Objective This study aims to explore the influence of different screening methods of evaluation indicators and machine learning models on the evaluation results of Benggang susceptibility. Methods Taking Benggang in Shicheng County as a case study, three approaches-non-screening, geodetector (GD) screening, and variance inflation factor (VIF) screening-were employed to construct the evaluation indicator system. The random forest (RF) and multilayer perceptron (MLP) models were applied to conduct a suitability study comparing different indicator sets and models for Benggang susceptibility evaluation. Results 1) The susceptibility evaluation results based on GD-screened indicators were superior to those obtained using VIF-screened indicators. The results of GD screening with a 90% cumulative q-value were basically the same as those of non-screening. 2) The RF model was better than MLP model in evaluating the vulnerability to Benggang. Under the three indicator systems (non-screening, GD screening, and VIF screening), the AUC values for the RF model were 0.847, 0.823, and 0.779, respectively. 3) The high-susceptibility zones and very-high-susceptibility zones in the study area were mainly distributed in the northwest of Shicheng County, which was consistent with the distribution of Benggang. Conclusion Employing the RF model with indicators screened by GD at a cumulative q-value contribution rate of 90% provides a highly rational and applicable approach for susceptibility evaluation. The research findings can provide a scientific reference for the evaluation of Benggang susceptibility in the Gannan region.
LIAOY S, TANGC Y, YUANZ J, et al. Research progress on Benggang erosion and its prevention measure in red soil region of southern China[J].Acta Pedologica Sinica,2018,55(6):1297-1312.
HANJ X, DENGZ H, WANGJ M, et al. Development of soil cracks in granite Benggang area under wetting-drying cycles[J].Journal of Soil and Water Conservation,2024,38(5):262-271.
[5]
HUANGM Y, SUNS J, FENGK J, et al. Effects of Neyraudia reynaudiana roots on the soil shear strength of collapsing wall in Benggang,southeast China[J].Catena,2022,210:e105883.
WENH, NIS M, WANGY T, et al. A study on silty soil shear strength and its influencing factors in different vegetation types in Benggang erosion area of southern Jiangxi[J].Acta Pedologica Sinica,2022,59(6):1517-1526.
[8]
LIUX L, QIUJ N, ZHANGD L. Characteristics of slope runoff and soil water content in Benggang colluvium under simulated rainfall[J].Journal of Soils and Sediments,2018,18(1):39-48.
[9]
LIUW P, OUYANGG Q, LUOX Y, et al. Moisture content,pore-water pressure and wetting front in granite residual soil during collapsing erosion with varying slope angle[J].Geomorphology,2020,362:e107210.
LUOL G, PEIX J, CUIS H, et al. Combined selection of susceptibility assessment factors for Jiuzhaigou earthquake-induced landslides[J].Chinese Journal of Rock Mechanics and Engineering,2021,40(11):2306-2319.
DENGY S, CAIC F. Progress of survey,monitoring,and control technology of Benggang erosion in red soil hilly area[J].Acta Pedologica Sinica,2025,62(2):322-333.
[14]
LIANGW C, HEL, LINZ, et al. Microtopography promoting Benggang erosion:Formation and stability of niche on collapsing wall[J].Geomorphology,2025,486:e109908.
GUOF, WUD, WANGX J, et al. Susceptibility assessment of Benggang based on random forests model and geodetector in Xingguo County of south Jiangxi[J].Journal of China Three Gorges University (Natural Sciences),2023,45(6):44-50.
[17]
王劲峰,徐成东.地理探测器:原理与展望[J].地理学报,2017,72(1):116-134.
[18]
WANGJ F, XUC D. Geodetector: Principle and prospective[J].Acta Geographica Sinica,2017,72(1):116-134.
[19]
LOMBARDOL, MAIP M. Presenting logistic regression-based landslide susceptibility results[J].Engineering Geology,2018,244:14-24.
XIAL, XIAW X, GUOF, et al. Effects of different spatial resolutions of grid cells on Benggang susceptibility using subjective and objective weighting methods[J].Research of Soil and Water Conservation,2025,32(6):208-215,224.
FENGC J, DENGY S, HEY J, et al. Evaluation of collapse erosion intensity based on principal component and cluster analysis[J].Research of Soil and Water Conservation,2019,26(1):41-46.
LIUN, ZHANGH, DENGC L, et al. Distribution characteristics of Benggang in the middle and upper reaches of the Han River[J].Tropical Geography,2024,44(3):415-428.
LIAOK T, LIUY, LIUQ, et al. Distribution characteristics and driving factors of Benggang erosion in Ganzhou City[J].Research of Soil and Water Conservation,2021,28(6):126-130.
LINX H, HUANGY H, LINJ S, et al. Risk assessment and spatial-temporal characteristics of Benggang erosion based on DPSIR model[J].Transactions of the Chinese Society of Agricultural Engineering,2023,39(18):123-131.
[30]
KAVZOGLUT, KUTLUG SAHINE, COLKESENI. Selecting optimal conditioning factors in shallow translational landslide susceptibility mapping using genetic algorithm[J].Engineering Geology,2015,192:101-112.
CHENGD B, ZHAOY L, ZHANGP C, et al. Risk assessment of collapse gully erosion in Jiangxi Province based on bivariate statistical analysis of entropy information[J].Journal of Changjiang River Scientific Research Institute,2019,36(2):27-32.
GUANJ L, HUANGY H, LINJ S, et al. Comparisons between Benggang risk assessments based on information model and random forest model[J].Mountain Research,2021,39(4):539-551.
LIC M, XUG L, LUY. Key influencing factors and susceptibility of collapse gully in southeast Guangxi, China[J].Journal of Yangtze River Scientific Research Institute,2020,37(3):131-136.
[37]
GAYENA, POURGHASEMIH R, SAHAS, et al. Gully erosion susceptibility assessment and management of hazard-prone areas in India using different machine learning algorithms[J].Science of the Total Environment,2019,668:124-138.
[38]
WEIY J, WUX L, WANGJ G, et al. Identification of geo-environmental factors on Benggang susceptibility and its spatial modelling using comparative data-driven methods[J].Soil and Tillage Research,2021,208:e104857.
GUOF, WUD, GEM R, et al.Influence of continuous variable factor classification and machine learning model on the accuracy of landslide susceptibility evaluation[J].Geomatics and Information Science of Wuhan University,2026,51(2):236-248.
GUOF, JIANGG H, HUANGX H, et al. Impact of environmental factor combinations and negative sample selection on Benggang susceptibility assessment in granite areas[J].Transactions of the Chinese Society of Agricultural Engineering,2024,40(1):191-200.
SUNK, CHENGD B, HEJ J, et al. Comparative study on risk assessment methods for collapse gully erosion∶ A case of Guangdong Province[J].Soil and Water Conservation in China,2018(3):51-54.
[45]
LIUZ, WEIY J, CUIT T, et al. Spatial scaling effects of gully erosion in response to driving factors in southern China[J].Journal of Geographical Sciences,2024,34(5):942-962.
GUOF, LAIP, HUANGF M, et al. Literature review and research progress of landslide susceptibility mapping based on knowledge graph[J].Earth Science,2024,49(5):1584-1606.
[48]
POURGHASEMIH R, YOUSEFIS, KORNEJADYA, et al. Performance assessment of individual and ensemble data-mining techniques for gully erosion modeling[J].Science of the Total Environment,2017,609:764-775.
LIH, WENA B, LIUT, et al. Spatial distribution of 137Cs reference inventory in Sichuan Province using geographically weighted regression Kriging combined with 137Cs reference inventory mathematical model[J].Science of Soil and Water Conservation,2018,16(5):57-66.
YANGS, LID Y, YANL X, et al. Landslide susceptibility assessment in high and steep bank slopes along Wujiang River based on random forest model[J].Safety and Environmental Engineering,2021,28(4):131-138.
[53]
LIUZ Q, GILBERTG, CEPEDAJ M, et al. Modelling of shallow landslides with machine learning algorithms[J].Geoscience Frontiers,2021,12(1):385-393.
CHENZ C, GUOC X, ZHANGZ R. Analysis of soil-water characteristics and their correlation with collapsing gully in the high-occurrence area of collapsing gully in Anxi,Fujian[J].Science of Soil and Water Conservation,2025,23(5):216-224.