Assessing trends in urban ground-level concentrations of nitrogen dioxide

Abstract

Tillerson, Clint. Master of Environmental Assessment. Assessing Trends in Urban Ground-level Concentrations of Nitrogen Dioxide A review of the National Ambient Air Quality Standards (NAAQS) is required of the U.S. Environmental Protection Agency every five years. Included in the review typically is an assessment of health effects associated with air quality adjusted to meet the existing and potential alternative standards. In 2010, a 1-hour standard for nitrogen dioxide (NO2) was established to protect the population from short-term exposure found to cause adverse health effects in sensitive populations. Ambient monitoring data collected over the last 30+ years indicate NO2 concentrations have progressively lowered throughout all urban areas in the United States such that no areas are currently above the existing daily maximum 1-hour (DM1H) standard of 100 parts per billion. For the current and likely subsequent reviews of the NO2 NAAQS, it is necessary to adjust monitored concentrations upward to meet the existing standard (or some alternative if greater than existing conditions) in a manner that estimates realistic concentrations across the entire distribution. A prior analysis of ambient air quality data strongly supported that concentrations could reasonably be adjusted proportionally across the entire distribution, though there was evidence that the relationship might be nonlinear in the lower and upper percentiles of the distribution. This study revisits the trends in the reduction of ambient NO2 over time and compares two methods for adjusting NO2 concentrations upward. The first method assumes a proportional relationship exists across the distribution of DM1H concentrations while a second method assumes the same proportional relationship up to and including the 98th percentile DM1H and a nonlinear relationship above the 98th percentile. While this study is not intended to definitively define the appropriateness of either adjustment (proportional vs. proportional with nonlinear above 98th percentile DM1H) in representing the hypothetical high concentration air quality scenario, there is evidence to suggest assuming proportionality across the entire distribution could result in a significant underestimation of concentrations at the upper percentiles of the distribution.