Physiological and ecological effects of fencing elephants in montane forests, Kenya

Morrison, Jacqueline (2019) Physiological and ecological effects of fencing elephants in montane forests, Kenya. Doctoral thesis (PhD), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (3MB) | Preview

Abstract

Worldwide, wildlife populations are declining at an unprecedented rate, anthropogenic influences including habitat-loss, poaching, and over-exploitation are driving many species to extinction. A rapidly increasing human population and subsequent conversion of land use, has intensified a competition of resources between humans and wildlife. Human elephant conflict (HEC) is a prominent concern and whilst many attempts have been made to mitigate conflict, many are unsuccessful, as such, Kenya embarked on the necessary strategy of fencing its montane forests to separate elephants from people. Whilst fencing successfully reduced conflict, the physiological and ecological effects of confining elephants in forests is unknown. This thesis is the first study examining the impact of fencing elephants in montane forests, on both the ecological impact on the integrity of the forest habitat, and the impact of isolation on the demographic structure, and physiology of the elephant population. The outcomes of this study will be used to inform the future management of elephant populations in forest habitats throughout their range. Chapters of this thesis address the following outcomes; (1) the impact of confining African elephants (Loxodanta africana) on the extent and quality of a forest habitat and (2) to provide a density estimate of elephants in a montane forest by employing the Random Encounter Model (REM) with camera traps. Additionally (3), it also assesses the impact of restricting dispersal on the demographic structure of elephant populations compared to free-roaming populations, and (4) examines the physiological effect of confinement on elephants. To provide an assessment of the impact of confinement on the forest habitat, changes in the forest canopy were mapped by applying the Breaks For Additive Season and Trend (BFAST) model using a time-series of MODIS satellite data on the 2 fenced forest habitats of varying sizes. Change maps produced from the BFAST model on MODIS timeseries data revealed that 13% of the Aberdare National Park had undergone a change in vegetation, with 0.9% loss to the forest. Similarly, the Shimba Hills National Reserve underwent significant change over the 12-year period, averaging a 9% change in the area per year with both positive ‘greening’ events and forest degradation. Density estimates calculated from the REM yielded results averaging 0.49 elephants/ km2 which falls in line with expected estimates in protected areas. Demographic results determined that the age viii class frequencies for the Aberdare Conservation Area (ACA) were significantly different to parity (x 2 = 147.24, df = 2, p-value < 0.001), showing a skewed age distribution towards adults and sub-adults and a lower proportion of juveniles. Comparison of the age class frequencies in the ACA to 3 populations from published data of known status (stable, and rapidly increasing) revealed that the ACA population were significantly different to both stable populations (x 2 = 495.05, p-value < 0.001; x 2 = 215.98, p-value < 0.001), and the rapidly increasing population (x 2 = 329.1, p-value < 0.001). The number of dependents to adult females was significantly lower compared to the control populations (x2 = 9.0872, df = 3, p-value = 0.02815). Endocrine analysis of adrenal activity found significant differences in FGM concentrations between the confined ACA population, a free-roaming savannah population, and a population from a montane forest that utilises a wildlife corridor to the lowland savannah (R2 =0.2331, F = 35.1, df= (2), 231, p-value < 0.001). At 3.73 ng/g (95%CI, 3.35-4.16), the confined ACA population is 91% higher than the Lewa Wildlife Conservancy (LWC) population with a wildlife corridor (1.95 ng/g, 95% CI, 1.72-2.21) and 68% higher than the Samburu National Reserve (SNR) population (2.22 ng/g, 95% CI, 1.98-2.59). No significant differences in FGMs were observed between the two free-roaming (LWC and SNR) populations. The greatest difference in FGM concentrations were observed in the subadult age class, with the ACA averaging 4.13 ng/gm (95% CI, 3.58-4.76) compared to 1.87 ng/g (95% CI, 1.59-2.1) in the LWC, and 2.74 ng/g, (95% CI, 2.28-3.30) in the SNR. Whilst the ACA exhibited only small-scale degradation, and the current density of elephants falls within the expected densities for protected areas. Significant differences were observed in the demographic structure and physiological condition of the isolated elephant population. The age-distribution was skewed towards the adult age classes, and concentrations of adrenal hormones were elevated. As such, these could have negative implications on population growth rates, and the subsequent viability of the population.