Age estimation of live arctic foxes Vulpes lagopus based on teeth condition

Aging individuals is a prerequisite to many studies in wildlife biology. In mammals, the most accurate methods are based on cement lines analysis and require tooth extraction. Here, we adapted a method based on tooth wear assessment in live individuals and calibrated against cement lines counts on dead individuals. We developed and tested this less invasive method in arctic foxes Vulpes lagopus. We photographed dentition of live individuals captured in 2006-2013 and of skulls collected opportunistically in 2003-2013. Five observers assessed twice the tooth wear of 245 sets of pictures (left, frontal, and right views of the dentition for 179 captures and 66 skulls, with a mean of 7.2 pictures per set) to obtain a teeth condition index (TCI) for each set. TCI was highly repeatable, with an intraclass correlation coefficient averaging 0.89 and 0.83 within and between observers, respectively. We then used 44 known-aged individuals to predict fox age from TCI through linear regression (age = 0.62 x TCI + 1.04, r(2) = 0.64). We tested through jackknife validation the ability of this model to accurately age foxes. The rate of correct classification of exact age was 39%, while 80% and 100% of predicted ages were within one and two years of the true age, respectively. The model predicted correctly the age class (young, prime-age adult, old) of individuals 75% of the time. We conclude that our proposed method, which is much less invasive than those based on tooth extraction, can allow progress in the study of demography of arctic foxes, and potentially other mammal species.