How fish negotiate culverts

Posted on March 27, 2017

While culverts are designed to facilitate the movement of water by allowing it to flow under roads or other obstructions, they can place physiological and/or behavioural restrictions on the movements of fish. Although several studies have considered the implications of culvert design for fish passage, surprisingly few have tried to identify the factors that influence successful passage once a fish enters a culvert. Researchers addressed this deficiency by implanting over 1000 wild brook trout (Salvelinus fontinalis) with transponder (PIT) tags and monitoring their success in negotiating 13 culverts in three watersheds of southern Quebec, Canada under a range of hydraulic and environmental conditions. The fish had fork lengths of 85–240 mm.  Most were released in a cage in the pool at the downstream end of the culvert and were allowed 24-48 hours to stage their own passage attempts, but some were released free in the downstream pool itself and given a longer time (42-72 hours). Culverts were made of metal, concrete or polyethylene and, except for one concrete box culvert, they were circular in cross section. Metal culverts were “rough”, with annular or spiral corrugations, the other culverts being smooth. Of 208 regression models tested, one emerged as clearly superior to the rest in predicting passage success: in this model the interaction between culvert type and fish length was particularly important, with smaller fish performing better in rough culverts, probably because the corrugations created areas of reduced flow where fish could rest periodically. Other factors influencing passage success were culvert slope (negative), depth of downstream pool (negative), water temperature (optimal around 14-15 oC), water velocity (negative), and trial duration (positive). The depth of the pool probably influenced a fish’s perception of the oncoming flow and available refuge, with deeper pools allowing fish access to slow-velocity zones well away from the surface. Caged fish staged more passage attempts. Interestingly, the length of the culvert was not an important factor, and this suggested that swimming capacity and fatigue were less influential than interactions between physiological and behavioural factors. The increasing tendency to replace metal culverts with smooth ones in order to increase culvert lifespan is likely to reduce connectivity between trout populations.

Reference: Goerig, E. et al. 2016. Brook trout passage performance through culverts. Canadian Journal of Fisheries and Aquatic Sciences 73, 94–104.