Research Document 2021/023

Evaluating otolith preparation methods for anadromous Arctic Char: establishing an age estimation protocol and comparing historical with contemporary data

By Gallagher, C.P., Wastle, R.J., and Howland, K.L.

Abstract

Reliable age estimation of anadromous Arctic Char,Salvelinus alpinus, is important for assessment and management of populations. We examined precision (coefficient of variation- CV% and percent agreement) and bias of within-reader age estimates between whole and thin-section, and whole and grind otolith preparation methods based on single reads using samples from two populations of Arctic Char (Tatik Lake and Hornaday River, Northwest Territories) collected between 2010 and 2012. Additionally, standardized criteria were used to assess between-method differences in age reader confidence to determine whether perceived improvement in confidence resulted in different age estimates. Whole vs thin-section CV was 4.6% while exact percent agreement was 51–56%. Bias between whole and thin-section methods was evident among older age classes where the whole method tended to under-estimate compared to thin-sectioning. A combination of whole and thin-section otolith preparation methods is recommended to minimize sample preparation time while maintaining confidence in age estimates. The age estimation protocol for Tatik Lake and Hornaday River would use the whole method for ≤ age-12 and ≤ age-9, respectively and use the thin-section method thereafter. Any otolith, regardless of age, would be thin-sectioned if the confidence associated with the whole read was low as thin-sectioning produced a greater frequency of high-confidence age estimates compared to whole. While age estimates from whole otoliths read with low confidence were not significantly different when thin-sectioned and read with high confidence, thin-sectioning improved reader confidence for 50.5–58.4% of the whole reads with low confidence. Similar findings were observed between whole and grind otolith preparation methods. Finally, we evaluated bias between the whole method (used historically by one age reader) and the combination of whole and thin-section method (used recently by another age reader), to assess the comparability of historical and contemporary age data. Between-reader CV was 6.0% for Tatik Lake and 6.7% for Hornaday River, while exact percent agreement was 45% and 54%, respectively. The contemporary method tended to produce older age estimates resulting in differences in age frequency distributions, maximum age, and annual mortality estimates. However, there was no significant difference in von Bertalanffy growth parameters for either location. Researchers should consider the implication of these differences when working with both historic and contemporary data sets for these locations and any others with a similar difference in the methods used to assign age estimates.

Research Document 2021/023 (PDF, 1,147 KB)

Accessibility Notice

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.