Interpretation of denitrification and stable isotope fractionation in a heterogeneous aquifer using backward particle tracking simulation

The mixing of groundwaters due to physical heterogeneity can complicate the inference of reactive transport characteristics from sample concentrations. This influence, however, has not been investigated systematically in heterogeneous aquifers. In this study, random walk particle tracking was used to evaluate the effects of mixing on inferred reaction rates and stable isotope fractionations. For a study site including a 1-km transect of multilevel wells near Merced, CA, flow and transport models and field observations were used to study the effects of physical heterogeneity on aquifer concentrations of environmental tracers including sulfur hexaflouride and CFC-12, reactive solutes including dissolved oxygen and nitrate, and stable isotopes of N in nitrate. A site-specific, 3-D flow model was built using MODFLOW with boundary conditions interpolated from a regional model. Realizations of geological variability were generated using transition- probability based geostatistics. Solute transport was simulated between the water table and well screens using a backward random walk particle tracking scheme which included age based input concentrations, spatially uniform reactions and isotopic fractionations. Results show that physical heterogeneity results in substantial mixing of waters in individual samples from short-screened monitoring wells. Consequently, the reaction rates of oxygen and nitrate are greater than is immediately apparent from the bulk sample concentrations. Oxygen reaction rates required to produce observed concentrations were higher for mixed samples than for unmixed samples by a factor of 1 to 2. Nitrate reaction rates for mixed samples were greater by a factor of 1 to 7. Rayleigh fractionation factors for mixed samples were greater in magnitude by a factor of 1 to 4. The fractionation factors for mixed and unmixed samples matched previously inferred values from homogeneous laboratory incubations and in-situ experiments, respectively. This similarity indicates that physical heterogeneity can feasibly explain the differences between isotopic fractionation factors inferred from laboratory versus field experiments.