Articles | Volume 8, issue 5
Research article
29 Oct 2012
Research article |  | 29 Oct 2012

A model–data comparison for a multi-model ensemble of early Eocene atmosphere–ocean simulations: EoMIP

D. J. Lunt, T. Dunkley Jones, M. Heinemann, M. Huber, A. LeGrande, A. Winguth, C. Loptson, J. Marotzke, C. D. Roberts, J. Tindall, P. Valdes, and C. Winguth

Abstract. The early Eocene (~55 to 50 Ma) is a time period which has been explored in a large number of modelling and data studies. Here, using an ensemble of previously published model results, making up "EoMIP" – the Eocene Modelling Intercomparison Project – and syntheses of early Eocene terrestrial and sea surface temperature data, we present a self-consistent inter-model and model–data comparison. This shows that the previous modelling studies exhibit a very wide inter-model variability, but that at high CO2, there is good agreement between models and data for this period, particularly if possible seasonal biases in some of the proxies are considered. An energy balance analysis explores the reasons for the differences between the model results, and suggests that differences in surface albedo feedbacks, water vapour and lapse rate feedbacks, and prescribed aerosol loading are the dominant cause for the different results seen in the models, rather than inconsistencies in other prescribed boundary conditions or differences in cloud feedbacks. The CO2 level which would give optimal early Eocene model–data agreement, based on those models which have carried out simulations with more than one CO2 level, is in the range of 2500 ppmv to 6500 ppmv. Given the spread of model results, tighter bounds on proxy estimates of atmospheric CO2 and temperature during this time period will allow a quantitative assessment of the skill of the models at simulating warm climates. If it is the case that a model which gives a good simulation of the Eocene will also give a good simulation of the future, then such an assessment could be used to produce metrics for weighting future climate predictions.