NEW PREDICTIONS OF MID-PLIOCENE ANTARCTIC CLIMATE AND BIOME DISTRIBUTIONS: APPLICATIONS OF COMBINED CLIMATE-VEGETATION MODELLING TO THE SIRIUS GROUP DEBATE

Significant debate exists surrounding the nature of the Antarctic ice-sheet(s) during the period of middle Pliocene warmth. This debate has, in part, been fuelled by the identification of 'Pliocene' Nothofagus remains with the Sirius Group Strata of the Transantarctic Mountains (Wilson et al., 1998). Results from vegetation modelling, with a mechanical biogeographical model using mean climatological data from a General Circulation Climate Model (GCM) and different Antarctic ice-sheet configurations for the mid-Pliocene, are now becoming available (Haywood et al., 2002; Francis et al., 2002). The predicted vegetation distributions from such modelling exercises provide us with a useful first approximation of an Antarctic biome distribution that is in equilibrium with the imposed Pliocene ice-sheet and GCM predicted climatology. These predictions can be compared to palaeobotanical data (Francis et al., 2002). However, such results may be model dependent, therefore it is good practice to utilise more than one vegetation model. Furthermore, to fully represent the role of climate-vegetation feedbacks on Antarctica during the middle Pliocene, it is desirable for the type of land-cover to be treated as an interactive element within the climate modelling. Recent advances in model design now make such an exercise possible. We present results from new coupled climate-vegetation simulations for the middle Pliocene, focussed on Antarctica, using the UK Meteorological Office's HadAM3 GCM and TRIFFID dynamic vegetation model. TRIFFID is run synchronously with the climate model (both models updating one another with climate and vegetation characteristics at regular intervals) until an equilibrium state is achieved. The new predicted Antarctic climatology and vegetation distributions are compared to those previously obtained in earlier modelling investigations and to available palaeobotanical data sets.