Using photo identification to estimate the population size of nyala (tragelaphus angasii) in Umkhuzi Game Reserve

Nakale, Gisbertus Shanyengange
Journal Title
Journal ISSN
Volume Title
Wildlife conservation and management requires an in-depth understanding of the demographics and dynamics of the population concerned to enable sound management decisions. Yet it is often very challenging to obtain reliable information of cryptic and highly migratory species. I used photo identification and capture-recapture methods to study the population of nyala, a highly secretive species, in the Umkhuzi game reserve. The nyala species is individually identifiable by the mark pattern on their body. Photographs used for this project were taken at a waterhole in Umkhuzi game reserve between June 23rd and 01 July 2014, representing nine sampling occasions. Identification of individuals was achieved with a computer-assisted technique using open source pattern identification software called Wild-ID version 1.0.1. A total of 652 photographs were taken at the sampling site and from these pictures wild-ID identified 372 distinct animals. An encounter history for each individual during the nine sampling occasions was also produced using Wild-ID. The encounter histories of all individuals were entered into Program MARK. I used the closed population models in Program MARK to obtain population estimates. Program MARK indicated that Mh was the most appropriate model to fit this data as indicated by the AICc ranking. Mh showed population estimates as follows: adult males: 111.90 ± 16.07, adult females: 298.01 ± 36.66; young adult males: 21.33 ±19.34; juvenile males: 37.15 ±16.84; yearling males: 37.73 ±8.51; yearling females: 96.48 ±22.75 and juvenile unidentified: 69.03. ±28.96 Closure test performed to ascertain demographic and geographic closure during the sampling period showed a χ2= 21.74, p= 0.08, df = 14, for the Stanley & Burnham test and a p and z-values of 0.06 and -1.51 respectively for the Otis et al test. These results shows marginal violation of population closure, nevertheless closed population models were used to estimate population abundance due to the fact the violations are marginal and the sampling period iii was very short, nine days. The study revealed that there is as much as twice the number of females compared to males.
A research report submitted to the Faculty of Science, in partial fulfilment of the requirements for the degree of Master of Science in Environmental Science, University of the Witwatersrand, Johannesburg, 2015.