Development of Risk Assessment and Early Damaged Area Estimation Methods for Large Scale Earthquake Disaster Using DMSP/OLS Night-time Imagery

点击翻译按钮获取中文摘要

In this research, we developed the following methods using time-series DMSP/OLS imagery data which capture night-time city light to monitor the changes in light intensity and geo-location patterns as an index of regional socio-economic activities before and after the earthquake. The spatial distribution patterns in city lights enable us to know the disaster risk before the event and the changes in city light patterns the potential damage area just after the event. These methods would be especially useful in the developing countries which do not have sophisticated disaster information system nor high resolution socio-economic statistics.1)Web-based Early Damaged Area Estimation System(EDES) was developed. The system worked on 168 earthquakes with M6.0 in the world in 2004. The results were published on the website, http://www.edm.bosai.go.jp/default.htm.2)A night-time city light mapping method whose procedure is suited to automation and is free from saturation caused by sensor gain limitation was proposed.3)A methodology to estimate spatial distribution of damage inventory such as population or buildings using night-time city light intensity distribution was proposed by examining the relationship between the spatial distribution of night-time city light intensity and that of population or buildings. It was clarified that the method has enough accuracy to know the big picture of possible damage situation and the distribution heavily damaged area of large scale earthquake by applying this method to Tokai earthquake damage assessment.4)A method for estimating the quantity of buildings with different seismic vulnerability according to GESI method using a high resolution satellite imagery data and geo-referenced digital video imagery data was developed. This method is useful to know detailed damage potential information for local seismic risk assessment