Researcher: Jeff Muntifering and Dr.Tara HarrisRegions: Namiba & South Africa
BackgroundHartmann’s mountain zebra (Equus zebra hartmannae) are a range restricted, lesser known zebra species, inhabiting the dry and mountainous regions of Namibia and South Africa. They are listed as Vulnerable on the Red Data List as they number only about 9 000 mature individuals and their populations are thought to be declining overall. While numerous scientific publications have focused on the Cape mountain zebra in South Africa, very little is known about Hartmann’s mountain zebra, which is almost exclusively found in Namibia.
A pilot for this project was carried out in 2009, whereby project coordinators identified and met with various individuals within organisations/institutions crucial to future collaboration. The team travelled through the Kunene Region to collect data about size, composition, and location (i.e., GPS coordinates) of mountain zebra groups, including breeding herds, bachelor groups and individuals.
ObjectivesThe overall aim and long term objective is to identify key conservation areas for Hartmann’s mountain zebras in Namibia and to work with the Namibian government, local people and conservation NGOs to craft a regional conservation action plan for this subspecies to better direct land use and management practices.
Information about mountain zebra seasonal movement patterns particularly in relation to human-induced pressures (trophy hunting and subsistence meat) and critical periods (droughts and early rains), association with different habitat-types, and gene flow would contribute data that will improve conservation strategies for this unique and valuable subspecies.
Investigate seasonal movement patterns of mountain zebra in north-western Namibia using telemetry data;
- Conduct a GIS-based regional and local-scale habitat suitability and capability assessment using observed Hartmann’s mountain zebra location data and mapped habitat metrics such as landscape/resource and human use spatial variables;
- Assess relationships between habitat suitability/capability and observed population performance;
- Characterise and map seasonal Hartmann’s mountain zebra core and corridor areas across the study area.
MethodologyIn order to adequately address our objectives stated above, we will be required to collect, assimilate and analyse both zebra movement data with respect to a suite of landscape, resource and human land use variables, as well as hematologic and serum biochemical parameters across the study area. We obtained spatial data for landscape, resource and human land use variables during our pilot study from a combination of assimilating available spatial data sets and filling observed gaps through targeted field surveys. However, the current level of zebra movement data is highly biased, both in terms of spatial and temporal sampling, and is not nearly sufficient for our proposed analysis.
The advent of GPS and satellite technology built into wildlife tracking devices (i.e. tags or collars) has revolutionized how scientists and managers are able to assess wildlife movement and make inferences concerning their conservation and management. This is particularly evident in species that are elusive and wide-ranging. The ability of such devices to log data at a much higher resolution allow much more accurate and precise inferences to be made regarding spatial and temporal space use dynamics such as patch-level resource selection and specific movement corridors. Such technology is perfectly suited to the presumed behaviour of Hartmann’s mountain zebra in Kunene (wide-ranging across an extremely rugged, remote landscape) while providing data of appropriate spatial and temporal resolution to adequately address our research questions. Thus we plan to deploy an additional 5-7 Satellite Collars to increase our sample to 8-10 (including the already deployed 3 GPS/UHF collars) in Focal Sites 1 & 2 (Appendix 3) during our second study phase. Collars will be programmed to obtain the most number of GPS fixes over a period of 2 years and will target breeding mares.
We will explore various relationships between zebra movement and various habitat metrics to identify; (1) key factors that drive zebra habitat selection at a local-scale, and (2) the level of variability within our sample population to assess how representative our current sample size is. We will assess habitat co-variates using a resource selection function (RSF) approach (Manly et. al., 2002; Johnson and Gillingham, 2005). This preliminary investigation will inform further collaring efforts (to increase our effective population sample) and will also identify where probable key core and corridor areas exist which will then enable us to target field excursions to characterize these key locations and their relative importance towards zebra population dynamics. We will incorporate these findings in proposals to acquire the necessary funds to acquire the needed collar type and frequency to adequately supplement our current study sample population.
Focal Survey Sampling
We will also continue conducting monthly observational surveys in each of our focal sites (Site 1 & 2) to collect data on demographic characteristics such as density of harems and bachelor herds, group size and structure (gender and age class), and number of mares with foals of the year. This work will be conducted in cooperation will local partners such as Save the Rhino Trust, and the Game Guards from Anabeb and Torra Conservancy.
We have submitted an annual update for inclusion in the Wilderness Wildlife Trust Annual report. Upon completion of the study, a full report documenting all activities specifically maps illustrating zebra collar data, and preliminary habitat use assessments, and a series of next steps will be produced and submitted to MET and the conservancy management committees in our study area.
In July 2010 the Minnesota Zoo, in partnership with Save the Rhino Trust, the Ministry of Environment and Tourism, and local communities, initiated the Habitat Assessment component of the Kunene Hartmann’s Zebra Programme. The overall aim is to identify key conservation areas (cores and corridors) for Kunene’s Hartmann’s zebra and to work with the Namibian government, local communities and conservation NGOs to craft a regional conservation action plan for this sub-population to inform land use and management practices. Our first field season focused primarily on piloting the deployment and use of GPS/UHF collars, and initiating a community-based monitoring programme with local game scouts.
The project team successfully deployed two GPS/UHF collars during the initial start-up phase. Using the remote UHF modem, we successfully downloaded over 3 000 locations from one GPS collar, programmed to record one location every 30 minutes, which was placed on a bachelor male in the rugged basalt escarpment mountains of Etendeka Concession. During a nine-week period, this individual surprisingly moved on average only about 5 kilometres per day across roughly 2 500 hectares venturing no further than 3km from water. As predicted, he has recently dispersed with the onset of the rains to the east. His next download, which may require aerial reconnaissance, will reveal his wet season whereabouts and transit pathways. We hope to deploy an additional 10 collars in 2011 now that we are confident in the system.
To initiate the community-based monitoring programme, a local Namibian was trained, who joined Save the Rhino Trust’s field teams, logging over 260km during eight patrols accumulating 344 Hartmann’s zebra observations across Palmwag and Etendeka concession areas. In the neighbouring Conservancy lands, we implemented an incentive-based monitoring programme that will provide a monetary payment to the Game Guards for the completion of targeted Hartamnn’s zebra monitoring forms during routine fixed-route patrols. The programme was initiated in November with Anabeb and Torra Conservancy Game Guards, committing to cover approximately 40 and 50 kilometres on foot each month, respectively.
April 2010 Update
The pilot project focuses on the Kunene Region of north-western Namibia, where several thousand mountain zebra are thought to exist. Based on insights from local experts, 630km of established track in two national parks, three concession areas, and two community conservancies was covered collecting data on zebra locations and group compositions. The locations of over 1200 mountain zebra were mapped and different methods for estimating relative population densities were tested. Preliminary results suggest that two of the ecotourism concessions (Hobatere and Etendeka) and western Etosha National Park had relatively high densities of the species in the 2009 dry season. Survey results also identified specific areas across the study region in which to focus future studies.
175 mountain zebra dung samples were collected from the seven study areas and the DNA from these samples will be used for genetic analyses. This will help understand whether mountain zebra in the Kunene Region belong to one large breeding population or whether they move around and breed within more localised areas. Genetic analyses on dung samples collected at permanent water springs will be analysed to better understand zebras’ use of these critical resources and to test whether dung counts at springs are useful for estimating local zebra population sizes.
The next phase will involve placing GPS and radio telemetry collars on zebra in different areas of the Kunene Region to track their seasonal movements and better understand the ecological and humaninduced factors that drive their habitat selection. Observational data will also be collected to document seasonal and yearly changes in mountain zebra distributions and densities across the study area.