MEPS 662:15-34 (2021)  -  DOI: https://doi.org/10.3354/meps13597

ABSTRACT: Understanding the spatial and temporal preferences of toxic phytoplankton species is of paramount importance in managing and predicting harmful events in aquatic ecosystems. In this study, we addressed the realised niche of the species Alexandrium minutum, Pseudo-nitzschia fraudulenta and P. australis to highlight distribution patterns at different scales and determine possible drivers. To achieve this goal, we developed original procedures coupling niche theory and habitat suitability modelling using abundance data in 4 consecutive steps: (1) estimate the realised niche applying kernel functions, (2) assess differences between the species’ niches as a whole and at the local level, (3) develop habitat and temporal suitability models using niche overlap procedures and (4) explore species temporal and spatial distributions to highlight possible drivers. We used data on species abundance and environmental variables collected over 27 yr (1988-2014), including 139 coastal water sampling sites along the French Atlantic coast. Results showed that the niches occupied by A. minutum and P. australis are very different, although both species exhibit a preference for warmer months. They both respond to decadal summer North Atlantic Oscillation (sNAO) but in an opposite way. P. fraudulenta’s realised niche lies in between that of the 2 other species; it also prefers warmer months but does not respond to decadal sNAO. The Brittany peninsula is now classified as an area of prevalence for the 3 species. The methodology used here will allow us to anticipate species distributions in the event of future environmental challenges resulting from climate change scenarios.

KEY WORDS: HABs · Time series · Monitoring · REPHY · Climatic index · Multivariate analysis · Prediction · Temperature