1961-2017年贵州降水时空变化特征

为研究贵州地区降水的时空演变规律，根据贵州省84个气象站1961-2017年的降水量资料，利用EOF和REOF方法、Mann-Kendall检验、Cramer法、YAMAMOTO法和滑动t检验及Morlet小波分析法，对贵州地区57年来降水的空间分布、时间系数变化特征和时间序列的趋势性、突变性及周期性进行了分析。结果表明:EOF方法可以很好地解释降水场的时空分布特征，前2个特征向量累计贡献率达50.08%，揭示了两种典型的分布场，即全局型和东西反向型。分析特征向量所对应的时间系数，贵州省的降水场主要表现为全省多雨、全省少雨、东多西少、东少西多4种类型，REOF可将贵州降水场分为东西部两个区域，通过Mann-Kendall检验发现均存在降水减少的突变，并通过Cramer法、YAMAMOTO法和滑动t检验对结果进行验证，发现Ⅱ区比Ⅰ区降水突变明显。Morlet小波分析表明，贵州地区降水存在着显著的28 a周期变化。

Temporal and Spatial Distribution of Precipitation in Guizhou Province from 1961 to 2017

In order to study the temporal and spatial evolution process of precipitation in Guizhou Province, according to the precipitation data of 84 weather stations from 1961 to 2017, we analyzed time series trends, periodicity, mutation and spatial distribution of the precipitation of the Guizhou Province in the recent 57 years, by using the methods of EOF, REOF, Mann-Kendall, Cramer, YAMAMOTO, t checking and Morlet analysis. The results show the EOF method is capable for revealing the spatial and temporal distributions of precipitation field. The accumulated contribution of the first two EOF can reach to 50.08%, and reflect two typical vector distribution fields, which is the whole region pattern and east-west reverse pattern. Analysis of the eigenvector of the corresponding time factor coefficient demonstrates that the precipitation fields of the Guizhou Province mainly exhibite four types, abundant rain in whole province, less rain in whole province, abundant rain in east and less rain in west, less rain in east and abundant rain in west. On the basis of the REOF of rainfall, the Guizhou Province could be divided into two regions, east part and west part. Mutation is found in rainfall trend by Mann-Kendall test. Verification by the methods of Cramer, YAMAMOTO and t checking, indicates that the abrupt change of precipitation in zone Ⅱ is more obvious than zone t. Morlet wavelet analysis suggests that there is a strong 28-year period of the precipitation in Guizhou Province.