Modelling and projecting the response of local terrestrial biodiversity worldwide to land use and related pressures: the PREDICTS project

The PREDICTS project (Projecting Responses of Ecological Diversity In Changing Terrestrial Systems) has collated ecological survey data from hundreds of published biodiversity comparisons of sites facing different land-use and related pressures, and used the resulting taxonomically and geographically broad database (abundance and occurrence data for over 50,000 species and over 30,000 sites in nearly 100 countries) to develop global biodiversity models, indicators and projections. After outlining the science and science-policy gaps that motivated PREDICTS, this review discusses the key design decisions that helped it to achieve its objectives. In particular, we discuss basing models on a large, taxonomically and geographically representative database, so that they may be applicable to biodiversity more broadly; space-for-time substitution, which allows estimation of pressure-state models without the need for representative time-series data; and collation of raw data rather than statistical results, greatly expanding the range of response variables that can be modelled. The heterogeneity of data in the PREDICTS database has presented a range of modelling challenges: we discuss these with a focus on our implementation of the Biodiversity Intactness Index, an indicator with considerable policy potential but which had not previously been estimated from primary biodiversity data. We then summarise the findings from analyses of how land use and related pressures affect local (?) diversity and spatial turnover (? diversity), and how these effects are mediated by ecological attributes of species. We discuss the relevance of our findings for policy, before ending with some directions of ongoing and possible future research.