The application of biodiversity indicators to infer ecosystem health in regenerating tropical forest

Allen, Laura (2019) The application of biodiversity indicators to infer ecosystem health in regenerating tropical forest. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2019Allenphd.pdf] PDF
Download (9MB)
Printed Thesis Information:


There are an overwhelming number of biodiversity indices and indicators available for ecologists and conservationists to use when seeking to understand how biodiversity responds to human disturbance. In choosing between measures there is often an underlying assumption that if a measure works well for one group it will be equally applicable to another. In this study, I use multiple taxa to explore the performance of a wide range of alpha and beta diversity measures for studying biodiversity responses to human disturbance in tropical forest. I sampled 18 sites along a gradient of human disturbance from primary tropical forest to banana monocultures in Peru. I chose three taxonomic groups and one audio approach, which have all been suggested to be useful indicators for studying biodiversity responses to disturbance: orchid bees (n = 1783), dung beetles (n = 3787), butterflies (n = 2506) and soundscape samples (n = 6600). This allowed me to identify how these groups responded to disturbance, which diversity measures were most sensitive for detecting those changes and whether the same measures were suitable for all groups. I used Hill numbers to measure alpha diversity and explored beta diversity by looking at changes in community composition and two new measures of beta diversity: redundancy and representativeness. To see how the diversity patterns changed when taxonomic similarity was considered, I used a recently developed family of similarity-sensitive diversity measures and compared the results of these against more traditional measures. I found that the diversity indices that were best for detecting disturbance patterns varied widely among taxonomic groups. For dung beetles, species richness and community composition were the most effective measures, whereas these performed poorly for orchid bees. Abundance and redundancy were more sensitive for detecting a response to disturbance in orchid bees. Using the butterfly dataset, I show that the inclusion of species similarity completely changed the diversity patterns found across the disturbance gradient. The similarity of species present in a community is likely to be important for the preservation of evolutionary adaptability and the provision of ecosystem functions and I therefore suggest that diversity measures based on similarity will be a useful additional tool for conservation and impact assessments. Acoustic diversity showed unintuitive responses to disturbance, with higher diversity detected in more disturbed forest, and more research is required to assess the performance of different acoustic indices in rainforest environments. Overall, my results demonstrate the importance of choosing diversity indices carefully to suit the taxa being studied to avoid missing important ecological responses, including a consideration of species similarity. I recommend that, where possible, multiple diversity indices and taxonomic groups should be used to reduce this risk and provide a comprehensive understanding of ecosystem patterns in response to environmental change.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: biodiversity, tropical forest, indicators, diversity measures, human disturbance, invertebrates, acoustics, orchid bees, dung beetles, butterflies.
Subjects: Q Science > QH Natural history
Q Science > QH Natural history > QH301 Biology
Q Science > QL Zoology
Colleges/Schools: College of Medical Veterinary and Life Sciences > Institute of Biodiversity Animal Health and Comparative Medicine
Supervisor's Name: MacLeod, Dr. Ross, Reeve, Dr. Richard and McGregor, Dr. Anna
Date of Award: 2019
Depositing User: Laura Allen
Unique ID: glathesis:2019-40949
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 23 Jan 2019 11:33
Last Modified: 05 Mar 2020 22:44
Thesis DOI: 10.5525/gla.thesis.40949

Actions (login required)

View Item View Item


Downloads per month over past year