A new approach to the modelling of the groundwater flow in fractured rock is presented in the paper. The empirical knowledge of the hydrogeologists is summarized first. There are three types of objects Important for the groundwater flow - small fractures, which can be replaced by blocks of porous media, large deterministic fractures and lines of intersection of the large fractures. These objects are by their nature 3D, resp. 2D, resp. 1D. The rest of the paper describes how to set up a numerical model representing all three types of objects. We use existing models based on the Mixed-Hybrid FEM and we connect them by equations representing the mass exchange between various types of elements. Our model uses two types of the connection of the elements, so-called compatible and incompatible type.
Numerical modelling of the hydraulic, geochemical and transport processes in the fractured rock attracts the attention of many scientists for more than forty years. The first numerical models of such processes Were created in late 60's of the 20-th century. According to , there existed more than thirty software packages claimed to solve problem of the fluid flow in fractured rock in 1994.
Despite these facts, there is a lot of open and unresolved problems in this field of research. The reason for such Situation lies in the nature of the problem.
Lack of input data, their uncertainty and often low accuracy, high computational costs are the main difficulties we encounter when we try to simulate processes in the fractured rock. Avoiding these difficulties is usually possible only at a price of simplifications of the problem.
Our research is motivated by the need of finding the most suitable locality for a permanent deep repository of the radioactive waste. There are two nuclear power plants in the Czech Republic, a construction of the repository is planned in the 30's of the 21-th century. Nevertheless, the process of selection of the most Suitable locality already began, as well as some other preliminary projects. Two of them are projects GAČR 102/04/P019 and MŽP VaV 660/2/03 focused on improving and testing existing numerical models a:n1i development of the new models. In this paper, we I show one of the results of these projects, a new approach for numerical models of groundwater flow which could be used for simulations of the processes in the large neighbourhood of the repository.
The radioactive waste repository will be situated in the compact crystalline rock massif. Of course, a good and reliable numerical model of the fluid flow and transport in such massif has to reflect its specific properties. The hydrogeologic research brought following empirical knowledge about the rock environment there and groundwater flow in them.
The rock matrix can be considered hydraulically impermeable.
Even the most compact massifs are disrupted by numerous fractures.