剔除“系统误差”对月尺度区域气候动力预测影响的研究

STUDY OF THE EFFECT OF ELIMINATING THE SYSTEMATIC ERRORS ON MONTHLY-SCALE REGIONAL CLIMATE DYNAMICAL PREDICTION SOUTHERN

通过大气环流模式提供侧边界条件,单向嵌套并入了水文模型VXM的区域气候模式RegCM3对我国华东地区进行月尺度气候预测。根据嵌套模式1991～2000年10年各月的积分结果与美国气候预测中心的降水综合分析资料(CMAP)、美国国家气象中心(NMC)的温度资料建立了一组针对各月的"系统误差",对2001～2005年5年各月的回报结果进行订正,通过对原始回报结果和订正结果进行评估发现,经剔除"系统误差"后的预测结果较原始预测结果改进较为明显,表明该方法可有效提高月尺度区域气候动力预测的准确率。

A general circulation model is embedded with a fine-resolution regional climate model, RegCM3, into which the hydrological model VXM is incorporated. Then the nested models system is used for predicting the monthly-scale regional climate over East China. In the system, the systematic bias/errors are formulate based on the results of the nested models from 1991 to 2000 with CMAP reanalysis precipitation data and the NMC reanalysis temperature data. Then, they are used for correcting the hindcast results during 2001 - 2005. After evaluating the original and the corrected results of the temperature and precipitation, it is found that the corrected results are better than the original results obviously. All these show that the method can be applied for enhancing the accuracy in monthly-scale regional climate prediction.