Knowledge Resource Center for Ecological Environment in Arid Area
| 丝绸之路核心区气候变化归因分析及预估 | |
| 其他题名 | Attribution and projection of the climate change in the Core Region of Silk Road Economic Belt |
| 彭冬冬 | |
| 出版年 | 2018 |
| 学位类型 | 博士 |
| 导师 | 周天军 ; 张丽霞 |
| 学位授予单位 | 中国科学院大学 |
| 中文摘要 | 丝绸之路核心区属于生态环境对气候变化十分敏感的干旱区。近五十年，整个丝绸之路核心区显著增温，且升温速率远高于全球平均。位于其东部的中国西北地区降水持续增加，气候由暖干转为暖湿。降水和温度的变化会对该区域水资源及干旱变化具有巨大影响。然而，当前关于丝绸之路核心区近五十年温度和降水变化及其成因的认识仍然不足。因此，本文先基于拉格朗日追踪方法给出了丝绸之路核心区降水事件的水汽来源，再利用水汽收支诊断方法分析了近五十年中国西北地区变湿原因，接下来对整个区域的降水和温度变化开展了检测归因分析，在此基础之上预估了该区域未来百年的气候变化特征，主要研究结论如下：一、各个水汽源地对丝绸之路核心区降水的贡献观测分析表明，丝绸之路核心区以75E为界，东、西部的干、湿季节刚好相反，东部的干季为冷季（当年11月至次年4月）、湿季为暖季（5-10月），西部的干季为暖季、湿季为冷季。据此，基于丝绸之路核心区东、西部的冷季和暖季分别挑选多个强降水事件，利用拉格朗日模式进行水汽追踪。结果表明，丝绸之路核心区的水汽源地为局地、欧亚大陆、大西洋、北非和印度半岛。丝绸之路核心区东部和西部各个季节降水事件的主要水汽来源均为局地蒸发，其对丝绸之路核心区东部和西部暖季降水的贡献分别为60%和57%，对冷季降水的贡献为45%和48%。欧亚大陆的水汽输送对丝绸之路核心区降水事件的贡献次之，该输送对东部和西部暖季降水的贡献分别为29%和31%，对冷季降水的贡献依次为34%和24%。此外，源自北非的水汽输送主要对丝绸之路核心区东、西部的冷季降水有贡献，分别为8%和24%，而印度半岛的水汽输送对核心区降水的贡献较小且主要集中在冷季（4%）。大西洋地区的水汽输送对核心区的降水贡献最小，不超过2%。二、近五十年中国西北夏季降水增加的原因基于中国台站及格点化观测资料的分析表明，位于丝绸之路核心区东部的中国西北地区近五十年夏季降水显著增加了14.56%。因此，本文基于JRA55再分析资料，利用水汽收支诊断方法揭示了近五十年中国西北夏季降水增加原因。结果表明，夏季气候平均的降水和蒸发基本相当，水汽垂直平流项同水平平流项的作用相互抵消。就长期趋势而言，西北地区降水减蒸发也呈显著增加趋势，且主要源于水汽垂直平流项的贡献。进一步的分析结果表明，降水减蒸发的增加主要源于水汽垂直平流项的热力贡献（同水汽变化有关），其次为动力贡献（同环流变化有关）。其主要机制如下：该区域大气温度升高，有利于蒸发增强，会使该区域上空大气可降水量增加（即热力贡献）；另一方面，近五十年夏季亚洲副热带西风急流显著南移，该区域上空有水平涡度平流正异常而引起了上升运动（即动力贡献），有利于该区域降水增加。三、人为活动使丝绸之路核心区近五十年夏季降水显著增加上述基于再分析资料的诊断结果有助于理解中国西北夏季降水增加的物理过程，却难以回答这一增加趋势究竟是由自然因子还是人为因素引起。与此同时，基于全球台站和多套格点资料的分析结果表明，近五十年夏季降水增加趋势不止局限于丝绸之路核心区东部（即中国西北），还可以扩展至核心区西部。观测中，整个丝绸之路核心区近五十年夏季降水显著增加了14.15%。本研究基于CLIVAR C20C+检测归因计划中CAM5.1模式的全强迫（CAM5-All）和自然强迫（CAM5-Nat）试验，针对整个丝绸之路核心区近五十年夏季降水增加趋势开展了检测归因。结果表明，在考虑了人为强迫的CAM5-All试验中合理的再现了该增加趋势，而仅考虑了自然强迫的CAM5-Nat试验无法模拟该特征。因此，该区域夏季降水增加可归因为人为活动的贡献，且人为活动使得该区域近五十年夏季降水增加了16.10%。基于JRA55再分析资料的水汽收支诊断分析表明，水汽垂直平流项的动力贡献主导了整个丝绸之路核心区的降水增加，而热力贡献主要令核心区东部降水增加。CAM5-All可以再现JRA55的分析结果，而CAM5-Nat却不能。进一步的分析结果表明，人为活动使得欧亚大陆对流层升温不均匀，高纬地区升温快于低纬地区。这一增暖通过引导丝绸之路核心区上空的西风急流南移，进而引起南风异常和暖平流异常输送，诱发局地异常的上升运动，有利于丝绸之路核心区夏季降水增加。四、人为活动使丝绸之路核心区近五十年显著增暖鉴于温度变化对降水变化的重要作用，有必要对丝绸之路核心区近五十年温度变化进行检测归因。因CMIP5模式历史气候模拟试验截至时间为2005年，当前研究时段设定为1961-2005年。观测事实指出，丝绸之路核心区近五十年的年平均温度显著升高（1.33℃），且该区域增温有明显的季节特征，显著增温主要集中在夏季（0.90℃）、秋季（1.22℃）和冬季（2.48℃），而春季增温不显著。为此，本文主要针对年平均和上述三个季节的显著增温进行检测归因。检测结果表明，利用指纹法可以检测到人为温室气体强迫（GHG）对丝绸之路核心区近五十年增暖的作用。GHG使得该区域年平均温度升高了1.25℃（0.52-2.00℃），使夏季、秋季和冬季温度分别升高了1.11℃（0.32-1.92℃）、1.11℃（0.40-1.83℃）以及2.50℃（0.91-4.34℃）。检测归因结果同时指出，CMIP5模式低估了1961-2005年丝绸之路核心区年平均温度变化趋势，该低估主要源于模式对冬季增暖趋势的低估。模式低估了丝绸之路核心区的历史温度变化，这很可能会使得该区域的未来温度变化被低估。经检测归因结果校正后的模式预估结果表明，21世纪丝绸之路核心区将以更快的速率升温，RCP45和RCP85排放情景下升温速率分别为0.32和0.74℃每十年，排放情景浓度越高升温速率越快。RCP85排放情景下，到21世纪末期丝绸之路核心区温度将升高7.00℃，较校正前高0.89℃。五、全球增温1.5℃和2℃背景下丝绸之路核心区的极端气候指数变化上述结果指出了人为活动对该区域气候变化的影响。因此，未来预估情景下，特别是1.5℃和2℃温升阈值下丝绸之路核心区的极端气候将如何变化亟待回答。为此，本文系统评估了CMIP5历史模拟试验对丝绸之路核心区极端气候指数的模拟能力，并在此基础之上展开了未来预估研究。历史试验分析结果表明，CMIP5模式对四个极端降水指数（最大连续无降水日数CDD、降水强度SDII、日最大降水量RX1day和连续五天最大降水量RX5day）及两个冷指数（日最高气温最小值TXn和日最低气温最小值TNn）的气候态空间分布有很好的模拟能力，而对两个暖指数（日最高气温最大值TXx和日最低气温最大值TNx）的模拟能力偏弱。模式模拟偏差表现为：大部分模式模拟的极端温度指数偏低，极端降水指数RX1day和RX5day偏多、SDII偏强，而模拟的CDD偏长和偏短的模式数各占一半。就空间分布而言，模式模拟的极端气候指数偏差最大区域为青藏高原区域。当分析结果不包括青藏高原时，模式对该区域的极端气候指数模拟能力显著提高，模拟偏差明显降低。为此，后续重点讨论了两种预估情景（RCP45和RCP85）下不包括青藏高原在内的丝绸之路核心区，两类极端气候指数在1.5℃和2℃温升阈值下的变化特征。分析结果指出，RCP45（RCP85）排放情景下21世纪丝绸之路核心区的极端温度指数TNn、TNx、TXn及TXx的升温速率分别为0.48℃（0.95℃）、0.28℃（0.64℃）、0.45℃（0.83℃）及0.29℃（0.64℃）每十年，极端降水指数RX1day、RX5day和SDII增加速率分别为1.05%（2.26%）、1.04%（1.94%）和0.67%（1.45%）每十年，排放情景浓度越高增加速率越快。冷指数升温速率快于暖指数，极端降水RX1day和RX5day增加速率快于极端降水强度。两种排放情景下，CDD变化均不显著。到达全球1.5℃和2.0℃温升阈值时，除CDD变化不显著外，该区域的其它极端气候指数均将显著增加。和1.5℃温升阈值相比，RCP85排放情景下到达2℃温升阈值时该区域的极端温度指数TNn、TNx、TXn和TXx分别将显著升高0.79℃（0.18-1.62℃）、0.65℃（0.41-0.86℃）、0.35℃（-0.21-1.36℃）和0.65℃（0.32-0.89℃），极端降水指数RX1day、RX5day和SDII将分别增加1.05%（-0.74-2.99%）、1.41%（0.09-2.08%）和1.91%（0.93-2.23%），CDD变化不确定性大。 |
| 英文摘要 | The ecosystem and development of society in the Core Region of Silk Road Economic Belt (SRCR) is very sensitive to climate change. During the past five decades, the average temperature in SRCR increased significantly at a rate much higher than that of the global average, and in the eastern part of SRCR (the Northwest China), precipitation has shown a significant upward trend with the climate shifting from warm and dry to warm and wet. The changes in temperature and precipitation have a great impact on water resource and aridity in SRCR. However, the corresponding physical mechanisms remain ambiguous. Thus, detection and attribution studies on the climate change in SRCR are applied in this study. Firstly, the moisture sources of precipitation events in SRCR are identified; secondly, the dynamic processes responsible for the increasing summer precipitation in the eastern part of SRCR are revealed based on the moisture budget analysis; thirdly, the human influence on the changes in precipitation and temperature over SRCR is explored; and lastly, the projected changes in extreme climate indices in SRCR are shown. The main conclusions are listed as follows: 1. Contributions of different moisture source regions to the precipitation events in SRCR. Observation shows that the dry and wet seasons for the western SRCR are opposite to that for the eastern SRCR. In the eastern SRCR, the wet season is from May to October (the warm season) and the dry season is from November to April in the following year (the cold season), while for the western SRCR, the opposite is true. The contributions of different moisture source regions to the precipitation events in warm and cold seasons for both eastern and western SRCR are revealed based on the lagrangian model. Local evaporation is the main contributor to the precipitation events in SRCR. The contribution of local evaporation to the precipitation events over eastern (western) SRCR is 60% (57%) in the warm season and is 45% (48%) in the cold season. The moisture from Eurasia is the second contributor to precipitation events, with a fraction of 29% (31%) in the warm season and 8% (24%) in the cold season over eastern (western) SRCR. The moisture from North Africa (Indian Subcontinent) mainly contributes to the precipitation in the cold season, with a fraction of 8% (4%) for eastern SRCR, and 24% (4%) for western SRCR. The contribution of moisture from the Atlantic Ocean is less than 2%. 2. Mechanisms responsible for the increasing summer precipitation in Northwest China during past five decades.Observations show that summer precipitation in Northwest China has significantly increased by 14.56% from 1961 to 2010. Thus, the moisture budget analysis based on JRA55 dataset is employed to explore the observed increasing trend. The difference between precipitation and evaporation shows a significant increasing trend, and is dominated by the thermodynamic (associated with changes in specific humidity) and dynamic contribution (associated with changes in atmospheric circulation) of the changes in vertical moisture advection, and the former is larger. The air temperature shows a significant increasing trend during the past five decades, which favors more evaporation, resulting in an increase in atmospheric precipitable water (thermodynamic contribution). On the other hand, the horizontal vorticity advection intensifies in association with the significant southward displacement of the Asian Subtropical Westerly Jet (ASWJ), inducing anomalous ascending motion (dynamic contribution) and hence increasing the precipitation. 3. Human contribution to the increasing summer precipitation in SRCR during the past five decades. The above budget analysis based on the JRA55 dataset has shown the dynamic mechanisms responsible for the wetting trend in Northwest China. However, whether this wetting trend is a result of natural or anthropogenic influence remains unknown. Meanwhile, based on the global stations and multi-observed gridded datasets, we found that the increasing summer trend is not limited to the eastern SRCR, but also can be extended to the western SRCR. From 1961 to 2010, the summer precipitation over SRCR has increased by 14.15% based on multi-observational datasets. The corresponding physical processes and contributions of anthropogenic forcing are investigated by comparing reanalysis dataset and experiments of Community Atmosphere Model version 5.1 (CAM5.1) from the CLIVAR Climate of the 20th Century Plus Project (C20C+). The observed wetting trend is well reproduced in the simulations driven by all radiative forcings (CAM5-All) but poorly reproduced in the simulations with natural forcings only (CAM5-Nat). Thus, the observed wetting trend is robustly attributed to human contribution. Moisture budget analysis based on JRA55 shows that the observed wetting trend is dominated by the dynamic contribution of increasing vertical moisture advection term, while the corresponding thermodynamic contribution only increases the precipitation in the eastern SRCR. The observed contributions of moisture budget components to the wetting trend are only captured by CAM5-All experiments. In response to human influence, the troposphere in Eurasia shows an uneven warming pattern, with a higher warming rate at higher latitude, which results in a significant southward displacement of the ASWJ. This results in an anomalous southwesterly which in-turn brings in much warm air, inducing anomalous ascending motion and increasing the summer precipitation. 4. Human contribution to the warming trend in SRCR during the past five decades.Temperature changes play an important role in modulating precipitation changes in SRCR. Thus, the attribution of temperature changes is conducted by using the optimal fingerprinting method. Considering that the CMIP5 historical simulations ended in 2005, here we focus on the period 1961-2005. From 1961 to 2005, the observed annual mean temperature has significantly increased by 1.33℃. The significant warming trend occurs in summer (0.90℃), autumn (1.22℃), and in winter (2.48℃). The changes of temperature in spring are not significant, thus are not discussed in this study. Results show that the influence of anthropogenic greenhouse gases (GHG) on the annual and seasonal significant warming trend in SRCR is robustly detected. From 1961 to 2005, the annual mean temperature has increased by 1.25℃ (0.52-2.00℃) in response to GHG. The contribution of GHG is 1.11℃ (0.32-1.92℃) in summer, 1.11℃ (0.40-1.83℃) in autumn, and 2.50℃ (0.91-4.34℃) in winter. Meanwhile, attribution results indicate that the historical annual mean temperature in SRCR is underestimated by the CMIP5 models, which is mainly caused by the underestimated warming trend in winter. These imply that the projected changes in temperature may also be underestimated. Thus, the CMIP5 projections are corrected based on the attribution results, showing a larger warming rate in SRCR in the 21st century, with a value of 0.32℃ decade-1 under RCP45 and 0.74℃ decade-1 under RCP85. To the end of 21st century, the annual mean temperature in SRCR will increase by 7.00 ℃ under RCP85, which is a further warming of 0.89 ℃ than the uncorrected result.5. Changes in the extreme climate indices over SRCR in a 1.5℃ and 2℃ warmer world.We have shown evidence of human influence on the climate change over SRCR. Therefore, the projected climate changes in SRCR under RCPs (especially in a 1.5℃ and 2℃ warmer world) need to be explored. The performance of CMIP5 models in simulating the mean state of extreme climate indices is firstly evaluated. CMIP5 models show good performance in simulating the spatial patterns for the mean state of four extreme precipitation indices (precipitation intensity on wet days (SDII), maximum 1-day precipitation (RX1day), and maximum consecutive 5-day precipitation (RX5day), maximum number of consecutive days (CDD)), and two cold indices (annual minimum of daily maximum (TXn) and minimum (TNn) temperature), but with low skill for two warm indices (the annual maximum of daily maximum (TXx) and minimum (TNx) temperature). Models tend to underestimate the four extreme temperature indices but overestimate the extreme precipitation indices SDII, RX1day, and RX5day. Half of the models overestimate (underestimate) the CDD. The model bias in simulating the extreme climate indices is mainly from the Tibet Plateau (TP).When the TP region is not included, the skill scores of models in simulating extreme indices increase, and the model biases reduce. Thus, the TP region is not included in the projection of climate change over SRCR. Under RCP45 (RCP85) scenario, the significant increasing rates per decade are 0.48℃ (0.95℃), 0.28℃ (0.64℃), 0.45℃ (0.83℃), and 0.29℃ (0.64℃) for the extreme temperature indices TNn, TNx, TXn, and TXx, respectively, and are 1.05% (2.26%), 1.04% (1.94%), and 0.67% (1.45%) for the extreme precipitation indices RX1day, RX5day, and SDII, respectively, showing higher increasing rates under higher RCPs. However, the changes in CDD are not significant under both RCP45 and RCP85. In a 1.5℃ and 2℃ warmer world, except for the slight changes in CDD, the other seven extreme indices increase significantly over the whole SRCR. With respect to the 1.5℃ warmer world, in the 2℃ warmer world under RCP85, the extreme temperature indices TNn, TNx, TXn, and TXx increase significantly by 0.79℃ (0.18-1.62℃), 0.65℃ (0.41-0.86℃), 0.35℃ (-0.21-1.36℃), and 0.65℃ (0.32-0.89℃), respectively, and the extreme precipitation indices RX1day, RX5day and SDII increase by 1.05% (-0.74-2.99%)，1.41% (0.09-2.08%), and 1.91% (0.93-2.23%), respectively. However, the changes in CDD are not significant. |
| 中文关键词 | 丝绸之路 ; 水汽追踪 ; 检测归因 ; 1.5℃和2℃ ; 极端气候指数 |
| 英文关键词 | Silk Road moisture source detection and attribution 1.5℃ and 2℃ extreme climate indices |
| 语种 | 中文 |
| 国家 | 中国 |
| 来源学科分类 | 气象学 |
| 来源机构 | 中国科学院大气物理研究所 |
| 资源类型 | 学位论文 |
| 条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/288070 |
| 推荐引用方式 GB/T 7714 | 彭冬冬. 丝绸之路核心区气候变化归因分析及预估[D]. 中国科学院大学,2018. |
| 条目包含的文件 | 条目无相关文件。 | |||||
| 个性服务 |
| 推荐该条目 |
| 保存到收藏夹 |
| 导出为Endnote文件 |
| 谷歌学术 |
| 谷歌学术中相似的文章 |
| [彭冬冬]的文章 |
| 百度学术 |
| 百度学术中相似的文章 |
| [彭冬冬]的文章 |
| 必应学术 |
| 必应学术中相似的文章 |
| [彭冬冬]的文章 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
