Functional traits, drought performance, and the distribution of tree species in tropical forests of Ghana

  Tropical forests occur along a rainfall gradient where annual amount, the length and intensity of dry season vary and water availability shapes therefore strongly the distribution of tree species. Annual rainfall in West Africa has declined at a rate of 4% per decade, and climate change models predict a further reduction in rainfall and an increase in frequency and intensity of drought. This will have large consequences for the diversity, composition and distribution of tropical tree species. Understanding the factors that shape tree species distribution will help to understand current forest functioning and to predict the potential impact of climate change on forests. In this thesis, I used a combination of forest inventory data, greenhouse and field experiments to determine the responses of 10-23 species to drought and shade, and analyse the underlying mechanisms. I addressed 4 questions: (1) What is the relative importance of rainfall and temperature on tree species distribution? (2) How do tree species acclimatise to drought and shade in terms of their physiology, morphology, growth and survival? (3) What morphological and physiological traits determine species drought performance and distribution? (4) How do seedling survival, growth and physiology vary between dry and wet forests, and does drought tolerance and growth determine species distribution along the rainfall gradient? Forest inventory data showed that the distribution of 95% of 20 species was significantly associated with annual rainfall, 60% with rainfall seasonality, 45% with isothermality and 40% with temperature seasonality. Thus, a reduction in annual rainfall, and an increase in frequency and intensity of drought as predicted by climate change models may affect the distribution of many tree species. A greenhouse experiment indicated that shade facilitated the survival of seedlings subjected to drought, rather than reducing it. This contrasts with the trade-off hypothesis that suggests a stronger impact of drought in shade conditions. Across 23 species, I found a trade-off between drought avoidance (by a deciduous leaf habit during drought) and physiological drought tolerance (by having tough and persistent tissues that allow plants to function during drought) strategies. These strategies were closely associated with species’ shade tolerance and growth rates. A suite of functional traits predicted drought survival and tree species position on the rainfall gradient. Across species, drought survival was enhanced by having less biomass allocation to transpiring leaves, a low leaf area per unit plant mass, and by dense and tough leaf and wood tissues that allow plants to function during drought. The field experiment showed that drought survival (and growth) in the dry forest relative to the wet forest correlated negatively with species position on the rainfall gradient. Hence, species that survive and grow relatively well in dry forests are found at the drier end of the rainfall gradient. This suggests that species sensitivity to low water availability determines the distribution of tree species. The predicted increase in drought frequency and intensity may, therefore, cause a shift in the distribution of tree species in tropical forests