Combining Push Pull Tracer Tests and Microbial DNA and mRNA Analysis to Assess In-Situ Groundwater Nitrate Transformations

Abstract:

Nitrogen transformation mechanisms in the Upper Floridan Aquifer (UFA) are still poorly understood because of karst aquifer complexity and spatiotemporal variability in nitrate and carbon loading. Transformation rates have not been directly measured in the aquifer. This study quantifies nitrate-nitrogen transformation potential in the UFA using single well push-pull tracer injection (PPT) experiments combined with microbial characterization of extracted water via qPCR and RT-qPCR of selected nitrate reduction genes. Tracer tests with chloride and nitrate ± carbon were executed in two wells representing anoxic and oxic geochemical end members in a spring groundwater contributing area. A significant increase in number of microbes with carbon addition suggests stimulated growth. Increases in the activities of denitrification genes (nirK and nirS) as measured by RT-qPCR were not observed. However, only microbes suspended in the tracer were obtained, ignoring effects of aquifer material biofilms. Increases in nrfA mRNA and ammonia concentrations were observed, supporting Dissimilatory Reduction of Nitrate to Ammonia (DNRA) as a reduction mechanism. In the oxic aquifer, zero order nitrate loss rates ranged from 32 to 89 nmol /L^*hr with no added carbon and 90 to 240 nmol /L^*hr with carbon. In the anoxic aquifer, rates ranged from 18 to 95 nmol /L*hr with no added carbon and 34 to 207 nmol /L^*hr with carbon. These loss rates are low; 13 orders of magnitude less than the loads applied in the contributing area each year, however they do indicate that losses can occur in oxic and anoxic aquifers with and without carbon. These rates may include, ammonia adsorption, uptake, or denitrification in aquifer material biofilms. Rates with and without carbon addition for both aquifers were similar, suggesting aquifer redox state and carbon availability alone are insufficient to predict response to nutrient additions without characterization of microbial response. Surprisingly, these rates are consistent with reported rates of denitrification in other PPTs performed in basin fill aquifers which range from 10 to 360 nmol /L^*hr. Thus, assuming equivalency between nitrate loss rates and denitrification rates may be an oversimplification when microbial response, nutrient spiraling, and all reaction byproducts are not evaluated.