利用我国160站的观测资料、NCEP/NCAR再分析资料以及英国气象局哈德莱中心的海表面温度（SST) 资料, 分析了我国冬季降水年际变化的主模态以及与其相联系的大气环流异常和海温异常。结果表明, 在年际变化的时间尺度上, 我国冬季降水主要表现为长江以南地区降水量的一致变化（EOF1) 和华南、西南地区与新疆北部、华北和长江中下游地区降水反位相振荡的变化（EOF2) 两个主模态, 它们分别可以解释年际变化总方差的49.6％和17.3％, 并具有显著的2～4年周期。此外, 这两个模态也有明显的年代际信号, EOF1表现为20世纪80年代中期以前南方地区冬季降水偏少, 1988年之后转为偏多, 而进入21世纪后逐渐接近历史平均值并有再次转为偏少的趋势, 而EOF2表现为1980年至2005年新疆北部和长江中下游地区的冬季降水出现年代际增多, 而华南地区的冬季降水出现年代际减少。进一步通过回归分析表明, 我国冬季降水年际变化的EOF1与ENSO循环以及东亚冬季风强度的变化有密切的关系, 当ENSO处于其暖（冷) 位相的冬季时, 东亚冬季风系统偏弱 （强), 来自孟加拉湾和南海的异常水汽输送在我国南方地区形成辐合（辐散), 从而有利于该地区降水异常偏多（少)。与我国冬季降水年际变化EOF2相联系的环流表现为欧亚大陆上空具有相当正压结构的一个波列, 该波列对应于东亚沿岸的异常北 （南) 风, 进而引起长江中下游地区水汽辐散（辐合) 和华南地区水汽辐合（辐散), 有利于南正（负) 北负（正) 的降水异常分布。赤道中东太平洋和北大西洋挪威海地区的SST异常分别与这两个模态具有紧密的联系, 并对这两个模态具有一定的潜在预报意义。
Based on the observed precipitation data from 160 China stations, the NCEP/NCAR reanalysis dataset and the sea surface temperature (SST) dataset from the Met Office Hadley Centre, this study investigates two major modes of wintertime precipitation over China through empirical orthogonal function (EOF) method and associated atmospheric circulation and SST through linear regression method. It reveals that the first two EOF modes account for 49.6% and 17.3% of the total variance, respectively, and can be well separated from the remaining modes. EOF1 reflects the strength of wintertime precipitation over southeastern China. The associated principal component (PC1) has significant interannual variability with 2－4 year period. Its interdecadal variations suggest that the wintertime precipitation over southeastern China was below (above) normal before (after) the mid-1980s, and has a slight decreasing trend in recent years. EOF2 delineates an out-of-phase relationship between South China and the middle and lower reaches of the Yangtze River as well as the northern part of Xinjiang. The associated time series (PC2) also has a clear 2－4 year period and an interdecadal component, which experiences a clear trend towards its negative polarity from 1980 to 2005. Further analysis suggests that EOF1 is closely related to the El Niño-Southern Oscillation (ENSO) and the East Asian winter monsoon (EAWM). When warm (cold) ENSO is in its mature phase, the EAWM is weak (strong), the anomalous water vapor transport from the Bay of Bengal and the South China Sea converges (diverges) over southeastern China and favors more (less) precipitation in this region. The SST anomalies over the tropical eastern Pacific have significant relationship with the variation of EOF1, and may act as a potential predictor for this mode. The variation of EOF2 is closely associated with a barotropic wave train across the Eurasian continent originating from the North Atlantic, which can induce anomalous northerlies (southerlies) along the coasts of southeastern China. It then causes the divergence (convergence) of water vapor flux in the lower reaches of the Yangtze River and opposite condition in South China, and therefore favors a positive (negative) phase of EOF2. The SST anomalies around the Norwegian Sea over the North Atlantic have the closest relationship with the variations of EOF2, and may act as a potential predictor with a leading time of 7 months.