Due to the diversity and complexity of mountain flood disasters, the research in key technical fields is still in the initial stage and cannot meet the actual needs of current mountain flood disasters. How to accurately forecast local heavy rainfall, develop a mountain flood monitoring technical system and a localized flood disaster prediction model, refine the dynamic warning indicators of mountain flood disasters, realize the transition from post-disaster relief to pre-disaster prevention, build a risk assessment platform, construct a mountain flood disaster prevention model, and improve mountain flood disaster warning and risk prevention capabilities are the frontiers and hot issues of current mountain flood disaster research, and the practical problems that must be faced in the construction of county-level non-engineering measures for mountain flood disaster prevention which strongly implemented by the country. Focusing on the above issues, the following three key research objectives are summed up. The first one is clarifying the dynamic process of mountain flood disasters, establishing a local rainstorm forecast model and a mountain flood dynamic simulation model. The second one is developing and constructing a multi-element monitoring technology and system for mountain flood, realizing real-time dynamic transmission and release of mountain flood disaster warning and risk assessment information, solving the problem of low accuracy of mountain flood peak flow forecast, short warning period and short-term emergency response. The last one is building a multi-level and multi-target dynamic early-warning and risk assessment platform for mountain flood disasters, refining the disaster prevention and mitigation mode of mountain flood disasters, and demonstrating and promoting that to the public. The following innovations will be achieved through the research. Starting from revealing the formation mechanism of small and medium-scale weather systems that induce flood, short-term near-storm forecasting techniques in mountainous areas is developed, the prediction accuracy of floods and torrential rains in mountainous areas is improved, and the warning period of mountain flood disasters is effectively extended. By researching multi-source remote sensing, radar, micro-nano sensing, intelligent recognition and other mountain rainwater monitoring data acquisition technology and integration of adaptive, practical and stable traditional technologies, a multi-element three-dimensional monitoring technology system for air and land integration is built, technological innovation and extend the mountain flood warning period is realized. Through the analysis of the response process of flood catastrophe process in mountainous areas, the formation process of rainstorm mountain floods and the dynamic mechanism of disasters are clarified, and the dynamic simulation model of mountain flood process based on rainfall prediction and dynamic monitoring of soil water content is constructed to improve the accuracy and reliability of mountain flood warning. Taking the migration process of mountain floods as the entry point, a multi-index warning model for mountain flood disasters is established, real-time dynamic grading warning technology for mountain flood disasters is developed, the warning period of disasters is effectively prolonged, and the aging time for emergency rescue responseis improved.
