The role of CO2, sea level, and vegetation during the Milankovitch-forced glacial-interglacial cycles

Sensitivity experiments have been made over the last glacial-interglacial climatic cycle using the Louvain-la-Neuve two-dimensional Northern and Southern Hemispheres climate model. The continental ice volume was simulated for the past 122 kyr in response to changes in both the insolation and the CO2 atmospheric concentration. The sensitivity of such a response to sea level changes and to the vegetation-snow albedo feedback indicates that the 100-kyr cycle cannot be sustained if these processes are not taken into account. The adoption of the factor separation method by Stein and Alpert allows the identification of the contribution of the processes involved in a climate model as well as their synergistic effects. Here, this technique was restricted to two variables the sea level and the albedo of vegetation when covered by snow - to quantify their individual impacts and mutual contributions to the global ice volume variations over the last glacial-interglacial cycle. The simulated sea level drop of about 100 m at the Last Glacial Maximum leads to an increase of emerged continental surfaces over the present-day value by about 13%. As a consequence, the growth of the ice sheets up to the shoreline leads to an ice volume change 20% larger than if the sea level would have been kept constant. On the other hand, if the vegetation-snow albedo feedback is not included in the simulation, the Northern Hemisphere ice volume is overestimated most of the time, especially in the experiment where the sea level is allowed to vary. Over the past 10 kyr, the difference between the control case (all processes are included) and the reference case (sea level fixed and no vegetation effect) is explained mainly by the vegetation-snow albedo feedback, the pure positive contribution of sea level being almost canceled by the negative contribution of the synergism between the sea level and vegetation impacts.