Table 1. Predictor variables and model AIC values for the probability that a Magellanic penguin chick died in a storm at Punta Tombo, Argentina, 1983–2010.
| Predictor variables | k | AIC | ΔAIC | LL | AIC weight |
| a, a2, r, a*r, a2*r, a2*l, a2*r*l (no Low, no interactions between Low and Age or Rain,3-way interaction uses Age2) | 8 | 622.79 | 0 | −303.4 | 0.47 |
| a, a2, r, l, a*r, a2*r, a2*l, a2*r*l (includes Low, but no interactions betweenLow and Age or Rain, 3-way interaction uses Age2) | 9 | 624.03 | 1.23 | −303.0 | 0.25 |
| a, a2, r, l, a*r, a*l, a2*r, a2*l, a2*r*l (no Rain-Low interaction, 3-way interaction uses Age2) | 10 | 625.28 | 2.48 | −302.6 | 0.14 |
| a, a2, r, a*r, a*l, r*l, a2*r, a2*l, a2*r*l (no Low, 3-way interaction uses Age2) | 10 | 626.77 | 3.98 | −303.4 | 0.06 |
| a, a2, r, l, a*r, a*l, r*l, a2*r, a2*l, a2*r*l (3-way interaction uses Age2) | 11 | 627.03 | 4.23 | −302.5 | 0.06 |
| a, a2, r, l, a*r, a*l, r*l, a2*r, a2*l, a*r*l, a2*r*l (full model) | 12 | 629.03 | 6.23 | −302.5 | 0.02 |
| a, a2, r, l, a*r, a*l, r*l, a2*r, a2*l, a*r*l (3-way interaction uses Age) | 11 | 635.45 | 12.65 | −306.7 | <0.001 |
| a, a2, r, l, a*r, a*l, r*l, a2*r, a2*l (no 3-way interactions) | 10 | 638.73 | 15.94 | −309.4 | <0.001 |
| a, a2, l, a*r, a*l, r*l, a2*r, a2*l, a2*r*l (no Rain, 3-way interaction uses Age2) | 10 | 648.55 | 25.76 | −314.3 | <0.001 |
| a, r, l, a*r, a*l, r*l, a2*r, a2*l, a2*r*l (no Age2, 3-way interaction uses Age2) | 10 | 651.61 | 28.82 | −315.8 | <0.001 |
| a, a2, r, l, a*r, a*l, r*l, a*r*l (no Age2 interactions) | 9 | 655.04 | 32.24 | −318.5 | <0.001 |
| a, a2, r, l, r*l, a2*r, a2*l (no Age interactions and no 3-way interaction) | 8 | 661.17 | 38.37 | −322.6 | <0.001 |
| a, a2, r, l, a*r, a2*r (no Low interactions) | 7 | 680.25 | 57.45 | −333.1 | <0.001 |
| a, r, l, a*r, a*l, r*l, a*r*l (no Age2 and no Age2 interactions) | 8 | 681.62 | 58.82 | −332.8 | <0.001 |
| a, a2, r, a*r, a2*r (no Low and no Low interactions) | 6 | 681.87 | 59.08 | −334.9 | <0.001 |
| a, a2, r, l, a*l, a2*l (no Rain interactions) | 7 | 721.75 | 98.95 | −353.9 | <0.001 |
| a, a2, r, l (no interactions) | 5 | 724.31 | 101.51 | −357.2 | <0.001 |
| a2, r, l, a*r, a*l, r*l, a2*r, a2*l, a2*r*l (no Age, 3-way interaction uses Age2) | 10 | 726.83 | 104.03 | −353.4 | <0.001 |
| a2, r, l, r*l, a2*r, a2*l (no Age and no Age interactions, no 3-way interaction) | 7 | 859.00 | 236.21 | −422.5 | <0.001 |
| r, l, r*l (no Age, Age2 , or interactions) | 4 | 923.56 | 300.77 | −457.8 | <0.001 |
| a, a2, l, a*l, a2*l (no Rain or Rain interactions) | 6 | 1054.93 | 432.13 | −521.5 | <0.001 |
Multiple logistic regression was used, grouping on nest, with robust standard errors on 2482 chicks in 590 nests with 233 storms. a = chick age (days) on date of storm (chicks that did not die in a storm were randomly assigned to a storm), a2 = age squared, r = rain, l = low temperature, * indicates interaction terms, k = number of parameters, AIC = Akaike’s Information Criterion, ΔAIC = the difference between AIC and the lowest AIC, LL = model log-likelihood, AIC weight = the probability that the model is the best model given the data and the set of candidate models. Pseudo R 2 of the best model was 0.57.