Landscape genetics indicate recently increased habitat fragmentation in African forest‐associated chafers

Abstract

Today, indigenous forests cover less than 0.6% of South Africa's land surface and are highly fragmented. Most forest relicts are very small and typically occur in fire-protected gorges along the eastern Great Escarpment. Yet, they hold a unique and valuable fauna with high endemism and ancient phylogenetic lineages, fostered by long-term climatic stability and complex microclimates. Despite numerous studies on southern African vegetation cover, the current state of knowledge about the natural extension of indigenous forests is rather fragmentary. We use an integrated approach of population-level phylogeography and climatic niche modeling of forest-associated chafer species to assess connectivity and extent of forest habitats since the last glacial maximum. Current and past species distribution models ascertained potential fluctuations of forest distribution and supported a much wider potential current extension of forests based on climatic data. Considerable genetic admixture of mitochondrial and nuclear DNA among many populations and an increase in mean population mutation rate in Extended Bayesian Skyline Plots of all species indicated more extended or better connected forests in the recent past (<5 kya). Genetic isolation of certain populations, as revealed by population differentiation statistics (GST'), as well as landscape connectivity statistics and habitat succession scenarios suggests considerable loss of habitat connectivity. As major anthropogenic influence is likely, conservational actions need to be considered.

The genetic consequences of habitat specificity for fig trees in southern African fragmented forests

“…craterostoma figs are also known to be eaten by monkeys (Basabose, 2002;Linden et al, 2015), their small size and the fact that they ripen to a bright yellow and sometimes even to a reddish colour suggests that birds are important dispersers (Lomáscolo et al, 2008), and figs in the canopy are more likely to attract birds that disperse over long distances . Continuous forest cover across parts of South Africa is believed to have started to fragment during the late Pleistocene (Lawes, 1990;Lawes et al, 2007a;Eberle et al, 2017). The extent to which the populations of the two forest-specialist fig tree species have diverged is consistent with such long-term fragmentation of their forest habitat and agrees with conclusions based on regional climatic conditions inferred from pollen studies (Scott et al, 1997;Scott, 2002) and climatic reconstructions (Eeley et al, 1999;Eberle et al, 2017 (White, 1983).…”

“…Continuous forest cover across parts of South Africa is believed to have started to fragment during the late Pleistocene (Lawes, 1990;Lawes et al, 2007a;Eberle et al, 2017). The extent to which the populations of the two forest-specialist fig tree species have diverged is consistent with such long-term fragmentation of their forest habitat and agrees with conclusions based on regional climatic conditions inferred from pollen studies (Scott et al, 1997;Scott, 2002) and climatic reconstructions (Eeley et al, 1999;Eberle et al, 2017 (White, 1983).…”

Oscillayers: A dataset for the study of climatic oscillations over Plio‐Pleistocene time‐scales at high spatial‐temporal resolution

“…Other potential applications include, for example, the generation of PalaeoENMs via georeferenced fossils (Myers et al, ); the testing of biodiversity‐related hypotheses about palaeo‐climatic stability in the tropics (e.g. Couvreur et al, ; Kissling et al, ; Rakotoarinivo et al, ); the testing of predictions of the glacial‐sensitive model of island biogeography (Fernández‐Palacios et al, ; Norder et al, ) or the facilitation of landscape connectivity (dispersal corridor) analyses over time in a conservation context (Eberle, Rödder, Beckett, & Ahrens, ; Yu et al, ).…”

The role of Sahara highlands in the diversification and desert colonization of the Bosc's fringe‐toed lizard

“…Ensembles were produced by averaging the downscaled outputs of all available scenarios (delta method; CCSM4, CNRM-CM5, FGOALS-g2, IPSL-CM5A-LR, MIROC-ESM, MPI-ESM-P and MRI-CGCM3; Fick & Hijmans, 2017) prior to the palaeoprojections (e.g. Eberle et al, 2017). Palaeoclimatic variables were downloaded at the same 2.5 arc-minute resolution and cropped to the study area to decrease computation time.…”