Measurements of nitrogen oxides and a simple model of NOy fate in the remote North Atlantic marine atmosphere

NO and NOy (total reactive oxidized nitrogen) were measured at a site in the Azores (21.322 degrees W, 38.732 degrees N, 1 km altitude) over a 3 week period in August and September 1993, during the first summer intensive of the North Atlantic Regional Experiment (NARE). These measurements were performed to determine background reactive nitrogen oxides levels and to assess the impact that long-range transport of reactive nitrogen oxides associated with human activities have on these levels. Median NOy mixing ratios during background marine boundary layer (MBL) periods ranged from 59 to 93 parts per trillion by volume (pptv), with an overall median of 73 pptv. Analysis of back trajectories, low and uncorrelated CO and O-3 levels, and low levels of MBL NOy indicate that the central North Atlantic region was not influenced by direct transport of anthropogenic emissions during the period of this study. Changes in NOy levels during two MBL periods with adjacent in and out-of-cloud events indicated that up to 45-47% of MBL NOy was scavenged by clouds. However, mean NOy levels during all in-cloud periods (similar to 70 pptv) and all out-of-cloud periods (similar to 80 pptv) were not significantly different, apparently because of variability in MBL NOy levels. In addition to the MBL periods, there were two periods when the site was within the free troposphere (FT), as indicated by vertical soundings and weather conditions at the site. FT NOy mixing ratios were greater than or equal to 280 pptv and greater than or equal to 400 pptv during these two periods. The median clear-sky FT NO level during the hour centered on solar noon was 16 pptv. A mass balance model considering FT/MBL exchange and MBL removal processes is used to find the NOy MBL effective first-order loss lifetime (similar to 1.2 days) and the NOy MBL e-folding response time due to both effective first-order loss processes and subsidence-induced ventilation of the MBL (similar to 0.9 days). The apparent rapid loss of MBL NOy implies that it will respond rapidly to changes in the overlying FT, but that correlations of NOy with trace gases with slower MBL removal, such as O-3 and CO, will be degraded within the subsidence-influenced MEL.