Global change effects on ecosystem species composition, functions and services of West Africa’s Sudanian savannas

Abstract

Savanna ecosystems cover an eighth of the world's land surface and are of immense ecological and economic importance. Ecologically, they help regulate climate, air quality, water quality, and soil erosion. Economically, savannas provide food, medicines and fiber to humans and animals; e.g. they support approximately 50% of global livestock production.<br /> Global environmental change has led to major impacts on savannas worldwide. Key determinants of species distribution, biodiversity and ecosystem functioning in rangelands include grazing and climate. In Africa’s savannas, rangelands provide important ecosystem services and contribute considerably to local livelihoods. Here, grazing by domestic herbivores is an important type of land-use. It is usually done on extensive basis thus highlighting the importance of rangelands. Despite drastic increases in land-use pressure and considerable climatic changes affecting Africa's savannas, there is still limited information about the importance of these global change agents for spatio-temporal patterns in ecosystem functions and services, a situation which hinders the development and implementation of effective land management strategies. In this context, understanding the underlying environmental drivers of herbaceous species composition, diversity, ecosystem functioning and ecosystem service provision is a crucial step. This thesis aimed at filling these critical research gaps by investigating the impacts of multiple environmental factors on savanna ecosystems. In section 2, this study aimed at determining the drivers of herbaceous plant species composition and distribution. In section 3, the study aimed at determining the drivers of plant species and functional diversity of the herbaceous layer while section 4 aimed at determining the drivers of major ecosystem services, namely forage provision and erosion control.<br /> This study was conducted within the framework of the West African Science Service Center on Climate Change and Adapted Land Use (WASCAL) project at the University of Bonn which was funded by the German Federal Ministry of Education and Research (BMBF). Field work was conducted in the Sudanian savannas of Ghana and Burkina Faso. Due to the existence of a steep regional gradient of climatic aridity, a space-time substitution approach was used in this study aiming to understand future climatic impacts on herbaceous vegetation. The sampling was stratified into three rainfall zones oriented along a south-north gradient of increasing climatic aridity. Within strata, sampling was designed to capture as much as possible of the variations in geology, grazing intensity and topography by choosing sites that maximise the range of these gradients. A nested plot design was used where three 1 m² circular subplots were randomly placed in each plot of 10 m x 10 m.<br /> For section 2, I collected data at two spatial scales; a regional scale comprised of long-term climatic data (averaged over a 50-year period) and a local scale comprised of plot characteristics, namely land-use intensity (related to grazing pressure), topo-edaphic conditions (topography, soil physical and chemical properties) and vegetation data (species’ name, species’ cover and species’ height). For section 3, I collected data on long-term climatic conditions, topo-edaphic conditions, disturbance (grazing pressure and fire frequency) and vegetation. The vegetation data comprised of species’ name, species’ cover and six plant traits, namely life history, height, life form, photosynthetic pathway, growth habit and nitrogen fixation. Vegetation data was used for the calculation of species and functional diversity indices. For section 4, I collected data on long-term climatic conditions, season’s antecedent precipitation, topo-edaphic conditions, land-use intensity and vegetation. The vegetation data comprised of species’ name, species’ cover and three plant traits, namely height, growth form and life form. These traits were used to form plant functional types. Additionally, I derived three proxies of the provisioning ecosystem service of forage provision, namely aboveground biomass, metabolisable energy, metabolisable energy yield and one proxy of the regulating ecosystem service of erosion control, namely perennial plant cover. Aboveground biomass was derived via allometric functions based on biomass measurements from representative sample plots across the study area. A portable field spectro-radiometer was used to measure plant reflectances and I then used a regression model, calibrated in the same area, to estimate the metaboblisable energy. Metabolisable energy yield was obtained as a product of aboveground biomass and metabolisable energy.<br /> In section 2, this study used; (1) Mantel tests and variance partitioning to identify the drivers of vegetation composition, (2) isometric feature mapping and partitioning around medoids (Isomap) to perform agglomerative cluster analysis and (3) non-metric multidimensional scaling (NMDS) to perform ordination. In sections 3 and 4, linear mixed-effect models with model selection procedures were applied to obtain the best set of predictors for each measure of diversity (section 3) and ecosystem service (section 4).<br /> In section 2, the results from variance partitioning showed that the regional scale drivers (long-term climate at a given site) were more important for determining vegetation composition than local scale drivers (topo-edaphic conditions and land-use intensity) and plot’s geographic location. The Mantel tests showed that long-term precipitation (averaged over a 50-year period) had the highest correlation with herbaceous vegetation composition. Ten herbaceous vegetation clusters were found, arranged along three NMDS axes, that mainly represent climatic and land-use (grazing) variations which have been found as major drivers of the spatial differentiation of species composition and vegetation clusters in the area. In section 3, the study found that taxonomic diversity and functional diversity mostly varied independently and were not strongly correlated to each other. Precipitation seasonality and grazing intensity were signicantly related to all functional diversity indices but not to any of the taxonomic diversity indices except species richness. Taxonomic diversity indices were significantly related to soil texture and topography. In section 4, the importance of long-term climate regime (averaged over a 50-year period) varied with ecosystem services: it was less important than antecedent precipitation for aboveground biomass, and metabolisable energy yield but was more important for perennial plant cover than antecedent precipitation. Land-use intensity (grazing pressure) was an important predictor for forage provision but not for perennial plant cover. Vegetation attributes (plant functional types, phenological stage and species diversity) were important predictors for all ecosystem services while topo-edaphic conditions were of secondary importance.<br /> In summary, long-term climate was found to mainly drive floristic composition and diversity on a regional scale. It also exerted (indirect) effects on ecosystem service provision via its effects on vegetation attributes and hence on ecosystem structure and function. Land-use (grazing) and topo-edaphic conditions acted mostly as modifiers of ecosystem structure and function at the local scale. These findings have two major implications for understanding climate change effects on ecosystem services provided by West Africa’s Sudanian savannas. First, local site conditions (in soil, topography, land-use etc) could determine to which extent climate change effects on plant communities are actually translated into changes in ecosystem structure and function and second, short-term (seasonal) variation in rainfall may mask effects of changing climate and land-use on forage provision. The findings of this study are useful and can serve as a decision-making support tool for policy makers, rangeland managers and conservationist within the context of ongoing climate change.