Back                 Printable Version

Predicting Change in Nitrogen Loading to Escambia Bay due to Additional Point Source
Controls on a Nearby Power Plant

Michele Cutrofello*, Jo Ellen Brandmeyer, Stephen Beaulieu, Randy Dodd
RTI International, 3040 Cornwallis Road, Research Triangle Park, NC 27709
Justin T. Walters, John J. Jansen
Southern Company, 600 N. 18th St., 14N-8195, Birmingham, AL 35203

Krish Vijayaraghavan
Atmospheric & Environmental Research, Inc., 2682 Bishop Dr, Ste 120, San Ramon, CA 94583


Pollution controls for NOx and SO2 are being installed at a power plant located near Escambia Bay in the Florida panhandle. The control technology is expected to change both the mass and species of nitrogen emitted to the atmosphere and, ultimately, reduce the nitrogen loading to the bay. The purpose of this modeling study was to estimate the difference in total nitrogen loading to the bay between the “no controls” and “controls” scenarios.

Wet and dry deposition of nitrogen species were simulated in a companion study using three configurations of the Community Multi-scale Air Quality (CMAQ) model and one configuration of the CALPUFF air dispersion model. An air quality simulation was performed for each scenario and each model configuration, resulting in eight sets of modeling results as inputs for the watershed modeling. The total nitrogen load to the bay attributable to plant emissions was the sum of (1) the direct deposition to Escambia Bay and (2) transport to the bay following deposition to the watershed.

In the initial set of watershed modeling runs, the nitrogen transfer following deposition to the watershed was calculated using a screening-level model based on EPA’s Pollutant Loading Model to provide a rapid estimate of the nitrogen load. The screening model provided a conservative estimate of the actual nitrogen load to the bay; that is, the model was set up and parameterized to ensure that the nitrogen load to the bay would not be underestimated. In the second set of watershed modeling runs, a more mechanistic model was used to improve the accuracy of nitrogen loading estimates to the bay. The ReNuMa (Regional Nutrient Management) Model developed by Cornell University was modified to use daily wet and dry atmospheric deposition fluxes of total nitrogen specific to land uses in the watershed. Following deposition, the model simulated a series of physical and chemical processes (e.g., soil retention; denitrification), and the loads from each watershed were routed from each watershed outlet to determine the total load to the bay. The contribution from atmospheric deposition, non-point source runoff, and point sources within each watershed was determined for both the screening and ReNuMa modeling runs.

This paper presents the methodology used to (1) estimate the total nitrogen from nitrogen species predicted by the atmospheric models, (2) calculate loads to the bay using the total nitrogen estimates, and (3) determine the change in nitrogen loading to the bay associated with emissions controls. Thus, the methodology is intended to fully support the evaluation of ecological benefits associated with changing nitrogen loads to the bay. In addition, the paper also describes how monitoring data were used to calibrate the watershed models and to measure model performance. We present ideas for how additional monitoring data could be used in this type of study to improve model performance.