One person’s vegetation anomaly is another person’s climate-change laboratory: The story of (South) Africa’s southernmost miombo woodland

By Brenden Pienaar (MSc candidate, WITS University) & Dr Dave Thompson (Biodiversity Scientist, SAEON Ndlovu Node)

Brachystegia spiciformis is the principal tree component of miombo, the colloquial term used to describe savanna woodlands dominated by three closely related genera (Brachystegia, Julbernardia and Isoberlinia; Figure 1). Miombo covers much of south-central Africa with a contemporary southern distributional limit at approximately 21^oS, sited in south-central Zimbabwe (Figure 2).

Figure 1. The extensive miombo woodlands of south-central Africa, here in central Zimbabwe, are dominated by Brachystegia spiciformis and allied genera (Photo: B Pienaar)

And yet, in the far north-eastern reaches of South Africa, local people refer to the tree as ‘Musiiwa’ - the one that was left behind.

Figure 2. Distribution of miombo woodland in Africa shown in grey. Adapted from White (1983)

To etymology scholars this suggests an understanding by the indigenous people of a species displaced within human, rather than ecological, time-frames. Sediment cores containing Brachystegia pollen, dated 6000-1000 years ago and recovered from Mookgophong, Tate Vondo and Pretoria in South Africa (Figure 4), confirm a widespread historic range for the species across central Africa, and south (to approximately 26^o south, in Gauteng) of the present-day distribution during the hotter and wetter conditions of the mid-Holocene.

Miombo woodland likely expanded and retracted repeatedly across the sub-continent in response to climatic fluxes over geological time and the approximate 450-km north-eastward range retraction of the species during the past 1000 years suggests either an abrupt change in climate, for which there is no evidence, or that minor shifts in temperature and moisture regimes triggered marked changes in Brachystegia population dynamics.

Could miombo again expand southwards?

The 1999 South African Plant Biodiversity Vulnerability and Adaptation Assessment (Rutherford et al., 1999) forecast that miombo woodland could again expand southwards, given suitable ecological habitat under future global climate change scenarios along the high rainfall savanna-grassland biome interface of north-eastern South Africa (Figure 3). However, such expansion towards lower latitudes back into South Africa across the Limpopo River Valley was thought unlikely given seed dispersal limitations.

Just two years later, the 2001 anomalous discovery of ~15 hectares of miombo woodland (known as Gundani locally) north of Thohoyandou (see Figure 4) placed the species well south of the previously recognised contemporary distribution - both within South Africa and within the region predicted by Rutherford et al. (1999) as comprising the suitable climate envelope for the species under climate change.

The only Brachystegia utilis individual in South Africa occurs within this B. spiciformis community, suggesting that this isolated patch of miombo is likely a climate relict from a time when, at least, the genus dominated the Soutpansberg massif in north-eastern South Africa.

Can climate relicts help us predict global change responses?

At global and regional scales, climate broadly determines the distribution of plant taxa, and the response of species to changing environments is consequently likely to be largely determined by population responses at range margins. Climate relicts - such as Gundani - form through the discontinuous retraction of the population’s trailing edge and persist in isolated areas of favourable environmental conditions within an inhospitable regional climate.

Studies of vegetation-climate dynamics, which are critical in informing our ability to predict likely community responses, and biodiversity consequences, of future global change, are vastly biased in favour of high-latitude (northern hemisphere), high-altitude range margins and temperate plant communities, which typically have minimum temperature controls.

Conversely, Musiiwa - the one that was left behind provides a unique opportunity to explore climatic constraints at low-latitude (southern hemisphere) range margins for an ecologically and economically significant African savanna species that is likely particularly responsive to present and future anthropogenic climate change.

Natural laboratory with enormous value in climate change detection

Using a predictive modelling approach, researchers from SAEON, the Global Change and Sustainability Research Institute and School of Animal, Plant and Environmental Sciences at the University of the Witwatersrand, and the School of Agricultural, Earth and Environmental Sciences at the University of KwaZulu-Natal, have been working with Master of Science candidate Brenden Pienaar to answer questions relating to the mechanism and control of Gundani’s isolation and the persistence and behaviour of the population under selected global climate change projections.

In so doing, the relict has been validated as a natural laboratory with enormous value in climate change detection and response monitoring studies regionally, and within Africa.

From among 19 regional bioclimatic variables, altitude and soil, the researchers show that precipitation of the wettest quarter and temperature seasonality - both with narrow ranges of suitability, are the two most important variables explaining the distribution of B. spiciformis woodland in southern Africa, highlighting the utility of the species as an indicator of climate change.

Based on these controls of distribution, multiple environmental niche models predict the occurrence of isolated B. spiciformis populations in South Africa under current climatic conditions, one of which is coincident with the Gundani climate relict (Figure 4). Although these models indicate suitable ecological conditions for the species elsewhere at disjunct locations along the high rainfall savanna-grassland interface of north-eastern South Africa, the species does not presently occur outside of the relict population - essentially a refinement of the 1999 Rutherford et al. projections (Figure 3).

Figure 3. The 1999 South African Plant Biodiversity Vulnerability and Adaptation Assessment forecast southward range extension for B. spiciformis into South Africa under predicted climate change. One model output – here highlighted in green, shows extensive range expansion along the grassland-savanna biome interface. Figure 4. Ecological niche of Brachystegia spiciformis woodlands in southern Africa under present-day climate conditions. Location of the Gundani climate relict population (black circle) and three Mid-Holocene pollen records (black triangles) is included.

What does the future hold for Musiiwa?

Three commonly used atmosphere-ocean general circulation models, applied under an intermediate emissions scenario, are consistent with a predicted decrease in the extent of the ecological niche of 30-47% for the continuous miombo woodlands of southern Africa. This reflects a net change regionally, with a decreased distribution resulting from larger retractions of present-day distributions, primarily in Zimbabwe, combined with comparatively smaller range increases at newly-favourable locations elsewhere (Figure 5).

For the most part, range retraction will be driven by temperature seasonality and maximum temperature of the warmest month under future climates exceeding the maximum tolerance of the species across its present-day range. Essentially temperatures will be too high to support B. spiciformis woodland in these areas. Considerable southward range expansion, as far as 29 ^ºS, is supported in South Africa due to increasingly favourable climatic conditions - particularly as the most important bioclimatic variable explaining B. spiciformis distribution, namely precipitation of the wettest quarter, increases for the species at the savanna-grassland biome interface.

There is disagreement between the models concerning the persistence of the Gundani miombo woodland, but the majority suggest that conditions will become increasingly unfavourable for B. spiciformis by 2050.

Figure 5. On-going research suggests suitable, albeit fragmented, habitat for B. spiciformis in South Africa, even under present-day climates. The location of the anomalous Gundani population (red circle) is recovered by this composite model. The three Mid-Holocene pollen records (yellow circles) are included. Figure 6. One of several predicted future (2050) climate envelopes for B. spiciformis in southern Africa has been developed as part of the current research project. Here the output from model HadCM3 is shown in dark green. Significant range retraction - shown in light green, is forecast in Zimbabwe and neighbouring Mozambique, whilst range expansion occurs along the eastern escarpment into South Africa.

Persistence of the relict, at least in the short- or medium-term and so avoiding immediate local extirpation, is plausible given that vegetation boundaries are not in equilibrium with climate as adult trees may persist in an area of previously suitable climate, particularly as the species is capable of clonal regeneration.

Figure 7. MSc student Brenden Pienaar collects Brachystegia spiciformis population demographic data at the ~15ha Gundani miombo woodland in north-eastern South Africa (Photo: B Pienaar)

Notwithstanding seed dispersal limitations, the vast hurdle of the Limpopo River Valley, which currently exceeds maximum temperature tolerance of the species, makes southward colonisation from beyond South Africa’s borders doubtful. Range expansion within South Africa southwards from Gundani is equally improbable, given the scatter of the predicted refugia, and the highly transformed nature of the eastern escarpment (Figure 5).

As Brachystegia woodland and associated biota form crucial socio-economic and biodiversity components of savannas in southern Africa, their suggested further range retraction is of concern.

Before we say farewell to Musiiwa - at least in South Africa, the suggested opposite responses of the continuous miombo woodlands relative to that of the South African relict demand an understanding of population dynamics and demographics to validate the predictions concerning migration, persistence and extinction of B. spiciformis. This field monitoring forms part of the MSc research (Figure 6) which has already shown that climate relicts, even if doomed to local extinction ‘Miri a itsha wanala – that which is lost forever’, have value as instructive models and natural laboratories for investigating population response to on-going climatic change.

Reference

Rutherford MC, Midgley GF, Bond WJ, Powrie LW, Roberts R, Allsopp J. South African country study on climate change. Plant Biodiversity: Vulnerability and Adaptation Assessment. 1999. SANBI.