Can reduced salmonid population abundance be detected in
time to limit management impacts?
Kenneth D. Ham and Todd N. Pearsons
Can. J. Fish. Aquat. Sci./J. Can. Sci. Halieut. Aquat. 57(1): 17-24 (2000)
Abstract: We
evaluated eight populations of native salmonids to determine if rapid,
sensitive detection of a reduction in abundance is possible in the Yakima River
basin, Washington, where a large-scale test of hatchery supplementation is
being conducted. Prospective power to detect impacts to abundance was estimated
from 3-16 annual baseline surveys conducted by electrofishing, trapping, or
snorkeling. High interannual variation in abundance estimates (CV=26-94%)
prevented detection of small impacts for most taxa. For three taxa, models of
environmental and biological influences accounted for between 42 and 49% of
temporal variation, increasing our ability to detect impacts of other
influences. Detectable impacts for a t test with alpha=0.1 and beta=0.1
were >18% for all eight taxa and >54% for four of eight taxa. We suggest
that population abundance monitoring may not provide feedback sufficiently
sensitive or rapid enough to implement corrective actions that prevent impacts
from causing harm or exceeding an acceptable level, especially for rare or
highly valued taxa with small acceptable impacts.