Construction And Application Of Debris Flow Early-warning And Forecast Model In The Upper Min River

The upper Min River is located in the transition zone of the first and second classes in China.The terrain is complex and dangerous,the forest coverage is low,and human activities frequently interfere with the mountain environment.In addition,it is affected by geological effects such as rock layer fractures.When encountering continuous rainfall,heavy rain,etc.In the weather,severe debris flow disasters often occur.Frequent debris flow disasters bring serious economic losses and huge casualties to people in the upper Min River.Debris flow forecasting and early warning is an important disaster prevention and mitigation work,and it has become one of the hot issues in debris flow research at home and abroad.Regional debris flow early warning and forecast is often an early warning of the activity and occurrence of dozens or hundreds of debris flows in a large area.It has a wide application range and can be called a macroscopic research method for debris flow forecasting and early warning.It has greater social application value and obvious economic value than single debris flow forecast and early warning.The regional debris flow early warning forecast is usually an early warning forecast for the activity and occurrence of dozens and hundreds of debris flow gullies in a larger area,and its application range is wider.Compared with single-ditch debris flow forecast and early warning,it has greater social application value and obvious economic value.Based on this,the susceptibility zoning in the river basin was completed,and a pre-alarm model for the upper Min River was established by using the gray system theory and determining the threshold value of rainfall.The research results obtained in this paper are as follows:(1)The debris flow in the upper Min River is mainly distributed in the valleys such as Zagunao River,Zhenjiang Pass,and the mainstream of the Min River in the north of the basin.It is spatially distributed along the river in a "leaf vein" pattern.The debris flow in Songpan county,Wenchuan county,and Mao county is relatively dense.It is mainly in the young and middle ages,and the possibility of mudslide outbreaks is high.(2)Based on the analysis of the formation conditions and inducing factors of debris flow disasters in the upper Min River,nine evaluation factors such as slope,elevation,engineering rock group,vegetation coverage,river network density,landform category,rainfall,distance from fault zone,and human engineering activities were established.Evaluation index system of debris flow susceptibility in the upper Min River.The information quantity method is used to determine the information value of each rating within the evaluation factor,and the information value of each evaluation unit is assigned in ArcGIS.Finally,the susceptibility is divided based on the divided 708 sub-watersheds,and the upper Min River are divided into 5 categories.(3)The gray-GM(1,1)model is used to establish a medium-long-term forecast model for debris flow in the upper Min River.The historical debris flow disaster time in Wenchuan and Songpan county is used as a statistical sample to predict the error.The model was verified with the years when large-scale debris flows actually occurred from 2011 to 2019.The final forecast result was consistent with the actual year,and the effect was great.(4)The study area was divided into one rainfall process,and two indicators of effective cumulative rainfall in the previous period and the amount of rainfall that was stimulated on the same day were selected.The rainfall data of 45 historical disaster points and the rainfall data of 25 non-disaster points were used as samples.A Logistic regression model was established for regression in SPSS.According to statistics,the influence range of the previous rainfall was determined to be the first 7 days,so as to determine the critical line of rainfall warning,and a short-term and early warning model of debris flow was established based on the susceptibility to zoning.The model was verified with two heavy rainfall events on July 10,2013 and August 20,2019,and the results obtained were similar to those released by the Meteorological Bureau,and the effect was great.