To meet the challenges of producing more while lowering impacts on ecosystems, new fish-farming systems have to be designed. To define development strategies, a multi-scale assessment method that generates consistent performance indicators based on the same set of input data is needed. Through two contrasting fish-farming systems and a combination of two methods of environmental assessment, we attempt to define the major components of environmental sustainability and ecological intensification of fish farming. Ecological intensification promotes the production of goods using ecological processes rather than human-induced inputs. The first system is a recirculating system (RAS) of Atlantic salmon that aims to maximize productivity with technology. This system depends highly on external inputs, especially feed and energy. The second one is an extensive fish polyculture (pikes, carps, roaches, etc…) in a pond (EP) that aims to minimize inputs while increasing natural productivity. These two systems were assessed during one production year using two methods Life Cycle Assessment (LCA) and Emergy. LCA (ISO, 2006) is designed to estimate potential environmental impacts throughout a product’s life, from raw material to final disposal, at global and regional scales. Results are presented as traditional midpoint indicators (e.g., GWP, total energy demand, Net Primary Product Use (NPPU), eutrophication, acidification, water dependence and land competition). Emergy accounting (Odum, 1996) includes the contributions of natural systems (sun, rain, groundwater, etc.) in the analysis and can provide indicators to evaluate the efficiency of energy use and its quality during the lifecycle. The traditional Emergy indicators are: the Emergy indicator “Percentage of renewability” that indicates the percentage of renewable resources used by the systems; the Emergy Yield Ratio (EYR) that measures the ability to rely on local resources; Emergy Investment Ratio (EIR) that evaluates the efficiency of the Emergy invested. Environmental Loading Ratio (ELR) that indicates of the level of exploitation of nonrenewable resources compared to renewable ones and Emergy Index of Sustainability (EIS) which is the EYR/ELR ratio.. For the two methods, the functional unit chosen was 1 tonne of living fish. For 1 tonne of living fish, RAS had higher potential impacts for NPPU and all the Emergy indicators. The difference between the two systems can be explained by the relative mass of feed (manufactured feed with high concentrations of fish protein for RAS, biomass and barley grain for EP). As a pond-based system, EP is integrated into the natural environment and less dependent on manufactured resources. However, RAS had lower potential impacts for climate change, eutrophication, land competition and water dependence than EP, which reflects the level of intensification of the systems. The consumption of energy (total cumulative energy demand calculated by LCA) to produce 1 tonne of fish was similar for the 2 systems. Nevertheless, the contributors to this impact differed: for EP 51% of the energy demand came from direct energy use and 48 % for transport, whereas for RAS the impact came from feed (39%) and direct energy use (51%), through electricity consumption. The difference in the percentage of renewability between systems was due to water origin, since water for RAS was pumped from an aquifer whereas water for EP came essentially from rain and water run-off, which are renewable resources from nature. EP has a higher EYR, which means that it depends less on market resources than RAS. The EYR value for EP is consistent with literature values for other extensive systems (Zhang et al., 2011). A high value of EIR, such as found for RAS, indicates that the system is essentially based on nonrenewable natural resources. In the future, with the depletion of non-renewable resources and the increase of their price, the systems with high EIR will be supplanted by systems that have a greater contribution of nature. RAS has a higher value of ELR than EP (7.98 vs. 0.95), its value indicates a moderate environmental impact (Brown and Ulgiati, 2004). The value 1/EIS is higher for RAS than for EP due to its use of external nonrenewable resources and its higher environmental impacts and economic dependency; thus, the RAS system depends less on local resources (renewable or nonrenewable) and induces a larger pressure on the environment. The combination of LCA and Emergy accounting on two contrasting systems provides a perspective of what ecological intensification could mean in aquaculture: a decrease in potential impacts per unit mass of final products, especially for global warming, eutrophication and acidification; a decrease in dependence on market-based and external resources (i.e., more autonomy); and an increase in the use of renewable natural resources and input efficiency. This is particularly true for choices regarding feed ingredients and the origin of energy sources.