Report No. : GEO Report No. 69
Report Title : Diagnostic Report on the November 1993 Natural Terrain Landslides on Lantau Island (1998), 98 p. plus 1 drg.
Author : H.N. Wong, K.C. Lam & K.K.S. Ho
In the severe rainstorm of 4 and 5 November 1993, over 800 landslides occurred on the natural terrain on Lantau Island. A systematic study of the natural terrain failures in three selected areas on Lantau was carried out by the Special Projects Division of the Geotechnical Engineering Office as a project under the R&D Theme on Landslide Field Studies. A factual documentation of the investigations and observations of the landslides in the three study areas is given by Wong et al (1996).
This Report analyses the data collected and diagnoses the landslides in the three study areas. The study has highlighted the importance of having a rational classification of the different types of failure and modes of debris movement for a detailed assessment of the risk posed by landslides.
The field inspections have shown that the majority of the landslides inspected were shallow failures involving loose bouldery colluvium. The mobility of the landslide debris varied greatly and was dependent on the mechanisms of the debris movement.
It has not been able to reliably assess from the November 1993 rainstorm the effects of rainfall on landslide propensity of comparable land units because the potential local variability in rainfall intensities is not known with sufficient accuracy. The difficulty in the assessment is further exacerbated by the lack of information on the times of occurrence of the landslides.
Correlations between landslide propensity and characteristics of natural terrain components based on the GASP data, including geology, slope gradient, signs of past instability and erosion and landform, have been examined. However, it is likely that the accuracy of the correlations is limited by the resolution of the GASP data and the way the parameters were collected, as well as possible interdependence of the factors. Nevertheless, such broad correlations may be useful for general zoning purposes or global risk assessment, provided that due consideration is given to the limits of applicability of the correlations.