Climatic effects of NOx emissions through changes in tropospheric O3 and CH4: A global 3D model study

As a result of chemical processes in the atmosphere, emissions of NOx can both damp and enhance the greenhouse effect. The two most important effects of NOx in this context are increased concentrations of tropospheric ozone and reduced levels of methane. The ozone response has a warming effect (positive indirect effect) and the methane response gives cooling (negative indirect effect). Previous studies using simplified models have shown that when NOx is emitted from ground sources, these effects may be of the same magnitude on a global scale. It is, however, important to be aware that since they have very different characteristics, these effects do not offset each other. In the free troposphere, the lifetime of ozone is 100-200 days and the changes in ozone and the resulting radiative forcing is limited to a regional scale. Changes in ozone and its radiative forcing occur relatively shortly after the emissions of NOx takes place. Methane, on the other hand, has a lifetime of 10-12 years and changes in methane degradation will therefore affect the concentration of methane on a global scale. The relatively long lifetime also leads to a corresponding time lag in the response of methane compared to changes in NOx emissions. Studies of the effects of NOx have usually focused on the effects of increased emissions. In this project we have studied the chemical and radiative effects of reduced NOx emissions, and how these effects vary among different regions of the world. The following areas have been chosen: Scandinavia, Central Europe, Southern Europe, USA, Southeast Asia, and Australia. In separate model tests the emissions of NOx have been reduced by 20% in each region, and the resulting changes in ozone and methane have been calculated. Changes in the concentrations of these gases have formed the basis for calculation of radiative forcing. In the case of Scandinavia it has been undertaken model tests where emissions of VOC and CO are also reduced. Calculations of the radiative forcing from both ozone and methane have been performed for Scandinavia (two tests), USA and Southeast Asia. The results confirm earlier findings that the indirect effects of NOx through methane and ozone are of the same magnitude, but with opposite signs. We also find that the effects vary significantly in magnitude among the various regions. In addition, we find that in a global and regional perspective, 20% reductions in NOx emissions in Scandinavia have a negligible effect on climate. The magnitude of NOx emissions in the studied regions varies substantially (a factor of 35 between USA and Scandinavia). This is due to differences in area size, population density, as well as economic and technological conditions. USA and Southeast Asia will experience the largest effect of a 20% reduction in NOx emissions on the concentrations of ozone and methane. This is due to the large emissions in USA and the chemical and meteorological conditions in Southeast Asia. Scandinavia separates from the rest by having the smallest effect. The very large differences in changes in NOx emissions in absolute numbers necessitate a normalisation of the effects of changes in emissions; i.e. that changes in ozone and methane are calculated per Tg (1012 g) change in NOx emissions. The same has been done for radiative forcing (given as W/m2/TgN/yr). A ranking of the regions according to changes in the most climate sensitive altitudes of ozone concentrations per reduction in NOx emissions, shows that Southeast Asia is the most sensitive area, followed by Australia. USA and Europe follow thereafter, with Scandinavia ranking slightly below. Concerning radiative forcing resulting from changes in ozone (radiative forcing per reduction in TgN), Southeast Asia is about 8 times more sensitive than Scandinavia, while USA is about 20% more sensitive. We find that the sensitivity of radiative forcing due to increased methane concentrations is about 6 times higher for NOx reductions in Southeast Asia than for NOx reductions in Scandinavia. The sensitivity of NOx reductions in the USA is about 30% higher than in Scandinavia. While the climate effect of a 20% reduction in NOx in Scandinavia is negligible, the radiative forcing of reductions in ozone resulting from a 20% reduction in NOx emissions in the USA is significant. However, if indirect radiative forcing of NOx reductions is given per unit reductions in NOx emissions, the estimates indicate that the sensitivities of radiative forcing from ozone and methane in Scandinavia and USA are comparable. The estimates also show that changes in the emissions of other gases are very important. Model experiments that at the same time reduced the emissions of VOC and CO in Scandinavia by 30% (test 2) resulted in a radiative forcing of around 60% lower (less negative) than if only NOx emissions were reduced. The reason for this is that oxidation products from these gases take part in chemical reactions that remove ozone when NOx levels are sufficiently low. This fact underlines the need for considering various measures for emission reductions jointly and to analyse how measures affect several gases simultaneously. A model experiment was also undertaken where emissions of VOC, CO and NOx were reduced by 30% in Scandinavia (test 3). The results showed only minor deviations from test 2 (20% reduction in NOx and 30% reduction in VOC and CO), but on a global scale the effect on ozone was somewhat lower in test 3. This is a result of complex relationships in the atmospheric chemistry as well as varying impacts of transport on the distribution of gases that leads to ozone formation. Reduced levels of NOx and O3 and hence lower concentrations of OH in one region result in lower oxidation of CO and hydrocarbons. This in turn increases the supply of these ozone precursors in areas further south, and enhances the ozone levels there. Because of the complex role of NOx in the chemistry of the atmosphere as well as large spatial and temporal variations in the NOx concentration, there are significant uncertainties associated with calculations of changes in ozone and methane. This implies that, at present, it is not possible to give any firm conclusions regarding the net effect of NOx on the radiative forcing, since the warming and the cooling effects are of the same magnitude. Estimations of net effects of NOx may be of limited value as the radiative forcing of ozone and methane have different characteristics and may lead to different effects on the atmospheric circulation patterns. Despite the uncertainties concerning quantification of indirect climatic effects of NOx emissions, several studies support our conclusion that NOx both has warming and cooling effects and that these probably are of the same magnitude. In order to study the chemical responses on changes in NOx emissions, we have used a global 3-dimensional model (longitude x latitude x elevation) with an extensive chemistry scheme as opposed to previous studies that used 2-D models. Radiative forcing resulting from changes in ozone is calculated by professor Wang and his group at the State University of New York at Albany, USA.