Rapid evolution of migratory life histories: From steelhead to trout in 25 generations

Background/Question/Methods

Quantifying rapid evolution will give insight into how populations respond to novel anthropogenic perturbations. Understanding how threshold traits respond to selection is particularly important and challenging as they are involved in important ecological processes, including dispersal and migration strategies (stay or go), mating tactics (fighter or sneaker), and life-history events (e.g. size and age at maturation). Threshold traits produce discrete traits conditional on the organism’s state; the alternative state is expressed when a continuously distributed, environmentally sensitive polygenic trait (the “cue”; e.g. body size) exceeds an associated threshold.  Thus, novel selection regimes may alter frequencies of ecologically important traits that were previously adaptive. We utilize a natural “experiment” established through the historical transplant of steelhead trout (Oncorhynchus mykiss) above a waterfall in a coastal California watershed to test whether the life-history decision between residence and migration is a growth-dependent threshold trait capable of responding to selection against migration in the above waterfall population. In a common garden experiment with fish from both populations, we use an ‘animal model’ to measure the additive genetic variation and heritability of growth rate and threshold size.

Results/Conclusions

O. mykiss rapidly evolved from the migratory to residency life-history. We found both components of the threshold trait are heritable and have evolved in the predicted direction in the above barrier population. Fish from the above fall population were both 19 mm smaller and 14 g lighter than fish from the below fall population. Heritability of both traits was ~0.5, however, there was high uncertainty in these estimates. These results are consistent with the prediction that selection above the waterfall favors slower growing fish, as faster growing fish are more likely to exceed their threshold and migrate out of the population. We also found support for larger threshold sizes in the above waterfall population, and this trait was highly heritable (> 0.85). This finding is consistent with the prediction that above the waterfall fish with smaller threshold sizes will be selected against as they are more likely to exceed their threshold size. Together, the evolved differences in growth rate (slower) and threshold size (larger) provide a mechanistic explanation for the significantly lower incidence of the migratory phenotype observed in the above waterfall population. Therefore, migratory barriers (e.g. waterfalls, dams) may elicit rapid evolution toward increased frequency of the resident life-history.