【作者】 周慧；

【导师】 章新平；

【作者基本信息】 湖南师范大学 ， 自然地理学， 2019， 硕士

【摘要】 气候变化是全球的热点话题之一。对于气候变化的研究,传统的器测记录仅能提供有限的视角,这在一定程度上制约了对于气候要素的变化和预测研究,而稳定同位素是可以从时间和空间上扩大相关天气气候过程研究的重要工具。降水稳定同位素与气候要素之间存在着相关关系,是定量反演古介质中蕴含的气候信息、及时掌握当前气候波动事实以及预测未来气候变化的重要依据。本文在空间上展现了我国降水中稳定同位素的分布特征,并基于我国36个站逐月降水稳定同位素资料,使用相关分析、聚类分析及逐步回归分析等方法分析我国大气降水中δ¹⁸O与7个气象因子之间的相关关系;在时间上,以典型台站（下称代表站）为例,基于引入水稳定同位素循环的全球气候模型（iGCMs）模拟输出产品数据,综合运用连续小波变换、交叉小波和相干小波变换方法,讨论不同气候背景下代表站降水中δ¹⁸O的年际变化及其与年平均温度和年降水量的相互关系;最后针对目前同位素研究的热点问题,利用长沙站2010²017年的日降水δ¹⁸O数据,分析了该站降水中δ¹⁸O的变化特征及与局地和上游气候变量的关系。结果表明:（1）我国降水中δ¹⁸O的变化具有明显的纬度效应和高程效应。具体而言,东南沿海为我国降水中δ¹⁸O的相对高值集中分布区,随着纬度的增加,降水中δ¹⁸O值逐渐降低,至华北和东北地区,降水中δ¹⁸O约是经向上我国降水中δ¹⁸O的最低值。此外,位于我国地势第一级阶梯上的青藏高原成为我国降水中δ¹⁸O低值的集中分布区。在降水中δ¹⁸O季节差的空间分布上,我国西北、东北及青藏高原北部地区降水中δ¹⁸O的季节变化与温度的变化同步,δ¹⁸O平均值暖半年（4⁹月）高于冷半年（10^翌年3月）,季节差为正值;但在我国西南及东南沿海地区,水汽来源的季节转换对δ¹⁸O的变化产生了重要影响,降水中δ¹⁸O呈现出夏低冬高的季节变化特点,季节差为负值。（2）通过计算我国36个站降水中δ¹⁸O与近地面气温、降水量、大气可降水量、外向长波辐射、500hPa高度风速、Nino 4区海表温度距平以及南方涛动指数之间的相关关系发现,秦岭-淮河一线南北两侧站点降水中δ¹⁸O与气象因子之间的关系差异显著,该线是我国一条重要的降水稳定同位素环境效应分界线。（3）聚类分析的结果表明,我国降水中δ¹⁸O可以分为3个区域,即北部区（包括西北和东北地区）、中部过渡区（含华北及青藏地区）和南部区。其中北部区和中部过渡区的分界线大致与我国西北地区和北方地区的分界线吻合,中部过渡区与南部区的分界线大体与我国北方地区和南方地区的分界线相一致。不同地区控制降水中δ¹⁸O的气象因子存在差异:北部区为温度,中部过渡区为温度和500hPa高度风速,南部区是500hPa高度的风速。对各分区当地大气水线的比较发现,受不同水汽来源及当地气候条件的影响,LMWL的斜率小于8的站点多集中分布在北部区和中部过渡区,斜率大于8的站点主要分布在南部区。（4）对我国具有较长时间序列的6个代表站年降水中δ¹⁸O的Morlet小波变换结果表明,乌鲁木齐站年降水中δ¹⁸O在8a时间尺度下负-正交替的周期约为5年。齐齐哈尔站¹⁸O在12a特征尺度下大致经历了3个富集→贫化的转换期,¹⁸O变化的平均周期约为8年。石家庄站年降水中δ¹⁸O变化的主周期为9a,在该时间尺度下¹⁸O共经历了4个完整的贫化→富集的周期变化,变化的平均周期约为6年。德令哈站在13a特征时间尺度上年降水中¹⁸O变化的周期为9a左右,大约经历了3个富集→贫化的转换期。昆明站在8a的主周期时间尺度下降水中¹⁸O大致经历了5个富集→贫化的转换期,¹⁸O变化的平均周期约为5年。香港站近30年来降水中¹⁸O变化的主周期为6a,大致经历了7个贫化→富集的转换期,¹⁸O变化的平均周期约为4年。（5）交叉小波和相干小波分析的结果表明,6个代表站年降水中δ¹⁸O与年平均温度、年降水量在时频域中均存在不同尺度的共振周期和时滞效应,稳定的温度效应或降水量效应在整个研究时段上并不存在。各站点降水中δ¹⁸O与温度和降水量的相关性也存在差异。其中,乌鲁木齐站和石家庄站年降水中δ¹⁸O的变化与年平均温度和年降水量之间均存在较为密切的关系,且石家庄站δ¹⁸O与降水量的相关关系稍强于它与温度的相关关系;德令哈站年平均温度对该站年降水中δ¹⁸O的影响较年降水量更大;在齐齐哈尔、昆明和香港站,年降水量对年降水中δ¹⁸O的影响更为显著。（6）日时间尺度下,长沙降水中δ¹⁸O的变化呈现暖半年的低值与冷半年的高值交替的特点,且δ¹⁸O与局地温度和降水量之间都不存在显著且稳定的相关关系。与“局地效应”相比,长沙降水中δ¹⁸O的变化能敏感地响应上游关键区降水量的变化。日尺度下δ¹⁸O与上游关键区前期区域平均降水量之间的最大相关系数暖半年变化在-0.79^-0.63之间,均通过0.001的信度检验;冷半年变化在-0.79^-0.38之间,除2012年外,其余年份也都通过0.001的信度检验。说明局地降水中δ¹⁸O的变化存在“上游效应”,且上游关键区的降水对下游地区降水稳定同位素组成的影响较“局地效应”显著。更多还原

【Abstract】 Climate change is one of the global hot topics.As the research on climate change always involves a long period of time and space range,traditional instrumental records can only provide a limited perspective.However,stable isotopes can expand such research in both spatial and temporal scales.It is because the close correlations between precipitation isotopes and climatic variables that stable isotopes in precipitation are often used to reconstruct past climatic information preserved in natural archives,master the fact of climatic fluctuation in time and predict the climate change in the future.In this paper,the spatial variation patterns ofδ¹⁸O in precipitation(δ¹⁸O_p)in China were displayed,and based on the precipitation stable isotopic data at 36 stations,the correlations betweenδ¹⁸O_p and multiple meteorological factors were analyzed by using methods of correlation analysis,cluster analysis and stepwise regression analysis.Meanwhile,based on the output data from the isotope enabled GCMs,the continuous wavelet transform was used to display the temporal variation patterns in annualδ¹⁸O_p at 6 representative stations,the cross-wavelet and coherent-wavelet methods were applied to explore the impacts of temperature and precipitation amount on annualδ¹⁸O_p.We finally used the dailyδ¹⁸O_p data at Changsha station from 2010 to 2017 to analyze the variation characteristics ofδ¹⁸O_p and the relationships betweenδ¹⁸O_p with local and upstream climatic variables.The results showed that,（1）There were obvious latitude effect and altitude effect in the variation ofδ¹⁸O_p in China.Specifically,the values ofδ¹⁸O_p were relatively higher in the Southeastern of China.With the increasing of latitude,they became lower in North and Northeastern China.In addition,the lower values ofδ¹⁸O_p mostly concentrated in the Tibetan Plateau because of its high elevation.In the spatial distribution of the seasonality ofδ¹⁸O_p(△δ¹⁸O_p),it was in agreement with the seasonal change of temperature in the Northwestern and Northeastern of China and the northern part of the Tibetan Plateau,the average values ofδ¹⁸O_p were higher in the warm half-year（from April to September）than that in the cold half-year（from October to the following March）,so the△δ¹⁸O_p displayed positive values.While in the Southwest and Southeast of China,the seasonal changes of moisture sources had significant influence onδ¹⁸O_p,the average values ofδ¹⁸O_p were lower in summer than that in winter,so that the△δ¹⁸O_p showed negative values.（2）After calculating the relationships of monthlyδ¹⁸O_p at 36 stations in China with seven meteorological variables including near-surface temperature,precipitation amount,atmospheric precipitable water,outgoing longwave radiation and wind speed at 500hPa,as well as with sea surface temperature anomaly in Nino 4 and southern oscillation index,we found that there were significant differences in the relations betweenδ¹⁸O_p and the seven meteorological variables to the north and south sides of the Qinling Mountains-Huaihe River line,an important line dividing the different environmental effects on the precipitation stable isotopes in China.（3）Based on the cluster analysis,variations of isotopes in precipitation were divided into three zones in China,namely,the north zone including Northwest China and Northeast China,the central transition zone covering North China and the Tibetan Plateau,and the south zone.The dividing line between the north and central transition zones was roughly the same as that between the Northwest and Northern China,while the dividing line between the central transition zone and the south zone aligned well with the boundary of the Northern and Southern China.The meteorological variables controllingδ¹⁸O_p differed among zones.The controlling factors were the temperature in the north zone,temperature and wind speed at 500hPa in the central transition zone,and wind speed at 500hPa in the south zone,respectively.Affected by different moisture sources and local climatic conditions,the slope of Local Meteoric Water Line（LMWL）at stations differ among zones.Stations with the slope of LMWL less than 8 mainly concentrated in the north zone and the central transition zone,while stations with the slope of LMWL more than 8 mostly lied in the north zone.（4）The Morlet wavelet analysis of annualδ¹⁸O_p at 6 representative stations showed that it alternated in a negative and positive way in about 5 years at the 8-year-time scale in Urumchi.The ¹⁸O almost experienced three enrichment-depletion conversion at the 12-year-time scale in Qiqihaer,the average transformation time were about 8 years.The principal period of annualδ¹⁸O_p in Shijiazhuang were 9 years,and¹⁸O nearly appeared four depletion-enrichment conversion under this time scale,the average transformation time were about 6 years.In Delingha,the ¹⁸O showed three enrichment-depletion conversion at the 13-year-time scale,the average transformation time were about 9 years.The ¹⁸O almost experienced five enrichment-depletion conversion at the 8-year-time scale in Kunming,the average transformation time were about 5 years.While in Hongkong,its ¹⁸O appeared seven depletion-enrichment conversion at the 6-year-time scale,the average transformation time were about 4 years.（5）There were different time-scale resonant periods and lag effect between annualδ¹⁸O_p with annual average temperature and annual precipitation amount at 6representative stations.The correlations between them also had differences among stations.There were close relationships between annualδ¹⁸O_p with both annual average temperature and annual precipitation amount at Urumqi and Shijiazhuang stations,but the correlation between annualδ¹⁸O_p with annual precipitation amount was stronger than that with annual average temperature in Shijiazhuang.The annual average temperature exerted more significance on annualδ¹⁸O_p at Delingha station.However,the annual precipitation amount was more important for the variation of annualδ¹⁸O_p at Qiqihaer,Kunming and Hongkong stations.（6）At the daily scale,δ¹⁸O_p exhibited a pronounced seasonal pattern of variation,with lowerδ¹⁸O_p in the warm half-year and higherδ¹⁸O_p in the cold half-year.There were no statistically significant and consistent negative correlations betweenδ¹⁸O_p with local temperature and precipitation amount.Whereas changes inδ¹⁸O_p in Changsha responded sensitively to the variation of precipitation in the key upstream area along air mass trajectories.Year-to-year,the strongest negative lagged correlations（r’）betweenδ¹⁸O_p and the preceding average precipitation amount varied from-0.79 to-0.63（all significant at the 0.001 level）in the warm half-year of 2010-2017.However,in the cold half-year,corresponding r’values varied from-0.79 to-0.38 that were all significant at the 0.001 level,except for the year 2012.The results emphasized the role of upstream effect in explaining the variations of stable isotopes in precipitation in Changsha.更多还原