Marine ecological abrupt shifts in a warming world: from monitoring to predictions

Impermanence is an ecological principle but there are times when changes occur nonlinearly as Abrupt Community Shifts (ACSs), altering the ecosystem structure. ACSs can also rapidly modify ecosystem services (the benefits people obtain from ecosystems), hence they can disrupt economies and societies. Understanding and predicting them is crucial for both scientists and managers. Here, we present a model based on the MacroEcological Theory on the Arrangement of Life (METAL) to explain and predict ACSs at the global scale. We test our model using 14 multi-decadal time series of marine metazoans from zooplankton to fish, spanning all latitudes, from shelf to open ocean, from 1960 to 2015. Predicted and observed fluctuations correspond, both identifying ACSs at the end of the 1980s and 1990s. Then we extend the predictions to the global ocean, beyond the range of time series monitoring. We show that ACSs coincide with changes in climate that alter local thermal regimes, which in turn interact with the thermal niche of species to trigger shifts at the community level. ACSs unprecedented in magnitude and extent are predicted after 2012, coinciding with a strong El Niño event and major shifts in Northern Hemisphere climate. Our results underline the sensitivity of marine communities to temperature, and suggest an increase in the size and consequences of ACS events in the upcoming warming world. They also indicate the vulnerability of the Arctic Ocean communities. Our ability to resolve the spatial extent of oceanic community shifts is severely constrained by a scant and unrepresentative coverage of observations. Most marine populations, in particular metazoans, are hidden from earth observation tools and adequate monitoring coverage for the entire ocean is logistically unlikely. Our framework could therefore be meshed with existing monitoring programs to provide a macroscopic tool for identifying regions likely to develop ACSs and to help anticipate biological perturbations, especially in function of global warming