The ecology and population dynamics of the Nile crocodile Crocodylus niloticus in the Flag Boshielo Dam, Mpumalanga Province, South Africa

by Botha, Petrus Johannes.

During the period January 2002 to November 2003 surveys were made of the Nile crocodile population in the Flag Boshielo Dam, Mpumalanga. The aim of these surveys was to estimate the size of the population and to determine the population structure based on size. The double-survey method was used to estimate the number of individual crocodiles in the population. The density in terms of biomass was compared to results of other studies on Nile crocodile populations in Africa. The size of every crocodile observed during spotlight counts and aerial surveys were estimated. This was used to group individual animals together in predetermined size classes. The number of crocodiles in each size class was expressed as a percentage of the total and used to establish the population structure. ii University of Pretoria etd – Botha, P J (2005) The entire shoreline of the Flag Boshielo Dam was divided into 14 segments using easily recognisable reference points. The number of times that crocodiles were observed per size class in each segment was entered into an Excel spreadsheet. The spreadsheet functions allowed for the number of crocodiles observed per segment to be expressed as a percentage of the total number of animals seen. The GPS locations of every crocodile counted during night-counts were plotted on a GIS map of the Flag Boshielo Dam. Maps showing the dispersal patterns of crocodiles in the Flag Boshielo Dam could thus be compiled. The number of times that crocodiles were observed per size class in each segment of shoreline over the total study period was noted and entered into a spreadsheet and grouped together by size class. Once grouped in this way, the frequency at which crocodiles were observed per segment of shoreline were determined and expressed as a percentage of the total number of animals seen, per size class, over the study period in that particular area. The data contained in the spreadsheet was manipulated to group together data of the colder months (May - August) and to group data together from the warmer months (January - April and September - December). The extent of any seasonal variation in dispersal of crocodiles in the Flag Boshielo Dam was determined by comparing the sets of data. A total of 13 crocodiles were captured and VHF transmitters attached to their tail scutes. A further 2 crocodiles were captured and had GPS/GSM transmitters attached to the nuchal scales directly behind their heads. Radio tracking of these animals was done to establish their daily movements and to try and evaluate the size of their territories. The distribution data of the radio-tagged crocodiles were plotted and analysed with the ArcView 3.2 GIS software package. Range use analysis was iii University of Pretoria etd – Botha, P J (2005) achieved by using the Spatial Analyst Extension of ArcView 3.2 together with the Animal Movement Analysis extension. Nest surveys were done each year from 2000 to 2004. Attempts were made to measure a number of environmental parameters at each identified nest. These included the estimated total length of the nesting female, the length of the hindfoot print (HF), the distance from the nest to water, the height of the nest above the water, the incline of the shoreline, estimated exposure of the nest to sunlight, height and species of vegetation near the nest and the type of substrate. iv University of Pretoria etd – Botha, P J (2005)