The connectivity of river drainages has been decreased by the installation of roadway culverts, particularly for the salmonids of the Pacific Northwest. Thousands of culverts within the State of Washington have been designated by the state DOT as fish passage barriers. Though it is well known that the anadromous salmon travel upstream to spawn, recent evidence suggests that juvenile salmon also travel upstream to seek preferred habitats for feeding, which may ultimately improve their survival at sea. Retrofitting culverts is an economical solution that has been initially implemented to improve adult salmon passage. Baffles increase water depth for low flow conditions and reduce velocities for higher flowrates. To determine the effect of baffles on upstream passage of juveniles, sloped-baffles were studied at a culvert test bed near Tenino, Washington. Using an Acoustic Doppler Velocimeter (ADV), 3-D velocity fields were collected in a full-sized 12.2 m (40’) long, 1.8 m (6’) diameter corrugated culvert. The culvert slope, baffle spacing, and baffle height were varied to observe flow regime trends that describe conditions suitable for fish passage. This project is unique from other hydraulic studies in that biological testing was conducted in conjunction with the hydrodynamic measurements. Biologists randomly selected 100 juvenile Coho salmon from the on-site rearing facility and allowed the fish to ascend the culvert during a three hour period. The movement of the fish was recorded with video cameras and the passage rate was determined.

Results indicate that there is considerable spatial variability in the flow created by the baffles within the culvert. The flow is asymmetric, consisting of a jet traveling over the low side of the baffle and an area of re-circulating water on the high side of the baffle. The asymmetry decreases as the discharge increases and the mean water height surpasses the baffle height. The diversity of flow structures created by this asymmetry is important because it increases the number of reduced velocity paths that fish may travel. The fish passage success rates are also consistent with the trends of asymmetry: as the culvert discharge increases fish are limited to fewer possible paths, and passage rates decrease. The results suggest that both the structure of the flow and the average speed of the flow affect the passage rate. We present a scaling equation that relates the occurrence of flow structures to the independent study parameters in order to provide guidance in baffle implementation. Recommendations for future work include further biological interpretation and testing, so that the hydraulic and biological results may be more closely coupled.