Prediction Runout Volume Of Gully Debris Flow Based On GIS And Flow-R

After the 8.0 magnitude earthquake in wenchuan in 2008,a large number of landslide disasters occurred in the earthquake area,resulting in the increase of loose debris in the basin.Subsequently,affected by continuous heavy rainfall,the amount of water collected in the region increased sharply,triggering a massive mass debris flow disaster.In this paper,Longxi River Basin in Longchi Town,Dujiangyan is taken as the study area,uses numerical simulation methods to predict the debris flow deposition area,and establishes a gully-type debris flow runout volume prediction model based on GIS and Flow-R,in order to provide reference for early debris flow identification and risk analysis,and reduce casualties and economic losses.Combined with the Flow-R numerical simulation method,based on DEM data and geological conditions,slope,aspect,lithology,plane curvature and cumulative accumulation are selected as the influencing factors for debris flow starting zone identification.Factor multi-collinearity analysis and Mann-Whitney U test were used to screen the impact factors.By analyzing the interpretation results of the 15 debris flow gully start areas in the study area,the thresholds of the identification factors of the start area are determined,and related literatures are consulted to obtain the debris flow motion simulation parameters.The Flow-R numerical simulation model is used to predict the range of 15 debris flow submerged areas in the study area,and the accuracy and sensitivity are used to verify the simulation results.On the basis of Flow-R numerical simulation,a prediction model of debris flow outflow based on GIS and numerical simulation is established,and the error analysis and applicability discussion of the model are made,and the following conclusions are drawn:(1)Factor multi-collinearity analysis can filter out unnecessary factors in multi-factor analysis to avoid the complexity and inaccuracy of the model caused by too many factors,and the Mann-Whitney U test uses a quantitative method to filter factors,it can be more intuitive and effective selection of factors that have an important impact on debris flow,improve the accuracy of the model.(2)Slope,lithology,and flow accumulation of confluence represent the three major conditions that affect the topography,provenance,and water source of the formation of debris flow initiation zone.The results of factor screening show that these three factors also have an important impact on the initiation of debris flow.Sex and the accumulation of confluence are used as debris flow initiation factors.(3)Use the interpretation results of the debris flow starting area to analyze the slope distribution characteristics,lithology characteristics and river network distribution characteristics extracted from the accumulation of the confluence of the debris flow starting area to provide a basis for the threshold setting of the debris flow starting factor.Refer to the relevant literature to obtain the debris flow motion parameters in the study area,and use Flow-R to perform numerical simulation to obtain the debris flow start area distribution map and debris flow inundation area probability map.The accuracy and sensitivity are used to verify the simulation results of the submerged area.The verification results show that the average accuracy is 86.79% and the average sensitivity is 78.64%,indicating that the simulation accuracy is high and the simulation effect is good.(4)Based on the simulation results of the debris flow submerged area,combined with the debris flow accumulation area division method,the numerical simulation prediction range of the debris flow accumulation area is determined.Based on the GIS platform and the accumulation area prediction results,a debris flow outflow prediction model is established to predict the outflow amount Compared with the measured results,the results show that the relative error is within-21.9% ? 32.76%,the error value is within the acceptable range,indicating that the prediction model can be used to predict the amount of debris flow.(5)An error analysis was carried out on the prediction model of the flushing amount.The results show that the profile type of the debris flow accumulation area and the debris flow accumulation surface simplified by the model are the reasons for the large deviation between the predicted value and the measured value of the model.Based on remote sensing images,the actual flushing volume of Bayi Ditch is estimated.The results show that the estimated results are close to the measured results,with a relative error of only 4.26%,indicating that the prediction model has certain application value in the estimation of the actual flushing volume of debris flow,and also shows that the value The area of the simulated prediction accumulation area is also one of the key reasons that affect the accuracy of the prediction model.