EP43A-3546:
Salmon Spawning Effects on Streambed Stability

Thursday, 18 December 2014
John M Buffington1, Todd H Buxton2, Elowyn Yager2, Alexander K Fremier3 and Marwan A Hassan4, (1)US Forest Service, Boise, ID, United States, (2)Univ of ID-Idaho Water Ctr, Boise, ID, United States, (3)Washington State University, Pullman, WA, United States, (4)University of British Columbia, Vancouver, BC, Canada
Abstract:
Female salmon build nests (“redds”) in streambeds to protect their eggs from predation and damage by bed scour. During spawning, streambed material is mixed, fine sediment is winnowed downstream, and sediment is moved into a tailspill mound resembling the shape of a dune. Redd surfaces are coarser and better sorted than unspawned beds, which is thought to increase redd stability because larger grains are heavier and harder to move and sorting leads to higher friction angles for grain mobility. However, spawning also loosens sediment and creates topography that accelerates flow, both of which may increase particle mobility. We address factors controlling the relative stability of redds and unspawned beds using simulated salmon redds and water worked (“unspawned”) beds composed of mixed-grain surfaces in a laboratory flume. Results show that simulated spawning lowered packing resistance to particle mobility on redds by an average of 32-39% compared to unspawned beds. Reductions in packing were sufficient to counter the higher inherent stability of relatively coarse, well sorted grains on redds, overall reducing critical shear stress by 8-20% relative to unspawned beds. In addition, boundary shear stress was 13-41% higher on redds due to flow convergence over the tailspill structure. Finally, redd instability relative to unspawned beds was observed in visual measurements of grain mobility, where bed-averaged shear stress was 22% lower at incipient motion and 29% lower at the discharge that mobilized all grain sizes on redds. Results of these complementary observations, along with sediment mass transport rates being nearly five times higher on a redd than an unspawned bed, indicate that redds are unstable compared to unspawned beds. Given these findings, further research is needed to investigate linkages between spawning disturbance and streambed mobility that may affect salmon reproduction in streams, and to assess whether a certain level of bed disturbance from spawning is required to restore ecosystem functions in streams with threatened populations of salmon.