Influence of migration timing and total length on relative reproductive success
Mean arrival date at the weir for males was earlier than females (mean Julian date of arrival: females = 162, males = 158; Figure 4A ). Total offspring production for males was highest for those that arrived midseason relative to early or late arriving males (Figure 4B ). We observed a greater variance in arrival date for males (s 2 = 63.7) than females (s 2 = 60.2). The total number of offspring produced per week peaked during the first week of June for both females and males, however, relative offspring production on a weekly basis (i.e., the sum of all offspring produced by adults that arrived within a given week) was significantly different as a function of sex (χ2 = 550.22, df =6, p -value = 2.2 x 10-16).
The relative number of offspring produced per adult increased as a function of size for both females and males (Figure 5 ), although the strength of this trend was greater for males. Additionally, we identified that the largest adult males arrived earlier than smaller males and were responsible for producing the greatest relative number of offspring (Figure 5 ). The best model which explained relative reproductive success was one that included both total length and date of arrival for both males and females (Table 1 ). The relative likelihood of the model which included length and date was 0.796 for females and 0.858, both of which were much larger than the next best fit model (∆AICc > 2).
Effective population size (N e) and number of breeders (N b)
The point estimate of N e was 868 (787 – 959, 95% CI) and N b was 576 (513 – 650, 95% CI). The ratio of N b/N e was 0.66.