当前位置: X-MOL 学术Geomat Nat. Hazards Risk › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A comparative study on the integrative ability of the analytical hierarchy process, weights of evidence and logistic regression methods with the Flow-R model for landslide susceptibility assessment
Geomatics, Natural Hazards and Risk ( IF 4.5 ) Pub Date : 2020-01-01 , DOI: 10.1080/19475705.2020.1846086
Hien Minh Do 1, 2 , Kun Long Yin 1 , Zi Zheng Guo 1
Affiliation  

Abstract The main purpose of the present study is to compare the results of the integrated models, the weights of evidence (WoE), analytical hierarchical process (AHP) and logistic regression (LR) models combine with a Flow-R model for landslide susceptibility assessment in Ha Giang city, Vietnam and the surrounding areas. First, three landslide susceptibility index (LSI) maps were calculated using thirteen landslide conditioning factor maps. Then, a runout map was generated by Flow-R model using the digital elevation model and the slope map of the study area. Secondly, the success rate curve, prediction rate curve and area under the curves (AUC) were calculated to assess prediction capability. The final landslide susceptibility maps were produced by the integration of the LSI maps with the runout map that was extracted from FlowR model. The validation results showed that the AUC of LSI maps in the integrated models, AHP-FlowR, WoE-FlowR and LR-FlowR, were 82.89%, 88.15% and 86.53% with prediction accuracies of 80.41%, 84.94% and 85.71%, respectively. The combined models of WoE-FlowR (88.15%) and LR-FlowR (86.53%) have shown the highest prediction capability. In general, the integrated models provide reasonable results that may be aided in prevention planning and land use planning purposes.

中文翻译:

滑坡敏感性评估中层次分析法、证据权重和逻辑回归方法与Flow-R模型整合能力的比较研究

摘要 本研究的主要目的是比较综合模型的结果,证据权重 (WoE)、层次分析法 (AHP) 和逻辑回归 (LR) 模型与 Flow-R 模型相结合进行滑坡敏感性评估。在越南河江市及周边地区。首先,使用十三个滑坡条件因子图计算三个滑坡敏感性指数(LSI)图。然后,利用数字高程模型和研究区的坡度图,通过 Flow-R 模型生成跳动图。其次,计算成功率曲线、预测率曲线和曲线下面积(AUC)以评估预测能力。最终的滑坡敏感性图是通过将 LSI 地图与从 FlowR 模型中提取的跳动图整合而成的。验证结果表明,集成模型AHP-FlowR、WoE-FlowR和LR-FlowR中LSI图的AUC分别为82.89%、88.15%和86.53%,预测准确率分别为80.41%、84.94%和85.71% . WoE-FlowR(88.15%)和LR-FlowR(86.53%)的组合模型显示出最高的预测能力。一般而言,综合模型提供了可能有助于预防规划和土地利用规划目的的合理结果。
更新日期:2020-01-01
down
wechat
bug