University of Twente Student Theses

Login

Elephant Foraging Behaviour : Application of Levy Flights in Geo-information Science and Remote Sensing

Kahiu, Milkah Njoki (2009) Elephant Foraging Behaviour : Application of Levy Flights in Geo-information Science and Remote Sensing.

[img] PDF
7MB
Abstract:The foraging behaviour of the African elephant (Loxodonta africana) along its movement path has been studied minimally due to lack of detailed movement data. However with advancement in Geographic information systems and remote sensing, data is available for understanding movement and resource use. Different methods have been used to understand animal movement, but random walks are among common models used to study animal forage search behaviour along their movement path. A recent and controversial type of random walk model is the Levy flights. Some researchers have disputed its existence in biological organisms, while others have faulted prior methods used in estimating it. Consequently, a more accurate method, Maximum Likelihood Estimation (MLE) has been proposed. This research strives to understand foraging behaviour of elephant population in Marsabit Protected area (Kenya) during the dry period. First, the research uses MLE method to determine whether elephants use Levy flights when searching for resources. Secondly it determines whether the foraging behaviour is influenced by forage density (vegetation), gender and time of feeding. Elephant movement data recorded for about 3years at one-hour time intervals was used for determining Levy flights. ASTER image was used to derive suitable foraging areas through Maximum Likelihood Classification and forage densities were estimated through Spectral Mixture Analysis using the Linear Mixture Model. For data organization and statistical analysis, Microsoft Excel and SPSS softwares were used. The research established that elephants use Levy flights in their search for resources in the Levy index range 1.2<µ≤1.3. Levy index had a positive relationship with the forage density, while males and females had different foraging behaviour. For the three elephant feeding peaks, the Levy indices were different. It was possible to understand the foraging behaviour of Marsabit Protected Area elephants during the dry season. However, there is need for further research to understand the elephant behaviour during the wet season. The map generated in the research for the suitable elephant foraging areas would be useful in management. An improvement in the determination of forage density would also be handy for improvement of the findings. Key words: Elephant, Foraging Behaviour, Forage, Levy Flights, Maximum Likelihood Estimation, Maximum Likelihood Classification, Spectral Mixture Analysis, Linear Mixture Model
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/93050
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page