Botswana Roan Antelope Ecology Project

The main aim of this study is to determine the home range use of various roan antelope herds occurring in the Linyanti Concession in northern Botswana, where they appear to be thriving. Together with home range, a number of other aspects of roan antelope ecology will be studied, including habitat use within the home range, the seasonal variation in nutritional status of roan populations and how predation and competition effects the movement of roan herds in the area.

Region: Linyanti Concession, Botswana
Organization: University of Pretoria, South Africa
Qualifications:
BScEnvironmental Sciences
BSc (Hons) Zoology

Roan antelope are considered one of the locally endangered mammal herbivores of southern Africa. Studies on roan populations that appear to be stable or increasing are thus vital for providing important information for use in areas where they appear to be declining. Home range-extent and habitat use are two of the main features that researchers focus on to gain vital information regarding an animal’s population status and behaviour.

Very little is known about the precise density, movement patterns and habitat use of roan antelope in northern Botswana. The results of this study will potentially provide valuable information regarding their general ecology and behaviour and ultimately help to gain a better understanding of their movement patterns and habitat selection. This information may aid in the proper management and conservation of this antelope species.

• To determine the density of roan antelope in northern Botswana.
• To determine the home range of different roan antelope herds.
• To create a vegetation map indicating the different habitat types present within each home range.
• To determine the habitat use of different roan antelope herds within their home range and how it may vary seasonally.
• To identify the habitat preferences of roan antelope within their home ranges and estimate the amount of suitable habitat available within the study area.
• To identify how habitat preference changes spatially across the various seasons.
• To relate roan antelope habitat use to predator movement within the study area.
• To determine the extent of competition between roan antelope and other herbivores across the seasons within the study area.
• To determine how the nutritional status of roan antelope changes seasonally.

Density of roan antelope in the area will be measured by conducting line transect counts, mark-recapture techniques using photographs of individual antelope and aerial transects. GPS collars will be used in this study, recording every hour, with roan herds being tracked on a daily basis. Only adult females will be collared due to their important role in leading the breeding groups in their daily activities. Roan herds are known to graze intensively between the periods of 09:00 and 12:00 and 16:00 and 17:00, with a rest period between 12:00 and 16:00. Coordinates recorded by the collars during these times will be considered as the foraging and rest sites used by the roan antelope herds.

A detailed vegetation map of the area will be constructed to indicate the variety of grass, shrub, and tree species occurring in the different habitat types. Knowing what plant species occur in the habitats will help determine the important characteristics that influence the roan antelope’s choice of locality.

Seasonal changes in the nutritional status (crude faecal protein and phosphorus levels) of roan will also be determined. Fresh faecal samples will be collected from foraging and resting sites and analysed.

Field work began towards the end of 2010, to continue until the beginning of 2012, thus ensuring that the two main seasons – the dry season (May – October) and the wet season (November to April) – are incorporated into the study for seasonal comparison. An MSc thesis will be written up amongst other published works.

April 2013
The abundance chapter has been very successful and preliminary results have estimated the population size at 80 individuals in the Linyanti. To date over 60 different roan individuals have been identified in both concessions, indicating that there may be a higher abundance of the species in northern Botswana than previously thought. The home ranges of the herds in the Linyanti area were much larger (72 sq. km, 82 sq. km and 71 sq. km respectively) compared to the Abu herd (32 sq. km).

Mopane (Colophospermum mopane) and Kalahari apple leaf (Philenoptera nelsii) woodlands were the preferred habitats chosen by the herds in Linyanti; the abundance of these habitat types throughout the area would indicate that the roan should thrive here. The Abu herd is providing valuable comparison data as they occur in a completely different environment – that of silver cluster-leaf (Terminalia sericea) island belts and floodplains.

Analyses of scat samples to determine the foraging ecology of the roan have begun.

November 2012
Following the collaring of three females last year in the Linyanti Concession, the home ranges of two herds were large (72 square km and 82 square km) over the duration of the year. Unfortunately the third collar failed and we only managed to record the wet season home range which was 72 square km. The two herds that recorded a full year of spatial data did show a seasonal difference in home range extent, utilizing a smaller range in the wet season compared to the dry season. Mopane (Colophospermum mopane) and Kalahari apple-leaf (Philenoptera nelsii) woodlands, especially the boundary (ecotone) between the two, were the preferred habitats chosen by the herds. The abundance of this habitat type throughout the concession indicates that roan antelope should thrive in this area.

The third satellite collar was deployed on a female in the Abu Concession in May 2012 and the herd is currently utilizing an area of approximately 31 square km. The Abu herd is providing valuable comparison data with the Linyanti herds as they occur in a completely different environment. The roan antelope alternate between using the silver cluster-leaf (Terminalia sericea) island belts and the floodplains during foraging, usually returning to the Terminalia belts at night to rest. In April/May the herd spent the majority of the day feeding knee-deep in water in the floodplains. During these feeding bouts the majority of the herd would feed on both sedge and grass species just above the water level; however three adults were observed submerging their entire heads under water to feed on the grasses below. Another unusual foraging behaviour observed in this particular herd, which has never before been recorded, is them feeding on the fruits of the sausage tree (Kigelia africana). While foraging, the Abu herd was observed walking purposely from one sausage tree to the next to feed on the fallen fruits and would usually spend an average of ten minutes chewing the fruits before swallowing them. This behaviour indicates that roan antelope foraging ecology may differ markedly between different regions of their range, with associated differences in other aspects of their ecology. The collars will remain on the female until May 2013 allowing us to record a full year of spatial movement.

Analysis of scat samples collected throughout the duration of the project will commence in 2013 to determine the foraging ecology of roan in both NG15 and NG26.

October 2011
The project on Roan antelope in the Linyanti concession started in December 2010. Three adult female roan antelope have been successfully collared (Figure 1, 2 and 3) to date and we are hoping to deploy another two satellite collars on roan herds before the end of the year. The first collared female occurs in the Savuti area of NG15 and is part of a herd of five individuals. The second collared individual occurs in the east of Kings Pool camp in the vicinity of the Linyanti (BDF) airstrip and forms part of a herd of nine individuals. The third collared female has only recently been collared, and we are therefore uncertain of the exact number of the individuals in her herd.

While roan antelope are still very wary of vehicles, during the last ten months that the project has been running we have started to build up a good database containing information about the movement patterns, habitat use and feeding data of this antelope species. Due to the skittish behaviour of this antelope species the collared females are only followed once a week. This is to ensure that the collars only capture the natural movement patterns of the females and not data influenced by the presence of a vehicle. Once GPS points have been downloaded from the collars, we randomly select several points where we conduct vegetation plots. This allows us to obtain a good representation of the grass composition and habitat types of the areas utilised by roan. In the last three months (August, September, October), roan antelope have shifted from being strict grazers to mixed feeders, including three woody plant species in their diets.

A good collection of scat has been obtained from roan throughout the concession. This, together with direct feeding observation, will give us a good indication of the seasonal changes in roan foraging ecology. We have recently started to create an extensive database of roan antelope occurring within the concession based on facial photography.These photographs, taken opportunistically whenever roan are sighted, will be collated into distinct sampling events and used in a mark-recapture analytical framework to estimate population abundance of roan. Dragging of transects along the cutline and around various pans throughout the concession for track and scat counts of roan antelope is conducted on a weekly basis and will be used as indices of population abundance together with the facial photography method.The scats and tracks of other large mammalian herbivores will similarly be counted during the scheduled sampling events to determine relative population sizes of potential competitors.

Annual Report 2011
The main aim of this study is to determine the home range use of various roan antelope herds occurring in the Linyanti Concession in northern Botswana, where they appear to be thriving. Together with home range, a number of other aspects of roan antelope ecology are being studied including habitat use within the home range, the seasonal variation in nutritional status of roan populations and how predation and competition effects the movement of roan herds in the area.

Fieldwork began towards the end of 2010, to continue until the beginning of 2012, thus ensuring that the two main seasons – the dry season (May – October) and the wet season (November to April) – are incorporated into the study for seasonal comparison.

Roan antelope were formerly one of the widest‐ranging antelope species in Africa occurring throughout most parts of the northern and southern savannahs. Today, although they appear to be widely distributed throughout Africa, only populations in West Africa appear to be thriving and locally abundant. Over the past decades, roan antelope numbers south of the Zambezi have been drastically reduced due to poaching, habitat destruction and predation and continue to remain threatened with extinction up until today. Currently, roan populations are confined to surviving in and around protected areas and in other areas that have low densities of people and livestock. Even though the status of roan antelope falls under the category of least concern, they are still considered one of the rare and locally endangered mammal herbivores of southern Africa.