Application of the Genetic Mark–Recapture Technique for Run Size Estimation of Yukon River Chinook Salmon

We present an application of the genetic mark-recapture technique to estimate salmon run size in a large river. Application of this technique requires modifications to estimation methodology. Under a typical Lincoln-Petersen mark-recapture estimation of salmon run size (N = M/p), individual fish are captured and marked (M) in the lower river and are recaptured (m) at escapement (E: the number of fish reached spawning ground) monitoring sites selected upriver where the proportion of marked individuals (p = m/E) is estimated. In this genetic mark-recapture technique, the marked individuals are not captured and recaptured, but rather the naturally distinctive genetic (marked) population is captured and recaptured. Genetically, the lower river population is a mixture of multiple genetic stocks, whereas the upriver escapement population consists of a single genetic stock. Hence, the mark-recapture experiment (N = M/p(m)) is reversed. The proportion of "marked" genetic stock (p(m)) is estimated in the lower river, and size of the "marked" stock in the lower river (M) is estimated by summing its upriver escapement (E-m) and harvest (C-m) between the lower and upper portions of river (M = E-m+C-m). The harvest is calculated as a product of total upriver harvest (C) and the proportion of the "marked" stock (p(cm)) in the harvest (C-m = C center dot p(cm)). Further, when the proportion of multiple genetic stocks (p(k)) is identified, stock-specific run size (N-k = N center dot p(k)), escapement (E-k = N-k-C-k, where C-k = C center dot p(ck)), and exploitation rate (Ex(k) = C-k /N-k) can also be estimated, which provides substantially more information than does the conventional approach. We illustrate an application of this technique for estimating run size of Chinook Salmon Oncorhynchus tshawytscha in the Yukon River, Alaska. Received June 6, 2013; accepted November 20, 2013