Water temperature affects somatic growth, body condition and oxygen and carbon otolith isotopes of stout whiting (Sillago robusta)

Ocean warming will continue to affect the growth, body condition and geographic distributions of marine fishes and understanding these effects is an urgent challenge for fisheries research and management. Determining how temperature is recorded in fish otolith carbonate, provides an additional chronological tool to investigate thermal histories, preferences and patterns of movement throughout an individual's life history. The influence of three water temperature treatments (22°C, 25°C, and 28°C) on hatchery-reared juvenile stout whiting, Sillago robusta, was tested using a controlled outdoor mesocosm system. Fish were measured for change in length and weight, and body condition was determined using bioelectrical impedance analysis. Sagittal otoliths were analysed for stable oxygen (δ¹⁸O_otolith) and carbon (δ¹³C_otolith) isotopes via isotope ratio mass spectrometry. Whiting kept at 22°C were significantly smaller and had diminished body condition compared to fish in 25°C and 28°C, which did not significantly differ from each other. The δ¹⁸O _otolith values of stout whiting demonstrated a negative temperature-dependent fractionation relationship which was similar in slope but had a different intercept to the relationships reported for inorganic aragonite and other marine fish species. The δ¹³C _otolith values also showed a negative relationship with water temperature, and the calculated proportion of metabolic carbon M in otoliths differed between fish reared in the coolest (22°C) and warmest (28°C) temperature treatments. Overall, the results suggest that stout whiting may have reached an upper growth threshold between 25°C and 28°C, and that growth and body condition may be optimised during warmer seasons and toward the northerly regions of their distribution. Otolith oxygen thermometry shows promise as a natural tracer of thermal life history, and species-specific fractionation equations should be utilised when possible to prevent errors in temperature reconstructions of wild-caught fish.