Abstract
Marine fish commonly move across distinct habitats throughout their lifetimes, particularly during larval stages, when they are particularly difficult to track. Such transitions are necessary as environmental demands and predation pressures change dramatically with increases in body size. Fish larvae suffer high natural mortality rates, and varying environmental experiences during early life can significantly impact population demographics. Understanding the nature of habitat use, habitat connectivity and ontogenetic timing of habitat transitions is therefore key to establishing effective management practices covering whole life histories. Tracing individual movements of tiny larvae is impossible using conventional tagging approaches. However, otolith chemistry is an attractive alternative where movements occur across suitable chemical gradients. The stable isotope composition of oxygen in otolith aragonite is a valuable tracer, predictably related to widely measured and/or modelled ocean variables (temperature and salinity). Emerging high‐resolution analytical methods enable larval otolith subsampling and the generation of time‐resolved isotopic records that potentially reveal larval movements. However, such analyses are logistically challenging and costly, preventing their routine use. A priori determination of when such approaches are likely to generate useful information would enable tailored and cost‐effective design for high‐resolution otolith isotope projects. Here, we develop isotope‐enabled particle drift models to predict whether high‐resolution otolith analyses of northeast Atlantic mackerel ( Scomber scombrus ) larvae have the potential to discriminate among potential spawning origins and associated larval drift trajectories. Larval drift simulations revealed distinct drift patterns associated with four spawning areas: South, West 1, West 2 and the North Sea. Drift pathways and timescales varied significantly between these regions, with larvae in the South remaining near natal spawning areas, while those in the West 1 and West 2 regions exhibited more complex and extensive northward movements. Predicted otolith δ 18 O values associated with these drift trajectories differed significantly between some spawning areas, with higher values observed in West 1 and West 2 compared with the North Sea. We identify key regions of the otolith (corresponding to periods of drift) most likely to discriminate among spawning origins, and therefore potentially focussing targeted high‐resolution analyses. Our results suggest that a combination of low‐resolution and high‐resolution otolith sampling could effectively discriminate between different spawning areas and improve our understanding of larval drift patterns. Our coupled model approach is directly transferrable to species with pelagic larval stages and may help to focus resources on species and regions where larval drift questions are reasonably tractable using stable isotope tracers.