Assessment of debris flow susceptibility based on different slope unit division methods and BP neural network

doi:10.13474/j.cnki.11-2246.2022.0234

Abstract

Abstract: Selecting appropriate assessment units is the key to the assessment of debris flow susceptibility. In order to explore the impact of different methods of slope unit division on the assessment results of debris flow susceptibility. Taking Dongchuan district as an example, this paper compares and analyzes the effects of two slope unit division methods: hydrological analysis method and curvature watershed method in the evaluation of debris flow susceptibility. Based on the interpretation of debris flow points, the slope units with different division methods are used as the assessment unit, and the preliminary selected index factors are analyzed for multicollinearity and contribution rate to improve the index factor system, and finally build the debris flow susceptibility evaluation model based on BP neural network. The results show that the very high and high susceptibility areas of debris flow are mainly distributed in the Xiaojiang river valley and the south bank of Jinsha river in the study area, where the geological environment is fragile and the risk is high; The AUC value of the susceptibility model based on the curvature watershed method is 0.8658, which is higher than that of the hydrological analysis method of 0.8153, indicating that the slope unit divided by the curvature watershed method is more suitable for debris flow susceptibility assessment in the study area.

Key words: debris flow, susceptibility assessment, slope unit division, BP neural network, index factor

CLC Number:

P208

LI Kun, ZHAO Junsan, LIN Yilin, ZHOU Bao. Assessment of debris flow susceptibility based on different slope unit division methods and BP neural network[J]. Bulletin of Surveying and Mapping, 2022, 0(8): 68-74.

References

[1] 张晨,王清,陈剑平,等.金沙江流域泥石流的组合赋权法危险度评价[J].岩土力学,2011,32(3):831-836.
[2] 陈鹏宇,乔景顺,彭祖武,等.基于等级相关的泥石流危险因子筛选与危险度评价[J].岩土力学,2013,34(5):1409￣1415.
[3] 赵晓燕,谈树成,李永平.基于斜坡单元与组合赋权法的东川区地质灾害危险性评价[J].云南大学学报(自然科学版),2021,43(2):299-305.
[4] 周超,殷坤龙,曹颖,等.基于集成学习与径向基神经网络耦合模型的三峡库区滑坡易发性评价[J].地球科学,2020,45(6):1865-1876.
[5] YU Lanbing,CAO Ying,ZHOU Chao,et al.Landslide susceptibility mapping combining information gain ratio and support vector machines:a case study from Wushan segment in the Three Gorges reservoir area,China[J].Applied Sciences,2019,9(22):4756.
[6] 田乃满,兰恒星,伍宇明,等.人工神经网络和决策树模型在滑坡易发性分析中的性能对比[J].地球信息科学学报,2020,22(12):2304-2316.
[7] 叶振南,田运涛,陈宗良,等.西藏芒康县斜坡地质灾害空间分布特征与易发性区划[J].自然灾害学报,2021,30(3):199-208.
[8] ZHANG Yonghong,GE Taotao,TIAN Wei,et al.Debris flow susceptibility mapping using machine-learning techniques in Shigatse area,China[J].Remote Sensing,2019,11(23):2801.
[9] XIONG Ke,ADHIKARI B R,STAMATOPOULOS C A,et al.Comparison of different machine learning methods for debris flow susceptibility mapping:a case study in the Sichuan province,China[J].Remote Sensing,2020,12(2):295.
[10] 黄启乐,陈伟,傅旭东.斜坡单元支持下区域泥石流危险性AHP-RBF评价模型[J].浙江大学学报(工学版),2018,52(9):1667-1675.
[11] 张曦,陈丽霞,徐勇,等.两种斜坡单元划分方法对滑坡灾害易发性评价的对比研究[J].安全与环境工程,2018,25(1):12-17.
[12] CHEVALIER G G,MEDINA V,HÜRLIMANN M,et al.Debris-flow susceptibility analysis using fluvio-morphological parameters and data mining:application to the Central-Eastern Pyrenees[J].Natural Hazards,2013,67(2):213-238.
[13] 武雪玲,任福,牛瑞卿,等.斜坡单元支持下的滑坡易发性评价支持向量机模型[J].武汉大学学报(信息科学版),2013,38(12):1499-1503.
[14] 李紫微,马庆勋,吕杰.BP神经网络的近地面臭氧估算及时空特征分析[J].测绘通报,2021(6):28-32.
[15] 龚旭,吕佳,皮家甜.结合信息增益率和K-means聚类的协同训练算法[J].重庆师范大学学报(自然科学版),2020,37(2):112-119.
[16] LIANG Zhu,WANG Changming,ZHANG Zhimin,et al.A comparison of statistical and machine learning methods for debris flow susceptibility mapping[J].Stochastic Environmental Research and Risk Assessment,2020,34(11):1887-1907.
[17] 李益敏,杨蕾,魏苏杭.基于小流域单元的怒江州泥石流易发性评价[J].长江流域资源与环境,2019,28(10):2419-2428.