Thesis
On the edge of the abyss : modelling the early marine migration of Atlantic salmon post-smolts
- Creator
- Rights statement
- Awarding institution
- University of Strathclyde
- Date of award
- 2026
- Thesis identifier
- T18116
- Person Identifier (Local)
- 202185768
- Qualification Level
- Qualification Name
- Department, School or Faculty
- Abstract
- Atlantic salmon (Salmo salar ) populations have suffered declines across their range in recent decades, largely attributed to decreasing marine survival rates. The first few months at sea are thought to be a time of particular vulnerability, but the drivers of mortality during this period are not yet fully understood. Characterising the conditions experienced by salmon at sea is challenged by a lack of knowledge of full migration paths, with empirical approaches limited to specific locations. Additionally, the mechanisms shaping migration paths and the extent of variability in migrations between years are unclear. To address knowledge gaps, the movement of juvenile salmon over the first three months at sea is simulated for a group of Irish, Northern Irish and Scottish rivers using a Lagrangian particle tracking model coupled with an active swimming model that is dependent on ocean currents, salinity, and compass direction. Differences in the speed of migrations, the proportion of time spent in different oceanographic regions, and the proportion of migrations reaching the Norwegian Sea are seen between years. These differences are related to changes in local ocean conditions: years with lower on-shelf salinity, stronger onshelf northwest currents, and stronger shelf-edge currents were associated with greater migration success. Motivated by the observed interannual variability, the value of behavioural variability in dampening interannual variance in modelled migration success is investigated. The migration model is expanded and parameterised for 38 rivers around the British Isles, with different parameter sets representing different navigation behaviours. Portfolios of behaviours were selected for each river by maximising a version of the Sharpe index, which balances portfolio return (mean migration success rate) and risk (interannual variability in migration success). For simulated salmon from some rivers, a single swimming behaviour is found to be optimal, while other river populations benefit from up to 24 behaviours. The use of an optimal portfolio of swimming behaviours for each river minimises stock-level variation (when considering all 38 rivers together). These findings highlight the importance of behavioural diversity both within and between salmon populations, which may be supported by the conservation of genetic diversity. To aid future research into the bycatch risk of juvenile salmon, a potential driver of mortality at sea, the exposure of juvenile Atlantic salmon to pelagic fisheries during their first few months at sea is determined. Landings (taken as a proxy for fishing intensity) of six commercially important pelagic finfish species (mackerel, herring, horse mackerel, capelin, sardine, chub mackerel) were quantified from 2000 to 2022 within salmon migration regions. Pelagic fisheries for herring and mackerel present the highest total exposure along the migration corridor between April and August. Additionally, there is potential for Atlantic salmon bycatch in the horse mackerel and sardine fisheries in April. Future investigations using Environmental DNA (eDNA) analysis and/or landings port observations of the fisheries identified in this study could quantify bycatch rates, which may be combined with the exposure values developed here to quantify bycatch risk to Atlantic salmon stocks. Two applications of the juvenile salmon migration model are presented in the final chapters. First, zooplankton energy derived from the Continuous Plankton Recorder (CPR) is used as a measure of energy in the juvenile salmon food chain (prey-of-prey energy), and occupancy weighted zooplankton energy estimates are generated using gridded densities of modelled juvenile salmon. CPR sampling is found to be highly variable within specific migration space-time regions, so time-mean energy over multiple years is considered within the spatial migration regions. Slower migrations allow access to more energy resources, but there may be a trade off between moving slowly and feeding, versus making it to the Norwegian Sea by a suitable time. Finally, a metric of similarity between modelled migrations from different rivers is developed and used to estimate marine survival rates for rivers where they are unknown, assuming that similarity in modelled migration densities drives similarity in experienced prey and predation conditions, and hence survival patterns. Estimated marine survival rates accurately depict known interannual trends, though some uncertainty is generated by differences in the absolute values of marine survival rates between rivers. The estimated marine survival rates are used in a decision support tool which visualises the life-cycle wide effects of changing river and ocean conditions, helping to inform appropriate management action.
- Advisor / supervisor
- Bull, Colin (Marine biologist)
- Banas, Neil S.
- Resource Type
- DOI
- Funder
Relations
Items
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
PDF of thesis T18116 | 2026-08-13 | Public | Download |