1. EPD and climate (Biome 6000 - PMIP) (Joel Guiot)

There is a demand from the climate community for past vegetation and climate information. In this respect, the Biome 6000 project which started in 1993 and made an important step this year, has utilized most of the continental pollen databases and has produced vegetation maps (at the biome level) for the 0, 6 and 21 ka BP periods. Other periods will certainly be reconstructed in the next years. These maps are extensively used for tha data-model comparaison in the frame of PMIP (Paleoclimate Modelling Intercomparison Project). From the EPD point of view, the projects have advantages and weaknesses. The main advantage is that Biome 6000 has developed an objective method to assign biomes to pollen spectra. This method has been genralized to all the continents, but the subsequent weakness is that the biome unit is too broad a one for a more local study. Starting from the same basic ideas, however, the method can be refined for the pollen-analyst's needs. In the frame of PMIP, paleoclimate maps have also been produced which question classical ideas of refuges. The reconstructed climate variables may be somewhat biased or even not compatible when using different statistical methods or additional proxies. That comparisons lead to fruitful discussions which can be tested, by climate and vegetation models which become more and more precise in resolution. So a continuous exchange of ideas within the framework of the EPD can lead not only to improve climate models but also to a more mechanistic understanding of the pollen diagrams themselves.

2. Oak Genetic Project and the EPD (Simon Brewer)

The project to construct and compare isopollen maps with the distribution of Quercus chloroplast DNA cytotypes at the European scale, moved into its second year. At the EPD, the addition of more data, in particular from the Mediterranean area, and the increase in the number of entities with chronologies based on radiocarbon dates has allowed the production of a series of preliminary maps showing the distribution of oak pollen percentages over the time period 13.5 to 8 ka BP. The maps, at 500 year timeslices, were presented to the EPD AB and EC. Whilst some problems exist with the maps, several points concerning the distribution of Quercus during this period could be clearly seen: notably, the importance of four regions in providing refugia for the taxa: Greece, Italy, Spain and the Balkans close to the Black sea, and the postglacial spread of the Quercus into northern Europe, from 13.5 ka BP onwards. However, some sites in the North of Europe showed the presence of Quercus at a time in contradiction with that known from regional experts suggesting chronological probable errors. A few sites 'disappeared' between maps, indicating a problem with the interpolation method used, linear interpolation between samples will be used in future to eliminate this problem Finally, these first maps were constructed using a single Quercus taxa, with no species differentiation. These problems are being tackled in the forthcoming year. The comparison between the genetic and pollen data sets should allow a refinement of our knowledge of postglacial migration of oaks, owing to the high spatial resolution of the geneticists study, and the genticists will benefit from having historical data to compare with their study. The project now has a successor in the CYTOFOR project which aims to provide similar genetic maps for a further twenty species. The EPD is involved in this project, with the goal of producing isopollen maps for those species recorded in the sedimentary sequence (Acer, Alnus, Betula, Calluna, Carpinus, Corylus, Fagus, Fraxinus, Ilex, Salix, Tilia and Ulmus).