Abstract/Summary

Tackling the threats to biodiversity and reversing declines requires an understanding of how it is changing and what the key drivers of change are. One of the largest data sets used to assess these trends globally is that underlying the Living Planet Index (LPI), a biodiversity indicator based on vertebrate population trends. This data set contains just under 40,000 population trends from over 5,000 species collected at monitoring sites in terrestrial, freshwater and marine habitats around the world. However, gaps in the geographic and taxonomic representation of the data set risk incorporating bias into our current understanding of biodiversity change. Assessing and addressing any potential bias in the LPI, coupled with an insight into how common drivers of biodiversity decline relate to the observed trends, can help advance knowledge of how and why vertebrate populations are changing regionally and globally. Using the database that populates the LPI, I first analysed trends in vertebrate populations in the Arctic and revealed an average increase in relative abundance across the region but varying trends among the High, Low and Sub Arctic. Critically, declines were recorded in sea-ice associated species, which are particularly vulnerable to impacts in this rapidly changing ecosystem. My second paper assessed the taxonomic and geographic bias in the global LPI. The lowest representation of known vertebrate species was found in tropical biogeographic realms and among herptiles and fish. A method for mitigating this bias was proposed which employs a proportional weighting for each taxa-realm combination informed by the estimated species richness within that subset. Applying this diversity-weighted approach to calculating the global LPI produced an index which suggests that trends in vertebrate populations are more negative than previously thought. The second two papers examine two of the most common drivers of biodiversity decline: habitat loss and exploitation. Firstly, I explored trends in species which are wholly reliant on forests as their habitat and found that their populations are declining more on average than terrestrial species in general. Along with my co-author we also used two forest data sets to examine the relationship between tree cover change and population trends. There was no evidence of a relationship between these two variables, but we did find a significant association with whether the population was threatened with overexploitation. For the final paper, I used information on whether or not a population was utilised to explore global and regional trends in populations that support people’s wellbeing or livelihoods. I found an average decline of 50% in utilised populations since 1970 and a more negative trend than populations which are not utilised. Crucially, a positive trend was associated with when a population was subject to targeted management. Both of these papers resulted in two new indicators which have been identified for use in monitoring progress towards global biodiversity goals and targets of the Convention on Biological Diversity: the Forest Specialist Index and the LPI for utilised populations. Together these four publications present a clearer picture of how vertebrate populations are fluctuating globally and regionally, and better evidence of how common drivers of change relate to these trends. Limitations remain as to how population time-series data are aggregated into a global indicator and how data bias and uncertainty is addressed. Whilst abundance data is a valuable and sensitive metric for measuring biodiversity change, advances in how the data are analysed to produce global assessments will be vital to ensure a better understanding of trends in vertebrate populations and how they are predicted to change under future environmental change scenarios.