PALEOCLIMATE MODELING | Data–Model Comparisons

Data–model comparison is the shorthand term used to describe the process of comparing output from climate or Earth-system model simulations of past time periods with paleoenvironmental observations. The article describes the iterative process whereby observations of past environmental conditions, or reconstructions of past climate states based on these observations, are used together with physically based models of major components of the Earth system to arrive at a fundamental understanding of the mechanisms of past climate and environmental changes. It also describes the more pragmatic use of environmental data to evaluate the performance of climate models. The global data sets that are currently available for such analyses are described and examples of systematic evaluation of model simulations made by the Paleoclimate Modeling Intercomparison Project (PMIP) are used to illustrate the general approach to data–model comparisons.