Major Research Themes

There are four major research themes in this research group

Dune fynbos-thicket mosaic ecology

Ecology of the Albany Subtropical Thicket

Restoration of spekboom-dominant subtropical thicket

Biome boundary drivers


Prospective students (from Mandela University or elsewhere) interested in terrestrial botany or molecular ecology should contact me directly to discuss academic interests and potential projects (3rd year to PhD). I strongly encourage postgraduate students to develop their own research systems in addition to collaborating with me on current study systems. A typical postgraduate student will have the opportunity to conduct fieldwork, potentially do molecular lab work, and run ecological laboratory or field experiments. Students should integrate across several of these fields in their thesis projects.

Please note that a list of of potential student funding sources is available here.

Please see this Google Document for a list of projects that I have proposed.

and see this Google Document for a list of projects underway in my research group.

Dune Thicket ecology: Underground trees

Background. Preliminary investigations of underground structures of subtropical thicket lineages suggests that these trees have traded growing tall for growing wide ​— underground! This study will aim at characterising this undescribed behaviour.

Methods. A range of different methods ​— from DNA to digging them up...

Status. Ms Sinenjongo Gcina has excavated many species in the dune landscapes and classified their resprouting type as part of her MSc thesis (2019-2020). However, the extent of individual plants remains to be explored. These are potentially some of the largest "trees" in the world, they just grow sideways under ground..

Dune Fynbos-Thicket ecology: Life history strategies and hydraulic traits

Background. The dune fynbos-thicket mosaic has species with many different life history strategies that likely differ in terms of hydraulic traits across different regeneration niches. This study would involve comparing hydraulic traits across postfire nonsprouters, facultative sprouters, obligate sprouters, and opportunists in these systems.

Methods. Pressure volume curves (generated using a Scholander pressure chamber) and similar methods.

Background reading: Pratt et al. (2012) Xylem transport safety and efficiency differ among Fynbos shrub life history types and between two sites differing in mean rainfall. Int. J. Plant Sci. 173, 474-483.

Dune Fynbos-Thicket ecology: Effect of wind on pollination

Background. Wind makes it difficult for pollinators to traverse landscapes and access flowers. But does it really?

Methods. By combining on the ground wind speed sensors and camera traps, the degree of pollination under various wind speeds will be assessed.

Dune Fynbos-Thicket Ecology: Effect of shading on thicket and fynbos plants

Background. Fynbos needs to grow in highlight conditions ​— and fire keeps the system open and keeps the light levels high. There are also thicket species that grow in the high fire frequency parts of the landscape. Dense tall vegetation species ​— from thicket and forest ​— can survive in low-light conditions. However, this has never been investigated! A simple concept, cited in the literature, and there is no data on it for fynbos and dune fynbos.

Methods. In-field and common garden shading experiments. For example, use shade cloth of different intensities planted across existing communities on the reserve and measure flowering, fruiting and seed production differences. This could also be done in a potted experiment.

Dune Fynbos-Thicket Ecology: Tracking plant species through the postfire environment

Background. Four trail camera have been established that take hourly photos since a fire that occurred in November 2019.

Aims. To use repeat photography coupled with in-field validation of species and individuals to assess life-history strategy classifications (e.g. using Grime's LHS classifications) for the various species in the photos at each site.

Dune Fynbos-Thicket ecology: Yoda's self-thinning rule and underground trees

Background. We have strong evidence of self-thinning of above-ground ramets (branches) of the underground tree, Olea exasperata. But does this also apply to other underground trees in the dune systems?

Methods. Measuring stem densities and sizes of a range of other dune thicket plant species ​— e.g. Rapaneae gilliana, Searsia laevigata, and Euclea racemosa ​— and developing allometric biomass models.

Thicket restoration: Soil temperatures

Background. Thicket degradation exposes the soils to high temperatures. To what degree these temperature affect root development, seedling establishment etc. etc. is largely unexplored (although Richard Lechmere-Oertel did explore this in his seminal PhD work) .

Study system. Areas of that have both intact and degraded thicket along fence-line contrasts.

Methods. This project compares below-ground and above-ground temperatures across intact, degraded and restored spekboom-veld using an array of iButtons.

Status. >1 year of data has been collected, and requires analysis.

Thicket restoration: Effects of herbivory

Background. Herbivory has been identified as a major contributing factor to failure in restoration efforts using Portulacaria afra (spekboom). This project will explore the different ways in which herbivores damage planted spekboom-cuttings, and test ways to limit such damage.

Study system. Areas of degraded thicket exposed to indigenous and/or domestic browsers.

Methods. Trial sites with different methods for planting spekboom will be monitored using camera traps to observe the fate of cuttings.

Thicket restoration: where do all the seeds go?

Background. Portulucaria afra ​— spekboom ​— seeds are very short-lived (a few months at best). Recruitment from seeds is considered to be exceptionally rare. The fate of seeds has not been explored ​— are they eaten, die soon after germinating, or fail to grow beyond a seedling.

Study system. Areas of intact and degraded thicket where spekboom is present and is flowering.

Methods. A range of experiments, including monitoring of seeds in petri dishes, experimental watering of seeds in the wild and in a greenhouse, revisiting sites with known seedling establishment.

Thicket restoration: return of biodiversity

Background. Degraded Thicket does not offers little structural habitat for fauna. Without this habitat, insects and small animals ​— such as chameleons ​— are lost from the system.

Study system. 15-old restored sites (Thicket-Wide Plots) ​— of which there are ~100 across the Eastern Cape.

Methods. Insect and small animal assessments will be conducted in restored thicket-wide plots and neighbouring degraded thicket to compare the faunal biodiversity.

Thicket restoration: the golden rooting ratio?

Background. The rate of growth of Portulucaria afra ​— spekboom ​— cuttings used for restoration varies through time and across the landscape. We suspect that the time lag between planting and when cuttings start expanding is the time the plant needs to establish an adequate root-to-shoot ratio. At present, we don't know what this ratio may be, or what we can do to increase the rate of root growth. We could also use this data to calculate the below-ground carbon stock based on above-ground biomass.

Study system. Areas of degraded thicket where cuttings have been established over the last fifteen years.

Methods. Excavating plants in different states and ages to estimate the root-ratio and link these to above-ground parameters (such as growth).

Thicket classification: Is mesic thicket actually Forest?

Background. There are three forms of Albany Subtropical Thicket: arid thicket, valley thicket and mesic thicket. Arid and valley thicket types are ecologically and floristically very similar (e.g. slow growing and rich in succulents). Mesic thicket, in contrast, lacks most of the succulent component and is fast growing ​— and generally shares more with the Forest biome than the Thicket biome. This study aims to provide quantitative floristic data across the various thicket forest types to determine whether the mesic thicket should be reclassified as Forest type.

Study system. The various forms of thicket and forest in the Eastern Cape.

Methods. Vegetation plot data will be extracted from existing datasets and new data will be collected where there are gaps.

Progress. Mr Abul-Lateef Ismail will be conducting these assessments for his MSc thesis, starting in 2021.

Thicket ecology: Under da ground ​— root functioning and water absorption

Background. Plants largely absorb water by creating a water pressure potential (Ψ) between their leaves and their roots. The strength of Ψ is dictated by wood density and strength. Spekboom has a low wood density and thus can only generate low Ψ. They hypothesis is that it has a different approach, specifically a sponge-like root structure that aids in the absorption of water (unrelated to Ψ).

Methods. Measure the ability of roots from different thicket species to absorb water without Ψ (i.e. roots removed and dried, and then rehydrated).

Thicket ecology: relative roles of termites and saprotrophic microorganisms as drivers of wood decay

Background. Deadwood and leaflitter represent an important carbon pool in Subtropical Thicket. Biological decay of of this pool is driven by macro-invertebrates and saprophytic microorganisms. The rates of decay across Thicket types and in degraded and intact vegetation may vary. Termite activity is largely unexplored in Thicket.

Methods. Following the methods proposed by Cheesman et al. (2018).

Thicket ecology: Leaf longevity of dominant tree species

Background. Almost all trees in the Albany Subtropical Thicket communities are evergreen ​— or more precisely, drought-deciduous (in the case of thicket trees, they lose their leaves in response to extended drought that could occur at any time of the year). We have very little understanding of the longevity of individual leaves ​— are leaves produced and shed continuously and quickly, or are their seasonal patterns of growth and shedding, or are leaves produced continuously but have a high longevity.

Methods. Individual leaves from on branches from different thicket tree species will be monitored. Also, specific leaf area (SLA) will be measured and compared with leaves from truly deciduous systems (e.g. Savannas)

Thicket ecology: Elephant damage comparisons between savanna and thicket

Background. Elephants are highly destructive feeders. However, the degree of destruction varies across biomes. This project would quantify branch and stem sizes, and bark stripping across savanna and thicket biomes.

Methods. Surveys in elephant areas measuring plant heights, size of damaged branches etc, and degree of bark stripping.

Thicket ecology: Elephants as major dispersal agents of "Spekboom"

Background. Elephants are highly destructive feeders ​— but they also walk and eat. On another tangent, the seedling recruitment of Portulacaria afra

Methods. Surveys in elephant areas measuring plant heights, size of damaged branches etc, and degree of bark stripping.

Progress. Lara Wattam (Hons, 2019). Strong preliminary support that elephants are important dispersal agents for P. afra, but further work is required.

Thicket ecology: The light niche across succulent plants

Background. Thicket has a rich succulent flora (and has been identified as one of South Africa's hotspots of succulent diversity). What is unusual about this succulent diversity is that much of it occurs under a canopy ​— i.e. in the shade. We usually associate succulents with open high light conditions. Thus, different thicket CAM species appear to specialise in different light intensities ​— from full sunlight to deep dappled shade. This study will explore whether there are CAM light specialists and light niches in thicket.

Methods. Surveying the below and above canopy light conditions for n>50 individuals per succulent species to characterise the light niche that each species occupies (e.g. open, edge, dappled shade, deep shade). This will help make predictions about how elephant impacts (i.e. the changing of the canopy structure) may impact different succulent species.

Thicket ecology: A comparison of seed dispersal syndromes across South African biomes

Background. Thicket has a lot of species that have bird- and mega-herbivore-dispersed seeds. Is this significantly higher than other biomes?

Methods. Primarily exploring the literature and conducting a meta-analysis of existing data, plus generating data for the dune fynbos-thicket mosaic on campus for a case study.

Thicket ecology: The effect of extra-limital top-down browsers on thicket trees

Background. Giraffe have been introduced into many subtropical thicket landscapes for the tourism industry. The effects of such a "top-down" browser on trees requires further exploration ​— Parker & Bernard highlighted potential tree species that may be of concern in 2005, but this topic has not been further explored since.

Methods. Exploring the changes in architecture between trees browsed by giraffe and those not, and determining whether giraffe-browsed trees are being negatively affected.

Biome boundaries: leaf size and frost resistance

Background. Many vegetation types are dominated by plant species with nanophyllous leaves, e.g. grassland and fynbos, and others dominated by mesophyllous leaves, e.g. forest and fynbos. This study will explore whether their are trade-offs between leaf size and frost-resistance.

Methods. Freezing experiments of species with different leaf sizes to determine reduction in photosynthetic efficiency. Methods will follow Duker et al. (2015).

Biome boundaries: leaf size and low-light tolerance

Background. Different vegetation types often have a characteristic range of leaf sizes. For example, fynbos is dominated by species with nanophyllous leaves, whereas forests are dominated those with mesophyllous leaves. There may be a trade-off between leaf size and tolerance to low-light levels.

Methods. Transplant and shading experiments to test the influence of shading on different measures of plant growth.

Status. Unexplored thus far in my group, but Dr Simon Power has done some seminal work on this topic. Nonetheless, experimental studies are still lacking.

Biome boundaries: predicted vs observed climate conditions from "leaf"-communities

Background. The Climate Leaf Analysis Multivariate Program is based on a global database of plant communities ​— characterised by their leaf morphologies ​— linked to local climate data. This database is then used to reconstruct potential climates from fossil-leaf communities. Here, we will use it to predict the climate of the different plant communities in and around Port Elizabeth (e.g. grassland, fynbos, thicket, and forest) and compare this with our observed climate.

Methods. Characterise the leaf morphology of a range of plant communities, use the CLAMP system to develop expected climate for each community, and compare this with our actual local climate.

Biome boundaries: grassland vs subtropical thicket

Background. Grassland and Subtropical Thicket co-exist across the Eastern Cape with grasslands restricted to upper mountain tops on sandstone and thicket on valley slopes on shales. Where precipitation increases above ~600 mm per annum, then grasses invade off the sandstone onto shales, and thicket retreats into fire refugia. These field observations have never been tested. In addition, an invasive grass species (Pennisetum setaceum ​— in the picture) appears to be able to grow on shale at low precipitation where our indigenous grass species cannot. This study will test the hypothesis that the absence of a high cover of indigenous grasses in thicket is geologically driven, and compare the traits of our indigenous grasses with the invasive grass.

Methods. Similar pot-based greenhouse experiments as conducted by Lechmere-Oertel and Cowling (2001), as well as field-based transplant experiments similar to Esler et al. (2015).

Fynbos/grassland ecology: Wand plant architecture

Throughout the Cape Floristic Region, in a range of local environments, can be found a distinctive growth form: “wand” plants. This curious plant architecture comprises perennial plants which have slender (wand-like) stems that extend high above the matrix vegetation. We have published this paper on the topic: Bailey et al. (2019), but there is still much work to be done, including:

  • Testing alternative hypotheses posited in Bailey et al. For example:

    • Using camera traps to assess insect visitation rates of raised flowers versus those close to the ground.

  • Developing specific metrics to for comparisons amongst wand plants and with non-wand plants.

Status. Mr Abdul-Lateef Ismail conducted field experimental work (Hons: 2020) exploring visitation rates of pollinators at different heights in a system that does not have wand plants.

Hybridisation between local and introduced Protea species: outbreeding depression and Witches Broom

The climatic stability in the Cape (in comparison to other parts of the world) has meant that Cape species have not had to shift around on the landscape (i.e. did not have to retreat from glaciers). This means that many species have never come into contact.

At the Van Stadens Wildflower Reserve, a range of local and non-local Proteaceae are grown together ​— and hybridise (Macqueen & Potts, 2018). Both field observations and preliminary investigations demonstrate that Witches Broom affect hybrids far more than the parent species. This project will:

  • Assess Witches Broom, and the resulting decline in fecundity of hybrids and non-hybrids

  • Identify hybrids and non-hybrids using cheap and quick PCR primer-tagging techniques.

Landscape ecology: Role of aeolian dust in shaping landscapes — a termite-free test of the heuweltjie-depositional hypothesis

Background. Heuweltjies are common features of the Western Cape landscape. Their origins remain debated. Some argue that they are ancient termite mounds, whereas others argue they are self-organising systems built up by dust deposits. In the Bontveld landscape of the Eastern Cape, driven by geological features (cracks in calcrete) we have an opportunity to test the level of dust input in driving heuweltjies by testing changes in mound height in a system that clearly is not termite-originated.