In a 16‐week study, we stocked duplicate groups of 3.3‐g Atlantic salmon (Salmo salar) in 25‐fish increments (25 to 100 fish/6.5‐L jar) and in 250‐fish increments (250 to 1,000 fish/144‐L trough), in an ozonated, 100% closed water reuse system, to examine changes in fish performance in response to rearing at different densities and loadings within the system. Maximum rearing density (kg!m³) reached 219 in jars and 125 in troughs. Maximum loading (ratio of fish weight to water flow) reached 0.45 kg/(L/min) in jars, and 0.79 kg/(L/min) in troughs. A rise in water temperature from 18.9 to 21°C during weeks 11 and 12 was accompanied by a drop in effluent dissolved oxygen (DO), followed by an increase in mortality and a decline in biweekly weight gain. Later declines in water temperature to 18.4°C, and reductions in biomass, were followed by reduced mortality and increased weight gain. In both jars and troughs, mortality increased when dissolved oxygen dropped to 6.5 mg/L or lower, water temperature reached 19.6°C, and biomass reached 50–75 kg/m³, or 0.1–0.2 kg/(L/min) flow. Changes in DO, weight gain, and mortality per unit change in density (kg biomass/m³), were lower in jars than in troughs, but changes in these responses in terms of flow (kg/[L/min]) were lower in troughs. Results of this study suggest that, for optimum growth, survival, and feed conversion of Atlantic salmon in an ozonated water reuse system, dissolved oxygen should be at least 6.5 mg/L. Maintenance of an acceptable level of oxygen should minimize detrimental effects offish density or elevated water temperature.