Species distribution model for the Roan antelope (Hippotragus equinus) in the West African savanna

Thalia Mathilde Pelegrin

Master's thesis

Abstract:

With constantly developing technologies, conservation of species has increasingly become a multidisciplinary approach, often using both on-the-field and computer-based methods to expand our understanding of different aspects such as behaviour, distribution, threats faced, and more. Species distribution modelling is one of these techniques that has become frequently used as it enables conservationists to comprehend where species occur, what drives them, and how this may change in the future. Within this discipline, several different methods have been developed, allowing the use of various predictors from a wide array of global databases. In this thesis, the distribution of the Roan antelope was studied in Niokolo-Koba National Park situated in Senegal, using various environmental variables. These were chosen based on knowledge of species ecology and included elevation, slope, distance to river and marshes, normalized difference vegetation index and its standard deviation, above ground woody biomass, mean annual temperature, temperature seasonality, maximum temperature of the warmest month, annual precipitation, and precipitation seasonality. Species data was collected from camera traps located within the centre of the park and used to create various species distribution models. Statistical evaluation and selection of these models was then performed to determine the most robust model with a good fit. This model (containing above ground woody biomass as a predictor of occupancy) was then used to extrapolate the probability of presence of Roan both in the park and a smaller area of interest within its borders. The model outputs predicted a high probability of Roan presence almost everywhere across the park, consistent with results expected from a generalist antelope with a high level of adaptability for a broad range of ecological conditions. However, in the absence of data available from the borders of the park, where other factors such as anthropogenic influence may affect predictions, precautions must be made when interpreting model results. Future studies to further increase our knowledge on Roan distribution would benefit from including more data sources of different factors, comprising ideally of multi seasonal data, as well as expanding the area of study to include a wider range of habitats.

Keywords

Environmental modelling, Large herbivore distribution, Roan antelope, Senegal, Species distribution modelling

Language used: English

Date on which the thesis was submitted / produced: 8. 8. 2023

Thesis defence

• Supervisor: prof. RNDr. Pavla Hejcmanová, Ph.D.
• Reader: Pierre Cyril Renaud, externi

Czech University of Life Sciences Prague

Faculty of Tropical AgriSciences

Master programme / specializace:
Wildlife and Livestock Production, Management and Conservation / Wildlife Management and Conservation