Recent disturbances in the Congo basin: an anthracological contribution to vegetation reconstructions

In contrast to the well-known vegetation history of the northern hemisphere, few is known about past vegetation change in Central Africa. However, recent palaeoecological and biogeographical studies suggest that early human disturbances had a substantial influence on Central Africa vegetation patterns, particularly allowing the expansion of light-demanding species. This interesting hypothesis is the basis of one of the main research questions of the ERA-net BIODIVERSA CoForChange project: what was and is the relationship between (increasing) human activity and vegetation change? A recently developed protocol for the identification of ancient Central African charcoal fragments opened the door for vegetation reconstructions with a high spatial and taxonomical resolution. Therefore, we chose to study macro-charcoals from pedoanthracological profiles situated in N Congo and SE Camero0n. In total, 48 taxa were found in nine radiocarbon dated profiles. At the moment, three taxa have been identified down to species level. First, we found numerous fragments of Gilbertiodendron dewevrei under a current monodominant forest of G. dewevrei (40 cm depth, 1421-1327 cal. BP). This seems to confirm the presumption that monodominant forests of G. dewevrei, shade-bearer species, are relatively stable. Second, we evidenced the lack of Triplochiton scleroxylon charcoals under a T. scleroxylon stand, which could confirm the hypothesis of the recent nature of those stands. Finally, our results suggest that taxonomic diversity of charcoal findings in open canopy Marantaceae forests is greater than in dense forests. Pterocarpus soyauxii and Millettia drastica have been found under an open Marantaceae forest at 40 cm depth (1184-1055 cal. BP). The abundance of the light-demanding species P. soyauxii appears to be decreasing over time (levels 20 to 40 cm depth) to the benefit of giant herbs. Anthracology in Central Africa is on the rise and the first results of the CoForChange project are promising. More identifications will follow, resulting in a better understanding of the evolution of Central African forests