Many processes involving atmospheric chemistry, and the coupling of atmospheric chemistry with other elements of global change, have been proposed in the scientific literature. These are generally based on sound physical and chemical principles, but unfortunately, there is no consensus on their quantitative role in atmospheric chemistry on a global scale (e.g., the effects of clouds on tropospheric ozone: Lelieveld and Crutzen (1990) vs. Liang and Jacob (1997)), on the magnitude of possible compensating effects (e.g., net settling of HNO3 on cloud particles: Lawrence and Crutzen (1999) vs. full cloud-scale dynamics), or even on how to implement them or whether these are already effectively included in many of the model calculations. While many of these processes may be important, there is inadequate information or consensus to make a quantitative evaluation in this assessment. This assessment is not a review, and so this section presents only a few examples of recent publications studying feedbacks or chemical processes, which are not included, but which are potentially important in this assessment.
Analyses and observations (see Section 4.2.6) continue to test and improve the chemistry and transport used in the global CTMs. In terms of the chemistry, recent studies have looked, for example, at the representation of NMHC chemistry (Houweling et al. 1998; Wang et al. 1998b), the role of halogens in the O3 budget of the remote marine troposphere, and the acetone source of upper tropospheric OH (see Sections 4.2.4 and 4.2.6). Most of these improvements in understanding will eventually become adopted as standard in the global CTMs, but at this stage, for example, the role of tropospheric halogen chemistry on the Y2100 predictions has not been evaluated in the CTMs.
Convection, as well as urban pollution and biomass burning plumes, occur on horizontal scales not resolved in global CTMs. These sub-grid features appear to be important in calculating OH abundances and O3 production for biomass burning emissions (Pickering et al., 1996; Folkins et al., 1997), for the remote upper troposphere (Jaeglé et al., 1997; Prather and Jacob, 1997; Wennberg et al., 1998), and in urban plumes (e.g., Sillman et al., 1990). Convection is represented in all CTMs here (e.g., Collins et al., 1999; Müller and Brasseur, 1999) but in quite different ways, and it still involves parametrization of processes occurring on a sub-grid scale. A substantial element of the differences in CTM simulations appears to lie with the different representations of convection and boundary layer transport, particularly for the short-lived gases such as NOx.
A change in the geographic emission pattern of the pollutants (NOx, CO and VOC) can by itself alter tropospheric O3 and OH abundances and in turn the abundances of CH4 and HFCs. In one study of regional NOx emissions and control strategy, Fuglestvedt et al. (1999) find that upper tropospheric O3 is most sensitive to NOx reductions in Southeast Asia and Australia and least to those in Scandinavia. Understanding trends in CO requires knowledge not only of the in situ chemistry of CO (e.g., Granier et al., 1996; Kanakidou and Crutzen, 1999), but also of how local pollution control has altered the global pattern of emissions (e.g., Hallock-Waters et al., 1999). These shifts have been included to some extent in the SRES emissions for year 2100 used here; however, the projected change in emission patterns have not been formally evaluated within the atmospheric chemistry community in terms of uncertainty in the Y2100 global atmosphere.
Other reports in this collection