KMS XINJIANG INSTITUTE OF ECOLOGY AND GEOGRAPHY，CAS
| 中亚极端气温时空演变与大气环流的关系研究 | |
| 冯如 | |
| Subtype | 博士 |
| Thesis Advisor | 于瑞德 |
| 2018-06-05 | |
| Degree Grantor | 中国科学院大学 |
| Place of Conferral | 新疆乌鲁木齐 |
| Degree Discipline | 理学博士 |
| Keyword | 中亚 时空变化 时间变异百分位数 极值分析 奇异值分解 Central Asia spatio-temporal change time-varying percentiles extreme value analysis Singular value decomposition |
| Abstract | 全球气候变暖不仅直接影响极端气温的变化同时也加速了极端气候事件的发生（暴雨、高温、热浪、干旱等），20 世纪中叶以来，极端气候事件的向着更强更频繁的发现发展。中亚地区是一个典型的干旱半干旱地区对气候变化极其敏感，是全球升温最快的地区之一，中亚的极端气候事件造成的损失逐年增加，如何应对极端气候事件的发生，不仅受到社会公众的普遍关注，而且也是气候变化科学研究的前沿问题。开展中亚地区的极端气候事件的研究，可以使我们更清楚地认识中亚地区的极端气候事件的变化规律，提高中亚地区对极端气候事件的预测水平，为区域应对灾害提供科学依据。论文基于中亚 108 个气象站点的 1981-2015 年的最高温和最低温数据和500hPa 位势高度场数据。首先，利用 ETCCDI 推荐的指数对中亚极端气温的时空变化特征进行分析；第二，分析了中亚最高温最低温和气温日较差的时间变异百分位趋势；第三，分析了最高温的年最大值和最低温的年最大值进行非平稳假设下的回归期和回归水平；最后，对中亚夏季最高温、冬季最低温与 500hPa 位势高度场进行奇异值分解，分析了极端气温与位势高度场的相互关系。（1）从 1981 年到 2015 年，TMAX，TMIN 和 DTR 在年尺度上都表现出上升趋势，但最高温上升高于最低温，DTR 在除新疆之外的大部分中亚地区呈现上升的趋势，而在新疆 DTR 呈现下降的趋势。季节上尺度上，TMAX，TMIN 和 DTR 主要在春季和夏季显著上升在，3 月份上升趋势大；暖极端指数（TX90p、TN90p 和 TXx）的上升趋势集中在春季；而冷极端指数（TX10p、TN10p 和 TNn）的趋势变化主要集中在秋季。TMAX、TMIN、TX90p、TN90p、TX10p 和 TN10p 在空间上表现为从中亚东南部（土库曼斯坦）到中亚东部（中国新疆）沿着天山一带呈现显著变化。（2）时间变异百分位趋势的研究结果显示 TMAX 在 3 月份上升趋势大，主要下降趋势发生在1-2月份的下百分位数趋势空间上。从空间分布上来看，在中亚西南部、东部和西部地区呈现上升的趋势；而在中亚北部地区，冬季和夏季最高温呈现明显下降趋势，春季和秋季则呈现明显的上升趋势。TMIN 上升趋势主要发生在 2-3 月份、5月份下百分位数和9-11月份；下降趋势主要发生在1月份的下百分位数趋势空间上。从空间分布上来看，中亚东部地区所有月份的最低温均呈现上升趋势，且上升趋势较中亚其他地区更大；而中亚广大南部地区最低温在春季和秋季呈上升趋势；中亚北部地区则呈现出冬季和夏季夜晚变冷的趋势，春季和秋季夜晚转暖趋势。DTR 在 3-9月份的百分位数趋势空间上呈现大于 1ºC/50a 的上升趋势；下降趋势主要发生在 11月份。从空间分布上来看，DTR 上升模式在中亚西天山一带和中亚西部地区更显著。（3）非平稳假设下极值分析显示 TXx 和 TNx 的回归水平在空间上有较好的地理分区，中亚的西部和西南部以及东部地区在 10a~50a 重现期上具有较高的回归水平。平稳假设和非平稳假设下，对显著上升（分散在中亚西南部和东部边缘）和下降（中亚中心位置）的 TXx、以及显著上升（中亚东部）的 TNx 的 GEV 拟合均是合理的，且非平稳假设具有较小的误差空间；在 TXx 和 TNx 显著上升的站点，随着 TXx 和TNx 的不断上升，不同重现期对应的回归水平也在显著上升；而在 TXx 和 TNx 显著下降的站点，随着 TXx 和 TNx 的显著下降，不同重现期的回归水平也呈现下降趋势。（4）中亚夏季最高温场和冬季最低温场与相应的 500hPa 位势高度场的奇异值分解（SVD）分解第一模态分别解释了总协方差的 89.66%和 75.22%；空间分布型时间系数之间的相关系数分别为 0.64 和 0.5；表明夏季夏季最高温和冬季最低温与相应的500hPa 位势高度场存在密切联系，二者正负相位的转化也和前人研究的西北地区气温突变结果一致。夏季，中亚中部的最高温可能与西伯利亚平原和亚速尔群岛的位势高度场呈负遥相关，与里海周边和阿尔泰山周边的 500hPa 位势高度场呈正遥相关；中亚东部的最高温与里海-黑海周边和阿尔泰山周边的位势高度场呈负遥相关，与西伯利亚平原和亚速尔群岛 500hPa 位势高度场呈正遥相关。冬季，中亚中部、北部和东部的最低温与欧洲中部和蒙古高原的 500hPa 位势高度场呈正遥相关，与亚速尔高压呈负遥相关。两场相互影响在夏季主要为中亚中部、南部和东部偏南地区最高温与里海-黑海附近和中国东北地区的 500hPa 位势高度场相互影响，当里海-黑海附近和中国东北地区的 500hPa 位势高度场呈正相位时，中亚的夏季最高温偏高。在冬季中亚中北部（哈萨克斯坦西部和新疆西北部）的最低温场与中西伯利亚高原和欧洲中部的 500hPa 位势高度场相互影响强烈，当中西伯利亚高原和欧洲中部的 500hPa 位势高度场呈负相位时，中亚东北部冬季最低温偏低，易发生低温灾害。 |
| Other Abstract | Global warming has directly influenced the changes of temperature extremes, andaccelerated the occurrences of extreme temperature events, such as storms, heatwave,drought, etc. Since the 20th century, both the intensity and frequency of temperatureextremes changed obviously. How to cope with extreme temperature events has becomenot only a focus of public attention, but only a preceding problem in climate changeresearch. Central Asia (CA), one of the areas experiencing rapid global warming, is verysensitive to climate change. Research on the temperature extremes in CA can helpunderstand the change characteristics of temperature extremes, enhance the forecastingability for extreme temperature events, and provide scientific references for handlingtemperature-related disasters. Based on the maximum temperature (Tmax) and minimumtemperature (Tmin) of 108 stations in CA from 1981~2015, and the global 500hPa height,this paper analyzed the spatio-temporal characteristics of temperature extremes changes inCA by using 11 indices recommended by ETCCDI (Expert Team on Climate Change andIndices), explored the trends in time-varying percentiles of monthly Tmax, Tmin and DTR(diurnal temperature range), assessed the return periods and return levels of annual TXx(warmest days) and TNx (warmest nights) under stationary and non-stationary assumptions,and discussed the teleconnections between temperature extremes and 500hPa height byusing SVD (singular values decomposition). The major conclusions are as follows:(1) From 1981 to 2015, Tmax, Tmin and DTR all showed increasing trends on annualscale, but Tmax increased faster than Tmin, thus leading to an increase in DTR. DTRincreased in most stations in CA, except for those in Xinjiang (China), where DTR mostlyexhibited reversely significant decreasing trends. Seasonally, Tmax, Tmin and DTR mainlyincreased in spring and summer; most warm temperature extremes (TX90p, TN90p andTXx) increased in spring, while most cold temperature extremes (TX10p, TN10p and TNn)increased in Autumn. Tmax, Tmin and the percentile-based warm and cold extreme indicesall exhibited significant trends from southwestern CA (Turkmenistan) to eastern CA(Xinjiang) via the Tianshan Mountains in CA.(2) According to the percentile trends for Tmax, Tmin and DTR, Tmax increased mostin March, and decreased most in January and February at the lower half of the distribution.Spatially, southwestern, eastern and western CA had Tmax increased; while in northern CA,Tmax increased in spring and autumn and decreased in summer and winter. For Tmin, theincrease mainly occurred in February and March, lower half of the distribution in May, andSeptember to November; while the decrease mainly occurred in the lower half of thedistribution in January. Spatially, Tmin in eastern CA exhibited increase trends in allmonths; Tmin in vast southern CA increased in spring and autumn; while in northern CA,Tmin decreased in summer and winter and increased in spring and autumn. DTR increasedin March to September with increase rates over 1ºC/50a, while decreased in November.Spatially, the increase trend of DTR was more obvious in western Tianshan Mountains andwestern CA.(3) Extreme value analysis under non-stationary showed there were clear spatialdifferentiations of TXx and TNx return levels, which were relatively high in western,southwestern and eastern CA. For TXx presenting significant increasing trends anddecreasing trends, and TNx presenting significant increasing trends, GEV simulations werereasonable under both stationary and non-stationary assumptions, and those undernon-stationary had less simulation errors. In stations with significant increasing TXx orTNx, return levels increased with the increase of TXx and TNx; in stations with significantdecreasing TXx or TNx, return levels decreased with the decrease of TXx and TNx.(4) The first modes of SVD of summer Tmax and winter Tmin with the corresponding500hPa height explained 89.66% and 75.22% of the total covariance, with the time seriescorrelations of 0.64 and 0.5, respectively. This indicates summer Tmax and winter Tminwere closely related with the corresponding 500hPa height. In summer, Tmax in middleCA was affected negatively by 500hPa height in Siberian Plain and Azores Islands, butpositively by 500hPa height in Caspian Sea and Altai Mountains; while Tmax in easternCA was affected negatively by 500hPa height in Caspian Sea and Altai Mountains, butpositively by 500hPa height in Siberian Plain and Azores Islands. In winter, Tmin in middle, northern and eastern CA was affected positively by 500hPa height in middleEurope and Mongolian Plateau, but negatively by 500hPa height in Azores Islands. Thekey interaction area was middle, southern and eastern by south CA for summer Tmax with500hPa height in Caspian Sea-the Black Sea and northern China, and middle and northernCA for winter Tmin with 500hPa height in Middle Siberian Plateau and middle Europe; theinteractions were both positive. |
| Subject Area | 自然地理学 |
| Language | 中文 |
| Document Type | 学位论文 |
| Identifier | http://ir.xjlas.org/handle/365004/14925 |
| Collection | 研究系统_荒漠环境研究室 |
| Affiliation | 中国科学院新疆生态与地理研究所 |
| First Author Affilication | 中国科学院新疆生态与地理研究所 |
| Recommended Citation GB/T 7714 | 冯如. 中亚极端气温时空演变与大气环流的关系研究[D]. 新疆乌鲁木齐. 中国科学院大学,2018. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment