【作者】 唐佳；

【导师】 武炳义；

【作者基本信息】 中国气象科学研究院 ， 气象学， 2012， 硕士

【摘要】 本文利用31年（1979–2009年）NCEP/NCAR再分析资料和JRA–25再分析资料，通过复矢量经验正交分析方法（CVEOF）揭示了东亚地区夏季（6–8月）850hPa风场变率的优势模态。本文详细分析了东亚夏季风前两个模态的时空分布特征，及其对夏季降水的影响，并讨论了外强迫对东亚夏季风异常的可能影响。得到以下主要结论：（1）NCEP/NCAR再分析资料揭示的东亚夏季风第一模态是由两个相似的子模态或它们的线性组合构成的，它们分别被命名为M11和M12。第一复主成分的实部和虚部可以分别作为强度指数来描述这两个季风子模态。M11和M12的相关系数为0.77（扣除趋势以后）。当M11和M12为正位相时，850hPa风场异常都呈现一个东亚/太平洋（EAP）遥相关波列，从西北太平洋经过日本地区到达鄂霍次克海。M11对应的500hPa高度场异常显示一个欧亚大陆北部纬向遥相关波列和一个EAP遥相关波列，而M12只体现一个EAP波列。与M11和M12联系的我国夏季降水异常都表现为长江中下游地区和东北北部降水显著减少，东北东部、东南沿海和西南西部降水增加。在M11和M12正位相时期，海平面气压场（SLP）异常在西北太平洋都表现为显著的低压异常，而向外长波辐射（OLR）异常在菲律宾和南海地区都为显著的负异常，对流加强。JRA-25再分析资料显示类似的结果。（2）M11与前期冬季ENSO有密切的联系。而M12为正位相时，前期冬季，热带印度洋海温异常和热带太平洋海温异常同时存在。此时ENSO处于最强盛阶段，印度洋海温异常也超过95%显著性水平。春末，ENSO衰减，而印度洋海温异常达强盛阶段。同期夏季，ENSO消亡，而印度洋海温异常持续对东亚夏季风产生影响。JRA-25再分析资料也显示类似的结果。通过偏相关方法分析得到，前期冬、春季，热带太平洋海温异常和热带印度洋海温异常基本上是相互独立的，对后期东亚夏季风变率都存在影响。（3）两套再分析资料所揭示的东亚夏季风第二模态在20世纪90年代初期均发生了年代际转型，与我国夏季降水的年代际转型时间一致。东亚夏季风第二模态与中高纬大气环流异常关系密切，对应的500hPa高度场异常呈现一个欧亚大陆北部准纬向遥相关波列。而对应我国夏季降水异常则表现为经向偶极型分布。伴随着东亚夏季风的年代际转型，90年代以后我国夏季北方大部分地区降水减少，特别是我国东北北部和黄河长江之间105oE附近地区显著减少，而华南地区和淮河流域夏季降水显著增加。从动力上解释我国夏季降水年代际转型特征，夏季500hPa高度场两个时段的差值分布（1993–2009年和1979–1992年）显示为欧亚大陆北部准纬向遥相关波列，而夏季850hPa风场的差值分布则表现为：贝加尔湖东南侧和日本以南地区存在两个异常反气旋式环流，而我国南方地区和鄂霍次克海附近则为异常气旋式环流。（4）夏季西北太平洋、北印度洋以及部分中高纬海洋（北大西洋和北太平洋）的海表面温度和春季欧亚大陆积雪水当量在20世纪90年代初出现显著的变化，春季北极海冰的年代际转型发生在20世纪90年代初期，这些外强迫因素都可能成为20世纪90年代初期东亚夏季风年代际转型的原因。这些外强迫因子在东亚夏季风模态年代际转型中的作用尚不清楚，需进一步研究。更多还原

【Abstract】 Using the NCEP/NCAR and JRA-25reanalysis data from1979to2009, this studyrevealed dominant modes of summer season (June-August)850hPa wind field variabilityover East Asia by means of the complex vector empirical orthogonal function method. Thisthesis deeply analyzed space-time distribution characteristics of the former two EASM modes,and its effect on the precipitation in summer. Moreover, the external forcing exerting possibleeffects on EASM were discussed. The main conclusions are as follows：(1) The leading EASM mode is constituted by two similar sub-modes or their linearcombination, which were revealed by NCEP/NCAR reanalysis data. The real part and theimaginary part of the leading complex principal component can be used as intensity index todescribe two similar sub-modes, which are named M11and M12respectively.The correlationcoefficient between M11and M12are0.77(excluding trend).When M11and M12are in thepositive phase,850hPa wind field anomalies potray a East Asia/Pacific (EAP)teleconnection pattern from the western North Pacific (WNP), across Japan to the OkhotskSea. M11corresponding to anomalous500hPa geopotential height fields show a anomalouszonal teleconnection pattern in northern Eurasia and a EAP teleconnection wave train, whileM12only correspond to an EAP wave train. The distribution of summer precipitationanomalies association with M11and M12both show a significant reduction in the middle andlower reaches of the Yangtze River and north of the Northeast, whereas precipitation over eastof the Northeast、the southeast coast and of the west Southwest increase. In M11and M12positive phase period, sea level pressure (SLP) are significantly lower in the WNP, whileoutgoing longwave radiation (OLR) anomalies in Philippines Sea and South China Sea (SCS)are significantly negative, the convection heating enhance. JRA-25reanalysis data showsimilar results.(2)M11is closely related to ENSO in the previous winter. When M12is in its positivephase, in the previous winter, tropical India Ocean (TIO) and tropical Pacific SST anomaliescoexist. Meanwhile ENSO is in the most robust stage, TIO SST anomalies also exceed95%significance level. Late spring, ENSO decay rapidly, whereas the TIO SST anomalies peak. In summer, ENSO vanished, while TIO SST anomalies sustain the influences on the EASM.JRA-25reanalysis data also show similar results. Using partial correlation analysis, weconcluded the tropical Pacific SST and TIO SST anomaly basically developed independentlyin the previous season, moreover they both exerted their influences on EASM。(3) The second EASM mode revealed by two sets of reanalysis data underwent oneinter-decadal shift in the early1990s, which were consistent with the inter-decadal shift ofChinese summer rainfall in the early1990s. The second EASM mode is closely related to themid-high latitude atmospheric circulation anomalies, corresponding anomalous500hPageopotential height fields show a anomalous quasi-zonal teleconnection pattern in northernEurasia, whereas the distribution of summer precipitation anomalies correspond to ameridional dipole pattern. Along with inter-decadal shift of EASM, after the early1990s themajority of summer precipitation in northern China decreased, especially in north of theNortheast and the area around105oE between the Yangtze and Yellow River region; whilesummer precipitation increased significantly in South China and the Huaihe River Basin.From the perspective of dynamic, the characteristics of inter-decadal shift of Chinese summerprecipitation are described. The difference distribution of summer500hPa geopotential heightfields between two period (1993–2009and1979–1992) show northern Eurasian zonalteleconnection pattern, then the difference distribution of summer850hPa wind fields showthis structure that there are two anomalous anti-cyclonic circulations in southeast of LakeBaikal and south of Japan, while there are two anomalous cyclonic circulations in southernChina and sea of Okhotsk.(4) We discuss the reasons for inter-decadal shift of EASM. Sea surface temperature(SST) in the northwestern Pacific, north Indian Ocean and part of high latitude ocean（NorthAtlantic and North Pacific）, as well as spring Eurasia snow water equivalent changedsignificantly in the early1990s, spring Arctic sea ice occurred one inter-decadal shift in theearly1990s. These external forcing might be the causes of interdecadal shift of EASM in theearly1990s. The role of these external forcing in interdecadal shift of EASM is unclear andfurther study is essential.更多还原