GCM降尺度预测泾河流域未来降水变化

利用统计降尺度模型SDSM确定预报因子，建立预报量与预报因子之间的统计关系，用独立的观测资料验证模型，并将其应用于HadCM3在A2和B2两种排放情景下的输出生成未来气候变化情景，预测了泾河流域未来3个时段（2020s，2050s，2080s）的降水变化，以期为流域未来气候变化情景的构建及当地农业的可持续发展提供参考依据。结果表明，泾河流域未来年降水量呈不显著的减少趋势且存在一定的季节差异。夏季降水减少趋势最明显，减幅随时间的推移逐渐增大；冬季呈现微弱增加趋势；春季与秋季变化趋势类似，均有增有减，减少的时期较增加的时期多。流域降水变化的空间差异大体呈现由东南部向西北部减小的趋势。

Downscaling GCMs to Project the Potential Changes of Precipitation in Jinghe Basin

In order to predict the future precipitation change to improve the construction of the future climate change scenario and sustainable development of local agriculture, we built the statistical relationship between large-scale climatic elements and precipitation variables, identified predictor variables used in the model, then tested the model reliability by applying independent observation data, applied previous statistical relation to the output of A2 and B2 emission scenarios of HadCM3, and to predict future precipitation change in the Jinghe Basin at last. Results show that future precipitation of Jinghe Basin will be bound to reduce; however, there was no significant difference compared with base period (1961—1990). There were some seasonal differences. The decreasing trend of summer precipitation was obvious, and the change increased gradually over time. Winter precipitation increased weakly. The precipitation change trend of spring was similar to the trend of autumn, and reductive months were more than increased months. There was spatial difference in precipitation change of watershed. The increase reduced from the southeast to the northwest part of the basin. It was shown that the future precipitation spatial difference would be larger.