Full Circle: upstream and downstream migration of Atlantic salmon (Salmo salar) in the northern Swedish river Vindelälven

Abstract: The life cycle of the anadromous Atlantic salmon (Salmo salar) can span large geographic, political, and socio-economic boundaries. Management strategies and regulations that only concentrate on small spatial scales often overlook larger basin-wide problems, especially relating to post-spawn seaward migration. In this master thesis one entire migration cycle of wild adult Atlantic salmon in the unregulated northern Swedish river Vindelälven was monitored by radio-telemetry tracking during upstream spawning migration and downstream seaward migration back to the Baltic Sea. The effect of ladder passage variables (time to pass, total time in the ladder, passage day over the ladder) at a fish ladder downstream in the river Umeälven, as well as fish size, were evaluated to determine if differences in upstream migration distance in the river Vindelälven could be observed. Ladder passage variables did not affect migration distance, but size exhibited a negative relationship to migration distance. Additional aspects of spawning migration and spawning sites were also described. Microhabitat variables of depth, substrate and velocity were measured at spawning locations to compare fish size with habitat usage. No correlation with depth and substrate size compared to fish size could be found, but larger fish were found at significantly higher velocity spawning locations with an average velocity of 0.45 m/sec. Salmon kelts were monitored after spawning with evaluations of over-winter and downstream migration. Six tagged salmon migrated downstream soon after spawning and 30 tagged salmon over-wintered in the river. Salmon that migrated further upstream to spawn were significantly more likely to over-winter than leave directly after spawning. 22 corresponding spawning and over-wintering kelts were located, with over-winter locations on average 16.9 km below spawning areas. No correlation was observed between upstream migration distance and the distance travelled between spawning and over-winter sites. No salmon over-wintered above their spawning sites, and over-wintering locations were in deep, slow moving ice-covered sections of the river. Seaward migrating kelts were monitored in the spring with evaluations on the effects of river discharge and temperature on initiation of migration, with additional descriptions of timing, speed and the efficacy of the fish ladder as a bypass around the Stornorrfors power station. Seaward migration appeared to coincide with an increase in both river discharge and temperature with a majority of downstream migration occurring in late May with an average downstream migration speed of 2.5 km/hr. No kelts descended the fish ladder. Genetic analysis discovered that nine percent of tagged salmon that passed the fish ladder were strayers of river Luleälven stock. These fish were monitored to ascertain any differences in size and upstream spawning migration from native river Vindelälven stock. No differences were found between the two stocks.