Assessing the environmental impacts of conventional and organic scenarios of rainbow trout farming in France
In France, rainbow trout (Oncorhynchus mykiss) farming traditionally used flow-through systems, which raised concerns about environmental impacts, including limited freshwater availability, and the use of ingredients from intensive agriculture and fishing. To address the growing demand for sustainable food products, there is an increasing interest in organic aquaculture. In this study, we employed an attributional life cycle assessment (LCA) to analyse the environmental impacts of rainbow trout production. We simulated conventional and organic production practices in a hypothetical fish farm to evaluate the differences in environmental impacts at the farm level. The potential impacts were calculated using a product-based functional unit (one tonne of trout) under the two production scenarios and were also expressed using a surface-based functional unit (m2y). Our life cycle impact assessment revealed that organic farming significantly reduced environmental impacts per tonne of trout in seven out of the nine selected impact categories. Notably, freshwater ecotoxicity exhibited the greatest difference, with organic systems showing a 55% decrease. The only exceptions were freshwater eutrophication and water dependence, where organic production led to higher impacts per tonne of trout. In conventional farming, emissions amounted to 14 kg of P eq./tonne, whereas in organic farming, the emissions were slightly higher (15 kg of P eq./tonne). For water dependence, one tonne of trout production in the conventional system mobilized 128 103 m3 vs. 185 103 m3 in the organic system. The environmental benefits of organic production were even more marked when using a surface-based functional unit (m2y). We demonstrated the benefits of organic trout production from an environmental perspective. However, our findings highlight the caution needed when interpreting LCA comparisons of such production systems that can be highly influenced by methodological choices such as the functional unit used.