Numerical Analysis of Potential Contaminant Migration from Abandoned In Situ Coal Conversion Reactors

Abstract

In the context of a potential utilisation of coal resources located in the Mecsek mountain area in Southern Hungary, an assessment of groundwater pollution resulting from a potential water-borne contaminant pool remaining in in situ coal conversion reactors after site abandonment has been undertaken in the scope of the present study. The respective contaminants may be of organic and inorganic nature. A sensitivity analysis was carried out by means of numerical simulations of fluid flow as well as contaminant and heat transport including retardation to assess spatial contaminant migration. Hereby, the main uncertainties, e.g., changes in hydraulic gradient and hydraulic contributions of the complex regional and local fault systems in the study area, were assessed in a deterministic way to identify the relevant parameters. Overall 512 simulations of potential groundwater contamination scenarios within a time horizon exceeding the local post-operational monitoring period were performed, based on maximum contaminant concentrations, cumulative mass balances as well as migration distances of the contaminant plume. The simulation results show that regional faults represent the main contaminant migration pathway, and that contamination is unlikely assuming the given reference model parametrisation. However, contamination within a simulation time of 50 years is possible for specific geological conditions, e.g., if the hydraulic conductivity of the regional faults exceeds a maximum value of 1 × 10−5 m s−1. Further, the parameter data analysis shows that freshwater aquifer contamination is highly non-linear and has a bimodal distribution. The bivariate correlation coefficient heatmap shows slightly positive correlations for the pressure difference, the fault permeability and the simulation time, as well as a negative correlation for the retardation coefficient. The results of this sensitivity analysis have been integrated into a specific toolkit for risk assessment for that purpose.