number (DN). Based on change detection methods, Advanced Land Observing Satellite (ALOS) Phased Array type L-band Synthetic Aperture Radar (PALSAR) data and TerraSAR-X data acquired for the Ms 8.0 Wenchuan earthquake were used to extract earthquake damage information. This study revealed that landslides showed stronger backscattering and barrier lakes showed lower backscattering in post-earthquake 10 m ALOS PALSAR images comparing to pre-earthquake, and collapsed buildings showed lower backscattering compared to un-collapsed buildings in 1 m TerraSAR-X image. Results showed that SAR data with different spatial resolutions are useful for different earthquake damage information extraction: medium spatial resolution SAR data, e.g. 10 m ALOS PALSAR data, were efficient for secondary geological disaster extraction; high-resolution SAR data, e.g. 1 m TerraSAR-X data, with the help of ancillary G1S data or high-resolution optical data, could be used to extract building collapse information in urban areas. This study indicates that SAR remote sensing data can provide earthquake damage information at early emergency stage and assist the field surveying, further damage assessment and post-earthquake reconstruction.