Disturbance in the Udzungwas: Responses of Monkeys and Trees to Forest Degradation

Research output: ThesisDoctoral Thesis


Tropical forests are one of the world’s most threatened biomes. As tropical forests are increasingly destroyed and degraded, there is increasing need for research into the implications for rare species and habitats, and for conservation management. This thesis is a series of four papers investigating species and community responses in the Udzungwa Mountains, an area of international importance for biodiversity. The focus is on both rare species and potential indicators of habitat degradation. The most direct impact of human disturbance is on vegetation structure. However the response of most tropical plant species to disturbance has not been quantified. The Udzungwa area contains many rare and restricted range plant species, and furthermore, is arguably Africa’s most important single site for primate conservation. Primates also have high potential as indicators. In particular, diurnal monkeys are easily observed, are mostly dependent on tropical forests, and may show a diverse range of demographic and behavioural responses to disturbance.
Human impacts on the monkeys and trees of the Udzungwas are here assessed at two levels: 1) habitat loss, and 2) habitat degradation. The first analytical chapter (Chapter 2) investigates the relationship between monkey species richness and forest fragment size among 22 sites. Multivariate techniques are used to consider the relationship together with other confounding variables. The results show that there is a log-linear species-area relationship, highlighting the importance of large forests (above 150 km2) for biodiversity conservation. The results also suggest that hunting and isolation have further influenced the species composition, particularly in the smaller fragments.
The second analytical chapter (Chapter 3) assesses vegetation responses to disturbance using 120 plots, along six transects in the heavily disturbed lowland forest of Matundu. Multivariate analyses are used to assess variation in community composition, species abundance, stem density and diversity in relation to disturbance, environmental and topographic variables. With the exception of diversity, all measures of vegetation structure and composition are shown to have been affected by disturbance. In particular, rare species diversity was negatively related to disturbance, and also the presence of large animal paths. Therefore both humans and elephants seem to have had a large impact on the Matundu ecosystem. Three common tree species, Funtumia africana, Vangueria volkensii and Parinari excelsa, that have significant negative correlation with disturbance, could be used as indicators of forest health. Several environmental and topographical correlates with vegetation structure and composition are also identified.
The third analytical chapter (Chapter 4) is a methodological chapter reviewing techniques for estimating density of clustered animals. This is included in the thesis for use in the monkey-habitat analysis in Chapter 5, and because of the continuing intense debate regarding method selection for primates. The aim of the chapter is to develop a simple guide to method selection. It begins with a summary of the main debate, and an introduction to four alternative methods that are currently employed for surveying primates. The main controversy in the literature surrounds the debate over prioritising for mathematical framework (perpendicular methods) or for minimising correction factors (animal-observer methods). Three of the four methods have shown reasonable accuracy compared to known primate densities, and the fourth has yet to be tested. Perpendicular methods are the most desirable given appropriate field conditions. However problems arising from the five criteria of visibility, habituation of animals, cluster spread, and available resources, may often preclude their use.
Finally, chapter 5 looks at the relationship between monkey relative abundance, density and social grouping versus habitat. Data are presented on four monkey species (Udzungwa red colobus Procolobus gordonorum, Angolan black and white colobus Colobus angolensis palliatus, Sykes monkey Cercopithecus mitis subsp. and yellow baboon Papio cynocephalus) counted along the six line transects used for vegetation survey in Chapter 3. In accordance with ecological theory, the four species respond to habitat disturbance as expected from their dietary specialisation, geographic range, and from previous studies on these and closely related species. The Udzungwa endemic red colobus showed the closest relationship with habitat, and may have the most potential as an indicator of forest habitat quality of all taxa in the Udzungwas. Monkey community composition is also suggested as a good indicator of habitat quality. The results suggest that those areas in the Udzungwa lowlands that contain high densities and large groups of red and black and white colobus monkeys contain the best quality forest in terms of vegetation structure and composition. Those containing mainly Sykes monkeys are the poorest quality.
Overall the thesis shows that monkey and tree communities in the Udzungwa Mountains have been negatively impacted by human activities. However, the results also emphasise the importance of disturbed forests for future conservation of biodiversity. Given that most tropical forests have been disturbed to some extent, management of these areas is of paramount importance, as discussed in Chapter 6. The indicator species and communities identified here may be useful for this, both for identifying priority conservation areas and for monitoring forest recovery.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • University of York
  • Lovett, Jon C., Supervisor, External person
  • White, Piran Crawfurd Limond, Supervisor
  • Struhsaker, Thomas T., Advisor, External person
Thesis sponsors
Publication statusPublished - 2007

Cite this