Damage detection for the 2004 Niigata-ken Chuetsu Earthquake using satellite SAR

Abstract

The building damage detection technique which we have developed has been successfully applied to past events such as the earthquakes in Kobe in 1995, India in 2001, and Bam in 2003 by using the compound index, z-value, a value derived from the correlation and difference in intensities between pre- and post-event SAR images. This technique was applied to the areas affected in the Niigata-ken Chuetsu earthquake of October 23, 2004 by using one pair of Radarsat images taken before and after the earthquake. However, it was not possible to identify any significant distribution of damaged buildings. In our study, we examined the reasons for that and proposed a new technique that uses two pairs (pre-seismic and co-seismic) of SAR images to identify smaller building damage ratios in less densely built-up areas as compared to the previous technique. The main idea is to minimize the effects of signal noise and temporal changes of the earth's surface on building damage estimation by calculating the difference values of the two pre-event images and one postevent image. From a macroscopic point of view, the distributions of both difference values of the z-value and the correlation coefficient in built-up areas in Niigata-ken Chuetsu region were in good agreement with damage reported in survey reports. In former Yamakoshi village, located in the highlands, we could also identify large-scale landslides with accuracy as good as that of interpretation from aerial photos.