Goto main content

Study on the drivers of distribution of Northeast Atlantic mackerel in the Northeast Atlantic ocean

In our newest publication ‘Drivers of the summer-distribution of Northeast Atlantic mackerel (Scomber scombrus) in the Nordic Seas from 2011 to 2017; a Bayesian hierarchical modelling approach’ by Nikos Nikolioudakis, Hans Julius Skaug, Anna Olafsdottir, Teunis Jansen, Jan-Arge Jacobsen and Katja Enberg, we analyze the observed distributional patterns of mackerel and identify the factors mainly controlling them.

Published 20.08.2018 - Updated 20.08.2018

Northeast Atlantic (NEA) mackerel is a species of great economic and ecological importance, whose habitat expansion in the last decade has altered the biomass dynamics in the pelagic realm of the Nordic Seas. Mackerel is caught as west as Greenland and as North as Svalbard and this has caused concerns on the effects this expansion can have on other pelagic species, such as the Norwegian spring spawning herring, that compete for the same prey resources. Utilizing data obtained during the International Ecosystem Summer Survey in the Nordic Seas from the period 2011 to 2017, we used state-of-the-art modelling techniques, such as Bayesian hierarchical spatiotemporal models, to identify the factors affecting mackerel’s distribution.

Figure 1. Observed mackerel densities (left panel) vs. predicted (mean posterior fitted) values for the positive density model (right panel) for 2017. Circle sizes in the right panel represent the estimated densities, whereas the colour scale indicates the uncertainty for each fitted value (K: thousands of fish).
Our findings showed that temperature in the upper 50m of the water column, food availability (approximated by mesozooplankton biomass), a proxy of herring abundance and longitude were the main factors influencing both the catch rates (proxy for fish density) and the occurrence of NEA mackerel. Stock size was not found to directly influence the distribution of the species; however, catch rates in higher latitudes during years of increased stock size were lower. Additionally, we highlight the improved performance of models with spatiotemporal covariance structures, thus providing a useful tool towards elucidating the complex ecological interactions of the pelagic ecosystem of the Nordic Seas.

Figure 2. Model predictions for the probability of occurrence (blue dotted line) and the positive density of mackerel (red solid line), based on the assumption of a quadratic relationship with the mean temperature of the top 50 m.
These results suggest that the distribution pattern of a pelagic fish species are far more complex than we imagine and not easily explained by our static picture of the ocean, as perceived from data derived standard scientific cruises. Indispensable as the latter may be, we need to improve our methods of capturing long term variability if we want to understand the complex ecological interactions in the pelagic ecosystems.

The open-access article can be found here