Are Morphometrics Sufficient for Estimating Age of Pre-Fledging Birds in the Field? A Test Using Common Terns (Sterna hirundo)

• Published in: PLoS ONE
• Main Authors: Wails, Christy N., Oswald, Stephen A., Arnold, Jennifer M.
• Format: Text
• Language: English
• Published: Public Library of Science 2014
• Subjects: Research Article; Common tern; Sterna hirundo
• Online Access: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4222966; http://www.ncbi.nlm.nih.gov/pubmed/25375105; https://doi.org/10.1371/journal.pone.0111987

Age is a key component of fitness, affecting survival and reproductive capacities. Where it is not possible to study known individuals from birth, morphometrics (predominantly patterns of plumage development for birds) are most often used to estimate age. Although criteria for age estimations exist for many species, the degree to which these criteria improve the precision of estimates remains to be tested, restricting their widespread acceptance. We develop a photographic tool for estimating ages of Common Tern (Sterna hirundo) chicks and test it using 100 human observers of varying prior experience across four breeding colonies (three North American sites and one European site) and under controlled laboratory conditions. We followed the design approach of other morphometric tools, expanding it to create a user-friendly guide (divided into six age groupings). The majority (86%) of observers improved in chick-aging accuracy when using the tool by an average of 20.1% (±1.4 SE) and correctly estimated 60.3% (±1.4) of chick ages. This was similar to the intrinsic aging ability of our best field observer (63.3%). Observers with limited experience showed the greatest increases in chick-aging accuracy over experienced observers who likely had established a method for estimating chick ages prior to using the tool. Even the best observers only correctly estimated ages of chicks 62.9% (±2.8) of the time in the field and 84.0% (±2.9) of the time in the lab when using the tool and typically underestimated ages. This indicates that developmental variation between individual chicks can prevent completely reliable age estimates and corroborates the few existing data that suggest that morphometric criteria fail to achieve robust levels of accuracy and may introduce error into studies that rely on them. We conclude that novel approaches for estimating age, not only morphometric criteria, must be pursued.