Landslide risk assessment along Sino-Nepal Highways in Nepal
文献类型：学位论文
| 作者 | NIRDESH NEPAL |
| 答辩日期 | 2020 |
| 授予单位 | 中国科学院大学 |
| 授予地点 | 北京 |
| 导师 | 陈剑刚 |
| 关键词 | 中尼公路 滑坡易发性 风险评估 尼泊尔地震 |
| 学位名称 | 硕士 |
| 学位专业 | 岩土工程 |
| 其他题名 | 中尼公路沿线的滑坡风险评估（尼泊尔境内段） |
| 英文摘要 | Landslide, one of the most prevalent phenomena in mountainous areas, is a major hazard in Nepal. After the commencement of Belt and Road Initiative (BRI), China plans to construct major highways and railways through seismically active and geologically fragile- Nepal Himalaya. Sino-Nepal highway in Nepal i.e. Pasang lhamu and Araniko highway selected for the current study are the only trade route between Nepal and China. It passes through deep river valleys and high slopes, and is highly susceptible to landslide. Three landslide dams were formed in past along this area, and it regularly suffers from damage and interruption due to slope failure. After Gorkha earthquake 2015, these highways have suffered huge losses due to seismically induced landslides. Subsequently, the presence of overwhelmed loose materials which can cause landslide threatens the safety of the people and infrastructure. The situation was further exacerbated due to the intervention of monsoon rainfall. This research, therefore, focuses on the assessment of landslide susceptibility and risk along these highway corridors. Landslide inventory and selection of safer areas was done interpreting high-resolution Google Earth imageries and was verified through the datasets from field investigation. Furthermore, other open source inventories available were also integrated in the present inventory. Categories of each variable factor was determined statistically using Shannon’s Entropy; and landslide susceptibility of the area was determined using the statistical models like Logistic Regression (LR), Artificial Neural Network (ANN) and Ensemble Method (EM). All of the three models applied resulted in accurate prediction based on the AUC (>0.9) of ROC curve, however ANN was slightly better than other models in terms of success rate curve. The conclusions are as follows:(1) From the visual inspection of the produced rainfall induced landslide susceptibility maps, most susceptible areas are concentrated on the debris influence zone. However, most susceptible areas for earthquake induced landslides are concentrated in the area with blind seismic rupture where maximum aftershock seismicity was observed after Gorkha earthquake. (2) In the case of altitude, altitude greater than 2300m was more susceptible in both cases.(3) Considering the land use patterns, landslide distribution and susceptibility is higher in the area closer to the major highways which may be correlated to the construction of non-engineered roads often referred to as “dozer road”. (4) Application of Shannon’s Entropy to determine categories of each variable factor has significantly improved the accuracy of landslide susceptibility of map.Furthermore, the division of susceptibility to different levels and identification of risky area has provided key insights to minimize the implications of landslide hazard in the area using different engineering measures and proper planning and prioritization of the protection work along the area. This can also serve as a pre-feasibility assessment tool for any development scheme in the area. Risk assessment was done based on the resource damage potential and its susceptibility based on the present scenario qualitatively. Integrated mitigation plan, including the establishment of monitoring and early warning system, based on economical bio-technical measures is proposed to be implemented in the risky area to cope with future risks. |
| 语种 | 英语 |
| 页码 | 172 |
| 源URL | [http://ir.imde.ac.cn/handle/131551/54998]  |
| 专题 | 成都山地灾害与环境研究所_山地灾害与地表过程重点实验室 |
| 作者单位 | 中国科学院成都山地灾害与环境研究所 |
| 推荐引用方式 GB/T 7714 | NIRDESH NEPAL. Landslide risk assessment along Sino-Nepal Highways in Nepal[D]. 北京. 中国科学院大学. 2020. |
入库方式： OAI收割
来源：成都山地灾害与环境研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。