| An effective parameter optimization method can reduce the uncertainty of the simulation and thus improve the accuracy of the hydrological model.The uncertainty analysis and rate determination methods of model parameters are important areas of hydrological model application research.In this paper,the distributed hydrological model SWAT is used to analyse the uncertainties caused by the model parameters in the catchment area of the Shikoumen Reservoir(hereinafter referred to as the Shikoumen basin)and to explore the optimization methods of the model parameters in order to improve the accuracy of the basin runoff simulation and forecasting.The research results will be applied to the simulation of runoff processes in the Shiqoumen basin under future climate scenarios,providing technical support to cope with climate change and having important implications for the formulation of water resources management policies,water resources protection and optimal allocation.The main research results are as follows.(1)Construction of the SWAT model for the Shiokoumen watershedBased on the basin meteorological and hydrological data,DEM,land use,soil type and other spatial and attribute data,the SWAT model database was established,and the study area was divided into 35 sub-basins and 252 hydrological respo NSE units to construct the SWAT model of the Shi Koumen basin.(2)Uncertainty analysis of basin SWAT model parametersThree algorithms,GLUE,SUFI-2 and PSO,were selected for the uncertainty analysis of the model parameters,and comparative analyses were conducted on the principles of the algorithms,the relevance of the a posteriori parameters,simulation capability,uncertainty size,computational efficiency and implementation difficulty.The simulation accuracy and uncertainty of all three algorithms can meet the expected accuracy,and the SUFI-2 algorithm is very efficient in searching for the parameter combination that is closest to the optimal value of the objective function.Therefore,SUFI-2 is selected as the optimal parameter determination algorithm for this basin.(3)Research on the intelligent optimisation of SWAT model parameters in the basinIn order to identify i NSEnsitive parameters and reduce synoptic interference,the key SWAT model parameters that affect simulation accuracy are selected using global sensitivity analysis,and the SOM intelligent clustering algorithm is introduced to optimise the SWAT model parameter intervals.The results show that the simulation results based on parameter optimisation intervals can obtain narrower uncertainty intervals and better simulation accuracy,effectively solving the problem of parameter values exceeding the initial intervals after iteration using the SUFI-2 algorithm,increasing the credibility of the simulation results,improving the simulation and forecasting accuracy,effectively reducing the heterogeneous and synoptic interference in the model,and effectively reducing the uncertainty of the runoff simulation process.(4)Simulation of runoff processes in the Shikoumen watershed under future climate scenariosBased on the four emission scenarios SSP 1-2.6,SSP 2-4.5,SSP 3-7.0 and SSP 5-8.5of CMIP6,the GFDL-ESM4 climate model data were selected and the parameter-optimized SWAT model was applied to simulate the hydrological processes in the basin under climate scenarios,and the results show that the hydrological processes in the basin are highly sensitive to climate change,and the annual The overall trend of annual runoff in the basin under the four scenarios is slightly increasing,and the future runoff changes are basically consistent with the trend of future precipitation changes,and the magnitude of the multi-year average runoff in the basin during 2024-2050 is ranked as SSP 3-7.0 < SSP 1-2.6 < SSP 5-8.5 < SSP 2-4.5,in which SSP 2-4.5,SSP 3-7.0,and SSP 5-8.5 scenarios are likely to have larger future The extreme changes in runoff will lead to a high incidence of flooding events in the basin,posing a great challenge to water resources security. |
PDF Full Text Request