Model-based mapping of assemblages for ecology and conservation management: A case study of demersal fish on the Kerguelen Plateau

Aim: Quantifying biological assemblages and their environment is a fundamental, yet statistically challenging task in conservation ecology. Here, we use a recently developed approach called Regions of Common Profile (RCP) to quantify and map the distribution of demersal fish assemblages in an ecologically significant region of the Southern Ocean to (1) gain ecological and management insights and (2) evaluate the utility of the new method for ecoregionalization. Location: Northern Kerguelen Plateau, Subantarctic Islands, Southern Ocean. Methods: The RCP approach is a multispecies, model-based approach that can overcome many limitations of traditional distance-based approaches. It simultaneously groups sites with a similar composition of species and describes the patterns of variation in assemblages using environmental data, allowing the prediction of assemblages across the study region. We apply RCP to a unique dataset of demersal fish occurrences across the northern Kerguelen Plateau to model and map the distribution of assemblages and examine the representativeness of the Heard Island and McDonald Island marine reserve. Results: We demonstrate that the RCP approach allows a direct and quantitative interpretation of the composition of assemblages as well as their environment Further, the model reasonably predicts the occurrence of individual species across the plateau as well as the species composition of sites. We distinguish and map seven assemblages defined by depth, surface temperature and chlorophyll a. Shallow-water assemblages contain a high proportion of endemic species, while deep-water assemblages contain more cosmopolitan species. With the exception of one deep-water assemblage, assemblages were well represented within the current Heard and McDonald Islands marine reserve. Main conclusions: The RCP is a valuable tool for classifying biological regions with a range of ecological and conservation management applications. Our results extend current ecological and biogeographic knowledge for the northern Kerguelen Plateau, and maps of the distribution of assemblages will be useful for ongoing spatial management.