Table 1.
Description of variables used in the global models.
Fig 1.
Body condition of adult beavers increased continuously with age.
The relationship between age and body weight (controlling for body length and time of year) of beavers ≥ 3 years old where exact age is known (A), or only minimum age is known (B). Lines represent predictions (thick lines) and their 95% prediction intervals (PIs, thin lines) averaged from the top set of LMM models. Points depict raw data. Ages 13–16 were combined. Data pertaining to males are coloured blue and females red.
Table 2.
Model estimates of effects on adult body weight.
Table 3.
Model selection tables examining the effects on body weight, controlling for body length and time of year.
Table 4.
Model selection for probability of reproduction and reproductive success.
Table 5.
Model estimates on the probability of reproduction and reproductive success.
Fig 2.
Adult females from low quality territories are significantly less likely to reproduce as they age.
The relationship between probability of reproduction and maternal minimum age (mmage) for the four levels of territory quality (TQ4). Colours denote TQ4 where blue = 1, green = 2, orange = 3 and red = 4. Lines present predictions and shading denotes areas within the 95% prediction intervals, averaged from the top set of GLMM models.
Fig 3.
During resource poor years offspring from younger mothers are smaller.
The relationship between offspring body length and maternal minimum age in dry (‘○’, red) and wet (‘+’, blue) years. Data points represent raw data (body length controlling for age) while lines provide predictions and 95% prediction intervals derived from the averaged top set of LMM models.
Fig 4.
Offspring size was comparable for all litter sizes irrespective of mother age.
The relationship between maternal age 3–8 and ≥9) and offspring body length (BL, cm) at different litter sizes (LS) where black indicates LS = 1; blue LS = 2 and red LS ≥ 3. Solid lines indicate predictions and dashed lines indicate 95% prediction intervals. Predictions are from model averaged estimates of the most supported (ΔAICc < 4) models. The large overlap in prediction intervals suggests that the effects of LS and maternal age are marginal.
Table 6.
Model selection for offspring body weight and body length.
Table 7.
Model estimates on offspring body weight and body length.