摘要: 本文采用GCMs和水文模型耦合的方式，基于SWAT在渭河流域构建分布式水文模型，并采用SUFI-2算法进行参数敏感性分析、参数率定、模型验证以及不确定性分析，从而对渭河流域1961~2008年的径流过程进行模拟。然后将GCMs降尺度生成不同情景下降水、最高、最低气温日序列输入SWAT模型，模拟流域未来径流量，从而分析未来不同气候变化条件下流域径流可能的变化。研究结果表明：2046~2065和2081~2100时期不同情景下流域多年平均径流量分别为80.4与104.3亿m3，较基准期增加12.4%和45%。未来两个时期，枯季流量较基准期更低，而洪峰流量则将较基准期更高，即流域内极端事件(干旱与洪水)在未来两个时期有加剧趋势。不同情景下渭河流域径流深空间变化较为一致，即上游部分子流域和北洛河上游地区径流量较基准期有所减少，而流域中下游地区径流量均呈一定的增加趋势。
Abstract:
In this study, Soil and Water Assessment Tool (SWAT) was selected to set up a hydrological model in the Wei River basin (WRB), calibrated and validated with Sequential Uncertainty Fitting program (SUFI-2) based on river discharge, then future daily precipitation, maximum and minimum air temperature data series at each station, generated by the Statistical Downscaling Method (SDSM), were inputted to drive the SWAT model for analyzing the spatiotemporal characteristics of runoff during the future periods (2046-2065 and 2081-2100) under three climate scenarios including CSIRO, INM and MRI. Two emission scenarios (SRES A2 and SRES B1) were also included. The results show that average values of mean annual runoff in the periods of 2046-2065 and 2081-2100 were 80.4 × 108 m3 and 104.3 × 108 m3, which were greater than runoff in the base period by 12.4% and 45%, respectively. In both of the future periods, low flows would be much lower, while high flows tend to be much higher than that in the base period. In other words, there would be more extreme events (droughts and floods) in the future. For the spatial distribution of runoff over the WRB, it showed consistency for runoff changes under most combined scenarios, with runoff decreased at some areas of upstream and the upstream of Beiluo River, while increased at mid-lower stream of the WRB.