Prediction model for maximum sediment carrying capacity of flash floods in the Wenchuan earthquake area
After the Wenchuan earthquake,a substantial amount of loose deposits was generated in the earthquake area.The flash flood disaster became more destructive under the influence of heavy rainfall,increasing the risk of disaster and complicating disaster prevention efforts.Studying flash flood disasters in the earthquake area can provide a foundation for assessing disater risk and providing a basis for subsequent management.Using 66 historical flash flood disasters as sample data—34 in Wenchuan and 32 in Dujiangyan—the study aimed for a more accurate model by considering comprehensive influencing factors.Information on five influencing factors,including basin area,main channel length,relative elevation difference,ditch bed vertical gradient,and source quantity,was collected from the sample data.Correlation analysis identified four main control factors:basin area,main channel length,relative elevation difference,and source quantity.The factor of vertical gradient was removed due to its extremely low applicability to the curve of regression simulation.Using SPSS software,nonlinear regression analysis was conducted to establish an estimation model based on the maximum sediment carrying capacity and maximum sediment carrying distance of one flood.To validate the sediment carrying capacity prediction model,23 flash flood disaster data points from Wenchuan and Dujiangyan were selected and compared with other models.The R2 of the maximum sediment carrying capacity model for the flood is 0.762,indicating a good explanatory power of 76.2%for sediment carrying capacity.The relative error of the sediment carrying capacity model validation data based on floods ranges between-52%and 47%,with 73.9%of the data having a relative error between 10%and 40%,and 17.4%of the basin having a relative error of around 52%.This indicates that the model prediction in this study is more stable and reliable.The R2 of the maximum sediment carrying distance model for the flood is 0.88,indicating good explanatory power of 88%for the maximum sediment carrying distance.The relative error of the validation data for the maximum sediment carrying distance model based on floods ranges between-47%and 46%.The prediction value of this model is more stable,with an overall relatively small relative error,suggesting higher accuracy.The results show that the model in this study is reliable and precise.The research results will aid in predicting potential flash flood disaster areas in Wenchuan earthquake area and provide a basis for flash flood disaster assessment.
NETL
NSTL
万方数据
