Evaluation of Susceptibility under Different Landslide Sample Points and Polygonal Expression Modes
-
摘要: 滑坡编录模式常为点和多边形面，滑坡点的定位及多边形的采样范围会给滑坡易发性评价结果产生影响.为研究不同点和多边形滑坡样本采样方式下的易发性结果差异，以四川省宁南县为例，采用滑坡多边形和陡坎缓冲区来比较不同多边形表达模式对易发性评价的影响，用滑坡陡坎点和滑坡质心点来比较不同点表达模式对易发性评价的影响，选取3种评价模型支持向量机（SVM）、随机森林（RF）和人工神经网络（ANN）进行滑坡易发性建模，采用ROC曲线、均值、标准差等分析建模的差异.结果如下：（1）在滑坡样本为多边形表达模式下，陡坎缓冲区的评价效果优于滑坡多边形.在滑坡样本为点表达模式下，滑坡质心点的评价效果优于滑坡陡坎点.（2）RF模型在不同采样方式下易发性评价效果更好，不同采样方式下基于RF模型的易发性结果差异性也较小，相比SVM和ANN模型有更好的泛化能力.（3）离散型因子是导致点表达模式下采样方式易发性结果差异的主要因素.陡坎缓冲区采样方式相比于滑坡多边形保留如岩组等离散型环境因子的空间信息，因此评价效果较好.可见在县级尺度下使用滑坡陡坎区域等精细化地形特征作为滑坡采样方式可以提高易发性评价精度.Abstract: Landslide cataloging modes are usually points and polygons. The location of landslide points and the sampling range of polygons will affect the results of landslide susceptibility evaluation. In order to study the differences in the susceptibility results of different points and polygonal landslide sample sampling strategies, taking Ningnan County, Sichuan Province as an example, landslide polygons and landslide steep sill buffer zones were used to compare the susceptibility evaluation of different polygon expression patterns. The influence of landslide sill point and landslide mass center point was used to compare the influence of different point expression patterns on susceptibility evaluation, and three evaluation models were selected, namely, support vector machine (SVM), random forest (RF) and artificial neural network (ANN). Landslide susceptibility modeling was performed, and differences in modeling were analyzed using ROC curve, mean, and standard deviation. The results are as follows: (1) When the landslide samples are in the polygonal expression mode, the evaluation effect of the steep sill buffer zone is better than that of the landslide polygon. When the landslide sample is in a point expression mode, the evaluation effect of the landslide mass center point is better than that of the landslide steep point. (2) The susceptibility evaluation effect of the RF model is better under different sampling strategies, and the susceptibility results based on the RF model under different sampling strategies are also less different, and have better generalization ability than the SVM and ANN models. (3) The discrete factor is the main factor leading to the difference in the susceptibility results of the sampling strategy under the point expression pattern. Compared with the landslide polygon, the sampling strategy of the steep sill buffer preserves the spatial information of discrete environmental factors such as rock formations, so the evaluation effect is better. It can be seen that using refined terrain features such as landslide steep ridge areas as landslide sampling methods at the county scale can improve the accuracy of susceptibility evaluation.
-
图 1 不同滑坡样本表达模式的易发性评价流程
Fig. 1. Flow chart of susceptibility evaluation of expression patterns of different landslide samples
图 3 宁南县历史滑坡点分布及灾害点组成
资料来源：2021年四川省宁南县地质灾害风险评价报告. a.宁南县历史滑坡点；b.宁南县各个部门滑坡隐患点组成
Fig. 3. Distribution map of historical landslide points and composition of disaster points in Ningnan County
图 4 地形地貌、土地覆被、水文以及地质因子（a~l）
距断层距离和地形湿度指数未列入
Fig. 4. Topography, land cover, hydrology and geological factors (a‒l)
图 5 各采样方式下环境因子相关性系数
a.高程；b.坡向；c.坡度；d.平面曲率；e.剖面曲率；f.岩组；g.距河流距离；h.土地类型；i.谷深；j. NDVI；k.距道路距离；l.地形粗糙指数；m.距断层距离；n.TWI
Fig. 5. Correlation coefficients of environmental factors under each sampling strategy
图 6 各采样方式下环境因子重要性
Fig. 6. The importance of environmental factors under each sampling strategy
图 7 基于SVM、RF和ANN的易发性制图
Fig. 7. Susceptibility mapping based on SVM, RF and ANN
a~d. SVM; e~h. RF; i~l. ANN
图 8 不同组合工况下的滑坡易发性建模ROC曲线
Fig. 8. ROC curves of landslide susceptibility modeling under different combination conditions
图 10 不同采样方式下坡度差异性规律
a.滑坡坡度值对比图（仅展示部分滑坡）；b.不同滑坡编录模式坡度值差异
Fig. 10. Law of gradient difference under different sampling strategies
表 1 不同滑坡样本采样方式下环境因子FR值
Table 1. Evaluation factor FR value under different landslide sample sampling strategies
因子 因子分级 类型 4种滑坡采样方式下的环境因子FR值 滑坡多边形 陡坎缓冲区 滑坡陡坎点 滑坡质心点 高程 575 ~1 070 m 连续型 1.643 5 1.336 6 1.544 8 1.699 3 1 070 ~1 435 m 2.082 9 1.956 1 1.741 4 1.963 4 1 435 ~1 774 m 1.746 0 1.902 7 1.882 5 1.710 1 1 774 ~2 113 m 0.828 6 0.956 9 1.006 5 0.894 7 2 113 ~2 439 m 0.304 4 0.290 5 0.341 3 0.361 3 2 439 ~2 752 m 0.123 4 0.147 4 0.153 0 0.131 2 2 752 ~3 900 m 0 0 0 0 坡向 平面(‒1°) 连续型 0.328 1 0.159 4 0 0.199 8 北(0°~22.5°, 337.5°~360°) 0.944 9 0.986 9 1.041 5 0.972 1 东北(22.5°~67.5°) 1.445 9 1.410 1 1.232 8 1.147 2 东北(67.5°~112.5°) 1.202 5 1.170 7 1.211 0 1.252 2 东南(112.5°~157.5°) 1.221 5 1.205 7 1.070 8 1.200 1 南(157.5°~202.5°) 0.874 3 0.889 1 1.062 9 0.903 5 西南(202.5°~247.5°) 0.449 9 0.498 6 0.562 7 0.660 6 西(247.5°~292.5°) 0.539 4 0.539 2 0.554 6 0.526 9 西北(292.5°~337.5°) 0.900 3 0.955 0 1.062 9 1.062 9 坡度 0°~ 8° 连续型 0.569 4 0.429 7 0.083 2 0.749 2 8°~ 16° 1.715 6 1.616 5 1.509 2 1.930 4 16°~ 23° 1.623 8 1.656 3 1.567 8 1.762 1 23°~ 29° 1.153 8 1.197 6 1.129 5 1.099 4 29°~ 34° 0.638 4 0.678 6 0.889 5 0.475 4 34°~ 40° 0.448 5 0.453 2 0.554 0 0.332 4 40°~ 48° 0.270 2 0.289 2 0.458 9 0.098 3 48°~ 75° 0.082 6 0.113 3 0.216 3 0.000 0 平面曲率 ‒11.15 ~ ‒2.17 连续型 0.175 3 0.195 9 0.284 7 0 ‒2.17 ~ ‒1.15 0.433 3 0.378 7 0.240 9 0.301 1 ‒1.15 ~ ‒0.43 0.918 4 0.905 6 0.986 7 0.800 5 ‒0.43~ ‒0.031 1.298 9 1.328 4 1.258 3 1.368 7 ‒0.03 ~ 0.37 1.135 9 1.090 5 1.056 7 1.032 6 0.37 ~ 1.09 0.806 3 0.849 9 0.946 0 1.089 3 1.09 ~ 2.11 0.370 2 0.452 1 0.439 6 0.164 8 2.11 ~ 14.97 0.112 1 0.106 5 0.756 5 0 剖面曲率 ‒58.53 ~ ‒1.82 连续型 0.253 0 0.340 9 1.199 6 0.599 8 ‒1.82 ~ ‒0.92 0.475 2 0.543 7 0.604 1 0.278 8 ‒0.92 ~ ‒0.46 0.731 2 0.757 6 1.032 7 0.794 4 ‒0.46 ~ ‒0.44 1.177 9 1.177 8 1.112 1 1.248 1 ‒0.44 ~ 2.26 0.737 6 0.694 5 0.695 5 0.452 9 2.26 ~ 57.61 0.212 7 0.162 1 0.259 1 0.259 1 岩组 厚层块状灰岩、白云岩 离散型 0.280 5 0.279 6 0.270 1 0.283 4 灰岩夹软弱泥、页岩 0 0 0 0 软弱松散岩 1.175 0 1.297 6 6.973 0 2.073 9 砂岩夹软弱泥岩、页岩或互层岩组 2.769 8 2.686 5 0.286 5 2.716 1 块状玄武岩组 0.603 7 0.570 6 5.718 8 0.498 1 泥岩、页岩 0.982 1 0.990 1 0.239 0 0.909 6 石英砂岩、粉砂岩和软弱页岩 2.066 8 2.153 9 2.192 0 2.173 6 河流距离 0 ~ 184 m 连续型 0.852 5 0.508 8 0.583 9 1.038 1 184 ~ 381 m 1.261 8 1.346 8 1.386 8 1.374 6 381 ~ 590 m 1.099 5 1.233 4 1.296 2 0.911 6 590 ~ 823 m 1.082 0 1.232 3 1.192 3 1.138 9 823 ~ 1 098 m 0.827 3 0.842 2 0.766 7 0.618 3 1 098 ~ 1 466 m 0.792 9 1.028 6 0.708 0 0.458 1 1 466 ~ 2 037 m 0.639 2 0.701 2 0.497 0 0.596 4 2 037 ~ 3 632 m 0.118 9 0.134 7 0.226 7 0.226 7 土地利用 林地 离散型 0.464 9 0.518 6 0.622 4 0.520 7 草地 0.149 0 0.145 2 0.112 3 0.075 3 耕地 2.312 4 2.184 0 1.715 8 0.029 2 公路用地 1.706 2 1.996 9 3.450 3 3.468 5 河流 0.697 3 0.515 9 0.191 7 0.385 5 农村居民地 4.670 1 4.380 0 4.617 4 17.240 9 裸岩、裸土 0.624 2 0.820 8 2.281 4 0 内陆滩涂 0.153 8 0 0 0 园地 2.579 9 2.496 5 2.609 0 5.081 6 其他 0.875 0 1.139 0 0 3.956 8 工业用地 0.683 3 0.520 8 0 3.347 2 水库、坑塘 0.593 0 0.736 2 0 0 水工建筑用地 0.000 0 0 0 0 城镇居民地 0.524 1 0.449 4 0 0 采矿用地 1.362 0 0.482 2 0 0 谷深 ‒130 ~ 79 m 连续型 0.458 1 0.634 9 0.725 3 0.400 2 79 ~ 179 m 0.861 7 0.952 9 1.147 6 0.974 8 179 ~ 274 m 1.065 7 1.211 0 1.170 5 1.231 3 274 ~ 369 m 1.569 8 1.574 4 1.342 8 1.477 1 369 ~ 469 m 1.333 4 1.008 6 0.837 1 1.100 2 469 ~ 584 m 1.349 9 1.149 3 1.089 7 1.463 3 584 ~ 724 m 0.941 2 0.495 0 0.509 6 0.611 5 724 ~ 1 149 m 0.537 6 0.196 4 0.267 2 0.801 6 NDVI ‒0.356 ~ 0.009 连续型 0.031 3 0.021 5 0 0 0.009 ~ 0.177 1.197 0 1.139 7 1.010 3 1.437 7 0.177 ~ 0.256 1.509 1 1.474 7 1.418 2 1.435 9 0.256 ~ 0.341 1.563 5 1.512 8 1.382 0 1.803 4 0.341 ~ 0.434 1.350 2 1.342 4 1.409 4 1.231 0 0.434 ~ 0.532 0.988 1 1.007 8 1.270 2 1.008 7 0.532 ~ 0.630 0.487 1 0.524 7 0.468 8 0.388 0 0.630 ~ 0.836 0.191 4 0.244 7 0.244 2 0.122 1 距道路距离 0 ~ 707 m 连续型 1.491 3 1.454 6 1.497 0 1.577 7 707 ~ 1 604 m 1.314 3 1.402 3 1.421 6 1.275 4 1 604 ~ 2 634 m 1.524 0 1.465 3 1.270 8 1.305 2 2 634 ~ 3 816 m 0.525 6 0.532 5 0.580 6 0.603 8 3 816 ~ 5 148 m 0.097 3 0.114 8 0.204 6 0.175 4 5 148 ~ 12 503 m 0 0 0 0 地形粗糙指数 0 ~ 5.82 连续型 1.454 5 1.340 0 1.126 1 1.696 0 5.82 ~ 10.18 1.456 3 1.492 0 1.418 0 1.522 7 10.18 ~ 16.00 0.604 4 0.628 7 0.752 7 0.460 9 16.00 ~ 21.82 0.300 0 0.314 6 0.460 4 0.145 4 21.82 ~ 29.10 0.101 1 0.133 2 0.460 5 0 29.10 ~ 40.74 0.052 4 0.035 9 0 0 40.74 ~ 372.49 0 0 0 0 断层距离 0 ~ 408 m 连续型 1.324 9 1.286 1 1.375 9 1.337 9 408 ~ 873 m 1.162 9 1.231 6 1.103 3 1.269 4 873 ~ 1 415 m 1.000 2 0.999 3 1.066 4 0.883 1 141 ~ 2 026 m 0.741 3 0.722 1 0.661 7 0.617 6 2 026 ~ 2 721 m 0.572 6 0.527 9 0.453 7 0.538 8 2 721 ~ 3 546 m 0.422 3 0.469 6 0.506 9 0.506 9 3 546 ~ 4 586 m 0.730 6 0.708 3 0.742 3 0.742 3 4 586 ~ 6 832 m 0.745 2 0.739 1 0.886 7 0.886 7 TWI 0.53 ~ 5.16 连续型 0.386 1 0.541 7 0.892 8 0.378 1 5.16 ~ 6.18 0.843 5 0.937 2 1.096 8 0.967 2 6.18 ~ 7.29 1.418 2 1.362 4 1.068 5 1.412 9 7.29 ~ 8.68 1.766 0 1.523 5 1.231 2 1.634 8 8.68 ~ 10.62 1.465 9 1.248 1 0.712 9 1.273 0 10.62 ~ 13.03 0.739 9 0.510 4 0.358 6 0.478 1 13.03 ~ 16.92 0.500 2 0.122 6 0 0.333 6 16.92 ~ 24.23 0.546 8 0.131 6 0 0.841 4 
下载: 导出CSV
表 2 不同采样方式下RF模型的混淆矩阵和准确率
Table 2. Confusion matrix and accuracy of RF model under different sampling methods
滑坡多边形SVM 真实值 陡坎缓冲区SVM 真实值 滑坡陡坎点SVM 真实值 滑坡质心点SVM 真实值 准确率 0.751 准确率 0.764 准确率 0.764 准确率 0.771 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 滑坡 23 986 9 345 滑坡 17 748 6 546 滑坡 265 115 滑坡 288 92 非滑坡 7 276 26 055 非滑坡 4 933 19 361 非滑坡 94 286 非滑坡 82 298 滑坡多边形RF 准确率 0.986 陡坎缓冲区RF 准确率 0.996 滑坡陡坎点RF 准确率 0.868 滑坡质心点RF 准确率 0.854 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 滑坡 32 633 698 滑坡 24 141 153 滑坡 335 45 滑坡 331 49 非滑坡 234 33 097 非滑坡 42 24 252 非滑坡 55 325 非滑坡 62 318 滑坡多边形ANN 准确率 0.715 陡坎缓冲区ANN 准确率 0.734 滑坡陡坎点ANN 准确率 0.675 滑坡质心点ANN 准确率 0.701 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 预测值 滑坡 非滑坡 滑坡 25 371 7 960 滑坡 15 154 9 140 滑坡 263 117 滑坡 353 27 非滑坡 11 013 22 318 非滑坡 3 763 20 531 非滑坡 130 250 非滑坡 200 180
下载: 导出CSV
表 3 不同采样方式下易发性面积和滑坡百分比（RF模型）
Table 3. Susceptible area and percentage of landslides under different sampling strategies (RF model)
滑坡多边形 陡坎缓冲区 滑坡陡坎点 滑坡质心点 分区面积
(km2)滑坡面积占比(%) 分区面积(km2) 滑坡面积占比(%) 分区面积(km2) 滑坡面积占比(%) 分区面积(km2) 滑坡面积占比(%) 极低易发 1 023.32 0.00 968.27 0.00 454.41 0.00 573.43 0.79 低易发 289.67 0.07 332.50 0.00 457.74 3.42 469.06 6.84 中易发 156.32 0.65 179.88 0.07 359.77 12.11 344.27 24.74 高易发 99.13 5.93 103.45 2.04 270.36 30.26 200.33 36.05 极高易发 96.59 93.34 80.94 97.88 122.76 54.21 77.94 31.58
下载: 导出CSV
表 4 不同采样方式下易发性面积和滑坡百分比（SVM模型）
Table 4. Susceptible area and percentage of landslides under different sampling strategies (SVM model)
滑坡多边形 陡坎缓冲区 滑坡陡坎点 滑坡质心点 分区面积(km2) 滑坡面积占比(%) 分区面积(km2) 滑坡面积占比(%) 分区面积(km2) 滑坡面积占比(%) 分区面积(km2) 滑坡面积占比(%) 极低易发 967.64 5.56 651.61 1.52 569.38 2.89 613.51 8.95 低易发 211.00 9.16 481.24 9.51 380.40 8.42 633.53 16.58 中易发 135.21 9.21 165.10 11.33 249.39 13.68 178.87 27.89 高易发 127.41 13.07 145.04 14.19 229.40 20.53 118.69 14.47 极高易发 223.77 63.00 222.05 63.44 236.45 54.47 120.43 32.11
下载: 导出CSV
表 5 不同采样方式下易发性面积和滑坡百分比（ANN模型）
Table 5. Susceptible area and percentage of landslides under different sampling strategies (ANN model)
滑坡多边形 陡坎缓冲区 滑坡陡坎点 滑坡质心点 分区面积
(km2)滑坡面积占比
(%)分区面积
(km2)滑坡面积占比
(%)分区面积
(km2)滑坡面积占比
(%)分区面积
(km2)滑坡面积占比
(%)极低易发 903.83 5.76 614.60 0.86 349.73 2.11 922.38 17.37 低易发 273.99 12.53 297.76 4.32 456.84 6.84 482.83 22.11 中易发 191.17 17.27 301.07 13.49 397.55 24.47 114.74 28.95 高易发 221.83 34.46 281.07 36.09 253.35 22.37 78.13 8.42 极高易发 74.21 29.98 170.52 45.24 207.55 44.21 66.96 23.16
下载: 导出CSV
-
Abraham, M. T., Satyam, N., Lokesh, R., et al., 2021. Factors Affecting Landslide Susceptibility Mapping: Assessing the Influence of Different Machine Learning Approaches, Sampling Strategies and Data Splitting. Land, 10(9): 989. https://doi.org/10.3390/land10090989 Ali, R., Kuriqi, A., Kisi, O., 2020. Human-Environment Natural Disasters Interconnection in China: A Review. Climate, 8(4): 48. https://doi.org/10.3390/cli8040048 Bayat, M., Ghorbanpour, M., Zare, R., et al., 2019. Application of Artificial Neural Networks for Predicting Tree Survival and Mortality in the Hyrcanian Forest of Iran. Computers and Electronics in Agriculture, 164: 104929. https://doi.org/10.1016/j.compag.2019.104929 Cortes, C., Vapnik, V., 1995. Support-Vector Networks. Machine Learning, 20(3): 273-297. https://doi.org/10.1007/BF00994018 Hu, T., Fan, X., Wang, S., et al., 2020. Landslide Susceptibility Evaluation of Sinan County Using Logistics Regression Model and 3S Technology. Bulletin of Geological Science and Technology, 39(2): 113-121 (in Chinese with English abstract). Huang, F. M., Cao, Y., Fan, X. M., et al., 2021. Influence of Different Landslide Boundaries and Their Spatial Shapes on the Uncertainty of Landslide Susceptibility Prediction. Chinese Journal of Rock Mechanics and Engineering, 40(S2): 3227-3240 (in Chinese with English abstract). Huang, F. M., Hu, S. Y., Yan, X. Y., et al., 2022. Landslide Susceptibility Prediction and Identification of Its Main Environmental Factors Based on Machine Learning Models. Bulletin of Geological Science and Technology, 41(2): 79-90 (in Chinese with English abstract). Huang, F. M., Ye, Z., Yao, C., et al., 2020. Uncertainties of Landslide Susceptibility Prediction: Different Attribute Interval Divisions of Environmental Factors and Different Data-Based Models. Earth Science, 45(12): 4535-4549 (in Chinese with English abstract). Huang, F. M., Yin, K. L., Jiang, S. H., et al., 2018. Landslide Susceptibility Assessment Based on Clustering Analysis and Support Vector Machine. Chinese Journal of Rock Mechanics and Engineering, 37(1): 156-167 (in Chinese with English abstract). Li, W. B., Fan, X. M., Huang, F. M., et al., 2021. Uncertainties of Landslide Susceptibility Modeling under Different Environmental Factor Connections and Prediction Models. Earth Science, 46(10): 3777-3795 (in Chinese with English abstract). Liu, F. Z., Wang, L., Xiao, D. S., et al., 2021. Evaluation of Landslide Susceptibility in Ningnan County Based on Fuzzy Comprehensive Evaluation. Journal of Natural Disasters, 30(5): 237-246 (in Chinese with English abstract). Mao, Y. K., 2020. Stability Evaluation of Landslide Based on Machine Learning and System Development (Dissertation). University of Electronic Science and Technology of China, Chengdu (in Chinese with English abstract). Pardeshi, S. D., Autade, S. E., Pardeshi, S. S., 2013. Landslide Hazard Assessment: Recent Trends and Techniques. Springer Plus, 2(1): 523. https://doi.org/10.1186/2193-1801-2-523 Pourghasemi, H. R., Kornejady, A., Kerle, N., et al., 2020. Investigating the Effects of Different Landslide Positioning Techniques, Landslide Partitioning Approaches, and Presence-Absence Balances on Landslide Susceptibility Mapping. CATENA, 187: 104364. https://doi.org/10.1016/j.catena.2019.104364 Sahin, E. K., Colkesen, I., Kavzoglu, T., 2020. A Comparative Assessment of Canonical Correlation Forest, Random Forest, Rotation Forest and Logistic Regression Methods for Landslide Susceptibility Mapping. Geocarto International, 35(4): 341-363. https://doi.org/10.1080/10106049.2018.1516248 Süzen, M. L., Doyuran, V., 2004. Data Driven Bivariate Landslide Susceptibility Assessment Using Geographical Information Systems: A Method and Application to Asarsuyu Catchment, Turkey. Engineering Geology, 71(3/4): 303-321. https://doi.org/10.1016/S0013-7952(03)00143-1 Wu, R. Z., Hu, X. D., Mei, H. B., et al., 2021. Spatial Susceptibility Assessment of Landslides Based on Random Forest: A Case Study from Hubei Section in the Three Gorges Reservoir Area. Earth Science, 46(1): 321-330 (in Chinese with English abstract). Xie, P., Wen, H. J., Ma, C. C., et al., 2018. Application and Comparison of Logistic Regression Model and Neural Network Model in Earthquake-Induced Landslides Susceptibility Mapping at Mountainous Region, China. Geomatics, Natural Hazards and Risk, 9(1): 501-523. https://doi.org/10.1080/19475705.2018.1451399 Xu, Q., Lu, H. Y., Li, W. L., et al., 2022. Types of Potential Landslide and Corresponding Identification Technologies. Geomatics and Information Science of Wuhan University, 47(3): 377-387 (in Chinese with English abstract). Yilmaz, I., 2010. The Effect of the Sampling Strategies on the Landslide Susceptibility Mapping by Conditional Probability and Artificial Neural Networks. Environmental Earth Sciences, 60(3): 505-519. https://doi.org/10.1007/s12665-009-0191-5 Zhang, Y., Wu, W. C., Qin, Y. Z., et al., 2020. Mapping Landslide Hazard Risk Using Random Forest Algorithm in Guixi, Jiangxi, China. ISPRS International Journal of Geo-Information, 9(11): 695. https://doi.org/10.3390/ijgi9110695 Zhou, C., Yin, K. L., Cao, Y., et al., 2020. Landslide Susceptibility Assessment by Applying the Coupling Method of Radial Basis Neural Network and Adaboost: A Case Study from the Three Gorges Reservoir Area. Earth Science, 45(6): 1865-1876 (in Chinese with English abstract). Zhou, X. T., Wu, W. C., Lin, Z. Y., et al., 2021. Zonation of Landslide Susceptibility in Ruijin, Jiangxi, China. International Journal of Environmental Research and Public Health, 18(11): 5906. https://doi.org/10.3390/ijerph18115906 Zhu, A. X., Miao, Y. M., Yang, L., et al., 2018. Comparison of the Presence-Only Method and Presence-Absence Method in Landslide Susceptibility Mapping. CATENA, 171: 222-233. https://doi.org/10.1016/j.catena.2018.07.012 胡涛, 樊鑫, 王硕, 等, 2020. 基于逻辑回归模型和3S技术的思南县滑坡易发性评价. 地质科技通报, 39(2): 113-121. https://www.cnki.com.cn/Article/CJFDTOTAL-DZKQ202002013.htm 黄发明, 曹昱, 范宣梅, 等, 2021. 不同滑坡边界及其空间形状对滑坡易发性预测不确定性的影响规律. 岩石力学与工程学报, 40(S2): 3227-3240. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX2021S2023.htm 黄发明, 胡松雁, 闫学涯, 等, 2022. 基于机器学习的滑坡易发性预测建模及其主控因子识别. 地质科技通报, 41(2): 79-90. https://www.cnki.com.cn/Article/CJFDTOTAL-DZKQ202202008.htm 黄发明, 叶舟, 姚池, 等, 2020. 滑坡易发性预测不确定性: 环境因子不同属性区间划分和不同数据驱动模型的影响. 地球科学, 45(12): 4535-4549. doi: 10.3799/dqkx.2020.247 黄发明, 殷坤龙, 蒋水华, 等, 2018. 基于聚类分析和支持向量机的滑坡易发性评价. 岩石力学与工程学报, 37(1): 156-167. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX201801016.htm 李文彬, 范宣梅, 黄发明, 等, 2021. 不同环境因子联接和预测模型的滑坡易发性建模不确定性. 地球科学, 46(10): 3777-3795. doi: 10.3799/dqkx.2021.042 刘福臻, 王灵, 肖东升, 等, 2021. 基于模糊综合评判法的宁南县滑坡易发性评价. 自然灾害学报, 30(5): 237-246. https://www.cnki.com.cn/Article/CJFDTOTAL-ZRZH202105023.htm 毛宇昆, 2020. 基于机器学习的滑坡稳定性评价及系统研发(硕士学位论文). 成都: 电子科技大学. 吴润泽, 胡旭东, 梅红波, 等, 2021. 基于随机森林的滑坡空间易发性评价: 以三峡库区湖北段为例. 地球科学, 46(1): 321-330. doi: 10.3799/dqkx.2020.032 许强, 陆会燕, 李为乐, 等, 2022. 滑坡隐患类型与对应识别方法. 武汉大学学报(信息科学版), 47(3): 377-387. https://www.cnki.com.cn/Article/CJFDTOTAL-WHCH202203007.htm 周超, 殷坤龙, 曹颖, 等, 2020. 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. 地球科学, 45(6): 1865-1876. doi: 10.3799/dqkx.2020.071 -
点击查看大图
图(10) / 表(5)
计量
- 文章访问数: 658
- HTML全文浏览量: 207
- PDF下载量: 104
- 被引次数: 0
