How temporal turnover in community composition is linked to anthropogenic pressure

Abstract

Understanding the mechanisms behind biodiversity is central to assessing and forecasting anthropogenic impacts on ecological communities. However, quite how intrinsic ecological processes and external environmental drivers act together in natural systems to influence local temporal turnover is currently largely unexplored. Here, we determine how human impacts affect multiple metrics of bird community turnover to establish the ecological mechanisms behind compositional change. We used US Breeding Bird Survey data to calculate transect-level rates of three measures of temporal species turnover: a) short-term (initial rate of decline of Sorensen similarity), b) long-term (asymptotic Sorensen similarity), and c) overall species accumulation rate (species-time relationship exponents) over 2692 transects across 27 habitat types. We then hierarchically fit linear models to estimate the effect on these turnover metrics of anthropogenic impact via the Human Modification Index proxy, while accounting for observed species richness, the size of the species pool and annual environmental variability. We found broadly consistent impacts of increased anthropogenic pressures across diverse habitat types. The Human Modification Index was associated with greater turnover at long-timescales, but marginally slower short-term turnover. The species accumulation rate through time was not notably influenced. Examining anthropogenic impacts on different aspects of species turnover in combination allows greater ecological insight. Observed human impacts on short-term turnover were the opposite of existing expectations and suggest humans are disrupting the background turnover of these systems, rather than simply driving rapid directed turnover. The increased long-term turnover was driven by more frequent species changes between core and occasional status rather than greater arrival of new species. These results highlight the role of intrinsic dynamics and caution against simple interpretations of increased species turnover as reflections of environmental change. Open Research StatementNo new empirical data are used in the manuscript as all primary data is publicly available, as cited in the manuscript. Our code repository (including fitted model objects and markdown documents detailing all steps) is available at https://figshare.com/s/f5b9152ff7643efb347d.