Suitable habitats of fish species in the Barents Sea

• Main Authors: Husson, Berengere, Certain, Gregoire, Filin, Anatoly, Planque, Benjamin
• Format: Text
• Language: English
• Published: Cold Spring Harbor Laboratory 2020
• Subjects: envir; geo; Arctic; Barents Sea; atlantic cod; Climate change
• Online Access: https://doi.org/10.1101/2020.01.20.912816; https://archimer.ifremer.fr/doc/00605/71682/70120.pdf; https://archimer.ifremer.fr/doc/00605/71682/70121.pdf; https://archimer.ifremer.fr/doc/00605/71682/70122.pdf; https://archimer.ifremer.fr/doc/00605/71682/70123.pdf; https://archimer.ifremer.fr/doc/00605/71682/70124.pdf; https://archimer.ifremer.fr/doc/00605/71682/

Many marine species are shifting their distribution poleward in response to climate change. The Barents Sea, as a doorstep to the fast-warming Arctic, is experiencing large scale changes in its environment and its communities. This paper aims at understanding what environmental predictors limit fish species habitats in the Barents Sea and discuss their possible evolution in response to the warming of the Arctic. Species distribution models usually aim at predicting the probability of presence or the average abundance of a species, conditional on environmental drivers. A complementary approach is to determine suitable habitats by modelling the upper limit of a species’ response to environmental factors. Using quantile regressions, we model the upper limit of biomass for 33 fish species in the Barents Sea in response to 10 environmental predictors. Boreal species are mainly limited by temperatures and most of them are expected to be able to expand their distribution in the Barents Sea when new thermally suitable habitats become available, in the limit of bathymetric constraints. Artic species are often limited by several predictors, mainly depth, bottom and surface temperature and ice cover, and future habitats are hard to predict qualitatively. Widespread species like the Atlantic cod are not strongly limited by the selected variables at the scale of the study, and current and future suitable habitats are harder to predict. These models can be used as input to integrative tools like end-to-end models on the habitat preference and tolerance at the species scale to inform resource management and conservation.