Coupled modeling of global carbon cycle and climate in the Neoproterozoic: links between Rodinia breakup and major glaciations

A coupled climate–geochemical model of new generation (GEOCLIM) is used to investigate the possible causes of the initiation of snowball glaciations during Neoproterozoic times. This model allows the calculation of the partial pressure of atmospheric CO₂ simultaneously with the climate at the continental surface with a rough 2D spatial resolution (10° lat. × 50° long.). We calculate that the breakup of the Rodinia supercontinent, starting 800 Myr ago, results in a global climatic cooling of about 8 °C triggered by enhanced consumption of atmospheric CO₂ resulting from increased runoff over continental surfaces. This increase in runoff is driven by the opening of oceanic basins resulting in an increase of soil moisture sources close to continental masses. This climatic effect of the supercontinent breakup is particularly strong within the 800–700 Ma interval since all continents are located in the equatorial area, where temperature and runoff conditions optimize the consumption of CO₂ through weathering processes. However, this effect alone is insufficient to trigger snowball. We propose that the efficient weathering of fresh basaltic surfaces that erupted during the Rodinia breakup, and were transported to the humid equatorial area through continental plate motion, contributed the necessary CO₂ sink that triggered the ca. 730-Ma Sturtian glacial event. Simulations of the GEOCLIM model for the ca 580-Ma Gaskiers ice age, where all continents are centered on the South Pole, shows that no snowball glaciation can be initiated. The calculated CO₂ partial pressure remains above 1000 ppmv, while a threshold of less than 80 ppmv is required to initiate a snowball glaciation. At that time, a polar configuration does not allow the onset of total glaciation. Nevertheless, a regional glaciation is simulated by the GEOCLIM when the climatic and geochemical (i.e. weathering related) effects of the Pan-African orogeny (～600 Ma) are taken into account. Finally, the question of the role of the paleogeographic setting in the Marinoan snowball event (～635 Ma) is still an open question, since no reliable Marinoan paleogeographic reconstruction exists due to the paucity of paleomagnetic data.