Disentangling the effects of climate, population dynamics, and fishing on the body size of key commercially important species in the Celtic Sea Ecoregion

Abstract Shrinking body size is proposed as a universal response to warming among marine ectotherms. However, in fish, body size is determined by multiple interacting factors which must be disentangled to predict how climate change will impact populations. This study investigated length variability and its drivers in commercially important demersal species across three areas within the Celtic Sea Ecoregion: the North of Ireland/Malin Sea, West of Ireland, and the Celtic Sea, over two decades (2003–2023). Mixed-effects models were used to partition length variability into its intrinsic (age, sex) and extrinsic components. Generalised additive models were used to link length variability to extrinsic factors (temperature, abundance, recruitment, fishing pressure and the North Atlantic Oscillation). Results revealed spatial heterogeneity in length-at-age and varying levels of temporal synchrony in length trends within and between species. Density-dependent mechanisms explained a significant proportion of the variability, with smaller mean sizes associated with increases in abundance and recruitment, while fishing effects were generally weak and mostly consistent with density release among survivors. Across species, length was sensitive to fluctuations in environmental variables at both regional (temperature) and broad (North Atlantic Oscillation) scales. The temperature-size rule predicts faster juvenile growth but smaller adult size under warming. Across species, temperature effects on adult size were consistently negative. However, evidence for positive juvenile responses to warming was limited in gadoids, potentially reflecting the exclusion of age-0 fish in early maturing species. These results highlight that the universality of shrinking body size under warming extends across demersal species in the Celtic Sea Ecoregion, while also underscoring the difficulty of disentangling true climate responses from population-based data in complex ecological systems.