| Influenced by the change of climate and underlying surface,flood disasters occur frequently in China,which seriously threaten the safety of people’s lives and property,and restrict the region sustainable development.Due to hilly area’s weak water storage capacity,strong sudden occurrence of mountain torrent and complex influencing factors,it is difficult to accurately predict and warn mountain torrent disaster.Therefore,it has become the top priority of mountain torrent disaster prevention and control.As an important index of mountain torrent early warning,the determination of critical rainfall in the past only consider soil moisture and early warning period,less or roughly consider the influence of rainfall time distribution,spatial distribution and other factors,the warning accuracy need to be further improved.Therefore,the critical rainfall set of disaster prevention objects under different scenarios was determined by comprehensively considering various uncertain factors of the uneven spatial and temporal distribution of rainfall,and the three-level transfer critical rainfall early warning model was constructed.Taking Luanchuan County as an application example,the mountain torrent warning model was studied based on the spatial and temporal uncertainty of rainfall.The main contents and achievements are as follows:(1)HEC-HMS model for small watershed was established.ArcGIS was used to process the digital elevation,soil and land use data of the basin,extract the river system and divide the molecular watershed.The SCS curve method,SCS unit line method,water retreat curve method and Mastingen method were set up to form the model of runoff generation and confluence scheme.Taking the minimum peak weighted root mean square error as the objective function,10 typical floods were selected for parameter calibration,and 4 floods were used for model verification.The simulation accuracy reached the level of class B,indicating that the HEC-HMS model has good applicability in the research area.(2)A method to determine the rainfall pattern set under the uncertainty of rainfall time and space was proposed.According to the Gini coefficient and the position coefficient of peak rainfall intensity,nine types of characteristic rain patterns were generalized,including the front uniform type and the back concentrated type,etc.Based on the marginal distribution function of total rainfall and peak rainfall intensity,the optimal Copula joint distribution function of each station were established,and the peak-volume relationship of risk combinations of five return periods of total rainfall and seven conditional probabilities of peak rainfall intensity were calculated.The spatial distribution of measured rainfall was analyzed and the spatial distribution of five rainfall scenarios were designed according to the shift of rainfall gravity center from upstream to downstream.On the basis of the above,the rain pattern set based on the spatiotemporal uncertainty of rainfall was constructed.(3)The three-level critical rainfall warning mode of mountain flood disaster transfer was constructed.On the basis of the above research,the critical rainfall set under different scenarios was determined by combining the factors of earlier influence rainfall and early warning period,etc.The results show that the more backward and concentrated the rainfall time distribution was,the closer the rainfall center was to the downstream,the smaller the critical rainfall was.The critical rainfall boxplots were drawn,according to the characteristics of boxplot and critical rainfall characteristics of early warning indicators,the first quartile,median,and the third quartile as Ⅰ,Ⅱ,Ⅲ level transfer critical rainfall warning value,respectively.Then the critical rainfall early warning model of three-level transfer of mountain flood disaster was established,and the early warning model was tested by four measured rainwater processes,the results showed that the effect of the early warning model was good,it can provide reference for mountain torrent warning. |
PDF Full Text Request