Ecological Consequences of Rapid Life-History Evolution in Rainbow Trout

There is growing appreciation for ecological and evolutionary processes to potentially overlap in time. For instance, studies of fishing induced selection suggest morphological and life-history traits can respond over only a few generations. However, the ecological consequences of rapid evolution are relatively understudied and may well be important, especially in cases where divergent phenotypes of the same species occupy different habitats. Faculatively anadromous species such as steelhead/rainbow trout (Oncorhynchus mykiss) represent a natural system in which to study such eco-evolutionary dynamics. Across their range recently diverged anadromous and resident forms of O. mykiss are often separated in space by impassable barriers (e.g. waterfalls, dams). Here we test the ecological consequences of rapid evolution by making use of a natural experiment where fish from a predominantly anadromous population were introduced above a barrier waterfall in 1910 and have subsequently evolved a resident life-history.

We present results from common garden experiments and field observation which provide evidence that hard selection against anadromy in the derived, above falls population has 1) resulted in lower incidence of smolting for a given growth rate relative to the ancestral, below falls population, and 2) led to a 5:1 difference in density between the below falls and above falls populations. We test the ecological consequences of these two evolutionary responses by performing an in-stream mesocosm study. Mesocosms were stocked with either above or below fall origin fish at high and low densities, or left fish free as a control. Community and ecosystem properties were measured over the two month study and results analyzed by planned contrasts to assess the effects of ecology (presence/absence of fish) and evolution (phenotype, density, and their interaction). Fish presence and density significantly affected several ecosystem properties, including accumulation of total solids, proportion substrate covered in silt, and decomposition rate of leaf litter. We did not observe direct ecosystem effects of the phenotypes; however, mortalities of above falls fish at high density were significantly greater than those of below falls fish. This suggests the evolutionary trade-off of lower densities in resident populations will indirectly alter ecosystem processes.

It is important for management actions to be evaluated for efficacy at timescales that account for the ecological and evolutionary processes involved. For this reason, efforts to reestablish gene flow between anadromous and resident populations should consider the rate at which life-histories will (re)evolve and the manner in which ecological processes will respond in concert.