Patterns of Hatchery-Produced Returns of American Shad in the James River, Virginia

Abstract American Shad Alosa sapidissima is an anadromous clupeid that once supported a robust fishery but has declined drastically throughout its native range due to overfishing, dam proliferation, and poor water quality. A hatchery program on the James River in Virginia was introduced in 1992 to support the recovery of stocks. Following a moratorium of the fishery enacted in 1994, a fisheries-independent survey was initiated in 1998 to monitor the population recovery efforts and status of American Shad stocks in Virginia. This paper examined 22 years of monitoring data for the James River and determined the effect of hatchery inputs on the James River stock of American Shad. The spawning stock index increased from 2.57 in 1998 to a peak of 9.33 in 2003 but has generally been declining since and has been at very low levels in most recent years. The hatchery prevalence for female American Shad (i.e., the percentage of fish derived from the hatchery) ranged between 3.6% and 60.5%. Years with higher spawning stock index values were significantly correlated to higher percentages of hatchery fish returning to spawn. The stock–recruitment relationship was best explained by the Ricker model, which had the lowest residual standard error and Akaike information criterion value. A threshold level of hatchery-released individuals (approximately 4 million larvae) was necessary to achieve the highest numbers of returning spawners, but stocking above 7 million larvae correlated with declining returns. Long-term monitoring of the James River American Shad spawning population allowed for the critical examination of the contribution of hatchery individuals to the yearly spawning run and the relative success rate of each hatchery year-class. From these data, we consider that the James River spawning stock of American Shad was dependent upon hatchery inputs, with ideal hatchery returns occurring during years of moderate levels of hatchery stocking.