Relier les variations interannuelles des indices d'abondance du capelan dans le golfe du Saint-Laurent aux indicateurs environnementaux de la régulation ascendante de la force des cohortes

Abstract

In this study, we applied a conceptual model relating capelin cohort strength to bottom-up processes using newly developed capelin abundance indices and descriptors of physical and biological oceanographic conditions in the Gulf of St. Lawrence (GSL). The main objective of the analyses was to provide evidence that abundance indices estimated from bottom-trawl surveys in the GSL track some of the variation expected from environmental conditions known to control variations in capelin cohort strength on the eastern Newfoundland shelf. In addition, we sought to improve our understanding of environmental drivers of capelin dynamics and productivity in the GSL over the last 3 decades. Our approach included (1) the development of a capelin larval abundance indices considered as a potential explanatory variable in the conceptual (2) analyses aimed at documenting the links between physical oceanographic conditions, spring bloom and Calanus dynamics, and capelin body condition in the GSL, and (3) multivariate non-linear models relating capelin abundance indices in different sub-regions of the GSL with environmental indices or capelin condition used as proxies of capelin survival potential. Our results showed that the seasonal phenology and June abundance of Calanus hyperboreus was associated with the ice and spring bloom dynamics, while the phenology and abundance of C. finmarchicus was mostly associated with sea-surface temperature (SST) indices in spring and early summer. High capelin body condition (Kn) in June was related to either early ice retreat or associated high C. hyperboreus abundance and early development timing, while high Kn in August-September was observed in years with a late timing of C. finmarchicus population development or a high C. finmarchicus abundance and SST during summer. Variations in capelin abundance indices were generally associated with variations in Kn in June and/or in August-September or their environmental proxies during the first 1-2 years of life in capelin, but not to capelin larval abundance indices. Timing of ice retreat was selected in 5 out of 6 models whereas proxies of Kn in August-September were selected in 3 of 4 models. Our results therefore identified sea ice dynamics (late winter-early spring) or SST (late spring-summer) as potential key drivers of Calanus species dynamics and capelin condition, confirming the bottom up hypothesis implicitly considered in the predictive model of capelin biomass on the eastern Newfoundland shelf. Our results also confirmed that capelin abundance indices derived from bottom-trawl surveys in the GSL generally track variation in abundance expected from known bottom-up processes regulating capelin’s cohort strength.