Close encounters with a fish screen II: delta smelt behavior before and during screen contact.

摘要

Fish screens reduce entrainment losses of fish at water diversions, but little is known about the behavior of individual fish near these structures, including the likelihood, consequences, and mechanics of involuntary screen contact. Recent studies using a large, laboratory-based flume, the Fish Treadmill, have shown that such contact can result in injury, stress, and mortality. We investigated key aspects of screen contact events using videotape records of Fish Treadmill experiments. We quantitatively described precontact behavior, impact velocity and angle, contact duration, distance traveled, and velocity of delta smelt Hypomesus transpacificus while in contact with the screen, and related these responses to postexperiment injuries and mortality. Immediately prior to contact, delta smelt generally exhibited positive rheotaxis and moderate swimming velocity (mean [plus or minus] SD = 20.2 [plus or minus] 9.7 cm/s). Most fish impacted the screen broadside (within 40[degree] of parallel to the screen) and tail first. Impact velocity generally increased as water velocity increased; however, at water velocities greater than 25 cm/s, impact velocity decreased with increases in fish swimming velocity and was less than the water velocity measured near the screen. Mean contact duration was short ( 1 s) and decreased with increasing water velocity. Distance traveled in contact with the screen was not related to flow, but the velocity at which fish were dragged along the screen surface increased with increases in water velocity. Only impact velocity affected injury rates and severity; injuries increased with increases in the velocity of the flow vector normal to the screen (approach velocity). Combined with previous Fish Treadmill results for delta smelt and other species, these findings suggest a mechanism for screen contact- and flow-related injuries and mortality, provide additional support for fish screen criteria designed to limit screen contacts, and demonstrate the value of experimental approaches that integrate hydraulic design with physiological and behavioral responses of fish.