1 Transactions of the American Fisheries Society 134:1–18, 2005 [Article]  Copyright by the American Fisheries Society 2005 Cohort-Specific Growth and Mortality of Juvenile American Shad in the Pamunkey River, Virginia JOEL C. HOFFMAN* AND JOHN E. OLNEY Virginia Institute of Marine Science, Post Office Box 1346, Route 1208 Greate Road, Gloucester Point, Virginia 23062, USA Abstract.—We estimated the variation in instantaneous rates of growth (G) and mortality (M) between intraannual cohorts of juvenile American shad Alosa sapidissima in the Pamunkey River, Virginia. The ages of juveniles captured by push net during the juvenile abundance index surveys in 1998 and 1999 were estimated by counting daily rings on the sagittal otoliths. Weight-at-age and abundance-at-age data were used to generate instantaneous daily rates of growth and mortality for 5-d cohorts. In 1998, the peak hatch date lagged behind the peak spawning periods that had been inferred from collections of American shad broodstock and occurred after peak spring flow events. In 1999, the correspondence between the hatch date distribution and the peak spawning periods was greater than in 1998. The instantaneous daily growth rate was relatively constant between cohorts, ranging from 0.037 to 0.064 in 1998 and from 0.046 to 0.066 in 1999. The instantaneous daily mortality rate was more variable between cohorts, ranging from 0.047 to 0.084 in 1998 and from 0.044 to 0.093 in 1999. The physiological mortality rate, or M/G, was calculated for all cohorts. Most cohorts in 1998 and 1999, including the largest cohorts in both years, had an M/G value close to 1.0, indicating that these cohorts were barely maintaining or losing biomass during the early juvenile stage. The results of this study indicate that in both 1998 and 1999 the year-class declined in biomass during the early juvenile stage and underwent demographic struc- turing that affected the composition of the population that ultimately migrated to sea. Variability in the growth and mortality rates of young of year fish is usually high, and small dif- ferences among years will determine the strength of a year-class (Houde 1987, 1989). Within a year- class, variability in growth (G) and mortality (M) between intraannual cohorts will determine the rel- ative strength of each cohort (Rutherford and Houde 1995). The progeny of temperate zone fishes that have protracted spawning periods may experience a wide range of temperatures and hydrographic con- ditions, leading to potentially large differences in recruitment between intraannual cohorts. Larval- stage biomass increases when M/G, the ‘‘physio- logical mortality rate’’ (Beyer 1989), is less than 1.0 and decreases when M/G is greater than 1.0. The size or time of transition, when M/G = 1.0, is an important stage-specific parameter for determin- ing year-class strength (Houde 1997). Cohorts that transition earlier will contribute more biomass to the year-class, assuming that the trend through suc- cessive stages of early life history is towards a lower M/G. Variability in recruitment between intraannual cohorts has been observed in American shad Alosa sapidissima, an anadromous alosine clupeid native to the Atlantic coast of North America. This var- * Corresponding author: jhoffman@vims.edu Received December 15, 2003; accepted June 14, 2004 iability has been attributed to dynamics during both the larval and juvenile stages. In the Con- necticut River, variability in mortality during the larval stage establishes the year-class strength of American shad (Crecco et al. 1983; Savoy and Crecco 1988). Crecco and Savoy (1984) found that the juvenile abundance index (JAI) of American shad in the Connecticut River correlated signifi- cantly with June river temperature and flow, evi- dence that survival depends on conditions that af- fect both the larval and transitional stages. These conditions are likely to favor survival of certain cohorts within the year-class. In the Hudson River, the 1990 year-class of juvenile American shad was composed mostly of fish hatched after June 1, even though spawning activity peaked in early to mid- May (Limburg 1996). Limburg concluded that that the low survivorship of larvae hatched early in the season was due to high river flow and low water temperature. In a retrospective analysis of the oto- liths of returning adults, Limburg (2001) found that survivorship was greatest among early and late juvenile cohorts. This finding suggested that there had been two stages of low survivorship— the first affecting larvae that were hatched early, and the second affecting juveniles that were hatched late. For age-0 American shad, instantaneous daily mortality during prejuvenile life stages decreases