Use of the hydrogeological model for the Pos Selim landslide to examine different recharge conditions

Abstract

In 2003, a landslide movement occurred at Gunung Pass in the Cameron Highlands. Total station monitoring was started in October of the same year and captured the post-failure peak velocity. The object of this study is to further develop the simple hydrogeological model with input rainfall time-series (Malone 2019 personal communication), to obtain a water-pressure time-series output for Basal Rupture Surface (BRS) that corresponds to the velocity time-series measured at three clusters of the landslide. The hydrogeological model for the southern area of the landslide was modified by providing a 10m wide recharge zone at the inner edge of Berm 11. It was assumed that rainwater enters the recharge zone and the BRS in the form of a slug each day. Other possible values, such as permeability (k=1x10-4m/s), and effective porosity (n=10%), were set according to the literature. The parameters "proportion of rainwater reaching BRS" and "thickness of the BRS" were varied using the single variable method as experimental groups Series H set 2 and Series 2 set 3, respectively. The results showed that when the proportion of rainwater reaching BRS was raised from 20% to 25%, the peak of the Basal Rupture Surface Water Table (BRSWT) time-series increased by 2.5m; when the thickness of the BRS was reduced from 3m to 1m, the peak grew by 7.7m to 1370.5m. However, this peak value of 1370.5m is still less than the previous prediction by Limit Equilibrium Analysis (Zhu 2021). It was concluded that adjusting these two parameters had an effect on the increase of the peak BRS water table but changing one of the parameters alone would not be sufficient to achieve the predicted BRS water table. The contribution of possible recharge areas to the east of Gunung Pass to the water table of the landslide was also considered in this study. By assuming the flow nets for different conditions in two directions and observing the topographic map, it can be conjectured that river infiltration on the east highland may act as a recharge for the landslide. Seepage is visible at the toe of the landslide, which can show how the water table in the BRS reacts to rainfall. Seepage from the surface of toe was observed according to the photographs of the site taken 1 km away, from which the wet episodes were determined and compared to the rainfall data. A weakly discernible pattern between wet episodes and 7-day rolling rainfall can be seen from the comparison.