Tên bài báo:
Application of Classification and Regression Trees for Spatial Prediction of Rainfall Induced Shallow Landslides in the Uttarakhand Area (India) Using GIS- Tác giả:
- Phạm Thái Bình
- Tham gia cùng:
- Tạp chí:
- Climate Change, Extreme Events and Disaster Risk Reduction
- Năm xuất bản:
- 2016
- Trang:
- Từ trang 1 đến trang 19
- Lĩnh vực:
- Kỹ thuật xây dựng công trình giao thông
- Phạm vi:
- Quốc tế
Tóm tắt:
Landslide is defined to be a mass movement of slope materials from up to downslope under various geo-environmental conditions. It is well known as one of the most serious geo-hazards causing loss of human life and properties throughout the world. In the present study, we present an application of Classification and Regression Trees (CART) for spatial prediction of rainfall induced shallow landslides in the Uttarakhand area (India) using GIS. A total of 430 historical landslide locations have been first identified to construct landslide inventory map. In addition, eleven landslide influencing factors (slope angle, slope aspect, elevation, curvature, lithology, soil type, land cover, distance to roads, distance to rivers, distance to lineaments, and rainfall) have been taken into account for analyzing the spatial relationship with landslide occurrences. Moreover, the predictive capability of the CART model has been validated using statistical analysis based evaluations and. Overall; the CART model performs well for spatial prediction of landslides. Its performance is even better than other landslide models (Naïve Bayes and Naïve Bayes Trees). Therefore, the CART indicates as encouraging alternative method which could be used for landslide prediction in landslide prone areas. The results obtained from this study would be helpful for landslide preventing and combating activities in the study area.
