Knowledge Resource Center for Ecological Environment in Arid Area
| 基于GIPL2模型的青藏高原冻土时空变化模拟研究 | |
| 其他题名 | A Research on the Spatiotemporal Variations of the Permafrost Based on the GIPL2 Model on the Qinghai-Tibetan plateau |
| 秦艳慧 | |
| 出版年 | 2018 |
| 学位类型 | 博士 |
| 导师 | 吴通华 |
| 学位授予单位 | 中国科学院大学 |
| 中文摘要 | 气候变暖导致近地表冻结/融化指数、活动层厚度和冻土温度等发生显著变化。了解青藏高原冻土现状、空间分布特征及其对气候变化如何响应对全球气候变化研究、生态环境研究、水资源管理、寒区工程规划和建设的辅助决策具有重要的科学与实践意义。青藏高原面积广阔，气候条件独特，发育着约1.06×106 km2的多年冻土，多年冻土主要分布在自然环境恶劣的无人区，导致多年冻土实际观测资料匮乏。仅靠有限的观测资料系统地研究青藏高原冻土存在一定困难，模型模拟的诸多优点为大面积长时间序列地研究青藏高原冻土的现状及变化提供了可能。本论文基于Geophysical Institute Permafrost Lab Version2（GIPL2）模型，系统地分析了青藏高原冻土空间分布特征及其对气候变化的响应，得到以下主要结论：1. 通过对青藏高原海拔2000 m以上的69个国家气象站1980 - 2013年空气和地面冻结/融化指数的综合分析，发现无论是在多年冻土区还是季节冻土区，在相同的站点，年平均空气冻结指数都要高于年平均地面冻结指数，年平均地面融化指数要高于年平均空气融化指数。多年冻土区冻结指数的降低速率快于季节冻土区，这意味着青藏高原多年冻土区对气候变暖的响应比季节冻土区更敏感。无论在多年冻土区还是季节冻土区，地面温度比气温对气候变暖的响应更敏感。青藏高原多年冻土区冻结指数的降低速率要快于中国东北地区、蒙古高原和北半球高纬度地区。在多年冻土区，冷季升温显著；而在季节冻土区，暖季升温更显著。2. 在不同海拔梯度订正ERA-interim地面温度后，年平均和季节ERA-interim地面温度与观测值都更为接近。与订正前相比，经过订正后，ERA-interim年平均地面温度的均方根误差（Root Mean Square Error：RMSE）值从7.7 °C降低为1.6 °C，平均误差（MBE：Mean Bias Error）从-7.5 °C变为0 °C，相关系数（correlation coefficient：R）从0.55增加到0.78。与订正前相比，订正后春季、夏季、秋季和冬季的RMSE有所降低的站点分别占到了气象站总数的98.6 %、97.1 %、97.1 % 和92.8 %，结果表明对ERA-interim的订正能够获得更为贴近青藏高原实际情况的地面温度产品，可以作为冻土模型的上边界条件。利用订正后的ERA-interim地面温度产品看区域上青藏高原冻土环境特征，可以看到，春季和秋季地面温度的空间分布特征与年平均地面温度的空间分布特征非常相似。青藏高原地面冻结指数的空间分布特征与冷季地面温度的空间分布特征相似，青藏高原地面融化指数的空间分布特征与暖季地面温度的空间分布特征十分相似。3. GIPL2模型在单点尺度的模拟结果显示：模型在高寒草原试验点（QT06）模拟效果较好，高寒沼泽草甸试验点（QT03）的模拟结果较差，高寒草原（QT01）、高寒荒漠草原（QT05）和高寒草原化草甸（QT04）试验点的模拟效果介于高寒草原和高寒沼泽草甸试验点之间。模型对上边界条件的选取非常敏感，以地面温度为上边界条件的模拟结果更接近土壤热状况的实际情况。4. 单点模型扩展到青藏高原区域后，区域模拟结果表明，青藏高原活动层厚度存在很大的空间差异，活动层厚度随着海拔的升高而降低，并且从东南向西北逐渐降低，平均活动层厚度为2.3 m（2.21 m - 2.40 m）；不同深度地温空间分布与活动层厚度空间分布非常相似，即地温高的地方，活动层厚度也相对较大；反之亦然。不同深度冻土地温都表现出随着海拔的升高而降低，并且不同深度冻土地温呈现从东南向西北逐渐降低的特点。青藏高原所占面积比例最大的是亚稳定型多年冻土，在青藏高原多年冻土和季节冻土交界地方，大多属于过渡型多年冻土和不稳定型多年冻土。1980 - 2013年青藏高原多年冻土区活动层厚度的平均增厚速率为3.1 cm/a，对气候变暖响应敏感，活动层厚度在青藏高原南部的增厚速度要高于青藏高原北部，在青藏高原西部的增厚速度要快于青藏高原东部。 |
| 英文摘要 | Under the background of global climate change, the permafrost environment will changes, such as the near surface freezing and thawing indices, activity layer thickness and the permafrost temperature. Therefore, the knowledge of the current permafrost environmental status, its spatial distribution and response to climate changes on the Qinghai-Tibetan Plateau（QTP）is of great significant to some researches, such as the global climate change research, ecological environment research, water resources management, auxiliary planning and construction. Permafrost area amounts to approximately 1.06 × 106 km2 over the QTP, which is mainly distributed in the harsh area that makes large area observations is difficult to be realized. It is difficult to systematically study the environment of QTP permafrost due to limited actual observation data. However, the advantages of model simulation can provide the possibility of studying the present current status and variation of permafrost environmental for a long time on the QTP. Based on the Geophysical Institute Permafrost Lab Version2（GIPL2）model, the purpose of this paper is to analyze the permafrost environmental characteristics, its spatial distribution and response to climate changes systematically in the permafrost of the QTP, and the main conclusions are as follows:1. Based on the daily ground surface temperature and daily air temperature from 69 meteorological stations, the spatial and temporal variation of ground surface freezing indices, ground surface thawing indices, air freezing indices, and air thawing indices under permafrost regions and seasonally frozen ground regions over the QTP were analyzed using Mann-Kendall test and Sen's slope estimate. The annual ground surface freezing indices was larger than air freezing indices and the annual ground surface thawing indices was larger than air thawing indices in both of the permafrost regions and the seasonally frozen ground regions. Moreover, the decreasing trend of ground surface freezing indices and air freezing indices in permafrost regions was faster than that in seasonally frozen ground regions, the increasing rate of ground surface thawing indices and air thawing indices in the seasonally frozen ground regions was larger than those of permafrost regions. In permafrost regions, the downward trend of freezing indices was more significant various than the upward trend of thawing indices. However in the seasonally frozen ground regions, the upward trend of thawing indices was more acute variation than the freezing indices. The results suggested that in permafrost regions, the warming was more pronounced in the winter than the other seasons, and the summer warming was more pronounced than the other seasons in the seasonally frozen ground regions. Decreasing trend of annual air freezing indices and ground surface freezing indices on permafrost regions of the QTP was pronounced than northeast of China, the Mongolia, and the northern Hemisphere. The results suggested that in permafrost regions, the warming was more pronounced in the winter than the other seasons, and the summer warming was more pronounced than the other seasons in the seasonally frozen ground regions. 2. We calibrated the ERA-interim reanalysis ground surface temperature data using the methods of elevation correction on the QTP. After calibration, the quality of reanalysis data has been improved significantly. For the annual time series, the Root Mean Square Error（RMSE）decreased from 7.7 °C to 1.6 °C , the absolute value of Mean Bias Error（MBE）varied from 7.5 °C to 0.0 °C, and the correlation coefficient increased from 0.55 to 0.78. After calibration, the decreased proportion of RMSE in spring, summer, autumn, and winter was 98.6 %, 97.1 %, 97.1 % and 92.8 %, respectively. The annual and seasonal spatial distributions of ground surface temperature was displayed by the calibrated data, it showed that the spatial distribution of spring and autumn ground surface temperature closely resembled the annual mean pattern.The spring and autumn climatology closely resemble the annual mean pattern. The long-term climatology of ground surface freezing indices strongly resembles the cold season temperature climatology of the QTP. The climatology of ground surface thawing indices was reminiscent of the long-term warm season temperature climatology, with high values corresponding to warm areas on the QTP.3. The simulated results at single point indicated that the GIPL2 model performed better in the alpine steppe experiment site（QT06）, at the alpine swamp meadow experiment site（QT03）performed worse, at the alpine meadow experiment site（QT01）, the alpine desert steppe experiment site（QT05）, and the alpine steppe experiment site（QT04）are performed between the experiment site of alpine steppe and alpine swamp meadow. The model is very sensitive to the selection of the upper boundary conditions. The simulation results based on the ground temperature as the upper boundary condition are closer to the actual situation of soil thermal conditions. Based on the single-point simulations, the model is upscaled to the entire QTP. 4. The simulated results showed that the upscaled GIPL2 model improved the accuracy of the permafrost thermal state simulations. The active layer thickness decreases with an increase in the altitude and decreases from the southeast to the northwest. The active layer thickness is thin in the central QTP, but it is thick in the high-elevation mountain areas and some areas surrounding glaciers and lakes. Due to our simulated results on the QTP, the average active layer thickness is of 2.30 m （2.21 m - 2.40 m）. The simulated results of the mean annual ground temperature indicated that most of the permafrost is sub-stable, which was sensitive to climate warming. The boundaries between permafrost and seasonally frozen ground areas belong to transitional or unstable permafrost. During 1980 to 2013, the active layer thickness for most of the regions showed linear increasing trends, over the entire permafrost areas, the area-average variation rate of the active layer thickness on the QTP is 3.1 cm/a. The increase rate of active layer thickness in the southern of QTP is higher than that in the northern of QTP. The thicken rate of the active layer thickness in the western of QTP is faster than the eastern of QTP. |
| 中文关键词 | 多年冻土 ; 青藏高原 ; 气候变化 ; GIPL2模型 ; 再分析数据 |
| 英文关键词 | Permafrost Qinghai-Tibetan Plateau Climate Change GIPL2 Model Reanalysis Data |
| 语种 | 中文 |
| 国家 | 中国 |
| 来源学科分类 | 地图学与地理信息系统 |
| 来源机构 | 中国科学院西北生态环境资源研究院 |
| 资源类型 | 学位论文 |
| 条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/288145 |
| 推荐引用方式 GB/T 7714 | 秦艳慧. 基于GIPL2模型的青藏高原冻土时空变化模拟研究[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 |
| [秦艳慧]的文章 |
| 百度学术 |
| 百度学术中相似的文章 |
| [秦艳慧]的文章 |
| 必应学术 |
| 必应学术中相似的文章 |
| [秦艳慧]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。
