The relationship between the geological setting and the groundwater condition in a natural hill area in Ma On Shan

Abstract

Engineering projects involving excavation such as mining, tunneling and dam construction require the understanding of the groundwater condition. The hydraulic conductivity of a fractured crystalline rock mass is a critical factor in such projects. Lugeon Test is the typical test to obtain the in-situ permeability of the rock mass. To obtain a more precise estimation of hydraulic conductivity, the test section cannot be too long in order to reduce the range of variation of the affecting parameters. Large amount of Lugeon Test has to be done to construct a detailed profile of hydraulic conductivity which can be costly and time consuming. This dissertation aims to study the relationship between the parameters including depth, Rock Quality Designation (RQD) and other fracture parameters that affect the hydraulic conductivity of the rock mass in order to create an empirical model to estimate the conductivity. Data from 13 vertical and 22 inclined boreholes with 76 Lugeon Tests were analyzed. Regression analysis was done to evaluate the correlation relationship between the measured hydraulic conductivity and the factors including depth, RQD, fracture density and lithology which are available from borehole logs and Borehole Televiewer data. For this particular study area, it was found that the depth was not a useful indicator on the hydraulic conductivity. Instead, the average RQD gave the best correlation relationship with coefficient of determination R^2 equals 0.42 among all the factors; and the weathering grade also has certain degree of effect on the conductivity. It was concluded that the hydraulic conductivity of a rock mass is affected by a combination of factors, therefore the empirical model has to be able to consider several parameters at the same time. An empirical method called HC-System was proposed by Hsu et al. to estimate the hydraulic conductivity based on four parameters: Depth Index (DI), Rock Quality Designation (RQD), Gouge Content Designation (GCD) and Lithology Permeability Index (LPI). Three case studies were done to evaluate the application and performance of the HC-System and to examine the feasibility to use it in this study area. It was found that the performance differs from site to site and modifications must be made on how the parameters are considered in order to create a more accurate and reliable model. Modifications were made based on the dependence of the measured hydraulic conductivity on different individual factors. With the modified HC-System, a general solution was obtained with a coefficient of determination R^(2 ) equals 0.45, which is not a very well correlation relationship. By cluster analysis based on RQD and weathering grade, it was found that the model works well in the sections with Grade III as the highest weathering grade, with the equation K=0.0006×〖(HC)〗^2+3×〖10〗^(-5)×(HC)+1×〖10〗^(-7) and a coefficient of determination R2 equals 0.738. The estimation of hydraulic conductivity of a fractured rock mass is complicated. The proposed model can only be used to estimate the hydraulic conductivity for sections with Grade III rocks as the highest weathered materials. Further study is recommended to examine the geological and geotechnical factors in details in order to generate a more general and accurate empirical model.