Paleogeography: a driver for past climate changes?

CO₂ is the most important greenhouse gas in the Earth’s atmosphere and has fluctuated considerably over geological time. However, proxies for past CO₂ concentrations have large uncertainties and are mostly limited to Devonian and younger times. Consequently, CO₂ modelling plays a key role in reconstructing past climate fluctuations.

Silicate weathering and subsequent carbonate deposition are widely recognized to compose the primary sink of carbon on geological timescales and are largely influenced by changes in climate, which in turn are linked to changes in paleogeography. The role of paleogeography on silicate weathering fluxes has been the focus of several studies in recent years and mostly aiming to constrain climatic parameters such as temperature and precipitation affecting weathering rates through time. However, constraining the availability of exposed land is crucial in assessing the theoretical amount of weathering on geological time scales. Associated with changes in climatic zones, the fluctuation of sea-level is critical for defining the amount of land exposed to weathering. The current reconstructions used inmodels tend to overestimate the amount of exposed land to weathering at periods with high sea levels. Through the construction of continental flooding maps, we have constrained the effective land area undergoing silicate weathering for the past 540 million years. Our maps not only reflect sea-level fluctuations but also contain climate-sensitive indicators such as coal (since the Early Devonian) and evaporites to evaluate climate gradients and potential weatherablity through time. This is particularly important after the Pangea supercontinent formed but also for some time after its break-up.

We here investigate the potential link between land availability dictated by paleogeography and climate changes during the Phanerozoic. Recent studies have shown that continental glaciations occur following periods of decreasing atmospheric CO₂, and these periods correspond to peaks in land availability at tropical latitudes. This link tends to attribute such changes to the paleogeographic evolution of our planet but this is not the case for the enigmatic end-Ordovician cooling where the increase in land availability prior to the Hirnantian glaciation appears not to have been enough to counteract the increase in solar energy and initiate cooling.