DIC is a measure of model fit, ΔDIC measures the relative difference between the best model (ΔDIC = 0) and all subsequent models in the model set, and is the relative likelihood of a model given the
a Covariates with prefix d are measured as distance to the environmental or anthropogenic feature. Results are shown for models with AICΔi ≤ 2.0 and all these models were used to develop model-average
Best fitting model structures considered for logistic regression predictive model with associated Akaike Information Criterion (AICc) scores (corrected for small sample sizes).