The ability of readers to correctly age rainbow trout was tested using scales from hatchery trout marked and released as spring yearlings. After an initial learning phase, accuracy averaged 78% (range 73-85%). Neither the differences in accuracy among readers nor a tendency to underage was significant. Readers tended to make the same errors on individual scales, but most errors could be detected by discrepancies among pairs of readers. Errors in ageing had little effect on estimates of age structure or length at age 2+ for this data set in which readers discussed the true age of each scale before proceeding to the next. Estimated age structure was 16% 1+, 57% 2+ and 27% 3+ and older, whereas the true age structure was 18% 1+, 50% 2+ and 32% 3+ and older. Estimated and true mean lengths at age 2+ were 304 and 301 mm, respectively. The small size of these discrepancies may be due to prior expectations and the continual feedback on true values for this data set.
Parkinson, Eric A.. 1989. Errors in Ageing Hatchery Rainbow Trout from Small Lakes in Southern British Columbia. Ministry of Environment. Fisheries Technical Circular. FTC87