Kostyuchenko Y. Integration of data from satellite observations and mathematical modeling in the tasks of assessing geoecological risks and safety.

Research is aimed to the development of methods for assessing the risks of disasters using the satellite monitoring. Theoretical and methodological basis for the application of satellite observation, analysis of spatially-distributed data and energy-mass exchange modeling of natural systems was proposed in the work. The multi-model approach for analyzing and forecasting the multi-scale hazards is proposed for optimization of decision-making in the field of sustainable nature management. Using the proposed nonlinear algorithms, all available temporal and spatial distributions of natural disasters are analyzed. On this basis, existing hypotheses have been confirmed and new trends in the natural disaster development have been discovered. New integrated models of energy-mass exchange in natural systems and models of formation of a useful signal of terrestrial objects and systems have been developed. They are oriented to application of satellite methods and analysis of spatially distributed data. Using well-known physical and statistical models, algorithms for estimating the parameters of the geological environment and natural systems are developed and adapted. In particular, models of snow cover, dynamics of soil cover moisture, and evapotranspiration have been developed. Approaches to assessing the long-term effects of changes are also developed. In particular, the impact of climate change to the escalation of natural disasters for the territory of Ukraine is assessed. In particular, a set of methods for nonlinear multidimensional multivariate analysis for the assessment of risk measures in multi-component heterogeneous systems is proposed and applied in the work. A number of methods of data analysis have been proposed and developed. In particular, the method of non-linear spatial-temporal regularization for the analysis of ground-based measurements to obtain regularized distributions of regional indicators; method of integrating data from local meteorological observations into global climate models; a method for evaluating complex measures of risk based on the results of a joint analysis of multivariate values based on nonparametric kernel copula were developed. Problem-oriented methods for coupling the satellite data are developed and improved. Methods of data integration for the analysis of safety indicators are justified, developed, regionally adapted and applied. In particular, the decoding indicators for the assessment and prediction of the local flood potential have been determined, the necessary size and qualitative characteristics of verification data have been evaluated, a comprehensive scientific rationale for optimizing the satellite methods, an integrated approach to the use of spatially distributed data, remote sensing systems, data from landfills , and models of energy-mass exchange in geo-systems for the assessment of the disaster threats have been proposed in its final form. According to the proposed methods, the necessary calculations were made; distributions of quantitative indicators of risks of certain hazardous processes in the regions of Ukraine were received; uncertainties were estimated and the reliability of the calculations is shown. Methods of using remote sensing in disaster monitoring and control systems and risk assessment are proposed. Thus, the scientific and methodological basis for the use of satellite observation and geospatial analysis in long-term hazard prediction problems associated with the emergency situations has been developed, and a number of applied methods and algorithms have been proposed.