Comparisons of in situ and long path measurements of NO2 in urban plumes

Measurements of NO2 were taken at Cornelia Fort Airpark in Nashville, TN, during the 1999 Southern Oxidant Study using three different techniques: photolysis to NO followed by chemiluminescence (PCL), laser-induced fluorescence (LIF), and differential optical absorption spectroscopy (DOAS). This was an informal comparison of these techniques conducted during the 1999 Southern Oxidant Study. The PCL and LIF instruments were connected to a common manifold that sampled at the top of a 10-m-walkup tower. The DOAS instrument sampled over a 1.37-km-long light path with end points at 2 and 35 m above ground. The range of NO2 mixing ratios measured was 0.75 ppbv to over 60 ppbv and the median value was nearly 3 ppbv. While preliminary data analysis showed overall agreement between the LIF and PCL instruments to within 1% (least squares slope = 0.99; r2 = 0.98), subsequent analysis revealed a discontinuous shift of about 12% in the PCL data, which was confirmed by comparison to the DOAS data. A leak in the PCL inlet system was the likely cause. After adjustment of the affected PCL data, a comparison of all the coincident measurements showed high correlation (r2 > 0.99) and overall agreement to within 5%. Analysis of the ratios of PCL NO2 to LIF NO2 showed that greater than 90% of individual data points agree to within the total combined instrumental uncertainties. However, the comparison over short time periods is more precise than the average over the campaign. We attribute this to the need for improved PCL instrument data reduction procedures. The two in situ instruments were also operated side by side a year later in Houston, TX, with similar results.