Assessing the impact of river barriers on successful seaward migration of Atlantic salmon, Salmo salar L. along the River Derwent, Cumbria

Green, Amy (2024) Assessing the impact of river barriers on successful seaward migration of Atlantic salmon, Salmo salar L. along the River Derwent, Cumbria. PhD thesis, University of Glasgow.

Abstract

The Atlantic salmon, Salmo salar, Linnaeus 1785, is well studied due to its economical, ecological, and cultural importance across Europe, Scandinavia, and North America. Nonetheless significant gaps in our understanding persists when it comes to their life history, particularly during the smolt phase which has been considered as one of the most vulnerable phases in the Atlantic salmon life cycle. In recent decades, management and research incentives have concentrated primarily on enhancing and understanding migration behaviour of Atlantic salmon within freshwater environments. Despite management efforts, Atlantic salmon populations continue to decline with some European populations becoming extinct. Previous research has suggested that the decline is at least in part, related to increased habitat fragmentation resulting from artificial river-spanning infrastructure within the riverine environment which partially or completely impedes migration. This thesis describes a series of studies using acoustic telemetry that have provided evidence to fill the knowledge gaps in literature regarding downstream migratory behaviour of Atlantic salmon smolts from a UK tributary (River Derwent, Cumbria) draining into the Solway Firth.

Although Atlantic salmon migration through riverine environments have been extensively researched, there are gaps in literature regarding the inter-annual spatial variation in migration success through the river system. Previous studies have mostly focussed on overall river migration success rates. In this study I focus with greater precision on potential bottleneck zones within a river system which contribute to reduced migration success rates of Atlantic salmon. By combining acoustic telemetry and statistical modelling, smolts were tracked through the entire River Derwent. A number of potential factors that might influence migration success were assessed to understand specific geographic areas in which success is reduced and identify the factors which change the likelihood of smolts completing a successful riverine migration. Upstream areas of the River Derwent had a greater impact on migrating smolts, although variation in migration success across the three years of the study were observed. Consistent with previous literature, smolts experiences high overall loss rates during river migration (2020: 8% (n=8), 2021: 27.3% (n=41), 2022: 41.7% (n=48)). Migration speed (Rate of Movement (m.s-1 ) and ground speed (km.day-1 ) varied across river section and year but was consistently slow. There was also variation between years. Speed of migration in 2021 was found to be significantly higher when predicting rate of movement compared to 2020 and 2022, which is consistent with high water levels observed during tagging and release in 2021. Further investigations into the spatial variation of riverine systems across longer time scales are required to fully identify and understand how individual river reach characteristics may impact on migration success rates and speed; to determine if these are site specific or if these reach effects can be applied more widely.

The natural standing water present in the River Derwent (Bassenthwaite Lake) was identified as a bottleneck zone for migration and was found to decrease the likelihood of migration success. The understanding of the effect of natural standing waters remains limited and speculation still remains as to the cues associated with successful smolt navigation though these systems. Previous studies have highlighted natural standing waters associated with reduced migration rates, where some studies suggest non-directional pathways are one of the direct causes of migration mortality. Using acoustic telemetry and statistical modelling, I investigated potential factors which may increase successful migration likelihood through the most northernly un-impounded lake in the Derwent catchment. The evidence from this study is consistent with previous research and showed that migrating Derwent smolts experience high loss rates within Bassenthwaite Lake (33%), slow migration (0.16 m.s-1 ) and nondirectional movements. There was no evidence of phenotypic, behavioural or environmental effects that distinguished successful from unsuccessful lake migrant smolts. This suggests that migration success in Bassenthwaite Lake was random. I determined that the concept of a “Goldilocks Zone” is applicable to Bassenthwaite Lake and estimated the average minimum distance of this zone to be 0.72 ± 0.6 km (mean ± SD; range: 0.13 – 1.24) from the lake exit, though it remains unclear if the “Goldilocks Zone” is more generally applicable to all standing waters or if its specific to only natural lake systems.

The effect of small river-spanning infrastructure (low-head weirs, >5m height) on downstream migrating smolts remains poorly understood. In Chapter 4, the success rates, behaviours associated with success and the ability to choose alternative passage around weir structures was investigated using a combination of acoustic telemetry and statistical modelling. Smolt migration failure rates differed across both weirs, with Coops Weir having a failure rate of 1.49-2.24%.km-1 and between 0–32.2%.km-1 for Yearl Weir. Both of these figures are considerable higher when compared to other studies and overall riverine failure rates elsewhere within the River Derwent, although there was evidence to suggest that smolts may choose to migrate by an alternative route around and not over the low head weirs studied in this work. Although it remains unclear as to whether migration failure rate across weirs was due to lack of preparedness to high salinity environments, injury or disorientation/stress as a result of passage or predation.

Once smolts successful migrate through the riverine environment across both natural and anthropogenic barriers, transitioning from freshwater to saltwater environments smolts must navigate through the early marine environment to reach key feeding grounds. However, despite the rapidly developing understanding of migration in freshwater systems, the knowledge gaps around marine environment migration trajectories remain. By combining data from five acoustic telemetry studies we are able to document the migration of post-smolts from the west coast of northern England. Migration success differed significantly across the three years of the study, although migration success rate (%.km-1 ) was found to be higher in freshwater systems compared with the early marine environment. It was found that in 2021 smolts, used a northward migration pathway through the Irish Sea, although this study provided evidence to suggest that the Derwent populations may be impacted by aquaculture sites around the coast of Arran through which smolts were found to migrate around.

The five data chapters presented in this thesis employed acoustic telemetry to model and assess Atlantic salmon smolt migration behaviour across a three-year study period through the riverine and early marine environment. The results obtained from this thesis will be developed by future research and used by government bodies to aid in identifying potential stressors on migration within the riverine environment which could impact Atlantic salmon populations.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: S Agriculture > SF Animal culture > SF600 Veterinary Medicine; S Agriculture > SH Aquaculture. Fisheries. Angling
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Biodiversity, One Health & Veterinary Medicine
• Supervisor's Name: Adams, Professor Colin and Honkanen, Dr. Hannele
• Date of Award: 2024
• Depositing User: Theses Team
• Unique ID: glathesis:2024-84389
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 25 Jun 2024 08:30
• Last Modified: 25 Jun 2024 08:37
• Thesis DOI: 10.5525/gla.thesis.84389
• URI: https://theses.gla.ac.uk/id/eprint/84389