渤、黄、东海夏季环流的三维斜压模型

基于拉格朗日时均观点描述环流，建立起潮流与准定常流共同占优势系统中的陆浅海环流模型，并诊断计算了夏季渤、黄、东海的三维环流图。模拟结果较好地再现了渤、黄、东海主要流系的特征。对照冬季结果，对渤、黄、东海环流的季节变化做了阐述。从环流垂向分量的分布图上，可发现渐闽近海、长江口外存在较明显的上升流区。另外，对夏季渤、黄、东海的热盐环流和潮致余流分别进行了模拟，发现它们均能在黄海构成一逆时针向的五流系统，这对形成和维持夏季黄海冷水团的存在有重要作用。热盐环流的模拟结果表明，黄海冷水团环流含有“热成流”的成分；通过Lagrange余流的计算发现环绕黄河冷水团的环流还含有“潮成流”的成分。     

菲律宾以西海域的高温暖水与南海夏季风爆发

为了揭示高温暖水在中国南海（文中简称南海）夏季风爆发中所起的作用,依据欧洲中期天气预报中心发布的第5代全球大气海洋再分析资料,发现气候平均意义下印度洋—太平洋暖池中30℃以上高温暖水会在5月出现移位：5月上旬高温暖水出现在孟加拉湾中部,而到下旬消退并移位到南海南部。通过分析局地天气尺度的海洋-大气相互作用过程,揭示了上述高温暖水月内移位的物理机制：在孟加拉湾夏季风爆发后,逐渐增强的潜热释放和减少的短波辐射会导致孟加拉湾高温暖水的面积逐渐缩小;与此同时,在副热带高压影响下,南海菲律宾岛西南高温暖水出现,并因其面积逐渐增大,并与泰国湾的高温暖水共同构成了南海南部的高温暖水。研究发现南海季风爆发几乎都出现在上述高温暖水移位之后,因此孟加拉湾中部和南海南部海表温度的差由正转负可以作为南海季风爆发的先兆。

     

辽宁降水分区变化特征及夏季降水影响因子分析

根据1961—2000年辽宁地区降水资料,将辽宁地区降水划分为4个区,分析各分区降水的年际、年代际、季节变化以及周期特征。结果表明:辽宁南部和东南部地区降水明显减少;自20世纪90年代开始,东南部和南部地区降水显著减少,而中部地区降水增加;东南部地区的夏冬季和南部地区的冬季降水减少,西部地区春季降水增多。周期特征为春季和冬季周期为2 a或4 a,夏季和秋季为10 a。各分区夏季降水的影响因子不同,辽南地区多雨年东亚槽明显加深,东南区多雨年东亚槽略有加深并且偏东;南部和东南部夏季降水与副热带高压系统相关显著;各区夏季降水与太阳黑子负相关较好,其中8月相关显著,南部和中部相关性均好于东南部和西部地区。     

欧亚地区夏季大气环流年际变化的关键区及亚洲夏季风的关联信号

本文利用资料分析和数值模拟方法研究了欧亚地区夏季大气环流的相关性及其与亚洲夏季风的关联信号,以期为欧亚地区的气候变异及可预测性研究提供科学依据.结果表明:欧亚区域同期(JJA)500 hPa高度场年际变化的关键区包括热带区、中纬度的贝加尔湖和巴尔喀什湖之间以及欧洲地中海附近地区;表面气温的关键区主要位于热带海洋;海平面气压的关键区包括热带的海洋性大陆区域、印度洋和非洲大陆赤道附近部分区域、中高纬的贝加尔湖与巴尔喀什湖之间的地区.另外,夏季大气环流年际变化的春季关键区明显西移/南退,特别是表面气温(其西太平洋区不再是关键区).公用气候系统模式CCSM4.0的大气模式在给定海温年际变化的情况下对于上述大气环流相关场及其关键区的模拟基本合理,其中500 hPa高度场的模拟结果较好,海平面气压场的结果逊之;对于同期和前期的结果,模式都有夸大西太平洋海温影响的倾向.对于东亚夏季风指数与大气环流的同期年际变化信号而言,其空间分布基本表现为以30°N为界呈西南东北向的波列状分布;其春季前期信号中,30°N以南的显著区几乎都位于海洋,30°N以北主要位于欧洲、巴尔喀什湖与贝加尔湖之间的地区.南亚夏季风指数的前期显著相关区比同期明显西移/南退.总之,模式的模拟结果和观测结果相当吻合,但其同期模拟结果比前期的更好一些.这些结果说明:模式对于大气环流年际变化的耦合变化信息的刻画是基本合理的,这为利用气候模式进行有关可预测性研究和降尺度预测研究奠定了基础.     

El Ni?o对东亚夏季风和夏季降水季节内变化的影响

基于1979~2012年候平均再分析资料,合成分析了El Ni?o对东亚夏季风和夏季降水季节内变化的影响。结果表明,在El Ni?o衰减年夏季,西太平洋副热带高压(副高)明显偏强,位置偏向西南。副高的这种异常特征随夏季的季节进程有明显变化,初夏异常较弱,盛夏期间异常达到最强。此外,根据东亚夏季风降水呈现阶段式北进的特征,将夏季分为华南前汛期、江淮梅雨期、华北和东北雨期以及华南后汛期来分析东亚夏季风和降水的季节内变化。在上述各个时期,大气对流层低层表现为一致的环流异常型,副高及其以南区域为异常反气旋,其北部为异常气旋。这种异常环流型加强了副高南部偏东风及其北部偏北风,增强了热带水汽输送和高纬度地区冷空气的入侵,二者结合造成主汛期地区降水增加。需要强调的是,上述环流异常型随东亚夏季风逐步向北推移,导致东亚各地区的主汛期降水增加,非主汛期降水减少,降水分布更为集中。     

夏季亚洲高空急流纬向非对称变异与北大西洋海温和欧亚陆面热力异常的可能联系

基于1979~2015年ERA-Interim再分析资料,分析了夏季亚洲高空急流纬向非对称变异特征及其可能的外强迫因子。研究发现夏季亚洲200 hPa纬向风异常EOF第二模态（方差贡献为16.4%）主要表现出了急流纬向非对称的空间异常形态,反映了西亚和东亚区域急流南北偏移的反位相变化。通过进一步的诊断分析,我们发现急流纬向非对称变异与北大西洋海表温度（简称海温）和欧亚陆面热力异常可能存在一定的联系。北大西洋三极型海温异常会激发出向下游传播的异常波列,夏季该波列在欧亚大陆上空的异常环流中心与急流纬向非对称相关的异常环流中心对应一致,其中东欧平原的异常反气旋和巴尔喀什湖附近的异常气旋对西亚急流变化存在影响,东亚地区急流的变化与贝加尔湖北部异常气旋和贝加尔湖南部的异常反气旋有关。对比欧亚土壤湿度关键区内垂直环流,陆面热力异常可能会改变局地环流进而影响急流变异,且这种影响存在区域差异。     

北大西洋涛动和南半球环状模不同位相配置对我国南方夏季降水的协同影响

利用我国气象台站观测资料和再分析格点数据,诊断研究了前期春季（4-5月份）北大西洋涛动（North Atlantic Oscillation,NAO）和南半球环状模（Southern Annular Mode,SAM）在不同位相配置下对我国南方夏季降水异常变化的协同影响作用.分析结果表明,在剔除ENSO最强信号影响后,我国南方夏季降水异常分布显著地依赖于前期春季两个较为独立的年际变率主模态（NAO和SAM）位相的不同配置,即降水异常型主要表现为两个因子单独作用的叠加效果.当前春SAM正位相偏强而NAO负位相偏强时,二者的影响呈现协同正效应,我国南方夏季（6-7月份）降水表现为全区正异常,特别在长江中下游及其以南附近地区最为显著;反之,当前春SAM呈偏强的负位相而NAO偏强的正位相配置时,二者的影响呈现协同负效应,对应我国夏季长江中下游地区降水表现为显著负异常.对其可能的影响机理研究表明,在SAM与NAO位相相反情况下,二者均会通过海气相互作用过程影响到热带大西洋北部海温的异常变化,进而形成协同作用,增强北大西洋海温三极子模态异常信号,从而通过欧亚大陆的遥相关波列对东亚夏季风和我国南方降水产生显著影响.相比之下,当SAM与NAO同位相时,可能表现为抵消效应,不利于北大西洋海温三极子发展,从而削弱对我国南方夏季降水的影响,此方面还有待进一步研究.

     

奇异值分解和奇异交叉谱分析方法在华北夏季降水诊断中的应用

利用1965~2000年华北5省市及相邻省73个地面观测站逐月平均降水场及北半球500 hPa高度场、北太平洋海温场资料, 采用奇异值分解 (SVD)、奇异交叉谱 (SCSA) 分析方法, 将华北夏季降水场分别与1月北半球500 hPa高度场、冬季北太平洋海温场进行了诊断分析, 得出奇异向量分布型及相互作用的耦合周期信号。在对前4对奇异向量的分析中发现, 华北夏季降水全区域为正距平时与1月北半球500 hPa高度场PNA遥相关型关系非常密切。ENSO对华北夏季降水的影响确实存在, 但华北夏季降水全区域为正距平时与冬季北太平洋ENSO关系并不明显。同时还找出了华北降水与北半球500 hPa高度、北太平洋海温场相互作用的关键区。在华北各型降水与高度场、海温场关键区相互作用的耦合周期中, 前者以准2~7年振荡为主; 后者则周期较长, 最短周期仍为准2年振荡, 最长周期为准10~11年振荡。以上结论为进一步研究华北夏季降水短期气候预测方法, 提供了参考依据。     

Dynamic Relationship Between Entrainment Rate and Summer Monsoon in the Transition Area of Monsoon in the East of Northwest China

Entrainment rate refers to the ratio of surrounding air quality to air quality involved in rising unit distance, including turbulent entrainment and dynamic entrainment, which are applied to the boundary layer parametrization of convective clouds, the improvement of numerical model, the observation of cloud droplet spectral dispersion and the study of tropical cyclones.Based on the daily data at 07:00 and 19:00 every 10 m of five stations such as Minqin, Yuchong, Pingliang, Yinchuan and Yan'an from May to September during 2006-2016, combined with the daily observation data on the ground, the Entrainment Rates(ER) of different heights were calculated, and the relationships between ER and height in different regions, precipitation as well as monsoon during the monsoon period were further obtained. The main results were as follows:\mathrm{①} The ER was proportional to air temperature and saturated water vapor pressure, but inversely proportional to relative humidity. The relative humidity threshold of cloud was 65%. The higher the relative humidity threshold was, the lower the cloud height of different orders of precipitation was, and the cloud height was higher with the increase of rainfall. \mathrm{②}ER had obvious diurnal changes and regional differences: It was obviously smaller at 07:00 than at 19:00 from ground to 3 km, which weakened with the increase of height in the near surface , but strengthened with the increase of height above 500 m; From small to large, the monsoon affected area, the monsoon swing area and the non-monsoon area were in turn, and there was no regional difference above 3 km. \mathrm{③} ER was closely related to the intensity and property of precipitation in monsoon period. The ER weakened with the enhancement of rain intensity from near ground to below 600 m, but strengthened with the enhancement of rain intensity from 500 m to 2~3 km.From near ground to below 700 m, the ER of stable precipitation was strong, but that of convective precipitation was strong above 700 m. The convective precipitation had big saturated water vapor pressure and strong ER , while the stable precipitation had big saturated water vapor density, rich water vapor but weak ER. \mathrm{④}The relationship between ER and monsoon as well as its duration: From no monsoon to monsoon ER was weakened, the strongest maximum height was also decreasing. There was no significant difference in the duration of ER between the non-monsoon area and the monsoon affected area, but the longer the monsoon swing area lasted in the near ground layer, the smaller the ER was, while the opposite was at 1~2 km in the high altitude. \mathrm{⑤}The relationship between ER and the APO monsoon intensity index showed that: At 07:00, the ER strengthened with height from near ground to below 800 m, but weakened with height above 800 m,and the monsoon intensity was not related to the ER. At 19:00, the ER strengthened with the height near ground but weakened with the height above 300 m, and the stronger the monsoon was, the smaller the ER was. The ER weakened with the decrease of boundary layer height.     

Link between convection and meridional gradient of sea surface temperature in the Bay of Bengal

We use daily satellite estimates of sea surface temperature (SST) and rainfall during 1998–2005 to show that onset of convection over the central Bay of Bengal (88–92°E, 14–18°N) during the core summer monsoon (mid-May to September) is linked to the meridional gradient of SST in the bay. The SST gradient was computed between two boxes in the northern (88–92°E, 18–22°N) and southern (82–88°E, 4–8°N) bay; the latter is the area of the cold tongue in the bay linked to the Summer Monsoon Current. Convection over central bay followed the SST difference between the northern and southern bay (ΔT) exceeding 0.75°C in 28 cases. There was no instance of ΔT exceeding this threshold without a burst in convection. There were, however, five instances of convection occurring without this SST gradient. Long rainfall events (events lasting more than a week) were associated with an SST event (ΔT ≥ 0.75°C); rainfall events tended to be short when not associated with an SST event. The SST gradient was important for the onset of convection, but not for its persistence: convection often persisted for several days even after the SST gradient weakened. The lag between ΔT exceeding 0.75°C and the onset of convection was 0–18 days, but the lag histogram peaked at one week. In 75% of the 28 cases, convection occurred within a week of ΔT exceeding the threshold of 0.75°C. The northern bay SST, T _N , contributed more to ΔT, but it was a weaker criterion for convection than the SST gradient. A sensitivity analysis showed that the corresponding threshold for T _N was 29°C. We hypothesise that the excess heating (∼1°C above the threshold for deep convection) required in the northern bay to trigger convection is because this excess in SST is what is required to establish the critical SST gradient.