The biogeography of brachystegia woodland relicts in Southern Africa

dissertation submitted in fulfilment of the academic requirements for the degree of Master of Science in the School of Animal, Plant and Environmental Sciences, University of the Witwatersrand, Johannesburg. May 2015In this study I investigated the climate ecology and population biology of Brachystegia spiciformis (miombo) woodland at the distributional range edge in southern Africa. The main aim of the study was to exploit miombo relict populations, isolated well beyond contemporary distribution, to establish likely spatial and population dynamic response to future global climate change. Miombo woodlands have expanded and contracted across the central African plateau over geological time, with palynological evidence supporting an extensive latitudinal range during the Holocene altithermal. It is hypothesised that small shifts in climate may have major impacts on woodland dynamics and distribution. However, miombo relict populations in southern Africa, one in Mozambique and one in South Africa, suggest refugia; physiographic settings that support a once prevalent regional climate that has been lost (or is being lost) due to climatic shifts. Inclusion of relict populations has been shown to improve the performance of model-based projections elsewhere and have value as natural laboratories for investigating how populations react to on-going climatic change. This study aims to contribute to a better understanding of miombo woodland - an ecologically and economically significant savanna community - response to global climate change in southern Africa. A niche modelling approach was used to produce present-day and select future B. spiciformis woodland ecological niche models. Precipitation of the wettest quarter and temperature seasonality were identified as the two most important bioclimatic variables explaining B. spiciformis woodland distribution in southern Africa. Both variables displayed a relatively narrow range of optimal suitability for the species; 422 - 576 mm and 2.6 - 3.0 °C, respectively. In addition, significantly high temperature seasonality and maximum temperature of the warmest month were identified as limiting factors at the periphery of the contemporary miombo woodland distribution. Considering future regional climate change projections, it is suggested that the B. spiciformis climate niche could retract by between 30.6 - 47.3% in southern Africa by 2050. v In addition, a meta-population analysis of miombo woodland at the southern distributional range edge was carried out to elucidate patterns in population dynamics that could validate theorized climate response. Whilst the continuous miombo woodlands occurring to the north of the southern African range edge are internally relatively homogenous in community composition, relict populations reflected a loss of natural biota and a restructuring of the vegetation unit resulting in clear divergence from the core and range edge communities, and from each other. Although B. spiciformis reproduction, population density and structural dimensions were significantly low at the recently discovered relict population in South Africa relative to populations occurring to the north, we suggest that the medium-term persistence of this population is plausible based on the longevity of genets, their historic resilience, and a traditional management regime. The vicariant B. spiciformis relict population located in southern Mozambique is greater in extent than its South African counterpart and data suggest suitable reproduction dynamics for medium-term persistence. However, this population is currently heavily impacted by unsustainable habitat transformation under a lack of formal ecological or conservation protection. This study uniquely highlights B. spiciformis woodland as a climate (precipitation) sensitive component of savanna ecosystems in southern Africa and provides important baseline data for population dynamics at the distributional range edge