Decision Support Approaches to Advance the Understanding and Management of European Polyculture Aquaculture

Abstract

Aquaculture in Europe is expected to grow while policy requires lower ecological impacts. Polyculture is promoted for efficient resource use, yet farm-scale ecological functioning and management effects remain uncertain and context-dependent. The thesis investigates two contrasting European polyculture systems and how management choices alter their performance. Nutrient connectivity of a coastal salmon-kelp IMTA was tested over two production years (one kelp-only, one kelp-salmon) using stable isotope analysis. The isotopic signal of salmon feed was apparent in farmed kelp during the kelp-salmon year and dissipated as salmon harvesting began. Bayesian mixing modelling identified fish feed as the dominant nitrogen source of farmed kelp in the IMTA. An integrated IMTA model was used to couple salmon metabolism, hydrodynamics, and kelp growth. Production scenarios were applied to assess bioremediation potential. Extending the kelp production cycle increased modelled kelp nitrogen assimilation to 3.07% of salmon metabolic nitrogen. Achieving 10% removal of salmon wastes required 36 kelp lines in the most favourable scenario, which remained within the licensed area. Several diversely managed earthen carp ponds were represented with mass-balanced Ecopath models to quantify differences in trophic structure and flows. Ponds varied in trophic structure according to management type. Several polyculture scenarios were applied to each pond management type and Ecological Network Analysis derived system-level ecological indicators. Out of 11 indicators, 5 were selected as most relevant to pond management. Low-level polyculture with herbivorous carp gave little benefit where macrophytes were scarce. Under a “maximum” polyculture scenario, mass balance remained feasible only by diverting grass carp consumption to detritus and feed; plankton resources capped silver carp; indicators showed ecosystem fragility under extreme herbivory. Across both systems, models serve as tools for decision-support to help understand system functioning. This enables the mapping of feasible operating space, quantifying trade-offs and focusing effort on achievable change.