Applications of Age–Length Keys for Assigning Ages to Juvenile Steelhead

Abstract

AbstractMany population metrics for wild steelhead Oncorhynchus mykiss are unquantified due to complex early life history strategies and a lack of juvenile age data. Juvenile steelhead can rear in freshwater systems for up to 7 years, with overlapping size distributions among age‐classes. As logistics for sampling and aging fish can be challenging, there are benefits to utilizing age–length keys (ALKs). We examined the potential benefits of applying ALKs to unaged juvenile steelhead from a rotary screw trap and to determine whether stratifying trapping data by season could improve accuracy in assigning ages to unaged fish. Additionally, we examined the effects of reducing samples of aged fish using a uniform sampling design in which there was a fixed number of aged fish across all length intervals of the size distributions. Using a leave‐one‐out cross validation method with aged fish, we found that ALKs assigned unbiased ages to unaged fish. When including fish with assigned ages from ALKs, results of mean length‐at‐age and brood year abundance estimates between using proportional‐sampled aged fish and uniform‐sampled aged fish were comparable. Typically, absolute differences in estimated brood year abundances were improved if age‐assigned fish were included, and mean absolute error did not exceed 0.4 years for any of the methods examined. Although seasons had a significant effect on mean length at age, overall brood year estimates of juvenile abundance derived from stratifying data by season did not notably differ from pooling summer and fall seasons together, likely because there was a higher portion of juveniles emigrating in the fall compared to the summer. Age–length keys can benefit trapping operations and aid in filling data gaps in age‐based movement and survival for steelhead by being able to accurately assign ages to unaged fish and include more data for population analyses.