|dc.description.abstract||Up until 1969 Nile Crocodiles (Crocodylus niloticus) were considered as vermin in South Africa and were actively persecuted throughout the country. In an effort to re-establish viable populations within protected areas in Zululand a restocking program was initiated in the late 1960‟s and early 1970‟s. Ndumo Game Reserve (NGR) in north eastern KwaZulu-Natal was one of the reserves involved in the restocking program and the Nile Crocodile population in the reserve increased from a minimum absolute abundance of 348 (SE ± 3.39; N = 3) in the early 1970‟s to a minimum absolute abundance of 992 (SE ± 58.70; N = 4) in the 1990‟s. However, in recent years there has been some concern that the NGR Nile Crocodile population may be on the decline, initiating the current investigation into the ecology and conservation of the NGR population.
We examined changes in relative abundance using aerial survey data from 1971 – 2009. The precision and accuracy of population estimates was affected by water level, season, aircraft type and the use of different observers. A correction factor was applied to survey data and the current NGR Nile Crocodile population is estimated at an absolute abundance of 846 (± 263).
Distribution data from the aerial surveys were also used to examine habitat use over the last 40 years and revealed that Nile Crocodiles were not evenly distributed in NGR and that crocodiles favoured the Phongola over the Usuthu floodplain systems. NGR is characterised by a floodplain mosaic landscape and crocodile distributions between the various habitat patches were influenced by landscape physiognomy and composition as well as connectivity and corridor quality. Anthropogenic disturbances influenced the functionality of the floodplain landscape negatively with impacts on habitat use and connectivity. To quantify the effects of environmental conditions on crocodile habitat use we conducted 40 diurnal counts at Lake Nyamithi between 2009 and 2012 and related changes in crocodile numbers here to temperature, rainfall and water level. Crocodile density in Lake Nyamithi was significantly and negatively related to average maximum ambient temperature and numbers increased in the lake over the cool, dry winter season. Water level and rainfall had strong but not significant (p >0.05) negative influence on crocodile density in Lake Nyamithi. Environmental variables influenced different size class of Nile Crocodiles differently and the density of crocodiles in the 1.5 – 2.5 m Total Length (TL) size class were significantly influenced by rainfall and average minimum monthly temperature.
Movement patterns of 49 Nile Crocodiles between 202 – 472 cm total length (TL) were followed over 18 months using mark-resight (n = 36), radio (n = 10) and satellite (n = 3) telemetry. The duration of radio transmitter attachment (131 days, SE ± 11.35) was significantly related to TL and reproductive status. Satellite transmitters stopped functioning after 15 (SE ± 12.53) days and home range was calculated for 7 crocodiles ranging in size from 202 cm TL – 358 cm TL. Sub-adults (1.5 - 2.5 m TL) occupied smaller, more localized home ranges than adults (> 2.5 m TL). Home ranges overlapped extensively suggesting that territoriality, if present, did not cause Nile Crocodiles to maintain spatially discrete home ranges in NGR during the dry season. A single large scale migration event occurs every year between October and November whereby the majority of the NGR crocodile population leaves the reserve and enters the Rio Maputo floodplain in adjacent Mozambique and only return in April/May.
Nesting effort (19 – 21 %) in NGR was comparable to other populations of Nile Crocodile in southern Africa. Nests are completely destroyed by floods once every 10 years and predation rates may range from 20 – 86 % per year. In addition to aerial surveys, nesting surveys and movement studies crocodiles (n = 103) were caught opportunistically to collect demographic data on population structure. The population structure of Nile Crocodiles in NGR is currently skewed towards sub-adults and adults suggesting an aging population that may decline naturally in the future. This could be due to low recruitment levels in NGR that are not able to sustain the artificially high population size created by the restocking program. Sex ratios were skewed towards females in the juvenile and sub-adult size classes and towards males in the adult size class while the overall sex ratio was even between males and females.
It is predicted that the NGR Nile Crocodile population will decline in the future and that this decline should be considered as a natural process. However, the rate of decline will be accelerated at an unnatural speed and to an unnatural extent due to poaching, uncontrolled harvesting and destruction of nesting habitat within NGR. Based on the findings of the current study, management recommendations for the conservation of the combined NGR – Rio Maputo Nile Crocodile population were made. It is important that further research takes place in the Rio Maputo floodplain in Mozambique to better quantify the nesting ecology of the NGR Nile Crocodile population and to identify possible threats facing Nile Crocodiles in this region.||en