Abstract
The mid-Holocene `green' Sahara represents the largest anomaly of the atmosphere-biosphere system during the last 12 000 years. Although this anomaly is attributed to precessional forcing leading to a strong enhancement of the African monsoon, no climate model so far has been able to simulate the full extent of vegetation in the Sahara region 6000 years ago. Here two atmospheric general circulation models (LMD 5.3 and ECHAM 3) are asynchronously coupled to an equilibrium biogeography model to give steady-state simulations of climate and vegetation 6000 years ago, including biogeophysical feedback. The two model results are surprisingly different, and neither is fully realistic. ECHAM shows a large northward extension of vegetation in the western part of the Sahara only. LMD shows a much smaller and more zonal vegetation shift. These results are unaffected by the choice of `green' or modern initial conditions. The inability of LMD to sustain a `green' Sahara 6000 years ago is linked to the simulated strength of the tropical summer circulation. During the northern summer monsoon season, the meridional gradient of sea-level pressure and subsidence over the western part of northern Africa are both much weaker in ECHAM than in LMD in the present as well as the mid-Holocene. These features allow the surface moist air flux to penetrate further into northern Africa in ECHAM than in LMD. This comparison illustrates the importance of correct simulation of atmospheric circulation features for the sensitivity of climate models to changes in radiative forcing, particularly for regional climates where atmospheric changes are amplified by biosphere-atmosphere feedbacks.
Similar content being viewed by others
Author information
Authors and Affiliations
Additional information
Received: 20 April 1999 / Accepted: 20 January 2000
Rights and permissions
About this article
Cite this article
de Noblet-Ducoudré, N., Claussen, M. & Prentice, C. Mid-Holocene greening of the Sahara: first results of the GAIM 6000 year BP Experiment with two asynchronously coupled atmosphere/biome models. Climate Dynamics 16, 643–659 (2000). https://doi.org/10.1007/s003820000074
Issue Date:
DOI: https://doi.org/10.1007/s003820000074