Correction to: Polar Biology (2021) 44:1–15 https://doi.org/10.1007/s00300-020-02766-4

A coding error resulted in omission of a candidate density surface model for goose abundance from AICc model comparisons. Inclusion of the omitted model resulted in it meeting the ∆AICc < 2 threshold which we used for model interpretation. Interpretation of the omitted model was very similar to the model presented in the paper, but differed in that it contained an interaction term for freshwater and year which indicated associations between geese and freshwater weakened in 2016 and 2017. This effect was opposite to our prediction that geese would associate with freshwater more strongly in warm and dry years; however, the conclusion is tentative due to overlap of the 95% confidence intervals for the terms with zero, and the low number of years of study. Summary information for the omitted model is provided in Table 1. Additionally, in the Materials and Methods section on density surface modelling (page 4), the Latin name for Sandhill Crane was incorrectly given as Cygnus columbianus. It should be Antigone canadensis.

Table 1 Model results for omitted candidate model for goose abundance from Hawkshaw et al. (2021)
Full size table