According to ICES [61], Central Baltic herring is exploited outside of safe biological limits, suffering from small fish size and decreasing stock biomass. Different well-justified hypotheses exist about the reasons behind this reduced growth and the variable productivity of the stock; these competing hypotheses can lead to totally different management conclusions

(e.g., advised increase or decrease of fishing pressure). The Baltic case study aimed at testing alternative probabilistic models and exploring issues around model uncertainty in discussions with stakeholders. Explicitly, the participatory modelling objectives of the Baltic case study were to: – integrate stakeholders’ knowledge into the modelling of Baltic herring population dynamics Six www.selleckchem.com/products/Lapatinib-Ditosylate.html stakeholders (representing managers, scientists, fishers and environmental

NGOs) from four Baltic Sea countries shared Tacrolimus in vivo their knowledge related to the stock assessment and management of the Central Baltic herring. The stakeholders were treated as experts, and everyone built an own model in a separate workshop, independently of the others. Six conceptual biological models (graphical causal system models) were built based on assumptions of the individual stakeholders about causalities and factors influencing the natural mortality, growth, and egg survival mafosfamide of the Central Baltic herring. The estimated strengths of the assumed causalities were expressed as probabilities [64]. The six individual stakeholder models were afterwards pooled by the researcher into a large meta-model using the techniques of Bayesian model averaging, and further combined with scientific data [50]. A parallel modelling task aimed at a better framing of the herring fishery management problem. The stakeholders were asked to extend their biological model by including additional factors they considered important for the Central Baltic herring stock assessment, management objectives, and measures to reach

these objectives [65]. The logic of Bayesian influence diagrams [64] was used to build a qualitative graphical model on herring fishery management with each stakeholder. The stakeholders participated in two workshops. The first was arranged for each stakeholder separately, to build the model independently of the others. The second took place at the end of the project, to present the analysed models to all stakeholders together, to discuss them, and to get systematic feedback. The Baltic case study focused mainly on structural uncertainties, i.e., the basic ignorance about the nature of a complex system, by acknowledging that there are alternative beliefs about the components, dynamics, and inherent internal interactions in the fishery [66].