Teaching fish new tricks: Repeated exposure to a velocity barrier improves passage performance

Abstract

Instream structures like culverts and dams can impede upstream fish migration, acting as environmental filters that only allow onward migration of individuals that can successfully pass them. Cognition and learning ability may be an important factor in determining if a fish can successfully traverse such structures. This study investigated the effect of repeated exposure on passage performance of juvenile Galaxias maculatus through an experimental raceway. Over five consecutive days, individual fish were subjected to the same high-speed (0.45-0.5 m s-1) conditions within the raceway, and performance on each day was recorded. The proportion of fish successfully passing the barrier increased significantly from 40% on Day 1-63% on Day 5. Time-to-event analysis further revealed that by Day 5, fish successfully passed the barrier at a significantly faster rate compared to Day 1. However, repeated exposure did not significantly improve approach or entry rates into the raceway. Fish length influenced approach rates, but not entry or passage rates. These findings suggest that cognition and spatial memory play a role in improving passage performance through velocity barriers, but other factors such as attraction flows may also play an important role in successful passage outcomes.

Fig 1. Aerial view experimental setup, focusing on the working section containing the raceway.

(A) Line marking the “approach” to the raceway, (B) line marking the “entry” to the raceway, (C) line marking successful “passage” of the fishway, (D) mesh screens (visible as dark bars) to confine fish within the flume, (E) removeable mesh screen, containing fish in the “starting section” for acclimation before the start of the trial. No change in bed elevation throughout the entirety of the flume. Water depth across the whole flume was 0.2 m throughout all trials.

Fig 2. ADV measurements throughout the raceway taken at 20%, 60%, and 80% depth of the water column.

Grey, dashed lines represent the fish, approach, and entry to the raceway from left to right.

Fig 3. Binomial logistic regression of approach success, entry success, and passage success across five days.

Each fish is represented by a point.

Fig 4. Event rate models for each raceway threshold, representing the proportion of fish crossing each threshold as a function of time.

Zero on the y-axis represents that 0% of the population have crossed the threshold at that time (t), and one represents that 100% of the fish have crossed the threshold at time (t). Data are stratified by exposure day (Day 1-5). When the lines diverge, this indicates a hazard ratio that differs from 1. Where the lines overlap, it suggests a hazard ratio of 1, indicating no significant difference between exposure days. Shaded regions in the corresponding colours represent the 95% confidence intervals.