UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Migration performance of Atlantic salmon post-smolts in a Norwegian fjord system Plantalech Manel-la, Núria 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-0559.pdf [ 3.52MB ]
Metadata
JSON: 831-1.0101117.json
JSON-LD: 831-1.0101117-ld.json
RDF/XML (Pretty): 831-1.0101117-rdf.xml
RDF/JSON: 831-1.0101117-rdf.json
Turtle: 831-1.0101117-turtle.txt
N-Triples: 831-1.0101117-rdf-ntriples.txt
Original Record: 831-1.0101117-source.json
Full Text
831-1.0101117-fulltext.txt
Citation
831-1.0101117.ris

Full Text

MIGRATION PERFORMANCE OF ATLANTIC SALMON POST-SMOLTS IN A NORWEGIAN FJORD SYSTEM by NURIA PLANTALECH MANEL-LA B.Sc. (Ocean Science) University of Las Palmas de Gran Canaria, Spain, 2003 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Animal Science) THE UNIVERSITY OF BRITISH COLUMBIA May 2007 © Niiria Plantalech Manel-la, 2007 ABSTRACT The migratory behaviour of hatchery-reared Atlantic salmon (Salmo salar) post-smolts during their seaward migration was examined using acoustic telemetry. The first study focused on the swimming depths of eight post-smolts relative to the water-column salinity and temperature using depth sensitive acoustic transmitters. Individuals were released at the mouth of the Eio River and manually tracked during the first hours of their sea migration in a Norwegian fjord system. Vertical salinity and temperature distributions were recorded along the tracked post-smolts' trajectory. Post-smolts were tracked for a mean of 11.4 ± 1.9 hours after release over a mean distance of 8.7 ±3.4 km from release into the fjord. Mean swimming depth was 1.6 m with a range of 0.5 - 2.1 m. The results of this study suggest no overall preference for salinity concentration with post-smolts spending on average 68% of the time in salinity concentrations lower than 20 practical salinity units (psu) during the sea migration. The second study compared the behaviour of two distinct populations with different inherent migratory distances to the ocean in a common-garden experiment. Atlantic salmon post-smolts of Rivers Flekke (20 km from the open ocean, n = 80) and Lasrdal (144 km from the open ocean, n = 79) origin were tagged -with acoustic transmitters and released in spring of 2005 and 2006 at the mouth of the river Opo, in S0rfjorden, Western Norway. The post-smolts' migratory behaviour was continuously monitored using acoustic listening stations at three areas in Sorfjorden (2.5 -35 km from the release point). The high percentage of post-smolts detected 35 km from release site (84% in 2005 and 81% in 2006) suggested a high survival and was independent of the fish origin. The post-smolts' progression rate from release to the end of area 3 (means, 2005: 0.6 bl s"1 and 2006: 0.7 bl s"1) was not dependent on fish origin in either of the two years. This study is the first to describe the swimming behaviour, including swimming depth, of fjord-migrating post-smolts. TABLE OF CONTENTS Abstract ii Table of contents iv List of tables v List of figures vi List of abbreviations vii Acknowledgements viii Dedication x Co-Authorship Statement xi CHAPTER 1 - General introduction 1 References 8 CHAPTER 2 - Migratory behaviour of Atlantic salmon (Salmo salar) post-smolts relative to the variation in salinity and water temperature 12 Introduction 12 Material and methods 13 Results 16 Discussion 18 References 22 CHAPTER 3 - A comparison of migratory performance between two stocks of Atlantic salmon post-smolts 38 Introduction 38 Material and methods 39 Results 41 Discussion 43 References 47 CHAPTER 4 General discussion 55 References 60 iv LIST OF TABLES Table 2-1: Descriptive data from the Atlantic salmon post-smolts tracked using acoustic transmitters in Eidfjord, Hardangerfjord system, Norway (n=8) 26 Table 2-2: Distance migrated, speed and vertical movements of Atlantic salmon post-smolts (n=8) tracked during May 2006 in Eidfjord, Hardangerfjord system, Norway. The percentage of time spent at salinities <20 psu are based on isohaline plots 27 Table 3-1: General characteristics of the post-smolts released and number offish detected in area 1 (2.5 km from release), area 2 (9.5 km from release) and area 3 (35 km from release). Values for total length (LT) and weight are: Mean ± s.d. (range) 50 Table 3-2: Progression rates (bl s"1) and time (h) spent in the release point for the groups of released post-smolts (Flekke and La^ rdal) in Hardangerfjord (2005 - 2006). Values are: Mean ± s.d. (range) 51 v LIST OF FIGURES Figure 2-1: Study area. Atlantic salmon post-smolts (n=8) were released and manually tracked using acoustic depth sensitive transmitters in Eidfjord between 10 May and 30 May, 2006 28 Figure 2-2: Salinity (A) and temperature (B) profiles at the reference station 2TS during the study period. Legend: Circles = 27 April; Squares = 19 May; Triangles = 31 May.... 29 Figure 2-3: The post-smolts' (n=8) swimming depth during transect is plotted over the contour maps of salinity and temperature. The dots indicate the time and the depth where the salinity and temperature profiles were taken. The isohalines and the isotherms were drawn with an interval of 2 psu and 1 or 0.5 °C respectively. The thick continuous line represents the depth of the post-smolt. Time scale shows local time. 30 Figure 3-1: Study area. Tagged post-smolts from the Rivers Flekke (n=80) and Laerdal (n=79) were released in spring 2005 - 2006 at the Opo River mouth, Odda, Hardangerfjord system, Norway 52 Figure 3-2: Salinity (A) and temperature (B) profiles in the release area during the study period. Legend: Squares = 19 April 2005; Triangles = 27 April 2006 54 vi LIST O F ABBREVIATIONS Bl.s"1 Body Lengths per Second Km Kilometres L T Total Length M Metres Mm Milimoles Psu Practical Salinity Units S.d. Standard Deviation vii ACKNOWLEDGEMENTS I would like to thank Scott McKinley for opening the door and giving me the opportunity and support to explore my interests through the completion of my M.Sc. thesis and also for providing me financial support. I am grateful also to Drs. Colin Brauner, Brian Klinkenberg and Marina von Keyserlingk for their support and guidance during my M.Sc. Funding was provided by MSERC and Fiskeri og Havbruksnasringens Forskningsfond (FHF). I would also like to thank the staff at the Statkraft Energy AS hatchery in Simadalen for providing the fish for our research, extensive help during the project and many social evenings. I am also very grateful to Kjell Arne Mo and Sigurd Larsen for offering the use of their balcony and fish tanks where I stored my fish used in chapter 2. Thanks also to Kari Sivertsen for drawing the study area figure used in chapter 2 and to Dr Bengt Finstad for analyzing the post-smolts' blood samples. I would not have been able to pursue my studies and research without Finn Okland, who introduced me to the world of telemetry. I am also deeply grateful to the rest of my Norwegian colleagues (Eva Thorstad, Bengt Finstad, Rolf Sivertsgard, and Jan Davidsen), thanks for your help during the field project in Hardangerfjord and for the emails and phone calls back and forth during the field data analysis and the manuscript writing process. I had a great time in Hardangerfjord, both working and having fun (having drinks in the local pub, eating raspaboller, watching movies in the Eidfjord cinema, visiting the surroundings and spending evenings together listening to some accordion music). Thanks to Fiona K. Cubitt, Kevin Butterworth, Susan Dean, Cedar Chittenden and Audun Rikardsen for their assistance in Hardangerfjord, it was great to have all of you over there. viii I wish to thank my friends in Vancouver for helping me combine the studies with a fun life as well. For me this is essential, my life cannot be only studies, it needs much more than that and you have provided it for me. Special thanks are owed to my parents, Jaume Plantalech Maya and Rosa Manel-la Llinas and to my sister Agnes Plantalech Manel-la for giving me support even if I am far away from home. Also Thanks to Jose Maria Sousa do Vale Marcal for his support and encouragement during the time I spent in Norway and in Canada. Finally, magic thanks are owed to the post-smolts used in chapters 2 and 3 for letting me tag them and follow them for many km from the river mouth towards the open ocean. Thanks everyone! Without all of you the completion of my thesis would have not been possible or it would have been much harder than it has been. Nuria ix To my parents and my sister CO-AUTHORSHIP S T A T E M E N T The studies were designed collaboratively by Nuria Plantalech Manel-la, Prof. R. Scott McKinley, Dr. Bengt Finstad, Finn Okland, Eva B. Thorstad, Jan G. Davidsen, Rolf Sivertsgard and Cedar M. Chittenden. Nuria Plantalech Manel-la conducted the research, analyzed all data and prepared the manuscripts under the guidance of Prof. R. Scott McKinley, Dr. Bengt Finstad, Finn Okland, Eva B. Thorstad, Jan G. Davidsen, Rolf Sivertsgard and Cedar M. Chittenden. xi CHAPTER 1 - General introduction The Atlantic salmon has been historically known as the "King of Fish" because of the outstanding characteristics of its life cycle that involve, for instance, undertaking long migrations and traveling through many obstacles when moving from freshwater to the marine environment and vice versa. Atlantic salmon can be found in the coastal countries of the North Atlantic Ocean, from the White Sea to Portugal on the European coast and from Ungava Bay to Connecticut in North America. This fish that has long fascinated biologists is, according to the US Endangered Species Act (ESA), now considered to be in danger of extinction. After two centuries of a slow decline that coincided with human industrial development, Atlantic salmon has nowadays reached its lowest historical numbers. Wild Atlantic salmon populations are only considered in healthy condition in four countries in the world and they have already disappeared from Germany, Switzerland, the Nederland, Belgium, the Czech Republic and Slovakia (WWF 2001). Their anadromous condition makes them sensitive to both environmental pressures in rivers and fjords and overfishing in the sea. Some of the major impacts on wild Atlantic salmon include: hydropower dams, variations in the river flow, pollution and, salmon aquaculture. All these factors affect the salmon in different ways. Hydropower dams and other river obstructions form obstacles to the upstream and downstream migration of salmon, reducing the individual's capacity to survive. Variations in the river flow, produced for example by deepening the channels or disconnecting the direct river flow from the tributaries, may result in habitat degradation or direct habitat loss. Pollutants from 1 agricultural runoff or industrial waste may result in inputs of excessive nutrients or an increase of the heavy metal concentration in the rivers which may impact the behaviour and survivability of the salmon stocks. Salmon aquaculture can act as a source of diseases and parasites for the migrating wild populations and escaped farmed salmon may threaten the wild stocks by competing for the spawning grounds and reducing the genetic variability on the wild populations. All these threats mentioned above impact different stages of the salmon's life cycle. Alevins and juveniles may be affected, for instance, by habitat loss or pollution and, smolts and adults by impediments during their in-river migration. In this thesis, I focus on the post-smolt stage. For the purposes of this work, the post-smolt stage starts when the salmon leave the river to navigate towards the open ocean. Telemetry techniques have been suggested for examining the migration behaviour of post-smolts (Moore et al. 2000). Migration from the gravel beds to the fjord Atlantic salmon embryos remain in the gravel bed of fresh water rivers until they have absorbed most of the yolk and then move to a nursery habitat that may be a few metres or kilometres away from the spawning grounds. Juvenile salmon migrate downriver, through estuaries and fjords and eventually arrive at the oceanic feeding areas (McCormick et al. 1998). Juvenile salmon, prior to leaving the rivers, need to transform physiologically and morphologically to prepare for seaward migration and life in the sea. They accomplish that by undergoing through a process called smoltification. Smoltification involves behavioural, physiological and morphological changes that prepare the fish to enter the hypersaline environment. Timing of smoltification and sea 2 migration are controlled by environmental and physiological variables. Water temperature, water discharge and photoperiod have been identified as the main environmental factors controlling the onset of seaward migration and the onset of smoltification (McCormick et al. 1998; Hvidsen et al. 1995). Several researchers have reported that water temperatures approaching 10 °C are an important stimulus in initiating downstream migration (McCleave 1978; Moore et al. 1995). When smoltification does occur, smolts initiate movement down the rivers and eventually enter the sea. In temperate rivers their displacement is largely nocturnal and affected by factors influencing the water movement such as currents and tides (Thorpe et al. 1981). During daytime the smolts migrate deeper than during nighttime in order to minimize predation by sight-feeding predators (Moore et al. 1995). In the estuaries, fish movement will change from being mostly nocturnal to diurnal and, at a particular salinity threshold, they switch from moving passively with the current to actively swimming seawards (Moore et al. 1995). In-fjord migration Once in the marine environment, the post-smolts encounter a wide range of environmental and physical conditions that influence their swimming depth and migration behaviour. Thus, the fish may face strategic choices for their seaward journey. Post-smolts travel with currents or tides, stay outside currents or use a combination of currents and active movements as they swim towards the sea (Lacroix and McCurdy 1996; Hansen et al. 2003). Fjord migration has been suggested to be one of the greatest challenges faced by the post-smolts as they are required to acclimatise to a "new" environment where predation may be high and the food sources scarce (Hansen et al. 2003). Predation in fjord areas is thought to be a major factor contributing regulating the local populations of migrating Atlantic salmon post-smolts. For example, heavy predation by cod of both hatchery-reared and wild post-smolts has been observed in the estuary of the river Orkla (Hvidsen and.Lund 1988) and the Eira River (Jepsen et al. 2006) in Norway. Predation by avian predators such as gulls on post-smolts in Eresfjord, Norway was also noted during a study on salmon migration (see: 0kland et al. 2006). Changes in the marine environment may also affect the migration performance of Atlantic salmon post-smolts. For example, a rise in river flow may influence the timing of smolt migration such that fish may move to an environment where the temperature and the salinity are not optimal for their survival. In addition, if the smolts migrate too early in the season, the food resources in coastal areas may be scarce due to low water temperatures. Increased water volume flowing into the fjord may also result in changes to the thickness of the brackish water layer that can have profound influences on food resources. For example, Atlantic salmon post-smolts feed mainly on small fish and invertebrates (Hansen et al. 2003) that live at depths where the water temperature and salinity are optimal for their survival. If the brackish water layer thickens, these small individuals may be forced to move further down the water column to areas where the post-smolts may not be able to find them; or alternatively, if the post-smolts also move down the water column they may be at increased risk for predation by other fish that habitat these depths. All of these factors may influence the survival of the post-smolts in the fjords and may cause a decline in fish numbers. However, to date there is no information available on the swimming patterns of post-smolts relative to their environment during their sea migration. Some studies have shown that post-smolts swim at a mean depth of 2 metres once in the fjord and are surface oriented (Doving et al. 1985; Sturlaugsson and Thorisson 1995). In 4 addiction, post-smolts have been recorded to dive regularly during the seaward migration and that the diving activity increases in areas with less linear current systems and hydrographical fronts (Holm et al. 2003). In a 2005 pilot study, during daytime, in Hardangerfjord en post-smolts dived several times from their regular swimming depth 2-3 m to 5-10 m (F. 0kland pers. comm.). The reason for this behaviour is not known, however, it may be a response to sudden physical changes in the water-column (e.g. light, current patterns, salinity, temperature). The influence of environmental factors on post-smolts swimming depth is limiting (Moore et al. 2000). Atlantic salmon post-smolts move rapidly away from coastal waters (Lacroix and McCurdy 1996; Finstad et al. 2005). Some studies have shown that the post-smolts move with the currents and head out of coastal systems on an ebb tide, holding positions during flood tides (Lacroix et al. 2004; Lacroix et al. 2005). However, Finstad et al. (2005) suggested that the post-smolts move actively out of coastal systems and that their movements as well as the orientation cues are independent of currents. Furthermore, Atlantic salmon move in all compass directions, with the lowest frequency towards the fjord. The rates of movement for hatchery-reared Atlantic salmon post-smolts in a Norwegian fjord system range from 0.53 - 0.77 body lengths per second (Finstad et al. 2005; Thorstad et al. 2007). Higher rates of movement have been recorded in estuaries (Fried et al. 1978; LaBar et al. 1978; Moore et al. 1995) and in Passamaquoddy Bay, Canada (Lacroix and McCurdy 1996) probably due to higher water velocities in these localities (Thorstad et al. 2007). Salmon populations adapted to different environments may differ in morphological, behavioural and physiological traits (Taylor 1991). For example, in a study investigating two salmon populations inhabiting tributaries of the Miramichi River, differences in timing 5 of their downstream migration and body morphology were noted (Riddell and Legget 1981). Also, there were differences in migratory routes among groups of post-smolts released in different rivers in Passamaquoddy Bay, Canada (Lacroix et al. 2004). Progression rate and survival may be dependent on the distance from natal locations to the sea. For instance, populations adapted to migrate long distances from the river mouth to the open ocean (e.g. long fjord systems) may have faster and more oriented swimming patterns than populations adapted to migrate short distances to the open ocean. However, there is very limited work available comparing the behaviour of distinct salmon populations in common garden-experiments (Garcia de Leaniz et al. 2007). As I have now described, many factors, such as predation, disease agents and parasites make the fjords to be a challenging environment for out-migrating post-smolts. In the view of all these perils, individuals need to orientate and rush towards the open ocean in order to reach the feeding grounds as soon as possible. The migratory behaviour displayed by the post-smolts has been identified as critical for the success of the Atlantic salmon populations (Hansen et al. 2003; Thorstad et al. 2007). However, many of the factors that affect the performance and survival of the post-smolts during their seaward migration are poorly understood. Thesis objectives The objective of this study was to investigate the overall migratory performance of Atlantic salmon post-smolts, including swimming depth, during their seaward migration and to assess how temperature and water column salinity changes affect their swimming 6 behaviour using biotelemetry techniques in a fjord system. I addressed this objective using a two step approach. In Chapter Two I describe a study that was undertaken to investigate the relationship among temperature, salinity and the post-smolts' swimming depth. The tagged post-smolts were manually tracked using a boat equipped with an acoustic receiver and a directional hydrophone. Ground speeds of the released fish were also calculated. In Chapter Three I have compared the migration performance of two stocks of Atlantic salmon post-smolts from the Flekke and Lasrdal Rivers, located 20 km and 144 km respectively from the open ocean. Progression rates and percent survival were monitored using acoustic listening stations deployed in various areas in the Hardangerfjord system, Norway. 7 R E F E R E N C E S Doving, K.B., Westerberg, H. and Johnsen, P.B. 1985. Role of olfaction in the behavioral and neuronal responses of Atlantic salmon, Salmo salar, to hydrographic stratification. Canadian Journal of Fisheries and Aquatic Sciences. 42: 1658-1667. Finstad, B., 0kland, F., Thorstad, E.B., Bjorn, P.A. and McKinley, R.S. 2005. Migration of hatchery-reared Atlantic salmon and wild anadromous brown trout post-smolts in a Norwegian fjord system. Journal of Fish Biology. 66: 86-96. Fried, S.M., McCleave, J.D. and LaBar, G.W. 1978. Seaward migration of hatchery-reared Atlantic salmon, Salmo salar, smolts in Penobscot River estuary, Maine: riverine movements. Journal of the Fisheries Research Board of Canada. 35: 76-87. Garcia de Leaniz, C, Fleming, LA., Einum, S., Verspoor, E., Jordan, W.C., Consuegra, S., Aubin-Horth, N., Lajus, D., Letcher, B.H., Youngson, A.F., Webb, J.H., Vollestad, L.A., Villanueva, B., Ferguson, A. and Quinn, T.P. 2007. A critical review of adaptive genetic variation in Atlantic salmon: implications for conservation. Biological Reviews. 82: 173-211. Hansen, L.P., Holm, M., Hoist, J.C. and Jacobsen, J.A. 2003. The ecology of post-smolts of Atlantic salmon. In Salmon at the edge, pp. 26-39. Ed by Mills, D. Blackwell Science, Oxford, UK. 8 Holm, M., Hoist, J.C., Hansen, L.P., Jacobsen, J.A., Niall O'Maoileidigh and Moore, A. 2003. Migration and distribution of Atlantic salmon post-smolts in the North Sea and North-East Atlantic. In Salmon at the edge, pp. 7-23. Ed. by Mills, D. Blackwell Science, Oxford, UK. Hvidsten, N.A. and Lund, R.A. 1988. Predation on hatchery-reared and wild smolts of Atlantic salmon, Salmo salar L, in the estuary of the river Orkla, Norway. Journal of Fish Biology. 33: 121 - 126. Hvidsten, N.A., Jensen, A.J., Vivaas, H., Bakke, O., Heggberget, T.G. 1995. Downstream migration of Atlantic salmon smolts in relation to water flow, water temperature, moon phase and social interaction. Nordic Journal of Freshwater Research. 70: 38-48. Jepsen, N., Holthe, E. and 0kland, F. 2006. Observations of predation on salmon and trout smolts in a river mouth. Fisheries Management and Ecology. 13: 341-343. LaBar, G.W., McCleave, J.D. and Fried, S.M. 1978. Seaward migration of hatchery-reared Atlantic salmon (Salmo salar) smolts in Penobscot River estuary, Maine: open-water movements. Journal Du Conseil. 38: 257-269. Lacroix, G.L., McCurdy, P. and Knox, D. 2004. Migration of Atlantic salmon post-smolts in relation to habitat use in a coastal system. Transactions of the American Fisheries Society. 133: 1455-1471. Lacroix, G.L. and McCurdy, P. 1996. Migratory behaviour of post-smolt Atlantic salmon during initial stages of seaward migration. Journal of Fish Biology. 49: 1086-1101. McCleave, J.D. 1978. Rhythmic aspects of estuarine migration of hatchery-reared Atlantic salmon (Salmo salar) smolts. Journal of Fish Biology. 12: 559-570. McCormick, S.D., Hansen, L.P., Quinn, T.P. and Saunders, R.L. 1998. Movement, migration, and smolting of Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences. 55: 77-92. Moore, A., Potter, E.C.E., Milner, N.J. and Bamber, S. 1995. The migratory behavior of wild Atlantic salmon (Salmo salar) smolts in the estuary of the river Conwy, North Wales. Canadian Journal of Fisheries and Aquatic Sciences. 52: 1923-1935. Moore, A., Lacroix, G.L. and Sturlaugsson, J. 2000. Tracking Atlantic salmon post-smolts in the sea. The ocean life of Atlantic salmon: environmental and biological factors influencing survival. Oxford. D. Mills. Fishing News Books, 49-64 pp. Riddell, B.E. and Leggett, W.C. 1981. Evidence of an adaptive basis for geographic-variation in body morphology and time of downstream migration of juvenile Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences. 38: 308-320. Sturlaugsson, J., Thorisson, K. 1995. Post-smolts of ranched Atlantic salmon (Salmo salar L.) in Iceland: The first days of sea migration. ICES CM. 1995/M:15. Taylor, E.B. 1991. A review of local adaptation in Salmonidae, with particular reference to pacific and Atlantic salmon. Aquaculture. 98: 185-207. 10 Thorpe, J.E., Ross, L.G., Struthers, G. and Watts, W. 1981. Tracking Atlantic salmon smolts, Salmo salar L., through Loch Voil, Scotland. Journal of Fish Biology 19: 519-537. Thorstad, E.B., 0kland, F., Finstad, B., Sivertsgard, R., Plantalech, N., Bjorn, P.A. and McKinley, R.S. 2007. Fjord migration and survival of wild and hatchery-reared Atlantic salmon and wild brown trout post-smolts. Hydrobiologia. 582: 99-107. 0kland, F., Thorstad, E.B., Finstad, B., Sivertsgard, R., Plantalech, N., Jepsen, N. and McKinley, R.S. 2006. Swimming speeds and orientation of wild Atlantic salmon post-smolts during the first stage of the marine migration. Fisheries Management and Ecology. 13: 271-274. Electronic Reference WWF (2001). The status of Wild Atlantic Salmon: A River by River Assessment. Available at: http://www.wwf.org.uk/filelibrary/pdf/atlanticsalmon.pdf 11 CHAPTER 2 - Migratory behaviour of Atlantic salmon (Salmo salar) post-smolts relative to the variation in salinity and water temperature1 I N T R O D U C T I O N Atlantic salmon (Salmo salar) smolts start their downstream migration mostly in spring (Thorpe 1988) and travel through fjords towards the open ocean to feed (Hansen et al. 2003; Klemetsen et al. 2003). The onset of migration is controlled by physiological and environmental variables (McCormick et al. 1998). Temperature (Davidsen et al. 2005) and water flow (Solomon 1978) have been suggested as the main factors triggering the migration of smolts. Prior to entering the fjord, juvenile salmon undergo a process called smoltification that enables them to live in the marine environment. These first few weeks as they acclimatize physiologically to the new environment are critical for their survival (e.g. Thorstad et al. 2007). The Norwegian fjord systems present additional challenges to the post-smolts in that there is an overlying freshwater layer from the spring freshet. Consequently, post-smolts encounter three water environments with different physical properties (e.g. salinity and temperature) during out-migration. River regulation may contribute to the spring freshet, for example, in that released water from the dams is put back into the rivers. Post-smolts feed on small invertebrates, mainly insects and crustaceans (Hansen et al. 2003) and these communities may alter their ' A version of this chapter has been submitted for publication in ICES Journal of Marine Science. N. Plantalech Manel-la, J.G. Davidsen, F. 0kland, E.B. Thorstad, R. Sivertsgard, B. Finstad and R.S. McKinley. Vertical movements of Atlantic salmon (Salmo salar) post-smolts related to the variation of salinity and temperature in the water column. 12 habitat due to unusual thickening of the brackish water layer. As a result, post-smolts may also change their swimming behaviour and choose different layers to swim in during their seaward migration. If the post-smolts change their swimming depth, they may encounter a higher number of predators that inhabit these depths and this may have an impact on their survivability. However, little is known about the behaviour of the post-smolts relative to the brackish water layer thickness during their first days at sea. Thus, the primary objective of this study was to examine the swimming depth of post-smolts during the first few days in the fjord relative to water-column salinity and temperature using telemetry techniques. A secondary objective was to calculate the ground speeds of the post-smolts. MATERIAL AND METHODS Study area The study was conducted in Eidfjord in the Hardangerfjord system (115 km long), western Norway (Figure 2-1). The mean depth of this fjord system is ca. 150 m and the maximum depth is 800 m. There is a continuous source of freshwater input extending 30 -40 km into the fjord throughout the year with maximum freshwater input levels achieved in June and July due to snow melt in the mountains. Handling and release of the smolts Eight two-year-old hatchery-reared Atlantic salmon smolts from the Lasrdal River were implanted with acoustic pressure transmitters (model ADT-9-short, 9*34 mm, 13 Thelma, Norway, mass in water/air 3.3/5.3 g). The smolts had a mean mass of 239 ± 32.0 g and a mean total length (LT) of 31.9 ± 3.4 cm (Table 2-1). A seawater tolerance test (Blackburn and Clarke 1987) was performed on 9 April. The mean plasma chloride level of the post-smolts was 146.4 mM at a temperature of 7 °C indicating that they had undergone the smoltification process (Sigholt and Finstad 1990). Acclimatisation to saltwater was achieved by transferring the smolts from freshwater to brackish water (15 - 18 psu); 3 days later, the salinity was increased to 26 - 29 psu for at least 10 days prior to release. To insert the transmitters, smolts were anaesthetized by immersing them in an aqueous solution of 2-phenoxy-ethanol (EC No 204-589-7, SIGMA Chemical Co., USA, 0.5 ml L" 1). The transmitter was then inserted through a 1.2 - 1.3 cm incision on the ventral surface posterior to the pelvic girdle and pushed forward into the body cavity. The incision was closed using two independent silk sutures (4.0 Ethicon). After surgery the tagged smolt was placed in a saltwater tank to recover for 1 - 4 days prior to being transported in a plastic bag to the release site at the outlet of the River Eio (transport time: 15-20 minutes). The tagged smolt was released together with 10-15 non-tagged hatchery-reared Atlantic salmon smolts. Manual tracking and recording of water-column physical variables Individual post-smolts were manually tracked from a boat with a VR60 receiver (VEMCO, Canada). The fish position was fixed every 10 minutes using a GPS receiver (Garmin GPSmap 75). Depth was continuously decoded based on the time delay between two acoustic pulses received; a delay of 1000 milliseconds corresponded to the fish being at the surface, the delay increased at a rate of 100 milliseconds per m below the sea surface. 14 On average, one depth measurement was recorded every 4 seconds. Post-smolts (n=8) were followed for a mean of 11.5 hours after release (range 6.8 - 12.5 hours) (Table 2-1). Six to 19 salinity and temperature profiles were taken while following individual fish. The number of profiles taken was dependent on the weather conditions and the fish movements. In addition, salinity and temperature measurements were also regularly taken at the fish swimming depth during tracking, and an average of 25 salinity and temperature measurements were taken for individual fish (range 10 to 44 measurements) (Table 2-1). To measure the brackish water layer variations during the study period, three temperature and salinity profiles were taken at a reference station (2ST, see Figure 2-1), on 27 April, 19 and 31 May. Data analyses As data from each fish were measured repeated times, the assumption for independence was contravened; therefore, each fish was regarded as an independent data set. Data were summarized accordingly and an average value was presented for each fish (single summary approach; see Grafen and Hails 2002). The swimming depth was averaged every 5 minutes and plotted over contour maps of the vertical salinity and temperature distributions over the fish transect using the program Minitab 14.0. The association between the fish swimming depth and the salinity distribution was analyzed by measuring the amount of time that the post-smolts stayed above (e.g. brackish water) and below the isohaline of 20 psu. In cases where it was not possible to gather profile information for periods longer than 2 hours the fish swimming depth data were excluded from analyses based on isoline plots. The accuracy of the 15 analyses based on isoline plots was evaluated by also analyzing salinity and temperature measurements recorded regularly at the fish swimming depth. The number of large amplitude vertical movements of each individual was counted (defined as > 1- metre movements up and down the water column in less than one minute). The transmitters were calibrated (conditions: 25 °C, 1000 hPa) by the manufacturer and any resulting corrections applied for the atmospheric pressure at the study site were applied at the time when the post-smolts were followed. Corrections were performed by adding or subtracting a centimeter from the depth for each unit of difference in the atmospheric pressure used to calibrate the tags and the hourly atmospheric pressure while each post-smolt was followed. Data indicating positive depths are possible because the transmitters' precision was ± 30 cm. Data were manually verified and any outliers indicating a vertical velocity greater than 1 m/s, and a swimming depth of more than 0.3 m above the sea surface were omitted. Based on these criteria on average, 21 % of the data for individual fish was deleted. RESULTS Environmental conditions The recordings at the reference station (2TS, see Figure 2-1) showed that a brackish water layer was present in the inner part of Eidfjord throughout the duration of the study period (Figure 2-2) and that this brackish water layer thickened and the temperature increased during the study period. The profiles taken while following the post-smolts 16 displayed a similar pattern (Figure 2-3). On 27 April, there were no salinity recordings below 20 psu. On 19 and 31 May, there was a brackish water layer of 2.5 m and 3.5 m respectively. Temperature stratification in the fjord was only recorded on 31 May, when the temperature was 12 °C at the surface but decreased to 9.5 °C at 7 m depth. Migratory Behaviour The migration distance of the post-smolts from the release point at Eira River mouth to where the tracking stopped was on average 8.7 ±3.4 km (Table 2-2). The post-smolts did not follow the shortest migration route; the mean distance from the release point to the outermost recording was 3.4 km, giving a mean migration efficiency of 39%. The average migratory speed, ground speed measured as body lengths per second (bl s"1), for individual fish was 0.7 bl s"1 (Table 2-2). Swimming Depth The mean swimming depth was 1.6 m (range of individual means: 0.5 - 2.2 m) (Table 2-1). The deepest recording for any individual was 5.6 m. The post-smolts performed an average of 2.1 (range of individual means, 0.7 - 3.5) large amplitude vertical movements through the water column per hour (Table 2-2). The mean salinity where the post-smolts migrated was 20 psu (range of individual means, 17 - 23) and the mean temperature was 10.5 °C (range of individual means, 9.5 -12.0) (Table 2-1). There were differences among individuals in the salinity and temperature where they migrated (ANOVA, salinity: F= 4313.9, PO.OOl, temperature: F=39.4, P<0.001). The fish were swimming in brackish water (<20 psu) on average 68% of the time 17 (range of individual means, 25 - 100%) (Figure 2-3, Table 2-2). The fish intersected the isohaline of 20 psu an average of 1.8 times h"1 (range of individual means, 0.0 - 8.4, Table 2-2). Based on the salinity measurements taken at the fish swimming depth, the post-smolts migrated in salinities <20 psu on average 61% of the time (Table 2-1), which is similar to the results obtained based on isohalines (68%, see above). DISCUSSION Our results are the first to show that Atlantic salmon post-smolts swam primarily in the top 1-3 m of the water column where the salinity was mostly below 20 psu during the first few hours after release into the fjord. These results are similar to those obtained by . other authors in previous studies (Fried et al. 1978; Sturlaugsson et al. 1995). Interestingly, the post-smolts did not seem to follow the isohalines, and the mean salinity at the post-smolts' swimming depth varied among individuals. These results indicate that there were no overall preferences among the post-smolts for certain salinities. In the present study we failed to observe any movements to higher salinity layers by the released post-smolts (e.g. individuals did not dive frequently seeking for higher salinity water underneath the top freshwater layer). This was not unexpected as previous work on smolt and post-smolt migratory behaviour suggests that there is no-apparent period of acclimation to the marine environment (e.g. McCleave 1978; Moore et al. 1995; Lacroix et al. 1996; Moore et al. 1998). It appears that the saltwater-acclimated post-smolts used in the present study behaved in a similar manner to that of wild salmon populations that migrate from freshwater to marine water. 18 The post-smolts in this study showed a vertical migration pattern characterized by small and large amplitude vertical movements. During this movement patterns they were clearly experiencing changes in the water salinity as they changed swimming depth. The reason for this behaviour is not known but we speculate that it may be important for either orientation through the fjord, to imprint the odours of the natal stream, to search for prey, to acclimate to the marine environment or more likely a combination of these factors. For adult Atlantic salmon it has been hypothesized that they perform large amplitude vertical movements crossing the pycnoclines to search for prey, and as a means of recognizing the way to their natal stream (Westerberg 1982; Doving et al. 1985). The post-smolts spent a considerable amount of time moving up and down the water column without crossing isotherms (range of temperatures: 9.5-12 °C). However, caution is warranted when interpreting these results as the water column was not stratified for temperature until the end of the study at which time temperature conditions were relatively uniform. Westerberg (1982) studied the swimming depth of adult Atlantic salmon and, in contrast with our results, showed that the salmon moved across isotherms. These contrasting results may be due to differences in isotherm spacing or simply a difference between post-smolts and adults. When the post-smolts monitored in the current study were identified as being 3.4 km from the release site, measured in a straight line, they had in fact been swimming on average about 2.5 times that distance during their migration to the sea indicating that they do not travel in a straight line towards the outer part of the fjord. These movement patterns were similar to the migration behaviour recorded in another Norwegian fjord system (Thorstad et al. 2007). 19 Although the post-smolts had an average ground speed of 0.7 bl s" the fastest swimming fish swam twice as fast as the slowest fish that swam only at a ground speed of 0.6 bl s"1. Interestingly, the ground speeds were considerably slower than the ground speeds (average 1.2-1.3 bl s"1) recorded for Atlantic salmon post-smolts in another fjord system in Norway (Thorstad et al. 2004; 0kland et al. 2006). Differences among studies may be due to differences in, for instance, current speeds, fjord characteristics or in the fish stock origin. The dramatic differences in ground speed noted in the present study may have consequences on the survivability of individual fish. The slowest migrating individuals may be more susceptible to being predated, as they spend greater amounts of time in the fjord and coastal areas where the predation pressure is usually high (e.g. Hansen et al. 2003; Jepsen et al. 2006; Okland et al. 2006). Handling and tagging may influence the behaviour of the fish. For Atlantic salmon post-smolts studies undertaken in the sea, Moore et al. (2000) recommended tags be less than 5% of fish mass to minimize effects on behaviour and survival. In our study, the tag-to-body-mass ratio in air ranged from 1.7 to 2.9% which is well below the recommendation by Moore et al. (2000). Transmitters weighing up to 12% of the body mass have been found not to affect the swimming performance of juvenile rainbow trout (Onchorhynchus mykiss) (Brown et al. 1999). Thus, the transmitters used in the present study were small compared to the fish tagged, and thus adverse effects influencing their behaviour were unlikely. Also, we ensured that all tagged smolts were given a minimum of 24 hours recovery time following surgery and observed for any adverse effects. During the recovery period, all the post-smolts appeared to swim normally and we were not able to distinguish between the tagged and non-tagged post-smolts during these observations. 20 For this study, we used hatchery-reared post-smolts and assumed, based on the work of 0kland et al. (2006) and Thorstad et al. (2007) who reported no differences in migratory behaviour between hatchery-reared and wild Atlantic salmon post-smolts, that they would have similar swimming depth behaviour. However, there is a lack of information concerning the swimming depth of wild Atlantic salmon post-smolts. Unfortunately, presently the pressure transmitters available are too large to be implanted in most of the wild Atlantic salmon post-smolts and thus we were not able to use wild Atlantic post-smolts. In conclusion we observed that the hatchery-reared post-smolts released at the Eio River mouth did not follow a straight migration route towards the open ocean. Our results clearly show that during their journey the vertical movements performed by the post-smolts were independent of water column salinity and temperature. However, interpretation of these results is limited to the brackish water layer conditions that were present during the data collection phase of our study. The stratification of the water column may impact the ability of the post-smolts to acclimatise to the marine environment and the time spent in the fjord areas as they migrate to the open ocean. Future research is needed to fully understand the implications of changes in water column salinity on post-smolt swimming behaviour. 21 R E F E R E N C E S Blackburn, J. and Clarke, W. C. 1987. Revised procedure for the 24 hour seawater challenge test to measure seawater adaptability of juvenile salmonids. Canadian Technical Report of Fisheries and Aquatic Sciences, 1515: 1-39. Brown, R. S., Cooke, S. J., Anderson, W. G. and McKinley, R. S. 1999. Evidence to challenge the "2% rule" for biotelemetry. North American Journal of Fisheries Management, 19: 867-871. Davidsen, J.G., Svenning, M.A., Orell, P., Yoccoz, N., Dempson, B.J., Niemela, E., Klemetsen, A., Lamberg, A., Erkinaro, J. 2005. Spatial and temporal migration of wild Atlantic salmon smolts determined from a video camera array in the sub-Arctic River Tana. Fisheries Research. 74: 210-222. Doving, K. B., Westerberg, H. and Johnsen, P. B. 1985. Role of olfaction in the behavioral and neuronal responses of Atlantic salmon, Salmo salar, to hydrographic stratification. Canadian Journal of Fisheries and Aquatic Sciences, 42: 1658-1667. Fried, S.M., McCleave, J.D. and LaBar, G.W. 1978. Seaward migration of hatchery-reared Atlantic salmon, Salmo salar, smolts in Penobscot River estuary, Maine: riverine movements. Journal of the Fisheries Research Board of Canada. 35: 76-87. Grafen, A. and Hails, R. 2002. Modern statistics for the life sciences. Oxford University Press, Oxford. 351 pp. 22 Hansen, L. P., Holm, M., Hoist, J. C. and Jacobsen, J. A. 2003. The ecology of post-smolts of Atlantic salmon. In Salmon at the edge, pp. 25-39. Ed. by Mills, D. Blackwell Science, Oxford. Klemetsen, A., Amundsen, P. A., Dempson, J. B., Jonsson, B., Jonsson, N., O'Connell, M. F. and Mortensen, E. 2003. Atlantic salmon Salmo salar L., brown trout Salmo trutta L. and Arctic charr Salvelinus alpinus (L.): A review of aspects of their life histories. Ecology of Freshwater Fish, 12: 1-59. Lacroix, G.L. and McCurdy, P. 1996. Migratory behaviour of post-smolt Atlantic salmon during initial stages of seaward migration. Journal of Fish Biology. 49: 1086-1101. McCleave, J.D. 1978. Rhythmic aspects of estuarine migration of hatchery-reared Atlantic salmon (Salmo salar) smolts. Journal of Fish Biology. 12: 559-570. McCormick, S.D., Hansen, L.P., Quinn, T.P. and Saunders, R.L. 1998. Movement, migration, and smolting of Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences. 55: 77-92. Moore, A., Potter, E.C.E., Milner, N.J. and Bamber, S. 1995. The migratory behavior of wild Atlantic salmon (Salmo salar) smolts in the estuary of the river Conwy, North Wales. Canadian Journal of Fisheries and Aquatic Sciences. 52: 1923-1935. Moore, A., Ives, S., Mead, T.A., Talks, L. 1998. The migratory behaviour of wild Atlantic salmon (Salmo salar L.) smolts in the River Test and Southampton Water, southern England. Hydrobiologia 371/372: 295-304. 23 Moore, A., Lacroix, G. L. and Sturlaugsson, J. 2000. Tracking Atlantic salmon post-smolts in the sea. In The ocean life of Atlantic salmon - environmental and biological factors influencing survival, pp. 49-64. Ed. by D. Mills. Fishing News Books, Oxford. Sigholt, T. and Finstad, B. 1990. Effect of low-temperature on seawater tolerance in Atlantic salmon (Salmo salar) smolts. Aquaculture, 84: 167-172. Sturlaugsson, J., Thorisson, K. 1995. Postsmolts of ranched Atlantic salmon (Salmo salar L.) in Iceland: The first days of sea migration. ICES CM. 1995/M:15. Solomon, DJ. 1978. Migration of smolts of Atlantic salmon (Salmo salar, L) and sea trout (Salmo trutta, L.) in a chalkstream. Environmental Biology of Fishes. 3 (2): 223-229. Thorpe, J. E. 1988. Salmon migration. Science Progress. 72: 345-370. Thorstad, E. B., Okland, F., Finstad, B., Sivertsgard, R., Bjorn, P. A. and McKinley, R. S. 2004. Migration speeds and orientation of Atlantic salmon and sea trout post-smolts in a Norwegian fjord system. Environmental Biology of Fishes, 71: 305-311. Thorstad, E.B., 0kland, F., Finstad, B., Sivertsgard, R., Plantalech, N., Bjorn, PA. and McKinley, R.S. 2007. Fjord migration and survival of wild and hatchery-reared Atlantic salmon and wild brown trout post-smolts. Hydrobiologia. 582: 99-107. Westerberg, H. 1982. Ultrasonic tracking of Atlantic salmon (Salmo salar L.). 2. swimming depth and temperature stratification. Drottningholm, 60: 102-115. 24 0kland, F., Thorstad, E. B., Finstad, B., Sivertsgard, R., Plantalech, N., Jepsen, N. and McKinley, R. S. 2006. Swimming speeds and orientation of wild Atlantic salmon post-smolts during the first stage of the marine migration. Fisheries Management and Ecology, 13: 271-274. 25 Table 2-1: Descriptive data from the Atlantic salmon post-smolts tracked using acoustic transmitters in Eidfjord, Hardangerfjord system, Norway (n=8). Post- L T Body Release Release Hours Number of Number of Mean swimming Mean salinity at Mean temperature at smolt (cm) mass date time followed salinity and salinity and depth (m) (s.d., fish depth (psu) fish depth (°C) (s.d., number (g) (hh:mm) temperature temperature range) (s.d., range) range) profiles measurements at fish depth (% of records <20 psu) 3 30.7 241 10 May 11:12 12:00 12 16(36%) 0.5(0.6,0.1 - 1.7) 22 (6.6,9-31) 10.0(1.2, 7.0- 14.0) 4 39.5 223 11 May 11:00 06:45 7 10(90%) 1.9(0.3,0.4-2.6) 18(9.2,0-30) 9.5(1.3,5.5- 10.5) 5 30.2 229 15 May 10:40 12:25 17 30 (53%) 1.7(0.3,0.9-2.6) 20 (5.4, 5 - 28) 10.5 (0.7, 7.5 - 12.0) 7 31.7 257 20 May 14:40 12:00 15 15 (33%) 0.9(1.1,-0.3-5.6) 23 (3.1, 13 -29) 12.0(1.1,9.0-13.0) 10 32.2 291 26 May 11:17 12:00 19 25 (36%) 2.1 (0.6, -0.3-3.4) 18(6.0,3 - 29) 10.0 (0.8, 10.0-11.0) 12 31.5 221 30 May 10:00 12:00 10 21 (38%) 1.7 (0.2,0.0-3.4) 18(0.8,5-29) 10.0 (0.1, 8.0- 11.0) 13 27.7 185 28 May 09:30 12:00 10 38 (100%) 1.7(0.5,0.4-2.4) 18(4.5,4-28) 11.5 (0.8, 11.0- 12.5) 14 31.5 262 29 May 10:00 12:00 6 44(100%) 2.0 (0.2,0.4-2.2) 17(6.7,4-29) 12.0 (0.8, 7.5 - 14.0) Mean 31.9 239 11:23 12 25 (61%) 1.6 20 10.5 (s.d.) (3.4) (32.0) (1:52) (4.7) (11.8) (1.0) (0.8) (3.0) Table 2-2: Distance migrated, speed and vertical movements of Atlantic salmon post-smolts (n=8) tracked during May 2006 in Eidfjord, Hardangerfjord system, Norway. The percentage of time spent at salinities <20 psu are based on isohaline plots. Post-smolt Total Distance from Ground Ground Percentage of time Times hour" Total number Maximum number distance release site to speed speed the post-smolts the post-smolts of large amplitude of migrated outermost point (km h"1) (bl s1) spent in salinities crossed the 20 amplitude vertical movements (km) in the most direct <20 psu psu isohaline vertical (m) line (km) movements 3 9.7 2.8 0.8 0.7 34 0.4 37 2.5 4 3.1 0.9 0.5 0.3 99 0.3 21 2 5 10.1 2.6 0.8 0.8 82 1.7 11 1.5 7 6.8 3.2 0.6 0.5 39 1.1 17 2.3 10 7.6 3.1 0.6 0.6 • 65 8.4 28 1.7 12 14.4 6.6 1.2 1.1 25 2.2 14 1.5 13 10.8 5.4 0.9 0.9 100 0.0 42 2.7 14 6.8 2.4 0.6 0.5 100 0.0 8 1.7 Mean (s.d.) 8.7 (3.4) 3.4(1.8) 0.7 (0.2) 0.7 (0.3) 68 (32) 1.8 (2.8) 22.3 (12.3) 2.0 (0.5) to —) Figure 2-1: Study area. Atlantic salmon post-smolts (n=8) were released and manually tracked using acoustic depth sensitive transmitters in Eidfjord between 10 May and 30 May, 2006. 28 Figure 2-2: Salinity (A) and temperature (B) profiles at the reference station 2TS during the study period. Legend: Circles = 27 April; Squares =19 May; Triangles = 31 May 29 Figure 2-3: The post-smolts' (n=8) swimming depth during transect is plotted over the contour maps of salinity and temperature. The dots indicate the time and the depth where the salinity and temperature profiles were taken. The isohalines and the isotherms were drawn with an interval of 2 psu and 1 or 0.5 °C respectively. The thick continuous line represents the depth of the post-smolt. Time scale shows local time. Salinity Post-smolt 3 1 0 / 0 5 / 2 0 0 6 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Time (hh:mm) Temperature Post-smolt 3 1 0 / 0 5 / 2 0 0 6 -i * ' i 16:00 17:00 .19:00 Time (hh:mm) 21:00 2.1:00 30 Salinity Post-smolt 4 11/05/2006 • • 1 • 1— 11:00 12:00 13:00 14:00 15:00 Time (hh:mm) Temperature Post-smolt 4 11/05/2006 - 7 * • • r * . * 11:00 12.00 i.S w 14:00 • 15:00 Time (hh:mm) Salinity Post-smolt 5 15/05/2006 p—i » 1 • 1 * — • — ^ — — 9 •—i • — r — • — r 11:00 12:00 15:00 14:00 lj.;00 16:00 17:00 18:00 19:00 20:00 21:00 Time (hh:mm) Temperature Post-smolt 5 15/05/2006 Time (hh:mm) Temperature Post-smolt 7 20/05/2006 20/03/2006 1 6:00 20/022006 18:00 20/05/20O6 2O.OO 20/05/2006 22.00 21/05/2006 0.00 21/05/2006 2.00 Day, time (dd/mm/yyyy, hh:mm) Salinity Post-smolt 10 26/05/2006 Time (hh:mm) 12:00 13:00 14:00 15:00. 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Time (hh:mm) Salinity Post-smolt 12 28/05/2006 — i * » • — ' — 1 « * • « • 1 = » 10:00 11:00 ' 12:00 13.00 14:00 15:00 16:00 Time (hh:mm) Temperature Post-smolt 12 28/05/2006 Time (hh:mm) Salinity Post-smolt 13 29/05/2006 Time (hh:mm) Temperature Post-smolt 13 29/05/2006 Time (hh:mm) Salinity Post-smolt 14 30/05/2006 «— — i 1 1 1 v 1 r » 10:00 11:00 12:00 .13:00 14:00 15:00 16:00 17:00 18:00 1!):00 20:00 21.00 22:00 Time (hh:mm) Temperature Post-smolt 14 30/05/2006 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:CK) Time (hh:mm) CHAPTER 3 - A comparison of migratory performance between two stocks of Atlantic salmon post-smolts2 I N T R O D U C T I O N In Norway, Atlantic salmon (Salmo salar) juveniles usually stay in the rivers for 2 - 4 years before migrating to the ocean (Klemetsen et al. 2003). In the spring smolts migrate downstream and then through fjords or coastal areas as they make their way to the open sea. The transition from freshwater to the marine environment is believed to be a critical period for the survival of Atlantic salmon (e.g. Thorstad et al. 2007). However, empirical observations of the migratory behaviour of post-smolts appear to be limited (Moore et al. 2000). Several studies suggest that Atlantic salmon in different river systems belong to genetically distinct stocks (e.g. Stabell 1984, Vespor 1997, Vespoor et al. 2005) and may exhibit their own characteristic distributions both in fresh and saltwater, and display different rates of natural mortality (e.g. Moller 1970). The genetic variability of Atlantic salmon among local environments may result in populations being distinguishable from one another with respect to different life history traits (Taylor 1991). For example, there were genetic differences noted among several Atlantic salmon populations in the same fjord system (Skaala et al. 2007). However, there is a lack of knowledge on the population-specific behaviour and performance of post-smolts during the seaward migration, and whether there are adaptations in the post-smolts' behaviour to different environmental conditions along the migration route. 2 A version o f this chapter will be submitted for publication in the proceedings o f the 7th conference on fish telemetry held in Europe. N . Plantalech Manel-la, J . G . Davidsen, F . 0kland, E . B . Thorstad, R. Sivertsgard, B . Finstad, C M . Chittenden and R. S. M c K i n l e y . A comparison o f migratory performance between two stocks o f Atlantic salmon post-smolts. 38 The objectives of the present study were to describe and compare the migratory performance of Atlantic salmon post-smolts from two different populations. The two populations differed in their migration routes from their original river to the sea: one having navigated through a short fjord and the other through a long fjord. We tested this objective using a common-garden experiment that enabled us to separate between genetic and environmental adaptations when comparing distinct populations' characteristics. To avoid any previous environmental adaptation of the stocks to the migratory route, stocks from two 'foreign' fjord systems were used in this study. M A T E R I A L A N D M E T H O D S Post-smolt groups and study area The study was conducted in the Hardangerfjord system, Western Norway. Post-smolts were released on 25 - 30 April in 2005 and 2006 at the Opo River mouth, in Sorfjorden, Odda (Figure 3-1). The post-smolts used for the study had wild parents that originated from the Rivers Flekke (126 km N-E of the Opo River mouth) and Lasrdal (153 km N-W of the Opo River mouth). The Flekke River empties its waters in the Dalsfjord system and the Lasrdal River outlet is located in the Sognefjord system (see Figure 3-2). A salinity and temperature profile was recorded at the release site each year (Fig. 3-3). Fish handling A total of 139 two-year-old hatchery-reared post-smolts from the Rivers Flekke (n= 80), located 20 km away from the open ocean, and Lasrdal (n=79) (Table 3-1), located 144 km away from the open ocean, were tagged with coded acoustic transmitters (VEMCO Ltd., Halifax, 39 Canada, V8SC-6L-R256 and V8SC-1H-R256 in 2005 and Thelma, Trondheim, Norway, LP-9-SHORT in 2006) using methods described in Chapter 2. Two weeks before tagging, smolts were acclimatised to marine water by initially transferring them from freshwater to brackish water (15 - 18 practical salinity units (psu)) for 2 days and thereafter increasing salinity to 25 - 29 psu. Smolts were maintained at 25 - 29 psu for 1 - 9 days before tagging. Smolts were then transported (90 min drive) directly from Eidfjord to Odda in a transportation tank. Upon arrival they were placed in tanks with circulated brackish water from the fjord (18-22 psu, 5.5 - 8.5 °C) for 24 - 36 hours to allow for recovery before release. Smolts were released at high tide in groups of 24 - 26 individuals (12-13 Flekke and 12 -13 Lasrdal). In 2006, an additional 10-15 non-tagged smolts were released with the tagged smolts. Ten Flekke and 10 Lasrdal smolts were randomly selected from the tanks, killed by a blow to the head, weighed and measured and immediately after blood was aspirated from the caudal vein using a heparin coated syringe. The blood was centrifuged at 1000 rev/min for 5 minutes and the plasma was frozen at -20 °C for chloride analyses. Monitoring system Post-smolts were monitored using 15 acoustic telemetry receivers (model: VR2, VEMCO Ltd, Halifax, Canada) deployed in different areas along Sorfjorden approximately. 2.5, 9.5 and 35 km from the release point (Figure 3-3). In addition, the departure of the tagged post-smolts was monitored using an acoustic receiver moored 0.9 km from the release site and placed at 3 m depth. The receivers monitored and recorded the code and the time of passage for every individual fish within the range of the receiver. The effective detection range for receivers varied 40 with salinity, water column stratification, temperature, weather conditions and wave action. As a result, detection ranges within the fjord could vary by as much as ±300 m. In area 1, two receivers were deployed in two moorings at 3 m depth. In areas 2 and 3, six receivers were deployed in four moorings (Figure 3-3). Each mooring had a receiver attached to a rope 3 m below the surface and the two middle ones had also a receiver positioned 8 m below the surface. This was to ensure maximum coverage in the middle area of the fjord, where the moorings had to be at least 500 m apart due to boat traffic. Data analysis Progression rate out of the fjord and subsequent survival was calculated for three predetermined areas of coverage (Fig. 3-3). The coverage provided by each receiver separating the three areas served as virtual gates. Percent survival for fish was calculated based on the number of tagged individuals detected at each of these virtual gates. The progression rate was calculated in body lengths per second (bl s"1) based on the time from release to the first detection at each of the virtual gates, measured as the shortest distance between each gate. Fish of different origins (Flekke vs. Lasrdal) were compared in terms of percent survival, progression rate and time spent near the release point. Data from progression rates and percentage survival of post-smolts did not follow a normal distribution, and non-parametric tests were used for comparing means between different groups (Flekke vs. Lasrdal stocks) using the program SPSS 10.0. R E S U L T S Sorfjorden salinity in the release area was lower in 2005 than in 2006 (<20 psu) but the surface temperature was similar averaging 7±1°C over the two years (Figure 3-2). 41 There were no differences in total length (LT) and weight between fish from different origins (Flekke vs. Laerdal) in either of the two years (Mann Whitney U-test, weight: P=0.467-0.579, length: P=0.104-0.611). However, the post-smolts were larger in 2006 than in 2005 (Table 3-1), and as a result the two years were analysed separately (Mann Whitney U-test, PO.005). The average levels of plasma chloride in 2005 and 2006 were 125.7 and 131.3 Mm respectively. In 2005 and 2006, 95.6% of the 159 tagged post-smolts were detected at least once while migrating through the fjord. There was no difference in the amount of time spent by Flekke and Lasrdal at the release point before dispersing over the two years (2005 - (Mann Whitney U test, P= 0.775, 2006 - (Mann Whitney U test, P= 0.464)). Detection rate decreased as tagged individuals progressed throughout the fjord. The percentage of post-smolts detected in area 1 was 94.3. In areas 2 and 3, detection rates equaled 91.8%) and 82.3% respectively (Table 3-1). There were differences on the proportion offish from different origins detected in area 2 in 2005 and area 3 in 2006 (2005 - (Fisher's exact test, P= 0.024, 2006 - (Fisher's exact test, P= 0.04)). Mean time spent in each respective area: area 1= 10.2 hours, area 2= 18.6 hours, area 3= 62.3 hours. The mean progression rate for the tagged post-smolts migrating through Sorfjorden (from release point to area 3) was 0.6 bl s"1 (2005) and 0.5 bl s"1 (2006) (Table 3-2) and did not depend on the fish origin in any of the years (2005 - (Mann Whitney U test, P = 0.275), 2006 - (Mann Whitney U test, P = 0.550)). The post-smolts' progression rates, calculated as bl s"1 from release point to area 1 and from release point to area 2, were not dependent on fish origin in any of the two study years (Mann Whitney U-tests: P = 0.12 - 0.41) (Table 3-2). The post-smolts' progression rate was lower in the inner parts Sorfjord, from release point to area 2, than in the outer parts of Sorfjord, from area 2 to area 3 (Krustal-Wallis test, 2005: p=0.002, 2006: p=0.006). 42 DISCUSSION Despite previous work suggesting that the navigation systems in Atlantic salmon vary among local populations (Hansen et al. 2003) our study comparing two stocks of Atlantic salmon post-smolts in a common-garden experiment showed very few differences in terms of survival and progression rate through the fjord. The lack of differences in migratory behaviour between the two populations described in this study appear to indicate that there were no specific behavioural adaptations for the post-smolts in the migratory route resulting in the Lasrdal fish progressing faster through the fjord. In the present study, post-smolts spent more time in the inner part of the fjord as compared to the outer part which was similar to the findings reported by Thorstad et al. (2004) on another fjord system in Norway. These latter authors suggested that such behaviour may be explained by the post-smolts needing to acclimatise or to orient themselves after being released to the marine environment. The need to acclimatise to sea water may not have been a factor in our study as the post-smolts were already habituated to saltwater prior to release. However, the post-smolts may have required more time in the inner part of the fjord system for orientation purposes. The post-smolts' progression rates (mean: 0.6 bl s"1) were similar to those recorded in another study in Romsdalsfjord, Norway (0.5 - 0.8 bl s'\ Thorstad et al. 2007). Ground speeds have been recorded in the same fjord system as the present study using manual tracking protocols, with post-smolts swimming at a mean ground speed of 0.7 bl s"1 over 10 min periods (see chapter 2). These results suggest that despite post-smolts not following a straight route towards area 3 (e.g. the outer part of Sorfjorden), they did not seem to swim a much larger distance than the length of the fjord. In contrast, higher ground speeds (average: 1.2-1.3 bl s"1) have been recorded for Atlantic salmon in another fjord system (Okland et al. 2006; Thorstad et al. 2004), suggesting that the post-smolts swim twice the distance of the fjord. The differences 43 between the two studies could be related to differences in the genetic make-up between the two stocks, or the fjord area or to the orientation of the post-smolts in foreign fjord systems versus local fjord systems or all three. The percentage of fish recorded on the receivers in area 3 was high during both years (2005: 81%, 2006: 84%) indicating high survival of post-smolts during the early marine phase. Our results are similar to those obtained by Lacroix et al (2004) in a study in Passamaquoddy Bay, Newfoundland, where 71 - 88 % of post-smolts survived migration through the Bay. In the present study, the high detection rates may be partly due to the geography of the fjord. Sorfjorden is a straight fjord and it has no "blind" arms where the fish can go through instead of migrating straight to the sea. The measure of survival was considered to be highly accurate because the distance between receivers in each virtual gate was relatively small (200 - 500 m). Although, the virtual gate in area 3 did not close the middle fjord area completely and therefore, the possibility that some post-smolts left Sorfjorden without being detected should not be disregarded. Consequently, the estimated survival proportions through areas 1 - 3 are expected to be higher than that observed. Tagging may have negative effects on survival and swimming performance of the post-smolts at sea. Moore et al. (2000) suggested that to minimize adverse effects on behaviour and survival of tagged fish, transmitters should not exceed 5% of the individual's body mass. In our study, the transmitters ranged from 0.9 to 3.1 % and we allowed the fish to recover from surgery for a minimum of 23 hours, thereby minimizing any potential negative effects of tagging. Majority of tagged individuals were detected in both years and most survived the journey to reach the outermost region of the fjord. Those not detected were likely lost due to predation in the fjord. The predation rate may be high in fjord systems and a challenge for the post-smolts that migrate towards the ocean (Hansen et al. 2003). Close to the Orkla River mouth, in Norway, cod 44 predation on post-smolts has been estimated to be 20% (Hvidsten and Lund 1988). High predation rates have also been observed in the Romsdalsfjord system, Norway (Jepsen et al. 2006). Therefore, the possibility of some of the tags passing the receiver lines in the stomach of a large predator should also be considered but unfortunately we were not able to verify this. However, the migratory speeds observed in our study were comparable to those recorded in other studies and moreover any predator species has been reported to only carry the tag for a short period of time (Andersen 1999, 2001). In both years, a few post-smolts (2005: n=7, 2006: n=12) moved back to areas 2 or 1 after being detected in area 3. This behaviour may be to re-orient themselves or in response to predation. When the fish leave Sorfjorden they can choose several routes, however, only one of them leads to the sea. If the fish choose the "wrong" way, they may need to return to Sorfjorden to re-orientate themselves. Regular coded transmitters were used to follow the fish making it difficult to know if some of the fish that passed the receiver sites were the released fish or a tag in the stomach of a large predator. To circumvent this problem we recommend that future studies make use of depth sensitive transmitters as swimming depth would aid in differentiating between post-smolts and predators. A common-garden experiment was performed to compare the migration performance of two distinct populations of Atlantic salmon post-smolts. The results obtained in this study clearly show that post-smolts' survivability was high at least until the end of area 3, and that they spend more time in the inner than in the outer area of the fjord. In addition, the failure to detect any differences between stocks indicated that there were no specific adaptations that made the Laerdal stock migrate faster or survive better than the Flekke stock through the long Hardangerfjord. Size is known to affect the migratory behaviour of salmon (Garcia de Leaniz et al. 2007) but there were no significant differences between the body lengths of the Flekke and Lasrdal individuals 45 used for the study. Thus, the similar progression rates and percentage survival of the two populations may be also partly due to the similar body size of the stocks. REFERENCES Andersen, N.G. 1999. The effects of predator size, temperature, and prey characteristics on gastric evacuation in whiting. Journal of Fish Biology. 54: 287-301. Andersen, N.G. 2001. A gastric evaluation model for three predatory gadoids and implications of using pooled field data of stomach contents to estimate food rations. Journal of Fish Biology. 59: 1198-1217. Garcia de Leaniz, C, Fleming, I.A., Einum, S., Verspoor, E., Jordan, W.C., Consuegra, S., Aubin-Horth, N., Lajus, D., Letcher, B.H., Youngson, A.F., Webb, J.H., Vollestad, L.A., Villanueva, B., Ferguson, A. and Quinn, T.P. 2007. A critical review of adaptive genetic variation in Atlantic salmon: implications for conservation. Biological Reviews. 82: 173-211. Hansen, L.P., Holm, M., Hoist, J.C. and Jacobsen, J.A. 2003. The ecology of post-smolts of Atlantic salmon. In Salmon at the edge, pp. 25-39. Ed. by Mills, D. Blackwell Science, Oxford, UK. Hvidsten, N.A. and Lund, R.A. 1988. Predation on hatchery-reared and wild smolts of Atlantic salmon, Salmo salar L, in the estuary of the river Orkla, Norway. Journal of Fish Biology. 33: 121 - 126. Klemetsen, A., Amundsen, P. A., Dempson, J. B., Jonsson, B., Jonsson, N., O'Connell, M. F. and Mortensen, E. 2003. Atlantic salmon Salmo salar L., brown trout Salmo trutta L. and Arctic charr Salvelinus alpinus (L.): A review of aspects of their life histories. Ecology of Freshwater Fish. 12: 1-59. 47 Jepsen, N., Holthe, E. and 0kland, F. 2006. Observations of predation on salmon and trout smolts in a river mouth. Fisheries Management and Ecology. 13: 341-343. Lacroix, G.L., McKurdy, P. and Knox, D. 2004. Migration of Atlantic Salmon post-smolts in relation to habitat use in a coastal system. Transactions of the American Fisheries Society. 133: 1455 - 1471. Moller, D. 1970. Genetic diversity in Atlantic salmon and salmon management in relation to genetic factors. The International Atlantic Salmon Foundation. Special Publication Series. 1. No 1: 7-29 Moore, A., Lacroix, G.L. and Sturlaugsson, J. 2000. Tracking Atlantic salmon post-smolts in the sea. In The ocean life of Atlantic salmon, pp. 49-64. Ed. by Mills, D. Fishing News Books, Oxford, UK. Skaala, 0., Wennevik, V., Glover, K.A. 2006. Evidence of temporal genetic change in wild Atlantic salmon, Salmo salar L., populations affected by farm escapees. ICES Journal of Marine Science, 63: 1224-1233. Stabell, O.B. 1984. Homing and olfaction in salmonids: a critical review with special reference to the Atlantic salmon. Biological reviews. 59: 333-388. Taylor, E.B. 1991. A review of local adaptation in salmonidae, with particular reference to Pacific and Atlantic salmon. Aquaculture. 98: 185-207. Thorstad, E. B., Okland, F., Finstad, B., Sivertsgard, R., Bjorn, P. A. and McKinley, R. S. 2004. Migration speeds and orientation of Atlantic salmon and sea trout post-smolts in a Norwegian fjord system. Environmental Biology of Fishes. 71: 305-311. 48 Thorstad, E.B., 0kland, F., Finstad, B., Sivertsgard, R., Plantalech, N., Bjorn, P.A. and McKinley, R.S. 2007. Fjord migration and survival of wild and hatchery-reared Atlantic salmon and wild brown trout post-smolts. Hydrobiologia. 582: 99-107. Vespoor, E. 1997. Genetic diversity among Atlantic salmon (Salmo salar L.) populations. ICES Journal of Marine Science. 54: 965-973 Vespoor, E., Beardmore, J.A., Consuegra, S., Garcia de Leaniz, C , Hindar, K., Jordan, W.C., Koljonen, M.-L., Mahkrov, A., Paaver, T., Sanchez, J.A., Skaala, 0., Titov, S. and Cross, T.F. 2005. Population structure in the Atlantic salmon: insights from 40 years of research into genetic protein variation. Journal of Fish Biology. 67 (Supplement A): 3-54. 0kland, F., Thorstad, E.B., Finstad, B., Sivertsgard, R., Plantalech, N., Jepsen, N. and McKinley, R.S. 2006. Swimming speeds and orientation of wild Atlantic salmon post-smolts during the first stage of the marine migration. Fisheries Management and Ecology. 13: 271-274. 49 Table 3-1: General characteristics of the post-smolts released and number offish detected in area 1 (2.5 km from release), area 2 (9.5 km from release) and area 3 (35 km from release). Values for total length (LT) and weight are: Mean ± s.d. (range). Number of Number of Number of fish detected fish detected fish detected in area 1 in area 2 in area 3 (%) (%) (%) 2005 Flekke 20.5 ± 2.7 77.3 ± 30.4 26 (87) 24 (80) 23 (77) (n=30) (16.1-26.3) (36-150) Lcerdal 20.4 ± 2.7 70.5 ± 26.2 29(100) 29(100) 26(90) (n=29) (16.6-25.5) (33-106) 2006 Flekke 30.0 ±2.7 232.3 ± 63.6 47 (94) 45 (90) 36 (72) (n=50) (23.7-35.9) (118-396) Lserdal 29.4 ± 2.4 224.9 ± 59.3 48 (96) 48 (96) 45 (90) (N=50) (25.3-34.7) (144-389) Year Fish group L T (cm) Weight (g) or release group 50 Table 3-2: Progression rates (bl s"1) and time (h) spent in the release point for the groups of released post-smolts (Flekke and Lasrdal) in Hardangerfjord (2005 - 2006). Values are: Mean ± s.d. (range). Year Group Release to Area 1 Release to Area 2 Release to Area 3 Mean time spent in (2.5 km) (9.5 km) (35 km) the release point (h) 2005 Flekke 0.6 ±0.4 (0.1 - 0.5 ±0.3 (0.2- 0.6 ±0.2 (0.2- 16.1 ± 16.3 (1.9-1.3) 1.2) 1.1) 58.6) Lserdal 0.4 ± 0.2 (0.2 - 0.6 ±0.3 (0.2- 0.7 ±0.2 (0.5- 13.8 ± 11.3 (1.5-0.9) 1.2) 1.3) 55.1) 2006 Flekke 0.5 ±0.5 (0.1 - 0.4 ±0.2 (0.1 - 0.5 ±0.3 (0.2- 7.8 ±7.4 (1.2-2.2) 2.2) 1.6) 38.3) Lasrdal 0.5 ±0.3 (0.2- 0.5 ±0.3 (0.1- 0.6 ±0.3 (0.2- 7.5 ±4.4 (1.1 -1.7) 1.1) 2.2) 20.1) 51 Figure 3-1: Study area. Tagged post-smolts from the Rivers Flekke (n=80) and Laerdal (n=79) were released in spring 2005 - 2006 at the Opo River mouth, Odda, Hardangerfjord system, Norway. Figure 3-2: Map of Sorfjorden with the receivers' locations. Tagged post-smolts (n=159) were released at the Opo river mouth, Odda in spring 2005 - 2006 and monitored using fifteen acoustic receivers. Legend: Squares = Location of acoustic receivers. 53 Figure 3-2: Salinity (A) and temperature (B) profiles in the release area during the study period. Legend: Squares = 19 April 2005; Triangles = 27 April 2006. A) B) Temperature (°C) 54 CHAPTER 4 General discussion The challenges that Atlantic salmon face as they develop from the egg to the reproductive stage have been described by Holliday (2003) using the analogy of thousands of holed Swiss cheese slices lined up in a row. He described their journey as having to navigate through a row of lined up Swiss cheese slices where some of the slices having more or bigger holes than others. Individual fish encounter numerous challenges as they navigate towards the open ocean and these challenges can be equated to navigating through the numerous holes in a Swiss cheese. Moreover, only individuals that are fortunate to select a route where the holes line-up are the ones that will survive to the reproductive stage. Unfortunately, the probability of lining up small holes is less than that for big holes, thus, the bigger the holes are, the more chances the salmon will have to survive and complete its' life cycle. The question facing biologists is how to increase the size of the holes; thereby, increasing the chances of success for the salmon to reach adulthood? If we broaden our understanding of these "holes" we may be able to "control" their size and therefore, increase the chances of success for the Atlantic salmon populations. In this thesis, I have focused on one of these holes: migration performance of post-smolts in a fjord system. The overall goal of my research was to examine the swimming pattern of individual Atlantic salmon post-smolts following release at the river mouth using depth sensitive acoustic transmitters (Chapter 2) and to compare the migratory behaviour of two distinct post-smolts' stocks from their release at the river mouth and throughout the fjord in a common-garden experiment (Chapter 3). With the advent of acoustic telemetry, a number of research studies have focused on the seaward migration of smolts. Most of this research has been conducted in rivers and estuaries (e.g. Moore et al. 1995, 1998), with very few studies undertaken in fjords and coastal areas 55 (Lacroix et al. 2004; Thorstad et al. 2004, 2007). However, these research initiatives have mainly focused on the two-dimensional behaviour of smolts and post-smolts. The swimming depth study (Chapter 2) represents the first accurate recording of three-dimensional movements of post-smolts during the first hours at sea. Results of my research show that out-migrating post-smolts spent the majority of time in the top 3 m of the water column, most of the time in brackish water (e.g. salinity <20 psu) performing irregular dives to more saline water. In addition, individuals did not show preference for sub-layers with a certain temperature or salinity concentration. However, the post-smolts were acclimatized to saltwater prior to release and therefore, in the study described in Chapter 2, it was assumed that they behaved in the same manner as fish found in nature. I would recommend undertaking further studies using post-smolts not previously acclimated to marine water or, more appropriately, studies comparing the behaviour of post-smolts previously acclimated to saltwater and post-smolts not acclimated to saltwater. The post-smolts' movements did not appear to be related to temperature or salinity gradients. Since the water column was not very stratified during the study, individuals may not have had a choice on changing salinity or temperature layers as they changed swimming depth. Unfavourable conditions of temperature or salinity in the fjord areas may result in migrating fish lingering around the release area to acclimatise to the marine environment. Further, river regulation may alter the river flow and increase the water temperatures. Such combination of factors may trigger early smolt migration and, if the temperatures in the fjord have not yet increased, the migrating individuals my encounter conditions less favorable to their survival. The findings presented in this thesis may provide additional information for use in the decision making process when it comes to control of river flows for hydro-power production or for water supply to recreational, domestic, industrial or agricultural activities. 56 Atlantic salmon populations from the Flekke and the Lasrdal Rivers have similar survival and progression rates through Sorfjorden. It appears that adaptation to migration route length from the natal river to the ocean does not affect the performance of these post-smolts during their seaward migration. However, in the study described in chapter 3, the behaviour of only two populations of salmon was compared and also, genetic analyses were not undertaken to verify that the post-smolts belonged to different local populations, instead, it was assumed that the stocks used for the study were genetically distinct based on the homing tendency of Atlantic salmon and the low percentage of strayed individuals that spawn in a different river from the natal one (Hindar et al. 2004). Future research efforts should focus on comparing the migration performance of many populations adapted to different characteristics of the migratory route and, at the same time, perform genetic analyses of the stocks. In Norway, 19% of the salmon rivers are considered to have populations in extinct or critical condition (WWF 2001). One strategy to 'recover' Atlantic salmon populations from such rivers would be to introduce salmon from a different river system. However the 'survival' of introduced Atlantic salmon populations in a foreign fjord system is a concern. For example, McGinnity et al. (2004) examined the lifetime success and performance of native and non-native Atlantic salmon stocks under communal conditions and observed that non-native salmon have much lower adult returns than native salmon. In the present study, the high survival of Flekke and Lasrdal groups through Hardangerfjord indicated that non-native salmon stocks may perform well, at least during their seaward migration. With the work described in this thesis I have contributed to our scientific knowledge of post-smolts' behaviour during their first days at sea. I have partially filled a few of the knowledge gaps required to increase the size of the Swiss' cheese holes by having discovered that out-migrating salmon swim most of the time close to the surface in low-salinity water and that different populations of salmon have similar survival and progression rates. However, there are 57 still many knowledge gaps that need to be solved. Many parts of the salmon behaviour at sea are still a mystery to us and if we want to have healthy populations again, further investigations are required. Additional areas of consideration for future research: 1. The navigation systems for Atlantic salmon may vary from population to population and the post-smolts' capacity to orientate through a fjord system may vary depending on the geography of the area. Thus, if we want to create a 'model' for Atlantic salmon migration, we should study the migration behaviour (e.g. survival, progression rates, swimming depth, orientation capacity) in fjord systems with distinct characteristics. 2. It is important that research efforts also focus on the effects of predators on the swimming behaviour of migrating post-smolts. Post-smolts may change their swimming behaviour in presence or absence of predators; they may swim in deeper water layers in presence of predators. One way to examine this hypothesis is by performing a laboratory experiment to compare the behaviour of post-smolts in the two situations. 3. Monitor the post-smolts' swimming depth for long distances to examine their behaviour during acclimation to seawater, from the river mouth, when a brackish water layer is present, to the sea. 58 4. Physiology and post-smolts' behaviour during acclimation to the marine environment. On a laboratory setting, simulate a brackish water layer and stepwise increase the salinity. Monitor the post-smolts' behaviour with a video-camera and take physiological samples several times during the experiment. 5. Some studies indicated that wild and hatchery-reared post-smolts have a similar horizontal migratory behaviour but yet it is unknown if their vertical migratory behaviour is similar or not. Thus, a study comparing the swimming depth between wild and hatchery-reared post-smolts would be necessary to test if the results obtained using hatchery-reared salmon reflect the vertical movements of wild salmon as well. 59 REFERENCES Garcia de Leaniz, C, Fleming, I.A., Einum, S., Verspoor, E., Jordan, W.C., Consuegra, S., Aubin-Horth, N., Lajus, D., Letcher, B.H., Youngson, A.F., Webb, J.H., Vollestad, L.A., Villanueva, B., Ferguson, A. and Quinn, T.P. 2007. A critical review of adaptive genetic variation in Atlantic salmon: implications for conservation. Biological Reviews. 82: 173-211. Hindar, K., Tufto, J., Sasttem, M.L. and Balstad, T. 2004. Conservation of genetic variation in harvested salmon populations. ICES Journal of Marine Science. 61: 1389-1397. Holliday, F. 2003. Eroding the edge - The future of coastal waters. In Salmon at the edge, pp. 1-4. Ed by Mills, D. Blackwell Science, Oxford, UK. Lacroix, G.L., McCurdy, P. and Knox, D. 2004. Migration of Atlantic salmon post-smolts in relation to habitat use in a coastal system. Transactions of the American Fisheries Society. 133: 1455-1471. McGinnity, P., Prodohl, P., Maoileidigh, N.6., Hynes, R., Cotter, D., Baker, N., O'Hea, B. and Ferguson, A. (2004). Differential lifetime success and performance of native and non-native Atlantic salmon examined under communal natural conditions. Journal of Fish Biology. 65: 173-187. Moore, A., Potter, E.C.E., Milner, N.J. and Bamber, S. 1995. The migratory behavior of wild Atlantic salmon (Salmo salar) smolts in the estuary of the River Conwy, North Wales. Canadian Journal of Fisheries and Aquatic Sciences. 52: 1923-1935. 60 Moore, A., Potter, E.C.E., Milner, NJ. and Bamber, S. 1998. The migratory behavior of wild Atlantic salmon (Salmo salar L.) smolts in the River Test and Southampton Water, southern England. Hydrobiologia. 371-372: 295-304. Thorstad, E. B., 0kland, F., Finstad, B., Sivertsgard, R., Bjorn, P. A. and McKinley, R. S. 2004. Migration speeds and orientation of Atlantic salmon and sea trout post-smolts in a Norwegian fjord system. Environmental Biology of Fishes. 71: 305-311. Thorstad, E.B., 0kland, F., Finstad, B., Sivertsgard, R., Plantalech, N., Bjorn, PA. and McKinley, R.S. 2007. Fjord migration and survival of wild and hatchery-reared Atlantic salmon and wild brown trout post-smolts. Hydrobiologia. 582: 99-107. Electronic Reference WWF (2001). The status of Wild Atlantic Salmon: A River by River Assessment. Available at: http://www.wwf.org.uk/filelibrary/pdf/atlanticsalmon.pdf 61 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0101117/manifest

Comment

Related Items