Geological discrete fracture network (DFN) model for the granite in Po Toi Island in Hong Kong

Abstract

In order to find the most suitable location for a water well on Po Toi Island, a discrete fracture network (DFN) of this island's granite was studied. The aim of this study is to analyse the geometric statistical patterns of fractures and then carry out a simulation of a DFN model for the granite in this area based on the obtained fracture data from multiple methods. The fractures, as an important channel for water transport, have a significant impact on the groundwater occurrence. In particular, the granite has extensive fracture development, it is impractical to obtain accurate information about each fracture. The three-dimensional DFN model is a stochastic simulation result which built on the statistical laws. This model does not require a quantitative description of each fracture and can serve to visualize the distribution of fractures within the rock mass and to grasp the structural features that are difficult to measure. A number of outcrops in the southeastern part of Po Toi Island were selected for fieldwork to obtain basic data for the study of fractured rock mass. Firstly, the geometric parameters of the 175 measured fractures were statistically analyzed, and the fractures were divided into 5 groups based on fracture attitude. The number and optimum trend and plunge of each group were obtained, then the fracture trace length and density were calculated. After that, Monte Carlo method was used to simulate the random fractures in the selected area according to the statistical probability density function of each parameter. Then a 3D fracture network model of the rock mass can be generated. Finally, the model’s accuracy and reasonableness are verified by means of data comparison.