How do Microbes Eat Rocks? Modelling Biogeochemical Reactions at the Rock/Water Interface in Fractured Limestone Aquifers

Abstract

Groundwater contamination of the fractured aquifers by agricultural pollutants has been happening worldwide since more than a century, boosting the scientific research to develop new modelling approaches for reactive transport simulation in aquifers. Spatially implicit methods like travel time approaches have been gaining interest, attracting by their lower computational demand and flexibility. At the same time, detailed mechanistic models of reactive transport in the fractured systems allow percipient understanding of underlying geochemical processes and provide veritable quantification of the latter. In this work, we connect a spatially explicit model of denitrification and isotope transport, and analytical solutions of atrazine transport in the fractured system, with a spatially implicit travel time approach. The work aims to study reactive behaviour of agriculturally produced contaminants on the catchment scale and quantify the fractured system parameters of the Muschelkalk aquifer. Reactive transport modelling was employed for this purpose with MIN3P and analytical solutions for a single fracture. Reactive transport models included advective flow in the fracture as well as possible (multicomponent) diffusive exchange with the rock matrix and redox processes taking place along the flow path. Determination of the travel times and transport parameters is done by direct modelling of tritium, helium, radiogenic helium, and argon-39 isotopes and accounting for multicomponent diffusion and radioactive decay within the streamline. Stochastic simulations of the atrazine transport under parametric uncertainty were employed for predictive quantification of travel time-dependent groundwater vulnerability. Conceptual model selection of denitrification and investigation of the redox evolution on the catchment scale was done with the MIN3P code. Results were verified with observations made across the Ammer catchment. The gained knowledge highlights the significance of comprehensive process-based hydrogeochemical modelling along with an uncertainty assessment on the catchment scale. It also demonstrates the relevance of reactive transport modelling for correct calibration as a prerequisite for prediction of the long-term evolution and transport of solutes in groundwater.