Multi-hazard susceptibility mapping based on Convolutional Neural Networks

• Published in: Di xue qian yuan. Vol. 13; no. 5; p. 101425
• Main Authors: Ullah, Kashif, Wang, Yi, Fang, Zhice, Wang, Lizhe, Rahman, Mahfuzur
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier B.V 01.09.2022; Elsevier Science Ltd; Elsevier
• Subjects: Algorithms; Artificial intelligence; Artificial neural networks; Convolutional Neural Network; Debris flow; Deep learning; Disaster management; Eastern Hindukush; Flash floods; Flood mapping; Flood predictions; Floods; Hazard assessment; Hazard mitigation; Hydrologic surveys; Hydrology; Land management; Landslides; Landslides & mudslides; Local government; Machine learning; Multi-hazard; Neural networks; Pakistan; Risk management; Statistical analysis; CNN; DFSM; LR; Multi-hazard; DL; STI; KNN; SPI; NDVI; AUC; ALOS; MDG; OA; Pakistan; ML; Convolutional Neural Network; LSM; TWI; RMSE; DEM; NDMA; Eastern Hindukush; MAE; PCC; Machine learning; FFSM; MCDM
• ISSN: 1674-9871; 2588-9192
• DOI: 10.1016/j.gsf.2022.101425

[Display omitted] •First-hand multi-hazard susceptibility mapping framework based on Convolutional Neural Network.•Evaluation of flash floods, landslides, and debris flows.•62.57% of the study area is prone to multi-hazards, of which 0.18% are prone to three main hazards. Multi-hazard susceptibility prediction is an important component of disasters risk management plan. An effective multi-hazard risk mitigation strategy includes assessing individual hazards as well as their interactions. However, with the rapid development of artificial intelligence technology, multi-hazard susceptibility prediction techniques based on machine learning has encountered a huge bottleneck. In order to effectively solve this problem, this study proposes a multi-hazard susceptibility mapping framework using the classical deep learning algorithm of Convolutional Neural Networks (CNN). First, we use historical flash flood, debris flow and landslide locations based on Google Earth images, extensive field surveys, topography, hydrology, and environmental data sets to train and validate the proposed CNN method. Next, the proposed CNN method is assessed in comparison to conventional logistic regression and k-nearest neighbor methods using several objective criteria, i.e., coefficient of determination, overall accuracy, mean absolute error and the root mean square error. Experimental results show that the CNN method outperforms the conventional machine learning algorithms in predicting probability of flash floods, debris flows and landslides. Finally, the susceptibility maps of the three hazards based on CNN are combined to create a multi-hazard susceptibility map. It can be observed from the map that 62.43% of the study area are prone to hazards, while 37.57% of the study area are harmless. In hazard-prone areas, 16.14%, 4.94% and 30.66% of the study area are susceptible to flash floods, debris flows and landslides, respectively. In terms of concurrent hazards, 0.28%, 7.11% and 3.13% of the study area are susceptible to the joint occurrence of flash floods and debris flow, debris flow and landslides, and flash floods and landslides, respectively, whereas, 0.18% of the study area is subject to all the three hazards. The results of this study can benefit engineers, disaster managers and local government officials involved in sustainable land management and disaster risk mitigation.