The Evolution of the Nocturnal Inversion in Denver using Multi-level 40 Meter Tower Data: Analysis and Comparison with nearby NWS Data and the Differences on Predicted Concentrations using the AERMOD Dispersion Model

Denver, Colorado has had a long and checkered history when it comes to air pollution. The topography of the region frequently contributes to nocturnal inversions which lead to degraded air quality during both winter and summer seasons. The Denver region is home to several atmospheric research labs and universities with sophisticated meteorological equipment deployed throughout the region. Unfortunately, Denver is also home to several Superfund sites that warranted significant cleanup over the past twenty years. One such site, the Rocky Mountain Arsenal (RMA), a former chemical weapons and pesticide manufacturing site, required extensive cleanup that occasionally produced tremendous odors in nearby communities. In anticipation of this problem and to address community concerns, a 40 m meteorological tower with a mixture of wind and temperature sensors at 2, 10, 25, and 40m was operational from 2003-2010. This study will present the results from an analysis of hourly meteorological data from 2004-2008 to show the evolution of the very shallow nocturnal boundary by season at the RMA. Denver has also developed an air quality management system that links meteorological data, emissions data, the AERMOD air dispersion model, and air monitoring data with a Geographic Information System (GIS) to produce visual maps of citywide air quality for a suite of pollutants. Meteorological data from standard 10 meter National Weather Service (NWS) stations at the former Stapleton Airport and the Denver International Airport have been used in previous iterations of AERMOD. These two sites are four miles and 12 miles away from the RMA multi-level tower, respectively. Recent advances in the dispersion model and the meteorological preprocessor, AERMET, allow for easier incorporation of onsite data and can handle data collected at one-minute averaging periods. This feature has reduced the frequency of calm winds as compared to traditional (i.e. hourly average) NWS data. Traditionally, calms hours were excluded from the dispersion model but the new data presents challenges in that it includes a higher percentage of low wind speed hours, which in turn can result in higher predicted concentrations. Denver will present the results from the AERMOD dispersion model generated using the same emissions but with three different meteorological datasets: 1) 1986-1990 Stapleton Airport surface and upper air data, 2) 2002-2006 surface data from Denver International Airport paired with upper air data collected at the former Stapleton Airport site, and 3) 2004-2008 multi-level tower data at RMA paired with upper air data collected at the former Stapleton Airport site. The differences in the meteorological data collected will be explained. Finally, the validation of AERMOD dispersion model predictions with observed ambient concentrations will be presented. Based on this analysis, recommendations for the appropriate dataset in Denver will be made.