Reconstructing the climate of the Earth through geological times : what can be expected from models, what can be used as proxies ? (Invited)

The climatic and biogeochemical evolution of the Earth surficial envelops are tightly coupled. Processes controlling the atmospheric partial pressure of the greenhouse gases over geological times are, at least partly, dependent on the climate, generally in a complex way. Since about 30 years, a bunch of numerical models have been built to reconstruct the climate of the last 800 million of years of the Earth history. The output of those models (atmospheric CO2 concentration, air temperature, carbonate deposition or isotope ratios of the seawater) are compared with available proxies, both in the continental and oceanic realms. Up to know, all those climate-carbon models were coupling simplistic climate models with a numerical description of the biogeochemical cycles. Owing to constraints on the computation times, the climate models often reduces to simple parametric laws linking the global mean air temperature to the CO2 content of the atmosphere, modulated by the long term evolution of the solar constant. Furthermore they do not include any explicit description of the water cycle, making them poorly reliable in terms of climate reconstruction. In the recent years, more complex numerical models of the climate system, such as general circulation models, have been increasingly used to explore the climate of the geological past. However, although they are able to calculate an oceanic and atmospheric circulation for timeslices in the past, they require numerous boundary conditions, one of them being the atmospheric CO2. Such studies do not account explicitly for the biogeochemical cycles, but they allow for the first time the comparison between spatialized data and model output. In the recent years, a new generation of models has been built. They couple realistic climate models (GCM or EBM) to the biogeochemical cycles at the Earth surface (GEOCLIM, GENIE). Those models open new fields of investigations, in which both the circulation and the chemical composition of the ocean and atmosphere together with the climate can be calculated and compared to available proxies. In this lecture, we will present an overview of the various kind of numerical models described above, and explore their abilities to reproduce the various proxies of the past climate and geochemistry.