Transition probabilities help identify putative drivers of community change in complex systems

Abstract

Understanding the role of larger-scale processes in modulating the assembly, structure, and dynamics of communities is critical for forecasting the effects of climate-change and managing ecosystems. Developing this comprehensive perspective is difficult though, because species interactions are complex, interdependent, and dynamic through space and time. Typically, experiments focus on tractable subsets of interactions that will be most critical to investigate and explain shifts in communities, but qualitatively base these choices on experience, natural history, and theory. One quantitative approach to identify the putative forces regulating communities, without reducing system complexity, is estimating transition probabilities among species occupying space (i.e., multispecies Markov chain models). Although not mechanistic, these models estimate the relative frequency and importance of ecological pathways in community assembly and dynamics, and can serve as a framework to identify how pathways change across large scales and which are most important to investigate further. Here, we demonstrate this method in the Gulf of Maine (GOM) intertidal zone, where research has largely focused on the local-scale processes that influence communities, while the mechanisms responsible for more regional shifts in communities are less clear. Transition probabilities of faunal elements were quantified bimonthly for ~2.5yr in local intertidal communities at three replicate sites in the southern, mid-coast, and northern GOM. Transitions related to mortality, colonization, and replacement by mussels, barnacles, red algae, and encrusting corallines differed regionally, suggesting specific pathways related to consumer pressure and recruitment vary across the GOM with shifting intertidal community structure. Combined with species abundance data and insights from previous research, we develop and evaluate the pathways by which communities likely change in the GOM. Species interactions in local communities can be complex, and this complexity should be incorporated into hypothesis building, experiments, theory, interpretations, and forecasts in ecology. Such a comprehensive approach will be critical to understand how regional shifts in local interactions can drive large-scale community change.