长江上游时空相关气候要素的区域趋势诊断

构建融合时空相关性的区域Spearman秩次相关检验方法,诊断长江上游CRU数据集中年降水量、年平均气温和年潜在蒸散发量在各子流域的区域变化趋势。结果表明:各气候要素均表现出明显的正向空间(互)相关性,气温和蒸散发还呈现出正向高阶时间(自)相关性,2种因素都会导致区域趋势显著性被高估。若忽略该影响,则判断上游各子流域均发生了显著气候变化。新方法有效削减了时空相关性的影响,发现近百年来上游气候变化趋势尚不显著,降水仅在金沙江区显著增加,在川江流域略有减少,气温和蒸散发在1960年以后才显著上升。研究结果表明考虑时空相关性对减少趋势误判起着重要作用。

Regional climatic trend analysis in upper Yangtze River with consideration of both temporal and spatial correlation

A regional average Spearman Rho method that takes into account both temporal and spatial correlation was developed. The method was used to assess regional variation trends of annual precipitation,annual mean air temperature,and annual potential evapotranspiration in each subcatchment in the upper Yangtze River based on the CRU dataset. The results show that the positive spatial cross-correlations between sites are prevalent for each climatic element. The annual mean air temperature and potential evapotranspiration also exhibit positive temporal autocorrelations. Both correlations can lead to overestimation of the significance of regional trends. If the correlations are ignored,it is concluded that a significant climatic variation occurs in each subcatchment in the upper Yangtze River. The new method effectively reduces the effect of temporal and spatial correlation,and concludes that the climatic trends have not been statistically significant over the last century,and that the annual precipitation only significantly increased in the Jinsha River area,whereas it slightly decreased in the Chuanjiang area. Meanwhile,the annual mean air temperature and potential evapotranspiration increased significantly after 1960. The results of the study indicate that consideration of temporal and spatial correlation plays an important role in decreasing trend misjudgment.