CMIP5模式降水订正法及未来30年中国降水预估

借助英国气候研究所（Climate Research Unit, CRU）全球陆地格点分析数据集（CRU TS v4.0）月降水资料和24个国际耦合模式比较计划第五阶段（Coupled Model Intercomparison Project Phase 5, CMIP5）模式历史气候模拟及RCP4.5情景下的降水预估数据,设计了多种回归方案并对模式降水预估偏差进行订正。这些方案包括一元回归、一元对数回归、一元差分回归、一元对数差分回归、多元回归、多元对数回归、多元差分回归、多元对数差分回归和简单移除气候漂移等。2006～2015年中国大陆模式降水预估的订正结果表明,一元回归订正法普遍优于多元回归订正和扣除气候漂移订正法,其中一元对数回归法的效果最好,其降水距平同号率（Anomaly Rate, AR）和降水距平百分率相关系数（Anomaly Percentage Correlation Coefficient, APCC）最高,分别达到69%和0.5;而降水距平相关系数（Anomaly Correlation Coefficient, ACC）最高的是一元对数差分回归法。不同回归订正法所得预估结果的距平同号格点分布显示,一元对数回归法在北方优于南方,而一元差分（年际增量）或对数差分回归法在南方优于北方。这直接导致在中国南方区域（95°E以东,35°N以南）一元对数回归或多元对数回归订正结果的AR、ACC和APCC均低于对应的差分/对数差分回归法,在北方和西部地区则与此相反。因此,模式降水的回归订正方案具有区域性,这可能源于不同区域降水序列统计性质的差异。用区域组合回归订正法,即在南方用一元差分回归订正,其余地区用一元对数回归订正,其降水预估场的AR提高到72%,但ACC和APCC均略有下降,原因是差分回归订正增加了预估降水场的方差。对RCP4.5情景下2016～2045年24个模式集合平均降水预估的组合回归订正结果显示,相对于1976～2005年平均,未来30年降水异常大致呈南北少,中间多的格局,其中长江中下游、江南中西部、西南东北部、华南沿海和海南省等地降水偏少10%～20%,淮河流域、三江源区和台湾省降水偏多10%～40%,西北东部、华北和东北大部降水正常或略偏少。从降水百分率方差看,模式群的离散度（不确定度）呈现东部小,西部大的分布特征,说明模式预估的西北中部和青藏高原西部等降水偏少区的不确定性较大;而河套北部、华北南部和江南东部等地对应于2006～2015年检验期的“盲区”（模式与观测降水距平反号）,其降水预估参考价值可能不大,需要引入他法加以改进。     

Summer monsoon indices on June rainfall in Hong Kong and its vicinity

Climatologically, June is usually the wettest month in Hong Kong. With significant interannual variation of the summer monsoon, the rainfall variability in June is also large. As Hong Kong is in close proximity to the peripheries of different monsoon regions, the variability of June rainfall largely depends on the relative strength of various monsoon systems. In the present study, a new index comparing the relative condition of the western North Pacific summer monsoon and the South China Sea summer monsoon is developed based on the difference between the respective monsoon indices WNPMI (western North Pacific summer monsoon index) and UMI (unified monsoon index). It is shown that June rainfall in Hong Kong and its vicinity is better correlated with this new index than either WNPMI or UMI alone. Based on the signs of the new index in conjunction with those of WNPMI and UMI, a framework to stratify the monsoon conditions into different configurations together with a simple scheme to summarize the associated rainfall responses is formulated. This study highlights how the rainfall variability on a local or regional scale can be quantified by the broad-scale conditions of different monsoon systems.