Insolation change from 132 to 112 ka BP



This time-slice comprises the last interglacial period and the transition to the following glacial. Because of the substantial changes in insolation and global climate, this period is ideal to test the model's ability to correctly simulate the feedbacks relevant for the transition from the last interglacial towards the glacial. Additionally, the Earth's orbital changes today are similar to what they were at the last interglacial (Kukla and Gavin, 1992).
For the study of feedbacks between the climate and the land biosphere the experiments were carried out with the ECHAM/LSG and the coupled ECHAM/LSG/LPJ model. In response to the modified insolation a rapid warming of the Northern Hemisphere is registered. The figure below shows a strong warming of the near-surface air temperature averaged over the first 2000 simulation years.

Click on the figure to proceed stepwise to the end of the simulation: The clip shows the differences between the near-surface temperature of the transient (132 - 112k) experiment and the mean state of the control run. Both the simulations were performend with the coupled ECHAM/LSG/LPJ model. Shown are 2000 year means.

In advance of the simulation the warm anomaly reaches also the Southern Hemisphere. Only over northern Africa and India the temperatures become cooler, wich is related to stronger cloud formation and precipitation over these areas (see below). The following cooling period is the response to the astronomical forcing. From about 18 ka onward this leads to a signficant cooling of near-surface temperatures on the northern Hemisphere.



Regional changes


The timeseries shown below display the climatic changes resulting from the varying astronomical forcing in selected regions of the Earth.

Locally, the insolation-changes lead to a significant increase of yearly mean precipitation (upper plot) in the sahara desert (compared to the two control runs with present-day boundary conditions). This effect is more pronounced in the coupled ECHAML/LSG/LPJ model including the dynamical vegetation (black curves). The reason for this stronger response is the greening of the sahara desert which strenghtens the regional ocean-land albedo contrast (thereby increasing the monsoon rain) in the ECHAML/LSG/LP model.

The middle plot shows the evolution of near-surface temperatures during the summer season in North America (48°-65°N, 65-126°W. Both model versions show only minor differences during the warmer climate up to 122.000 years. Significant differences are developed during the following colder climate from about 122 ka onward. The stronger cooling simulated with ECHAM/LSG/LPJ model is explained with a stronger summer albedo feedback in this model (lower plot). Here, the albedo increases considerably stronger, when with proceeding cooling darker forest areas are replaced by lighter areas covered by grass and snow.


Selected time-frames
The following figures refer to the mean climate state between 126 - 124k and 116 - 114k (see the dashed boxes in the figure above)


The last interglacial - 125k



During this period a warmer climate on the Northern Hemisphere has been inferred from many geological climate archives. The eccentricity and obliquity of the Earths' orbit were slightly higher compared to today. This resulted in a stronger (weaker) insolation during the northern hemisphere summer (winter) and in an amplification (weakening) of the seasonal cycle on the Northern (Southern) Hemisphere. Considering the yearly mean, the changes in insolation almost compensate for each other (see figure below).

The global mean temperature (not shown) is about 0.5°C higher compared to the control run. The zonal averaged yearl mean temperatures (blue curve)show a significant warming north of about 20°N.


The 2-dimensional plot shows a strong warming of the Northern Hemisphere and on the southern continents as well during the north summer (upper plot below). Maximum anomalies are restricted to the contents. Only over Central Africa and southern parts of Asia a regional cooling of 2 - 3°C is registered.

During the north winter (lower plot) a slight cooling is seen south of about 30°N. The relative strong warming in the highest northern latitudes is mainly related to a significant reduction in sea-ice thickness.
Because of the different heat capacities, temperature anomalies over the continents are by far greater (with maximun deviations of more than 6°C) than those observed in the sea-surface temperatures (see below). Here, the yearly mean temperatures show an overall warming north of 30°N. In the North Atlantic the warming is slightly lower compared to the North Pacific, because a larger part of warmer surface waters is transported to the deep ocean by convection. In the deeper layers of the Atlantic (not shown) a warming of 0.25 - 0.5°C is observed below 2500 m.

The temperature anomalies considerably influence the global precipitation pattern. The figure below exhibits strong increases in the north summer mean precipitation on the Northern Hemisphere and over the southern continents as well. The most pronounced anomalies occur over North Africa and over India which evidences an increase in the north summer monsoon rain.

About 70% of the yearly precipitation falls from July to September.
Also here, it is obvious that the climate response is much stronger simulated by the coupled ECHAM/LSG/LPJ model than with the fixed vegetation in the ECHAM/LSG model. This is especially true for the Sahara desert, where the establishment of a vegetation cover causes strong albedo feedbacks. In the ECHAM/LSG/LPJ model the precipitation during August and September is twice as much than in the ECHAM/LSG model (see figure below).


The transition to the glacial - 115k



Here, the modified orbital parameters cause a weakening of seasonal contrasts compared to the 125k experiment. Generally, the same pattern of climatic changes are registered as in the 125k experiment, but with reversed premises.

Contrary to the 125k experiment, a distinct cooling of near-surface air temperatures is observed during the north summer on the Northern Hemisphere and on the southern continents as well (upper plot). Only over Central Arica a positive temperature anomaly is found. The relative strong cooling in the northeastern Pacific during the north winter is related to a large increase in sea-ice thickness. Also here, the temperature anomalies over land are significantly weaker than those found over land. In agreement with this, anomalies observed in the sea surface temperature field are much waeker than in the near-surface air temperature. Considering here only the yearly mean anomalies, the large-scale warming of the Southern Ocean varies between 0.5-1.0°C. Likewise, the North Atlantic cools only locally by more than 0.5 °C. Only the northern North Pacific undergoes a substantial cooling by more than 2°C. This is due to an increased entrainment of deep water from the Antarctic Circumpolar Current into the deep Pacific which causes an enhanced upwelling of cooler deep waters in the North Pacific region.

The transition toward a colder climate in the subsequent glacial stage is displayed also by a substantial increase in the northern hemisphere sea-ice thickness. This increase is strongest in the Arctic region close to the adjacent North Pacific, where a maximum increase by more than 10 m is observed. Also in the North Pacifc more ice accumulates - and the seawater - ice boundary moves southward (not shown). In the high latitudes of the North Atlantic the observed changes are less extensive. Here, a persistent strong overturning and stable deep convection maintain a strong northward heat transport, which counteracts the effect of astronomical forcing. Only locally, in Ross Sea and around Antarctica are found decreases of sea-ices thickness, which is related to an enhanced deep convection in these regions.

The temperature anomalies cause strong changes in the regional precipitation patterns. Due to colder summers, less precipation is found primarily over the continents. The north summer moonsoon rain weakens. Most pronounced anomalies occur over Middle Africa and southern parts of Asia - a pattern which is similar to that found in the 125k experiment. The warmer north winters lead to generally higher precipitation over the continents.

Another important climate factors that take influence on the Earths' radiation budget are (besides the variations of solar radiation on the top of the atmosphere) internal feedbacks of the climate system. The figure below shows the mean land-albedo of the Northern Hempisphere between 60° and 90°N. Critical for the growth of continental ice sheets are the melting periods during summer. Here, both the model versions simulate a higher albedo compared to the control run (blue curves). This indicates the presence of more expanded areas covered by lighter snow, which in turn amplifies the astronomical signal and favors a colder climate. The reverse effect is seen in the 125k experiment (red curves). Only minor deviations can be seen between the two control runs (black curves). The climate feedbacks, however, seen in both, the 115k and the 125k experiment are by far stronger in the ECHAM/LSG/LPJ model compared to the ECHAM/LSG model.

In summary, the above described experiments demonstrate important climate feedbacks provided by the dynamical biosphere.