Climate refugia for kelp within an ocean warming hotspot revealed by stacked species distribution modelling.

Canopy forming macroalgae are declining globally due to climate change and the identification of refuges for these habitats is crucial for their conservation. This is particularly pertinent in ocean warming hotspots where significant range contractions of kelp have occurred and are projected to continue. We developed a stacked urchin-kelp species distribution model (SDM) to predict climate refugia for kelp (Ecklonia radiata) in an ocean warming hotspot, south-eastern Australia. The optimal stacked-SDM incorporated biotic and abiotic explanatory covariates and was validated using an independent dataset. Density of the urchin Centrostephanus rodgersii, summer bottom temperature and photosynthetically available radiation at the seabed were significant predictors of kelp cover, highlighting the physiological and ecological influence of these variables on the distribution of kelp. Our optimal stacked-SDM predicted three spatially distinct refuge areas, where kelp occurs in deeper waters than surrounding seascapes. The presence of kelp at two of these refuge areas was confirmed using independent data. The identification of these refuge areas is crucial for conservation, as they are likely to facilitate the persistence of ecologically and economically important kelp forests as waters warm in shallow areas and kelp retreat to depth under climate change. Furthermore, identification of refugia will enable proactive spatial planning that prioritises new locations for protection to ensure that key kelp habitats can persist in a future of increasing stress.