Extended predictability of metabolic constraints on fish habitat

Predicting the migration of marine life due to climate change is important not only ecologically but also for the management of valuable living marine resources. Anticipating how the habitable area of species will change on decision-relevant time scales, however, remains a challenge. Here we assess the predictability of a normalized metabolic index that expresses habitat-relevant anomalies in metabolic constraints using a coupled physical-biogeochemical prediction system based on an Earth system model. The normalized metabolic index was generally more predictable than temperature, especially in the subsurface tropics where persistent lateral oxygen advection anomalies at the boundary of oxygen minimum zones extended the predictability horizons. Further investigations suggest that interannual catch variations of bigeye tuna in the tropical exclusive economic zones can be anticipated from the predicted metabolic constraints a year in advance, supporting the potential utility of Earth system model-based physiological prediction for the proactive climate-informed management of living marine resources.