Meet us


Professor Rudi van Aarde
Chair of Conservation Ecology and Director of CERU
I direct the Conservation Ecology Research Unit (CERU) as a self-sufficient research entity funded via grants from national and international organisations and private industries. These grants also support research fellows, support staff, and post-graduate bursaries. My research focuses predominantly on two lines of inquiry.
My research on elephants focuses on the drivers of demographic variability and heterogeneity in spatial utilization, with emphasis on finding solutions for the causes rather than symptoms of so called 'elephant problems'. This research is conducted across gradients of environmental and management conditions in southern African protected areas in Mozambique, Malawi, Zambia, Botswana, Namibia, and South Africa. CERU's development of the 'megaparks for metapopulations' concept is an innovative platform for elephant management that emphasizes spatial structuring of populations and demographic responses to heterogeneity in environmental resources. This management option has important implications for the continuing and controversial debate on elephant management.
My research on the restoration of coastal dune forest in northern KwaZulu-Natal (South Africa) started 21 years ago. Coastal dune forest is rare in South Africa and falls within the Maputaland-Pondoland-Albany biodiversity hotspot. Some 10% of this forest is protected, but 43% has been transformed. The remaining 57% is threatened by tourism, dune mining, and clearing for subsistence living. The narrow and linear nature of the forest further contributes to its sensitivity to transformation, isolation, and fragmentation, and its restoration makes both conservation and economic sense. The project is now well established and CERU's scientific achievements here provide a recognised scientific foundation for restoration ecology in southern Africa. View Professor van Aarde's CV

Dr Robert Guldemond
Research Fellow
The rehabilitation of disturbed sites to sustainable and functioning ecosystems is one way to mitigate our negative impact on the environment. However, successful restoration requires continuous research and monitoring. Our study site north of Richards Bay relies on natural processes such as succession and patch-dynamics to drive recovery of dune forest after mining. One component of our research focuses on community structure and composition of birds, small mammals, trees, herbs, millipedes, and dung beetles in regenerating dune forest during early, mid, and late successional stages, as well as in un-mined old growth forests. In my current work, I aim to fit different candidate statistical, niche-based, and neutral models to rank abundance distributions. Model selection routines then determine which candidate model, along with its particular set of assumptions, best explains the community structures of these six taxa. We may then conclude which ecological mechanisms are the most likely determinants of the community re-assembly process in dune forests.

Camilla Nørgaard
Research Assistant
CERU's development of a routine to extract demographic information from free-ranging elephants relies on getting accurate measures of the back lengths of elephants in breeding herds. We use digital images that we make during dedicated aerial sorties from a helicopter. I measure elephants' back lengths and convert these to age using elephant growth models that we have developed at CERU. This enables me to determine estimates for age at sexual maturity, calving intervals, and age structures for each sub-population we survey. This information is used to model age-specific survival and fecundity rates as well as the intrinsic growth rates of populations. Each year we do this for some five to six thousand elephants in five to six populations.

Glenda Varrie
Research Assistant
The old-growth and regenerating coastal dune forests along the coast of KwaZulu-Natal provide an 'outdoor laboratory' for testing a variety of ecological theories. I assist students and staff with their research projects in these forests. My particular focus is of a botanical nature, and I specialise in vegetation based surveys and assessments. I also assist in the maintenance and logistics relating to CERU's Richards Bay field station.

Alida de Flamingh
Research Assistant
Animal populations are structured spatially and genetically, often correspondingly. We investigate the genetic structure and gene flow of spatially distinct populations by means of landscape genetics. We aim to determine whether elephant populations are structured into genetically distinct groups, and if there is historical, contemporary and/or gender-mediated gene flow between national parks and between genetically distinct groups. Ultimately we aim to identify genetically functional landscape linkages that may be important for conservation incentives. I am responsible for the laboratory and analytical work associated with this project and manages CERU’s genetic database.

Pieter Olivier
Post-doctoral Fellow
Habitat loss apparently drives the current extinction crisis. The processes through which it affects biodiversity, however, are complex and poorly understood. This is especially true for spatially complex regions that comprise a mosaic of land-use types, which often range from protected areas to dense human settlements. To meet conservation targets or regain ecosystem services, it is important to evaluate patterns, processes, threats and opportunities that relate to biological diversity within such affected landscapes. For my PhD, I studied the influence of forest loss and fragmentation on tree and bird communities that occur within coastal forests in northern KwaZulu-Natal. My current research builds on this foundation and further explores how coastal forest biodiversity respond to land-use change, and how these responses may vary at different spatial scales. I specifically focus on understanding rarity, extinction risk and the role of matrix habitats that surround coastal forest fragments (e.g. sugarcane and agroforestry plantations versus natural grasslands and woodlands), in supporting regional diversity. I am also interested in the drivers of community assembly, biodiversity conservation and ecological restoration.

Victor Rolo Romero
Post-doctoral Fellow
Coastal dune systems are among the most damaged ecosystems by human activities worldwide. Restoration efforts in these systems often imply the reconstruction of historic abiotic conditions and relying on successional processes to recover the vegetation layer. However the predictability of successional processes is hindered by a suit of concurrent biotic and abiotic factors that develop in the degraded state. Current ecological theory argues that community assembly mechanisms can be inferred from the study of species functional traits, but little is known about its applicability in these novel systems.
My research interests are related to the study of the replacement of species over time in dune forest after a major disturbance. At present we do not know to what extent the replacement of species in coast dune forest is driven by senescent species that open gaps in the canopy when falling. Specifically, are gap dynamics driven by stochastic processes or, by contrast, by a combination of mechanisms ranging from environmental selection to ecological interaction? Can we use current ecological theories based on functional traits to predict the processes of community assemblage after gap formation? A better understanding of the processes that controls the replacement of species in highly human-disturbed environments is of major importance for prioritizing conservation efforts and optimizing restoration activities.


Gabriel Albano
PhD Student
Inhaca Island, located on the northern limit of the Maputaland ecoregion in southern Mozambique, was separated from the mainland about 7 000 years BP. This separation isolated once continuous habitats and may have created unique assemblages on the island. For my PhD, I investigate whether tree assemblages on Inhaca Island are unique when compared with those on the mainland. I also aim to unravel the processes that may drive the observed differences among island and mainland tree assemblages. Specifically, I ask the following questions: (i) Is the coastal dune forest assemblage on Inhaca similar/dissimilar to the forest assemblage on the mainland? (ii) Do community and/or environmental variables explain variation in community composition, structure and function of the coastal dune forest on Inhaca? (iii) What effect has isolation had on the community composition, structure and functioning of the coastal dune forest on Inhaca? This study aims to provides ecological guidelines for the management of small, isolated coastal forest fragments in southern Mozambique.

Andrew Purdon
MSc Student
The potential deleterious effects of elephants on vegetation and management options to mitigate these effects have long been debated. Kruger National Park management recently opened fences, closed waterholes and stopped culling to promote spatial structuring of the elephant population, recognising that spatio-temporal heterogeneity in resources drives changes in elephant spatial use and demography. Spatial structuring may allow the elephant population to be governed by metapopulation dynamics permitting self-regulation. Additionally heterogeneity in spatial utilization of elephants potentially mediates the impact of elephants on other species. I aim to disentangle the multi-scale causes and mechanisms driving elephant movement patterns as well as subsequent consequences for biodiversity using location time series data combined with habitat and climate information. Linking these results with repeated Rapid Elephant Population Assessments of focal herds will allow us to determine how density, spatio-temporal heterogeneity in resources, and spatial utilization influence reproductive success. This research could aid in future adaptive management decisions to achieve the desired outcome of maintaining biodiversity.

Ashley Robson
MSc Student
Elephants are influenced by both extrinsic and intrinsic factors. Variations in resource availability and population density over time and space induce responses that can be assessed at the population level, allowing insight into the processes and mechanisms that drive elephant population dynamics. Kruger National Park has employed a number of elephant management approaches over the last 40 years, from culling and water provisioning to the recently introduced “metapopulation management” paradigm. This has resulted in a unique set of elephant population time-series data that span a population density and resource heterogeneity gradient. Using this data I aim to better understand the role extrinsic and intrinsic factors play in temporally and spatially structuring elephant populations by examining the effects of climatic fluctuations, resource availability and heterogeneity, and density on elephant population dynamics. The results of my research will contribute to the on-going evaluation of the “metapopulation management” approach.

Oscar Mohale
MSc Student
Gap dynamics, the natural creation of openings in the forest canopy and the trees that fill those gaps, play a major role in regenerating forests. Succession can lead towards a mature forest if secondary or climax species replace pioneer species in canopy gaps. However, theory dictates that large canopy gaps can arrest succession if no species replacement occurs or if pioneer species are self-replacing. Post-mining rehabilitation of dune forests north of Richards Bay initiates development of single cohort, monospecific stands of Acacia karroo. Previous research showed that other shade-intolerant species replace A. karroo within gaps, while A. karroo recruitment is poor, thus suggesting that regenerating coastal dune forest will diversify towards mature forest.
I am using an experimental approach to assess proposed mechanisms to explain replacement of A. karroo by secondary species. Higher seedling mortality of A. karroo partly explains species replacement in gaps. Proposed mechanisms include competition with the pre-existing understory vegetation, selective seedling herbivory that results in differential losses, and seedling inhibition by increased nitrogen levels. Understanding gap dynamics will allow us to better predict future trajectories of species composition.

Marc Freeman
Honours Student
Forests along the eastern coast of southern Africa have been naturally fragmented over time as a result of climate change and other stochastic events. In response to this, many forest dependent bird species may have become adapted to live within a fragmented landscape, making the composition of the surrounding matrix important for their survival. However, in recent times these matrices have been transformed – sugarcane and agroforestry plantations now surround forest fragments instead of natural grasslands and woodlands. For my Honours project, I therefore aim to determine whether the anthropogenic transformation of matrix habitats will alter the structure of coastal forest bird communities.

Kelly Green
Honours Student
My honours project aim to determine if the fine-scale movement of savanna elephants is a function of ambient temperature. I specifically aim to determine the temporal resolution where we can record changes in elephant movement patterns due to changes in ambient temperature. This will allow us to identify the optimum temporal resolution to track elephants, and provide insights into how these animals may respond to future climate change.

Puseletso Motsomane
Honours Student
Relying on a single ‘benchmark site’ while ignoring nearby old-growth forests to assess the efficiency of post-mining restoration efforts may be problematic. This approach assumes that regeneration develops in the same direction across sites, while we know that connectivity between source (i.e. undisturbed, old-growth forest fragments) and rehabilitating sites may influence community composition due to limitations imposed on species dispersal. A single benchmark approach also assumes that regenerating sites are colonised from only one source, thereby overlooking the many sources of potential colonisers scattered across the landscape. My honours project focus on the heterogeneous landscape that surround regenerating forests and aim to determine if the rate of regeneration across sites is a function of distance to source areas.

Elena Reljic
Honours Student
Assessing the patterns of species dissimilarity among rehabilitating communities is an important step in the evaluation of restoration outcomes. Indices of beta diversity can be used to measure dissimilarity among ecological communities and the changes in community composition across a region or along an environmental gradient. For my honours project, I aim to determine patterns of tree and bird community dissimilarity along a chronosequence representing 38 years of dune forest recovery. I also aim to infer whether local or regional processes give rise to the observed patterns of dissimilarity across space and time.

Michael Voysey
Honours Student
The manipulation of available surface water to reflect its natural distribution has been suggested as the most promising tool that managers have at their disposal to control elephant impacts on other species. The spatial arrangement of surface water in the landscape imposes a major limitation on elephant distribution patterns and ultimately has consequences for demography. Through my honours project I aim to contribute a better understanding to how elephants adjust their movement patterns in response to spatial and temporal surface water availability by studying focal elephant herds in northern Botswana. The results of this research could enhance the basis on which science-based decisions are made regarding water management practices that aim to spatially structure elephant populations.