Integrating Data Modality and Statistical Learning Methods for Earthquake-Induced Landslide Susceptibility Mapping

Zelang Miao,Minghui Pu,Changhao Hu,Ke Li,Renfeng Peng,Qinqin Liu,Qirong Li,Anshu Zhang,Shuai Chen,Wei Wang

doi:10.3390/app12031760

Abstract

Earthquakes induce landslides worldwide every year that may cause massive fatalities and financial losses. Precise and timely landslide susceptibility mapping (LSM) is significant for landslide hazard assessment and mitigation in earthquake-affected areas. State-of-the-art LSM approaches connect causative factors from various sources without considering the fusion of different information at the data modal level. To exploit the complementary information of different modalities and boost LSM accuracy, this study presents a new LSM model that integrates data modality and machine learning methods. The presented method first groups causative factors into different modal types based on their intrinsic characteristics, followed by the calculation of the pairwise similarity of modal data. The similarities of different modalities are fused using nonlinear graph fusion to generate a unified graph, which is subsequently classified using different machine learning methods to produce final LSM. Experimental results suggest that the presented method achieves higher performance than existing LSM methods. This study provides a new solution for producing precise LSM from a fusion perspective that can be applied to minimize the potential landslide risk and for sustainable use of erosion-prone slopes.

Highlights

Introduction iationsLandslide, one of the most destructive geological hazards in the world, often results in massive casualties and property losses [1,2,3,4]
Experimental results suggested that the data modality remarkably affects
Landslide susceptibility mapping (LSM) accuracy, and the presented LSM model effectively integrates the complementary information of different data modalities and achieves more satisfactory results than mainstream LSM models without consideration of data modalities

Summary

Introduction

Introduction iationsLandslide, one of the most destructive geological hazards in the world, often results in massive casualties and property losses [1,2,3,4]. The earthquake is a vital causative factor that triggers numerous landslides throughout the earthquake-affected area [5]. Landslide susceptibility mapping (LSM) refers to predicting potential landslides in an area depending on a range of causative factors [8,9,10,11,12]. It provides a beneficial reference to reflect the spatial distribution and the susceptibility level of landslide hazards, and has become a common tool in addressing landslide risk reduction [13]. A considerable amount of literature related to LSM has been published and can be grouped into three main categories: (1) physically-based methods, (2) knowledge-based methods, and (3) data-based methods [3,14,15,16].

Methods

Results

Conclusion