Detection of eco-evolutionary dynamics in communities using Joint Species Distribution Models

Abstract

Biodiversity at the metacommunity scale is typically influenced by a number of environmental, spatial, biotic, and stochastic factors. At the same time, these factors impact the evolution of individual species, as sites present different local selection pressures and connectivity can impact gene flow and genetic drift. Identifying the relative impacts of environmental, spatial, biotic, and other drivers on community composition across spatial and temporal scales has been greatly facilitated by joint species distribution models, but these models have yet to consider the impact of microevolution on community composition. We used Heirarchical Models of Species Communities (HMSC) to analyze simulated data of an populations and communities, including in a large evolving metacommunity model, to establish whether HMSC can sufficiently quantify the contribution of phenotypic evolution for metacommunity composition. The models successfully partitioned variance contributed by environmental, spatial, and evolving phenotypic drivers, and also estimated site- and year-specific covariance. We also applied the HMSC with trait evolution model to an existing dataset studying trait change and community dynamics in an experimental aquatic plant system. The study of eco-evolutionary dynamics may require data that reflects numerous complex, interacting processes and it is necessary to have flexible, generalized statistical models to analyze this data. JSDM models such as HMSC present one promising path for analysis of eco-evolutionary dynamics in multi-species communities.