Monte Carlo simulation of the effects of human intrusion on groundwater contaminant transport

Regulations for the long-term disposal of high-level radioactive wastes command that the performance of the disposal facility be assessed for very long periods of time and quite extensive spatial scales. By nature, then, the analysis of the performance of a disposal facility relies on the application of predictive models. However, these models are inherently affected by uncertainties which concern the future states of the disposal system, the assumptions on the relevant physicochemical processes and the numerical values of the parameters. In particular, the future states of the system are typically determined by climate-driven environmental changes, seismic and volcanic activities, and also by human actions. In this paper we perform a Monte Carlo simulation of the effects of human drilling on the transport of contaminant in groundwater. The framework provided by the Monte Carlo simulation is such as to allow incorporation of aspects of both stochastic and dynamic nature which are typical of these problems. The simulated time and location of drilling are described as aleatory variables. The stochastic occurrence of human drilling affects the dynamic process of contaminant transport in groundwater which is modelled, in our case, by means of the Kolmogorov-Dmitriev approach of branching stochastic processes recently proposed by the authors. A Monte Carlo estimate of the distribution of the time at which the contaminant level exceeds a prescribed safety threshold is obtained. (C) 1998 IMACS/Elsevier Science B.V.