Article Details

Abstract

Taking the landslide triggered by the 2022 Maerkang earthquake in Sichuan as an example, 12 types of landslide impact factors are selected according to the characteristics of earthquake, topography and geological environment in the earthquake area. The BP Neural Network model is used to analyze the risk of earthquake landslides, and then the road trafficability prediction model in the earthquake affected area is constructed, and the performance of the model is evaluated. The results show that the AUC of the earthquake landslide risk evaluation model based on BP Neural Network algorithm is 0.92, and the prediction accuracy of the model is high. The number and density of landslides distributed in each risk zone are in line with the law of disaster development, and the evaluation results are reliable. The evaluation results of the road trafficability prediction model show that the extremely low trafficability sections are mainly the sections of the Provincial Highway S220 Jiaomuzu Township to Longtoutan and Longtoutan to Caodeng Township, as well as the sections of Ribu Township and Rong'an Township to Caodeng Township, which have a high degree of coincidence with the actual severely affected sections.

Keywords

Maerkang earthquake; coseismic landslide; Machine Learning; road trafficability

References

1. [1] Qi Fengjiao, Li Wen, Su Hejun, etc A study on the assessment model of death in earthquake disaster zoning in Chinese Mainland [J]. Earthquake, 2022, 42 (1): 70-84
2. [2] Luo Hua Research on Geological Hazard Risk Assessment and Ecological Restoration Strategies for Earthquake Landslides [J]. Environmental Science and Management, 2024, 49 (8): 180-184
3. [3] Bai Huilin, Jiang Guowei, Liu Fenghui, etc Mechanism of Highway Slope Damage and Post earthquake Secondary Disaster Effects in the Markang Earthquake [J]. Journal of Earthquake Engineering, 2025, 47 (3): 554-565
4. [4] Bai Xianfu, Yang Zhiquan, Luo Weidong, etc Emergency assessment model for earthquake landslide road interruption risk based on GIS and multivariate decision tree [J]. Earthquake Research, 2023, 46 (3): 343-353
5. [5] Jia Zhaoliang, Zheng Chuan, Wu Yanmei, etc Risk Assessment of Earthquake Landslide Bearing Bodies Based on Newmark Model: A Case Study of the 6.5-magnitude Ludian Earthquake in Yunnan Province in 2014 [J]. Earthquake Research, 2023, 46 (3): 366-375
6. [6] Xia Yangde Long, Luo Weiqi, Gao Junhua, etc Optimal coupling model for landslide susceptibility assessment based on logistic regression [J]. Geospatial Information, 2025, 23 (6): 61-64, 78
7. [7] Liu Jincang, Wang Huanhuan, Li Yun, etc Evaluation of the susceptibility of slope geological hazards in South China based on support vector machine: a case study of Huaiji County, Zhaoqing City [J]. Journal of Spatiotemporal Information, 2024, 31 (6): 785-794
8. [8] Du Peng, Chen Ningsheng, Wu Kanglin, etc Landslide susceptibility evaluation and main control factor analysis in southeastern Tibet based on random forest model [J]. Journal of Chengdu University of Technology: Natural Science Edition, 2024, 51 (2): 328-344
9. [9] Zhou Tianyou, Liu Chang, Xue Peng, etc Research on modeling the susceptibility of post earthquake landslides based on different machine learning methods [J]. Journal of Natural Disasters, 2023, 32 (5): 177-185
10. [10] SHAO X Y，XU C，WANG P，et al. Two Public Inventories of Landslides Induced by the 10 June 2022 Maerkang Earthquake Swarm, China and Ancient Landslides in the Affected Area [J].Natural Hazards Research，2022，2（4）：269-272.
11. [11] CHEN J，BAN Y F，LI S N. China: Open Access to Earth Land-cover Map [J/OL].Nature，2014，514（7523）：434[2025-06-07].https://www.nature.com/articles/514434c.
12. [12] Li Chenyang, Wang Xinchun, He Chunzhen, etc National 1:200000 Digital Geological Map (Public Version) Spatial Database [J]. China Geology, 2019, 46 (S1): 1-10
13. [13] YANG Z Q，XU C，SHAO X Y，et al. Landslide Susceptibility Mapping Based on CNN-3D Algorithm with Attention Module Embedded [J/OL].Bulletin of Engineering Geology and the Environment，2022，81（10）：412[2025-06-11].https://doi.org/10.1007/s10064-022-02889-4.
14. [14] Xu Weiping, Yang Hui, Zhang Jiming, etc Assessment of landslide geological hazard susceptibility based on GIS and BP neural network algorithm [J]. Land and Resources Guide, 2025, 22 (1): 98-107

Funding

2025 Zhejiang Province University Student Science and Technology Innovation Activity Plan (New Talent Plan) (2025R481A014)

About the authors

Qiu Kefeng (2005-), male, Han ethnicity, from Wenzhou, Zhejiang, research direction: risk assessment of earthquake landslide disaster chain; Corresponding author: Chen Zhaoning (1997-), male, Han ethnicity, from Jinhua, Zhejiang, Ph.D., lecturer, research direction: mechanism and evaluation of earthquake geological hazards.