利用ECHO-G海气耦合气候模式模拟的中国东部过去近1000年的降水资料,通过经验正交函数分解方法得到了中国东部夏季降水分布的主要空间型态: 华北和华南地区与长江中下游地区反位相,即华北和华南地区降水多(少)时长江中下游地区降水少(多)。对31年滑动平均后3种空间分布型的时间系数进行Morlet小波分析和功率谱分析发现,3个地区均存在准60年的年代际振荡周期,长江中下游和华南地区均存在准200年的百年际振荡周期,华南地区还存在120年和80年的变化周期。全球海温场与3个地区夏季降水31年滑动平均后的相关分析表明,在40°~70°S,170°~60°W海域海温与华北和长江中下游地区夏季降水负相关,与华南地区夏季降水正相关除该区外的南北半球(30°~60°)中高纬海域海温与华北和长江中下游地区夏季降水正相关、与华南地区夏季降水负相关,说明在年代际和百年际时间尺度上,中高纬海域海温是影响长江中下游地区和华南地区夏季降水反相的主要因子之一。最后通过对太阳常数、火山活动等外强迫因子与中国东部夏季降水的功率谱分析发现: 中国东部夏季降水的年代际与百年际变化周期大都与外强迫因子的变化周期一致,说明外强迫因子,尤其是太阳常数和火山活动的变化是影响中国东部夏季降水年代际和百年际振荡的主要原因。
Summer precipitation of the last 1000 years in East China( 18°- 42°N, 105°- 125°E)was simulated by an atmosphere-ocean coupled climate model ECHO-G. The reliability of this model was validated by comparing the simulated results with the reanalysis results of Global Precipitation Climatology Project (GPCP) and the results reconstructed by tree ring in Delingha,Qinghai Province. The correlation coefficient between the simulated and the reanalyzed annual precipitation is 0.82, and that between the simulated and reconstructed results is O. 44; both have a confidence level more than 95%. The main spatial patterns of summer (JJAS) precipitation (SP) in East China were obtained by the Empirical Orthogonal Function (EOF). The spatial distribution of SP in the three regions of North China(NC) ,the middle and lower reaches of Yangtze River (YZ)and South China(SC) shows the "positive" ," negative" and "positive" patterns. That is,when precipitation in YZ is more than the normal,that in NC and SC will be less than the normal,vice versa. The Morlet wavelet analysis and the power spectrum analysis on the 31-year running mean time series of SP show that there is a quasi-60a periodicity in the three regions. And there is a quasi-200a periodicity in YZ and SC and a 120a periodicity and a 80a periodicity in SC. The correlation analysis on the 31-year running mean series of SP anomaly ( in inspect to 1000 - 1990A. D. ) and SST in the previous winter( DJF) shows that SST in the region of 170° -60°W and 40° - 70°S is negatively correlated with SP in NC and YZ and positively correlated with SP in SC. SST in other mid-high latitude regions (30° - 60°N and 30° -60°S)except the above one is positively correlated with SP in NC and YZ and negatively correlated with SP in SC. These suggest that on muhi-deeadal and centennial scales,SST in mid-high latitude regions is one of the major factors causing the opposition of SP in YZ to SP in SC. The power spectrum analysis on the