Search

The Barents Sea ecosystem components and services are under pressure from climate change and other anthropogenic impacts. Following an Ecosystem-based management approach, multiple simultaneous pressures are addressed by using integrative strategies, but regular prioritization of key issues is needed. Identification of such priorities is typically done in a 'scoping' phase, where the characterization of the social-ecological system is defined and discussed. We performed a scoping exercise using an open and flexible multi-stakeholder approach to build conceptual models of the Barents Sea social-ecological system. After standardizing vocabulary, a complex hierarchical model structure containing 155 elements was condensed to a simpler model structure containing a maximum of 36 elements. To capture a common understanding across stakeholder groups, inputs from the individual group models were compiled into a collective model. Stakeholders' representation of the Barents Sea social-ecological system is complex and often group specific, emphasizing the need to include social scientific methods to ensure the identification and inclusion of key stakeholders in the process. Any summary or simplification of the stakeholders' representation neglects important information. Some commonalities are highlighted in the collective model, and additional information from the hierarchical model is provided by multicriteria analysis. The collective conceptual stakeholder model provides input to an integrated overview and strengthens prioritization in Ecosystem-based management by supporting the development of qualitative network models. Such models allow for exploration of perturbations and can inform cross-sectoral management trade-offs and priorities. • Sector-based stakeholder groups characterize social-ecological systems differently. • Open and flexible approach promote detailed and novel stakeholder input for EBM. • Stakeholders identify cross-sectoral benefits in flexible scoping process. • Pollution and governance are considered key concerns by stakeholders. • Cross-sector conceptual models support an integrated overview and EBM implementation. [ABSTRACT FROM AUTHOR]

A food web topology describes the diversity of species and their trophic interactions, i.e. who eats whom, and structural analysis of food web topologies can provide insight into ecosystem structure and function. It appears simple, at first sight, to list all species and their trophic interactions. However, the very large number of species at low trophic levels and the impossibility to monitor all trophic interactions in the ocean makes it impossible to construct complete food web topologies. In practice, food web topologies are simplified by aggregating species into groups termed trophospecies. It is not clear though, how much simplified versions of food webs retain the structural properties of more detailed networks. Using the most comprehensive Barents Sea food web to date, we investigate the performance of methods to construct simplified food webs using three approaches: taxonomic, structural and regular clustering. We then evaluate how topological properties vary with the level of network simplification. Results show that alteration of food web structural properties due to aggregation are highly sensitive to the methodology used for grouping species and trophic links. In the specific case of the Barents Sea, we show that it is possible to preserve key structural properties of the original complex food web in simplified versions when using taxonomic or structural clustering combined with intermediate 25% linkage for trophic aggregation. [ABSTRACT FROM AUTHOR]

Understanding and predicting patterns arising from the dynamics of marine food webs is central to trophic and community ecology and numerical models of food webs constitute a primary tool to simulate these dynamics. Food web simulation models are often highly complex while at the same time often too constrained to reproduce the level of variability observed in real systems. The recently developed non-deterministic network dynamics (NDND) modelling framework has been suggested as a simulation alternative, which can generate multiple patterns of food web variability despite great structural simplicity. Two important aspects of the NDND modelling framework remain unexplored: first the derivation of model input parameters from empirical or theoretical studies and second the evaluation of the model simulations against observations. We provide a methodology for the derivation of model parameters based on empirical observations, the metabolic theory of ecology and life-history theory and apply it to the specific case of the Barents Sea food web. We then evaluate the ability of the NDND simulations to reproduce a wide range of patterns of food web dynamics against observations collected in the Barents Sea during 28 years. Patterns emerging from the simulations include trends and cycles in biomass, trophic levels and transfer efficiency, density-dependent growth, top-down vs bottom-up oscillations, ecosystem level stability and synchrony and trophic functional responses. The ability of the NDND to generate so many patterns observed empirically in the Barents Sea is remarkable given that it is based only on random trophic interactions operating within few constraints set by ecological rules. Our results show that investigations of food web dynamics in marine ecosystems, including the definition of reference states and responses to climate and exploitation pressures, may be achieved with models that are structurally simple and based on few well-established assumptions. [ABSTRACT FROM AUTHOR]

• Complex ecological models are increasingly used for societal purpose. • It is challenging for users to appraise the performance of model applications. • The OPE protocol (Objectives, Patterns, Evaluation) documents model evaluation. • It promotes standardisation and transparency of model evaluation. • The OPE can guide modelling developments and reviews of ecological modelling studies. Numerical models of ecological systems are increasingly used to address complex environmental and resource management questions. One challenge for scientists, managers, and stakeholders is to appraise how well suited these models are to answer questions of scientific or societal relevance, that is, to perform, communicate, or access transparent evaluations of ecological models. While there have been substantial developments to support standardised descriptions of ecological models, less has been done to standardise and to report model evaluation practices. We present here a general protocol designed to guide the reporting of model evaluation. The protocol is organised in three major parts: the objective(s) of the modelling application, the ecological patterns of relevance and the evaluation methodology proper, and is termed the OPE (objectives, patterns, evaluation) protocol. We present the 25 questions of the OPE protocol which address the many aspects of the evaluation process and then apply them to six case studies based on a diversity of ecological models. In addition to standardising and increasing the transparency of the model evaluation process, we find that going through the OPE protocol helps modellers to think more deeply about the evaluation of their models. From this last point, we suggest that it would be highly beneficial for modellers to consider the OPE early in the modelling process, in addition to using it as a reporting tool and as a reviewing tool. [ABSTRACT FROM AUTHOR]

A novel approach to model food-web dynamics, based on a combination of chance (randomness) and necessity (system constraints), was presented by Mullon et al. in 2009. Based on simulations for the Benguela ecosystem, they concluded that observed patterns of ecosystem variability may simply result from basic structural constraints within which the ecosystem functions. To date, and despite the importance of these conclusions, this work has received little attention. The objective of the present paper is to replicate this original model and evaluate the conclusions that were derived from its simulations. For this purpose, we revisit the equations and input parameters that form the structure of the original model and implement a comparable simulation model. We restate the model principles and provide a detailed account of the model structure, equations, and parameters. Our model can reproduce several ecosystem dynamic patterns: pseudo-cycles, variation and volatility, diet, stock-recruitment relationships, and correlations between species biomass series. The original conclusions are supported to a large extent by the current replication of the model. Model parameterisation and computational aspects remain difficult and these need to be investigated further. Hopefully, the present contribution will make this approach available to a larger research community and will promote the use of non-deterministic-network-dynamics models as ‘null models of food-webs’ as originally advocated. [ABSTRACT FROM AUTHOR]

Abstract: Modern fisheries research and management must understand and take account of the interactions between climate and fishing, rather than try to disentangle their effects and address each separately. These interactions are significant drivers of change in exploited marine systems and have ramifications for ecosystems and those who depend on the services they provide. We discuss how fishing and climate forcing interact on individual fish, marine populations, marine communities, and ecosystems to bring these levels into states that are more sensitive to (i.e. more strongly related with) climate forcing. Fishing is unlikely to alter the sensitivities of individual finfish and invertebrates to climate forcing. It will remove individuals with specific characteristics from the gene pool, thereby affecting structure and function at higher levels of organisation. Fishing leads to a loss of older age classes, spatial contraction, loss of sub-units, and alteration of life history traits in populations, making them more sensitive to climate variability at interannual to interdecadal scales. Fishing reduces the mean size of individuals and mean trophic level of communities, decreasing their turnover time leading them to track environmental variability more closely. Marine ecosystems under intense exploitation evolve towards stronger bottom–up control and greater sensitivity to climate forcing. Because climate change occurs slowly, its effects are not likely to have immediate impacts on marine systems but will be manifest as the accumulation of the interactions between fishing and climate variability — unless threshold limits are exceeded. Marine resource managers need to develop approaches which maintain the resilience of individuals, populations, communities and ecosystems to the combined and interacting effects of climate and fishing. Overall, a less-heavily fished marine system, and one which shifts the focus from individual species to functional groups and fish communities, is likely to provide more stable catches with climate variability and change than would a heavily fished system. [Copyright &y& Elsevier]