PDF(5347 KB)
过去30 a青藏高原东部多年冻土退化区地表水体变化特征
杨友刚, 郭子龙, 柴明堂, 冯建伟, 张航, 申梁, 李国玉, 齐舜舜
长江科学院院报 ›› 2026, Vol. 43 ›› Issue (4) : 52-60.
PDF(5347 KB)
PDF(5347 KB)
过去30 a青藏高原东部多年冻土退化区地表水体变化特征
Surface Water Dynamics in Degrading Permafrost Regions of Eastern Qinghai-Xizang Plateau over Past 30 Years
青藏高原地区被称为气候变化的“敏感区”以及全球气候变化的“驱动器”和“放大器”,其地表水体作为冻土退化的敏感指示因子有着重要研究意义。研究利用1995—2024年Landsat5/8卫星遥感影像,采用年平均地温反演经验模型和NDWI水体指数阈值法,对多年冻土区在退化状态下地表水体演变的时空特征及其与气候变化的响应关系进行了探究。结果表明:研究区内多年冻土区和季节冻土区面积占比分别为63%和37%,根据验证反演精度达88.1%。1995—2024年,研究区水体数量增长40%(≤0.01 km2水体贡献为主),面积增长29%((1,100]km2水体贡献为主);多年冻土区水体数量增加85%(≤0.01 km2水体为主),面积增幅28%;季节冻土区水体数量增长16%,面积增长28%((1,100]km2水体驱动)。研究区水体增加的主导因素是气温升高,降水在多数情况下不是主要驱动因素,且多年冻土区水体对气候的响应更加敏感。研究可为理解全球变暖下高原冻土区水资源和环境变化提供新的案例和数据支撑。
[Objective] As a region characterized by extensive permafrost, the Qinghai-Xizang Plateau has undergone significant environmental changes under global climate change. Surface water dynamics serve as sensitive indicators of permafrost degradation. This study investigates surface water changes over the past 30 years in a typical permafrost degradation area of the eastern Qinghai-Xizang Plateau, distinguishes variations between permafrost and seasonally frozen ground zones, and analyzes their relationships with temperature and precipitation. [Methods] Landsat 5 and Landsat 8 satellite images from 1995 to 2024 (August data only) were processed using Google Earth Engine (GEE) to remove clouds and high-reflectance interference through median-pixel compositing. An empirical annual mean ground temperature model, corrected for slope and aspect, was applied to classify permafrost and seasonally frozen ground zones. Surface water bodies were extracted using the Normalized Difference Water Index (NDWI) with an Otsu global-local thresholding method. The results were further refined using slope and hillshade data derived from the ASTER Global Digital Elevation Model (GDEM). Surface water bodies were classified by area into four categories: ≤0.001,(0.001,0.01],(0.01,1],and (1,100] km2. Monthly temperature and precipitation data from local meteorological stations were used to analyze correlations with water body metrics. [Results] Permafrost and seasonally frozen ground zones accounted for approximately 63% and 37% of the study area, respectively, with a classification accuracy of 88.1% as confirmed by field surveys. Between 1995 and 2024, the total number of water bodies increased by 40%, mainly driven by small water bodies (≤0.01 km2), while the total water surface area expanded by 29%, dominated by large water bodies ((1,100] km2). In permafrost zones, the number of water bodies increased by 85%, primarily due to small water bodies formed by thaw-induced subsidence, whereas the area increased by 28%. Seasonally frozen ground zones showed a moderate 16% increase in the total number of water bodies and a 28% increase in area, largely attributable to larger water bodies. Correlation analysis revealed significant positive relationships between temperature and water body metrics (r>0.75), with smaller water bodies exhibiting the highest temperature sensitivity. Conversely, precipitation generally had weak or negative correlations with water dynamics, particularly in permafrost zones, where heavy rainfall often promoted drainage and lake outflow. Seasonally frozen ground zones showed limited sensitivity to precipitation due to higher infiltration rates. [Conclusion] Rising temperatures primarily drive the expansion of surface water, exceeding the effects of precipitation. Permafrost zones are highly sensitive to warming, as indicated by rapid increases in small water bodies, whereas seasonally frozen ground zones maintain stable water body counts with area expansion driven by larger lakes. Precipitation plays a secondary or even negative role in water dynamics. The distinct responses of water bodies under different freeze-thaw conditions highlight the complexity of hydrological changes driven by climate warming, providing crucial insights for future environmental predictions and resource management on the Qinghai-Xizang Plateau.
冻土退化 / 地表水体 / 时空演变 / 气候响应 / NDWI / 冻土分布模型
permafrost degradation / surface water body / spatiotemporal evolution / climate response / NDWI / permafrost distribution model
| [1] |
程国栋, 赵林, 李韧, 等. 青藏高原多年冻土特征、变化及影响[J]. 科学通报, 2019, 64(27): 2783-2795.
(
|
| [2] |
姚檀栋, 邬光剑, 徐柏青, 等. “亚洲水塔”变化与影响[J]. 中国科学院院刊, 2019, 34(11): 1203-1209.
(
|
| [3] |
|
| [4] |
|
| [5] |
闾利, 张廷斌, 易桂花, 等. 2000年以来青藏高原湖泊面积变化与气候要素的响应关系[J]. 湖泊科学, 2019, 31(2):573-589.
(
|
| [6] |
苏红军. 高光谱遥感影像降维:进展、挑战与展望[J]. 遥感学报, 2022, 26(8):1504-1529.
(
|
| [7] |
|
| [8] |
刘沭岍, 刘凯, 曾繁轩, 等. 河流水文遥感及其在青藏高原应用研究进展[J]. 遥感学报, 2024, 28(10): 2427-2447.
(
|
| [9] |
刘苏宜, 黄对, 王建军, 等. 基于NDWI及扩展指数的地表水分变化监测: 以宿鸭湖水库为例[J]. 水利水运工程学报, 2024(6): 103-114.
(
|
| [10] |
常家豪, 姚晓军, 王玉, 等. 新藏公路沿线流域冰湖变化与冰湖溃决洪水危险性评估[J]. 湖泊科学, 2025, 37(4):1457-1469.
(
|
| [11] |
傅铭哲, 郑渊茂, 钱长照, 等. 基于多源遥感数据的洞庭湖流域时空演变研究[J]. 遥感信息, 2024, 39(4): 174-180.
(
|
| [12] |
李均力, 盛永伟, 骆剑承, 等. 青藏高原内陆湖泊变化的遥感制图[J]. 湖泊科学, 2011, 23(3): 311-320.
(
|
| [13] |
闫立娟, 齐文. 青藏高原湖泊遥感信息提取及湖面动态变化趋势研究[J]. 地球学报, 2012, 33(1):65-74.
(
|
| [14] |
孟恺, 石许华, 王二七, 等. 青藏高原中部色林错区域古湖滨线地貌特征、空间分布及高原湖泊演化[J]. 地质科学, 2012, 47(3): 730-745.
(
|
| [15] |
|
| [16] |
李治国. 近50 a气候变化背景下青藏高原冰川和湖泊变化[J]. 自然资源学报, 2012, 27(8):1431-1443.
论文综述了近年来青藏高原冰川和湖泊变化研究取得的成果,并特别着重于冰川和湖泊变化的相互关系论述。在全球变暖背景下,近几十年青藏高原冰川以退缩为主,湖泊水量以增加为主。论文一方面对青藏高原冰川末端退缩、冰川面积和冰川储量变化方面的研究成果进行了综合分析,探讨了冰川变化的时空特征;另一方面从湖泊面积和水位与水量变化探讨了湖泊变化的时空规律。结果表明青藏高原冰川退缩的幅度总体上呈从青藏高原外缘向内陆呈减小的变化态势,受冰川融水补给比较大的湖泊近期面积扩张、水位上升明显。最后指出了青藏高原冰川、湖泊变化研究中存在的问题及今后的发展趋势。
(
In this paper, we reviewed the recent plateau glaciers and lakes change research achievements, with special emphasis on discussion of the relationship between glacier shrinkage and lake change. In the context of global climate change, the glaciers in the Tibetan Plateau have been generally retreating while the lakes have been generally expanding. Firstly, the researches about glacial terminal retreat, areas and volume variations in the Tibetan Plateau in recent decades have been reviewed and analyzed, and tempo-spatial change characteristics of glaciers have been discussed. Secondly, the lake area, volume and water level changes have been reviewed and analyzed, and tempo-spatial change characteristics of lakes have been discussed. The results show that the retreat speed in the outer edge of the Tibetan Plateau was larger than that of the inland on the whole, the area and water level of the lakes fed by glacial water increased. Finally, the existing problems of the present studies on glaciers and lakes changes and the future tendency were analyzed.
|
| [17] |
刘佳丽, 周天财, 于欢, 等. 西藏近25 a湖泊变迁及其驱动力分析[J]. 长江科学院院报, 2018, 35(2): 145-150.
作为全球气候变化敏感区,分析气候变暖与经济发展背景下西藏湖面动态变化,对揭示区域环境演变特征与规律具有重要意义。基于1990—2015年Landsat遥感影像获取湖泊水域边界信息,分析近25 a西藏湖泊的时空分布及变化特征,结合气候因素与人为因素变化特征分析湖泊变化的主要驱动力,及湖泊变迁对气候的响应。结果表明:近25 a,西藏新增湖泊261个,总面积从24 161.1 km<sup>2</sup>增加到了30 549.2 km<sup>2</sup>。时间上,流域内湖泊总面积1990—1995年严重萎缩,1996—2006年迅速扩张,2007—2013年湖泊扩张速度减缓,降雨量、气温、蒸散发的变化趋势较好地反映了湖泊由萎缩到扩张的变化原因。空间上,不同地区湖泊变迁存在明显的空间差异:西藏中部地区温度和蒸散发相对较低,降雨量较大,湖泊面积扩张迅速;西藏北部地区温度偏高,降雨量少,蒸散发大,但温度升高加速冰川冻土融化成为湖泊扩张的可能因素;喜马拉雅山脉温度较高,蒸发量远大于降雨量,导致湖泊萎缩。人类活动加剧了当地对湖泊等水资源的需求,湖泊却呈扩张趋势,说明西藏地区的湖泊变迁主要受到气候因素的驱动。
(
Analyzing the dynamic changes of lakes in Tibet, the world’s climate sensitive region, in the background of global warming and economic development, is of great significance to revealing the environment evolution features. According to the lake boundary information obtained by Landsat remote sensing images from 1990 to 2015, we analyzed the dynamic changes of spatial and temporal distribution of lakes in Tibet in the past 25 years. We also investigated into the main driving factors of lake change in terms of climate factor and human factor, and further analyzed the lake change in response to climate change. Results show that in the past 25 years, the number of lakes in Tibet has increased by 261 and the area of lakes has increased from 24 161.1 km<sup>2 </sup>to 30 549.2 km<sup>2</sup>. In temporal scale, lake area experienced a severe shrinkage in 1990-1995, and then expanded rapidly in 1996-2006; in 2007-2013, the expansion alleviated. The temporal transformation can be well reflected by precipitation, temperature, and evapotranspiration. In spatial scale, the variation of lake area differs apparently: in the middle of Tibet, low temperature and evapotranspiration as well as high precipitation gave rise to the rapid expansion of lake area; whereas in north Tibet, despite high temperature, low precipitation and large amount of evapotranspiration, the melting of glaciers and frozen soil caused by high temperature could be a possible factor of lake expansion; in the Himalayas, evapotranspiration far exceeding precipitation led to lake shrinkage. In addition, although human activities contributed to the local demand for lakes and other water resources, the area of lake has expanded, indicating that climate is a main driving factor of lake change in Tibet.
|
| [18] |
周黎明, 张杨. 基于GPR探测的长江源地区冰川与冻土厚度研究[J]. 长江科学院院报, 2024, 41(3):1-8.
长江源地区的冰川变化揭示了青藏高原气候变化趋势。冰下地形探测作为冰川发育和运动过程研究的基础,对长江地区水土保持和淡水资源储量研究具有指导意义。长江科学院在长达10 a的江源科考基础上,分别于2022年、2023年采用探地雷达(GPR)技术对长江正源沱沱河发源地格拉丹东主峰的冰川厚度进行精准探测,并对查旦湿地冻土厚度上限进行了探测研究。结合多种冰川和冻土地质模型的GPR波场模拟结果,提高了GPR技术在长江源地区冰川和冻土探测的有效性和精准度。探测结果表明,格拉丹东主峰冰川厚度和查旦湿地冻土厚度上限均有不同程度降低,冰川厚度和冻土厚度上限观测是一个常年积累的结果,后续仍需持续进行观测,积累更多数据,分析变化趋势,以估算探测区域内冰储量,研究气候变化对冰川的影响效果。
(
|
| [19] |
鲁嘉濠, 牛富俊, 林战举, 等. 考虑局地因素坡向影响的青藏高原工程走廊冻土分布与制图研究[J]. 地理与地理信息科学, 2012, 28(3): 63-67.
(
|
| [20] |
. Monitoring the thermal state of permafrost (TSP) is\nimportant in many environmental science and engineering applications.\nHowever, such data are generally unavailable, mainly due to the lack of\nground observations and the uncertainty of traditional physical models. This\nstudy produces novel permafrost datasets for the Northern Hemisphere (NH),\nincluding predictions of the mean annual ground temperature (MAGT) at the\ndepth of zero annual amplitude (DZAA) (approximately 3 to 25 m) and active\nlayer thickness (ALT) with 1 km resolution for the period of 2000–2016, as\nwell as estimates of the probability of permafrost occurrence and permafrost\nzonation based on hydrothermal conditions. These datasets integrate\nunprecedentedly large amounts of field data (1002 boreholes for MAGT and\n452 sites for ALT) and multisource geospatial data, especially remote\nsensing data, using statistical learning modeling with an ensemble\nstrategy. Thus, the resulting data are more accurate than those of previous\ncircumpolar maps (bias = 0.02±0.16 ∘C and RMSE = 1.32±0.13 ∘C for MAGT; bias = 2.71±16.46 cm and\nRMSE = 86.93±19.61 cm for ALT). The datasets suggest that the areal\nextent of permafrost (MAGT ≤0 ∘C) in the NH, excluding\nglaciers and lakes, is approximately 14.77 (13.60–18.97) × 106 km2 and that the areal extent of permafrost regions (permafrost\nprobability >0) is approximately 19.82×106 km2. The areal fractions of humid, semiarid/subhumid, and arid permafrost regions are 51.56 %, 45.07 %, and 3.37 %, respectively. The\nareal fractions of cold (≤-3.0 ∘C), cool (−3.0 ∘C\nto −1.5 ∘C), and warm (>-1.5 ∘C)\npermafrost regions are 37.80 %, 14.30 %, and 47.90 %, respectively.\nThese new datasets based on the most comprehensive field data to date\ncontribute to an updated understanding of the thermal state and zonation of\npermafrost in the NH. The datasets are potentially useful for various\nfields, such as climatology, hydrology, ecology, agriculture, public health,\nand engineering planning. All of the datasets are published through the\nNational Tibetan Plateau Data Center (TPDC), and the link is https://doi.org/10.11888/Geocry.tpdc.271190 (Ran et al.,\n2021a).
|
| [21] |
|
| [22] |
马伟东, 刘峰贵, 周强, 等. 1961—2017年青藏高原极端降水特征分析[J]. 自然资源学报, 2020, 35(12): 3039-3050.
基于青藏高原78个气象站点的逐日降水数据,采用百分位阈值法确定极端降水阈值,计算极端降水指数并分析其时空分布特征,以期为区域气候变化预测及防灾减灾对策的制定提供参考。结果表明：（1）1961—2017年青藏高原年降水量表现出上升趋势,上升速率为8.06 mm/10 a,多年平均降水量达472.36 mm。78个站点的年降水量倾向率最小值为-25.46 mm/10 a,最大值为43.02 mm/10 a,有15.38%的站点降水在下降,较为集中地分布在高原的东部和南部,其余84.62%的站点降水量在上升。（2）青藏高原各站点极端降水阈值的平均值为23.11 mm,取值范围为7.84~51.90 mm。高值中心出现在横断山区的贡山和木里,低值中心出现在柴达木盆地及昆仑山北翼区。（3）青藏高原各站点的极端降水量、极端降水日数和极端降水贡献率均表现出了明显的上升趋势,极端降水强度虽然也在上升但趋势并不明显,表明青藏高原极端降水量的上升并非是极端降水的强度引起的,而是由极端降水频次的上升引起的。柴达木盆地的极端降水量和极端降水日数虽然并没有表现出高值水平,但该地区的极端降水贡献率却表现出较高水平,表明该区域虽然降水量较少,但是降水往往以极端降水的形式产生。
(
Using the daily precipitation data of the long-term series of meteorological stations on the Qinghai-Tibet Plateau, the percentile threshold method is used to determine the extreme precipitation threshold, calculate the extreme precipitation index and analyze its spatial and temporal distribution characteristics, in order to provide reference for regional climate change prediction and disaster prevention and mitigation countermeasures. The results show that: (1) From 1961 to 2017, the annual precipitation of Qinghai-Tibet Plateau showed an upward trend, with a rate of 8.06 mm/10 a, and the average annual precipitation reached 472.36 mm. The minimum precipitation tendency rate of 78 stations is -25.46 mm/10 a, and the maximum value is 43.02 mm/10 a. The precipitation of 15.38% of the stations is decreasing, which is mainly distributed in the east and south of the plateau, and the precipitation of the remaining 84.62% of the stations is increasing. (2) The average threshold value of extreme precipitation in the Qinghai-Tibet Plateau is 23.11 mm, with error values ranging from 7.84 mm to 51.90 mm. The high value centers are located in Gongshan and Muli of Hengduan Mountains, while the low value centers are located in the northern flank of Qaidam Basin and Kunlun Mountains. (3) The extreme precipitation, the number of days of extreme precipitation and the contribution rate of extreme precipitation at all the stations in the Qinghai-Tibet Plateau show an obvious upward trend. Although the intensity of extreme precipitation is also rising, the trend is not obvious, which shows that the increase of extreme precipitation in the plateau is not caused by the intensity of extreme precipitation, but by the increase of the frequency of extreme precipitation. Although the extreme precipitation and days of extreme precipitation in the Qaidam Basin do not show a high value level, the contribution rate of extreme precipitation is larger, which suggests that although there is less precipitation, extreme precipitation events frequently occur in this area. |
| [23] |
吴青柏, 张中琼, 刘戈. 青藏高原气候转暖与冻土工程的关系[J]. 工程地质学报, 2021, 29(2):342-352.
(
|
| [24] |
李琳, 谭德宝, 文雄飞, 等. 气候变化对可可西里盐湖流域湖泊水量变化的影响分析[J]. 长江科学院院报, 2022, 39(10):16-23.
湖泊是气候变化的敏感指示器。为了研究气候变化对湖泊水量的影响,以盐湖流域为研究区,应用统计方法对1989—2018年降雨、气温、蒸发进行线性趋势和突变分析,采用多源卫星遥感技术对湖泊面积等水文要素进行监测,分析湖泊面积与气象要素、湖泊面积与湖泊水量之间的相关性。利用VIC模型模拟径流并结合计算的冰川水量得到盐湖径流组成,定量探讨气象要素对湖泊水量变化的影响,综合分析2011年前后气象要素影响流域湖泊水量的差异。结合统计分析与水文模型定量计算可知:年降雨量、年平均气温显著升高,年蒸发量呈下降趋势,且与湖泊面积有较好的相关性。湖泊面积与湖泊水量间相关性较高,可间接体现气象要素对湖泊水量变化的影响。2011年前卓乃湖和盐湖水量变化主要受降雨量影响,库赛湖和海丁诺尔湖水量变化主要受气温影响;2011—2014年4个湖泊水量变化主要受降雨量影响;2015—2018年4个湖泊水量变化中降雨增加量、冻土释水和地下水补给增加量、冰川融水量对湖泊扩张的贡献约为34.48%、57.66%、7.86%,气温变化成为影响湖泊水量变化的主要因素,降雨量影响次之。
(
Lakes are sensitive indicators of climate change.In the aim of studying the impact of climate change on lake water volume,statistical methods were applied to examine the linear trend and abrupt changes of rainfall,temperature,and evaporation from 1989 to 2018 in the Hoh Xil Salt Lake Basin.Hydrological elements such as lake area were monitored by using multi-source satellite remote sensing technology,and the correlation between lake area and meteorological elements,lake area and lake water volume changes were analyzed.Moreover,the impact of meteorological elements on the changes in lake water volume was quantified by obtaining the composition of the Salt Lake runoff in association with VIC model-simulated runoff and calculated glacier water volume.The differences in the impact of meteorological elements on lake water volume in the basin before and after 2011 were also comprehensively scrutinized.The statistical analysis and quantitative calculation of hydrological models manifest that the annual rainfall and annual average temperature in the Hoh Xil Salt Lake Basin have increased significantly,while annual evaporation has presented a downward trend,all in good correlation with lake area.Lake area is highly correlated with lake water volume,which indirectly reflects the impact of meteorological elements on lake water volume changes.Before 2011,the changes in the water volume of Zhuonai Lake and Salt Lake were mainly affected by rainfall,and the water volume changes in Kusai Lake and Hading Knoll were mainly affected by temperature;from 2011 to 2014,changes in the water volume of the four lakes were mainly affected by rainfall;from 2015 to 2018,the increase in rainfall,the release of water from frozen soil and the increase in groundwater recharge,and the amount of glacial melting water contributed about 34.48%,57.66%,and 7.86%,respevtively,to the expansion of the four lakes.Temperature changes have become the major factor,followed by rainfall,affecting the changes in lake water volume.
|
| [25] |
朱立平, 张国庆, 杨瑞敏, 等. 青藏高原最近40年湖泊变化的主要表现与发展趋势[J]. 中国科学院院刊, 2019, 34(11): 1254-1263.
(
|
/
| 〈 |
|
〉 |
