MSc., Faculty of Science, University of the Witwatersrand, 2011 / Abundance counts of ungulate species which are carried out using an aerial census are susceptible to measurement errors. These measurement errors result from environmental factors such as vegetation cover and the resultant effect that factors, such as the annual rainfall pattern, have on the accuracy of the count. Given these measurement errors, models of population abundance that are based on these counts should not only account for the population dynamic process - they need to model the measurement error process simultaneously in order to produce plausible estimates. In addition to this, the recorded counts from these censuses do not give a breakdown of the total count into the different gender and life-history stages of the animals counted. This research report investigates the usefulness of a hierarchical Bayesian hidden process modelling approach at explicitly including the measurement error process and a sub-model for the gender and life-history stage of the animals counted into a population dynamics model based on the aerial census counts. The data used are aerial counts of Kudu in the Kruger National Park from 1983 to 1993. The result is a model which estimates the measurement error in each year of the census and also breaks down the overall count into the numbers in each gender and life-history stage. The usefulness of the model is evaluated based on statistical model diagnostics.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/10981 |
| Date | 16 January 2012 |
| Creators | Gatawa, Tatenda |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | application/pdf |
Page generated in 0.0018 seconds