NATURAL DISASTER RISK ASSESSMENT BASED ON THE ENSEMBLE PROCESSING AND TECHNOLOGY OF HETEROGENEOUS GEOSPATIAL DATA FUSION

The natural disaster risk assessment problem is stated which is based on heterogeneous geospatial data, namely, satellite data, ground-based observations and simulation data. A method for solving the problem is proposed. The heart of the method is the ensemble data processing and technology of the heterogeneous data fusion with respect to the unknown disaster probability density estimation based on a sample of data. This probability density depends on the finite number of parameters. The sources of heterogeneous geospatial data are analyzed which are used in the developed operational flooding risk mapping service for the territory of Namibia. We consider a conceptual sketch of the probability density estimation system to determine the flooding risk for the territory of Namibia. It is constructed in accordance with the method proposed. To continue the investigation according to the international pilot project <<Sensor Web Project for Flood Monitoring in Namibia>>, the staff of SRI NASU-NSAU will elaborate an operational flood risk mapping service with the use of modern Internet and GIS technologies. The operational service will satisfy the international standards of Open Geospatial Consortium (OGC) to provide geospatial information.