M-304A-8
Do Environmental Factors Guide Selection of Overwintering Locations in Atlantic Herring When Learning Opportunities Are Few?

Monday, August 18, 2014: 4:40 PM
304A (Centre des congrès de Québec // Québec City Convention Centre)
Jed Macdonald , Faculty of Life ad Environmental Sciences, University of Iceland, Reykjavík, Iceland
Kai Logemann , Faculty of Life ad Environmental Sciences, University of Iceland
Niall McGinty , Faculty of Life ad Environmental Sciences, University of Iceland
William Butler , Faculty of Life ad Environmental Sciences, University of Iceland
Þorsteinn Sigurgursson , Marine Research Institute, Iceland
Guðrún Marteinsdottir , Faculty of Life ad Environmental Sciences, University of Iceland
Accurate predictions on how fish behave in response to changes in their physical environment are central to the effective management of commercially harvested stocks. Here, we explored how oceanographic forces have influenced the distribution of the Icelandic summer spawning herring (Clupea harengus) stock over a 22-year time series (1991-2012). Large shifts in the overwintering (non-feeding) location of the stock have been observed during this time; however, the factors governing these shifts remain uncertain. Commercial catch data from the autumn/winter purse seine fishery were compiled and boosted regression tree models were used to examine the influence of local- and regional-scale oceanographic features on herring capture probability and abundance patterns. Herring were more likely to be captured in warmer, shallower locations at salinities between 33.8 to 34.4. Models fitted for each fishing season independently showed best predictive performance in years coinciding with changes in overwintering location, and performed poorly when traditional migration patterns were maintained. The adopted migrant hypothesis contends that guidance from older cohorts drives selection of wintering locations. Our data support this hypothesis, but also suggest that when learning opportunities are limited, environmental factors may become ‘unmasked’ and play an important role in determining new migration routes.