Climate change causes a redistribution of water resources by changing the global hydrological cycle.The main influecing factors are temperature and precipitation,which directly affect the surface runoff and thus lead to variations in ecological water demand.This thesis calculates the ecological water demand for the current year(2020),simulates the changes in ecological water demand from 2025 to 2100 by constructing an ecological water demand prediction model,analyzes the trend of changes in ecological water demand based on the ecological water demand calculation results for the current year.In addition,the existing model is improved to increase the simulation accuracy of ecological water demand prediction.The specific results are as follows:(1)Calculate the current annual ecological water demand.The ecological base flow,evaporation water demand,leakage water demand,and vegetation ecological water demand were estimated using the Tennant method,Penman formula,and other methods,and then the ecological water demand of Haoxi Watershed in the current year was determined.The results show that the ecological water demand of the Haoxi Watershed is 67 million m~3 from September of the current year to February of the following year and785 million m~3 from March to August.According to the calculation of ecological water demand under different precipitation frequencies in the Haoxi Watershed,among the ecological water requirements in the wet,normal,and dry years,the dry year is the highest,followed by the wet year,the normal year is the lowest.(2)Establish a prediction model for ecological water demand.Using the weather prediction data from Can ESM2,an SDSM downscale model is constructed to extract the location of the Haoxi Watershed,establish three scenarios,RCP2.6,RCP4.5 and RCP8.5,and perform downscale processing to obtain climate prediction data within the Haoxi Watershed.After calibration and validation,it is proved that the model was suitable for hydrological simulation in the Haoxi Watershed.Using the constructed SWAT model in the study area,predict the runoff from 2025 to 2100 under future climate scenarios and then calculate the ecological water demand within this time scale.The calculation results show that the ecological water demand presents a stable upward trend under the three scenarios,with the RCP8.5 scenario having the most significant upward trend and the RCP4.5 scenario having the slowest upward trend.(3)Improved model simulation accuracy.Using GF1-WFV remote sensing image data and using ENVI software for remote sensing interpretation,the land use data of the study area after fine division is obtained.The SWAT model of the study area constructed from this data is compared with the original model:the simulated R~2 value of the original model is 0.8,and the constructed model after the fine division of land use is 0.85,indicating that the fine division of land use is helpful to improve the accuracy of the model in the study area.The improved model was applied to estimate the ecological water demand of the watershed.Compared with the original model,the changing trend of ecological water demand in the RCP8.5 scenario was similar,with only a few years experiencing abrupt changes;Under the RCP4.5 scenario,the ecological water demand changes more sharply,but the overall increase remains unchanged;The ecological water demand changes are more stable under the RCP2.6 scenario. |