Application of Artificial Intelligence in Glacier Studies: A State-of-the-Art Review
<p>Flowchart for illustration of the methods used in the current review work.</p> "> Figure 2
<p>Glacier regions in the world (modified from [<a href="#B38-water-16-02272" class="html-bibr">38</a>]).</p> "> Figure 3
<p>Classification of the collected research works based on glacier regions, type of glacier study, and AI models using a Sankey Diagram.</p> "> Figure 4
<p>Yearly classification of AI algorithms applied for glacier studies.</p> "> Figure 5
<p>Flowchart for glacier mapping by Zhang et al. [<a href="#B40-water-16-02272" class="html-bibr">40</a>].</p> "> Figure 6
<p>Outline of the methodology by Mohajerani et al. [<a href="#B41-water-16-02272" class="html-bibr">41</a>].</p> "> Figure 7
<p>Training and testing sites [<a href="#B42-water-16-02272" class="html-bibr">42</a>].</p> "> Figure 8
<p>Flowchart of methodology by Khan et al. [<a href="#B43-water-16-02272" class="html-bibr">43</a>].</p> "> Figure 9
<p>Proposed flowchart of the methodology [<a href="#B45-water-16-02272" class="html-bibr">45</a>].</p> "> Figure 10
<p>The illustration of CNN with a heatmap output for RG evaluation [<a href="#B46-water-16-02272" class="html-bibr">46</a>].</p> "> Figure 11
<p>Study areas: La Laguna catchment, Chile, and Poiqu catchment, Central Himalaya, by Robson et al. [<a href="#B46-water-16-02272" class="html-bibr">46</a>].</p> "> Figure 12
<p>Selected area for the study by Xie et al. [<a href="#B48-water-16-02272" class="html-bibr">48</a>], Central Karakoram.</p> "> Figure 13
<p>GlacierNet architecture [<a href="#B48-water-16-02272" class="html-bibr">48</a>].</p> "> Figure 14
<p>ANN architecture U-Net used to map rock glaciers in Austria by Erharter et al. [<a href="#B50-water-16-02272" class="html-bibr">50</a>].</p> "> Figure 15
<p>RG examples from North Tyrolean “Wurmeskar”, Austria (left), and RG probability map based on ANN (right) developed by Erharter et al. [<a href="#B50-water-16-02272" class="html-bibr">50</a>].</p> "> Figure 16
<p>A general representation of the workflow for the GLNet technique [<a href="#B17-water-16-02272" class="html-bibr">17</a>].</p> "> Figure 17
<p>Glacial lakes in the Eastern Himalaya’s test site 3 were mapped and compared with the reference data to identify errors in false positives and false negatives [<a href="#B17-water-16-02272" class="html-bibr">17</a>].</p> "> Figure 18
<p>The flowchart of the model [<a href="#B52-water-16-02272" class="html-bibr">52</a>].</p> "> Figure 19
<p>Flowchart of the hybrid feature selection mechanism for automatic object-based glacier mapping [<a href="#B53-water-16-02272" class="html-bibr">53</a>].</p> "> Figure 20
<p>Flow chart of the developed approach. The steps include dataset pre-processing, reference vector dataset generation, convolutional neural network classification, and object-based image analysis refinement [<a href="#B54-water-16-02272" class="html-bibr">54</a>].</p> "> Figure 21
<p>Workflow diagram [<a href="#B55-water-16-02272" class="html-bibr">55</a>].</p> "> Figure 22
<p>Structure and workflow of the ALPGM by Bolibar et al. [<a href="#B57-water-16-02272" class="html-bibr">57</a>].</p> "> Figure 23
<p>Workflow of this study by Yang et al. [<a href="#B60-water-16-02272" class="html-bibr">60</a>].</p> "> Figure 24
<p>DeepLabv3+ semantic segmentation model and ResNet-50 residual unit [<a href="#B60-water-16-02272" class="html-bibr">60</a>].</p> "> Figure 25
<p>General flowchart of the proposed method [<a href="#B61-water-16-02272" class="html-bibr">61</a>].</p> "> Figure 26
<p>Connections between the model elements and the input data of IGM by Jouvet et al. [<a href="#B62-water-16-02272" class="html-bibr">62</a>].</p> ">
Abstract
:1. Introduction
2. Review Approach and Overview of the Collected Works
3. AI-Based Glacier Studies
3.1. AI for Glacier Inventory and Mapping
3.2. AI for Monitoring of Glacier Evolution
3.3. AI for Snow/Ice Differentiation
3.4. AI for Ice Dynamics Modeling
4. Discussion
5. Conclusions
- All the reviewed works are classified by the purpose of their research. Among them, glacier mapping is the most studied area, followed by glacier evolution, ice/snow differentiation, and ice dynamic modeling.
- For AI-based glacier evolution studies, the availability of glacier data in terms of time-frequency and overall measured duration is highly important to accurately capture the temporal evolution of glaciers.
- Ice/snow differentiation and ice dynamic modeling are in their early stages regarding AI-based studies. However, the methods developed so far show promising accuracy and require further advancements.
- Methods such as random forest (RF), K-nearest neighbors (KNN), support vector machines (SVMs), and decision trees (DTs) have been foundational. Among them, RF often outperforms other traditional methods in accuracy and robustness, especially for glacier mapping studies.
- Recent studies in glacier mapping have developed CNN-based models, notably U-net and DeepLabV3+, which showed enhanced accuracy in glacier mapping. However, the robustness of these models needs to be tested with appropriate methods, such as hyperparameter optimization, to fine-tune parameters like the learning rate, batch size, number of layers, and dropout rate.
- Hybrid methods that combine two ML and/or DL methods generally show better performance compared to single methods. However, the compatibility and integrability of different methods in hybrid solutions have not been thoroughly studied yet, and comparative studies among hybrid methods are lacking.
- Overall, AI-based glacier research has notably been gaining the attention of scientists and requires more detailed studies. The consistency of AI-based methods needs to be further evaluated, particularly when training on one glacier dataset and testing on a different dataset. Additionally, the impact of training and testing dataset sizes, as well as the remote sensing technologies used to obtain these datasets, should be assessed.
- More generalized AI-based glacier assessment tools, particularly for worldwide glacier mapping and inventory, appear to be a promising direction for future research.
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Zemp, M.; Huss, M.; Thibert, E.; Eckert, N.; McNabb, R.; Huber, J.; Barandun, M.; Machguth, H.; Nussbaumer, S.U.; Gärtner-Roer, I.; et al. Global Glacier Mass Changes and Their Contributions to Sea-Level Rise from 1961 to 2016. Nature 2019, 568, 382–386. [Google Scholar] [CrossRef] [PubMed]
- Wang, S. Opportunities and Threats of Cryosphere Change to the Achievement of UN 2030 SDGs. Humanit. Soc. Sci. Commun. 2024, 11, 44. [Google Scholar] [CrossRef]
- Rasul, G.; Molden, D. The Global Social and Economic Consequences of Mountain Cryospheric Change. Front. Environ. Sci. 2019, 7, 91. [Google Scholar] [CrossRef]
- Imdieke, A.; Pearson, P. Accelerated Glacier Retreat in the Himalayas Jeopardizes South Asian Agriculture—ICCI—International Cryosphere Climate Initiative. Available online: https://iccinet.org/accelerated-glacier-retreat-in-the-himalayas-jeopardizes-south-asian-agriculture/ (accessed on 13 March 2024).
- Puspitarini, H.D.; François, B.; Zaramella, M.; Brown, C.; Borga, M. The Impact of Glacier Shrinkage on Energy Production from Hydropower-Solar Complementarity in Alpine River Basins. Sci. Total Environ. 2020, 719. [Google Scholar] [CrossRef]
- Losapio, G.; Cerabolini, B.E.L.; Maffioletti, C.; Tampucci, D.; Gobbi, M.; Caccianiga, M. The Consequences of Glacier Retreat Are Uneven Between Plant Species. Front. Ecol. Evol. 2021, 8. [Google Scholar] [CrossRef]
- Milner, A.M.; Khamis, K.; Battin, T.J.; Brittain, J.E.; Barrand, N.E.; Füreder, L.; Cauvy-Fraunié, S.; Gíslason, G.M.; Jacobsen, D.; Hannah, D.M.; et al. Glacier Shrinkage Driving Global Changes in Downstream Systems. Proc. Natl. Acad. Sci. USA 2017, 114, 9770–9778. [Google Scholar] [CrossRef] [PubMed]
- WGMS, and National Snow and Ice Data Center (comps.). 1999, Updated 2012. World Glacier Inventory, Version 1. Boulder, Colorado USA. NSIDC: National Snow and Ice Data Center. Available online: https://nsidc.org/data/g01130/versions/1 (accessed on 5 August 2024).
- NSIDC. National Snow and Ice Data Center RGI 7.0 Consortium. Randolph Glacier Inventory—A Dataset of Global Glacier Outlines, Version 7.0; NSIDC: Boulder, CO, USA, 2023. [Google Scholar]
- Friedl, P.; Weiser, F.; Fluhrer, A.; Braun, M.H. Remote Sensing of Glacier and Ice Sheet Grounding Lines: A Review. Earth-Sci. Rev. 2020, 201. [Google Scholar] [CrossRef]
- Winsvold, S.H.; Kääb, A.; Nuth, C. Regional Glacier Mapping Using Optical Satellite Data Time Series. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2016, 9. [Google Scholar] [CrossRef]
- Chouksey, A.; Thakur, P.K.; Sahni, G.; Swain, A.K.; Aggarwal, S.P.; Kumar, A.S. Mapping and Identification of Ice-Sheet and Glacier Features Using Optical and SAR Data in Parts of Central Dronning Maud Land (cDML), East Antarctica. Polar Sci. 2021, 30. [Google Scholar] [CrossRef]
- Lu, Y.; Zhang, Z.; Kong, Y.; Hu, K. Integration of Optical, SAR and DEM Data for Automated Detection of Debris-Covered Glaciers over the Western Nyainqentanglha Using a Random Forest Classifier. Cold Reg. Sci. Technol. 2022, 193. [Google Scholar] [CrossRef]
- Peng, Y.; He, J.; Yuan, Q.; Wang, S.; Chu, X.; Zhang, L. Automated Glacier Extraction Using a Transformer Based Deep Learning Approach from Multi-Sensor Remote Sensing Imagery. ISPRS J. Photogramm. Remote Sens. 2023, 202, 303–313. [Google Scholar] [CrossRef]
- Malenovský, Z.; Rott, H.; Cihlar, J.; Schaepman, M.E.; García-Santos, G.; Fernandes, R.; Berger, M. Sentinels for Science: Potential of Sentinel-1, -2, and -3 Missions for Scientific Observations of Ocean, Cryosphere, and Land. Remote Sens. Environ. 2012, 120, 91–101. [Google Scholar] [CrossRef]
- Rott, H.; Mätzler, C. Possibilities and Limits of Synthetic Aperture Radar for Snow and Glacier Surveying. Ann. Glaciol. 1987, 9, 195–199. [Google Scholar] [CrossRef]
- Kaushik, S.; Singh, T.; Joshi, P.K.; Dietz, A.J. Automated Mapping of Glacial Lakes Using Multisource Remote Sensing Data and Deep Convolutional Neural Network. Int. J. Appl. Earth Obs. Geoinf. 2022, 115, 103085. [Google Scholar] [CrossRef]
- Rastner, P.; Strozzi, T.; Paul, F. Fusion of Multi-Source Satellite Data and DEMs to Create a New Glacier Inventory for Novaya Zemlya. Remote Sens. 2017, 9, 1122. [Google Scholar] [CrossRef]
- Zhang, L.; Zhang, Q.; Du, B.; Huang, X.; Tang, Y.Y.; Tao, D. Simultaneous Spectral-Spatial Feature Selection and Extraction for Hyperspectral Images. IEEE Trans. Cybern. 2018, 48, 16–28. [Google Scholar] [CrossRef] [PubMed]
- Zhang, L.; Zhang, L.; Du, B. Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art. IEEE Geosci. Remote Sens. Mag. 2016, 4, 22–40. [Google Scholar] [CrossRef]
- Jamil, A.; Bayram, B. Tree Species Extraction and Land Use/Cover Classification from High-Resolution Digital Orthophoto Maps. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2018, 11, 89–94. [Google Scholar] [CrossRef]
- Qian, Y.; Zhou, W.; Yan, J.; Li, W.; Han, L. Comparing Machine Learning Classifiers for Object-Based Land Cover Classification Using Very High Resolution Imagery. Remote Sens. 2015, 7, 153–168. [Google Scholar] [CrossRef]
- Zarea, A.; Mohammadzadeh, A. A Novel Building and Tree Detection Method from LiDAR Data and Aerial Images. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2016, 9, 1864–1875. [Google Scholar] [CrossRef]
- Zhang, L.; You, J. A Spectral Clustering Based Method for Hyperspectral Urban Image. In Proceedings of the 2017 Joint Urban Remote Sensing Event, JURSE 2017, Dubai, United Arab Emirates, 6–8 March 2017. [Google Scholar]
- Wu, Y.; Jiang, B.; Wang, Y. Incipient Winding Fault Detection and Diagnosis for Squirrel-Cage Induction Motors Equipped on CRH Trains. ISA Trans. 2020, 99, 488–495. [Google Scholar] [CrossRef]
- Amin, A.A.; Sajid Iqbal, M.; Hamza Shahbaz, M. Development of Intelligent Fault-Tolerant Control Systems with Machine Learning, Deep Learning, and Transfer Learning Algorithms: A Review. Expert Syst. Appl. 2024, 238, 121956. [Google Scholar] [CrossRef]
- Pal, N.R.; Pal, S.K. A Review on Image Segmentation Techniques. Pattern Recognit. 1993, 26, 1277–1294. [Google Scholar] [CrossRef]
- Paul, F.; Huggel, C.; Kääb, A. Combining Satellite Multispectral Image Data and a Digital Elevation Model for Mapping Debris-Covered Glaciers. Remote Sens. Environ. 2004, 89, 510–518. [Google Scholar] [CrossRef]
- Bibi, L.; Khan, A.A.; Khan, G.; Ali, K.; Hassan, S.N.U.; Qureshi, J.; Jan, I.U. Snow Cover Trend Analysis Using Modis Snow Products: A Case of Shayok River Basin in Northern Pakistan. J. Himal. Earth Sci. 2019, 52, 145–160. [Google Scholar]
- Nijhawan, R.; Garg, P.; Thakur, P. A Comparison of Classification Techniques for Glacier Change Detection Using Multispectral Images. Perspect. Sci. 2016, 8, 377–380. [Google Scholar] [CrossRef]
- Alifu, H.; Tateishi, R.; Johnson, B. A New Band Ratio Technique for Mapping Debris-Covered Glaciers Using Landsat Imagery and a Digital Elevation Model. Int. J. Remote Sens. 2015, 36, 2063–2075. [Google Scholar] [CrossRef]
- Gindraux, S.; Boesch, R.; Farinotti, D. Accuracy Assessment of Digital Surface Models from Unmanned Aerial Vehicles’ Imagery on Glaciers. Remote Sens. 2017, 9, 186. [Google Scholar] [CrossRef]
- Han, W.; Zhang, X.; Wang, Y.; Wang, L.; Huang, X.; Li, J.; Wang, S.; Chen, W.; Li, X.; Feng, R.; et al. A Survey of Machine Learning and Deep Learning in Remote Sensing of Geological Environment: Challenges, Advances, and Opportunities. ISPRS J. Photogramm. Remote Sens. 2023, 202, 87–113. [Google Scholar] [CrossRef]
- Barillaro, L. Deep Learning Platforms: PyTorch. In Reference Module in Life Sciences; Elsevier: Amsterdam, The Netherlands, 2024. [Google Scholar] [CrossRef]
- Janardhanan, P.S. Project Repositories for Machine Learning with TensorFlow. Procedia Comput. Sci. 2020, 171, 188–196. [Google Scholar] [CrossRef]
- Barillaro, L. Deep Learning Platforms: Keras. In Reference Module in Life Sciences; Elsevier: Amsterdam, The Netherlands, 2024. [Google Scholar] [CrossRef]
- Zhang, H.K.; Qiu, S.; Suh, J.W.; Luo, D.; Zhu, Z. Machine Learning and Deep Learning in Remote Sensing Data Analysis. Reference Module in Earth Systems and Environmental Sciences; Elsevier: Amsterdam, The Netherlands, 2024. [Google Scholar] [CrossRef]
- Intergovernmental Panel on Climate Change (IPCC). Climate Change 2013: The Physical Science Basis; Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Eds.; Cambridge University Press: Cambridge, UK; New York, NY, USA, 2013; 1535p. [Google Scholar]
- Fenza, G.; Gallo, M.; Loia, V.; Orciuoli, F.; Herrera-Viedma, E. Data Set Quality in Machine Learning: Consistency Measure Based on Group Decision Making. Appl. Soft Comput. 2021, 106, 107366. [Google Scholar] [CrossRef]
- Zhang, J.; Jia, L.; Menenti, M.; Hu, G. Glacier Facies Mapping Using a Machine-Learning Algorithm: The Parlung Zangbo Basin Case Study. Remote Sens. 2019, 11, 452. [Google Scholar] [CrossRef]
- Mohajerani, Y.; Wood, M.; Velicogna, I.; Rignot, E. Detection of Glacier Calving Margins with Convolutional Neural Networks: A Case Study. Remote Sens. 2019, 11, 74. [Google Scholar] [CrossRef]
- Baumhoer, C.A.; Dietz, A.J.; Kneisel, C.; Kuenzer, C. Automated Extraction of Antarctic Glacier and Ice Shelf Fronts from Sentinel-1 Imagery Using Deep Learning. Remote Sens. 2019, 11, 2529. [Google Scholar] [CrossRef]
- Khan, A.A.; Jamil, A.; Hussain, D.; Taj, M.; Jabeen, G.; Malik, M.K. Machine-Learning Algorithms for Mapping Debris-Covered Glaciers: The Hunza Basin Case Study. IEEE Access 2020, 8, 12725–12734. [Google Scholar] [CrossRef]
- Zhang, E.; Liu, L.; Huang, L.; Ng, K.S. An Automated, Generalized, Deep-Learning-Based Method for Delineating the Calving Fronts of Greenland Glaciers from Multi-Sensor Remote Sensing Imagery. Remote Sens. Environ. 2021, 254, 112265. [Google Scholar] [CrossRef]
- Alifu, H.; Vuillaume, J.F.; Johnson, B.A.; Hirabayashi, Y. Machine-Learning Classification of Debris-Covered Glaciers Using a Combination of Sentinel-1/-2 (SAR/Optical), Landsat 8 (Thermal) and Digital Elevation Data. Geomorphology 2020, 369, 107365. [Google Scholar] [CrossRef]
- Robson, B.A.; Bolch, T.; MacDonell, S.; Hölbling, D.; Rastner, P.; Schaffer, N. Automated Detection of Rock Glaciers Using Deep Learning and Object-Based Image Analysis. Remote Sens. Environ. 2020, 250, 112033. [Google Scholar] [CrossRef]
- Lu, Y.; Zhang, Z.; Shangguan, D.; Yang, J. Novel Machine Learning Method Integrating Ensemble Learning and Deep Learning for Mapping Debris-Covered Glaciers. Remote Sens. 2021, 13, 2595. [Google Scholar] [CrossRef]
- Xie, Z.; Asari, V.K.; Haritashya, U.K. Evaluating Deep-Learning Models for Debris-Covered Glacier Mapping. Appl. Comput. Geosci. 2021, 12, 100071. [Google Scholar] [CrossRef]
- Xie, Z.; Haritashya, U.K.; Asari, V.K.; Bishop, M.P.; Kargel, J.S.; Aspiras, T.H. GlacierNet2: A Hybrid Multi-Model Learning Architecture for Alpine Glacier Mapping. Int. J. Appl. Earth Obs. Geoinf. 2022, 112, 102921. [Google Scholar] [CrossRef]
- Erharter, G.H.; Wagner, T.; Winkler, G.; Marcher, T. Machine Learning—An Approach for Consistent Rock Glacier Mapping and Inventorying – Example of Austria. Appl. Comput. Geosci. 2022, 16, 100093. [Google Scholar] [CrossRef]
- Tian, S.; Dong, Y.; Feng, R.; Liang, D.; Wang, L. Mapping Mountain Glaciers Using an Improved U-Net Model with cSE. Int. J. Digit. Earth 2022, 15, 463–477. [Google Scholar] [CrossRef]
- Sood, V.; Tiwari, R.K.; Singh, S.; Kaur, R.; Parida, B.R. Glacier Boundary Mapping Using Deep Learning Classification over Bara Shigri Glacier in Western Himalayas. Sustainability 2022, 14, 13485. [Google Scholar] [CrossRef]
- Sharda, S.; Srivastava, M.; Gusain, H.S.; Sharma, N.K.; Bhatia, K.S.; Bajaj, M.; Kaur, H.; Zawbaa, H.M.; Kamel, S. A Hybrid Machine Learning Technique for Feature Optimization in Object-Based Classification of Debris-Covered Glaciers. Ain Shams Eng. J. 2022, 13, 101809. [Google Scholar] [CrossRef]
- Thomas, D.J.; Robson, B.A.; Racoviteanu, A. An Integrated Deep Learning and Object-Based Image Analysis Approach for Mapping Debris-Covered Glaciers. Front. Remote Sens. 2023, 4. [Google Scholar] [CrossRef]
- Hu, Y.; Liu, L.; Huang, L.; Zhao, L.; Wu, T.; Wang, X.; Cai, J. Mapping and Characterizing Rock Glaciers in the Arid Western Kunlun Mountains Supported by InSAR and Deep Learning. J. Geophys. Res. Earth Surf. 2023, 128, e2023JF007206. [Google Scholar] [CrossRef]
- Bolibar, J.; Rabatel, A.; Gouttevin, I.; Zekollari, H.; Galiez, C. Nonlinear Sensitivity of Glacier Mass Balance to Future Climate Change Unveiled by Deep Learning. Nat. Commun. 2022, 13, 409. [Google Scholar] [CrossRef]
- Bolibar, J.; Rabatel, A.; Gouttevin, I.; Galiez, C.; Condom, T.; Sauquet, E. Deep Learning Applied to Glacier Evolution Modelling. Cryosphere 2020, 14, 565–584. [Google Scholar] [CrossRef]
- Ambinakudige, S.; Intsiful, A. Estimation of Area and Volume Change in the Glaciers of the Columbia Icefield, Canada Using Machine Learning Algorithms and Landsat Images. Remote Sens. Appl. Soc. Environ. 2022, 26, 100732. [Google Scholar] [CrossRef]
- Rajat, S.; Rajeshwar Singh, B.; Prakash, C.; Anita, S. Glacier Retreat in Himachal from 1994 to 2021 Using Deep Learning. Remote Sens. Appl. Soc. Environ. 2022, 28, 100870. [Google Scholar] [CrossRef]
- Yang, S.; Mei, G.; Zhang, Y. Susceptibility Analysis of Glacier Debris Flow by Investigating Glacier Changes Based on Remote Sensing Imagery and Deep Learning: A Case Study. Nat. Hazards Res. 2023; in press. [Google Scholar] [CrossRef]
- Prieur, C.; Rabatel, A.; Thomas, J.B.; Farup, I.; Chanussot, J. Machine Learning Approaches to Automatically Detect Glacier Snow Lines on Multi-Spectral Satellite Images. Remote Sens. 2022, 14, 3868. [Google Scholar] [CrossRef]
- Jouvet, G.; Cordonnier, G.; Kim, B.; Lüthi, M.; Vieli, A.; Aschwanden, A. Deep Learning Speeds up Ice Flow Modelling by Several Orders of Magnitude. J. Glaciol. 2022, 68, 651–664. [Google Scholar] [CrossRef]
Author | Location | Glacier Location Name | Studied Glacier Types | Classification by GLIMS Manual | AI Model | Parameters | Dataset Size | Accuracy | Software |
---|---|---|---|---|---|---|---|---|---|
Glacier inventory and mapping | |||||||||
2019 Zhang et al. [40] | Parlung Zangbo Basin, China | Tibetian Plateau glacier |
| N/A | Random forest (RF) |
| 2755 | RF-98.6% (ovearall) | EnMAP-Box + DLL |
2019 Mohajerani et al. [41] | Greenland | Jakobshavn, Sverdrup, Kangerlussuaq, Helheim |
| N/A | U-Net |
| Training data: images from Jakobshavn, Sverdrup and Kangerlussuaq. Test data: images from Helheim glacier | Mean deviation of 96.3 m from the true calving fonts | Python |
2019 Baumhoer et al. [42] | Antarctica |
|
| N/A | Modified U-Net |
| 38 pre-processed Sentinel-1 scenes 90m resolution TanDEM-X | Average f1-score = 90% | N/A |
2020 Khan et al. [43] | Hunza Basin, Pakistan | Batura glacier |
| N/A |
|
| 2,688,723 pixels Training: 70% Testing: 30% | Kappa: SVM = 0.89 ANN = 0.92 RF = 0.95 f-measure: SVM = 91.86% ANN = 92.05% RF = 95.06% | N/A |
2021 Zhang et al. [44] | Greenland | Jakobshavn Isbræ, Kangerlussuaq, Helheim glaciers | Tidewater outlet glaciers | Tidewater outlet glacier |
| Optical:
| Training: 110 Landsat-8, 13 ALOS-1, 76 TSX, 140 Sentinel-1 Testing: 74 Landsat-8, 52 Sentinel-2, 48 Envisat, 17 TSX, 90 Sentinel-1, 14 ALOS-2 | Test-error studies: DRN-DeepLabv3+ is the lowest Refer to Table 3 from [44] for full test results | Python Open-source in GitHub: https://github.com/enzezhang/FrontDL3 (accessed on 7 July 2024) |
2020 H. Alifu et al. [45] | Karakoram—Pakistan Shaksgam Valley, China | North-western Karakoram region and Shaksgam Valley glaciers | Debris-covered glaciers | Valley, Mountain glaciers | Machine learning classifiers (MLC):
|
| Area 1: 2000 to 20,000 points. Area 2: 20,000 points | RF-97% | Python |
2020 Robson et al. [46] | Chilean Andes, Chile Central Himalaya | La Laguna catchment Poilu catchment | Rock glaciers | Mountain glaciers | CNN with OBIA |
| Not clear |
| Google Tensorflow |
2021 Lu et al. [47] | China | High Mountain Asia | Debris-covered glaciers | Mountain glacier | RF CNN |
| Eastern Pamir: 7499 samples Nyainqêntanglha: 3099 samples | Eastern Pamir and Nyainqentanglha User’s accuracy:
| Python |
2021 Xie et al. [48] | Kashmir Region. Nepal region | Karakoram glaciers Nepal glaciers | DCG | Mountain, Valley glaciers |
|
| N/A | IOU:
| N/A |
2022 Xie et al. [49] | Northern Pakistan | Central Karakoram | DCG | Mountain, Valley glaciers | CNN |
| Accucary:
| ||
2022 Erharter [50] | Austria Apls | Vienna, Burgenland, Lower Austria, Upper Austria | RG | Mountain glaciers | ANN with U-net |
| 5769 RGs:
|
| Python, Keras |
2022 Kaushik et al. [17] | 12 sites across Himalaya | Himalayan glaciers | Glacier lake | N/A | GLNet—Deep convolutional neural network |
| 660 images | Accuracy = 0.98 Precision = 0.95 REcall f-score = 0.95 | |
2022 Tian et al. [51] | Pamir Plateau | RG | Mountain glaciers | Channel attention U-net (U-net+cSE) |
| 7821 images | Accuracy: U-net = 0.9756 GlacierNet = 0.9689 U-net + cSE = 0.9774 | ||
2022 Sood et al. [52] | Bara Shigri, Himachal Pradesh, India | Valley glacier | ENVINet5 |
| Accuracy = 91.89% Kappa = 0.8778 | ||||
2022 Sharda et al. [53] | Karakoram Range, Pakistan | DCG | Mountain, Valley, Icefields |
|
| up to 99.8% |
| ||
2023 Peng et al. [14] | Qilian Mountains, China | Not specified | U-net with LGT encoder and LGCB decoder |
| 2072 glaciers:
| Accuracy: U-Net: 0.725 DeepLab V3+: 0.924 Attention DeepLab V3+: 0.960 Swin Transformer: 0.962 Proposed model: 0.972 | NA | ||
2023 Thomas et al. [54] | Khumbu—Nepal, China Manaslu—Nepal Hunza—Pakistan | DCG | Valley, Mountain, Icefields, Cirque | CNN with OBIA classification |
| 69,500 samples Supraglacial debris-20,000 Non-glacial material-20,000 Vegetation-10,000 Lakes-7500 Clean ice glacier-5000 Snow cover-5000 Shadows-2000 |
| Trimble’s eCognition Developer 10.2 TensorFlow library | |
2023 Hu et al. [55] | Western Kunlun Mountains, China | Western Kunlun Mountains | Rock glaciers | N/A | DeepLabv3+ with Xception71 backbone |
| Training (90%): 2007 images; Validation (10%): 223 images; | N/A | N/A |
Monitoring of glacier evolution | |||||||||
2022, 2020 Bolibar et al. [56,57] | French Alps | Écrins, Vanoise, Mont-Blanc glaciers | Mountain Glaciers | Mountain Glacier | ALpine Parameterized Glacier Model (ALPGM) based on ANN |
| 32 glaciers in French Alps | 47% in space 58% in time | Python |
2022 Ambinakudige and Intsiful [58] | Columbia Icefields, Canada | Icefields | SVM RF MLC |
| 1985, 1991, 2013, and 2020 Landsat satellite images 70% training 30% validation | Accuracy: RF = 99.8% MLC = 99.7% SVM = 99.7% Kappa: RF = 0.995 MLC = 0.993 SVM = 0.994 | N/A | ||
2022 Rajat et al. [59] | Himachal Pradesh, India | Himalayan mountains | Mountain glaciers | U-Net |
| 75% training 25% validation | F1 score: 95% | N/A | |
2023 Yang et al. [60] | Southeast Tibet | Zelongnong ravine | Glacier Debris Flow susceptibility | Valley, Cirque |
|
|
| ||
Snow/ice differentiation | |||||||||
2022 Prieur C. [61] | Zermatt, Switzerland | Mont Rose massif | Temperate glacier/snow lines | Temperate glaciers |
|
|
|
| - |
Ice dynamics modeling | |||||||||
2021 Jouvet et al. [62] |
| Icefields, Valley glaciers | Instructed Glacier Model (IGM) using CNN | - | ≈20 direct speedup using CNN | Python |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Nurakynov, S.; Merekeyev, A.; Baygurin, Z.; Sydyk, N.; Akhmetov, B. Application of Artificial Intelligence in Glacier Studies: A State-of-the-Art Review. Water 2024, 16, 2272. https://doi.org/10.3390/w16162272
Nurakynov S, Merekeyev A, Baygurin Z, Sydyk N, Akhmetov B. Application of Artificial Intelligence in Glacier Studies: A State-of-the-Art Review. Water. 2024; 16(16):2272. https://doi.org/10.3390/w16162272
Chicago/Turabian StyleNurakynov, Serik, Aibek Merekeyev, Zhaksybek Baygurin, Nurmakhambet Sydyk, and Bakytzhan Akhmetov. 2024. "Application of Artificial Intelligence in Glacier Studies: A State-of-the-Art Review" Water 16, no. 16: 2272. https://doi.org/10.3390/w16162272