Corrigendum to “asynchrony of senescence among phenotypic traits in a wild mammal population” [Exp. Gerontol. 71 (2015) 56–68]

The authors regret to report a miscalculation of two of the twenty traits analysed in this paper, namely Home Range Size (HRS) and Home Range Quality (HRQ). The methods used to calculate the new values were the same as the original paper (Hayward et al., 2015), while the error arose as a result of misalignment of the grids used for the census data (on which is based sheep location) and vegetation cover (on which is based the quality of the habitat). Having re-aligned the grids and recalculated both HRS and HRQ, we repeated all the analyses from the original manuscript.

First, we used generalized linear mixed-effects models (GLMMs) to determine the pattern of age-specific change in both traits. As in the original paper, the best-supported model suggested that the age-specific change differed between the sexes for both HRS (ΔAIC relative to a model with no sex interaction = − 11.13) and HRQ (ΔAIC = − 28.83). As expected, home range size shows the same age-specific pattern as before, with HRS increasing with age in males but declining in females (Fig. 1). However, in contrast to the original results, HRQ increased with age in females and remained constant with age in males, until declining sharply in the oldest age category (Fig. 1).

Fig. 1.

Age-dependent variation in home range size and home range quality wild Soay sheep. Points and bars are means and standard errors estimated from generalized linear mixed-effects models, with females represented by solid lines and symbols, and males represented by broken lines and open symbols. Units are given in the main text of the original manuscript.

Second, we applied generalized additive models (GAMs) to test for variation in ageing trajectories among all of the twenty traits we analysed. We re-ran all of the models from the original paper and once again found that the best-supported model suggested that all traits followed a different ageing trajectory (ΔAIC relative to the next best model, which fitted different ageing trajectories to the different traits, but fitted the same trajectory to the same trait in either sex = − 96.59). GAM estimates supported the same ageing patterns as the GLMMs above (Fig. 2).

Fig. 2.

Predictions from the best-fitting GAM which supported total synchrony with all traits measured followed a different ageing trajectory in males and females. Solid lines show predicted trait values across age; broken lines show predicted standard errors. Black lines represent traits measured in females and grey lines represent traits measured in males.

Overall, our new results provide the same conclusion as the original paper: ageing trajectories vary widely between traits, highlighting the complexity of ageing in natural populations and questioning the hypothesis that fitness related-traits should senesce in a synchronous manner. Our new results suggest that a decline in home range size with age in females is compensated for by an increase in home range quality, while the opposite is true for males. More detailed analyses in the future will be able to determine the causes and consequences of variation in senescence in home range sizes.

The authors would like to apologise for any inconvenience caused.