秋季土壤湿度和积雪异常对冬季AO及NAO变率的影响研究

发布时间：2018-02-24 15:09

本文关键词： 土壤湿度雪盖北极涛动北大西洋涛动可预报性　出处：《南京大学》2017年硕士论文　论文类型：学位论文

【摘要】：除海洋外,陆地表面是气候系统中的另一重要下边界,对大气和气候变化也有着显著的影响。许多研究表明前期陆地下垫面状态异常尤其是地表水状态(土壤湿度及雪盖)异常与后期天气气候异常有着显著关联和影响。目前,土壤湿度反馈作用研究多集中在区域层面上,主要讨论土壤湿度异常对局地气温和降水的影响,而较少讨论土壤湿度对大尺度大气环流的可能影响;过去已较为详细讨论秋季欧亚雪盖异常对冬季北极涛动的显著影响,但考虑到秋季欧亚积雪在过去几十年中一直有增加趋势,雪盖对北极涛动影响的关键区域及关键时间是否发生改变是我们的关注重点之一;过去对土壤湿度和雪盖的研究多侧重两者的单独作用,较少检查两者对大尺度大气环流的综合作用。北极涛动(AO)和北大西洋涛动(NAO)是北半球冬季大气环流最主要两个模态,对北半球气候变率有显著影响。本文主要展开以下三方面的研究(1)秋季土壤湿度对冬季AO/NAO的影响,(2)秋季欧亚雪盖对冬季AO、NAO影响的关键区域及关键时间的变化,(3)秋季土壤湿度和雪盖异常对冬季AO的协同可预报性。本文选用ERA-Interim再分析的土壤湿度资料、美国冰雪中心的雪盖卫星观测资料及NCEP再分析的海平面气压场资料。研究方法如下:(1)对秋季土壤湿度或积雪两个预报因子的异常场与北半球或北大西洋冬季海平面气压场进行奇异值分解(SVD),找出两个因子影响冬季AO/NAO的关键区域;(2)利用上述关键区域预报因子的异常场和北半球或北大西洋海平面气压场采用交叉验证的SVD方法建立一元及多元线性回归模型,对冬季观测AO及NAO指数进行单独回报和联合回报,计算出AO及NAO指数的预报值与实际观测值之间的典型相关系数(ACC)。论文有以下三点主要结论:第一,1979/80-2014/15年SVD分析结果显示秋季中西伯利亚高原和部分西西伯利亚平原地区土壤湿度的正异常与冬季DJF正位相AO以及JFM正位相NAO显著关联;进一步SVD交叉验证模型的结果显示10、11月以上区域土壤湿度对DJF AO指数回报的ACC分别为0.42及0.54,而1O、11、12月以上区域的土壤湿度对JFM NAO指数回报的ACC分别为 0.35、0.46 及 0.45。第二,为了探讨雪盖趋势变化对冬季AO及NAO环流作用的关键区域及关键时间的影响,本文分 3 个时段(1972/73-2001/02 年、1977/78-2006/07 年、1982/83-2011/12年)来计算秋季雪盖趋势增加对冬季AO及NAO环流影响及其可预报性的改变。SVD分析结果显示,在这3个时段中,10月雪盖对冬季AO及NAO环流影响的关键区域均在60°N左右的西伯利亚及东欧平原北部地区,11月雪盖对冬季AO及NAO环流影响的关键区域均在50°N左右的蒙古高原及东欧平原南部,但是关键区域的范围及变率强度在3个时段有所不同。同时,在这3个时段中,10月雪盖对冬季AO(NAO)环流影响的显著性从100%(96%)逐渐降低至92%(84%),对应回报的ACC从0.43(0.38)逐渐降低至0.19(0.20);而11月雪盖对冬季AO(NAO)环流影响的显著性从92%(70%)逐渐上升至100%(100%),对应回报的ACC从0.35(0.19)逐渐上升至0.58(0.50)。因此,在1972/73-2011/12年这40年间,秋季雪盖对冬季AO及NAO可预报性影响的最优时间及大值变率区域均发生了改变,可预报性的最佳月份均从10月逐渐转变为11月,且雪盖的大值变率区域也均从60°N左右的西伯利亚及东欧平原北部地区南移至50°N左右的蒙古高原及东欧平原南部。第三,由于秋季土壤湿度和积雪的变化具有相当的独立性,同时利用秋季土壤湿度与雪盖异常信息比考虑单独影响因子能够显著提高DJF AO的回报效果。11月土壤湿度及11月雪盖的综合运用对1979/80-2014/15年冬季AO回报的ACC达0.67,能够较好地反演出实际AO指数在观测中产生的异常振幅及低频变动。我们的研究一方面揭示了秋季欧亚土壤湿度异常对北半球冬季大尺度大气环流的显著关联和重要的可预报性,另一方面揭示了由于秋季欧亚积雪增加导致欧亚雪盖对北极涛动影响的关键区域及关键时间的变化,有助于掌握积雪大尺度气候效应的规律,从而为冬季气候提供了具有一定意义的预报新角度和理论基础。
[Abstract]:In addition to the oceans, the land surface is another important boundary in the climate system, the atmosphere and climate change have significant effect. Many studies show that the abnormal state of land surface especially state of surface water (soil moisture and snow cover anomaly) and weather late abnormal have significant correlation and influence. At present, soil moisture feedback studies focus on the regional level, mainly discusses the influence of soil moisture anomaly on air and precipitation, and less to discuss the possible effect of soil moisture on the large scale atmospheric circulation; the past has detailed discussion of autumn Eurasian Snow Cover Anomaly had significant impact on the Arctic Oscillation, but considering the fall of Eurasia the snow in the past few decades there has been increasing trend of snow cover on the key region and key time of the Arctic Oscillation influence whether the change is one of our focus on the past; Separate effect of soil moisture and snow cover on the two, two fewer checks comprehensive effect on the large scale atmospheric circulation. The Arctic Oscillation (AO) and the North Atlantic Oscillation (NAO) is the main Winter Atmospheric Circulation in northern hemisphere two modes, on the Northern Hemisphere climate variability has a significant impact. This paper study the following three aspects (1) effects of soil moisture on the autumn winter AO/NAO, (2) the fall of Eurasia snow winter AO, change the key region and key time on NAO (3), autumn soil moisture and snow cover anomalies on the winter AO collaborative predictability. This paper selected soil moisture data analysis ERA-Interim, SLP data analysis of satellite data and NCEP snow snow America Center. The research method is as follows: (1) abnormal field and the northern hemisphere in autumn two soil moisture or snow forecast factor or the North Atlantic winter Quaternary sea-level pressure field by singular value decomposition (SVD), to identify the two key factors affecting regional winter AO/NAO; (2) the key factor prediction of regional anomalies and the northern hemisphere or the North Atlantic sea surface pressure field using cross validation method to establish a SVD element and a multiple linear regression model, separate returns and return to winter observations of AO and NAO index, calculated AO and NAO index forecast canonical correlation coefficient between the actual and observed values (ACC). This paper has three main conclusions: first, 1979/80-2014/15 SVD analysis of positive anomaly in winter and the positive phase of DJF AO and JFM NAO showed significant positive phase correlation the soil moisture in the autumn of Siberia plateau and the west part of Siberia plain; further SVD cross validation model showed that 10,11 months of regional soil moisture return on DJF AO index ACC respectively. 0.42 and 0.54, and 1O, 11,12 months of regional soil moisture returns on JFM NAO index ACC were 0.35,0.46 and 0.45. second, in order to investigate the influence of snow cover changes of winter AO NAO circulation and the role of the key areas and key time, this paper is divided into 3 periods (1972/, 73-2001/02, 1977/78-2006/07, 1982/83-2011/12 autumn snow cover years) to calculate the trend of increasing impact on the winter AO and NAO circulation and can change the results of.SVD forecast shows that in the 3 period, the northern area in key areas of October winter snow cover on the AO and NAO circulation are at about 60 DEG N in Siberia and the plains of Eastern Europe, southern key regional influence November winter snow cover on the AO and NAO circulation are at about 50 DEG N in the Mongolia plateau and the plains of Eastern Europe, but the range of key areas and the change rate of intensity in 3 different periods. At the same time, in the 3 periods In October, the winter snow cover on AO (NAO) significantly influence the circulation from 100% (96%) decreased to 92% (84%), the corresponding return from 0.43 ACC (0.38) decreased to 0.19 (0.20); and in November the winter snow cover on AO (NAO) significantly influence the circulation from 92% (70%) gradually increased to 100% (100%), the corresponding return from 0.35 ACC (0.19) gradually increased to 0.58 (0.50). Therefore, in the 40 years from 1972/73-2011/12, the optimal time of autumn snow predictability effects on winter AO and NAO and the value of the rate of regional change, can be the best month forecast are gradually from October into November, and the southern snow big value rate region were also from about 60 DEG N in Siberia and the eastern plain of northern South to about 50 DEG N in the Mongolia plateau and the plains of Eastern Europe. In third, due to changes in soil moisture and snow fall are independent phase when, at the same time the use of wet soil in autumn And the snow cover anomaly information than separately considering the impact factor can significantly improve the comprehensive use of DJF AO.11 month return effect of soil moisture and snow cover on November winter 1979/80-2014/15 AO return ACC 0.67, abnormal amplitude and low-frequency changes can better show the actual anti AO index in observation. Our study reveals a significant association between autumn Eurasian soil moisture anomaly in the northern hemisphere large scale atmospheric circulation in winter and important predictability, on the other hand it due to the increase in autumn Eurasian Snow resulting in Eurasia snow cover on the change of the Arctic Oscillation in key areas and key time effect, helps to grasp the large scale climate effects of snow rules. In order to provide a new perspective of prediction has certain significance and theoretical basis for the winter weather.

【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P461

【相似文献】