Landslide susceptibility assessment and zonation based on landslide classification and improved Convolutional Neural Networks
-
摘要: 滑坡敏感性区划是根据研究区域的滑坡调查数据和地质环境条件，分析致灾因子的组合特征对滑坡发生的影响并将研究区域划分为不同等级的敏感区，为土地利用规划和滑坡防治政策制定提供理论依据。对山东省淄博市博山区开展了工程岩组和地质灾害调查，结合多种数据源和地质调查资料建立了博山区滑坡数据库；采用Pearson相关系数法筛选了滑坡致灾因子，基于信息量法分别对全部滑坡、自然滑坡和工程滑坡的致灾因子进行分级；建立了包含2轮卷积和池化的7层改进卷积神经网络，分别对全部滑坡、自然滑坡和工程滑坡进行了评价模型的训练和验证，采用AUC法验证了模型的准确性；基于ArcGIS12.0计算了博山区所有栅格的滑坡敏感性概率，绘制了博山区滑坡敏感性区划图。研究结果表明：滑坡致灾因子包括高程、坡度、坡向、剖面曲率、平面曲率、距河流距离、地表流沙输送量、地形湿度指数、距道路距离、土地利用类型、距断层距离、工程岩组和归一化植被覆盖指数；博山区滑坡敏感性概率最小为0.136、最大为0.841；极高敏感区、高敏感区、中敏感区、低敏感区和极低敏感区分别占博山区总面积的8.08%（56.4 km2）、17.62%（123.0 km2）、25.33%（176.8 km2）、32.87%（229.4 km2）和16.10%（112.4 km2），极高敏感区主要分布在博山区西北部、南部、东北部及其他地区，其中，自然滑坡主要分布于西北部极高敏感区，工程滑坡主要分布于东北部极高敏感区。Abstract: Landslide susceptibility regionalization aims to analyze the combined characteristics of hazard-inducing factors and their impact on the probability of landslide occurrence, thereby dividing the study area into different susceptible zones. This provides a theoretical basis for land use planning and the formulation of policies for landslide prevention and control. A landslide database was established for Boshan District by integrating various data sources and geological survey data. Hazard-inducing factors were selected using the Pearson correlation coefficient method, and these factors were classified for all landslides, natural landslides, and engineering landslides using the information content method. An improved 7-layer Convolutional Neural Network (CNN) with two rounds of convolution and pooling was constructed. Evaluation models were trained and verified separately for all landslides, natural landslides, and engineering landslides, and the accuracy of the models was assessed using the AUC method. The landslide susceptibility probabilities of all grid cells in Boshan district were calculated, and a landslide susceptibility regionalization map was generated using ArcGIS12.0. The results indicate that the hazard-inducing factors include elevation, slope, aspect, profile curvature, plane curvature, distance from rivers, STI, TWI, distance from roads, land use, distance from faults, lithology, and normalized difference vegetation index (NDVI). The minimum and maximum landslide susceptible probabilities in Boshan district are 0.136 and 0.841, respectively. The extremely high susceptibility, high susceptibility, medium susceptibility, low susceptibility, and extremely low susceptibility zones account for 8.08% (56.4 km2), 17.62% (123.0 km2), 25.33% (176.8 km2), 32.87% (229.4 km2) and 16.10% (112.4 km2) of the total area. The extreme high susceptibility zones are primarily distributed in the northwest, south, northeast and other areas. Natural landslides are mainly concentrated in the extremely high susceptibility zones, while engineering landslides are mainly found in the extremely high susceptibility zones in the northeast.
-
图 16 二维方阵结构形式
Figure 16. Two dimensional matrix structure form a） Based on All Landslides b） Based on Natural Landslides c） Based on Engineering Landslides
表 1 数据源和地质调查资料
Table 1. Data sources and geological survey data
数据源和地质调查资料 可提取的信息 数据提供者或下载地址 Landsat TM影像 土地利用类型、NDVI 地理空间数据云（ http://www.gscloud.cn/ ）博山区数字高程模型（DEM） 高程、坡度、坡向、剖面曲率、平面曲率、
距河流距离、SPI、STI、TWI山东省断层数据图 距断层距离 地质专业知识服务系统（ http://geol.ckcest.cn/ ）博山区道路数据 距道路距离 https://www.openstreetmap.org/#map=5/34.574/113.247 山东省工程岩组图 工程岩组 中国科学院地理科学与资源研究所资源环境科学与数据中心（ http://www.resdc.cn/ ）博山区降水数据、博山区地质灾害普查数据、
博山区地质灾害防治方案年平均降水量、滑坡数据 淄博市自然资源和规划局、淄博市交通运输局、博山区气象局 博山区工程岩组和地质灾害调查数据 工程岩组、滑坡数据 山东理工大学 
下载: 导出CSV
表 2 Pearson相关系数计算结果
Table 2. Calculation results of Pearson correlation coefficient
影响
因子高程 坡度 坡向 剖面
曲率平面
曲率年平均
降水量距河流
距离SPI STI TWI 距道路
距离土地
利用距断层
距离工程
岩组NDVI 高程 1.000 坡度 0.370 1.000 坡向 −0.052 −0.019 1.000 剖面曲率 −0.037 0.119 0.019 1.000 平面曲率 0.246 0.099 −0.092 −0.358 1.000 年平均降水量 0.269 0.163 −0.274 0.537 −0.325 1.000 距河流距离 0.190 −0.112 0.162 0.069 0.044 −0.638 1.000 SPI −0.034 0.002 0.091 0.257 −0.261 0.062 0.040 1.000 STI 0.163 0.369 0.047 0.366 −0.360 0.173 −0.053 0.826 1.000 TWI 0.060 −0.234 0.060 0.229 −0.333 −0.227 0.206 0.478 0.222 1.000 距道路距离 0.287 0.074 0.052 −0.027 0.060 −0.082 0.312 0.001 0.040 0.087 1.000 土地利用 0.356 0.368 −0.022 0.204 0.053 0.261 0.112 0.083 0.210 −0.164 0.171 1.000 距断层距离 0.085 0.098 0.113 0.050 −0.055 0.173 −0.021 0.037 0.149 0.012 0.287 0.041 1.000 工程岩组 −0.164 −0.104 0.006 0.005 −0.093 0.372 −0.093 −0.101 −0.105 0.006 −0.243 −0.085 −0.009 1.000 NDVI 0.191 0.110 −0.191 0.020 0.027 0.219 0.104 −0.018 0.012 −0.176 0.175 0.139 −0.128 −0.068 1.000
下载: 导出CSV
表 3 样本集概况
Table 3. Overview of the sample dataset
样本类别 自然滑坡栅格数 工程滑坡栅格数 无明显滑坡迹象的
普通边坡栅格数全部滑坡 619 841 1460 自然滑坡 619 0 619 工程滑坡 0 841 841
下载: 导出CSV
表 4 改进CNN的结构参数
Table 4. Structure parameters of the improved CNN model
层号 层名 层尺寸 过滤器尺寸 步长 零补方式 1 Input 13×13 0 0 NONE 2 Conv-1 13×13 3×3×4 1 SAME 3 Pool-1 13×13×4 2×2 2 VALID 4 Conv-2 6×6×4 3×3×8 1 SAME 5 Pool-2 6×6×8 2×2 2 VALID 6 F-layer 72 NONE NONE NONE 7 SoftMax 2 NONE NONE NONE
下载: 导出CSV
-
[1] BUI D T,TSANGARATOS P,NGUYEN V T,et al. Comparing the prediction performance of a Deep Learning Neural Network model with conventional machine learning models in landslide susceptibility assessment[J]. CATENA,2020,188:104426. doi: 10.1016/j.catena.2019.104426 [2] BERHANE G,KEBEDE M,ALFARAH N,et al. Landslide susceptibility zonation mapping using GIS-based frequency ratio model with multi-class spatial data-sets in the Adwa-Adigrat Mountain chains,northern Ethiopia[J]. Journal of African Earth Sciences,2020,164:103795. doi: 10.1016/j.jafrearsci.2020.103795 [3] YIN Chao. Hazard assessment and regionalization of highway flood disasters in China[J]. Natural Hazards,2020,100(2):535 − 550. doi: 10.1007/s11069-019-03824-9 [4] LIN Qigen,LIMA P,STEGER S,et al. National-scale data-driven rainfall induced landslide susceptibility mapping for China by accounting for incomplete landslide data[J]. Geoscience Frontiers,2021,12(6):101248. doi: 10.1016/j.gsf.2021.101248 [5] 吴森,张文,刘民生. 西藏地区崩塌滑坡影响因子敏感性分析[J]. 中国地质灾害与防治学报,2023,34(3):109 − 117. [WU Sen,ZHANG Wen,LIU Minsheng. Susceptibility analysis on influencing factors of rockfalls and landslides in Tibet[J]. The Chinese Journal of Geological Hazard and Control,2023,34(3):109 − 117. (in Chinese with English abstract) doi: 10.16031/j.cnki.issn.1003-8035.202201017WU Sen, ZHANG Wen, LIU Minsheng. Susceptibility analysis on influencing factors of rockfalls and landslides in Tibet[J]. The Chinese Journal of Geological Hazard and Control, 2023, 34（3）: 109-117. （in Chinese with English abstract） doi: 10.16031/j.cnki.issn.1003-8035.202201017 [6] ZHANG Tingyu,MAO Zhongan,WANG Tao. GIS-based evaluation of landslide susceptibility using a novel hybrid computational intelligence model on different mapping units[J]. Journal of Mountain Science,2020,17(12):2929 − 2941. doi: 10.1007/s11629-020-6393-8 [7] ZHOU Xinzhi,WEN Haijia,ZHANG Yalan,et al. Landslide susceptibility mapping using hybrid random forest with GeoDetector and RFE for factor optimization[J]. Geoscience Frontiers,2021,12(5):101211. doi: 10.1016/j.gsf.2021.101211 [8] YE Chengming,WEI Ruilong,GE Yonggang,et al. GIS-based spatial prediction of landslide using road factors and random forest for Sichuan-Tibet Highway[J]. Journal of Mountain Science,2022,19(2):461 − 476. doi: 10.1007/s11629-021-6848-6 [9] WANG Yi,FANG Zhice,HONG Haoyuan. Comparison of convolutional neural networks for landslide susceptibility mapping in Yanshan County,China[J]. Science of the Total Environment,2019,666:975 − 993. doi: 10.1016/j.scitotenv.2019.02.263 [10] MANDAL K,SAHA S,MANDAL S. Applying deep learning and benchmark machine learning algorithms for landslide susceptibility modelling in Rorachu River Basin of Sikkim Himalaya,India[J]. Geoscience Frontiers,2021,12(5):101203. doi: 10.1016/j.gsf.2021.101203 [11] FANG Zhice,WANG Yi,PENG Ling,et al. Integration of convolutional neural network and conventional machine learning classifiers for landslide susceptibility mapping[J]. Computers & Geosciences,2020,139:104470. [12] PHAM V D,NGUYEN Q H,NGUYEN H D,et al. Convolutional neural network—optimized moth flame algorithm for shallow landslide susceptible analysis[J]. IEEE Access,2020,8:32727 − 32736. doi: 10.1109/ACCESS.2020.2973415 [13] YIN Chao,WANG Zhanghua,ZHAO Xingkui. Spatial prediction of highway slope disasters based on convolution neural networks[J]. Natural Hazards,2022,113(2):813 − 831. doi: 10.1007/s11069-022-05325-8 [14] OUYANG Chaojun,WANG Zhongwen,AN Huicong,et al. An example of a hazard and risk assessment for debris flows—a case study of Niwan Gully,Wudu,China[J]. Engineering Geology,2019,263:105351. doi: 10.1016/j.enggeo.2019.105351 [15] 陈战军,杨静,宋庆华. 鲁中山区城市水生态文明建设模式探讨[J]. 中国水利,2017(16):34 − 36. [CHEN Zhanjun,YANG Jing,SONG Qinghua. Studies on construction model of water ecological civilization in central mountainous region of Shandong Province[J]. China Water Resources,2017(16):34 − 36. (in Chinese with English abstract) doi: 10.3969/j.issn.1000-1123.2017.16.010CHEN Zhanjun, YANG Jing, SONG Qinghua. Studies on construction model of water ecological civilization in central mountainous region of Shandong Province[J]. China Water Resources, 2017（16）: 34-36. （in Chinese with English abstract） doi: 10.3969/j.issn.1000-1123.2017.16.010 [16] 孙强,史清明. 基于GIS的城市地震灾害风险区划研究—以淄博市为例[J]. 四川地震,2020(2):19 − 24. [SUN Qiang,SHI Qingming. Study on urban earthquake disaster risk zoning based on GIS:A case study of Zibo city[J]. Earthquake Research in Sichuan,2020(2):19 − 24. (in Chinese) doi: 10.13716/j.cnki.1001-8115.2020.02.004SUN Qiang, SHI Qingming. Study on urban earthquake disaster risk zoning based on GIS—a case study of Zibo city[J]. Earthquake Research in Sichuan, 2020（2）: 19-24. （in Chinese） doi: 10.13716/j.cnki.1001-8115.2020.02.004 [17] 闫举生,谭建民. 基于不同因子分级法的滑坡易发性评价—以湖北远安县为例[J]. 中国地质灾害与防治学报,2019,30(1):52 − 60. [YAN Jusheng,TAN Jianmin. Landslide susceptibility assessment based on different factor classification methods:A case study in Yuan’an County of Hubei Province[J]. The Chinese Journal of Geological Hazard and Control,2019,30(1):52 − 60. (in Chinese with English abstract) doi: 10.16031/j.cnki.issn.1003-8035.2019.01.06YAN Jusheng, TAN Jianmin. Landslide susceptibility assessment based on different factor classification methods—a case study in Yuan'an County of Hubei Province[J]. The Chinese Journal of Geological Hazard and Control, 2019, 30（1）: 52-60. （in Chinese with English abstract） doi: 10.16031/j.cnki.issn.1003-8035.2019.01.06 [18] MUKHAMMADZODA S,SHOHNAVAZ F,ILHOMJON O,et al. Application of frequency ratio method for landslide susceptibility mapping in the surkhob valley,Tajikistan[J]. Journal of Geoscience and Environment Protection,2021,9(12):168 − 189. doi: 10.4236/gep.2021.912011 [19] PALAU R M,HÜRLIMANN M,BERENGUER M,et al. Influence of the mapping unit for regional landslide early warning systems:comparison between pixels and polygons in Catalonia (NE Spain)[J]. Landslides,2020,17(9):2067 − 2083. doi: 10.1007/s10346-020-01425-3 [20] Yang ZJ,Qiao JP. Entropy weight-based hazard degree evaluation of typical landslide[J]. Journal of Natural Disasters,2009,18(4):31 − 36. [21] SUN Deliang,XU Jiahui,WEN Haijia,et al. An optimized random forest model and its generalization ability in landslide susceptibility mapping:application in two areas of Three Gorges Reservoir,China[J]. Journal of Earth Science,2020,31(6):1068 − 1086. doi: 10.1007/s12583-020-1072-9 [22] KIEN N T,TRAN T V,LIEN V T H,et al. Landslide susceptibility mapping based on the combination of bivariate statistics and modified analytic hierarchy process methods:a case study of tinh tuc town,nguyen Binh district,Cao Bang Province,Vietnam[J]. Journal of Disaster Research,2021,16(4):521 − 528. doi: 10.20965/jdr.2021.p0521 [23] YESILNACAR E,TOPAL T. Landslide susceptibility mapping:A comparison of logistic regression and neural networks methods in a medium scale study,Hendek region (Turkey)[J]. Engineering Geology,2005,79(3/4):251 − 266. [24] LEE S,PRADHAN B. Landslide hazard mapping at Selangor,Malaysia using frequency ratio and logistic regression models[J]. Landslides,2007,4(1):33 − 41. doi: 10.1007/s10346-006-0047-y [25] ZÊZERE J L,PEREIRA S,MELO R,et al. Mapping landslide susceptibility using data-driven methods[J]. Science of the Total Environment,2017,589:250 − 267. doi: 10.1016/j.scitotenv.2017.02.188 [26] HUSSAIN M A,CHEN Zhanlong,KALSOOM I,et al. Landslide susceptibility mapping using machine learning algorithm:a case study along karakoram highway (KKH),Pakistan[J]. Journal of the Indian Society of Remote Sensing,2022,50(5):849 − 866. doi: 10.1007/s12524-021-01451-1 [27] 陈建平,王春雷,王雪冬. 基于CNN神经网络的煤层底板突水预测[J]. 中国地质灾害与防治学报,2021,32(1):50 − 57. [CHEN Jianping,WANG Chunlei,WANG Xuedong. Coal Mine floor water inrush prediction based on CNN neural network[J]. The Chinese Journal of Geological Hazard and Control,2021,32(1):50 − 57. (in Chinese with English abstract) doi: 10.16031/j.cnki.issn.1003-8035.2021.01.07CHEN Jianping, WANG Chunlei, WANG Xuedong. Coal Mine floor water inrush prediction based on CNN neural network[J]. The Chinese Journal of Geological Hazard and Control, 2021, 32（1）: 50-57. （in Chinese with English abstract） doi: 10.16031/j.cnki.issn.1003-8035.2021.01.07 [28] YAO Jingyu,QIN Shengwu,QIAO Shuangshuang,et al. Application of a two-step sampling strategy based on deep neural network for landslide susceptibility mapping[J]. Bulletin of Engineering Geology and the Environment,2022,81(4):148. doi: 10.1007/s10064-022-02615-0 [29] MA Xiangyun,ZHANG Shilei,SUN Jinglai,et al. A TFA-CNN method for quantitative analysis in infrared spectroscopy[J]. Infrared Physics & Technology,2022,126:104329. [30] WEI Ruilong,YE Chengming,SUI Tianbo,et al. Combining spatial response features and machine learning classifiers for landslide susceptibility mapping[J]. International Journal of Applied Earth Observation and Geoinformation,2022,107:102681. doi: 10.1016/j.jag.2022.102681 [31] LI Jiaying,WANG Weidong,CHEN Guangqi,et al. Spatiotemporal assessment of landslide susceptibility in Southern Sichuan,China using SA-DBN,PSO-DBN and SSA-DBN models compared with DBN model[J]. Advances in Space Research,2022,69(8):3071 − 3087. doi: 10.1016/j.asr.2022.01.043 [32] CIURLEO M,FERLISI S,FORESTA V,et al. Landslide susceptibility analysis by applying TRIGRS to a reliable geotechnical slope model[J]. Geosciences,2021,12(1):18. doi: 10.3390/geosciences12010018 [33] 朱楚雄,徐金明,钟传江. 基于全卷积神经网络的花岗岩中不同组分分布特征分析[J]. 中国地质灾害与防治学报,2021,32(1):127 − 134. [ZHU Chuxiong,XU Jinming,ZHONG Chuanjiang. Distributions of various compositions in granite specimen using fully convolutional network[J]. The Chinese Journal of Geological Hazard and Control,2021,32(1):127 − 134. (in Chinese with English abstract) doi: 10.16031/j.cnki.issn.1003-8035.2021.01.17ZHU Chuxiong, XU Jinming, ZHONG Chuanjiang. Distributions of various compositions in granite specimen using fully convolutional network[J]. The Chinese Journal of Geological Hazard and Control, 2021, 32（1）: 127-134. （in Chinese with English abstract） doi: 10.16031/j.cnki.issn.1003-8035.2021.01.17 [34] BOURENANE H,MEZIANI A A,BENAMAR D A. Application of GIS-based statistical modeling for landslide susceptibility mapping in the city of Azazga,Northern Algeria[J]. Bulletin of Engineering Geology and the Environment,2021,80(10):7333 − 7359. doi: 10.1007/s10064-021-02386-0 [35] FADHILLAH M F,HAKIM W L,PANAHI M,et al. Mapping of landslide potential in Pyeongchang-Gun,South Korea,using machine learning meta-based optimization algorithms[J]. The Egyptian Journal of Remote Sensing and Space Science,2022,25(2):463 − 472. doi: 10.1016/j.ejrs.2022.03.008 [36] HU Yicheng,ZHANG Weiqi,TOMINAC P,et al. ADAM:a web platform for graph-based modeling and optimization of supply chains[J]. Computers & Chemical Engineering,2022,165:107911. [37] 邹军华,段晔鑫,任传伦,等. 基于噪声初始化、Adam-Nesterov方法和准双曲动量方法的对抗样本生成方法[J]. 电子学报,2022,50(1):207 − 216. [ZOU Junhua,DUAN Yexin,REN Chuanlun,et al. Perturbation initialization,adam-nesterov and quasi-hyperbolic momentum for adversarial examples[J]. Acta Electronica Sinica,2022,50(1):207 − 216. (in Chinese with English abstract) doi: 10.12263/DZXB.20200839ZOU Junhua, DUAN Yexin, REN Chuanlun, et al. Perturbation initialization, adam-nesterov and quasi-hyperbolic momentum for adversarial examples[J]. Acta Electronica Sinica, 2022, 50（1）: 207-216. （in Chinese with English abstract） doi: 10.12263/DZXB.20200839 [38] 陈水满,赵辉龙,许震,等. 基于人工神经网络模型的福建南平市滑坡危险性评价[J]. 中国地质灾害与防治学报,2022,33(2):133 − 140. [CHEN Shuiman,ZHAO Huilong,XU Zhen,et al. Landslide risk assessment in Nanping City based on artificial neural networks model[J]. The Chinese Journal of Geological Hazard and Control,2022,33(2):133 − 140. (in Chinese with English abstract) doi: 10.16031/j.cnki.issn.1003-8035.2022.02-16CHEN Shuiman, ZHAO Huilong, XU Zhen, et al. Landslide risk assessment in Nanping City based on artificial neural networks model[J]. The Chinese Journal of Geological Hazard and Control, 2022, 33（2）: 133-140. （in Chinese with English abstract） doi: 10.16031/j.cnki.issn.1003-8035.2022.02-16 [39] HUANG Faming,CAO Zhongshan,JIANG Shuihua,et al. Landslide susceptibility prediction based on a semi-supervised multiple-layer perceptron model[J]. Landslides,2020,17(12):2919 − 2930. doi: 10.1007/s10346-020-01473-9 -
下载:
点击查看大图
图(18) / 表(4)
计量
- 文章访问数: 1015
- HTML全文浏览量: 183
- PDF下载量: 40
- 被引次数: 0
邮件订阅
RSS