Study On Rainfall Early Warning Model Of Mountain Flash Flood Based On Characteristic Rainfall Patterns

At present,the rainfall warning index is an important indicator in the early warning way of mountain flash flood in small watersheds of the hilly area in China.The traditional rainfall warning index has not considered the randomness and diversity of rainfall,and only the regional design rainfall pattern represents the rainfall process,and then the unique rainfall warning index value is determined.However,when the flood is calculated by rainfall,the different rainfall processes will lead to different flood processes,and even the peak flow will have a large difference.Therefore,it will indirectly affect the accuracy of the rainfall warning indicator and the degree of coincidence with the actual situation.Aiming at the problem of lower coincidence between the design rainfall pattern and the actual diversity rainfall process,and then affecting the accuracy of calculating rainfall early warning indices,this paper probes into the characteristics of rainfall distribution over time,and proposes a characteristic rainfall pattern determination method based on joint control of rainfall type feature parameters to determine the rainfall pattern set which represents the precipitation features in the regions.Finally,the rainfall characteristics are added into the flash flood warning model to construct a real-time dynamic rainfall warning method based on characteristic rainfall pattern,which provides reference value for improving the accuracy of rainfall warning indices,and further perfect the flash flood warning model.The main contents and research results of the paper are as follows:(1)The characteristic rainfall pattern determination method based on joint control of rainfall type feature parameters.According to the characteristics of rainfall distribution over time,the quantitative rainfall type characteristic parameters are proposed,and the characteristics of “precipitation intensity change trend” and “rain peak position” are combined to determine the diversity rainfall pattern that meets the regional rainfall characteristics.(2)Rainfall early warning index calculation and rationality analysis.The HECHMS hydrological model is used to simulate the runoff and concentration process to ensure accurate peak flow calculation.Considering the distribution of different rainfall patterns and the soil moisture in the early stage,the hydrological-hydraulics method is used to calculate the rainfall early warning indices for different warning time of warning objects.According to the disaster information,the time-interval characteristic rainfall assessment method is used to analyze the reasonableness of the rainfall warning indicators to improve the reliability of the rainfall warning indices.(3)Research on early warning way of mountain flash flood based on characteristic rainfall pattern.The cumulative rainfall-duration curve of different rainfall distributions has different variation characteristics.This paper proposes a rainfall warning model based on characteristic rainfall type.The real-time rainfall information is identified by the characteristic rainfall type,and the rainfall process that may occur in the current and future short-duration time is predicted,and the rainfall warning indices are matched to realize the real-time dynamic rainfall warning way.(4)Case application research.Taking the small watershed in Xinxian County of Henan Province as an example,the diversity representative rainfall patterns with small watershed rainfall characteristics were determined.Based on this,the rainfall warning indices of each warning object were analyzed and calculated.And using the disaster floods to make the reasonable check of rainfall warning indices,the results are reasonable.The application of the rainfall flood warning model based on the characteristic rainfall pattern is applied.The results showed that the rainfall characteristics are added to the mountain flash flood warning model,which can further improve the agreement between the theoretical warning and forecast information and the actual situation.It provides reference value for improving the early warning and forecast system of mountain flash flood.