Analysis On Fitting Result Of SCS-CN Runoff Model In Yaoxiang Small Watershed

Water erosion in Shandong province district,and its caused by runoff scour district,convenient for soil erosion prevention district accurately predict runoff,soil have important practical significance.The American SCS-CN hydrological model is one of the models for runoff prediction.However,its application depends on the underlying surface environment in specific areas,and its direct application will inevitably cause great errors and reduce the work efficiency of soil and water conservation.Taking Yaoxiang Small Watershed as the research object,using the method of plot runoff dynamic monitoring of rainfall and runoff data,five kinds of land use type,using the least squares fitting SCS-CN model,using four evaluation index to evaluate the result of the fitting,and using TOPSIS method to comprehensive evaluation index to judge.The SCS-CN model suitable for this region is sought to provide references for the study of regional production flow and hydrological forecast.The results are as follows:(1)The sensitivity of different model types is different.:Rainfall(P)and runoff curve number(CN)have greater impact on runoff prediction.In the standard model,the parameter sensitivity is the runoff number curves(CN),rainfall(P),and initial loss rate(λ).In the preceding affected precipitation type,the parameter sensitivity were P,CN,M andλ,and in the rainfall intensity type,the parameter sensitivity were CN,P,P_e,λ.(2)Under the same land use type,the best fitting models determined by different evaluation indexes are different.For example,in the slope farmland,the optimal RMSE index is the SM-1 and the SM-2 model,the optimal E index is the MSC model,the best of the (?)index is the MA-1 model,and the optimal q_r index is the OR-2 model.In bare land,the optimal NRMSE index is RI model,the optimal E index is SM-2 model,the optimal (?)index is the MA-1 mode and the q_r index is SM-2 model.In the woodland,the optimal NRMSE index is the RI model,the optimal E index is SM-1,the optimal (?)index is OR-2 model,and the q_r index is the MSV model.(3)When SCS-CN models were fitted,the rainfall size should be distinguished.Under the different rainfall levels,the analysis of RMSE,E,(?)and q_r showed that:except the OR-1model,RMSE,E,QR index showed that the fitting effect of other model rainfall grade 2 was better than the rainfall grade 1,and except OR-1 model,(?)indicator showed the fitting effect of other model rainfall grade 1was better than the fitting effect of the rainfall grade 2.(4)Topsis comprehensive evaluation method can determine the best SCS-CN model in the study area.The rainfall grade 2,the optimal model slope farmland is MA-1 model,terrace is SM-1 model,bare land is RI model,grassland is MA-1 model,woodland is RI model,and the other standard model,that is,OR-1 model is not suitable for research area.The SCS-CN model takes into account the characteristics of the lower surface of the small watershed and the influence of human activities.In addition,the model has the advantages of simple structure,few parameters and convenient use.It is more compatible with other models and has a wide range of applications.The work of this paper provides a new method for predicting runoff in the study area,and helps to improve the efficiency of soil and water conservation.However,in this model,the CN value is the only parameter that responds to the characteristics of the basin,and it lacks the overall consideration of the microclimate,vegetation hydrological feedback mechanism and interaction mechanism of small watershed.Therefore,the internal mechanism and micro response of vegetation,climate and hydrology need to be studied in depth so as to improve the accuracy and accuracy of the model prediction..