Identification of Anthropogenic Impact on Nitrogen Cycling Using Stable Isotopes and Distributed Hydrologic Modeling

Nitrogen contamination of water resources is a significant problem worldwide. Drastic shifts in agroeconomic policy over the past few decades have changed the manner and intensity of anthropogenic nitrogen inputs in the United States and China. It is crucial to gain an understanding of nitrogen cycling within varying environmental contexts and land uses and the interactions among them. Analysis of the spatial distribution of sources and processes affecting the concentration of NO3- and NH4+ is a useful exercise; however, unexpected characteristics of the system can be revealed through reactive modeling. It has been shown that management practices intended to attenuate nitrate in surface and subsurface waters, in particular the establishment of riparian buffer zones, are variably effective due to spatial heterogeneity of soils and preferential flow through buffers. Accounting for this heterogeneity in a fully distributed biogeochemical model allows for more efficient planning of management practices and enforcement of rules regarding releases. Highly sensitive areas within a watershed can be identified based on a number of spatially variable parameters, and by varying those parameters systematically to determine conditions under which those areas are under more or less critical stress. Responses can be predicted at various scales to stimuli ranging from local changes in cropping regimes to global shifts in climate.
A distributed hydrologic model, TREX, is presented that provides opportunities to study multiscale effects of nitrogen inputs, outputs, and changes. The model is adapted to run on parallel computing architecture and includes the geochemical reaction module PhreeqcRM, which enables calculation of δ15N and δ18O from biologically mediated transformation reactions in addition to mixing and equilibration.
This work presents simulations of conditions showing low antecedent nitrogen retention versus significant contribution of old nitrate. Nitrogen sources are partitioned using dual isotope ratios and temporally varying concentrations. In these two scenarios, we are able to evaluate the efficiency of source identification based on spatially explicit information.