The constraint of ignoring the subtidal water climatology in evaluating the changes of coralligenous reefs due to heating events.

Predicting community-level responses to seawater warming is a pressing goal of global change ecologists. How far such predictions can be derived from a fine gradient of thermal environments needs to be explored, even if ignoring water climatology does not allow estimating subtidal marine heat waves. In this study insights about the influence of the thermal environment on the coralligenous community structure were gained by considering sites (Sardinia, Italy) at different temperature conditions. Heating events were measured (by loggers at 18 m, 23 m, 28 m, 33 m and 38 m deep) and proxies for their duration (the maximum duration of events warmer than the 90th percentile temperature), intensity (the median temperature) and variability (the number of daily ΔT larger than the mean daily ΔT, and the number of heating events larger in ΔT than the 90th percentile ΔT) were selected by GAM models. Reliable predictions of decrease in coralligenous richness of taxa/morphological groups, with relevant increment in turfs and encrusting coralline algae abundance at the expenses of bryozoans were made. Associations to the different types of heating descriptor have highlighted the aspect (intensity, duration or variability) of the heating events and the threshold for each of them responsible for the trajectories of change.