You are here:
“Transference Ratios” to Predict Total Oxidized Sulfur and Nitrogen Deposition – Part II, Modeling Results
Citation:
Sickles, J., D. Shadwick, Vasu Kilaru, AND W. Appel. “Transference Ratios” to Predict Total Oxidized Sulfur and Nitrogen Deposition – Part II, Modeling Results. ATMOSPHERIC ENVIRONMENT. Elsevier Science Ltd, New York, NY, 77:1070-1082, (2013).
Impact/Purpose:
The US Environmental Protection Agency (EPA) is considering the use of “transference ratios” in the formulation of Secondary National Ambient Air Quality Standards (NAAQS) for oxides of nitrogen and oxides of sulfur as a means to estimate their total atmospheric deposition (US EPA, 2011). The concept of a “transference ratio,” introduced in US EPA (2011), was defined as the ratio of total (i.e., dry plus wet) deposition to the airborne concentration of the species of interest. This expresses the notion that the amount of an acidifying species deposited to the earth’s surface within a given area over a given period of time is proportional to the average airborne concentration of the acidifying species over the same area during the same time period, where the proportionality constant is called a transference ratio.
Description:
The current study examines predictions of transference ratios and related modeled parameters for oxidized sulfur and oxidized nitrogen using five years (2002-2006) of 12-km grid cell-specific annual estimates from EPA’s Community Air Quality Model (CMAQ) for five selected sub-regions in the eastern US. Modeled total deposition and modeled airborne concentrations are moderately to poorly correlated in the selected sub-regions of the eastern US. The monitored oxidized nitrogen species (OxN) considered in the current study include airborne gaseous nitric acid (HNO3) and particulate nitrate (NO3 -), and nitrate ion (NO3 20 -) in precipitation. Modeled airborne OxN accounts for approximately 20% of the modeled airborne concentration of the total reactive oxidized oxidized nitrogen (NOY) but is responsible for approximately 80% of the modeled total deposition of NOY. Airborne OxN concentration may be a better predictor of total deposition of NOY in some locations; whereas, airborne NOY concentration may be better in others. Modeled airborne concentration and total deposition for both oxidized sulfur and OxN tend to be higher than corresponding monitoring-based results, suggesting a need for both model refinement and more comprehensive comparisons with monitoring results. Model estimated annual transference ratios have both spatial variability (grid cell-to-grid cell) and temporal variability (across the five modeled years). Two approaches are explored to investigate the impacts of modeled spatial variability of transference ratios on estimates of regional total deposition. Assuming proportionality rather than equality between cell-specific and regional deposition appears to reduce the impact of modeled spatial variability 32 of transference ratios on estimates of regional total deposition by a substantial margin, to perhaps ±20% for oxidized sulfur and ±10% for oxidized nitrogen. Overall estimates of the variability of estimated regional total deposition were made, considering modeled cell-to-cell spatial variability and modeled year-to-year temporal variability along with the variability of monitored airborne concentration. The variability of monitored airborne concentration was found to influence strongly the resulting variability in regional total deposition estimated with input from CMAQ. Examination of five years of annual model predictions (along with an assumed annual CV of 10% for the monitored airborne concentration) suggests that in the sub-regions that were considered, estimates of modeled regional total deposition have maximum 95% CIs of ±26% for oxidized sulfur and ±25% for NOY. These findings should be considered with caution because they are based almost entirely on modeled annual results (i.e., modeled spatial and modeled temporal variability along with variability of monitored airborne concentrations), and they fail to consider several sources of uncertainty, including discrepancies between model predictions and monitoring results as well as important deposition processes.