Baobabs on Kubu Island, Botswana – A dendrochronological multi-parameter study using ring width and stable isotopes (δ13C, δ18O)

Abstract

According to the Intergovernmental Panel on Climate Change (IPCC), all of Africa is very likely to warm up more than the global average during this century. Especially (semi-)arid regions are expected to experience particularly high warming and possibly catastrophic droughts. However, assessments of the impacts of climate change on these regions are currently impeded by a lack of transregional high temporal resolution proxy data for the African continent. Baobab trees are widely distributed in (semi-)arid Africa and can reach ages of up to 2000 years. This pilot study was aimed at investigating African baobabs, Adansonia spp., from a site in Botswana using multiple dendroclimatological methods. Increment cores from 16 individual baobabs growing on Kubu Island (20°53’ S, 25°49’ E), a granite pluton located in the Kalahari, were collected in June 2011 to test for annual growth and the species’ utility for palaeoclimatic studies. Due to the particular wood fabric and relatively high water content, baobab increment cores were packed in air-tight opaque tubes and stored in a freezer to prevent drying and mould formation. The complicated wood anatomical structure was found to be analysed best using a microscope with incident UV light, allowing tree-ring boundaries to be distinguished. Nonetheless, potential differences in individual site conditions, as well as diverse tree ages, caused conventional dendrochronological crossdating to fail. Missing and false tree rings could be identified due to the strong relationship between tree-ring width and annual precipitation amount allowing the development of a preliminary 50 year-long baobab chronology (1960–2009). Subsequently, stable carbon and oxygen isotope analyses revealed significant correlations of Δ13C and δ18O of tree rings with climate data. Year-to-year isotope variability and trends were found to be in good agreement with established models of fractionation. Intrinsic water-use efficiency has mainly increased over the study period (2–30 %). Despite the demonstrated high potential of African baobabs as a valuable high-resolution climate archive, we conclude that more dendrochronological calibration studies are required at various sites in southern Africa. Furthermore, ecophysiological monitoring of climate and stable isotope signal transfer from the atmosphere, through soil and leaves into the tree rings is necessary to fully understand tree-ring formation and climate response of the African baobab.