Air quality trends in U.S. western mountain states

The purpose of the study is to identify and quantify trends in measured air quality in Class I Areas that have IMPROVE and CASTNET monitoring sites in Colorado, New Mexico, South Dakota, Utah and Wyoming. Analysis of the monitoring data focuses on fine particulates, NO/NO2 and ozone. Observations suggest that concentrations are sometimes close to the current or proposed revised NAAQS. Various scientific reasons have been advanced for these relatively high concentrations (e.g., advection from far upwind source regions, stratospheric intrusions associated with vertical motion in frontal zones, and local sources) but our study is not intended to address the scientific causes. We first acquired the IMPROVE and CASTNET observations and associated meteorological observations for ten sites in the region for the 1989-2008 time period and placed the data in a master data archive suitable for analysis. The QA/QC procedures and uncertainties of the data were assessed. For example, the uncertainty in ozone design values for CASTNET sites is about 2 ppbv. Regarding trends, several analysis procedures have been tested, with the main focus on removing the effects of meteorology and known seasonal and diurnal cycles. This paper gives examples of preliminary analyses of trends at one site - Mesa Verde National Park. An upward trend in ozone at the Mesa Verde site was found to be robust and consistent, regardless of which statistical method was used. For PM, the data suggest a significant downward trend in sulfur and an upward trend in NO3. However, the trend direction (upward or downward) and significance is not so clear at other sites and for other pollutants. The next step is to understand the cause for high concentrations seen in both fine mass and nitrate in 2002 and 2003.