利用云南124个气象观测站逐月降水量资料、 NCEP/NCAR再分析大气环流资料和海表面温度（SST）资料， 分析了云南冬季降水时空演变特征及其相对应的大气环流异常和海温异常。结果表明： （1）云南冬季降水第一模态表现为一致性变化模态, 方差贡献为53%, 具有显著的准5年周期, 该模态具有一定的年代际变化特征。（2）第二模态表现为滇中及以东、 以南与滇西及滇西北地区反位相振荡模态, 方差贡献为13.4%， 具有显著的准3年周期， 也具有一定的年代际变化特征。（3）云南冬季降水与大气环流异常密切相关。当云南冬季降水偏多（少）时， 中国大陆大部分地区海平面气压偏高（低）， 近地面冷高压活动频繁（偏少）， 冷空气易（不易）南下影响云南； 500 hPa高度场上, 中高纬度贝加尔湖附近高度场偏高（低）， 该处的脊强（弱）， 有（不）利于引导北方冷空气南下； 同时, 中低纬度在孟加拉湾北部高度场偏低（高）， 南支槽偏强（弱）， 有（不）利于南方暖湿空气向北输送。来自南海和孟加拉湾的异常水汽输送在云南形成辐合（辐散）， 从而造成该地区降水异常偏多（少）。（4）云南冬季降水与海温异常也存在密切关系。云南冬季降水偏多、偏少年, 太平洋海温差异场呈类似El Nino年的分布： 赤道中东太平洋的SST异常偏高， 而菲律宾以东的西太平洋SST异常偏低。El Nino年冬季降水以偏多为主； 而La Nina 年， 则以偏少为主。赤道中东太平洋海温异常对云南冬季降水具有一定的潜在预报意义。（5）北极涛动（AO）与云南冬季降水存在密切相关关系。当AO为正（负）位相时， 云南冬季降水偏多（少）。同时, AO与云南冬季降水的关系在一定程度上受ENSO事件的制约, 在ENSO暖位相时， 冬季AO与云南冬季降水密切相关, 而在ENSO冷位相时两者间几乎没有联系。

Using the monthly observed precipitation data of 124 stations of Yunnan Province from 1961 to 2010, the monthly mean atmospheric circulation data and SST data from NCEP/NCAR reanalysis, the evolutive features of the winter precipitation in Yunnan and correspondingly atmospheric circulation anomaly and SST anomaly were studied by correlation analysis method and synthesis method. The results indicate that: （1）The winter precipitation mainly shows the consistency change mode. This mode accounts for 53% of the total variance and has significant period of 5 years or so. It also has obvious interdecadal variations. （2）The second main mode of the winter precipitation in Yunnan shows an reverse phase oscillation between the middle area, the south area and the east area of Yunnan and the west area and the northwest area of Yunnan. This mode accounts for 13.4% of the total variance and has significant period of 3 years or so. It also has obvious interdecadal variations. (3）It is founded that the wintertime precipitation in Yunnan has intimate correlation with atmospheric circulation by the correlation method. When the wintertime precipitation in Yunnan is above(below) normal, the SLP in the most of China is higher(lower), the activity of cold high near surface is more(less) frequent and the cold air is (not) easy to influence southward Yunnan Province. The height near Lake Baikal is higher(lower), the high ridge is stronger(weaker) in 500 hPa height field, so it is (not) useful to lead the cold air toward the south. At the same time, the height of the north of Bay of Bengalis lower(higher), the southern branch trough is stronger(weaker), so it is (not) useful to lead the warm and wet air toward the north. The anomalous water vapor transport from the south China sea and the Bay of Bengal converges(diverges) in Yunnan and leads to more(less) precipitation in this region. （4）Through the synthesis method, it is founded that the wintertime precipitation in Yunnan also has intimate correlation with SST anomalies. The SST difference field between the above normal years and below normal years of the wintertime precipitation in Yunnan province is similar with the distribution of SST anomaly of El Nino years: The SST in the tropical eastern Pacific is higher and the SST in the western Pacific Ocean of the east of the Philippines is lower. In El Nino years, the wintertime precipitation in Yunnan is mainly above normal. On the contrary, in La Nina years, the wintertime precipitation in Yunnan is mainly below normal. The SST anomaly in the tropical eastern Pacific may act as a potential predictor for the wintertime precipitation in Yunnan. （5）There is closely relationship between AO and the wintertime precipitation over Yunnan. When the AO index is positive(negative) phase, the wintertime precipitation over Yunnan is above(below) normal. This relationship is modulated by ENSO. In the warm ENSO phase, this relationship between AO and the wintertime precipitation over Yunnan is much close, while in the cold ENSO phase, there is scarely relationship between AO and the wintertime precipitation over Yunnan.