基于VIC模型的金沙江流域水文要素分析和径流预测研究

Research on Hydrological Element Analysis and Runoff Prediction in the Jinsha River Basin Based on the VIC Model

[目的]金沙江流域是西南高山峡谷区的重要生态区域。在全球气候变暖背景下,预测并评估金沙江气候及未来径流变化对流域尺度的水土流失防治、防洪减灾和水土资源规划具有重要意义。[方法]基于VIC模型模拟金沙江流域的水文过程,对CMIP6全球气候模式下的2种排放情景(SSP2-4.5、SSP5-8.5)的20个气候模式及多模式集合平均(MMEA)的数据进行降尺度处理,预测2015-2100年径流变化趋势。[结果](1)历史径流模拟显示,模型模拟值与实测值的纳什效率系数超过0.6,BIAS(相对偏差)绝对值控制在0.15以内,表明VIC模型适用于模拟金沙江流域水文过程;(2)VIC模型模拟的水文过程分量显示,降水量从上游到下游递增,蒸散发在中下游较高,地表径流和基流的多年平均值从上游到下游递增,且存在显著差异;(3)基于EEMD(集合经验模态分解)的统计降尺度方法对比单模式和(MMEA)的模拟结果发现,观测数据与MMEA匹配度最高。在未来SSP2-4.5和SSP5-8.5情景下,模拟结果显示高辐射高排放情景(SSP5-8.5)下水文要素变化更显著,尤其是蒸散发增加趋势明显。[结论]径流量模拟显示,SSP2-4.5情景下年平均径流量变化不明显,而SSP5-8.5情景下年平均径流量呈小幅增长趋势。未来年内月径流量在SSP5-8.5情景下大多数月份高于SSP2-4.5情景。研究结果以期为金沙江流域水资源评估与管理、生态综合治理提供一定参考依据。

[Objective]The Jinsha River Basin is an important ecological area in the southwestern mountainous gorge region.Predicting and assessing climate and future runoff changes in the Jinsha River Basin under the background of global warming is of great significance for soil and water conservation,flood control,disaster reduction,and water resource planning at the basin scale.[Methods]This study simulates the hydrological processes of the Jinsha River Basin using the VIC model and downscales data from 20 climate models and the Multi-Model Ensemble Average(MMEA)under two emission scenarios(SSP2-4.5,SSP5-8.5)from the CMIP6 global climate models to predict runoff trends from 2015 to 2100.[Results](1)The historical runoff simulation indicates that the Nash efficiency coefficient between the simulated and observed values exceeds 0.6,and the absolute value of BIAS is controlled within 0.15,demonstrating that the VIC model is suitable for simulating the hydrological process in the Jinsha River Basin.(2)The hydrological process components simulated by the VIC model show that precipitation increases from the upstream to the downstream regions,evapotranspiration is higher in the middle and lower reaches,and the multi-year average values of surface runoff and base flow increase from the upstream to the downstream regions with significant differences,further validating the model's applicability.(3)A comparison of single-model and MMEA simulation results using the EEMD-based(Ensemble Empirical Mode Decomposition)statistical downscaling method reveals that the MMEA matches the observed data the best.Under future SSP2-4.5 and SSP5-8.5 scenarios,the simulation results indicate that hydrological elements change more significantly under the high-radiation and high-emission scenarios(SSP5-8.5),with a particularly noticeable increase in evapotranspiration.[Conclusion]The runoff simulations show that the annual average runoff remains relatively unchanged under the SSP2-4.5 scenario,while it shows a slightly increasing trend under the SSP5-8.5 scenario.Future monthly runoff simulations indicate that most months under the SSP5-8.5 scenario have higher runoff compared to the SSP2-4.5 scenario.