Wednesday, 12 January 2000: 2:29 PM
For two reasons it is important to study the sensitivity of the global climate towards changes in the vegetation cover over land. In the real world, regional and global implications may arise from changes in the vegetation cover. Second, in numerical simulations, the degree of the sensitivity of the simulated climate may depend on the specific parameterization schemes employed in the model and also on the large-scale systematic errors of the GCM. These aspects have been studied with the Max-Planck-Institutes climate model. The model version ECHAM4 was used to identify the sensitivity of the local and global climate during a full annual cycle to desertification and afforestation in the Mediterranean. The deforestation scenario represents an extreme desert which would be the worst case scenario for this region. The changes in the afforestation experiment are spread over a relatively larger area according to the pattern of the vegetation cover 2000 years before present and also include points at mid-latitudes. The comparison of the two experimental simulations to the control run shows a slight cooling at the surface and reduced precipitation after deforestation as a result of less evapotranspiration of plants and less bare soil evaporation from the assumption of eroded soils. There is no significant signal during the winter season due to the higher influence of baroclinic disturbances. In general the results of the afforestation experiment are of the opposite sign. A significant response was found in the vicinity of gridpoints, where the land surface characteristics had been changed, but there are also downstream amplifications of the local disturbances. Such effects were found in the Sahara in the afforestation experiment and agree with the response of GCM experiments by Reale and Shukla (1998).
![](images/flag.png)
![](images/blue_ribbon.png)