Drivers of biodiversity change in the Anthropocene

Across the globe, the populations of species and the biodiversity of ecological communities are changing, including declines, gains and stable trends over time. Against a backdrop of accelerating global change, a critical research challenge is to disentangle the sources of the heterogeneous patterns of population and biodiversity change over time. In this thesis, I linked population and biodiversity change with species traits like rarity and commonness, and with global change drivers like forest loss. I synthesised global biodiversity databases with gridded driver datasets to quantify how species’ populations and biodiversity are being impacted by human activities in the Anthropocene. The rise of open-access data in ecology has produced databases with millions of records which have launched large-scale syntheses of how Earth’s biota is changing over time and space. However, our knowledge of biodiversity change is limited by the available data and their biases. In Chapter 1, I tested the representation of three worldwide biodiversity databases (Living Planet, BioTIME and PREDICTS) across geographic and temporal variation in global change over land and sea and across the tree of life. I found that variation in global change drivers is better captured over space than over time and in the marine realm versus on land. I provided recommendations on how to improve the use of existing data, better target future ecological monitoring and capture different combinations of global change. In Chapter 2, I tested whether vertebrate species from specific biomes, taxa or with certain species traits are more likely to increase or decrease in a time of accelerating global change. I analysed nearly 10 000 population abundance time series from over 2000 vertebrate species part of the Living Planet Database. I integrated abundance data with information on geographic range, habitat preference, taxonomic and phylogenetic relationships, and IUCN Red List Categories and threats. I found that 15% of populations declined, 18% increased, and 67% showed no net changes over time. Amphibians were the only taxa that experienced net declines in the analysed data, while birds, mammals and reptiles experienced net increases. Despite this variation among broad taxonomic groups, surprisingly I did not detect phylogenetic patterns in which species were more likely to decline versus increase. Population trends were poorly explained by species’ rarity and global-scale threats. I found that incorporating the full spectrum of population change, including declines, gains and stable trends, will improve conservation efforts to protect global biodiversity. In Chapter 3, I explored land-use change to fill the gap in empirical evidence of how habitat transformations such as forest loss and gain are reshaping biodiversity over time. I quantified how change in forest cover has influenced temporal shifts in populations and ecological assemblages from over 6000 globally distributed time series across six taxonomic groups. I found that local-scale increases and decreases in abundance, species richness, and temporal species replacement (turnover) were intensified by as much as 48% after forest loss. Larger amounts of forest loss did not always correlate with higher population and biodiversity change across sites, highlighting the mediating effects of local context and historical baselines. Temporal lags in population- and assemblage-level shifts after forest loss extended up to 50 years and increased with species’ generation time. My findings indicate that forest loss amplified population and biodiversity change, with effects on both short and long temporal scales. A mix of immediate and lagged biodiversity change following land-use change emphasises the need for temporally explicit biodiversity scenarios to accurately estimate progress towards conservation goals. Together, my thesis findings demonstrate the wide spectrum of population and biodiversity change happening across varying amounts of global change and different realms, taxa and species traits. These heterogeneous impacts of global change on population and biodiversity spanned temporal scales from immediate effects in a couple of years to lagged responses decades after disturbance. The links between global change drivers and shifts in species’ abundance, species richness and compositional turnover depended on historical context and species’ characteristics like generation time. I documented both immediate and temporally delayed effects of global change drivers on species’ populations abundance and the biodiversity of ecological assemblages which highlights the importance of long-term ecological monitoring. The main implications of my thesis findings are that first, any inferences drawn from biodiversity syntheses reflect the types of species and places represented by the data and the global change that is experienced. To create accurate scenarios, we need biodiversity data that span not only different taxa and locations, but also the spectrum of global change variation around the world. Second, biodiversity predictions should incorporate both positive and negative impacts of global change drivers as well as lagged responses. Finally, ecosystems and the species within them are usually simultaneously exposed to a suite of global change drivers and a key future research step is to test the synergy and/or antagony in the effects and interactions among multiple types of environmental change on populations and biodiversity. Overall, my thesis research demonstrates that the drivers of biodiversity change in the Anthropocene have both immediate and temporally-delayed effects which depend on species’ traits and the sites’ historical context. My findings suggest that by incorporating the full spectrum of biodiversity change and the nuance around interacting global change drivers we can improve projections of future ecological shifts and enhance local and international conservation policies.