Aim Understanding the driving forces and mechanisms of changes in past plant distribution and abundance will help assess the biological consequences of future climate change scenarios. The aim of this paper is to investigate whether modelled patterns of climate parameters 6000 years ago can account for the European distribution of Fagus sylvatica at that time. Consideration is also given to the role of non-climatic parameters as driving forces of the Holocene spread and population expansion of F. sylvatica.
Methods European distributions were simulated using a physiologically-based bioclimatic model (STASH) driven by three different atmospheric general circulation model (AGCM) outputs for 6000 years ago.
Results The three simulations generally showed F. sylvatica to have potentially been as widespread 6000 years ago as it is today, which gives a profound mismatch with pollen-based reconstructions of the F. sylvatica distribution at that time. The results indicate that drier conditions during the growing season 6000 years ago could have caused a restriction of the range in the south. Poorer growth conditions with consequently reduced competitive ability were modelled for large parts of France.
Main conclusions Consideration of the entire European range of F. sylvatica showed that no single driving force could account for the observed distributional limits 6000 years ago, or the pattern of spread during the Holocene. Climatic factors, particularly drought during the growing season, are the likely major determinants of the potential range. Climatic factors are regionally moderated by competition, disturbance effects and the intrinsically slow rate of population increase of F. sylvatica. Dynamic vegetation modelling is needed to account for potentially important competitive interactions and their relationship with changing climate. We identify uncertainties in the climate and pollen data, as well as the bioclimatic model, which suggest that the current study does not identify whether or not climate determined the distribution of F. sylvatica 6000 years ago. Pollen data are better suited for comparison with relative abundance gradients rather than absolute distributional limits. These uncertainties from a study of the past, where we have information about plant distribution and abundance, argue for extreme caution in making forecasts for the future using equilibrium models.