Movement, home range and habitat use in leopard tortoises (Stigmochelys pardalis) on commercial farmland in the semi-arid Karoo.

Abstract

Given the ever-increasing demand for resources due to an increasing human population, vast ranges of natural areas have undergone land use change, either due to urbanisation or production and exploitation of resources. In the semi-arid Karoo of southern Africa, natural lands have been converted to private commercial farmland, reducing habitat available for wildlife. Furthermore, conversion of land to energy production is increasing, with areas affected by the introduction of wind energy, solar energy, or hydraulic fracturing. Such widespread changes affects a wide range of animal and plant communities. Southern Africa hosts the highest diversity of tortoises (Family: Testudinidae), with up to 18 species present in sub-Saharan Africa, and 13 species within the borders of South Africa alone. Diversity culminates in the Karoo, whereby up to five species occur. Tortoises throughout the world are undergoing a crisis, with at least 80 % of the world’s species listed at ‘Vulnerable’ or above. Given the importance of many tortoise species to their environments and ecosystems—tortoises are important seed dispersers, whilst some species produce burrows used by numerous other taxa—comparatively little is known about certain aspects relating to their ecology: for example spatial ecology, habitat use and activity patterns. Understanding an animal’s use of an environment is important in learning more about certain ecosystem functions and offering information to guide future conservation management. We studied spatial ecology and habitat use of the Leopard Tortoise, Stigmochelys pardalis: the largest and most abundant species in the region. Ten Global Positioning System (GPS) transmitters were placed eleven adult tortoises (one was redeployed following death of one individual), providing a minimum of 12 months of bihourly movement data. We used these data to estimate home ranges, indicate important predictor variables to movement, and investigate differences in space use between seasons, sex, and time of day. Using modern home range estimation techniques—such as GPS telemetry and Kernel Density Estimation (KDE)—we provided evidence that Leopard Tortoises have very large home ranges (n = 9, mean ± SE: 121.86 ± 28.12 ha, range 40.53—258.52 ha), with no significant differences between males and females. This large home range size supports previous research in the region, and supports the theory that populations in more arid regions have larger home ranges than in regions of higher rainfall. However, we also found that some Leopard Tortoises do not hold a home range: site fidelity tests suggested that two individuals exhibited apparent nomadic behaviour (hence home range estimated for only 9 of 11 tortoises above). When investigating seasonal changes in home ranges with generalised linear mixed models (GLMMs), we found important individual (sex and body mass), and weather (temperature and rainfall) predictor variables. We also used GLMMs to investigate bihourly and daily movement in Leopard Tortoises. Several important predictor models were identified, including temperature, rainfall, habitat type, availability of water, time of year, and time of day. We found a negative association for movement with distance from water sources, indicating that tortoises are more likely to move larger distances when closer to these resources. We attributed this behaviour to tortoises’ ability to supplement much of their water intake from water-rich food resources (e.g. succulents, grasses and forbs), as shown in previous studies. In contrast, as tortoises are still required to drink water to maintain water balance and excrete electrolytes and nitrogenous wastes, tortoises that have knowledge of drinking water resources within their home range are likely to make regular long distance movements to these areas. Movement data suggests that some Leopard Tortoises make nocturnal movements, despite tortoises being strongly diurnal animals. As adult Leopard Tortoises are large enough to avoid predation, are subject to reasonable night-time temperatures, and are able to maintain core temperatures above ambient temperatures, visibility of surroundings may be the largest limitation to movement. Further research is required, but we found higher nocturnal movement associated with periods of higher lunar illumination: e.g. full moon phase vs new moon phase. As stated above, there is a great importance in learning more about tortoises and contributing to conservation. One of these areas is improving genetic contributions that assists in identifying species. However, previous genetic research using a common mitochondrial primer site—cytochrome c oxidase I (COI)—has shown poor success rates with respect to tortoises. Using all tortoise COI sequence information available on the Barcode of Life Database (BOLD), and six of our own Leopard Tortoise samples, we recommend primer sites for the production of a mini-barcode specific to tortoises. Such a mini-barcode can be used to improve success rates in identifying specimens based on DNA, and increase extraction success with degraded DNA: e.g. museum specimens or environmental DNA. Information from this study can be used to further understanding of environmental and weather conditions that influence movement and space use in tortoises. We have identified several important resources and predictor variables which can affect how a tortoise uses its environment. Given that environments continue to be fragmented, degraded, or lost, better understanding of potential impacts on tortoises is required. We make recommendations on future research into fracking in the region, as well recommendations for use of electric fencing, which has been shown to cause large numbers of mortalities in Leopard Tortoises.