Abstract
Otolith trace element analysis is a powerful tool for studying fish migration between juvenile and adult habitats. To track catchment-scale migration, trace element signatures in juvenile otoliths are compared to the otolith natal region of adults, with the aim of matching adults to their natal habitat. However, the stability of signatures in natal habitat over time, crucial for this method to work across generations, has rarely been tested. To evaluate spatiotemporal stability of natal stream trace element signatures in otoliths, a four-year study in New Zealand's lower Clutha River/Mata-Au was executed. 602 brown trout juveniles from 20 tributaries were sampled and trace element analysis was performed, enabling comparisons of the signatures within each stream over time. The findings suggest the signatures of the elements Mn, Sr, and Ba are generally stable over four-years. This means that, it is not necessary to analyse juvenile otoliths from each year. Spatial variability of Mn, Sr and Ba was tested by using Random Forest and Linear Discriminant Analysis models, which assigned juvenile trout to the correct stream in 61 % and 52 % cases, respectively. Grouping the sites using Nearest-Neighbour Cluster Analysis resulted in 93 % accuracy for both models. Collection of juveniles from key spawning streams across 2–3 years should be sufficient to establish reference signatures for each tributary that will facilitate tracking of adult migration over several years.
• Trace elements Mn, Sr, and Ba in brown trout otoliths showed both temporal stability and spatial variability in this catchment.
• Random Forest and Linear Discriminant Analysis models achieved an accuracy of 93 % after grouping the sample locations.
• Temporal variation does not prevent using juvenile otoliths from different years to determining natal origins in this catchment.
• Trace element analysis is a powerful tool for tracking brown trout migration in New Zealand.