Understanding the 2500 yr BP rainforest crisis in West and Central Africa in the framework of the Late Holocene: Pluridisciplinary analysis and multi-archive reconstruction
Résumé
Numerous palaeoclimatic and prehistoric reconstructions have been reported at both the local scale and across the very large area of tropical Central Africa, clearly highlighting the Late Holocene Rainforest Crisis (LHRC) which mainly developed from ca. 2500 to 2000 yr BP. The broad synchroneity of this interval is striking and has been revealed in many different deep lakes from humid or montane forest, lakes or swamps marginal to tropical forest, and swamp forests where the oscillations of the surface water table produce temporary emersions. In parallel, a chronological review is presented of the Bantu arrival in these areas, including indicators for burning, metallurgy and land clearance. Nevertheless, evidence of human occupation, such as artefacts or deposits containing charcoal is exceptional and generally absent in most areas. Although the archaeological data exhibit a gradual southward densification of human occupation throughout Central Africa, the increase of settlements clearly began after 2350 yr BP, not before 2500 BP. However, some authors have interpreted the geochemical signal of increased erosion in the Congo Basin or the opening up of the forest around Lake Barombi Mbo in Cameroon as being attributed to agricultural clearance, or even the supply of charcoal required for metallurgy. In short, these early Bantu settlers (in such modest densities) may have been responsible for some local landscape degradation (clearance, fire, metallurgy), but these same settlers could not, under any circumstances and across all Central Africa, be held responsible directly or indirectly for the synchronous changes of lake levels, draining vast swamps and opening up of the tropical forest canopy, which was due to an increased dry season, while the recovery with the recolonization of light demanding species, including oil palms (Elaeis guineensis) assisted by dispersers such as chimpanzees, was likewise due to natural processes. Therefore, it can be concluded that no data available validate the hypothesis that the major erosion or vegetation destruction ca. 2500 yr BP, was the result of large population movements. Indeed, the evolution of the environments of Central Africa are linked to the natural responses induced by general palaeoclimatic processes, observed synchronously not only in northern and eastern Africa, but globally. The natural recovery and resilience of these forests until the last centuries contrasts with the situation currently being faced.