Ten “best” models selected in the process of model calibration and evaluation, showing predictor variables; “^2” indicates quadratic terms in a model.

Values for AUC, sensitivity, specificity, and TSS are averages deriving from the 10 k-fold evaluation process. Threshold indicates the value used to test model sensitivity and specificity. AUC = area under the receiver operative characteristic curve; TSS = true skill statistic; AIC = Akaike information criterion; Δ AIC = delta AIC; AIC γ = AIC weight.