Use of MODIS Vegetation Data in Dynamic SPARROW Modeling of Reactive Nitrogen Flux

SPARROW models are widely used to identify and quantify the sources of contaminants in watersheds and to predict their flux and concentration at specified locations downstream. Conventional SPARROW models are steady-state in form, and describe the average relationship between sources and stream conditions based on non-linear regression of long-term water quality monitoring data on spatially-referenced explanatory information. But many watershed management issues involve intra- and inter-annual changes in contaminant sources, hydrologic forcing, or other environmental conditions which cause a temporary imbalance between watershed inputs and outputs. Dynamic behavior of the system relating to changes in watershed storage and processing then becomes important. We describe the results of dynamic statistical calibration of a SPARROW model of total reactive nitrogen flux in the Potomac River Basin based on seasonal water quality and watershed explanatory data for 80 monitoring stations over the period 2000 to 2008. One challenge in dynamic modeling of reactive nitrogen is obtaining frequently-reported, spatially-detailed input data on the phenology of agricultural production and growth of other terrestrial vegetation. In this NASA-funded research, we use the Enhanced Vegetation Index (EVI) and gross primary productivity (GPP) data from the Terra Satellite-borne MODIS sensor to parameterize seasonal uptake and release of nitrogen. The spatial reference frame of the model is a 16,000-reach, 1:100,000-scale stream network, and the computational time step is seasonal. Precipitation and temperature data are from PRISM. The model describes transient storage and transport of nitrogen from multiple nonpoint sources including fertilized cropland, pasture, urban/suburban land, and atmospheric deposition. Removal of nitrogen from watershed storage to stream channels and to "permanent" sinks (deep groundwater and the atmosphere) occurs as parallel first-order processes. Point sources of nitrogen bypass storage and flow directly to stream channels. Model results indicate that, on average, a little more than half of the reactive nitrogen flux comes from transient storage; but in some sub-watersheds a large majority of the flux comes from stored nitrogen input to the watershed in previous seasons and years.