Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran
The identification of landslide-prone areas is an essential step in landslide hazard assessment and mitigation of landslide-related losses. In this study, we applied two novel deep learning algorithms, the recurrent neural network (RNN) and convolutional neural network (CNN), for national-scale land...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-03-01
|
| Series: | Geoscience Frontiers |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1674987120301687 |
| _version_ | 1797712881293197312 |
|---|---|
| author | Phuong Thao Thi Ngo Mahdi Panahi Khabat Khosravi Omid Ghorbanzadeh Narges Kariminejad Artemi Cerda Saro Lee |
| author_facet | Phuong Thao Thi Ngo Mahdi Panahi Khabat Khosravi Omid Ghorbanzadeh Narges Kariminejad Artemi Cerda Saro Lee |
| author_sort | Phuong Thao Thi Ngo |
| collection | DOAJ |
| description | The identification of landslide-prone areas is an essential step in landslide hazard assessment and mitigation of landslide-related losses. In this study, we applied two novel deep learning algorithms, the recurrent neural network (RNN) and convolutional neural network (CNN), for national-scale landslide susceptibility mapping of Iran. We prepared a dataset comprising 4069 historical landslide locations and 11 conditioning factors (altitude, slope degree, profile curvature, distance to river, aspect, plan curvature, distance to road, distance to fault, rainfall, geology and land-sue) to construct a geospatial database and divided the data into the training and the testing dataset. We then developed RNN and CNN algorithms to generate landslide susceptibility maps of Iran using the training dataset. We calculated the receiver operating characteristic (ROC) curve and used the area under the curve (AUC) for the quantitative evaluation of the landslide susceptibility maps using the testing dataset. Better performance in both the training and testing phases was provided by the RNN algorithm (AUC = 0.88) than by the CNN algorithm (AUC = 0.85). Finally, we calculated areas of susceptibility for each province and found that 6% and 14% of the land area of Iran is very highly and highly susceptible to future landslide events, respectively, with the highest susceptibility in Chaharmahal and Bakhtiari Province (33.8%). About 31% of cities of Iran are located in areas with high and very high landslide susceptibility. The results of the present study will be useful for the development of landslide hazard mitigation strategies. |
| first_indexed | 2024-03-12T07:28:19Z |
| format | Article |
| id | doaj.art-82e1bd5fb67b48f2b7287fb174c8c94a |
| institution | Directory Open Access Journal |
| issn | 1674-9871 |
| language | English |
| last_indexed | 2024-03-12T07:28:19Z |
| publishDate | 2021-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Geoscience Frontiers |
| spelling | doaj.art-82e1bd5fb67b48f2b7287fb174c8c94a2023-09-02T21:59:15ZengElsevierGeoscience Frontiers1674-98712021-03-01122505519Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of IranPhuong Thao Thi Ngo0Mahdi Panahi1Khabat Khosravi2Omid Ghorbanzadeh3Narges Kariminejad4Artemi Cerda5Saro Lee6Institute of Research and Development, Duy Tan University, Da Nang, 550000, Viet NamDivision of Science Education, Kangwon National University, Chuncheon-si, Gangwon-do, 24341, Republic of Korea; Geoscience Platform Division, Korea Institute of Geoscience and Mineral Resources (KIGAM), 124, Gwahak-ro Yuseong-gu, Daejeon, 34132, Republic of KoreaDepartment of Watershed Management Engineering, Sari Agricultural Science and Natural Resources University, Sari, IranDepartment of Geoinformatics–Z_GIS, University of Salzburg, Salzburg, 5020, AustriaDepartment of Watershed and Arid Zone Management, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, 49189-434, IranSoil Erosion and Desertification Research Group, Department of Geography, University of Valencia, Valencia, SpainGeoscience Platform Division, Korea Institute of Geoscience and Mineral Resources (KIGAM), 124, Gwahak-ro Yuseong-gu, Daejeon, 34132, Republic of Korea; Department of Geophysical Exploration, Korea University of Science and Technology, 217 Gajeong-ro Yuseong-gu, Daejeon, 34113, Republic of Korea; Corresponding author. Geoscience Platform Division, Korea Institute of Geoscience and Mineral Resources (KIGAM), 124, Gwahak-ro Yuseong-gu, Daejeon, 34132, Republic of KoreaThe identification of landslide-prone areas is an essential step in landslide hazard assessment and mitigation of landslide-related losses. In this study, we applied two novel deep learning algorithms, the recurrent neural network (RNN) and convolutional neural network (CNN), for national-scale landslide susceptibility mapping of Iran. We prepared a dataset comprising 4069 historical landslide locations and 11 conditioning factors (altitude, slope degree, profile curvature, distance to river, aspect, plan curvature, distance to road, distance to fault, rainfall, geology and land-sue) to construct a geospatial database and divided the data into the training and the testing dataset. We then developed RNN and CNN algorithms to generate landslide susceptibility maps of Iran using the training dataset. We calculated the receiver operating characteristic (ROC) curve and used the area under the curve (AUC) for the quantitative evaluation of the landslide susceptibility maps using the testing dataset. Better performance in both the training and testing phases was provided by the RNN algorithm (AUC = 0.88) than by the CNN algorithm (AUC = 0.85). Finally, we calculated areas of susceptibility for each province and found that 6% and 14% of the land area of Iran is very highly and highly susceptible to future landslide events, respectively, with the highest susceptibility in Chaharmahal and Bakhtiari Province (33.8%). About 31% of cities of Iran are located in areas with high and very high landslide susceptibility. The results of the present study will be useful for the development of landslide hazard mitigation strategies.http://www.sciencedirect.com/science/article/pii/S1674987120301687CNNRNNDeep learningLandslideIran |
| spellingShingle | Phuong Thao Thi Ngo Mahdi Panahi Khabat Khosravi Omid Ghorbanzadeh Narges Kariminejad Artemi Cerda Saro Lee Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran Geoscience Frontiers CNN RNN Deep learning Landslide Iran |
| title | Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran |
| title_full | Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran |
| title_fullStr | Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran |
| title_full_unstemmed | Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran |
| title_short | Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran |
| title_sort | evaluation of deep learning algorithms for national scale landslide susceptibility mapping of iran |
| topic | CNN RNN Deep learning Landslide Iran |
| url | http://www.sciencedirect.com/science/article/pii/S1674987120301687 |
| work_keys_str_mv | AT phuongthaothingo evaluationofdeeplearningalgorithmsfornationalscalelandslidesusceptibilitymappingofiran AT mahdipanahi evaluationofdeeplearningalgorithmsfornationalscalelandslidesusceptibilitymappingofiran AT khabatkhosravi evaluationofdeeplearningalgorithmsfornationalscalelandslidesusceptibilitymappingofiran AT omidghorbanzadeh evaluationofdeeplearningalgorithmsfornationalscalelandslidesusceptibilitymappingofiran AT nargeskariminejad evaluationofdeeplearningalgorithmsfornationalscalelandslidesusceptibilitymappingofiran AT artemicerda evaluationofdeeplearningalgorithmsfornationalscalelandslidesusceptibilitymappingofiran AT sarolee evaluationofdeeplearningalgorithmsfornationalscalelandslidesusceptibilitymappingofiran |