Human activities have long been known to influence tropospheric O3, not only in urban areas where O3 is a major component of smog', but also in the remote atmosphere (e.g., IPCC, l990, l994; the SAR and references therein). The current state of scientific understanding of tropospheric O3 chemistry and trends is reviewed in Chapter 4 of this report, where it is emphasised that tropospheric O3 has an average lifetime of the order of weeks. This relatively short lifetime implies that the distribution of tropospheric O3, as well as the trends in that distribution (which in turn lead to radiative forcing) are highly variable in space and time. Studies relating to the evaluation of radiative forcing due to estimated tropospheric O3 increases since pre-industrial times are discussed here. While there are a number of sites where high quality surface measurements of O3 have been obtained for a few decades, there are fewer locations where ozonesonde data allow study of the vertical distribution of the trends, and fewer still with records prior to about l970. A limited number of surface measurements in Europe date back to the late l9th century. These suggest that O3 has more than tripled in the 20th century there (Marenco et al., l994). The lack of global information on pre-industrial tropospheric O3 distributions is, however, a major uncertainty in the evaluation of the forcing of this key gas (see Chapter 4).
Biomass burning plays a significant role in tropospheric O3 production and hence in tropical radiative forcing over large spatial scales, particularly in the tropical Atlantic west of the coast of Africa (e.g., Fishman, l99l; Fishman and Brackett, l997; Portmann et al., l997; Hudson and Thompson, l998) and in Indonesia (Hauglustaine et al., l999). Export of industrial pollution to the Arctic can lead to increased O3 over a highly reflective snow or ice surface, and correspondingly large local radiative forcings (Hauglustaine et al., 1998; Mickley et al., 1999).
Chapter 4 and Sections 6.6.2 and 6.6.3 discuss the chemistry responsible for the tropospheric O3 forcing; here we emphasise that the tropospheric O3 forcing is driven by and broadly attributable to emissions of other gases. The observed regional variability of O3 trends is related to the transport of key precursors, particularly reactive nitrogen, CO, and NMHCs (see Chapter 4). However, the chemistry of O3 production can be non-linear, so that increased emissions of, for example, the nitric oxide precursor do not necessarily lead to linear responses in O3 concentrations over all ranges of likely values (e.g, Kleinman, l994; Klonecki and Levy, l997). Further, the relationship of precursor emissions to O3 trends may also vary in time. One study suggests that the O3 production efficiency per mole of nitrogen oxide emitted has decreased globally by a factor of two since pre-industrial times (Wang and Jacob, l998). Because of these complex and poorly understood interactions, the forcing due to tropospheric O3 trends cannot be reliably and uniquely attributed in a quantitative fashion to the emissions of specific precursors.
For the purposes of this report, several evaluations of the global radiative forcing due to tropospheric O3 changes since pre-industrial times have been intercompared. It will be shown that the uncertainties in radiative forcing can be better understood when both the absolute radiative forcing (Wm-2) and normalised forcing (Wm-2 per Dobson Unit of tropospheric O3 change) are considered. The results of this intercomparison and the availability of numerous models using different approaches suggest reduced uncertainties in the radiative forcing estimates compared to those of the SAR. Furthermore, recent work has shown that the dependence of the forcing on the altitude where the O3 changes occur within the troposphere is less pronounced than previously thought, providing improved scientific understanding. Finally, some estimates of the likely magnitude of future tropospheric O3 radiative forcing are presented and discussed.
Other reports in this collection