We tested 436 adult American shad Alosa sapidissima in an experimental louver bypass system, which was similar to a system operating at Holyoke Dam, Massachusetts, to determine guidance and passage efficiency and to study fish response to stimuli from physical structures, light intensity, and water velocity. Groups of 5–29 fish were exposed to combinations of two bypass exits (wide-shallow and vertical-slot sharp-crested weirs) and two louver arrays (7.6- and 15.2-cm slat spacing) oriented 20° to water flow direction. Underwater video observations showed fish responded to louvers as a physical barrier during the day, when they stayed 30–55 cm (1.3 cm/5 klx) away from and oriented parallel to louvers, and as a behavioral barrier at night, when they moved closer to louvers and oriented into the current. Both louver arrays guided fish effectively, (i.e., prevented fish from passing through the slats) 100% for narrow spacing and 97% for wide spacing. Adults avoided moving closer than 0.5 m to either exit type; instead, fish remained 0.8–1.4 bodylengths upstream, depending on light intensity (farther upstream during daytime, similar to behavior at louvers). At exits, water velocity increased from 0.4 m/s to 0.8 m/s or more in a distance of 0.9 m (rate of velocity increase, 0.44 m/s per meter). This rapid velocity increase elicited an avoidance response by fish that resulted in few fish (5%) passing. Our results provide behavioral explanations for the efficient guidance of adult American shad by louvers and for the fishes' avoidance of the exit at the Holyoke Dam. From this, we provide suggestions on how to prevent fish avoidance of exits.
