Regional and Global Temperature Projections using the Fractional Energy Balance Equation

Abstract:

We present climate projections based on a new class of atmospheric model, the Fractional Energy Balance Equation (FEBE), derived by updating Budyko-Sellers models. The FEBE has two distinct power law regimes with a transition at the relaxation time (of the order of a few years): its long memory governs responses to both internal and external sources of variability.

We model and project the multidecadal responses to external CMIP5, CMIP6 and aerosol forcing series. We use a Bayesian parameter estimation framework based on historical temperatures and natural and anthropogenic forcing series. The high frequency FEBE regime not only allows for modelling responses to volcanic forcings but also the response to internal white noise forcings: a theoretically motivated error model. The low frequency part uses historical data and long memory for climate projections, constraining both equilibrium climate sensitivity and historical aerosol forcings.

We present climate sensitivity estimates and projections using both the Representative Carbon Pathways and Shared Socioeconomic Pathways scenarios shown alongside the CMIP5/6 MME, at both the global and regional (5°x5°) scale.

For almost all scenarios, our 90% confidence interval lies within the corresponding GCM limits so that our projections give them full support while significantly reducing the uncertainties.