Validation of Age Estimation in the Harp Seal, Phoca groenlandica, Using Dentinal Annuli

• Published in: Canadian Journal of Fisheries and Aquatic Sciences
• Main Authors: Bowen, W. D., Sergeant, D. E., Øritsland, T.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Canadian Science Publishing 1983
• Subjects: Harp Seal; Phoca groenlandica
• Online Access: http://dx.doi.org/10.1139/f83-165; http://www.nrcresearchpress.com/doi/pdf/10.1139/f83-165

We investigated the validity and accuracy of age estimation in harp seals, Phoca groenlandica, using a sample of 155 known-age teeth from seals age 3 mo to 10 yr. Under transmitted light, transverse sections of harp seal canine teeth showed distinct incremental growth layers (IGLs) in the dentine. The first growth-layer group (GLG), representing Ist-year growth, consists of two IGLs: an outer layer of opaque dentine, bounded by the neonatal line, and an inner layer of translucent dentine. Subsequent GLGs, each representing 1 yr of growth, generally consist of three IGLs: an outer layer of interglobular dentine deposited during the annual molt in April, a middle layer of opaque dentine formed during the northward spring migration (May–June), and an inner layer of translucent dentine formed from July to March. We show that dentinal GLGs can be used to estimate the absolute age of harp seals. The accuracy of the method decreases with age. Only 72.4% of estimates of 0-group seals were correct using only transverse sections. These errors were virtually eliminated (99.0% correct age determination) when the tooth root was examined. Based on a single examination of a transverse section, the probabilities of correctly estimating age are 0.983, 0.889, 0.817, and 0.553 at ages 1, 2, 3, and 4 + yr, respectively, when clearly inaccurate tag-tooth associations are omitted. The respective probabilities are only slightly higher when age is based on the average of five blind readings, being 1.0, 0.889, 0.833, and 0.625. Beyond age 3 yr, existing data are insufficient to estimate reliably the accuracy of age determined by counting GLGs.