Climate of the Past Discussions www.clim-past-discuss.net 1814-9340 1814-9359 1 3 2005 10.5194/cpd-1-255-2005 http://www.clim-past-discuss.net/1/255/2005/ http://www.clim-past-discuss.net/1/255/2005/cpd-1-255-2005.html http://www.clim-past-discuss.net/1/255/2005/cpd-1-255-2005.pdf 255 285 2005-12-07 Effect of land albedo, CO₂, orography, and oceanic heat transport on extreme climates V. Romanova G. Lohmann K. Grosfeld Department of Physics, University of Bremen, Otto-Hahn-Allee, 33040 Bremen, Germany Alfred Wegener Institute for Polar and Marine Research, 27515 Bremerhaven, Germany Using an atmospheric general circulation model of intermediate complexity coupled to a sea ice-slab ocean model, we perform a number of sensitivity experiments under present-day orbital conditions and geographical distribution to assess the possibility that land albedo, atmospheric CO₂, orography and oceanic heat transport may cause an ice-covered Earth. Changing only one boundary or initial condition, the model produces solutions with at least some ice-free oceans in the low latitudes. Using some combination of these forcing parameters, a full Earth's glaciation is obtained. We find that the most significant factor leading to an ice-covered Earth is the high land albedo in combination with initial temperatures set equal to the freezing point. Oceanic heat transport and orography play only a minor role for the climate state. Extremely low concentrations of CO₂ also appear to be insufficient to provoke a runaway ice-albedo feedback, but the strong deviations in surface air temperatures in the Northern Hemisphere point to the existence of a strong nonlinearity in the system. Finally, we argue that the initial condition determines whether the system can go into a completely ice covered state, indicating multiple equilibria, a feature known from simple energy balance models.