Analysis of Variation Characteristics of Precipitation in Nearly 100 Years of China

中国气象台站资料具有时间较短(1951年以前观测站缺测资料较多)和覆盖面不均匀(东部站点密集，西部站点稀少)的特点，不利于中国降水年代际变率的研究。本文利用英国East Anglia大学气候研究中心(CRU)全球陆地逐月平均降水数据集TS3.22，分辨率为0.5˚ × 0.5˚，时间长度为1901.1~2013.12的CRU资料，德国气象局全球降水气候中心(Global Precipitation Climatology Centre)全球陆地逐月平均降水数据集，分辨率为0.5˚ × 0.5˚，时间长度为1901.1~2010.12的GPCC以及中国2782台站逐月降水资料，插值为0.5˚ × 0.5˚格点资料，时间长度为1901.1~2013.12的台站资料，使用趋势分析，小波分析，M-K突变分析，功率谱分析等方法分析不同区域降水的变化特征和规律，揭示中国百年来降水的时空变化特征。结果表明：1) CRU和GPCC资料与中国台站资料降水的趋势基本吻合。2) 东北和华北地区近百年的降水呈线性上升趋势；西南地区呈线性下降趋势。3) 小波分析表明，东北地区存在2~3年和22左右的周期特征；华北地区有1~2年和3年左右的周期特征；江南地区有2年左右的周期特征；西南地区有1~2年和2~4年左右的周期特征。4) 通过功率谱分析，西北地区降水存在4~6年的周期振荡，其它区域没有显著的周期特征。5) 通过Z指数反映出了各个区域近百年的旱涝程度。Chinese meteorological station data have limitations such as relatively short records (with significant data gaps before 1951) and uneven spatial coverage (denser stations in the east but sparser in the west), which hinder the study of decadal precipitation variability in China. This paper utilizes the monthly average precipitation dataset of global land (TS3.22) from the Climatic Research Unit (CRU) of the University of East Anglia, UK, with a resolution of 0.5° × 0.5° and a time span from January 1901 to December 2013; the monthly average precipitation dataset of global land from the Global Precipitation Climatology Centre (GPCC) of the German Meteorological Service, with a resolution of 0.5° × 0.5° and a time span from January 1901 to December 2010; as well as the monthly precipitation data from 2782 stations in China, which are interpolated into grid data with a resolution of 0.5° × 0.5° and cover the time period from January 1901 to December 2013. Methods such as trend analysis, wavelet analysis, Mann-Kendall mutation analysis, and power spectrum analysis are employed to analyze the variation characteristics and laws of precipitation in different regions, aiming to reveal the spatiotemporal variation characteristics of precipitation in China over the past century. The results show that: 1) The precipitation trends in the CRU and GPCC datasets are basically consistent with those in the Chinese station data. 2) Precipitation in Northeast China and North China has shown a linear upward trend over the past century, while that in Southwest China has shown a linear downward trend. 3) Wavelet analysis indicates that there are periodic characteristics of about 2~3 years and 22 years in Northeast China; about 1~2 years and 3 years in North China; about 2 years in the Jiangnan region; and about 1~2 years and 2~4 years in Southwest China. 4) By power spectrum analysis, there are periodic oscillations of precipitation in Northwest China, and there is no significant periodic characteristic in the other regions of the precipitation in the northwest of China. 5) The Z index reflects the drought degree of nearly one hundred years in each region.