Anthracology: past disturbances and vegetation evolution

One of CoForChange aims is to highlight the relationship between human settlement and plant distribution. Recent studies show that anthropogenic disturbances have had an impact on vegetation, including enabling the expansion of light-demanding species (see Newsl. No 4). To contribute to this analysis in a context of deep scarcity of anthracological data in Africa, we have been studying the charcoals from nine soil pits of the CoForChange project as part of a Master’s. The objective was to understand past and present species composition of the forests in relation with past disturbances. Two anthracological protocols of identification were tested: 1) at species level, by describing the anatomical characteristics of charcoals, by carrying investigations with InsideWood online database, and by comparing anatomical data at XylariumTervuren; 2) at community level, by searching for statistical relationships between ecological traits of species and wood anatomy. The results show that the taxonomic diversity is greater in Marantaceae forests than in lowland forests. Three identifications were obtained from 48 described taxa: Gilbertiodendron dewevrei under G. dewevrei monodominant forest (GIB1 at 40 cm depth, 1510 ± 30 years cal. BP), Millettia drastica and Pterocarpus soyauxii inMarantaceae open forests (F9 at 40 cm, 1200 ± 30 cal. BP). Our hypothesis is that monodominant stands of G. dewevrei are relatively stable. With regard to P. soyauxii, the proportion of this lightdemanding species seems to be decreasing over time (at levels from 20 to 40 cm) for the benefit of Marantaceae. Another result highlights the absence of Triplochiton scleroxylon under T. scleroxylon, which may confirm the hypothesis of the recent settlement of this species. Results on the relationship between wood anatomy and functional traits are still preliminary, and further research will be conducted through a PhD programme. The on-going developments of anthracology in tropical Africa will help to identify a larger number of collected samples, and thus to understand better the evolution of tropical forests.