Cargando…

History of the fragmentation of the African rain forest in the Dahomey Gap: insight from the demographic history of Terminalia superba

Paleo-environmental reconstructions show that the distribution of tropical African rain forests was affected by Quaternary climate changes. They suggest that the Dahomey Gap (DG)—the savanna corridor that currently separates Upper Guinean (UG, West Africa) and Lower Guinean (LG, western Central Afri...

Descripción completa

Detalles Bibliográficos
Autores principales: Demenou, Boris B., Doucet, Jean-Louis, Hardy, Olivier J.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Springer International Publishing 2017
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5943585/
https://www.ncbi.nlm.nih.gov/pubmed/29279603
http://dx.doi.org/10.1038/s41437-017-0035-0
Descripción
Sumario:Paleo-environmental reconstructions show that the distribution of tropical African rain forests was affected by Quaternary climate changes. They suggest that the Dahomey Gap (DG)—the savanna corridor that currently separates Upper Guinean (UG, West Africa) and Lower Guinean (LG, western Central Africa) rain forest blocks—was forested during the African Humid Holocene period (from at least 9 ka till 4.5 ka), and possibly during other interglacial periods, while an open vegetation developed in the DG under drier conditions, notably during glacial maxima. Nowadays, relics of semi-deciduous forests containing UG and LG forest species are still present within the DG. We used one of these species, the pioneer tree Terminalia superba (Combretaceae), to study past forest fragmentation in the DG and its impact on infraspecific biodiversity. A Bayesian clustering analysis of 299 individuals genotyped at 14 nuclear microsatellites revealed five parapatric genetic clusters (UG, DG, and three in LG) with low to moderate genetic differentiation (Fst from 0.02 to 0.24). Approximate Bayesian Computation analyses inferred a demographic bottleneck around the penultimate glacial period in all populations. They also supported an origin of the DG population by admixture of UG and LG populations around 54,000 (27,600–161,000) years bp, thus before the Last Glacial Maximum. These results contrast with those obtained on Distemonanthus benthamianus where the DG population seems to originate from the Humid Holocene period. We discuss these differences in light of the ecology of each species. Our results challenge the simplistic view linking population fragmentation/expansion with glacial/interglacial periods in African forest species.