XVI INQUA Congress

Paper No. 5
Presentation Time: 9:30 AM

VEGETATION-OCEAN-ATMOSPHERE CLIMATE INTERACTIONS DURING THE LAST 21,000 YEARS


VALDES, Paul J, Meteorology, Univ of Reading, Earley Gate, Whiteknights, Reading, RG6 6BB, United Kingdom, BREWER, Simon, CEREGE Europole, Mediterraneen de l’Arbois, BP 80, Aix-en-Provence, F-13545, France and DAVIS, Basil, Geography, Newcastle Univ, Newcaste upon Tyne, NE1 7RU, United Kingdom, P.J.Valdes@reading.ac.uk

We use a set of "snapshot" climate model simulations to investigate climate change during the last glacial interglacial cycle. The model simulations use the Hadley centre atmospheric model (HadAM3) coupled to a simple slab ocean model with prescribed (and unchanging) ocean heat flux convergence. This ocean model is unable to represent changes in the ocean circulation and thus part of any differences between model and data can be interpreted as indicating the importance of ocean circulation changes. The model is forced by changes in carbon dioxide, methane, ice sheets, and orbital parameters, and simulations are performed for every 1000 years from the LGM to present day. An additional set of simulations coupled the climate model to a biogeographical model (BIOME4) so that the role of vegetation changes can be quantified.

Broadly, the set of model simulations compare favourably to data, except during the glacial interglacial transition (i.e. from approximately 15 - 10 kyr BP) where the model is unable to represent the Bolling-Allerod or Younger Dryas. This suggests that the ocean circulation changes are important for the onset of the Bolling-Allerod as well as the Younger-Dryas.

During the Holocene, vegetation feedback can be very important. The model results are compared to new data compiled from several databases of pollen sites on the European continent. A pft-based modern analogue method has been used to reconstruct various climatic parameters from this data-set for the last 12,000 years at high frequency (up to every 100 years), including mean temperature of the coldest month and annual precipitation. We use this data to evaluate the model simulations for Europe. In general, the model-data comparison suggests that the large-scale patterns are being reproduced by the model but that much regional detail is missing. We will discuss the reasons for these discrepancies.