An. Palmisano et Ne. Elder, Standardized seawater rearing of chinook salmon smolts to evaluate hatchery practices showed low statistical power, T AM FISH S, 130(3), 2001, pp. 409-416
We examined, under standardized conditions, seawater survival of chinook sa
lmon Oncorhynchus tshawytscha at the smolt stage to evaluate the experiment
al hatchery practices applied to their rearing. The experimental rearing pr
actices included rearing fish at different densities; attempting to control
bacterial kidney disease with broodstock segregation, erythromycin injecti
on, and an experimental diet; rearing fish on different water sources; and
freeze branding the fish. After application of experimental rearing practic
es in hatcheries, smolts were transported to a rearing facility for about 2
-3 months of seawater rearing. Of 16 experiments, 4 yielded statistically s
ignificant differences in seawater survival. In general we found that high
variability among replicates, plus the low numbers of replicates available,
resulted in low statistical power. We recommend including four or five rep
licates and using alpha = 0.10 in 1-tailed tests of hatchery experiments to
try to increase the statistical power to 0.80.