Soil nitrogen oxide emissions and their impacts on air quality over the United States.

Soil emissions of nitric oxide (NO) and nitrous acid (HONO) contribute to ozone, which is becoming more sensitive to non-fossil fuel nitrogen oxides (NOx = NO, NO2) as fossil fuel emissions decline. Soil emissions of NO and/or HONO (SNO/HONO) arise from complex biogeochemical processes and are intensified by nitrogen additions from fertilizers. Despite this, most air quality models use simple parameterizations of these processes to quantify soil NO emissions. In this study, we compare the Yienger & Levy (YL) and the Berkeley-Dalhousie Soil NOx Parameterization (BDSNP) soil NO models. These models are compared with the mechanistic Fertilizer Emissions Scenario Tool for CMAQ (FEST-C*) that simulates agricultural soil NO and HONO emissions. We quantify the impacts of each emission estimate on ozone and particulate matter with the Community Multiscale Air Quality (CMAQ) model over the United States. We compare modeled nitrogen dioxide (NO2) concentrations with observations at rural monitoring sites and modeled NO2 vertical column densities with satellite measurements. Agricultural SNO/HONO from FEST-C* were double that of YL but two-thirds that of BDSNP. FEST-C* improved CMAQ's agreement with some rural NO2 observations. Satellite NO2 columns showed FEST-C* having intermediate correlation and bias with the lowest bias over agricultural areas. We also examine how each SNO/HONO estimate affects ozone pollution in several cities near agricultural areas. Higher SNO/HONO increased the number of days exceeding ozone regulatory thresholds in some cities. This study highlights needs for field SNO/HONO observations, better fertilizer use estimates, and mitigating the impact of agricultural emissions on ambient pollution.