Working Group I: The Scientific Basis

Other reports in this collection Nitrogen oxides (NOx)

Nitrogen oxides (NOx = NO + NO2) do not directly affect Earth’s radiative balance, but they catalyse tropospheric O3 formation through a sequence of reactions, e.g.,

OH + CO + O2 CO2 + HO2
HO2 + NO NO2 + OH
NO2 + h NO + O(3P)
O(3P) + O2 + M O3 + M

net: CO + 2O2 + h CO2 + O3

By rapidly converting HO2 to OH, NO enhances tropo-spheric OH abundances and thus indirectly reduces the atmospheric burdens of CO, CH4, and HFCs. Much of recent understanding of the role of NOx in producing tropospheric O3 and changing OH abundances is derived from in situ measurement campaigns that sample over a wide range of chemical conditions in the upper troposphere or at the surface (see Section 4.2.6 on tropospheric OH). These atmospheric measurements generally support the current photochemical models. There is substantial spatial and temporal variability in the measured abundance of NOx, which ranges from a few ppt near the surface over the remote tropical Pacific Ocean to >100 ppb in urban regions. The local chemical lifetime of NOx is always short, but varies widely throughout the troposphere, being 1 day or less in the polluted boundary layer, day or night, and 5 to 10 days in the upper troposphere. As with VOC, it is not possible to derive a global burden or average abundance for NOx from measurements of atmospheric abundances.

Most tropospheric NOx are emitted as NO, which photochemically equilibrates with nitrogen dioxide (NO2) within a few minutes. Significant sources, summarised in Table 4.8, include both surface and in situ emissions, and only a small amount is transported down from the stratosphere. NOx emitted within polluted regions are more rapidly removed than those in remote regions. Emissions directly into the free troposphere have a disproportionately large impact on global greenhouse gases. The major source of NOx is fossil fuel combustion, with 40% coming from the transportation sector. Benkovitz et al. (1996) estimated global emissions at 21 TgN/yr for 1985. The NOx emissions from fossil fuel use used in model studies here for year 2000 are considerably higher, namely 33 TgN/yr. The large American and European emissions are relatively stable, but emissions from East Asia are increasing by about +4%/yr (Kato and Akimoto, 1992). Other important, but more uncertain surface sources are biomass burning and soil emissions. The soil source recently derived from a bottom-up compilation of over 100 measurements from various ecosystems is 21 TgN/yr (Davidson and Kingerlee, 1997), much higher than earlier estimates. Part of the discrepancy can be explained by the trapping of soil-emitted NO in the vegetation canopy. Inclusion of canopy scavenging reduces the NOx flux to the free troposphere to 13 TgN/yr, which is still twice the flux estimated by another recent study (Yienger and Levy, 1995). Emissions of NOx in the free troposphere include NOx from aircraft (8 to 12 km), ammonia oxidation, and lightning (Lee et al., 1997). Estimates of the lightning NOx source are quite variable; some recent global estimates are 12 TgN/yr (Price et al., 1997a,b), while other studies recommend 3 to 5 TgN/yr (e.g., Huntrieser et al., 1998; Wang et al., 1998a). Recent studies indicate that the global lightning frequency may be lower than previously estimated (Christian et al., 1999) but that intra-cloud lightning may be much more effective at producing NO (DeCaria et al., 2000). In total, anthropogenic NOx emissions dominate natural sources, with fossil fuel combustion concentrated in northern industrial regions. However, natural sources may control a larger fraction of the globe. Overall, anthropogenic NOx emissions are expected to undergo a fundamental shift from the current dominance of the Northern Hemisphere to a more tropical distribution of emissions. Asian emissions from fossil fuel are expected to drive an overall increase in NOx emissions in the 21st century (Logan, 1994; Van Aardenne et al., 1999).

Table 4.8: Estimates of the global tropospheric NOx budget (in TgN/yr) from different sources compared with the values adopted for this report.
Reference: TAR Ehhalt (1999) Holland et al. (1999) Penner et al. (1999) Lee et al. (1997)
Base year 2000 ~1985 ~1985 1992  
Fossil fuel 33.0 21.0 20 - 24 21.0 22.0
Aircraft 0.7 0.45 0.23 -0.6 0.5 0.85
Biomass burning 7.1 7.5 3 - 13 5 - 12 7.9
Soils 5.6 5.5 4 - 21 4 - 6 7.0
NH3 oxidation - 3.0 0.5 - 3 - 0.9
Lightning 5.0 7.0 3 - 13 3 - 5 5.0
Stratosphere <0.5 0.15 0.1 - 0.6 - 0.6
Total 51.9 44.6     44.3
Anthropogenic emissions by continent/region Y2000 Y2100(A2p)      
Africa 2.5 21.8      
South America 1.4 10.8      
Southeast Asia 1.2 6.8      
India 1.7 10.0      
North America 10.1 18.5      
Europe 7.3 14.3      
East Asia 5.6 24.1      
Australia 0.5 1.1      
Other 2.3 2.6      
Sum 32.6 110.0      
The TAR column was used in OxComp model calculations for year 2000; fossil fuel includes bio-fuels, but surface sources only; stratospheric source in TAR is upper limit and includes HNO3 ; the range of values used in modelling for IPCC aviation assessment (Penner et al. 1999) is given.

The dominant sink of NOx is atmospheric oxidation of NO2 by OH to form nitric acid (HNO3), which then collects on aerosols or dissolves in precipitation and is subsequently scavenged by rainfall. Other pathways for direct NOx removal occur through canopy scavenging of NOx and direct, dry deposition of NOx, HNO3, and particulate nitrates to the land surface and the ocean. Dry deposition can influence the surface exchanges and can thus alter the release of NOx and N2O to the atmosphere. Peroxyacetyl nitrate (PAN), formed by the reaction of CH3C(O)O2 with NO2, can transport HOx and NOx to remote regions of the atmosphere due to its stability at the cold temperatures of the upper troposphere. In addition tropospheric aerosols provide surfaces on which reactive nitrogen, in the form of NO3 (nitrate radical) or N2O5, is converted to HNO3 (Dentener and Crutzen, 1993; Jacob, 2000).

Some CTM studies argue against either the large soil source or the large lightning source of NOx. A climatology of NOx measurements from aircraft was prepared by Emmons et al. (1997) and compared with six chemical transport models. They found that the processes controlling NOx in the remote troposphere are not well modelled and that, of course, there is a paucity of global NOx measurements. For short-lived gases like NOx, resolution of budget discrepancies is even more challenging than for the long-lived species, because the limited atmospheric measurements offer few real constraints on the global budget. However, an additional constraint on the NOx budget is emerging as the extensive measurements of wet deposition of nitrate over Northern Hemisphere continents are compiled and increasing numbers of surface measurements of dry deposition of HNO3, NO2, and particulate nitrate become available, and thus allow a much better estimate of the NOx sink.

Other reports in this collection

IPCC Homepage