【作者】 张敏；

【导师】 田文寿；

【作者基本信息】 兰州大学 ， 大气物理学与大气环境， 2012， 博士

【摘要】 青藏高原特殊大地型的动力热力作用不仅对区域及全球的天气气候有显著影响,而且对全球平流层和对流层物质交换有重要的调节作用。研究青藏高原上空的平流层-对流层物质交换(STE)对于我们更好的理解人类和自然排放的污染物质对区域乃至全球气候变化的影响具有重要指示意义。本文聚焦于青藏高原上空的中尺度对流层顶折卷(Tropopause Fold)事件,用中尺度数值模式(WRF),结合NCEP和欧洲中心再分析资料、CALIPSO和AIRS卫星资料、多种地基观测资料以及中国气象站点常规观测资料,研究了发生在青藏高原及其周边区域的对流层顶折卷的细节以及与折卷有关的STE特征,分析了该区域不同强度对流层顶折卷的多年气候学分布特征,并在此基础上进一步模拟诊断了影响对流层顶折卷的主要因子。所得主要结果和结论如下：(1)论文首先利用中尺度模式(WRF)模拟了发生在青藏高原及其周边区域(25°N-45°N,70°E-110°E)的两次对流层顶折卷个例。模拟结果表明,WRF模式能够很好地捕捉、重现对流层顶折卷过程,而模式中积云参数化方案的选取对折卷过程的模拟影响不大。与NCEP再分析资料相比,高分辨率的模式模拟结果能更好的反应对流层顶折卷时空变化特征,尤其在复杂下垫面上空,对对流层顶折卷事件的详细细节体现更为完整。模式模拟的水汽混合比和位势涡度场都能很好的体现对流层顶折卷的形态与结构特征,但是折卷的形态和结构在水汽场中比在位势涡度场中体现的更加清晰,而由地形所导致的对流层顶畸变则在位势涡度场中体现得更好。通过对两次对流层顶折卷事件的对比分析发现,如果折卷事件受到高原环流系统的直接影响而不是由高空天气系统生成,折卷事件发展演变会有显著不同,由折卷引起的穿越对流层顶物质输送量也差别很大。受高原环流系统影响的对流层顶折卷事件持续时间相对较长,穿越对流层顶物质输送量也明显较大。高原地形对于对流层顶折卷引起的穿越对流层顶的物质输送有显著影响。当折卷过程从高原上空移过后,由于平流层干空气的快速下沉侵入以及大气动量的垂直输送,会在折卷系统的后方产生一个对流层中层的湿空气层,同时在山体背风坡形成一个低空急流,两者相互作用使背风坡大气的不稳定性加强从而容易产生深厚的垂直上升运动,将对流层物质直接带入平流层低层。此外,当气团被地形强迫抬升到折卷形成的不稳定湿空气层高度,就会激发深对流活动,进而在山体迎风坡产生持续稳定的对流层向平流层的物质输送。值得指出的是,如果折卷受到高空锋面的影响,则会在高空锋面上产生持续的穿越对流层顶向上的物质输送,该输送量大于地形强迫抬升所产生的向上物质输送量。(2)论文进一步利用ERA-Interim再分析资料和中国753个常规气象观测站点的日平均温度资料,通过Q矢量散度极大值和PV极大值配合的判别方法,分析了1989至2009年间发生于青藏高原及其周边区域的对流层顶强、弱折卷事件的气候学分布特征及可能成因。通过大量折卷个例的对比验证,发现用Q矢量散度极大值和PV极大值配合的判别法来对折卷事件进行确认的方法是科学有效的,对同一折卷被重复统计两次的几率不超过5%。分析也表明折卷强度和400hPa等压面上Q矢量散度场极值大小密切相关,极值越大,折卷强度越强。统计分析显示1989至2009年间青藏高原及其周边区域上空的强、弱折卷发生次数都有显著增加的趋势,这一变化趋势可能与高原上空西风急流强度增强,冷空气入侵次数增加及上对流层稳定性降低有关。在强、弱折卷事件的季节分布方面,强折卷的发生次数一年中有明显的两次峰值,分别在冬季风盛行期(12-3月)和夏季风盛行期(6-8月),且冬季强折卷发生频率高于夏季。弱折卷只在夏季风盛行期发生频率极高,但全年弱折卷发生次数高于强折卷。从所有折卷事件的季节分布上看,青藏高原上空对流层顶折卷夏季发生频繁,其他三季发生频率差别不大。进一步对有、无折卷个例以及强、弱折卷个例的高空风场和温度场进行合成分析后表明,在青藏高原上空,高空急流是冬季风盛行期间影响对流层顶折卷发生频率及强度的主要因素；而在夏季风盛行期间高空急流对对流层顶折卷的影响较弱。(3)论文还利用中尺度模式WRF,通过一系列敏感性数值试验诊断分析了地表温度及地形高度等因素对对流层顶折卷的影响。分析表明,把模式中地表温度人为升高3K和6K后,在较短的积分时间内(6小时)对整层大气温度场和高空纬向风场没有显著变化,对对流层顶折卷也没有明显影响。而将模式中地形高度人为降低10%,20%和50%,并与真实地形条件下的模拟结果进行对比,发现青藏高原下垫面高度变化可以通过影响高空风场,间接地对对流层顶折卷产生显著影响,特别是会影响对流层顶折卷在区域内的传播速度。同时地形高度的降低会导致高原上空的重力波信号减弱,进而可能对该地区STE产生影响。(4)本文最后利用1989至2009年的ERA-Interim再分析资料,对青藏高原区域(70(?)-110(?),20(?)-45(?))和西太平洋区域(120(?)-160(?),20(?)-45(?))对流层顶折卷统计特征进行了对比分析,分析结果表明,西太平洋区域与青藏高原区域的冬季折卷都受到副热带西风急流的显著影响。但与青藏高原不同的是,西太平洋区域对流层顶折卷发生次数更多。且西太平洋区域以强折卷发生为主,强折卷发生次数是弱折卷的3倍,造成这种差异的原因和机理值得进行进一步深入的研究。更多还原

【Abstract】 The thermal and dynamical effects of the Tibetan Plateau (TP) have an important impact not only on regional and global weather and climate but also on the global Stratosphere-Troposphere Exchange (STE). Detailed investigations of the STE over the TP is important for understanding the impact of anthropogenic and natural emissions on the regional and global climate. Using a meso scale model (WRF) togather the ERA-Interim of ECMWF (European Center for Medium Range Weather Forecasting) reanalysis data and FNL (Final Operational Global Analysis) of NCEP (National Centers for Environmental Prediction) reanalysis data, satellite observations, various ground based measurements, and surface observations at routine meterologcal stations of China, the detailes of a few tropopause events and climatological characteristics of tropopause fold as well as the associated STE over the TP and its surrounding areas are analyzed in this thesis. In addition, the factors affecting the tropopause fold over the TP are also investigated. The main conclusions are summarized as bellow:(1) To verify the performance of WRF model in simulating tropopause fold events over the TP and its surrounding areas (25-45(?),70(?)-110(?)), the two tropopause events occurred in the middle of April and May2007were simulated. The result shows that the high-resolution meso scale model WRF can well capture the properties of the tropopause fold events and is able to provide more detail of the tropopause fold structures than FNL data, especially over regions with complex terrain. It is also found that the cumulus parameterization schemes have little impact on the tropopause fold simulation over TP. Both the modeled water vapor field and potential vorticity field can capture the morphology and structure of tropopause fold. The tropopause structure is more clearly defined in the modeled water vapor field while the tropopause distortion is more obvious in the modeled potential vorticity field. A comparison of these two tropopause fold events reveals that the fold duration will be longer when a fold is influenced directly by the TP-induced circulation systems and indirectly by the upper level synoptic systems, and the STE induced by fold will be more significant accordingly. The analysis also shows that orography has a significant impact on the cross-tropopause mass exchange. The leeside jet stream and a layer of wet air in the middle troposphere tend to develop when folds passed an elevated surface. The low-level jet was mainly due to downward transportation of momentum from the higher levels associated with the fold and mesoscale descent of air on the lee of the orography. The layer of wet air in the middle troposphere was lifted from the lower troposphere by the descent of dry air from the stratosphere. The leeside jet and the wet layer with high potential instability (PI) can give rise to deep upward motion on the leeside and inject tropospheric air into the lower stratosphere. On the other hand, when the flow encounters an elevated surface, forced lifting together with mid level wet layer can trigger deep convective motion on the windward slope. The troposphere-to-stratosphere transport was found to be persistent and almost stationary over a windward slope of high orography during the evolution of the fold. In particlular, if a fold is affected by the upper level front, the persistent cross-tropopause mass exchange will occur along the frontal surface, and the cross tropopause mass exchange is stronger than that caused by the forced of lifting air over the windward slope.(2) The climatlogical charateristics and factors of impacting on the long-term trend of tropopause fold over the TP during the time period from1989-2009are further analysed. A pattern maching method through seeking for collocated maxima of potential vorticity and Q-vector divergence is used for searching tropopause folds from the reanalysis data. A large number of folds selected by this method are compared and verified by the corresponding folds determined by the other means. The comparison shows that this pattern maching method is robust and scientificaly effective, and the rate of overlaped selections for the same fold event is below5%. The statistical analysis shows that the fold intensity is closely related to the Q-vector divergence on400hPa level with a larger Q-vector divergence corresponding to a stronger tropopause fold. The frequency of both strong and weak folds over the TP shows a increasing trend from1989to2009. This trend is possibly caused by the increased intensity of the subtropical westerly jet, enhanced frequency of cold air intrusion and decreased atmosphere stability in the upper troposphere over the TP. The annual variation of strong fold events have two obvious peaks with one in the winter monsoon season (Dec-Mar) and the other in the summer monsoon season (Jul-Aug). There are more the strong folds in winter than in summer while the weak folds mostly occur in summer monsoon season. The number of weak folds is larger than the number of strong folds over the TP. In addition, the seasonal distribution of tropopause fold events shows that the number of folds in summer over TP is the largest and there are no significant differences in fold numbers in the other seasons. A composite analysis of the wind and temperature fields with and without fold and corresponding fields associated with strong and weak folds is performed. The analysis indicates that the upper level westerly jet is the mainly factor influencing the frequency and intensity of fold during winter monsoon season while in summer monsoon season, the effect of the upper level westerly jet on folds over the TP is not significant.(3) The effects of the surface temperature and topography height on folds over the TP are investigated through a series of WRF sensitivity simulations. The results demonstrate that in a short integration time of6hrs there are little changes in potential temperature field and the upper level zonal wind field, as well as in the tropopause fold when the model surface temperature is arbitrarily increased by3K and6K. Nevertheless, changes in the topography height in the model can significantly influence the propagation speed of folds via changing intensity of upper level jet. Meanwhlie, the lower terrain results in weaker gravity waves, and this also has a possible impact on the STE over the TP and its surronding areas.(4) Finally, the number of tropopause events during the time period from1989to2009over the TP (70°E-110°E,20°N-45°N) is compared to that over the Western Pacific (120°E-160°E,20°N-45°N). The result indicates that the folds over both regions are mainly affected by the subtropical westerly jet in winter. But the number of folds over the Western Pacific is larger than that over the TP. Folds over the Weatern Pacific are mainly strong folds and the number of the strong folds is three time larger than the number of weak one. The possible reason and mechanisms of these differences between the two areas need further study.更多还原