基于降雨时空变化的山洪预警临界雨量

为提高山洪灾害预警精度，以彭坊小流域为例，基于概率分布等控制条件生成随机雨型，探究不同前期影响雨量下随机雨型对山洪预警临界雨量的影响，进一步分析降雨空间分布对山洪预警临界雨量的影响。结果表明：在雨峰位置偏后、雨峰峰值较大的雨型集下，临界雨量波动较大，有较高的不确定性；雨型对临界雨量的影响小于前期影响雨量对临界雨量的影响，在前期土壤较干燥时，雨峰位置系数较小，临界雨量值相对较大，雨型峰值倍比较大时，临界雨量值相对较小，更容易致灾；在降雨空间分布上，当降雨集中在下游时，临界雨量值最小，越容易达到警戒流量，但降雨空间分布对临界雨量的影响小于前期影响雨量的影响，且这种影响会随前期影响雨量的增加而不断减小。

Critical rainfall for flash flood warning based on rainfall uncertainty

Mountain flood disasters generally refer to floods, landslides, and debris flows hazards caused by rainfall and snowmelt. They are characterized by their suddenness and devastating impact. Critical rainfall is a commonly used indicator for mountain flood disaster early warning, indicating that when the rainfall reaches a certain threshold within a certain period, the flow at the outlet of the watershed will exceed the warning flow, resulting in disastrous flooding. These disasters have caused severe social problems. It poses a significant threat to people''s lives and property. So optimizing mountain flood early warning and addressing the shortcomings in flood control and disaster reduction are crucial steps in social development. Critical rainfall, as the most commonly used indicator for mountain flood early warning, has been studied by numerous scholars both domestically and internationally. However, there is limited research on the comprehensive consideration of the influence of random rainfall patterns and spatial distribution under different antecedent rainfall conditions on the critical rainfall for mountain flood early warning.To address these issues, the Pengfang small watershed is focused on as the research area and random rainfall patterns are generated, based on control conditions such as probability distribution functions. Based on the historical rainfall data from the Pengfang small watershed, the optimal distribution functions for the five sets of random rainfall patterns were as follows: Normal, Normal, Exponential, Normal, Exponential and Gamma distributions. The control conditions for these distribution functions were (0.50, 0.32), (0.50, 0.42), (0.50, 0.52), (0.17, 0.32), and (0.83, 0.32) respectively. The influence of different sets of random rainfall patterns is explored, considering different antecedent rainfall conditions, on the 6-hour critical rainfall for mountain flood early warning. Furthermore, the impact of rainfall spatial distribution on the 6-hour critical rainfall is also analyzed.The results are as follows: (1) Under rainfall patterns with a later peak position and larger peak values, critical rainfall exhibits significant fluctuations, leading to diverse disaster scenarios with higher uncertainty. (2) The influence of rainfall patterns on critical rainfall is smaller compared to the influence of antecedent rainfall. When the soil is relatively dry in the antecedent period and the coefficient of the peak position of the rainfall pattern is small, the critical rainfall value is relatively large. The change in the peak position has a smaller effect on the critical rainfall when the soil is moist in the antecedent period. While the peak value ratio of the rainfall pattern is larger, the critical rainfall value is relatively small, indicating a higher likelihood of causing mountain flood disasters. (3) In terms of rainfall spatial distribution, the rainfall is concentrated downstream, and the critical rainfall value is the smallest, indicating a higher likelihood of reaching the warning flow. However, the impact of rainfall spatial distribution on the critical rainfall is less significant compared to the impact of antecedent precipitation. Furthermore, as antecedent precipitation increases, the impact of rainfall spatial distribution on the critical rainfall gradually diminishes.The findings have significant implications for improving the accuracy of mountain flood disaster early warning and enhancing the early warning system. These research results can provide a scientific basis for the optimization and improvement of the mountain flood early warning system, aiming to better protect people''s lives and properties and effectively respond to the challenges posed by mountain flood disasters.