Use of Parentage Assignment and DNA Genotyping to Validate Scale Analysis for Estimating Steelhead Age and Spawning History

Abstract

Sound fisheries management depends on understanding life history characteristics, such as age at maturity, migration, and spawning history; hard parts, such as scales and otoliths, are commonly used to estimate values for these traits. Validation of these techniques is often difficult but critical because errors in mean values for a population can result in erroneous estimates of sustainable exploitation rates. In this study, we compared information on age and spawning history derived from genetic analysis with information from scale analysis for 1,836 individual steelhead Oncorhynchus mykiss representing two life history stages (smolt and mature adult); these fish were of hatchery and wild origin and were sampled from two rivers in Washington State over a period of 19 years. Aging error rates were less than 5% for both wild smolts and hatchery adults, but the error rate for wild adults was much higher (13%; 95% confidence interval [CI] = 1.82–29.22%). Adult scale ages were biased; scale readers typically overestimated the ages of younger fish and underestimated the ages of older fish. In addition to the errors in age estimation, repeat spawners were misclassified as virgin spawners 6.5% of the time (95% CI = 0.79–20.26%). The higher error rates obtained in using scales for aging and identifying repeat spawning in wild adults may partly stem from the more complex life history relative to that of smolts or hatchery-origin adults; these results highlight the need for verification of scale analysis, especially for wild steelhead populations.