Comparison of Rainbow Smelt Age Estimates from Fin Rays and Otoliths

Abstract

Rainbow smelt Osmerus mordax, although nonnative, are an important component of the offshore food web in the Laurentian Great Lakes. In Lake Ontario, we estimate ages of rainbow smelt annually to study population dynamics such as year-class strength and age-specific growth and mortality. Since the early 1980s, we have used pectoral fin rays to estimate rainbow smelt ages, but the sectioning and mounting of fin rays are time and labor intensive. Our objective was to assess the feasibility of using otoliths rather than fin rays to estimate rainbow smelt ages. Three readers interpreted the ages of 172 rainbow smelt (60–198 mm total length) based on thin sections of pectoral fin rays, whole otoliths with no preparation, and whole otoliths that had been cleared for 1 month in a 70:30 ethanol : glycerin solution. Bias was lower and precision was greater for fin rays than for otoliths; these results were consistent for comparisons within readers (first and second readings by one individual; three readers were used) and between readers (one reading for each reader within a pair). Both otolith methods appeared to misclassify age-1 rainbow smelt. Fin ray ages had the highest precision and provided the best approximation of age estimates inferred from the Lake Ontario population's length frequency distribution and from our understanding of this population.