Mapping the susceptibility to landslides based on the deep belief network: a case study in Sichuan Province, China
- PDF / 3,515,233 Bytes
- 23 Pages / 439.37 x 666.142 pts Page_size
- 105 Downloads / 208 Views
Mapping the susceptibility to landslides based on the deep belief network: a case study in Sichuan Province, China Weidong Wang1,2 · Zhuolei He1 · Zheng Han1 · Yange Li1 · Jie Dou3 · Jianling Huang1 Received: 6 May 2019 / Accepted: 19 June 2020 © Springer Nature B.V. 2020
Abstract A dataset of landslides from Sichuan Province in China, containing 1551 historical individual landslides, is a result of two teams’ effort in the past few years to map the susceptibility to landslides. Considering complex internal relations among the triggering factors, logistic regression (LR) and shallow neural networks, such as back-propagation neural network (BPNN), are often limited. In this paper, we make a straightforward development that the deep belief network (DBN) based on deep learning technology is introduced to map the regional susceptibility to landslides. Seven factors with respect to geomorphology, geology and hydrology are considered and verified through the collinearity test. A DBN model containing three pre-trained layers of restricted Boltzmann machines by stochastic gradient descent method is configured to obtain the susceptibility to landslides. Susceptibility results evaluated by DBN model are compared with those by LR and BPNN in the receive operator characteristic (ROC) analysis, showing that DBN has a better prediction precision, with a lower rate of false alarms and fake alarms. The case study also indicates different sensitivities of the triggering factors to the landslide susceptibility, that the factors of altitude, distance to drainage network and average annual rainfall have significant impact in mapping the susceptibility to landslides in the region. This research will contribute to a better-performance model for regional-scale mapping for the susceptibility to landslides, in particular, at the area where triggering factors show complex relations and relative independence. Keywords Landslides mapping · Susceptibility · Deep learning · Deep belief network · Sichuan area
* Zheng Han [email protected] 1
School of Civil Engineering, Central South University, 22 Shaoshan Road, Changsha 410075, Hunan, China
2
The Key Laboratory of Engineering Structures of Heavy Haul Railway, Ministry of Education, Changsha 410075, China
3
Department of Civil and Environmental Engineering, Nagaoka University of Technology, Niigata 940‑2188, Japan
13
Vol.:(0123456789)
Natural Hazards
1 Introduction Landslides are the most widespread geological hazard worldwide (Alexander 2004). In China for instance, landslides in the past years accounted for almost 76% of the annual geological disasters. This kind of landscape forming process often transports large volumes of deposits, disrupting traffic, blocking rivers, burying villages, and therefore posing serious risks to humans and local society (Wang et al. 2012; Han et al. 2015a). Given the severe risk from landslides, the issue of landslide prevention and mitigation has received considerable attention. A key aspect regarding to mapping the susceptibility to landsli
