Natural disaster risk assessment based on the ensemble processing and technology of heterogeneous geospatial data fusion

1Zyelyk, Ya.I, 1Kussul, NM, 1Skakun, SV, 2Shelestov, AYu.
1Space Research Institute of the National Academy of Sciences of Ukraine and the State Space Agency of Ukraine, Kyiv, Ukraine
2Space Research Institute of the National Academy of Science of Ukraine and the State Space Agency of Ukraine, Kyiv; National Technical University of Ukraine «Kyiv Polytechnic Institute», Kyiv, Ukraine
Kosm. nauka tehnol. 2011, 17 ;(1):60-64
https://doi.org/10.15407/knit2011.01.060
Publication Language: Russian
Abstract: 
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.
Keywords: ensemble data processing, geospatial data, natural disaster
References: 
1.  Bishop C. M. Pattern recognition and machine learning, 738 p. (Springer, New York, 2006).
2.  Haykin S. Neural networks. A comprehensive foundation, 768 p. (Prentice Hall, New Jersey, 1994).
3.  Jaakkola T. Course materials for 6.867 machine learning, Fall 2006. MIT OpenCourseWare, Massachusetts Institute of Technology, 10 p. (2006). Retrieved from  http://ocw.mit.edu/
4. Jonkman S. N., van Gelder P. H. A. J. M., Vrijling J. K. An overview of quantitative risk measures for loss of life and economic damage. J. Hazardous Materials, A99, 1—30 (2003).
https://doi.org/10.1016/S0304-3894(02)00283-2
5. Kussul N., Shelestov A., Skakun S. Grid and sensor web technologies for environmental monitoring. Earth Sci. Informatics, 2 (1-2), 37—51 (2009).
https://doi.org/10.1007/s12145-009-0024-9
6. Mitchell H. B. Multi-sensor data fusion — An introduction, 282 p. (Springer-Verlag, Berlin, 2007).

7. Vapnik V. Statistical learning theory, 740 p. (Wiley, New York, 1998).