Simple climate models are widely used in most climate mitigation pathways studies, including AVOID and the Committee on Climate Change’s carbon budget analysis, but there are questions regarding their reliability and how key assumptions influence the results for aggressive mitigation scenarios. The aim of this report is to examine the uncertainty resulting from key assumptions in the climate model used to produce climate projections within AVOID with the view to increase the robustness of future projections. Improved estimates of uncertainty in the climate response of emissions pathways is critical for the AVOID top level question “2.What emissions pathways will avoid “dangerous” climate change?”
The simple climate model, MAGICC, was used in AVOID to estimate the climate response to a large number of emissions pathways to establish whether a particular pathway remained below the policy relevant 2°C global temperature target. These climate projections of warming have associated uncertainties arising from our incomplete knowledge of the climate system but the uncertainties can be captured by varying the adjustable parameters within the model over the uncertainty range. The warming estimates provided in policy relevant work need to have an accurate representation of the uncertainties involved. Following a detailed analysis of the uncertainties in the model parameters, climate sensitivity, ocean diffusivity and aerosol radiative forcing and emissions, the consequences on how best to represent uncertainty in warming estimates in AVOID includes:
- Climate sensitivity (T2x) gives an estimate of how much the climate will warm to a doubling of atmospheric carbon dioxide concentrations. The estimate of climate sensitivity used in the previous AVOID emissions pathways studies (AV/WS1/D1/01 03) was that of Murphy et al., (2004) which provides a high central estimate of warming in simple climate models compared to most other T2x estimates. We also note that the Murphy estimate gives results that agree well with the central estimate of attributed warming. However, given the significant spread in the central estimates of T2x in the published literature and with no clear “best estimate” of T2x, it is recommended that the warming for a “high” and “low” central estimate of T2x should be provided to convey the uncertainty in T2x.
- Atmospheric aerosols are small particles that either cool or warm the planet. On a global scale the cooling effect from sulphate aerosols tends to outweigh warming from other aerosol species, such as black carbon aerosols, and so can offset some of the potential warming from greenhouse gases. How aerosol emissions will change in the future is another large source of uncertainty in the projected warming and the aerosol emissions in the AVOID emissions pathways uses one set of assumptions. Most alternate aerosol emissions pathways lie below those used in the previous AVOID work and because it is not clear which particular pathway is more representative, providing alternate estimates of aerosol pathways in future AVOID work is recommended. We also recommend further consideration of the uncertainty in how effective aerosol particles are in modifying the climate.
- Ocean diffusivity is a measure of the speed with which heat is transferred from the ocean surface to the deep ocean. The estimate of the ocean diffusivity, k, used in previous AVOID MAGICC climate projections was based on MAGICC tuned to the complex atmosphere–ocean global climate models (AOGCMs) in the IPCC Third Assessment Report (2001). The equivalent estimate for the IPCC Fourth Assessment Report (2007) MAGICC tuned values has a lower central k value and in light of more evidence that AOGCMs have more efficient mixing than observed (Hansen et al., 2011 Knutti 2008, Forest et al., 2006), we recommend that AVOID adopt the estimate with the lower central k value for future studies.
- iv) The findings of this study suggest that the warming projections in previous studies such as the Committee of Climate Change December 2008 report (Smith et al., 2008) and the early AVOID work on emissions pathways (AV/WS1/D1/01-03) would be slightly higher if the recommendations in this report are adopted but would remain within the uncertainty range of the new estimate.