Reactive Nitrogen

Reactive nitrogen (NOy) consists of NO and its oxidation products, the most important of which are NO, NO, HNO and aerosol NO, peroxyacetyl nitrate (PAN) and other organic nitrates, NO, NO5, HNO, and HNO. The fraction of reactive nitrogen which is present as NO (the sum of NO and NO) is particularly important in both the global troposphere and local urban areas. This is because NO plays a major role in the formation of tropospheric O and smog. In regions of high NO, photochemical sequences initiated by the reaction of CO with OH or the reaction of nonmethane hydrocarbons (NMHC's) with OH lead to O formation. In regions of low NO (as observed over remote ocean areas), the photochemical sequences lead to O destruction. NOx concentrations also partly control the concentration of OH. At low NMHC concentrations, increases in NO lead to increases in OH up to NOx concentrations of a few tenths of a ppb. Increases in NO above these levels lead to decreases in OH. In environments with significant NMHC concentrations the turnover level can be higher.

A significant fraction of NO in the troposphere is present as nitric acid (HNO) and, in the marine boundary layer, as aerosol nitrate (NO). These two components are important because they are major contributors to acid deposition. Also, their deposition to nitrogen-poor ecosystems and ocean areas can provide an important nutrient for these systems, leading to local, but short-lived blooms. PAN is a long-lived and important component of NOy, especially at the colder temperatures present at high altitudes and latitudes and in winter. It's ability to carry reactive nitrogen from distant sources at higher altitudes and then to decompose to form NOx may influence the production of ozone in relatively clean environments. Clearly, the contribution of anthropogenic NO to the total abundance of NO and its chemical form are of great environmental concern.

A series of measurement programs have increased our understanding of the abundance of NO and its chemical partitioning into specific components. Total reactive nitrogen was measured at mid to high altitudes in the troposphere and lower stratosphere in the Stratosphere-Troposphere Exchange Program (STEP), the Airborne Antarctic Ozone Experiment (AAOE), and the Airborne Arctic Stratospheric Expeditions (AASE). These have been summarized by Murphy et al. (1992) who indicate that tropospheric NO is far more variable than stratospheric NO when measured relative to the ozone concentration. These observations point to the importance of local sources (aircraft, lightning, or convective mixing of lower altitude air) in determining the nitrogen content of this region. The nitrogen content of this region and, in particular, the NO concentration in the upper troposphere, is of great interest due to the controlling effect of NOx on O and the fact that O changes at these altitudes contribute to the greenhouse forcing of climate. Honrath and Jaffe (1992) measured the seasonal cycle of NO and NO at Barrow, Alaska and suggest that at this northern location PAN decomposition during spring contributes to both a pulse of NO and a decay of NO. The decay was associated with the formation HNO which is more rapidly removed than PAN. The partitioning of NO has been explored in several measurement campaigns. Arctic air masses in the lower troposphere were sampled during the Arctic Boundary Layer Experiments (3A and 3B) (Harriss et al., 1992; Harriss et al., 1994). Concentrations of NO increased from a median of 350 ppt at low altitudes to 600 ppt at higher altitudes (Sandholm et al., 1992). The largest single component at the highest altitudes measured (6 km) was PAN and its increase with altitude explained the increase in NO. Lower troposphere marine air masses were sampled during the Mauna Loa Observatory Photochemistry Experiment (MLOPEX) (Ridley and Robinson, 1992). Median NOy concentrations were 262 ppt during downslope winds (Hubler et al., 1992) and the major component was HNO. PAN concentrations explained less than 10% of total NO (Atlas et al., 1992) although it was the dominant component of reactive nitrogen in the equatorial Pacific (Atlas et al., 1993). These campaigns all point to a shortfall of varying magnitude between the sum of the individually measured major nitrogen containing species and the concentration of total NO. This shortfall ranges from 25% at Mauna Loa, to between 30 and 60% in over Alaska and Hudson Bay, but good closure has been found over the Eastern U.S. in summer (Atlas et al., 1992; Sandholm et al., 1992; Parrish et al., 1993; Sandholm et al., 1994). The identity of the nitrogen compound(s) that may explain the shortfall is still unknown. Alternatively, the shortfall may result from an instrumental error which has not yet been quantified.

Global models of the tropospheric nitrogen budget are still unable to explain the measured abundance and composition of NO. Penner et al. (1991) used simplified chemistry to treat the cycle of reactive nitrogen in a three-dimensional model. They included a representation of all the known major sources of reactive nitrogen (fossil fuel combustion, biomass burning, soil microbial activity, lightning, and transport from the stratosphere). Kasibhatla et al. (1991, 1993) used a simplified but more complete chemistry in their three-dimensional study. However, only stratospheric or fossil fuel sources were treated. Both studies predict higher nitrate concentrations in remote regions in the Northern Hemisphere than in the Southern Hemisphere consistent with observations, but the predicted levels of nitrate in remote locales in both models are too low. Kasibhatla et al. (1993) were also unable to reproduce observed levels of PAN at high Northern latitudes. Singh et al. (1994) suggest that previously unincluded sources of acetone may explain some of the discrepancy in modeled and predicted PAN concentrations observed at high northern latitudes.

Work to refine the soil source of reactive nitrogen has continued (Williams et al., 1992; Bakwin et al., 1992; Hutchinson and Brams, 1992; Valente and Thornton, 1993). Tropical grasslands remain the most important source on a global basis in undisturbed ecosystems, but fertilized croplands can be quite important regionally. The sources of NH from ecosystems, a component of the tropospheric nitrogen cycle whose importance lies mainly in its ability to neutralize acid aerosols, have also received study (Langford et al., 1992).