研究论文

基于机器学习和动力过程的南极冰架崩解特征分析

  • 龙清云 ,
  • 张通 ,
  • 王彻 ,
  • 车涛 ,
  • 效存德
展开
  • 1.北京师范大学 地表过程与资源生态国家重点实验室,北京 100875
    2.首都师范大学 水资源安全 北京实验室,北京 100048
    3.中国科学院西北生态环境资源研究院,甘肃 兰州 730000
龙清云,硕士研究生,主要从事冰川研究. E-mail:qingyun@mail.bnu.edu.cn
张通,副教授,主要从事山地冰川和极地冰盖动力模拟和数值模式研发. E-mail:tzhang@bnu.edu.cn

收稿日期: 2024-04-09

  修回日期: 2024-07-13

  网络出版日期: 2024-08-23

基金资助

国家自然科学基金项目(42271133);中国科学院国际合作项目(121362KYSB20210024)

Analysis of the Features of Antarctic Ice Shelf Calving Based on Machine Learning and Dynamic Processes

  • Qingyun LONG ,
  • Tong ZHANG ,
  • Che WANG ,
  • Tao CHE ,
  • Cunde XIAO
Expand
  • 1.State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University, Beijing 100875, China
    2.Beijing Laboratory of Water Resources Security, Capital Normal University, Beijing 100048, China
    3.Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou 730000, China
LONG Qingyun, Master student, research area includes the study of ice sheets. E-mail: qingyun@mail.bnu.edu.cn
ZHANG Tong, Associate professor, research areas include the dynamics simulation and numerical model development of mountain glaciers and polar ice sheets. E-mail: tzhang@bnu.edu.cn

Received date: 2024-04-09

  Revised date: 2024-07-13

  Online published: 2024-08-23

Supported by

the National Natural Science Foundation of China(42271133);The International Cooperation Project of the Chinese Academy of Sciences(121362KYSB20210024)

摘要

冰架崩解对南极质量损失和动力过程有着直接影响,因此研究其变化的空间特征、环境条件和受控因子尤为重要。基于机器学习算法和冰盖动力模式,利用2005—2020年南极冰架崩解遥感数据、冰架表面裂隙数据、冰架支撑值、南极冰架损伤空间分布数据以及表面融化数据,结合机器学习二元分类算法,分析了18种影响冰架动力过程的特征要素的重要性,并测算7种不同机器学习算法的准确性。结果表明,随机森林算法在冰架崩解事件的二元分类中具备最高准确率,其中,冰架表面流速和冰架表面融水对冰架崩解具有较高的影响,表明利用冰架自身动力性质和外部环境影响因子进行冰架崩解的预测具有一定的可行性。后续需进一步耦合动力模式和机器学习算法,并构建相应的数值模式体系,来刻画更高时空分辨率的冰架崩解事件强度和范围。

本文引用格式

龙清云 , 张通 , 王彻 , 车涛 , 效存德 . 基于机器学习和动力过程的南极冰架崩解特征分析[J]. 地球科学进展, 2024 , 39(8) : 837 -846 . DOI: 10.11867/j.issn.1001-8166.2024.058

Abstract

Ice-shelf calving has a direct impact on Antarctic mass loss and dynamic processes, and it is particularly important to study its spatial characteristics, environmental conditions, and controlling factors. Based on the machine learning algorithms and ice sheet dynamic models, utilizing remote sensing data on Antarctic ice shelf calving from 2005 to 2020, ice shelf surface fracture data, ice shelf buttressing value, spatial distribution data of Antarctic ice shelf damage, and basal melting data, combined with machine learning binary classification, the importance of 18 characteristic elements influencing ice shelf dynamic processes was analyzed, and the accuracy of seven different machine learning algorithms was calculated. The results indicate that the random forest algorithm achieves the highest accuracy in the binary classification of ice shelf calving and that surface meltwater has a significant impact on ice shelf collapse, indicating the feasibility of using both the intrinsic dynamics of the ice shelf and external environmental factors for prediction. Subsequent efforts should further couple dynamic models with machine learning algorithms and establish corresponding numerical modeling systems to depict ice-shelf calving events with higher spatiotemporal resolutions in terms of intensity and extent.

参考文献

1 IPCC. Sea level rise and implications for low-lying islands, coasts and communities[M]// PO?RTNER H O, ROBERTS D C, MASSON-DELMOTTE V, et al. IPCC special report on the ocean and cryosphere in a changing climate. Cambridge,UK and New York,NY,USA: Cambridge University Press, 2019: 321-445.
2 RETWELL P, PRITCHARD H D, VAUGHAN D G, et al. Bedmap2: improved ice bed, surface and thickness datasets for Antarctica [J]. The Cryosphere, 2013, 7(1): 375-393.
3 RIGNOT E, MOUGINOT J, SCHEUCHL B, et al. Four decades of Antarctic Ice Sheet mass balance from 1979-2017[J]. Proceedings of the National Academy of Sciences of the United States of America, 2019, 116(4): 1 095-1 103.
4 IPCC. Ocean, cryosphere and sea level change[M]// MASSON-DELMOTTE V, ZHAI P, PIRANI A, et al. Climate change 2021: the physical science basis. Contribution of working group I to the sixth assessment report of the intergovernmental panel on climate change.Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press, 2021: 1 211-1 362.
5 EDWARDS T L, NOWICKI S, MARZEION B, et al. Projected land ice contributions to twenty-first-century sea level rise[J]. Nature, 2021,593: 74-82.
6 PRITCHARD H D, LIGTENBERG S R M, FRICKER H A, et al. Antarctic ice-sheet loss driven by basal melting of ice shelves[J]. Nature, 2012,484: 502-505.
7 DUPONT T K, ALLEY R B. Assessment of the importance of ice-shelf buttressing to ice-sheet flow[J]. Geophysical Research Letters, 2005,32. DOI: 10.1029/2004GL022024 .
8 DEPOORTER M A, BAMBER J L, GRIGGS J A, et al. Calving fluxes and basal melt rates of Antarctic ice shelves[J]. Nature, 2013,502: 89-92.
9 LIU Y, MOORE J C, CHENG X, et al. Ocean-driven thinning enhances iceberg calving and retreat of Antarctic ice shelves[J]. Proceedings of the National Academy of Sciences of the United States of America, 2015, 112(11): 3 263-3 268.
10 GREENE C A, GARDNER A S, SCHLEGEL N J, et al. Antarctic calving loss rivals ice-shelf thinning[J]. Nature, 2022, 609: 948-953.
11 NYE J F. The distribution of stress and velocity in glaciers and ice-sheets[J]. Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences, 1957,239(1 216): 113-133.
12 NICK F M, van der VEEN C J, VIELI A, et al. A physically based calving model applied to marine outlet glaciers and implications for the glacier dynamics[J]. Journal of Glaciology, 2010, 56(199): 781-794.
13 TODD J, CHRISTOFFERSEN P. Are seasonal calving dynamics forced by buttressing from ice mélange or undercutting by melting?Outcomes from full-stokes simulations of store glacier, West Greenland[J]. The Cryosphere, 2014,8(6): 2 353-2 365.
14 TODD J, CHRISTOFFERSEN P, ZWINGER T, et al. Sensitivity of a calving glacier to ice-ocean interactions under climate change: new insights from a 3-D full-Stokes model[J]. The Cryosphere, 2019,13(6): 1 681-1 694.
15 DUDDU R, JIMéNEZ S, BASSIS J. A non-local continuum poro-damage mechanics model for hydrofracturing of surface crevasses in grounded glaciers[J]. Journal of Glaciology, 2020,66(257): 415-429.
16 KACHUCK S B, WHITCOMB M, BASSIS J N, et al. Simulating ice-shelf extent using damage mechanics[J]. Journal of Glaciology, 2022,68: 987-998.
17 LEVERMANN A, ALBRECHT T, WINKELMANN R, et al. Kinematic first-order calving law implies potential for abrupt ice-shelf retreat[J]. The Cryosphere, 2012,6(2): 273-286.
18 SELBESO?LU M O, BAKIRMAN T, VASSILEV O, et al. Mapping of glaciers on horseshoe island, Antarctic peninsula, with deep learning based on high-resolution orthophoto[J]. Drones, 2023,7(2). DOI:10.3390/drones7020072 .
19 RAMANATH T S. Mapping the grounding line of Antarctica in SAR interferograms with machine learning techniques[Z]. Munich: Technische Universit?t München, 2022.
20 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). DOI:10.3390/rs11212529 .
21 BOLIBAR J, RABATEL A, GOUTTEVIN I, et al. Deep learning applied to glacier evolution modelling[J]. The Cryosphere, 2020,14(2): 565-584.
22 JOUVET G. Inversion of a Stokes glacier flow model emulated by deep learning[J]. Journal of Glaciology, 2023,69(273): 13-26.
23 DONG S, TANG X Y, GUO J X, et al. EisNet: extracting bedrock and internal layers from radiostratigraphy of ice sheets with machine learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022,60. DOI:10.1109/TGRS.2021.3136648 .
24 NIU L H, TANG X Y, YANG S H, et al. Detection of Antarctic surface meltwater using sentinel-2 remote sensing images via U-net with attention blocks: a case study over the amery ice shelf[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023,61. DOI: 10.1109/TGRS.2021.3136648 .
25 LAI C Y, KINGSLAKE J, WEARING M G, et al. Vulnerability of Antarctica’s ice shelves to meltwater-driven fracture[J]. Nature, 2020, 584: 574-578.
26 HOFFMAN M J, PEREGO M, PRICE S F, et al. MPAS-Albany Land Ice (MALI): a variable-resolution ice sheet model for Earth system modeling using Voronoi grids[J]. Geoscientific Model Development, 2018,11(9): 3 747-3 780.
27 BASSIS J N, MA Y. Evolution of basal crevasses links ice shelf stability to ocean forcing[J]. Earth and Planetary Science Letters, 2015,409: 203-211.
28 LAI C Y, KINGSLAKE J, WEARING M G, et al. Vulnerability of Antarctica’s ice shelves to meltwater-driven fracture[J]. Nature, 2020, 584(7 822): 574-578.
29 GUDMUNDSSON G H. Ice-shelf buttressing and the stability of marine ice sheets[J]. The Cryosphere, 2013,7(2): 647-655.
30 QI M Z, LIU Y, CHENG X, Annual iceberg calving dataset of the Antarctic ice shelves (2005-2020)[DS]. National Tibetan Plateau Data Center, 2021.DOI:10.11888/Glacio.tpdc.271250 .
31 QI M Z, LIU Y, LIU J P, et al. A 15-year circum-Antarctic iceberg calving dataset derived from continuous satellite observations[J]. Earth System Science Data, 2021, 13(9): 4 583-4 601.
32 QI M Z, LIU Y, LIN Y J, et al. Efficient location and extraction of the iceberg calved areas of the Antarctic ice shelves[J]. Remote Sensing, 2020,12(16). DOI:10.3390/rs12162658 .
33 NO?L B, van WESSEM J M, WOUTERS B, et al. Higher Antarctic ice sheet accumulation and surface melt rates revealed at 2 km resolution[J]. Nature Communications, 2023,14.DOI:10.1038/s41467-023-43584-6 .
34 RIGNOT E, JACOBS S, MOUGINOT J, et al. Ice-shelf melting around Antarctica[J]. Science, 2013,341(6 143): 266-270.
35 COX D R. The regression analysis of binary sequences[J]. Journal of the Royal Statistical Society: Series B (Methodological), 1958, 20(2): 215-232.
36 CORTES C, VAPNIK V. Support-vector networks[J]. Machine Learning, 1995, 20(3): 273-297
37 ZHANG H. The optimality of naive Bayes[C]// Proceedings of 17th international Florida artificial intelligence research society conference. Miami Beach Florida USA, 2004: 562-567.
38 BREIMAN L. Random forests[J]. Machine Learning, 2001,45: 5-32.
39 CHEN T Q, GUESTRIN C. XGBoost: a scalable tree boosting system[C]// Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. San Francisco California USA: ACM, 2016: 785-794.
40 KE G, MENG Q, FINLEY T, et al. Lightgbm: a highly efficient gradient boosting decision tree[C]// Proceedings of the 31st annual conference on neural information processing systems. Long Beach California USA, 2017: 3 146-3 154.
41 LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015,521: 436-444.
42 SCAMBOS T, FRICKER H A, LIU C C, et al. Ice shelf disintegration by plate bending and hydro-fracture: satellite observations and model results of the 2008 Wilkins ice shelf break-ups[J]. Earth and Planetary Science Letters, 2009, 280(1/2/3/4): 51-60.
43 FRANCIS D, MATTINGLY K S, LHERMITTE S, et al. Atmospheric extremes caused high oceanward sea surface slope triggering the biggest calving event in more than 50 years at the Amery Ice Shelf[J]. The Cryosphere, 2021, 15(5): 2 147-2 165.
文章导航

/