Spawning tactics of summer chum salmon Oncorhynchus keta in relation to channel complexity and hyporheic exchange

• Published in: Environmental biology of fishes 2014-10, Vol.97 (10), p.1095-1107
• Main Authors: Mouw, Jason E. B, Tappenbeck, Tyler H, Stanford, Jack A
• Format: Article
• Language: English
• Subjects: Agnatha. Pisces; Animal and plant ecology; Animal reproduction; Animal Systematics/Taxonomy/Biogeography; Animal, plant and microbial ecology; Animals; Autoecology; Bed load; Biological and medical sciences; Biomedical and Life Sciences; Environment; Flood channels; Floodplains; fluid mechanics; Freshwater & Marine Ecology; Fundamental and applied biological sciences. Psychology; Gravel; groundwater; Groundwater flow; habitats; Hydraulic gradient; Hydraulics; Life Sciences; Nature Conservation; Oncorhynchus keta; river water; Rivers; Salmon; sediments; Site selection; Spawning; Spring; Summer; Surface water; Topography; Upwelling; Vertebrata; Vertebrates: general zoology, morphology, phylogeny, systematics, cytogenetics, geographical distribution; Water exchange; Zoology
• ISSN: 0378-1909; 1573-5133
• DOI: 10.1007/s10641-013-0200-0

Vertical hydraulic gradient, channel hydraulic, and substratum data were collected at active chum salmon redds to evaluate their influence on redd site selection in the Kwethluk River, a wandering gravel-bedded tributary of Alaska’s Lower Kuskokwim River. Two distinct summer chum salmon spawning tacitcs: primary- and off-channel groups spawning in different hydraulic and thermal environments. Channel hydraulics, ground-surface water exchanges, and bed sediment data were collected throughout the range of topographic conditions at sites occupied and unoccupied by spawners. Analyses of these data showed that redd site selection was incoherent with channel hydraulics and substratum size, when all channel types were considered. Rather, spawning was associated with the exchange of river and groundwater at four spatial scales. In primary channels, chum salmon spawned only in reaches where the regional hydraulic gradient (regional scale) promoted the penetration of river water, or downwelling, into the bed sediments. Within these reaches, primary-channel spawning was influenced by bed topography (fluvial scale) that promoted strong localized downwelling; a logit model, fitted to constituent hydraulics, discriminated between selected and unselected sites with 89 % accuracy. In contrast, spawners in off-channel habitats (flood and spring channels) selected areas of upwelling groundwater, regardless of channel hydraulics and the regional hydraulic gradient. In parafluvial flood channels, spawners selected localized upwelling associated with short groundwater flow paths (parafluvial scale). In spring channels, within the floodplain’s forested riparia (orthofluvial scale), spawners eschewed areas of localized downwelling and uniformly selected upwelling zones feeding these channels. Thus, redd selectivity differentiated two spawning tactics (groups) in this complex channel network where ground and surface water exchange is an important predictor of spawning habitat.