Thursday, August 9, 2007

PS 61-86: Ten years of varying lake level and selection on size-at-maturity in sockeye salmon

Stephanie M. Carlson and Thomas P. Quinn. University of Washington

Despite the ubiquity of studies quantifying the strength and form of selection in nature, rarely is the ecological context for contemporary selection understood. Here we report a case where lake level is a selective factor acting on sockeye salmon body size-at-maturity because low lake levels cause large salmon to strand and die rather than reach the breeding grounds. As a result of a semelparous life history, death for salmon at this stage results in a lifetime fitness of zero. We combined information on the level of Lake Aleknagik (southwestern Alaska, U.S.A.) from 1952 – 2006 with a detailed comparison of the body size of mature salmon that died at the mouth of Hansen Creek versus individuals that successfully ascended to the spawning grounds over 10 breeding seasons (1997 – 2006). The proportion of salmon stranding at the mouth varied among years from 2% – 42% in males and <1% – 26% in females. Formal selection analyses indicated that the largest individuals were most susceptible to stranding mortality, especially in years when many salmon stranded, and these were years with low lake levels. Taken together, these results suggest that lake level was a strong and consistent selective force acting on this salmon population, acting synergistically with size-selective predation by bears. Salmon breeding in Hansen Creek tend to be smaller, younger, and more stream-lined than conspecifics from neighboring populations, suggesting that selection against large individuals could be driving these patterns.