1.School of Engineering and Technology,State Key Laboratory of Deep Earth Exploration and Imaging,China University of Geosciences (Beijing),Beijing 100083,China
2.Technical Innovation Center of Mine Geological Environmental Restoration Engineering in Southern Karst Area,Ministry of Natural Resources,Nanning,Guangxi 530028,China
Current geological hazard susceptibility assessments often neglect differences in the formation mechanisms of different hazard types, limiting model accuracy. This paper proposes a dynamic weight adjustment strategy based on a “classification evaluation-zonal weighting” comparison to improve the accuracy of comprehensive susceptibility assessments for rockfalls and landslides. Taking the Ningqiang-Yangxian area in Shaanxi Province as an example, and based on identifying the distribution characteristics and main controlling factors of 64 rockfalls and 610 landslides in the region, separate susceptibility models for rockfalls and landslides were first constructed. The proposed Dynamic Weight Adjustment Method (DWAM) was then applied to develop a comprehensive model, which was compared with three other combination strategies, namely the direct combination method, the maximum value method, and the probabilistic statistical method. Through zonal statistics, ROC curves, and susceptibility index distribution patterns, the four methods were compared and optimized to achieve comprehensive susceptibility zoning. The results show: (1) Rockfall distribution is mainly controlled by Normalized Difference Built-Up Index (NDBI), road construction, and faults, while landslides are primarily influenced by land use type, elevation, and engineering geological rock group. (2) Classification-based models significantly outperform mixed-type modeling, with DWAM yielding the best zoning results and predictive performance. (3) The extremely high and high susceptibility zones identified by DWAM account for 8.97% and 16.05% of the total area, respectively, primarily distributed in the northwestern part, showing a zonal pattern clearly controlled by roads and faults. The study demonstrates that DWAM effectively improves the accuracy of multi-hazard susceptibility assessments, providing scientific support for regional hazard prevention and mitigation planning.
YUL B, PRADHANB, WANGY. A comparative study of various combination strategies for landslide susceptibility mapping considering landslide types[J]. Geoscience Frontiers, 2025, 16(2): 101999.
HUANGF M, XIONGH W, YAOC, et al. Uncertainties of landslide susceptibility prediction considering different landslide types[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2023, 15(11): 2954 - 2972.
[10]
ADO M, AMITABK, MAJIA K, et al. Landslide Susceptibility Mapping Using Machine Learning: A Literature Survey[J]. Remote Sensing, 2022, 14: 3029.
PRADHANB. Landslide susceptibility mapping of a catchment area using frequency ratio, fuzzy logic and multivariate logistic regression approaches[J]. Journal of the Indian Society of Remote Sensing, 2010, 38(2): 301 - 320.
[16]
LOMBARDOL, TANYASH, HUSERR, et al. Landslide size matters: A new data-driven, spatial prototype[J]. Engineering Geology, 2021, 293: 106288.
[17]
LIW B, FANX M, HUANGF M, et al. Uncertainties Analysis of Collapse Susceptibility Prediction Based on Remote Sensing and GIS: Influences of Different Data-Based Models and Connections between Collapses and Environmental Factors[J]. Remote Sensing, 2020, 12(24): 4134.
MOUSAVIS Z, KAVIANA, SOLEIMANIK, et al. GIS-based spatial prediction of landslide susceptibility using logistic regression model[J]. Geomatics, Natural Hazards and Risk, 2011, 2(1): 33 - 50.
[21]
CHENT Q, GUESTRINC. XGBoost: A Scalable Tree Boosting System[C]// Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. 2016: 785 - 794.
NIRAJK C, SINGHA, SHUKLAD P. Effect of the Normalized Difference Vegetation Index (NDVI) on GIS-Enabled Bivariate and Multivariate Statistical Models for Landslide Susceptibility Mapping[J]. Journal of the Indian Society of Remote Sensing, 2023, 51: 1739 - 1756
[32]
THOMASA V, SAHAS, DANUMAHJ H, et al. Landslide Susceptibility Zonation of Idukki District Using GIS in the Aftermath of 2018 Kerala Floods and Landslides: a Comparison of AHP and Frequency Ratio Methods[J]. Journal of Geovisualization and Spatial Analysis, 2021, 5: 21.
MAS, CHENJ, WUS E, et al. Landslide Susceptibility Prediction Using Machine Learning Methods: A Case Study of Landslides in the Yinghu Lake Basin in Shaanxi[J]. Sustainability, 2023, 15: 15836.
LINQ G, LIMAP, STEGERS, et al. National-scale data-driven rainfall induced landslide susceptibility mapping for China by accounting for incomplete landslide data[J]. Geoscience Frontiers, 2021, 12, 101248.
[38]
HUANGF M, MAOD X, JIANGS H, et al. Uncertainties in landslide susceptibility prediction modeling: A review on the incompleteness of landslide inventory and its influence rules[J]. Geoscience Frontiers, 2024, 15(6): 101886.
[39]
HUANGF M, CAOZ S, GUOJ F, et al. Comparisons of heuristic, general statistical and machine learning models for landslide susceptibility prediction and mapping[J]. CATENA, 2020, 191:104580,