Epsilon Archive for Student Projects

Abstract

This thesis aims to identify the driving factors of the movements and the direction of movements of
the African savannah elephant (LA) based on data obtained from camera traps. Thereto, the
methodology involves two main steps: In a first step, the pictures of the camera traps taken between
the years 2015 and 2019 were analysed. Demographic data such as age and sex as well as
information on movement direction were collected. In a second step, a multivariate generalized
linear model (GLM) is estimated to examine the relationship between the response variables
movement and direction and the explanatory variables temperature, precipitation, mean rainfall over
the last 30, 60 and 90 days, vegetation index as well as group constellation.
This thesis has three main findings: Firstly, temperature, precipitation, mean rainfall over the
last 30 days (X30) and 90 days (X90), and group constellation have a significant influence on the
movement activity of LA. The likelihood of movement events happening increases on average with
a rise in temperature as well as a rise in X30 and X90. Contrariwise, movement activity decreases
on average with higher precipitation. The importance of the temperature and precipitation variables
suggests that there is an immediate response of LA´s movement activity to daily environmental
conditions. Furthermore, the importance of the variables X30 and X90 indicate that there is also a
seasonal component, influencing the LA activity. Moreover, variation in the likelihood of LA
movements can be explained by variation in the group constellation. Male groups occur to have the
highest likelihood to move, whereas female groups were least likely to cross the corridor. This
suggest that sex is one of the driving factors of movement activity. Secondly, the direction of the
movement of LA is determined by the following variables: temperature, group constellation and
vegetation index. Higher temperature increases the likelihood of movement inside the conservancy.
Male and female groups have the highest likelihood to move inside the park and single females and
males the highest to move outside. When the vegetation shows a negative growth rate, LA are more
likely to move inside the park. This controversial behaviour can be explained by, among other
things, man-made water resources and the availability of alternative food sources within the park.
Overall, this thesis illustrates that camera traps are an effective method to determine driving
factors of the LA movements and their direction of movements in Kenya. The results of this study
underline the correlation between the movement activity and direction of LA and various
demographic and environmental variables. Therefore, in times of changing environmental
conditions and increasing human-wildlife conflicts it is of utmost importance to understand these
dynamics for effective wildlife conservation and management strategies.