Using rates of oxygen and nitrate reduction to map the subsurface distribution of groundwater denitrification

Widespread fertilizer application over the last 70 years has caused serious ecological and socioeconomic problems in aquatic and estuarine ecosystems. When surplus nitrogen leaches as nitrate (a major groundwater pollutant) to the aquifer, complex flow dynamics and naturally occurring degradation processes control its transport. Under the conditions of depleted oxygen and abundant electron donors, microorganisms reduce NO3- to N2 (denitrification). Denitrification rates vary over orders of magnitude among sites within the same aquifer, complicating estimation of denitrification capacity at the catchment scale. Because it is impractical or impossible to access the subsurface to directly quantify denitrification rates, reactivity is often assumed to occur continuous along flowlines, potentially resulting in substantial over- or underestimation of denitrification. Here we investigated denitrification in an unconfined crystalline aquifer in western France using a combination of common tracers (chlorofluorocarbons, O2, NO3-, and N2) measured in 16 wells to inform a time-based modeling approach. We found that spatially variable denitrification rates arise from the intersection of nitrate rich water with reactive zones defined by the abundance of electron donors (primarily pyrite). Furthermore, based on observed reaction rates of the sequential reduction of oxygen and nitrate, we present a general framework to estimate the location and intensity of the reactive zone in aquifers. Accounting for the vertical distribution of reaction rates results in large differences in estimations of net denitrification rates that assume homogeneous reactivity. This new framework provides a tractable approach for quantifying catchment and regional groundwater denitrification rates that could be used to improve estimation of groundwater resilience to nitrate pollution and develop more realistic management strategies