Revisit the role of upslope natural hillside catchments in the stability of the chase cut slopes

Abstract

The CHASE study was initiated by the Geotechnical Control Office in May 1982 to study some selected soil cut slopes. Under this study, some natural terrain related attributes were also collected, including upslope catchment area and slope gradient, hydrological and terrain characteristics, vegetation cover, and geology. This dissertation Study is aimed at revisiting the CHASE study, focusing on the role of the upslope natural hillsides in the stability of the CHASE slope. The methodology of using LiDAR data analysis to identify anthropogenic features has been investigated. Upon matching the CHASE slopes with those in the Slope Catalogue of the Geotechnical Engineering Office, the catchments areas and upslope gradients have been re-assessed using GIS techniques on the LiDAR data. Some large discrepancies are found regarding the catchment areas between the CHASE data and the LiDAR-derived data whereas the discrepancy in the upslope gradients is relatively less. Despite the discrepancies, the patterns of frequency distribution are similar between the two data sets. Broadly speaking, hillsides with greater upslope gradients and larger catchment areas account for a higher percentage of the failed CHASE slopes. This is probably related to the intricate interplay between the amount of infiltration and the flow rate of the surface runoff associated with the hillside conditions. Of the various attributes assessed qualitatively, the following observations have been made: (i) A relatively high number of failed slopes belong to the “Exposed” infiltration class and “Sheet flow” drainage class. They may play a role in the stability of the CHASE slopes; (ii) Upslope terrain characteristics in the form of “Terrain component”, “Morphological attributes” and “Erosion” are not considered significant with respect to affecting the stability of the CHASE slopes; (iii) The number of failed slopes is significantly higher that of the stable slopes for volcanic rocks, suggesting that this geology type may play a role in the stability of the CHASE slopes; and (iv) Types of vegetation cover appear to have a role in the stability of the CHASE slopes, with grass cover showing a relatively higher number of stable CHASE slopes. A GIS-based methodology in analyzing LiDAR data has been developed for identification of anthropogenic features. Using LiDAR data of 0.5x5m2resolution, features such as roads, walls and excavations can be identified with confidence. With higher resolution LiDAR data, other smaller features could also be identified.