Early Post-Release Movements, Prey Preference and Habitat Selection of Reintroduced Cheetah (Acinonyx jubatus) in Liwonde National Park, Malawi

Sievert, Olivia (2020-03)

Thesis (MScConsEcol)--Stellenbosch University, 2020.

Thesis

ENGLISH ABSTRACT: Once widespread throughout Africa and southwestern Asia, the cheetah has disappeared from the majority of its historical range, making it Africa’s most endangered large felid. Scenario modelling has demonstrated the survival of the cheetah is highly dependent on protected areas and woodland habitats. Reintroduction into protected areas of recoverable range has the potential to assist in the conservation of the species. However, sizeable knowledge gaps regarding the behavioural ecology of this species within its historical range remain and must be filled to assist in reintroduction success. In 2017, African Parks in partnership with the Malawi Department of National Parks and Wildlife and the Endangered Wildlife Trust reintroduced seven cheetah into Malawi after a 20-year extirpation. This study aimed to provide an overview of the post-release movements, settlement and behavioural ecology of these reintroduced cheetahs to inform future pre- and post-release management techniques, long-term population management and assist in identifying other reintroduction sites in the country. Post-release movements were assessed using data collected from five GPS collared founder individuals who were tracked for two years after their release into Liwonde National Park (LNP). Pre-release holding periods greater than 23 days were shown to not affect post-release movements. All cheetah demonstrated release site fidelity; however, males experienced more extensive post-release movements and settled later than females. Reintroduction success was defined for both the individual and the population level. An individual success rate of 57 % was recorded (80 % for GPS collared animals). All females birthed their first litter within four months post-release and, within two years, the population began to conformto demography levels documented in the source population. Therefore, the overall reintroduction was considered successful. Using scat analysis and carcass observations, 13 prey species were recorded. Cheetah showed the highest preference for greater kudu (Tragelaphus strepsiceros) when considering prey populations. Four species comprised the bulk of cheetah diet, namely; kudu, impala (Aepyceros melampus), waterbuck (Kobus ellipsiprymnus) and bushbuck (Tragelaphus sylvaticus), all of which experienced asymmetric predation across their demography. Asymmetric predation, coupled with increasing predator densities in the park, may have long-term implications for the demography of certain prey species. The spatial distribution of GPS collared cheetah appeared restricted, and individuals experienced high levels of both home range (95 % isopleths) and core area (50 % isopleths) overlap. All cheetah lacked exclusivity of both their home range and core areas (>10 % overlap). Intrasexual overlap in females may be indicative of den site selection. The high overlap of females may have long-term implications on both cub and adult male survival. Cheetah used all habitat types in LNP. However, females selected for open woodland habitat with moderate prey frequency of occurrence within their home range. All cheetah demonstrated a preference towards open floodplains with high prey frequency of occurrence for kill sites. The success of this reintroduction is encouraging for continued range expansion of cheetah in Malawi. However, given the small founder population and low two-year recruitment rate, this population still requires intensive management. Genetic supplementation should be implemented to maintain genetic diversity. It is recommended that a metapopulation node for cheetah in Malawi is developed to assist in the long-term management of this population. It is further recommended that additional research into the effects of intraguild competition with cheetah in LNP is conducted once the full carnivore guild is restored.

AFRIKAANSE OPSOMMING: Die jagluiperd was voorheen wydverspreid oor Afrika en Suidwes-Asië, maar het grootliks verdwyn vanuit hul historiese verspreidingsgebied. Die gevolge is dat die jagluiperd vandag die mees bedreigde groot katsoort op die Afrika-kontinent is. Scenario-modellering het getoon dat die oorlewing van die jagluiperd hoogs afhanklik is van beskermde areas en bosveldhabitat. Hervestiging in beskermde areas van herkrygbare verspreidingsgebiede het groot potensiaal om die bewaring van dié spesie te ondersteun. Dit is egter nodig om kennisgapings aangaande die jagluiperd se gedragsekologie in die historiese verspreidingsgebied aan te vul om suksesvolle hervestiging in die hand te werk. In 2017 het African Parks, in samewerking met die Malawi Departement van Nasionale Parke en Natuurlewe en die Trust vir Bedreigde Natuurlewe (EWT), sewe jagluiperds ná ’n 20-jaar lange afwesigheid in Malawi hervestig. Hierdie studie was gemik daarop om ’n oorsig te bied rakende jagluiperds se aktiwiteitspatrone, vestiging en gedragsekologie na vrylating, om toekomstige voor- en na-vrylatingsbestuurstegnieke en langtermyn populasiebestuur toe te lig, en om ander hervestigingsgebiede in die land te identifiseer. Na-vrylatingsaktiwiteitspatrone was geassesseer deur verspreidingsdata vir vyf individue met GPS-halsbande oor ‘n periode van twee jaar na vrylating in die Liwonde Nasionale Park (LNP) in te samel. Voor-vrylatingswagperiodes langer as 23 dae het nie na-vrylatingsaktiwiteitspatrone beïnvloed nie. Al die jagluiperds het in die omgewing van die vrylatingsgebied gebly. Mannetjies het egter uitgebreide na-vrylatingsaktiwiteitspatrone getoon, en hulself later as die wyfies in ’n tuisgebied gevestig. Hervestigingssukses was omskryf op die individuele- sowel as die populasievlak. ’n Individuele suksessyfer van 57 % was aangeteken (80% vir jagluiperds met halsbande). Binne vier maande na vrylating het alle wyfies hul eerste werpsel gehad, en binne twee jaar was die populasie-demografievlakke soortgelyk aan dié wat in die bronpopulasie aangeteken is. Gevolgtelik word die hervestiging in sy geheel as ’n sukses beskou. Dertien prooispesies is met behulp van misanalise en karkasobservasie geïdentifiseer. Jagluiperddieet het hoofsaaklik uit vier prooispesies bestaan: koedoe (Tragelaphus strepsiceros), impala (Aepyceros melampus), waterbok (Kobus ellipsiprymnus) en bosbok (Tragelaphus sylvaticus), met asimmetriese predasie oor elke prooispesie se demografie en ’n sterk dieetvoorkeur vir koedoe. Asimmetriese predasie tesame met toenemende roofdierdigtheid in die park kan langtermyn nagevolge inhou vir die demografie van sekere prooispesies. Die ruimtelike verspreiding van jagluiperds met GPS-halsbande blyk om beperk te wees, en individue het hoë vlakke van tuisgebied- (95% isoplete) en kernarea-oorvleueling (50% isoplete) ervaar. Alle jagluiperds het ’n gebrek aan eksklusiwiteit in die tuisgebied en kernareas ervaar (>10 % oorvleueling). Hoë ruimetlike oorvleueling by wyfies kan aanduidend wees van lêplekseleksie en kan langtermyn nagevolge vir die oorlewing van welpies en volwasse mannetjies inhou. Die jagluiperds het alle habitatsoorte in LNP gebruik, maar wyfies het egter oop bosveld met matige prooifrekwensie in hul tuisgebied verkies. Alle jagluiperds het ’n voorkeur getoon vir oop vloedvlaktes met hoë prooifrekwensie. Hoë hervestigingssukses is aanmoedigend vir die voortdurende tuisgebiedsuitbreiding van jagluiperds in Malawi. Gegewe die klein stigterspopulasie en lae twee-jaar aanwinskoers, benodig hierdie populasie egter steeds intensiewe bestuur. Genetiese aanvulling moet geïmplementeer word om genetiese diversiteit te onderhou. Dit word aanbeveel om ’n metapopulasienodus vir jagluiperds in Malawi te ontwikkel om langtermynbestuur van dié populasie aan te vul. Dit word voorts aanbeveel dat addisionele navorsing rakende die effekte van interspesie-kompetisie met jagluiperds in LNP gedoen moet word soos wat algehele roofdiergetalle toeneem.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/108351
This item appears in the following collections: