How tropical trees in Africa adapted to cope with seasonal drought? Exploring the phylogenetic and climatic distribution of the species leaf habit, maximum height, and wood density

How tropical trees have evolved to cope with drought is still poorly known, especially in tropical Africa, which experiences a drier and more seasonal climate compared to other tropical regions. To address this gap, we characterized the phylogenetic and climatic distribution of three key traits associated with desiccation avoidance or tolerance, focusing on self-supporting forest and savanna species. We assembled a large database with newly compiled data on species leaf habit (evergreen vs deciduous) and maximum height, along with data on wood density, climatic niche, and phylogenetic relationships. Bayesian phylogenetic mixed models were used to assess the phylogenetic signal in niche and traits, to explore individual trait responses to climate and to investigate trait covariations, both in general and for each climatic niches. We identified significant phylogenetic signals, specifically for biome and leaf habit, and also major evolutionary changes. Relying solely on climate was ineffective for the prediction of wood density and insufficient for accurate predictions of species leaf habit or maximum height, despite the higher abundance of small deciduous trees in savannas than in forests. However, the effect of climate on leaf habit became evident when considering covariation with maximum height or wood density. We modelled the shift in species traits across forest layers, from an abundance of drought-tolerant evergreen species with dense wood in the understorey to drought-avoidant deciduous species with softer wood in the canopy. This pattern varies according to aridity as in the wettest African forests, it tends to be more common to find evergreen species that reach the canopy. This study represents a pioneering effort to model the leaf habit in tropical Africa for numerous species, considering the phylogenetic structure. Our models could be used to create detailed maps of leaf habit, leveraging phylogenetic relationships to fill missing data in existing datasets.