Marine ecosystems, particularly in high‐latitude regions such as the Arctic, have been significantly affected by human activities and contributions to climate change. Evaluating how fish populations responded to past changes in their environment is helpful for evaluating their future patterns, but is often hindered by the lack of long‐term biological data available. Using otolith increments of Northeast Arctic cod (Gadus morhua) as a proxy for individual growth, we developed a century‐scale biochronology (1924–2014) based on the measurements of 3,894 fish, which revealed significant variations in cod growth over the last 91 years. We combined mixed‐effect modeling and path analysis to relate these growth variations to selected climate, population and fishing‐related factors. Cod growth was negatively related to cod population size and positively related to capelin population size, one of the most important prey items. This suggests that density‐dependent effects are the main source of growth variability due to competition for resources and cannibalism. Growth was also positively correlated with warming sea temperatures but negatively correlated with the Atlantic Multidecadal Oscillation, suggesting contrasting effects of climate warming at different spatial scales. Fishing pressure had a significant but weak negative direct impact on growth. Additionally, path analysis revealed that the selected growth factors were interrelated. Capelin biomass was positively related to sea temperature and negatively influenced by herring biomass, while cod biomass was mainly driven by fishing mortality. Together, these results give a better understanding of how multiple interacting factors have shaped cod growth throughout a century, both directly and indirectly.

Marine spatial planning (MSP) is considered a valuable tool in the ecosystem-based management of marine areas. Predictive modelling may be applied in the MSP framework to obtain spatially explicit information about biodiversity patterns. The growing number of statistical approaches used for this purpose implies the urgent need for comparisons between different predictive techniques. In this study, we evaluated the performance of selected machine learning and regression-based methods that were applied for modelling fish community indices. We hypothesized that habitat features can influence fish assemblage and investigated the effect of environmental gradients on demersal fish diversity (species richness and Shannon-Weaver Index). We used fish data from the Baltic International Trawl Surveys (2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014) and maps of six potential predictors: bottom salinity, depth, seabed slope, growth season bottom temperature, seabed sediments and annual mean bottom current velocity. We compared the performance of six alternative modelling approaches: generalized linear models, generalized additive models, multivariate adaptive regression splines, support vector machines, boosted regression trees and random forests. We applied repeated 10-fold cross-validation, using accuracy as the measure of model quality. Finally, we selected random forest as the best performing algorithm and implemented it for the spatial prediction of fish diversity from the Baltic Proper to the Kattegat. To obtain information on the data reliability and confidence of the developed models, which are essential for MSP, we estimated the uncertainty of predictions with standard deviation of predictions obtained from all the trees in the ensemble random forest method. We showed how state-of-the-art predictive techniques, based on easily available data and simple Geographic Information System tools, can be used to obtain reliable spatial information about fish diversity. Our comparative work highlighted the potential of machine learning method to reduce prediction error in modelling of demersal fish diversity in the framework of MSP.

Otolith biochronologies combine growth records from individual fish to produce long-term growth sequences, which can help to disentangle individual from population-level responses to environmental variability. This study assessed individual thermal plasticity of Atlantic cod (Gadus morhua) growth in Icelandic waters based on measurements of otolith increments. We applied linear mixed-effects models and developed a century-long growth biochronology (1908–2014). We demonstrated interannual and cohort-specific changes in the growth of Icelandic cod over the last century which were mainly driven by temperature variation. Temperature had contrasting relationships with growth—positive for the fish during the youngest ages and negative during the oldest ages. We decomposed the effects of temperature on growth observed at the population level into within-individual effects and among‐individual effects and detected significant individual variation in the thermal plasticity of growth. Variance in the individual plasticity differed across cohorts and may be related to the mean environmental conditions experienced by the group. Our results underscore the complexity of the relationships between climatic conditions and the growth of fish at both the population and individual level, and highlight the need to distinguish between average population responses and growth plasticity of the individuals for accurate growth predictions.

Native to the Ponto-Caspian region, the benthic round goby (Neogobius melanostomus) has invaded several European inland waterbodies as well as the North American Great Lakes and the Baltic Sea. The species is capable of reaching very high densities in the invaded ecosystems, with not only evidence for significant food-web effects on the native biota and habitats, but also negative implications to coastal fishers. Although generally considered a coastal species, it has been shown that round goby migrate to deeper areas of the Great Lakes and other inland lakes during the cold season. Such seasonal movements may create new spatio-temporal ecosystem consequences in invaded systems. To seek evidence for seasonal depth distribution in coastal marine habitats, we compiled all available catch data for round goby in the Baltic Sea since its invasion and until 2017. We furthermore related the depths at capture for each season with the ambient thermal environment. The round goby spend autumn and winter at significantly deeper and offshore areas compared to spring and summer months; few fish were captured at depths < 25 m in these colder months. Similarly, in spring and summer, round goby were not captured at depths > 25 m. The thermal conditions at which round goby were caught varied significantly between seasons, being on average 18.3 °C during summer, and dropping to a low 3.8 °C during winter months. Overall, the fish sought the depths within each season with the highest possible temperatures. The spatial distribution of the round goby substantially overlaps with that of its main and preferred prey (mussels) and with that of its competitor for food (flatfish), but only moderately with the coastal predatory fish (perch), indicating thereby very complex trophic interactions associated with this invasion. Further investigations should aim at quantifying the food web consequences and coupling effects between different habitats related to seasonal migrations of the round goby, both in terms of the species as a competitor, predator and prey.

No abstract

No abstract

The assignment of individual fish to its stock of origin is important for reliable stock assessment and fisheries management. Otolith shape is commonly used as the marker of distinct stocks in discrimination studies. Our literature review showed that the application and comparison of alternative statistical classifiers to discriminate fish stocks based on otolith shape is limited. Therefore, we compared the performance of two traditional and four machine learning classifiers based on Fourier analysis of otolith shape using selected stocks of Atlantic cod (Gadus morhua) in the southern Baltic Sea and Atlantic herring (Clupea harengus) in the western Norwegian Sea, Skagerrak, and the southern Baltic Sea. Our results showed that the stocks can be successfully discriminated based on their otolith shapes. We observed significant differences in the accuracy obtained by the tested classifiers. For both species, support vector machines (SVM) resulted in the highest classification accuracy. These findings suggest that modern machine learning algorithms, like SVM, can help to improve the accuracy of fish stock discrimination systems based on the otolith shape.

Species distribution models for marine organisms are increasingly used for a range of applications, including spatial planning, conservation, and fisheries management. These models have been constructed using a variety of mathematical forms and drawing on both physical and biological independent variables; however, what might be called first‐generation models have mainly followed the form of linear models, or smoothing splines, informed by data collected in the context of fish surveys. The performance of different classes of variables were tested in a series of species occurrence models built with machine learning methods, specifically evaluating the potential contribution of lower trophic level data. Random forest models were fitted based on the classification of the absence/presence for fish and macroinvertebrates surveyed on the US Northeast Continental Shelf. The potential variables included physical, primary production, secondary production, and terrain variables. For accepted model fits, six variable importance measures were computed, which collectively showed that physical and secondary production variables make the greatest contribution across all models. In contrast, terrain variables made the least contribution to these models. Multivariable analyses that account for all performance measures reinforce the role of water depth and temperature in defining species presence and absence; however, chlorophyll concentration and some specific zooplankton taxa, such as Metridia lucens and Paracalanus parvus, also make important contributions with strong seasonal variations. Our results suggest that lower trophic level variables, if available, are valuable in the creation of species distribution models for marine organisms.