Seismic Landslide Susceptibility Assessment Using Newmark Displacement Based on a Dual-Channel Convolutional Neural Network
Landslide susceptibility assessment (LSA) is an essential tool for landslide hazard warning. The selection of earthquake-related factors is pivotal for seismic LSA. In this study, Newmark displacement (<i>D<sub>n</sub></i>) is employed as the earthquake-related factor, provid...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/16/3/566 |
| _version_ | 1797318288655515648 |
|---|---|
| author | Yan Li Dongping Ming Liang Zhang Yunyun Niu Yangyang Chen |
| author_facet | Yan Li Dongping Ming Liang Zhang Yunyun Niu Yangyang Chen |
| author_sort | Yan Li |
| collection | DOAJ |
| description | Landslide susceptibility assessment (LSA) is an essential tool for landslide hazard warning. The selection of earthquake-related factors is pivotal for seismic LSA. In this study, Newmark displacement (<i>D<sub>n</sub></i>) is employed as the earthquake-related factor, providing a detailed representation of seismic characteristics. On the algorithmic side, a dual-channel convolutional neural network (CNN) model is built, and the last classification layer is replaced with two machine learning (ML) models to facilitate the extraction of deeper features related to landslide development. This research focuses on Beichuan County in Sichuan Province, China. Fifteen landslide predisposing factors, including hydrological, geomorphic, geological, vegetation cover, anthropogenic, and earthquake-related features, were extensively collected. The results demonstrate some specific issues. <i>D<sub>n</sub></i> outperforms conventional earthquake-related factors such as peak ground acceleration (<i>PGA</i>) and Arias intensity (<i>I<sub>a</sub></i>) in capturing seismic influence on landslide development. Under the same conditions, the <i>OA</i> improved by 5.55% and <i>AUC</i> improved by 0.055 compared to the <i>PGA</i>; the <i>OA</i> improved by 3.2% and <i>AUC</i> improved by 0.0327 compared to the Ia. The improved CNN outperforms ML models. Under the same conditions, the <i>OA</i> improved by 4.69% and <i>AUC</i> improved by 0.0467 compared to RF; the <i>OA</i> improved by 4.47% and <i>AUC</i> improved by 0.0447 compared to SVM. Additionally, historical landslides validate the reasonableness of the landslide susceptibility maps. The proposed method exhibits a high rate of overlap with the historical landslide inventory. The proportion of historical landslides in the very high and high susceptibility zones exceeds 87%. The method not only enhances accuracy but also produces a more fine-grained susceptibility map, providing a reliable basis for early warning of seismic landslides. |
| first_indexed | 2024-03-08T03:50:15Z |
| format | Article |
| id | doaj.art-e59b27e90d734b92984948517361e110 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-08T03:50:15Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-e59b27e90d734b92984948517361e1102024-02-09T15:21:30ZengMDPI AGRemote Sensing2072-42922024-02-0116356610.3390/rs16030566Seismic Landslide Susceptibility Assessment Using Newmark Displacement Based on a Dual-Channel Convolutional Neural NetworkYan Li0Dongping Ming1Liang Zhang2Yunyun Niu3Yangyang Chen4School of Information Engineering, China University of Geosciences Beijing, Beijing 100083, ChinaSchool of Information Engineering, China University of Geosciences Beijing, Beijing 100083, ChinaSchool of Information Engineering, China University of Geosciences Beijing, Beijing 100083, ChinaSchool of Information Engineering, China University of Geosciences Beijing, Beijing 100083, ChinaChina Aero Geophysical Survey and Remote Sensing Center for Natural Resources, Beijing 100083, ChinaLandslide susceptibility assessment (LSA) is an essential tool for landslide hazard warning. The selection of earthquake-related factors is pivotal for seismic LSA. In this study, Newmark displacement (<i>D<sub>n</sub></i>) is employed as the earthquake-related factor, providing a detailed representation of seismic characteristics. On the algorithmic side, a dual-channel convolutional neural network (CNN) model is built, and the last classification layer is replaced with two machine learning (ML) models to facilitate the extraction of deeper features related to landslide development. This research focuses on Beichuan County in Sichuan Province, China. Fifteen landslide predisposing factors, including hydrological, geomorphic, geological, vegetation cover, anthropogenic, and earthquake-related features, were extensively collected. The results demonstrate some specific issues. <i>D<sub>n</sub></i> outperforms conventional earthquake-related factors such as peak ground acceleration (<i>PGA</i>) and Arias intensity (<i>I<sub>a</sub></i>) in capturing seismic influence on landslide development. Under the same conditions, the <i>OA</i> improved by 5.55% and <i>AUC</i> improved by 0.055 compared to the <i>PGA</i>; the <i>OA</i> improved by 3.2% and <i>AUC</i> improved by 0.0327 compared to the Ia. The improved CNN outperforms ML models. Under the same conditions, the <i>OA</i> improved by 4.69% and <i>AUC</i> improved by 0.0467 compared to RF; the <i>OA</i> improved by 4.47% and <i>AUC</i> improved by 0.0447 compared to SVM. Additionally, historical landslides validate the reasonableness of the landslide susceptibility maps. The proposed method exhibits a high rate of overlap with the historical landslide inventory. The proportion of historical landslides in the very high and high susceptibility zones exceeds 87%. The method not only enhances accuracy but also produces a more fine-grained susceptibility map, providing a reliable basis for early warning of seismic landslides.https://www.mdpi.com/2072-4292/16/3/566landslidelandslide susceptibility assessmentconvolutional neural network (CNN)Newmarkearthquake |
| spellingShingle | Yan Li Dongping Ming Liang Zhang Yunyun Niu Yangyang Chen Seismic Landslide Susceptibility Assessment Using Newmark Displacement Based on a Dual-Channel Convolutional Neural Network Remote Sensing landslide landslide susceptibility assessment convolutional neural network (CNN) Newmark earthquake |
| title | Seismic Landslide Susceptibility Assessment Using Newmark Displacement Based on a Dual-Channel Convolutional Neural Network |
| title_full | Seismic Landslide Susceptibility Assessment Using Newmark Displacement Based on a Dual-Channel Convolutional Neural Network |
| title_fullStr | Seismic Landslide Susceptibility Assessment Using Newmark Displacement Based on a Dual-Channel Convolutional Neural Network |
| title_full_unstemmed | Seismic Landslide Susceptibility Assessment Using Newmark Displacement Based on a Dual-Channel Convolutional Neural Network |
| title_short | Seismic Landslide Susceptibility Assessment Using Newmark Displacement Based on a Dual-Channel Convolutional Neural Network |
| title_sort | seismic landslide susceptibility assessment using newmark displacement based on a dual channel convolutional neural network |
| topic | landslide landslide susceptibility assessment convolutional neural network (CNN) Newmark earthquake |
| url | https://www.mdpi.com/2072-4292/16/3/566 |
| work_keys_str_mv | AT yanli seismiclandslidesusceptibilityassessmentusingnewmarkdisplacementbasedonadualchannelconvolutionalneuralnetwork AT dongpingming seismiclandslidesusceptibilityassessmentusingnewmarkdisplacementbasedonadualchannelconvolutionalneuralnetwork AT liangzhang seismiclandslidesusceptibilityassessmentusingnewmarkdisplacementbasedonadualchannelconvolutionalneuralnetwork AT yunyunniu seismiclandslidesusceptibilityassessmentusingnewmarkdisplacementbasedonadualchannelconvolutionalneuralnetwork AT yangyangchen seismiclandslidesusceptibilityassessmentusingnewmarkdisplacementbasedonadualchannelconvolutionalneuralnetwork |