Predicting how the juvenile life-stages of anadromous Atlantic salmon Salmo salar influence their migration phenology and marine survival.

Abstract

Atlantic salmon, Salmo salar (hereafter ‘salmon’), is a species of high commercial, recreational, and cultural importance. Native to rivers of both sides on the North Atlantic basin, this species has experienced dramatic population declines since the 1970s. The reasons for this phenomenon are likely to be complex and multifactorial. Thus, striving to understand factors related to salmon survival, as well as other important aspects of their life cycle, such as growth rates and migration phenology, is crucially important. Salmon are an anadromous species that undertake a marine migration from freshwater to oceanic feeding grounds, returning as mature adults to their natal river to reproduce. Undertaking lengthy migrations represents an evolutionary trade-off, where the risks associated with migration (such as novel environments and predators) are outweighed by the benefits (access to greater resources). For salmon, parts of their migration, such as through estuaries, are often considered a survival bottleneck. Capture-mark-recapture (‘CMR’) studies have a long history in ecology, particularly for assessing survival patterns between different time periods. A sample of a population of wild animals is captured, marked by some form of tag, released, and subsequently resampled. I used a long-term CMR dataset (2005 to present) of the salmon population from the River Frome, southern England, where approximately 10 000 juvenile salmon (known as ‘parr’) are captured throughout the river each autumn by electric-fishing. The individuals are each measured for their body length and then fitted with a passive integrated transponder (‘PIT’) tag. Each PIT tag has a unique code that allows for each individual fish to later be re-identified on tag detectors and readers. In the spring, the Frome parr metamorphose into ‘smolts’ and commence their emigration downstream to the sea. Previously PIT-tagged individuals are resampled in a trap, remeasured, scanned for their tag code, and then released. PIT-tagged individuals that successfully complete their migration and return to the river as adults are then redetected a final time by PIT-tag reading antennae in the river, enabling the marine return rate, ‘MRR’, to be estimated. It has been hypothesised that juvenile body size is an important driver of life history events. Body size may be a proxy for overall fitness and can influence, for example, somatic growth rates, migration timing, and survival between different life stages. For salmon, many other factors may also contribute to this, particularly environmental factors, such as temperature. Thus, the aim of this research was to test the hypothesis that juvenile body size achieved in freshwater is an important determinant of individual survival and marine return success. The hypotheses were tested in in four studies using the long-term CMR dataset of the River Frome. Firstly, I assessed how body size and environmental variables affect overwinter growth rates in the river. I found that smaller individuals grew more during the winter than expected given their initial autumn body size, and that individuals with the fastest winter growth rates were those that experienced warmer winters and more variation in their daily water temperatures. Secondly, I assessed the factors affecting variation in smolt migration timing. I detected that larger smolts migrated earlier than smaller smolts, and that while water temperature and discharge affected migration timing, the importance of these effects varied throughout the migration period. I also found that smolts were more likely to migrate in schools later in the migration period, and during the daytime instead of at night. Thirdly, I assessed factors that affect marine return rates, with an emphasis on conditions experienced by smolts during the early part of their marine migration. I found that smolt body size was the most important determinant of the probability of individual survival and their probability as returning as an adult, although water temperature and the presence of piscine predators may also play a role. Finally, I combined the Frome dataset with smolt tagging data from six other European rivers with PIT-tag programmes to assess whether smolt body size is an important determinant of marine return rates across a substantial portion of the European range of salmon. I found that, as on the Frome, smolt body size was an important determinant of marine return rates across Europe, with larger smolts more likely to return as adults than smaller smolts. The results of this thesis should be of importance for conservation efforts attempting to bolster population abundances of Atlantic salmon. In lieu of focusing predominantly on increasing the number of juvenile Atlantic salmon in the river to attempt to increase the number that return as adults, efforts should be made to ensure excellent growth conditions in nursey areas to maximise smolt length, given juvenile body size plays an important role in subsequent survival.