Modelling the spatial–temporal distributions and associated determining factors of a keystone pelagic fish

• Published in: ICES Journal of Marine Science
• Main Authors: Andrews, Samantha, Leroux, Shawn J, Fortin, Marie-Josée
• Other Authors: Kaplan, David M, Natural Sciences and Engineering Research Council of Canada’s Canadian Healthy Oceans Network, Department of Fisheries and Oceans Canada, Institut Nordique de Recherche en Environnement et en Santé au Travail
• Format: Article in Journal/Newspaper
• Language: English
• Published: Oxford University Press (OUP) 2020
• Subjects: Ecology; Aquatic Science; Ecology, Evolution, Behavior and Systematics; Oceanography; North Atlantic; North Atlantic oscillation
• Online Access: http://dx.doi.org/10.1093/icesjms/fsaa148; http://academic.oup.com/icesjms/article-pdf/77/7-8/2776/35589035/fsaa148.pdf

Abstract Mobile pelagic species habitat is structured around dynamic oceanographic and ecological processes that operate and interact horizontally and vertically throughout the water column and change over time. Due to their extensive movements, pelagic species distributions are often poorly understood. We use the Maxent species distribution model to assess how changes in the relative importance of modelled oceanographic (e.g. temperature) and climatic variables (e.g. the North Atlantic Oscillation) over 17 years affect the monthly average horizontal and vertical distribution of a keystone pelagic forage species, Atlantic Canadian capelin (Mallotus villosus). We show that the range and distribution of capelin occurrence probabilities vary across horizontal and vertical axes over time, with binary presence/absence predictions indicating capelin occupy between 0.72% (April) and 3.45% (November) of the total modelled space. Furthermore, our analysis reveals that the importance of modelled oceanographic variables, such as temperature, varies between months (44% permutation importance in August to 2% in May). By capturing the spatial dynamics of capelin over horizontal, vertical, and temporal axes, our analysis builds on work that improves our understanding and predictive modelling ability of pelagic species distributions under current and future conditions for proactive ecosystem-based management.