【作者】 刘芸芸；

【导师】 何金海； 俞永强；

【作者基本信息】 南京信息工程大学 ， 气象学， 2006， 硕士

【摘要】 本文从观测分析和数值模拟两个方面入手分析了热带大气季节内振荡。首先利用1958～1999年ECMWF的era40逐日再分析资料分析热带大气季节内振荡的时空特征变化;进而与IPCC第四次评估报告中的六个耦合模式模拟的热带大气季节内振荡的时空特征进行了对比分析,评估模式对热带季节内振荡的模拟能力,及其全球变暖对热带季节内振荡可能的影响。在研究过程中,运用了各种综合诊断方法（功率谱分析、Lanczos滤波、时空谱分析、自然正交函数分析、合成分析等）,并采取了实测资料和数值模拟相结合的方法,从而对热带大气季节内振荡进行了全面的研究。得到如下主要结论: （1）通过功率谱分析发现,由于大气中的CO₂浓度增加所导致的全球增暖,使得大气低频振荡有分别向更高频率和更低频率两极分化的趋势。 （2）为了定量反映ISO的活动强度,选择热带地区200hPa纬向风季节内变化的方差作为ISO指数,研究发现全球变暖背景下,ISO的活动强度有所加强,ISO指数的变化幅度也增加,说明后20年中ISO活动更加活跃:且ISO季节变化明显,冬、春季强,夏、秋季弱。ISO的年际变化与热带东太平洋的SST关系密切,ISO在El Nino事件发生期间出现减弱的倾向。 （3）通过分析20～100天带通滤波后200hPa纬向风解释总方差的百分比的空间分布及对EOF分解得到的第一特征向量,发现近20年ISO的活跃空间分布变化不大,但在后20年中,印度洋及孟加拉湾地区ISO的强度明显加大,说明随着全球温度的升高,使得印度洋及孟加拉湾地区的ISO活动对整个大气运动来说显得越发重要了。 （4）对200hPa纬向风做空间-时间功率谱计算,反映出ISO以东传的纬向1～2波为主,也有西传成分,以2～4波为主,但其能量值远远小于东传部分,后20年纬向风的能量更集中于1～3波,而小于30天的高频成分有加强的倾向。 （5）通过比较模式对气候平均态的模拟,发现六个耦合系统模式都较准确更多还原

【Abstract】 Both the observed and simulated tropical Intraseasonal Oscillation （ISO） is analyzed and compared in this study. Firstly, the temporal and spatial structures of tropical ISO are characterized by the use of daily mean ECMWF reanalysis data;then the characteristics of ISO simulated by 6 coupled GCMs from IPCC AR4 are compared in order to evaluate the ability of coupled GCMs to reproduce the tropical ISO and explore the possible effect of the global warming on tropical ISO. Various diagnosis methods including power spectra analysis, Lanczos filter, space-time spectra analysis, EOF, composite analysis etc. are applied in this study, and we also compared the simulated ISO with the observed one. Thus, the comprehensive syntheses about tropical ISO are performed in this study. The main conclusions are described as follows.（1）Through power spectra analysis, it is found that the global warming resulted from the increase in concentration of CO₂ enhances ISO activity for both the high and low frequency components.（2）ln order to describe the intensity of ISO quantificationally, the variance of 200hPa zonal wind with 20-100 day filter in the tropical area is defined as the ISO index. The results show that The intensity and the amplitude of ISO index are increased under the global warming background, which indicates that tropical ISO is more active than before, and with more obvious seasonal cycle, e.g., stronger in winter and spring, but weaker in summer and autumn. The interannual variability of ISO is related to the sea surface temperature （SST） anomalies averaged in the Nino3 region of central and east equatorial Pacific. When positive SST anomalies occur, ISO activity is weak.（3）Analyzing spatial patterns of percent of daily variance explained zonal wind at 200 hPa with 20-100 day filter and EOF1, it is found that the spatial distribution of ISO hardly changed during the last 20 years, but the ISO in Indian Ocean and the Bay of Bengal is more active, which indicates that ISO in the Indian Ocean and the Bay of Bengal becomes more important to the general atmosphere motion under the global warming background.（4） The space-time spectra of 200 hPa zonal wind indicate that ISO is mainly eastward propagation waves, in which the energy of ISO with wave number 1-2 is the largest. There are also has westward propagation waves with wave number 2-4 mostly, but the energy of westward propagating ISO are much weaker. The energy of ISO mainly concentrates in the wave number 1-3, and the frequency of ISO tends to increase.（5）Through the intercomparison of climatological mean state by 6 coupled GCMs, it is shown that all the coupled models have accurate description of the characteristics of the space structure of 200hPa zonal wind, which is a prerequisite for producing a realistic intraseasonal oscillation. Analyzing the period, the intensity, propagation and the spatial structure of tropical ISO, it is found that the coupled models captures the salient features of ISO in the tropics, especially in the period, the seasonal cycle and the spatial structure of ISO. However, there are also placesome systematic errors needed to be improved. For example, all the models underestimates the strength of ISO, and the simulated relationship between interannual variability of ISO and El Nino is not remarkable such as observed.更多还原