19.4.3 Synthesis

The studies reviewed in this section diverge widely in their methodological approach, in the sophistication with which uncertainties are considered in geophysical, biological and social systems, and in how closely they approach an explicit examination of key vulnerabilities or DAI. The models involved range from stand-alone carbon cycle and climate models to comprehensive integrated assessment frameworks describing emissions, technologies, mitigation, climate change and impacts. Some frameworks incorporate approximations of vulnerability but none contains a well-established representation of adaptation processes in the global context.

It is not possible to draw a simple summary from the diverse set of studies reviewed in this section. The following conclusions from literature since the TAR, however, are more robust.

• A growing literature considers response strategies that aim at preventing damage to particular key elements and processes in geophysical, biological and socio-economic systems that are sensitive to climate change and have limited adaptation potential; policy-makers may want to consider insights from the literature reviewed here in helping them to design policies to prevent DAI.
• In a majority of the literature, key impacts are associated with long-term increases in equilibrium global mean surface temperature above the pre-industrial equilibrium or an increase above 1990-2000 levels. Transient temperature changes are more instructive for the analyses of key vulnerabilities, but the literature is sparse on transient assessments relative to equilibrium analyses. Many studies provide global mean temperature thresholds that would lead sooner or later to a specific key impact, i.e., to disruption/shutdown of a vulnerable process. Such thresholds are not known precisely, and are characterised in the literature by a range of values (or occasionally by probability functions). Assessments of whether emissions pathways/GHG concentration profiles exceed given temperature thresholds are characterised by significant uncertainty. Therefore, deterministic studies alone cannot provide sufficient information for a full analysis of response strategies, and probabilistic approaches should be considered. Risk analyses given in some recent studies suggest that there is no longer high confidence that certain large-scale events (e.g., deglaciation of major ice sheets) can be avoided, given historical climate change and the inertia of the climate system (Wigley, 2004, 2006; Rahmstorf and Zickfeld, 2005). Similar conclusions could also be applied to risks for social systems, though the literature often suggests that any thresholds for these are at least as uncertain.
• Meehl et al., 2007 Table 10.8 provide likely ranges of equilibrium global mean surface temperature increase for different CO₂-equivalent stabilisation levels, based on their expert assessment that equilibrium climate sensitivity is likely to lie in the range 2-4.5°C (Meehl et al., 2007 Executive Summary). They present the following likely ranges (which have been converted from temperature increase above pre-industrial to equilibrium temperature increase above 1990-2000 levels – see Box 19.2); 350 ppm CO₂-equivalent: 0-0.8°C above 1990-2000 levels; 450 ppm CO₂-equivalent: 0.8-2.5°C above 1990-2000 levels; 550 ppm CO₂-equivalent: 1.3-3.8°C above 1990-2000 levels; 650 ppm CO₂-equivalent: 1.8-4.9°C above 1990-2000 levels; 750 ppm CO₂-equivalent: 2.2-5.8°C above 1990-2000 levels. Some studies suggest that climate sensitivities larger than this likely range (which would suggest greater warming) cannot be ruled out (Meehl et al., 2007 Section 10.7.2), and the WGI range implies a 5-17% chance that climate sensitivity falls above 4.5°C (see Key caveat in Section 19.4.2.2 for further information).
• While future global mean temperature trajectories associated with different emissions pathways are not projected to diverge considerably in the next two to four decades, the literature shows that mitigation activities involving near-term emissions reductions will have a significant effect on concentration and temperature profiles over the next century. Later initiation of stabilisation efforts has been shown to require higher rates of reduction if they are to reduce the likelihood of crossings levels of DAI (Semenov, 2004a,b; Izrael and Semenov, 2005, 2006). Substantial delay (several decades or more) in mitigation activities makes achievement of the lower range of stabilisation targets (e.g., 500 ppm CO₂-equivalent and lower) infeasible, except via overshoot scenarios (see Figure 19.2, bottom panel). Overshoot scenarios induce higher transient temperature increases, increasing the probability of temporary or permanent exceedence of thresholds for key vulnerabilities (Hammitt, 1999; Harvey, 2004; O’Neill and Oppenheimer, 2004; Hare and Meinshausen, 2005; Knutti et al., 2005; Schneider and Mastrandrea, 2005).
• There is considerable potential for adaptation to climate change for market and social systems, but the costs and institutional capacities to adapt are insufficiently known and appear to be unequally distributed across world regions. For biological and geophysical systems, the adaptation potential is much lower. Therefore, some key impacts will be unavoidable without mitigation.