Denitrification in heterogeneous aquifers: Relevance of spatial variability and upscaling

Denitrification is a key process in the fate of groundwater nitrate and an important, spatially distributed parameter to accurately assess the fate of groundwater nitrate from non-point source contamination. The spatial variability in nitrate reduction within an aquifer is particularly challenging to describe given the large scale of NPS contamination and the diversity and complexity of the biochemical and physical processes in place.

This work investigates the relevance of an explicit description of the spatial variability in denitrification rate when assessing nitrate concentration in extraction wells located in a large aquifer system. We simulate stochastically heterogeneous aquifer flow and transport conditions and compare different conceptual approaches to characterizing the spatial variability of denitrification rates.

Our results show that the spatial variability in the denitrification rate, described at the scale of aquifer structural heterogeneity, has significant impact on concentration statistics recorded at extraction wells. The conceptualization of the heterogeneity in the aquifer denitrification capacity further controls concentration statistics behavior. On the other hand, we develop an upscaling process to evaluate an effective denitrification rate that reproduces virtually perfectly matching nitrate concentration statistics, regardless of the model chosen to represent the denitrification rate heterogeneity. Upscaled, effective rates are semi-analytically evaluated for each realization and each extraction well. This is an important result given the uncertainty in characterizing spatially distributed biochemical conditions over a groundwater basin at a fine resolution.

Still, the accurate estimation of effective denitrification rates even at larger scales remains a challenging task since the parameter integrates complex interactions between aquifer physics and biochemistry along the solute path.