Spatial improvement of information on population distribution using GIS approaches: An input for tsunami people exposure assessment

Kurzfassung

The information about population distribution is a crucial component in disaster risk management. Every disaster such as flood, drought, volcanic eruption, storm, earthquake and tsunamis implies threats to people with respect to loss of live, injury, and suffer. Therefore, the information on detailed population distribution in the disaster or hazard zone is important in order to mitigate the impact of natural disasters. Moreover, accurate information on people exposure will help the government to improve the evacuation planning and to decrease the amount of people at risk. The available information on population distribution is mostly based on statistical data related to administrative boundaries, e.g. village, municipal, district, province, or national level. However, within the border of administra-tive boundaries the real people distribution is rather heterogeneous. Hence this research focuses on the improve-ment of the available data on population distribution for the area along the west coast of Sumatera, south coast of Java, and Bali. The results are used as an input for the tsunami risk assessment in the framework of the GITEWS (German-Indonesian Tsunami Early Warning System) project. To reach the aim of a high resolution of people distribution for this area a combination of the amount of people at village level (Census Data) and land use maps is used. This approach is performed by GIS and remote sensing techniques and is based on an analysis of people activities in a certain land use. In order to assign appropriate weighting factors to distribute the amount of people per village to diverse land use classes, information from Census and PODES (potential of village) data were analysed. The results of this approach are detailed population distribution maps for day and night time along the regarded southwest coast of Sumatera, Java, and Bali. By comparing the results with other available data on people distribution, the spatial improvement could be shown. Moreover, in order to estimate the accuracy of this approach an error and sensitivity analysis was performed.