Understanding uncertainty in the effect of lowhead dams on fishes of Great Lakes tributaries

Publication Date



dams, Great Lakes, small dams, structures, barriers, sea lamprey, lamprey, upstream, species richness, streams, sampling design, design, regression, models, sound

Journal or Book Title

Ecological Applications


Small dams represent one of the most widespread human influences on riverscapes.Greater understanding of how these structures affect aquatic organisms is needed to ensure thatdecisions regarding their construction and removal strike an appropriate balance betweencomponents of human and ecosystem services. Within the basin of the Laurentian Great Lakes,the effects that in-stream barriers (dams) used to control the non-native, parasitic sea lamprey(Petromyzon marinus) on the diversity of non-target fishes is a significant concern for fisherymanagers. A previous study indicated that upstream changes in the species richness of nontargetfishes observed in 24 streams with a sea lamprey barrier relative to paired referencestreams (a measure of effect size) was variable across the basin. We examined the degree towhich the variance in effect size could be attributed to imprecision in the field sampling protocolused to estimate effect sizes, differences in catchment-scale landscape attributes between barrierand reference streams within pairs, and differences in landscape attributes at different spatialscales among barrier streams. Simulation modeling and analyses of repeated fieldmeasurements made for a subset of streams demonstrated that a large variance in effect size isexpected for the field sampling design and that estimates of effect size measured for individualbarrier streams are imprecise. Regression models and multimodel inference methods based onAkaike's Information Criterion provided less support for hypotheses linking effect size tolandscape attributes. Mean effect size, adjusted for the influences of landscape characteristicswithin and across stream pairs, provides the most reliable and least biased estimate of the effectof sea lamprey barriers on the richness of nontarget fish species. With the information currentlyavailable, landscape characteristics of catchments cannot be used to help decision makersanticipate effects sizes for candidate streams being considered for future barrier construction. Ourfindings will help fishery managers in the Laurentian Great Lakes make more informed decisionsregarding the use and placement of sea lamprey barriers and achieve their objective of deliveringan integrated pest management plan for sea lamprey control that is environmentally andeconomically sound and socially acceptable. http://dx.doi.org/10.1890/06-1417.1







This document is currently not available here.