开展区域冰川变化的高精度识别与动态监测研究对于理解全球气候变化及水资源管理具有重要意义。近年来，深度学习技术在冰川自动识别领域的应用显著提升了冰川监测的效率与精度，然而冰川临近的基岩、积雪以及山体阴影等因素仍然对冰川识别造成了显著干扰。针对上述现象，本研究选取绒布河流域及周边区域作为研究区，基于多源遥感数据构建了冰川识别数据集，结合深度学习方法提出一种改进型U-Net网络结构（SAU-Net），并对2000年、2015年与2024年三期冰川边界进行提取与分析，系统探讨冰川面积变化的时空特征、地形因子与气候因子的响应关系。以下是论文的主要内容和研究结果：

（1）针对传统U-Net网络因多次卷积与池化操作而致空间信息损失，以及易将冰川与周边基岩误提取的问题，通过引入残差U型结构与注意力机制，对传统U-Net模型加以改进，提出SAU-Net网络。随后采用多源遥感数据制作冰川数据集对其进行训练。结果表明，SAU-Net的准确率为0.946，Fβ 值为0.835，MAE值为0.099。精度评价与提取效果均优于传统U-Net模型，能更精准地界定冰川范围。消融实验进一步验证了SimAM（Simple, Parameter-Free Attention Module）模块与残差U型结构在提高模型识别精度方面的显著贡献。

（2）对研究区冰川分布现状进行识别提取。基于多源遥感数据，提取了2000年，2015年和2024年的冰川边界，并对2024年冰川数量及分布进行分析，结果显示：绒布河流域及周边地区内共发育有276条冰川。根据统计分析，研究区内冰川覆盖总面积达到了671.46km²，平均每条冰川面积约为2.43km²。此外，冰川发育的平均坡度为25.60°，冰川分布的海拔中值为5916.46m。分析了2000-2024年绒布河流域及其周边地区冰川面积的时空变化特征。近25年间来研究区冰川数量从280条减少至276条，冰川面积共减少了38.71km²（5.45%），年均变化率为-0.23%/a。2000-2015年和2015-2014年两个时段对比分析表明，近25年间来绒布河流域及其周边地区年均变化率从-0.22%/a变至-0.25%/a，显示出研究区内冰川退缩率在增加。

（3）研究区域内冰川整体呈现显著的退缩趋势，不同规模冰川面积均呈减少态势，随着冰川规模增大和海拔升高，冰川面积先增后减；随着冰川平均坡度增加，冰川面积退缩量亦先增后减。面积退缩最为剧烈的冰川集中在1-2km²和2-5km²区间，冰川的面积退缩幅度分别高达6.34%和25.01%。从海拔分布看，冰川面积减少主要集中于5000m至6400m的海拔带，该区间退缩总量达28.01km²，占研究区冰川总面积退缩量的72.96%。西北、北和东北三个朝向的冰川面积退缩量显著高于其他朝向，总面积退缩了26.82km²。结合气象数据对冰川变化进行了分析，结果显示气温的持续上升是驱动冰川退缩的核心因素。

关键词：深度学习；冰川识别；冰川变化；绒布河区域

研究类型：应用研究

High-precision identification and dynamic monitoring of regional glacial changes are of great significance for understanding global climate change and water resource management. In recent years, the application of deep learning techniques in automated glacier recognition has significantly improved the efficiency and accuracy of glacier monitoring. However, factors such as adjacent bedrock, snow cover, and mountain shadows still pose significant interference to glacier identification. To address these challenges, this study selects the Rongbuk River basin and its surrounding areas as the research region. A glacier identification dataset was constructed based on multi-source remote sensing data, and an improved U-Net network structure (SAU-Net) was proposed using deep learning methods. Glacier boundaries from 2000, 2015, and 2024 were extracted and analyzed to systematically investigate the spatiotemporal characteristics of glacier area changes, as well as the response relationships between topographic and climatic factors. The main contents and results of the paper are as follows:

(1) To mitigate the spatial information loss caused by multiple convolutions and pooling operations in traditional U-Net networks, as well as the tendency to misclassify glaciers and adjacent bedrock, this study introduces a residual U-shaped structure and an attention mechanism to improve the traditional U-Net model, resulting in the SAU-Net network. The model was trained using a glacier dataset created from multi-source remote sensing data. The results show that SAU-Net achieves an accuracy of 0.946, an Fβ  score of 0.835, and mean absolute errors (MAE) of 0.099. Both precision evaluation and extraction effectiveness outperform the traditional U-Net model, enabling more accurate delineation of glacier boundaries. Ablation experiments further validate the significant contributions of the SimAM (Simple, Parameter-Free Attention Module) and the residual U-shaped structure in enhancing model recognition accuracy.

(2) The current distribution of glaciers in the study area was identified and extracted. Based on multi-source remote sensing data, glacier boundaries from 2000, 2015, and 2024 were extracted, and the number and distribution of glaciers in 2024 were analyzed. The results reveal that the Rongbuk River basin and its surroundings host 276 glaciers, with a total coverage area of 671.46 km² (average area per glacier: 2.43 km²). Additionally, glaciers develop at an average slope of 25.60°, and the median elevation of glacier distribution is 5916.46 m. Analysis of spatiotemporal variations in glacier area from 2000 to 2024 indicates that the number of glaciers decreased from 280 to 276 over 25 years, with a total area reduction of 38.71 km² (5.45%) and an average annual change rate of -0.23%/a. Comparative analysis of two periods (2000–2015 and 2015–2024) shows that the annual retreat rate increased from -0.22%/a to -0.25%/a, indicating an accelerating glacier shrinkage trend in the region.

(3) Glaciers in the study area exhibit a significant retreat trend overall, with reduced area across all size classes. Glacier area initially increases with size and elevation before declining, while retreat magnitude first rises and then decreases with increasing average slope. The most severe area losses occur in glaciers ranging from 1–2 km² (6.34% loss) and 2–5 km² (25.01% loss). Elevationally, area reductions are concentrated between 5000–6400 m, accounting for 72.96% of the total retreat (28.01 km²). Glaciers facing northwest, north, and northeast display significantly higher retreat volumes (26.82 km² total). Meteorological data analysis confirms that sustained temperature rise is the core driver of glacier retreat.

Key words: Deep learning; Glacier identification; Glacier change; Rongbuk River Region

Thesis: Application Research

[1] COGLEY J G. A more complete version of the World Glacier Inventory [J]. Annals of Glaciology, 2009, 50(53): 32-38.

[2] 秦大河, 姚檀栋, 丁永建. 冰冻圈科学概论 [M]. 北京：科学出版社, 2017.

[3] Yao T, Thompson L, Yang W, et al. Different glacier status with atmospheric circulations in Tibetan Plateau and surroundings [J]. Nature climate change, 2012, 2(9): 663-667.

[4] 施雅风. 2050 年前气候变暖冰川萎缩对水资源影响情景预估 [J]. 冰川冻土, 2001, 23(4): 333-341.

[5] 张九天, 何霄嘉, 上官冬辉, 等. 冰川加剧消融对我国西北干旱区的影响及其适应对策 [J]. 冰川冻土, 2012, 34(4): 848-854.

[6] Carrivick J L, Tweed F S. A global assessment of the societal impacts of glacier outburst floods [J]. Global and Planetary Change, 2016, 144: 1-16.

[7] Song C, Sheng Y, Wang J, et al. Heterogeneous glacial lake changes and links of lake expansions to the rapid thinning of adjacent glacier termini in the Himalayas [J]. Geomorphology, 2017, 280: 30-38.

[8] 孙美平, 刘时银, 姚晓军, 等. 近 50 年来祁连山冰川变化——基于中国第一, 二次冰川编目数据 [J]. 地理学报, 2015, 70(9): 1402-1414.

[9] 冀琴, 张翠兰, 丁悦凯, 等. 基于多源遥感数据的珠峰自然保护区冰川监测研究 [J]. 干旱区地理, 2023, 46(10): 1591-1601.

[10] 张聪, 姚晓军, 肖建设, 等. 2000—2020 年青海省冰川和湖泊演化特征 [J]. 自然资源学报, 2023, 38(3): 822-838.

[11] 冀琴. 1990-2015年喜马拉雅山冰川变化及其对气候波动的响应 [D]; 兰州大学, 2018.

[12] 徐浩栋, 李佳, 吴立新, 等. 西藏天摩沟冰川遥感监测及泥石流触发成因分析 [J]. 地球物理学报, 2023, 66(6): 2370-2385.

[13] 张国梁. 贡嘎山地区现代冰川变化研究 [D]; 兰州大学, 2012.

[14] Liu T, Dai J, Zhao Y, et al. Using remote sensing technology to monitor salt lake changes caused by climate change and melting glaciers: insights from Zabuye Salt Lake in Xizang [J]. Journal of Oceanology and Limnology, 2023, 41(4): 1258-1276.

[15] 丁悦凯, 刘睿, 张翠兰, 等. 喜马拉雅地区叶如藏布流域冰川和冰湖变化遥感监测研究 [J]. 干旱区地理, 2023, 45(6): 1870-1880.

[16] 冉伟杰, 王欣, 郭万钦, 等. 2017–2018年中国西部冰川编目数据集 [J]. 中国科学数据(中英文网络版), 2021, 6(02): 195-204.

[17] 邢武成, 李忠勤, 张慧, 等. 1959 年来中国天山冰川资源时空变化 [J]. 地理学报, 2017, 72(9): 195-204.

[18] 车正, 王宁练, 梁倩, 等. 祁连山托来南山 6 号冰川雷达测厚与冰储量分析 [J]. 冰川冻土, 2022, 44(5): 1409-1418.

[19] 李欢, 万玮, 冀锐, 等. 中国卫星遥感地表水资源监测能力分析与展望 [J]. 遥感学报, 2023, 27(7): 1554-1573.

[20] 陆艺杰. 基于机器学习与遥感影像的冰川边界提取方法研究 [D]; 安徽理工大学, 2021.

[21] 庞书剑, 柯长青, 周兴华, 等. InSAR 与激光雷达测高集成的马兰山冰川物质平衡变化 [J]. 遥感学报, 2022, 26(10): 2094-2105.

[22] 李海. 喜马拉雅山脉中段地区冰川冰湖遥感变化监测 [D]; 长安大学, 2021.

[23] 吕卉. 近 40 年喜马拉雅山冰川波动对气候变化的响应 [D]; 兰州大学, 2013.

[24] 秦大河. 青藏高原的冰川与生态环境 [M]. 中国藏学出版社, 1999.

[25] Kaushik S, Joshi P, Singh T. Development of glacier mapping in Indian Himalaya: A review of approaches [J]. International Journal of Remote Sensing, 2019, 40(17): 6607-6634.

[26] Nuimura T, Sakai A, Taniguchi K, et al. The GAMDAM glacier inventory: a quality-controlled inventory of Asian glaciers [J]. The Cryosphere, 2015, 9(3): 849-864.

[27] 李志杰. 基于遥感和 GIS 的 2000-2017 年帕米尔高原冰川变化研究 [D]; 西北大学, 2020.

[28] Khan A A, Jamil A, Hussain D, et al. Machine-learning algorithms for mapping debris-covered glaciers: the Hunza Basin case study [J]. Ieee Access, 2020, 8: 12725-12734.

[29] Garg P K, Shukla A, Jasrotia A S. On the strongly imbalanced state of glaciers in the Sikkim, eastern Himalaya, India [J]. Science of the Total Environment, 2019, 691: 16-35.

[30] Veh G, Korup O, Walz A. Hazard from Himalayan glacier lake outburst floods [J]. Proceedings of the National Academy of Sciences, 2020, 117(2): 907-912.

[31] Azam M F, Wagnon P, Berthier E, et al. Review of the status and mass changes of Himalayan-Karakoram glaciers [J]. Journal of Glaciology, 2018, 64(243): 61-74.

[32] Lemke P, Ren J, Alley R B, et al. Observations. Changes in Snow, Ice and Frozen Ground. Chapter 4 [M]. 2007.

[33] 王宗太, 苏宏超. 世界和中国的冰川分布及其水资源意义 [J]. 冰川冻土, 2003, 25(5): 498-503.

[34] Oerlemans, J. Quantifying global warming from the retreat of glaciers [J]. Science, 1994, 264(5156): 243-245.

[35] 段建平, 王丽丽, 任贾文, 等. 近百年来中国冰川变化及其对气候变化的敏感性研究进展 [J]. 地理科学进展, 2009, (2): 231-237.

[36] Kargel J. Compositional controls on the geological behavior of icy satellites, and a call for more lab data [J]. AGU Fall Meeting Abstracts, 2006: P42B-03.

[37] 上官冬辉, 刘时银, 丁良福, 等. 1970-2000 年念青唐古拉山脉西段冰川变化 [J]. 冰川冻土, 2008, 30(2): 204-210.

[38] Hall D K, Bayr K J, Bindschadler R A, et al. Changes in the Pasterze Glacier, Austria, as measured from the ground and space [J]. AGU Fall Meeting Abstracts, 2001: 187-193.

[39] Andreassen L, Paul F, Kääb A, et al. Landsat-derived glacier inventory for Jotunheimen, Norway, and deduced glacier changes since the 1930s [J]. The Cryosphere, 2008, 2(2): 131-145.

[40] Sidjak R. Glacier mapping of the Illecillewaet icefield, British Columbia, Canada, using Landsat TM and digital elevation data [J]. International Journal of Remote Sensing, 1999, 20(2): 273-284.

[41] 刘志超. 合成孔径雷达图像目标识别方法研究 [D]; 江南大学, 2020.

[42] 张雯. 基于多源遥感影像数据的冰川信息提取方法研究 [D]; 兰州交通大学, 2016.

[43] 张齐民, 吕明阳, 闫世勇. 基于高分三号影像的2019–2020年高亚洲地区典型冰川表面流速数据集 [J]. 中国科学数据(中英文网络版), 2021, 6(03): 158-169.

[44] 叶庆华, 程维明, 赵永利, 等. 青藏高原冰川变化遥感监测研究综述 [J]. 地球信息科学学报, 2016, 18(7): 920-930.

[45] Paul F, Frey H, Le Bris R. A new glacier inventory for the European Alps from Landsat TM scenes of 2003: challenges and results [J]. Annals of Glaciology, 2011, 52(59): 144-152.

[46] 刘时银, 姚晓军, 郭万钦, 等. 基于第二次冰川编目的中国冰川现状 [J]. 地理学报, 2015, 70(01): 3-16.

[47] 韩惠, 杨晓辉, 赵井东. 西昆仑山崇测冰川区多源遥感影像的冰川信息提取方法研究 [J]. 冰川冻土, 2018, 40(5): 951-959.

[48] 田洪阵, 肖月, 杨太保, 等. 1973—2020 年阿尔金山冰川面积变化及其对气温变化的响应 [J]. 冰川冻土, 2021, 43(5): 1424-1434.

[49] 陈辉, 李忠勤, 王璞玉, 等. 近年来祁连山中段冰川变化 [J]. 干旱区研究, 2013, 30(4): 588-593.

[50] 李成秀, 杨太保, 田洪阵. 近 40 年来西昆仑山冰川及冰湖变化与气候因素 [J]. 山地学报, 2015, 33(02): 157-165.

[51] 王祎婷, 陈秀万, 柏延臣, 等. 多源 DEM 和多时相遥感影像监测冰川体积变化——以青藏高原那木纳尼峰地区为例 [J]. 冰川冻土, 2010, 32(1): 126-132.

[52] Narama C, Kääb A, Duishonakunov M, et al. Spatial variability of recent glacier area changes in the Tien Shan Mountains, Central Asia, using Corona (~ 1970), Landsat (~ 2000), and ALOS (~ 2007) satellite data [J]. Global and Planetary Change, 2010, 71(1-2): 42-54.

[53] 王玉哲, 任贾文, 秦大河, 等. 利用卫星资料反演区域冰川冰量变化的尝试——以祁连山为例 [J]. 冰川冻土, 2013, 35(3): 583-592.

[54] 汪秋昱, 易爽, 孙文科. 应用 GRACE 与 ICESat 数据研究青藏高原冰川与湖面变化 [J]. 2014 年中国地球科学联合学术年会——专题 3: 地球重力场及其地学应用论文集, 2014.

[55] Arnold N, Willis I, Sharp M, et al. A distributed surface energy-balance model for a small valley glacier. I. Development and testing for Haut Glacier d’Arolla, Valais, Switzerland [J]. Journal of Glaciology, 1996, 42(140): 77-89.

[56] Braithwaite R J, Zhang Y. Sensitivity of mass balance of five Swiss glaciers to temperature changes assessed by tuning a degree-day model [J]. Journal of Glaciology, 2000, 46(152): 7-14.

[57] 王帅帅, 周石硚, 郑伟. 喜马拉雅山中东段蒙达扛日冰川水文与物质平衡观测研究 [J]. 冰川冻土, 2011, 33(05): 1146-1152.

[58] 杨威, 姚檀栋, 徐柏青, 等. 青藏高原东南部岗日嘎布地区冰川严重损耗与退缩 [J]. 科学通报, 2008, 53(17): 2091-2095.

[59] 吴倩如, 康世昌, 高坛光, 等. 青藏高原纳木错流域扎当冰川度日因子特征及其应用 [J]. 冰川冻土, 2010, 32(05): 891-897.

[60] 周广鹏, 姚檀栋, 康世昌, 等. 青藏高原中部扎当冰川物质平衡研究 [J]. 冰川冻土, 2007, 29(3): 360-365.

[61] Matsuo K, Heki K. Time-variable ice loss in Asian high mountains from satellite gravimetry [J]. Earth and Planetary Science Letters, 2010, 290(1-2): 30-36.

[62] Beveridge A. Measuring the Changing Mass of Glaciers on the Tibetan Plateau Using Time-variable Gravity from the GRACE Mission [D]; Princeton University, 2016.

[63] 聂琳娟, 超能芳, 晁定波. GRACE 卫星的冰川消融与水储量变化探讨 [J]. 测绘科学, 2016, 41(8): 82-86.

[64] 王永前, 施建成, 胡小工, 等. 关于利用重力卫星对青藏高原水储量年际变化和季节性变化进行监测并用微波数据产品进行验证的研究 [J]. 地球物理学进展, 2009, 24(4): 1235-1242.

[65] 晋锐, 车涛, 李新, 等. 基于遥感和 GIS 的西藏朋曲流域冰川变化研究 [J]. 冰川冻土, 2004, 26(3): 261-266.

[66] 叶庆华, 陈锋, 姚檀栋, 等. 近 30 年来喜马拉雅山脉西段纳木那尼峰地区冰川变化的遥感监测研究 [J]. 遥感学报, 2007, 11(4): 511-520.

[67] 聂勇, 张镱锂, 刘林山, 等. 近 30 年珠穆朗玛峰国家自然保护区冰川变化的遥感监测 [J]. 地理学报, 2010, 65(01): 13-28.

[68] 上官冬辉, 刘时银, 丁永建, 等. 中国喀喇昆仑山, 慕士塔格公格尔山典型冰川变化监测结果 [J]. 冰川冻土, 2004, 26(3): 374-375.

[69] 上官冬辉, 刘时银, 丁永建, 等. 喀喇昆仑山克勒青河谷近年来发现有跃动冰川 [J]. 冰川冻土, 2005, 27(5): 641-644.

[70] 刘景时, 王迪. 2009 年夏季喀喇昆仑山叶尔羌河上游发生冰川跃动 [J]. 冰川冻土, 2009, 31(5): 992.

[71] 牛竞飞, 刘景时, 王迪, 等. 2009 年喀喇昆仑山叶尔羌河冰川阻塞湖及冰川跃动监测 [J]. 山地学报, 2011, 29(03): 276-282.

[72] Kääb A, Winsvold S H, Altena B, et al. Glacier remote sensing using Sentinel-2. Part I: Radiometric and geometric performance, and application to ice velocity [J]. Remote sensing, 2016, 8(7): 598.

[73] Paul F, Winsvold S H, Kääb A, et al. Glacier remote sensing using Sentinel-2. Part II: Mapping glacier extents and surface facies, and comparison to Landsat 8 [J]. Remote Sensing, 2016, 8(7): 575.

[74] Lei H, Zhen L, Min Z J, et al. An automatic method for clean glacier and nonseasonal snow area change estimation in High Mountain Asia from 1990 to 2018 [J]. Remote Sensing of Environment, 2021, 258.

[75] Keshri A K, Shukla A, Gupta R P. ASTER ratio indices for supraglacial terrain mapping [J]. International Journal of Remote Sensing, 2009, 30(2): 519-524.

[76] 刘时银. 气候变化影响与风险: 气候变化对冰川影响与风险研究 [M]. 科学出版社, 2017.

[77] 姚明煌. 随机森林及其在遥感图像分类中的应用 [D]; 华侨大学, 2014.

[78] Subasi A, Tuncer T, Dogan S, et al. EEG-based emotion recognition using tunable Q wavelet transform and rotation forest ensemble classifier [J]. Biomedical Signal Processing and Control, 2021, 68: 102648.

[79] Zhang J, Jia L, Menenti M, et al. Glacier facies mapping using a machine-learning algorithm: The Parlung Zangbo Basin case study [J]. Remote Sensing, 2019, 11(4): 452.

[80] 张静潇. 基于遥感数据的藏东南冰川和湖泊动态变化监测及其对气候变化的响应研究 [D]; 中国科学院大学, 2019.

[81] Lu Y, Zhang Z, Huang D. Glacier mapping based on random forest algorithm: A case study over the Eastern Pamir [J]. Water, 2020, 12(11): 3231.

[82] Hussain D, Khan A A. Machine learning techniques for monthly river flow forecasting of Hunza River, Pakistan [J]. Earth Science Informatics, 2020, 13(3): 939-949.

[83] 肖湘文, 沈校熠, 柯长青, 等. 基于 Sentinel-1A 数据的多种机器学习算法识别冰山的比较 [J]. 测绘学报, 2020, 49(4): 509-521.

[84] Alifu H, Vuillaume J-F, Johnson B A, et al. Machine-learning classification of debris-covered glaciers using a combination of Sentinel-1/-2 (SAR/optical), Landsat 8 (thermal) and digital elevation data [J]. Geomorphology, 2020, 369: 107365.

[85] Xie F, Liu S, Wu K, et al. Upward expansion of supra-glacial debris cover in the Hunza Valley, Karakoram, during 1990∼ 2019 [J]. Frontiers in Earth Science, 2020, 8: 308.

[86] Marochov M, Stokes C R, Carbonneau P E. Image Classification of Marine-Terminating Outlet Glaciers using Deep Learning Methods [J]. The Cryosphere Discussions, 2020, 2020: 1-45.

[87] Mohajerani Y, Jeong S, Scheuchl B, et al. Automatic delineation of glacier grounding lines in differential interferometric synthetic-aperture radar data using deep learning [J]. Scientific reports, 2021, 11(1): 4992.

[88] Zhang E, Liu L, Huang L, et al. An automated, generalized, deep-learning-based method for delineating the calving fronts of Greenland glaciers from multi-sensor remote sensing imagery [J]. Remote Sensing of Environment, 2021, 254: 112265.

[89] Baumhoer C A, Dietz A J, Kneisel C, et al. Automated extraction of antarctic glacier and ice shelf fronts from sentinel-1 imagery using deep learning [J]. Remote Sensing, 2019, 11(21): 2529.

[90] Chu X, Yao X, Duan H, et al. Glacier extraction based on high-spatial-resolution remote-sensing images using a deep-learning approach with attention mechanism [J]. The Cryosphere, 2022, 16(10): 4273-4289.

[91] Chudley T R, Willis I C. Glacier surges in the north-west West Kunlun Shan inferred from 1972 to 2017 Landsat imagery [J]. Journal of Glaciology, 2019, 65(249): 1-12.

[92] Guo W, Liu S, Xu J, et al. The second Chinese glacier inventory: data, methods and results [J]. Journal of Glaciology, 2015, 61(226): 357-372.

[93] Zhang J, Jia L, Menenti M, et al. Interannual and seasonal variability of glacier surface velocity in the parlung zangbo basin, tibetan plateau [J]. Remote Sensing, 2020, 13(1): 80.

[94] Friedl P, Seehaus T, Braun M. Global time series and temporal mosaics of glacier surface velocities, derived from Sentinel-1 data [J]. Earth System Science Data Discussions, 2021, 2021: 1-33.

[95] Drusch M, Del Bello U, Carlier S, et al. Sentinel-2: ESA's optical high-resolution mission for GMES operational services [J]. Remote sensing of Environment, 2012, 120: 25-36.

[96] 纪显琛. 基于支持向量机的林业遥感影像分类器开发 [D]; 北京林业大学, 2018.

[97] 秦臣臣. 基于 ICESat/GLAS 的山东省 SRTM, ALOS 与 ASTER GDEM 高程精度评价与修正 [D]; 山东科技大学, 2020.

[98] 徐迪峰. 海量遥感影像管理系统的研究与实现 [D]; 苏州大学, 2009.

[99] Julzarika A, Aditya T, Subaryono S, et al. Vertical accuracy evaluation of digital terrain model (Dtm) alos palsar-2 in rote dead sea area–Indonesia [J]. Geod List, 2021, 75(1): 9-28.

[100]Bolch T, Menounos B, Wheate R. Landsat-based inventory of glaciers in western Canada, 1985–2005 [J]. Remote sensing of Environment, 2010, 114(1): 127-137.

[101]Kienholz C, Hock R, Arendt A A. A new semi-automatic approach for dividing glacier complexes into individual glaciers [J]. Journal of Glaciology, 2013, 59(217): 925-937.

[102]Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation [J]. CoRR, 2015: 234-241.

[103]Yang L, Zhang R-Y, Li L, et al. Simam: A simple, parameter-free attention module for convolutional neural networks; proceedings of the International conference on machine learning, F, 2021 [C].

[104]李林峰. 基于深度学习的贡嘎山冰川边界提取研究 [D]; 西南科技大学, 2024.

[105]Sharma A, Liu X, Yang X, et al. A patch-based convolutional neural network for remote sensing image classification [J]. Neural Networks, 2017, 95: 19-28.

[106]Huss M, Farinotti D. Distributed ice thickness and volume of all glaciers around the globe [J]. Journal of Geophysical Research: Earth Surface, 2012, 117(F4).

[107]Shukla A, Arora M, Gupta R. Synergistic approach for mapping debris-covered glaciers using optical–thermal remote sensing data with inputs from geomorphometric parameters [J]. Remote Sensing of environment, 2010, 114(7): 1378-1387.

[108]Kaushik S, Singh T, Bhardwaj A, et al. Automated delineation of supraglacial debris cover using deep learning and multisource remote sensing data [J]. Remote Sensing, 2022, 14(6): 1352.

[109]Chowdhury A, Sharma M C, Kumar De S, et al. Glacier changes in the Chhombo Chhu Watershed of the Tista basin between 1975 and 2018, the Sikkim Himalaya, India [J]. Earth System Science Data, 2021, 13(6): 2923-2944.

[110]Scherler D, Bookhagen B, Strecker M R. Spatially variable response of Himalayan glaciers to climate change affected by debris cover [J]. Nature geoscience, 2011, 4(3): 156-159.

[111]Pfeffer W T, Arendt A A, Bliss A, et al. The Randolph Glacier Inventory: a globally complete inventory of glaciers [J]. Journal of glaciology, 2014, 60(221): 537-552.

[112]Zlateski A, Jaroensri R, Sharma P, et al. On the importance of label quality for semantic segmentation; proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, F, 2018 [C].

[113]Bolch T, Kulkarni A, Kääb A, et al. The state and fate of Himalayan glaciers [J]. Science, 2012, 336(6079): 310-314.

[114]Kingma D P, Ba J J a p a. Adam: A method for stochastic optimization [J]. arXiv preprint arXiv, 2014, 1412: 6980.

[115]Zhou J, Li Z, Xing Q. Deriving glacier border information based on analysis of decorrelation in SAR interferometry [J]. Journal of Glaciology and Geocryology, 2010, 32(1): 133-138.

[116]Sakai A. Brief communication: Updated GAMDAM glacier inventory over high-mountain Asia [J]. The Cryosphere, 2019, 13(7): 2043-2049.

[117]李志杰, 王宁练, 陈安安, 等. 1993-2016 年喀喇昆仑山什约克流域冰川变化遥感监测 [J]. 冰川冻土, 2019, 41(4): 770-782.

[118]Hagg W, Mayer C, Lambrecht A, et al. Glacier changes in the big Naryn basin, Central Tian Shan [J]. Global Planetary Change, 2013, 110: 40-50.

[119]Le Bris R, Paul F. An automatic method to create flow lines for determination of glacier length: A pilot study with Alaskan glaciers [J]. Computers Geosciences, 2013, 52: 234-245.

[120]Shrestha A B, Aryal R J R e c. Climate change in Nepal and its impact on Himalayan glaciers [J]. Regional environmental change, 2011, 11: 65-77.

[121]Baolin L, Yichi Z, Chenghu Z. Remote sensing detection of glacier changes in Tianshan Mountains for the past 40 years [J]. Journal of Geographical sciences, 2004, 14: 296-302.

[122]刘时银, 上官冬辉, 丁永建, 等. 基于 RS 与 GIS 的冰川变化研究 [J]. 冰川冻土, 2004, 26(3): 244-253.

[123]Hjørland B. Indexing: concepts and theory [J]. KO Knowledge Organization, 2018, 45(7): 609-639.

[124]田洪阵, 杨太保, 刘沁萍. 基于遥感技术的近 40a 来敦德冰川变化和气候变化的关系研究 [J]. 冰川冻土, 2012, 34(2): 277-283.

[125]何毅, 杨太保, 冀琴, 等. 中亚阿拉套地区冰川退缩与气候变化关系研究 [J]. 干旱区地理, 2014, 37(5): 908-915.