Insolation forced Equilibrium climate simulations at 125k and 115k B.Ps

1. Vegetation modelling and climate-vegetation interactions


The influence of vegetation changes on the climate system will be illustrated using a set of five experiments that have been performed with the coupled atmosphere-ocean-vegetation model. The set consists of a control run for present-day climate, two runs for 125 ky B.P. and 115 ky B.P. with vegetation feedbacks included and two runs for 125 ky B.P. and 115 ky B.P. without vegetation feedbacks (the fixed vegetation that is prescribed correspond to the mean state of the control run). For the last two runs, the vegetation model has been used to model the vegetation corresponding to the climate of the atmosphere-ocean simulation, but the vegetation is not used to update the surface conditions of the atmosphere model. Results from the experiments on the occurrence of ecosystems and the storage of carbon are shown below. The vegetation feedback mechanisms are illustrated for the Sahara and the northern high latitudes.


Ecosystem distributions

With a simple scheme, the modelled plant functional types from LPJ are converted into the ecosystems mentioned above. The distribution of ecosystems is shown in the figure below.

Ecosytem distribution for the control run and for 125 ky B.P. and 115 ky B.P., both with and without vegetation feedbacks.

Clear shifts can be seen for the regions with positive feedbacks, i.e. Sahara desert and northern high latitudes. In general, changes are smaller for the experiments without feedback from the vegetation changes to the climate.

Terrestrial carbon storage

Changes in ecosystems, as well as temperature changes, have considerable effects on the stocks of carbon. In general, at 125 ky B.P. total terrestrial carbon stocks are predicted to be higer than present, whereas at 115 ky B.P. stocks are predicted to be lower (see the figure below).

Total terrestrial carbon storage for the 125 ky B.P. and 115 ky B.P. experiments, compared to the control run. Contributions of soil carbon changes (orange) and vegetation carbon changes (green) are distinguished.

Both soil carbon and vegetation carbon play an important role in this. The largest changes in carbon storage occur at the high latitudes, where low temperatures cause a slow carbon decomposition in soils and vegetation, resulting in large stocks (see figure below).

Difference in carbon storage for 125 ky B.P. (left) and 115 ky B.P. (right) compared to the control run, shown for vegetation (top) and soil and litter (bottom) carbon. For both 115 and 125 ky B.P., the experiments with feedback are used.


The albedo is the driving factor behind the two positive feedbacks in the desert regions and in the high latitudes, as has been described above. For the high latitudes, the albedo is lower for the 125 ky B.P. experiments than for the control run and for 115 ky B.P. This results in higher temperatures throughout the year. The temperature effect of an interactive vegetation is largest in spring and autumn:

Albedo (top) and resulting temperature (bottom) for North America, 124-96E, 58-69N. The annual cycle is shown for the control run, 125 ky B.P. with and without vegetation feedback, and 115 ky B.P. with and without vegetation feedback.
The albedo of the Sahara region changes from roughly 0.38 in the control run to 0.16 in the coupled 125 ky B.P. experiment. This results in an enhancement of the precipitation increase in this region:

Precipitation for the Sahara, 6°E-28W, 19-30N. The annual cycle is shown for the control run and for 125 ky B.P. with and without vegetation feedback.


An interactive vegetation model is a valuable tool for modelling the effect of changes in vegetation on paleoclimate. Largest changes occur in regions where positive feedbacks can enhance the effect of changes in solar insolation, e.g. the Sahara region or the boreal forests at high latitudes. Besides the biophysical coupling with surface parameters that has been performed here, biogeochemical feedbacks will have to be taken into account as well. A coupling with the ocean biogeochemistry model HAMOCC in our earth system model will enable modelling CO2 concentrations.

2. Climate Change

The figure below shows the climate response to insolation forcing by means of temperature and precipitation in the light of the 125k experiments. Both model versions show strong warming together with positive precipitation anomalies in the northern hemisphere with maximum response in each case over the continents (fig. 1). The stronger warming seen in the model with interactive vegetation (especially in the boreal regions of North America and Asia) results from a northward expansion of the vegetation which reduces the local albedo thereby amplifying the astronomical forcing. Southeast off the Mediterranean the density of vegetation coverage increases in the model with interactive vegetation (upper panel). Both models simulate significant cooling in the subtropics caused by intensified summer monsoon and cloud formation. Again, this negative feedback is stronger and spatially more extended in the model with interactive vegetation. In both model versions, intensified insolation during summer leads to higher precipitation over most of the continents (see figure, right). The maximum response is registered in the tropics and in the african-asien monsoon belt due to higher land-sea temperature contrasts. This is accompanied by intensified cloud formation leading to the cooling in the northern hemisphere monsoon belt. In the Sahara desert vegetation is established when interactive vegetation allowed. The replacement of lighter desert surface by darker vegetation reduces the local albedo considerably which again amplifies the land-sea temperature contrast leading to stronger summer monsoon. The resulting change in monsoonal precipitation exceeds the response simulated by the model with non-dynamic vegetation by more than the two-fold in this area. The anomal precipitation/temperature pattern obtained for 115 ka (not shown) is similar but with reversed premises compared to the 125 ka simulation. Furthermore the vegetation/albedo feedback mechanisms that were already identified in the 125k experiment are recognized at 115k resulting in a stronger cooling in the high latitudes and lower precipitation in the subtropics. Again, the response is generally stronger when coupling the vegetation interactively.