Towards population estimates for birds of central Africa’s Albertine Rift

Abstract

Globally, biodiversity faces severe pressures, which are likely to increase with ongoing climate change and a growing human population. Despite birds being one of the world’s best-monitored taxa, data on population sizes, distributions and demography remain patchy, rendering threat evaluation and mitigation planning difficult. Few studies comparing impacts of data collection methods on population estimation exist. Accurate population data are fundamental for conservation applications. In this thesis, I first use distance sampling, a method of density estimation accounting for individuals’ detectability, to compare the efficacy of two widely-used data collection methods, line transects and point counts. I demonstrate point counts obtain systematically higher density estimates, and show there are marked differences in estimates between methods related to species behavioural and morphological traits. The results demonstrate that survey method should be evaluated case-by-case, considering issues such as target species and landscape characteristics. Following this, I apply distance sampling to estimate population sizes of birds across regions of central Africa’s Albertine Rift, a global biodiversity hotspot lacking baseline data. This region harbours the continent’s highest levels of vertebrate species richness and endemism but faces major threats from climate and land-use change. This work reveals non-uniform patterns of species richness and densities, helping identify priority regions and habitats to protect individual endemic and threatened species. Finally, I investigate the effects of competitor abundance and habitat on abundance of the endemic and threatened bird species in the region. These analyses indicate little effect of competitor abundance, with habitat being a more important determinant of population sizes. In fact, species identified as competitors largely occur at higher abundances when co-occurring, something possibly attributable to habitat quality effects dominating over competitive interactions in driving density patterns. These findings suggest that translocation of species between mountain peaks in the Albertine Rift, a possible solution to extinction risk in this region where many populations occur on isolated high elevation sites, is unlikely to be detrimental to potential competitor species already present at recipient sites.