UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Activity and behaviour in spawning sockeye salmon Lake, Randal Gary 1999

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

Item Metadata


831-ubc_1999-0543.pdf [ 4.93MB ]
JSON: 831-1.0099409.json
JSON-LD: 831-1.0099409-ld.json
RDF/XML (Pretty): 831-1.0099409-rdf.xml
RDF/JSON: 831-1.0099409-rdf.json
Turtle: 831-1.0099409-turtle.txt
N-Triples: 831-1.0099409-rdf-ntriples.txt
Original Record: 831-1.0099409-source.json
Full Text

Full Text

ACTIVITY A N D BEHAVIOUR IN SPAWNING SOCKEYE SALMON by RANDAL GARY LAKE Technical D i p l o m a , B r i t i s h Columbia Institute o f Technology, 1989 B . S c , Simon Fraser University, 1994  A THESIS SUBMITTED IN P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Department o f Forest Sciences)  We accept this thesis as conforming to the^equired standard  T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A June 1999 © Randal Gary L a k e , 1999  UBC  Special Collections - Thesis Authorisation Form  Page 1 of 1  In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the head o f my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s understood t h a t copying o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n .  The U n i v e r s i t y o f B r i t i s h Columbia Vancouver, Canada  http ://www. library .ubc. ca/ spcoll/thesauth. html  8/26/99  ABSTRACT Total energy expenditures and muscle activity have been measured in spawning salmon but behaviour specific energy-use has never been been measured directly. This research used electromyogram ( E M G ) telemetry, combined with behaviour observations to assess activity levels and estimate relative energy use during spawning in sockeye salmon, Oncorhynchus  nerka.  M y main objectives  were to: assess the ability of E M G transmitter technology to discriminate between the rate o f muscle activity during short duration specific behaviours and general activity, and; use these data to compare the total activity between spawning stages, sex, and year. E M G data were converted to rate of tail beats and, using synchronized clocks, were matched with specific spawning ground behaviours. Rate o f tail beats, behaviour duration, and behaviour frequency were multiplied to estimate the total tail beats for the mean duration of male and female fish on the spawning grounds. Sequential reproductive stages were observed in the spawning stream. Males and females were given a 'status' designation corresponding to their reproductive stage. Behaviour specific analysis o f total tail beats for each behaviour revealed that males and females o f different reproductive status had different mean total tail beats for some behaviours.  Total tail beats for males and  females in two consecutive years revealed that females utilized more tail beats in one year and equal tail beats to the males in the other year.  Total tail beats were  strongly associated with duration of each fish in the spawning stream. This along  ii  with other observations suggests that the majority o f energy expended on the spawning ground was due to non-specific general behaviours such as "holding position" (eg. behaviours not involved in act o f spawning or defense/competition for spawning areas). E M G technology appears to be a highly effective tool for studying spawning ground activities in salmon.  iii  T A B L E OF C O N T E N T S Title  Page  ABSTRACT  ii  L I S T o f T A B L E S and F I G U R E S  v  ACKNOWLEDGEMENTS...  vi  DEDICATION  vii  INTRODUCTION.  1  Background:  Life history and reproductive behaviour of  Background:  Spawning activity and energetics in  METHODS  fish  fish  4 18 32  Study Site Description  32  Transmitter Implantation  34  Behaviour Observations  37  Data Organization and Analysis  41  RESULTS  45  DISCUSSION  65  REFERENCES  .....73  A P P E N D I X I: D E F I N I T I O N OF T E R M S  82  A P P E N D I X II: F I S H E S C A P E M E N T T O G L U S K I E C R E E K  83  iv  L I S T OF T A B L E S A N D F I G U R E S Title  Page  Table 1: Specific behaviours observed  39  Table 2: Summary of study fish  44  Figure 1: L o c a t i o n o f the study site  33  Figure 2: Tailbeat expenditure rates for specific behaviours  53  Figure 3: Duration of specific behaviours  54  Figure 4: Tailbeats used for specific behaviours  55  Figure 5: Frequency for specific behaviours  56  Figure 6: Tailbeats hr" for specific behaviours  57  1  Figure 7: Tailbeats hr" for spawning fish  58  1  Figure 8: Baseline, schooling, and spawning tailbeats hr"  1  59  Figure 9: Duration of fish on the spawning ground  60  Figure 10: Total tailbeats by sex and year  61  Figure 11: Upstream movement o f study fish  62  Figure 12: D i e l activity levels of fish  63  V  ACKNOWLEDGEMENTS I would like to thank my thesis committee - D r . Scott H i n c h , D r . M i c h a e l Healey, and D r . Kathy M a r t i n for their valuable advice in the design and development o f my thesis.  To D r . Scott Hinch I extend my gratitude and  appreciation for his experience and support while I developed my thesis. I would like to thank Pier van Dishoeck, an excellent field technician, for his enthusiastic assistance in the  field.  The Department of Fisheries and Oceans assisted by  allowing me to obtain sockeye salmon from the fish fence in Gluskie Creek and providing ancillary field data. M y field research and thesis development was supported by the Fraser River A c t i o n Plan through Canada's Green Plan and an N S E R C grant to D r . Scott H i n c h .  R.G.L.  vi  In memory of Dr. Donald  Guthrie  "Beyond is the great and w i l d sea with its living things too many to number, creatures both small and large. Here move the ships, and when come those great beasts, you chase them for the sport o f it. A l l o f them look to you to give them their due season. Y o u give it to them; they gather; you open your hands, and they are filled with good things. Y o u show your face, and they are fearless; you take away their breath, and they return to ashes. Y o u send forth your Spirit, and they are renewed, and you renew the face of the earth".  - Randal Lake, adapted from Psalm  vii  104.  INTRODUCTION  Pacific salmon stop feeding prior to initiation o f their upstream spawning migration. Completion o f the migration and successful spawning are dependent on energy reserves that the fish possess at the start o f migration. T w o studies have explored energetic costs o f upstream migration in sockeye salmon (Idler and Clemens 1959, Gilhousen 1980). This limits analysis to quantifying the total energy used during stages o f Pacific salmon migration but not behaviour-specific analysis o f energy expenditure associated with spawning. A l t h o u g h qualitative descriptions o f the behaviour of Pacific salmon, Oncorhynchus,  on the spawning  grounds have been made, measures o f spawning ground energy expenditure are rare.  Spawning studies could not look in detail when and where these costs were  incurred. Recent developments in radio transmitter technology allows the quantification o f muscle contractions.  H i n c h et al. (1996), H i n c h and Rand  (1998), Rand and H i n c h (1998) used Electromyogram ( E M G ) radio transmitters to study the behaviour and lateral swim muscle activity during upstream migration o f the early Stuart sockeye salmon through specific reaches and areas of difficult passage in the Fraser River. Laboratory and field models o f muscle activity were developed. Results o f these studies show that E M G radio telemetry is successful in identifying reaches and features o f difficult passage, and provides insight into the behaviour and muscle activity o f fish during upstream migrations.  1  The field activity of lake trout (Salvelinus namaycush) during their reproductive period in White Pine L a k e , Ontario, was used to infer spawning activity o f lake trout (Weatherley et al. 1996). The E M G s were characteristic of similar E M G s obtained from spawning lake trout in a laboratory study (Kaseloo et al. 1996). This laboratory study was used to calibrate E M G s to activity for assessment o f activity in the field, and led to the successful identification o f spawning activity in at least one of two lake trout implanted with E M G transmitters in the field study done by Weatherley et al. (1996). Although activity measures are not measures o f potential energy, energy is defined as the capacity to do work (Betts 1976), and tail beats do work. Therefore, there is a relationship between activity (tail beats) and energy. Standardizing lateral swim muscle contractions for all activities as tail beats, allows comparison o f ' t a i l beats' between different behaviour factors. A l t h o u g h the actual rate o f energy expenditure is not yet know for this experimental approach, it is assumed that a standardized measure o f work done (rate o f muscle contractions regressed to rate o f tail beats in a lab analysis) can be used as an index o f activity among behaviour factors and may be associated with rates o f energy expenditures.  This requires an assumption that energy expenditure and  activity are positively correlated, and that the rate o f energy expenditure is the same at specific rates and duration of activity. B o d y constituent analysis i f done frequently during stages on the spawning grounds, or measures o f oxygen consumption in an appropriate laboratory study, may eventually allow the  2  development o f an accurate model to convert different rates o f muscle activity to actual energy expenditure in spawning sockeye salmon. The general objective of this thesis is to measure muscle activity and standardize it as rate of tail beats ( T B s" ) and total tail beats ( T T B ) . 1  following are specific objectives:  The  1) I w i l l assess the ability o f E M G technology  to discern the muscle activity for very short duration behaviours.  Specific  behaviours, which are typically of short duration, w i l l be correlated to E M G values to show i f muscle activity can be quantified for these short duration behaviours.  2) M u s c l e activity w i l l be compared between specific behaviors and  general activity, and also among sexes, spawning stage, and spawning years. This w i l l show how activity is used by different fish throughout their duration on the spawning ground. 3) This information w i l l be used as a relative index o f energy expenditure, to assess i f the Fraser River sockeye salmon 'body constituent' conclusions o f Gilhousen (1980), that energy expenditures are greater in males and consistent among years, are reasonable. Gilhousen (1980) did not state assumptions that were used to calculate energy utilized by kilogram/fish or kilogram/fish/day. B a c k calculations revealed two possible unstated assumptions.  It was found that Gilhousen used the same  spawning ground duration between sexes in all years, and sometimes between years.  A more precise evaluation o f activity and life duration on the spawning  grounds may provide greater understanding of sockeye salmon spawning activity.  3  BACKGROUND  Life H i s t o r y and Reproductive Behaviour of Fish. Fish may partially or fully stop feeding during reproduction. Fasting may be adaptive when an absence or variability in food resources occurs. Energy utilized while searching for prey when it is scarce or absent, especially in an environment where there are many other spawning fish to compete with, may not be offset by energy obtained from captured prey. In a food abundant environment fasting may prevent spawning fish from eating their own embryos. Reproductive success depends on utilization and re-allocation of stored somatic energy and effective behaviour tactics to support reproductive effort. Reproductive effort includes energetic costs o f migration, mating, spawning, parental care, and costs to future reproduction ( C l u t t o n - B r o c k 1984). This section w i l l examine reproductive effort and associated costs to fish. Fish have various life history strategies which ultimately are adaptations to optimize reproductive success.  Reproductive success may be defined as the  number o f offspring surviving to reproductive maturity relative to those produced by other fish in the population (Wootton 1985). Reproductive effort can be defined as the accumulated investments, or costs, during a single reproductive event.  Examples o f reproductive effort include: migration, gamete development,  spawning behaviours, and parental care (Roff 1992). Reproductive effort has two types o f associated costs.  These are: the energy utilized during reproduction,  4  and costs to future reproduction and survival ( C l u t t o n - B r o c k 1984; R o f f 1992). When surplus somatic energy is insufficient for reallocation to reproduction, future growth, fecundity, and survival may be reduced. Salmonids are a diverse family o f fishes including both semelparous (Pacific salmon) and iteroparous species. The life history and evolution o f salmonid species are among the most studied. However, life history studies rarely quantify behaviours and none have quantified specific behaviours energetically. The life history of Pacific salmon species are similar in a general sense, and sockeye salmon are typical o f the other species (Burgner, 1991). M o s t sockeye salmon juveniles rear in lakes for 1 or 2 years, therefore spawning streams have close access to lakes. However, some juveniles may rear in rivers or their backwater pools. A few stocks are prevented by river flows from migrating upstream to lake rearing habitat and migrate downstream to the river estuary instead. F o r example, in B r i t i s h Columbia, emergent fry in the Harrison River tributary in the lower Fraser River migrate out to the Fraser River estuary.  Alternately, some sockeye salmon stocks, such as the Cultus L a k e stock  beach spawn in the rearing lake itself. Upper Fraser R i v e r stocks are typical of sockeye salmon in that fry emerge from streams and migrate upstream or downstream, in the appropriate direction, to lake rearing habitat.  Sockeye  salmon are distinctive from other Pacific salmon species in their predominant requirement for lake rearing habitat with associated spawning streams (Burgner 1991).  5  Sockeye salmon juveniles smolt and migrate down to coastal ocean waters in the spring. B r i t i s h Columbia stocks of southern origin migrate northward along the coast of B r i t i s h Columbia and A l a s k a while feeding in the nutrient rich waters o f the continental shelf. Juveniles move offshore into the deep Alaskan gyre sometime during late fall or winter. Depending on the stock and individual they circulate in the gyre from one to four years, while developing into sub-adults (Burgner 1991). F o o d abundance is often patchily distributed and variable between years. G r o w t h rate effects survival because predation on sub-adults is size dependent (Stearns 1976).  Swim speed is correlated with fish length and lateral swim  muscle mass. Burst swimming away from predators is an important method o f escape in the open ocean. It retains its importance during upstream spawning migrations through high velocity obstacles in rivers. As two to five year old maturing adults, British Columbia sockeye salmon exit the Alaskan gyre into coastal waters.  Southward migration follows the coast  using homing cues until the river estuary is reached.  Schooling in the estuary is  common and the duration is greater with later arriving stocks. Feeding may stop weeks prior to fresh water entry (Foerster 1968). Time o f schooling, fresh water entry, migration distance, and time to the spawning ground varies between stocks. The 'upper river' early Stuart River sockeye salmon enter the Fraser River from the ocean in July and migrate 1200 kilometers. In contrast, the Tower  river' Cultus L a k e stock enters the Fraser River from the ocean in September and has a short migration of 88 kilometers (Gilhousen 1960). Upstream migration o f sockeye salmon typically consists of mixed stocks for most o f the multi-stock systems. Fraser River stocks have exceptionally discrete peaks o f migrations such that peak migration timing allows identification of the different stocks while still in the Fraser River mainstem (Gilhousen 1960; K i l l i c k 1955). These discrete migration peaks are an adaptive effect o f the extremely different environmental conditions existing in each stocks' migration route and spawning location. Average egg incubation temperature is often different between spawning locations and effects the rate o f egg development. Timing o f spawning and subsequent fry emergence is critical for fry survival (Brannon 1987). Temperature and flow conditions encountered by adult sockeye salmon vary between migration routes and years, effecting energy utilization, migration behaviour, and survival (Idler and Clemens 1959; Gilhousen 1980; Rand and Hinch 1998; H i n c h and Rand 1998). Many large rivers have mean velocities which exceed the maximum sustained swim speed o f sockeye salmon (Osborne 1961; Brett 1965). H i n c h and Rand (1998) studied the relationship o f local environment and fish characteristics to swim speeds and energy use in adult sockeye salmon. They used identical electromyogram telemetry on the same years and stock o f early Stuart sockeye salmon as I used in this thesis.  They  demonstrated that upriver migrating fish selected l o w velocity pathways and  7  reverse flow pathways (i.e. eddies). Fastest migration speeds (m s" ) were 1  through l o w velocity reaches and slowest through constricted reaches such as H e l l s ' Gate canyon, considered the most difficult passage point in the river. Only differences in reaches significantly accounted for differences in migration speed. Differences in swim speeds (tailbeats s" ) and energy use between years appeared 1  most correlated with river flow dynamics. When reaching the spawning grounds some stocks may school for a variable time before beginning to spawn. L o w e r river stocks may school for weeks. Their fresh water residence is almost as lengthy as upriver migrating stocks.  Early  river entry and holding is variable in sockeye salmon stocks and other Pacific salmon species. M o s t salmonids spawn in streams.  The behaviour o f lake spawning  populations is rarely observed and studied, so it w i l l not be further discussed. Stream spawning behaviour of Pacific salmon species is highly similar, but differences exist with respect to some male spawning tactics. N o single study appears to encompass the full compliment of behaviour in sockeye salmon, or other salmon species.  Often, contests and territory defense behaviours are  lumped together as 'aggressive acts' ( M c C a r t 1971). A general review of studies is required to describe the behaviour o f spawning salmon. Riffles are preferred sites for establishment o f spawning territory and redds.  Hoopes (1972) found streambed morphology was the single most  important factor in selection o f a spawning territory, or redd. Gravel-cobble  8  substrate found in riffles is preferred.  Higher gradient sites with cobble bottoms  are used less often, and low gradient pools are used least.  The tail-spill is used  most often in pools, and it should be noted that water accelerates out o f the pool at these sites. In riffles, females chose gently sloping sites where the excavated tail-spill created while nest digging provided a slight rise. Spawning occurred almost always in riffle gradients between 0.5% to 2.0%.  Sockeye salmon  spawning in Four M i l e creek, tributary to Babine L a k e , had a mean water flow over the nests of 21.9 cm-s" ( M c C a r t 1971). Establishing territory for spawning 1  may occur by selection o f an unoccupied site, or displacement o f another fish by aggression.  When a male or female is seeking a spawning territory, they may  instead j o i n a territory held by a fish of the other sex.  Spawning territory is  usually established by the female and subsequently joined by a male (Hanson and Smith 1967). Hanson and Smith (1967) studied mate selection in sockeye salmon and found that males prefer to mate with a larger female, possibly due to her vigor and fecundity. Large males usually defeat smaller males during contests for females.  Therefore, the largest males mate with the largest females and the  smaller males mate with the smaller females.  This has the effect o f mated males  being the same size or slightly larger than their female. When the males greatly out number the spawning females, the largest males w i l l mate with females of all sizes. Smaller and senescing males may associate with spawning pairs as subdominant satellites.  Size was show to be an important determinant in male  9  dominance in sea trout which may have up to 10 males in association with a spawning female (Evans 1994). Behaviour interactions between a male and female are occasionally called 'courtship' behaviours (ex. quivering), but I have not seen any study that provides evidence that these are courtship behaviours (i.e. no studies have defined and identified courtship behaviour in sockeye salmon).  Size assortive  mating does occur, but by dominance contests between males for possession of a territory and access to a female. Females may chase away smaller males and accept equal size or larger males. Female sockeye salmon spawning with larger males appear to extrude all their eggs in less time than other females (Foote 1989). Mate choice by males and females may be by morphological characteristics, with body size being strongly assessed by a potential mate. Younger jack males occur in spawning sockeye salmon populations. They are older and larger than the jacks associated with coho salmon populations, and have some development o f secondary sexual characteristics including a bright red or pink color. Because o f their larger size and lack of crypsis, they may use a satellite strategy, spawn in extremely marginal spawning habitat, or spawn with female kokanee (the non anadromous form of sockeye salmon) (Foote and L a r k i n 1988). M a l e kokanee often spawn as satellites with anadromous sockeye salmon. Pacific salmon usually dig a series o f 3 to 5 successive nests known as a redd (Foote 1989). One nest precedes another in an upstream direction. Tautz and Groot (1975) reported chum salmon, O. keta, and rainbow trout, O. mykiss,  10  spawning in a flume started digging at sites with an upwelling flow.  The  upwelling could be created by placing a rise o f stones or w o o d on the bottom. The female would drift backwards until its tail encountered the acceleration o f the upwelling flow.  D i g g i n g would commence immediately upstream.  When  chum salmon females were offered a site with upwelling but no substrate, and a site with gravel substrate but no upwelling, digging occurred in the site with upwelling and no substrate. Sites with substrate and upwelling were preferred most. Rainbow trout initiated digging at sites o f upwelling. Heard (1972) observed the spawning behaviour of pink salmon in a tank with varied flow rates. During low flows (0.9 L s" ) the fish milled about the tank or held at the side of 1  the tank, and completely ignored previously established redds.  The flow was  generally similar to a pool. During higher flows (7.65 L s" ), with a velocity 1  'upstream' o f the redd o f about 15 cm/s, females moved back onto their redd and defended it from other females.  Female digging commenced and males appeared  more active. It should be noted from the synthesis o f information that the female appears to be oriented to the point o f commencement o f digging by an acceleration of water flow.  Each subsequent excavation creates an upwelling and  acceleration o f water flow.  Given a serial pattern o f excavating and backfilling  nests, the direction o f redd digging would always proceed in an orderly upstream direction. M c C a r t (1969) studied the nest digging behaviour o f sockeye salmon in tributary streams o f Babine L a k e , B r i t i s h Columbia.  11  The females constructed the  nest by repeated digging over the same area.  A ' d i g ' consisted o f a continuous  series of body flexures while in a horizontal position. Between digs the female was positioned immediately below the nest excavation. Occasional probing o f the nest occurred by the female settling into the nest with pectoral, pelvic, and anal fin extended. approaches.  The frequency o f this activity increases as gamete release A dig begins with the female moving upstream over the nest and  turning on her side with pectoral and pelvic fins extended.  A rapid series o f body  flexures occurs, dislodging substrate with the caudal peduncle and fin. The dislodged substrate drifts and resettles slightly downstream o f the dig site. Successive flexures often carries the female forward and upward. After the dig, the female returns to a position downstream o f the nest. The frequency of digging changes before and after a gamete release (Belding 1934; White 1942; Briggs 1953; Sheridan 1960; Mathisen 1962; M c C a r t 1969; Tautz and G r o o t 1975). Tautz and Groot (1975) recorded diel periodicity of chum salmon digging, and noted an abrupt increase associated w i t h gamete release.  Sheridan (1960) studied the frequency o f digging o f female pink salmon  before and after gamete release, in two streams in south-east A l a s k a . observation was made in each stream.  One  In both events the females dug at a regular  rate o f about 1 dig min." prior to gamete release. 1  A s gamete release approached,  digging slowed or stopped and the female increased probing. Immediately after gamete release each female dug at 7 and 9 digs min." . A s the nest became 1  backfilled, the digging slowed to frequencies similar to before gamete release.  12  This also coincides with the backfilling o f the nest becoming the excavation o f the next nest. Increased digging after gamete release was reported for sockeye salmon (Mathisen 1962), Atlantic salmon, Salmo salar, (Belding 1934), and for coho salmon, O. kisutch,  (Briggs 1953).  M c C a r t (1969) reported the duration, number o f flexures per dig, and frequencies o f digs. D i g frequency was about 1-min." before gamete release and 1  about 5.4-min." after gamete release. 1  The number of flexures and duration per  dig was inversely related to the frequency o f digs. to 2 per dig after gamete release.  Flexures decreased invariably  A n increase in digs and decrease in flexures  after gamete release is suggested as a quicker but more careful covering o f eggs to prevent predation or dislodgment of eggs downstream.  The force o f the final  flexure is used to turn the body, facilitating a rapid return to the bottom o f the nest. It is suggested this is adapted to increase the speed with which the female can begin the next dig. Males also dig but do not appear to contribute to preparing the nest. Substrate was rarely lifted and was not always oriented to the nest. M a l e digging appeared to occur after intruders swam past the males' territory, after aggressive interactions, in response to aggression, in response to incomplete spawning acts, and often without any interactions. M c C a r t (1971) suggests this is adapted as an alternative to aggression and therefore fits the definition o f a displacement activity as defined by Tinbergen (1952). Between digging a nest and backfilling it, a female and male(s) must synchronize a gamete release for optimal fertilization o f eggs.  13  A s a spawning  female develops a redd, the male regularly quivers the female.  The male  positions himself downstream to the side o f the female and near her body. Quivering appears to the observer as a short duration, high frequency vibration of the males' body. This behaviour is reported to provide a stimulus to the female which provides feedback to the male indicating the females' reproductive state (Satou et al. 1991). As the nest becomes more developed, female digging slows and probing the nest occurs more often (Needham and Taft 1934; Sheridan 1960; M c C a r t 1971). This often stimulates the male(s) to approach the nest. The female w i l l often follow the probe with a dig. A s female egg release approaches, she appears to be probing in the bottom o f the nest but digging does not occur. One or more males rush in and jostle for position. If the female does not release eggs the males leave the nest. This behaviour may occur repeatedly until the female releases eggs.  The female gapes her mouth open and vibrates convulsively. Eggs are  expelled from the vent and within seconds, the male(s) expel milt from their vents.  This is visible to the observer as a white ' p u f f or plume drifting out o f  the nest. Sometimes when the female does not expel eggs, males may still expel milt, creating a false spawn. The repeated probing or ' c r o u c h i n g ' in the bottom of the nest immediately before egg release may ensure male attendance and synchrony o f egg and sperm release.  The female may use the stimulus o f males  rushing into the nest to assure fertilization w i l l occur when she expels eggs.  14  Males dominant on a redd territory with a female have the best advantage for fertilization o f eggs due to proximity (Chebanov et al. 1983). Other males may rush in and jostle for position near the female, which correlates with greater fertilization success for the male. Subdominant satellites, jacks, male kokanee, and other dominants have all been observed to participate in the rush to fertilize eggs.  The occurrence and magnitude of multiple male morphologies and  associated reproductive tactics varies between species, populations, and years. This suggests both genetic and environmental factors effect the success and frequency o f the various male tactics for achieving fertilizations o f eggs. After the gamete release, males immediately disperse in the stream (Greeley 1932; Burgner 1991). They may be exploring other fertilization opportunities. A dominant male salmon may hold to one side o f its territory and may not contest intrusions by other males for a period o f time (Keenleyside and Dupuis 1988). Dominant coho salmon males have been reported to leave and j o i n another female in the males' territory (Prince 1977). Females begin rapid nest backfilling typified by the frequent short duration digs which occur after gamete release.  Once the female has extruded and buried  her eggs the dominant male w i l l usually leave and seek other fertilization opportunities.  The female remains to guard the finished redd until death.  w i l l defend the redd territory against intrusions by females and all males, regardless o f dominance ( M c C a r t 1971).  15  She  Contests occur between spawning salmon to defend or take over redd territories.  Territories include substrate for nesting, an attending dominant male  for females, and primary access to a spawning female and opportunities for fertilizations in adjacent redds for dominant males. T w o categories o f contesting behaviours exist: displays, which involve no physical contact but may initiate flight or fight responses in other fish, and, physical aggression, which involve physical contact with another  fish.  Lateral displays are the raising o f the dorsal fin to another fish, or raising the dorsal fin combined with a paired lateral upstream swim between two males. I w i l l clarify these two behaviours for my study fish, and suggest two separate descriptions that accurately define the meaning o f ' l a t e r a l ' . Females may raise their dorsal fin, but have not been reported engaging in lateral upstream swims. Contest behaviours occur throughout spawning. M a l e s contest to gain and retain access to egg fertilization opportunities. Females contest to gain and retain access to spawning substrate, and continue after spawning to guard the completed redd with its fertilized eggs against disturbance by other fish. From the general review o f salmon behaviour in this chapter, one conclusion can be made. M a l e and female salmon behaviours are adaptations to optimize fertilization and development o f eggs, and thereby reproduce successfully (Stearns, 1976). A l l species partition energy into reproduction or growth depending on their reproductive strategy, life history stage, and ecological circumstances ( R o f f  16  1992). When sockeye salmon begin spawning, they have partitioned stored somatic energy into reproductive growth. Remaining somatic energy is not replenished due to the absence o f feeding. A l l reproductive behaviours including the completion o f gamete releases must be done with remaining somatic energy. Therefore, the amount o f energy utilized during specific spawning behaviours may effect spawning behaviours used.  17  Spawning Behaviour. A c t i v i t y , and Energetics in Fish. Fasting mature salmon have often been used for energetic  expenditure  studies because feeding (energy intake), the main confounding variable is absent. Somatic proteins and lipids are metabolized during upstream migration. The earliest known study o f fish bioenergetics was by chemical analysis o f the changes in muscle and gonads o f female Atlantic salmon reproductively maturing in the Rhine River (Miescher-Rusch 1880). This seminal study recognized the role of energy intake and allocation to maintenance metabolism, growth, and reproduction including the energy re-allocation from somatic tissue to gonads in fasting mature salmon. Since then, fish bioenergetics has been studied by various methods including: 1) 'direct calorimetry' in which heat production is measured in a calorimeter chamber, 2) 'indirect calorimetry' in which oxygen consumption is measured, and 3) 'body composition changes' measured by bomb calorimetry or chemical analysis o f serial samples of body tissues from a fasting population (Brett 1973). M a n y of the early fish energetics studies were done by R. Brett and primarily on salmonids. Brett (1965) used a tunnel respirometer to measure oxygen consumption of juvenile and sub-adult sockeye salmon at standard (resting) and active (maximum oxygen uptake) metabolism. With large fish, tail-beat frequency was correlated with oxygen consumption rates. M a x i m u m 60 minute sustained swim speeds were higher as fish size increased;  however, when the swim speeds o f all fish were  expressed in body lengths s" a decrease in relative swim speed was indicated for 1  18  larger fish. Despite an increase in body musculature and metabolic scope accompanying increased size, increased hydrodynamic drag is only partially compensated by increased size. M e t a b o l i c processes do not keep pace with the increasing energetic cost o f higher sustained swim speed for larger fish. In a later study, Brett (1973) examined the relationship o f weight, length, and temperature to swim speed and metabolic rate. Relative swim speed was reduced in larger and longer fish. Optimal sustainable swim speeds for all sizes of fish was attained at about 15°C. M e t a b o l i c rate increased with temperature up to 15°C, which correlated with the optimal sustained rate for swim speed.  Fish  weight had least effect on active metabolic rate at temperatures below 10°C. Fish weight had about equal effect on standard metabolism at all temperatures (520°C). Brett (1973) used a flume to exercise juvenile and maturing adult sockeye salmon for long periods of sustained swimming (10-20 days). A t the end o f the trial, the fish were sacrificed and frozen for proximate analysis and calorimetry. Similarly, fish were also exercised for the same period in a respirometer with identical flow rate. Oxygen consumption rates in relation to the performance by the fish in the respirometer was measured.  Water, protein, fat, and carbohydrate  constituents o f the flume exercised fish were determined.  Calorimetry from the  same sample population was determined either with a bomb calorimeter or calculated from proximate analysis using known caloric equivalencies for the constituents.  Subsequent analysis of data showed caloric loss from body  19  constituents exceeded estimates from oxygen uptake by an average o f 19.8%. E x c r e t i o n of partially metabolized constituents from the vent and gills is suggested as the source o f difference from the estimate.  These partially  metabolized products may be byproducts of anaerobic activity. Anaerobic activity could occur during the exercise regime or periodic bursts o f activity, such as aggressive interactions between fish. Idler and Clemens (1959) did one o f the first comprehensive studies on energy utilization by spawning salmon. Gilhousen (1980) did a detailed reevaluation o f Idler and Clemens (1959) data. Using serial analysis o f separate body constituents, they examined energy partitioning to gonads and l o c o m o t i o n during river migration and spawning in sockeye salmon. The sampling years were between 1956 to 1958 with some years not sampled for some stocks. They examined three upper river stocks and three lower river stocks from fresh water entry to their spawning grounds. The upper river stocks had correlated longer migration times, shorter schooling times on arrival at the spawning grounds, and greater energy expenditure and re-allocation to gonads during migration. L o w e r river stocks underwent maturation while schooling for weeks near their spawning grounds. Analysis was also done upon arrival at the spawning grounds and again at death for the early Stuart sockeye salmon in 1956, 1957, and 1958. D u r i n g the upstream migration and in the spawning stream somatic energy is depleted.  Gilhousen (1980) reported that fat provided more than 70% of the  migration and spawning energy in the upper Fraser River sockeye salmon.  20  M u s c l e protein was increasingly utilized later in the migration and spawning. R e allocation o f stored soma to gonads during migration was also recognized. A portion o f utilized protein was replaced with water, reducing loss o f mass and increasing relative gravity. M u s c l e protein was reduced and the trimmings gained weight during migration in males, apparently for secondary sexual development. In females, an equal amount o f muscle protein appeared to be re-allocated to ovaries which gained weight and developed during migration. Testes were nearly fully developed early in migration. Gilhousen (1980) also reported on energetics of spawning activity. The samples o f the early Stuart sockeye salmon on their spawning grounds provided analysis o f energy use by males and females for spawning ground activity. Samples were taken from fish entering the spawning stream and fish recently dead. Males are reported to have greater absolute and daily energy expenditure than females in both years. Early Stuart sockeye salmon females re-allocate more somatic fat to gonad development during migration than males (Gilhousen 1980). In addition, males arrived at the spawning stream with more somatic fat than females, and dead males had more somatic fat than dead females.  Jonsson et al. (1991) did a serial  constituent analysis o f migrating and spawning Atlantic salmon in south-western Norway. D u r i n g spawning the utilization of somatic fat was greater in males than females, similar to sockeye salmon. U n l i k e sockeye salmon, spent female Atlantic salmon retained more somatic fat than males. This suggests the iteroparous  21  Atlantic salmon females utilize less somatic fat and/or re-allocate less somatic fat into gonad development than the semelparous sockeye salmon females. Fishery scientists have long sought a means o f accurate assessment o f the amount and nature o f locomotor activity in free ranging fish. Such information would permit precise estimates of the metabolic costs o f daily activity and specific behaviours o f a fish. This would serve as a useful tool for the evaluation of numerous ecological problems, including feeding, migrating, and spawning (Rogers et al. 1984). Other methods besides body constituent analysis have been used to determine relative energy expenditure in fish. A c t i v i t y , an index o f energy use, has been measured using cameras, and remote sonic and radio telemetry. Cameras record physical activity such as tailbeats or pectoral fin sculling. Events can be visually measured by counts, duration, and degree o f flexion o f body or fins.  A c t i v i t y can then be correlated with calorimetric methods. F o r example,  tailbeats s" has been converted to known oxygen consumption rates for a range 1  of tailbeats s" (Brett 1973) and then to calorimetric equivalents. 1  V i d e o cameras have been used to estimate field energetics. K r o h n and B o i s c l a i r (1994) used a video method to record spontaneous activity in free swimming brook char, Salvelinus  fontinalis,  while simultaneously measuring  oxygen consumption. Metabolic costs of spontaneous activity were estimated by combining these measures.  Energy expenditure, measured by respirometry,  during spontaneous swimming was, on average, six times higher than forced  22  swimming at constant speed and direction for juveniles weighing 5 - 1 1  grams.  Fish may develop a more efficient swimming behaviour during forced swimming, and spontaneous swimming involves acceleration which may incur greater short duration metabolic costs. M e t a b o l i c costs o f turning and accelerating may be an important component of activity energy expenditure in large fish (Webb 1995). This has implications for the study o f spawning Pacific salmon, which turn and accelerate frequently during reproductive behaviour.  A c t i v i t y metabolism on  sockeye salmon spawning grounds was found to equal and in some cases exceed river migration by 14% for females and 59% for males (Gilhousen 1980). H i n c h and Collins (1991) used video observations o f free ranging male smallmouth bass, Micropterus  dolomieui,  defending nests during o v i p o s i t i o n and  until swim up larvae appear. T w o nests, each guarded by a male and located within 15 m o f each other, were continuously monitored using time lapse video and infra-red light for illumination. Guarding males were found to be active day and night, sculling pectoral and caudal fins constantly over the nest. Caudal fin beat rates were used to relate the costs o f guarding swimming to non-guarding swimming. U s i n g a respiration model for largemouth bass, M. salmoides,  they  estimated the energetic expenditure of nest guarding small mouth bass to be up to 60% higher than non-guarding. Predation o f offspring was l o w for the nest guarding male, even though the male left the nest briefly.  In an abandoned nest,  predators took about 1 day to discover the offspring and another day before they were completely consumed. The male that abandoned this nest may have been  23  under energetic constraints.  It did not feed during 11 days o f observation,  whereas the male that remained in the nest was observed to feed at a rate averaged 1 - 2% o f its wet body weight per day. Energetic constraints of 24 hour per day nest guarding, continuous in-place swimming, reduced food intake, and brood defense may greatly reduce reproductive success in smallmouth bass. Besides video observations, telemetry has also been used to estimate energetic costs in w i l d spawning fish. Remote sonic and radio telemetry requires a transmitter that measures biophysical and/or environmental parameters, then transmits them to a remote receiver. The receiver logs the data and can be downloaded into a computer.  Transmitters are diverse in their capabilities. F o r  example, accelerometers have been used to measure tailbeat frequency and amplitude, electrocardiogram ( E C G ) sensors to measure heart rate, and electromyogram ( E M G ) sensors to measure muscle contraction rate. The advent o f underwater biotelemetry since the 1960s has made it possible to track the movements o f fish by simple location transmitters (radio or ultrasonic) attached to the fish. However, using this information to evaluate the metabolic costs o f activity is difficult given the intricacies o f fish behaviour and environmental conditions (Priede and Y o u n g 1977; Weatherley 1976; Weatherley et al. 1982). Progress in other aspects of underwater biotelemetry has led to the monitoring of physical and physiological variables such as tailbeat frequency (Stasko and P i n c o c k 1976; Ross et al. 1981), swimming speed ( V o g e l i and Pincock 1980) and heart rate (Priede and Y o u n g 1977) in free ranging fish.  24  Attempts to deduce the metabolic costs of activity from data on these variables have also proved error prone (Rogers et al. 1981; Weatherly et al. 1982). The technique of ultrasonic tagging was first used successfully by Trefethen (1956), and Johnson (1960) to track the movements o f migrating salmon. Larger superficial transmitters were used by Poddubnyi et al. (1966) to observe the migratory movements o f sturgeon, Acipenser  baeri, in the River  V o l g a . Henderson et al. (1966) and Hasler et al. (1969, 1970) used small stomach transmitters to investigate detailed orientation movements in freshwater fish.  A similar technique was used by Yuen (1970) to track the movements of  skipjack tuna, Katsuwanus  pelanus.  In general the transmitters used in these  experiments were large, with the exception o f Hasler et al. (1970), so only large fish could be used. Tracking was done by following the fish in motor boats fitted with hydrophones and receivers ( Y o u n g et al. 1971). Y o u n g et al. (1972) developed an ultrasonic tracking system to monitor the activity levels o f small fish (> 200 grams) in their natural environment.  Activity  measurement units were designed to conform to units used in laboratory studies to approximate the metabolic cost o f locomotion in fish (Brett 1964), so that it would be possible to estimate the relative importance o f activity in free ranging fish.  Sonic transmitters were attached externally to the lateral surface o f the fish,  below the dorsal fin. Four brown trout, Salmo trutta, with transmitters were released into Airthrey L o c h , Scotland and tracked for 24 hour periods ( Y o u n g et al. 1971). Swim speeds, fish locations, and activity levels were determined for 24  25  hour periods. T w o fish were caught by anglers and were reported to have fought vigorously and one caught fish had insects in its stomach suggesting that fish had behaved normally after transmitter attachment and release. There have been attempts to use heart rates o f fish as indices o f swimming activity, using ultrasonic methods (Kanwisher et al. 1974; Wardle and Kanwisher 1974; Priede and Y o u n g 1977) and radio telemetry (Frank 1968; N o m u r a and Ibaraki 1969; N o m u r a et al. 1972; Weintraub and M a c k a y 1975). Priede and Y o u n g (1977) have correlated trout heart rates with oxygen consumption in laboratory forced swims; therefore heart rate might be usable in an index o f activity energetics.  B u t heart rate in fish responds readily to environmental  factors (Randall 1970), and because of this, and also because heart output is related to stroke volume, heart rate may not be a reliable index o f physical activity or oxygen consumption (Weatherley et al 1980). F o r a further review o f physiological telemetry studies, including heart rate telemetry, see Lucas et al. (1993). Oxygen demands o f muscular activity in fish at any given temperature are determined by biochemical processes at the tissue level (Weatherley et al. 1982). A t the level o f the whole muscle, strength, frequency, duration o f contraction, and the bulk o f the active muscle, w i l l combine to determine its oxygen consumption. In most fish, the axial muscles, which are the main swimming muscles, consist o f myomeres arranged in a bilaterally symmetrical series. Myomeres comprise most o f the total bulk of the body muscle and are involved in  26  most body movements.  F o r many species o f fish, the demand for oxygen  associated with the bodily muscular activity o f steady state swimming can be explored by having a fish swim against a water, current at constant velocity in a swim tunnel (Fry 1971). However, in addition to its respiratory requirements, a contracting muscle generates a characteristic bio-electrical signal, the E M G , the configuration o f which is related to the strength and duration o f the muscle contraction. In the case of the fish myomere, which is one of a linear series, the E M G o f any one myomere might be assumed to be representative of other myomeres.  The  configuration o f the E M G o f a representative myomere might be assumed to be highly correlated with oxygen consumption resulting from the activity o f the entire myomere series (Weatherley et al. 1982). Uematsu et al. (1979) described E M G recordings from spawning female chum salmon in a laboratory flume. B i p o l a r electrodes were implanted in both dorsal and ventral axial muscles. The electrodes were attached to an amplifier at the other end and displayed on an oscilloscope and photographed.  T w o spawning  events were recorded and showed both dorsal and ventral axial muscles contracted convulsively during oviposition. A l l muscles where the electrodes were implanted fired simultaneously and corresponded with the time o f egg discharge. Rogers et al. (1981) and Weatherley et al. (1982) described a laboratory based radio telemetry system for the 'remote' detection o f E M G s from the main  27  swimming muscles o f rainbow trout.  E M G s correlated well with activity and  oxygen consumption rates under routine and constant swimming conditions. The system previously developed for field use (Rogers et al. 1984) had major problems associated with it. Transmitter functioning was too short, malfunctions occurred, externally attaching transmitters often resulted in dislodgment, and additional drag was imposed by the transmitter.  A n additional problem was that  the E M G signal was difficult to store and process with computers into mean values o f u.V. These problems have recently been addressed (Kaseloo et al. 1991). Signal transmission o f up to 7 months has been obtained using a radio transmitter packaged in laboratory fish. Problems with drag have been eliminated with internal implantation. Changes in transmitter design and packaging has greatly reduced transmitter malfunctioning. Kaseloo et al. (1991) reported recent improvements o f E M G telemetry. The technique employed in the past used radio telemetry transmitters to transmit E M G s from fish in the laboratory (Weatherley et al. 1982) or an ultrasonic version to do the same for free ranging fish (Rogers et al. 1984). The E M G s were detected by sensing electrodes implanted in the fish axial muscles.  The  E M G values could be used directly as indicators of the intensity o f fish activity. A l s o , they could be calibrated to fish oxygen consumption measured over the same times in swims o f selected velocities and duration, or over periods o f spontaneous activity (Weatherley et al. 1982). The latter procedure gives, in  28  principle, the possibility o f obtaining quantitative estimates o f the metabolic costs of activity by free ranging fish. M c K i n l e y and Power (1992) studied the activity and oxygen consumption of free ranging adult lake sturgeon, Acipenser  fulvescens,  using E M G radio  telemetry. E M G s from the axial muscles were successful in long term monitoring of locomotor activity in the field, and subsequent estimates o f correlated energy utilization. Demers et al. (1996) used E M G radio telemetry to determine the activity patterns o f largemouth and smallmouth bass.  T w o largemouth bass and  two smallmouth bass were captured in Ranger L a k e , Ontario, and surgically implanted with E M G transmitters by the method of K a s e l o o et al. (1992). The bass were released back into the lake with a shore based Y a g i antenna to collect E M G signals for the whole lake. E M G s were collected for a period o f months and allowed monitoring o f daily activity levels for each fish.  The fish were most  active during the day, and tracking the movement o f each fish allowed matching of distance traveled to activity level. A l l fish survived beyond the study period. After recapture, the incision site was completely healed and sutures were absorbed. N o internal damage was noticed and the transmitters were encapsulated in connective tissue. Kaseloo et al. (1996) used E M G radio telemetry in a laboratory study of the activity o f spawning lake trout.  One male and one female were surgically  implanted with E M G transmitters and allowed to spawn in an artificial laboratory spawning flume. E M G values were successfully calibrated to spawning activity.  29  Weatherley et al. (1996) used the laboratory E M G values calibrated to activity values to monitor the spawning activity of free ranging lake trout. They were successful in determining one fish engaged in similar activity patterns as the laboratory fish, during the spawning period and at a known spawning location. One other fish was shown to not engage in spawning activity, and migrated to several areas o f the lake during the spawning period. N o n e o f these areas were specifically associated with spawning.  H i n c h et al. (1996) used E M G radio telemetry to assess the energy utilization o f Fraser River sockeye salmon through areas o f difficult passage. Their study used the same E M G transmitter and receiver technology used by Kaseloo et al. (1992) and Demers et al. (1996), and the same implant techniques, described by M c K i n l e y and Power (1992). Early Stuart sockeye salmon migrate a long distance upstream to their spawning grounds while fasting, thus likely imposing strong selection for energetic efficiency (Bernatchez and Dodson 1987). Rand and H i n c h (1998) developed an energy use model for migrating adult early Stuart sockeye salmon. They simulated total energy use o f individual fish during upstream migration using activity measures by E M G radio telemetry.  M o s t energy use models use  large time scale estimates, which average energy use over a period o f hours or days. However, fish migration involves highly variable activity at fine time scales. Large time scale estimates may not accurately predict energy use by using  30  mean swim speeds, and may underestimate the use o f anaerobic metabolism occurring during short periods o f burst swimming. Combining E M G estimates of swim speed with a laboratory model converting swim speeds to energy equivalents, they developed a fine time scale energy use model that accounted for energy use as a exponential function o f swim speed.  The fine time scale activity  measures used by Rand and H i n c h (1998) allows more precise and accurate estimates o f aerobic and anaerobic energy use and how it effects migratory success. M y first o f three objectives is to assess i f E M G telemetry can be used to discriminate levels o f activity between spawning ground behaviours.  Thereby,  relative differences in muscle activity could be used for studying behaviour on spawning grounds.  M y second objective is to compare activity estimates between  spawning ground behaviours as spawning progresses, and make inferences as to how muscle activity is used by spawning male and female sockeye salmon. M y third objective is to use the activity estimates to generate an estimate of total muscle activity in 'units' of total tail beats by male and female sockeye salmon on the spawning grounds, and use my relative comparison o f muscle activity as an 'index' to compare with the actual measures o f use o f body energy reported by previous researchers on the spawning grounds o f the early Stuart sockeye salmon.  31  METHODS  Study Site D e s c r i p t i o n The spawning streams of the early Stuart sockeye salmon are the subject of an ongoing fish-forestry interaction study (Steve M a c d o n a l d , Department o f Fisheries and Oceans, personal communication 1994). The watersheds around several o f the spawning streams are designated for logging by different techniques.  Thorough studies of these streams are being done both before and  after logging to analyze and interpret the impacts o f different logging techniques. To further the development of the fish-forestry interaction study I chose to use Gluskie Creek, one o f the early Stuart sockeye salmon spawning streams for my study o f muscle activity associated with spawning behaviour (Figure 1). I chose this stream for the following reasons: 1. Forestry data loggers record stream temperature 24 hours per day. 2. Easy access exists from a fisheries camp to a pathway along the stream. 3.  G r i z z l y bears, Ursus arctos,  utilize this stream less often than nearby streams,  reducing the chances o f conflict. 4.  A fish fence operated by the Canadian Department o f Fisheries and Oceans in  the lower reaches confines the movement o f study fish to within the stream, provides samples, and known dates of stream entry. 5. The total fish return for Gluskie Creek is known since 1975.  32  Figure 1: M a p of the Fraser River showing the location o f early Stuart sockeye salmon spawning grounds and their migration route up the Fraser, Nechako, and Stuart River systems. The Gluskie Creek spawning grounds are indicated on the upper left-hand corner. 33  Transmitter Implantation The E M G radio transmitters used in this study are cylindrical, 5 cm long, 1.5 cm diameter, and weigh 20 grams in air. They have two steel wire electrodes, 20 cm long, fitted with 18 carat gold tips 1cm long and 1 mm diameter. A wire antenna, 40 cm long, trails from the radio transmitter's posterior end along with the wire electrodes. Each transmitter has a unique frequency which the radio receiver can selectively detect. E M G radio telemetry requires the placement o f the gold tipped electrodes in the red axial swimming muscle to detect voltage changes associated with the electrical activity of muscle contraction. B o t h dorsal and ventral axial swimming muscles are used during muscle contractions and all muscles contract simultaneously during spawning (Uematsu et al., 1979). The transmitter's internal electronics detect all voltages over 1-2 u V in amplitude. V o l t a g e signals are multiplied 2 7 , 0 0 0 X , rectified and integrated. Whenever two successive integrated pulses exceeds the predetermined threshold (150 u,V) an analogue signal is transmitted to the radio receiver (Lotek model S R X - 4 0 0 ) . integrator resets and the process is repeated.  The  B y this means, the time interval  between E M G pulses is correlated with the frequency o f muscle contractions (Hinch et al, 1997). E M G signals are "strongly" negatively correlated with strength and duration of muscle contractions (Sullivan et a l . , 1963). L o c a t o r radio transmitters also have unique frequencies and are used for tracking and monitoring fish location, and have no electrodes or E M G capability.  34  These radio transmitters are 4 cm long, 1.5 cm diameter, weigh 15 grams in the air, and have an antenna 3 5 cm long. They send a radio pulse at regular intervals which is strongest when the receiving antenna is pointed in the direction o f the transmitter.  Fish carrying either locator or E M G transmitters can be located to  within 1 meter o f their true location. Tracking requires w a l k i n g in the direction of the strongest pulse until the fish is found and identified visually. A l l fish to be implanted with E M G transmitters were anesthetized using C O 2 generated by a mixture o f sodium bicarbonate and glacial acetic acid (method described in Prince et al., 1995). After anesthesia, fish were removed from the anesthetic bath and measured for length (nose-fork), weight, and scale samples were taken. Once the surgery was complete, fish were tagged on their dorsal surface with numbered spaghetti tags for visual identification externally and then released. In 1994, fish were caught on their redds for the surgery and released back on their redds. In 1995, all fish were caught as they arrived at the fish fence. After surgery, fish were released above the fish fence. The E M G transmitters were surgically implanted by two methods; abdominal implantation (Hinch et a l . , 1996; Lucas et a l . , 1993; M c K i n l e y and Power, 1992), or subcutaneous implantation which I developed during this study. Abdominal implantation followed methods described in H i n c h et al. 1996. Subcutaneous implantation differs from abdominal implantation in that a horizontal scalpel incision is made just through the skin, about 3 cm long and 5 cm anterior to the pelvic fins. A cannula or other narrow blunt instrument was  35  used to separate the skin from muscle, opening a ' p o c k e t ' 5 cm long anterior to the scalpel incision.  The transmitter is pushed through the incision and forced  anterior until the incision can be closed. The electrodes are implanted by the same method as for abdominal implantation, but the abdominal cavity was not punctured.  Excess electrode wire was tucked under the skin with the transmitter.  The antenna is positioned to emerge from the incision in the ventral and posterior direction and does not require cannulation. The pocket was sutured closed with #1 sterile sutures leaving only the antenna wire trailing out.  To prevent the  transmitter from slipping out i f skin sutures fail, the transmitter electrodes may be sutured to the ventral muscle and around the electrodes emerging from the transmitter. Fish were captured for locator radio transmitter implantation as they arrived at the fish fence. N o anesthetic was necessary because this is a quick, relatively non-invasive procedure during which I could hold the fish by one hand in my lap. E a c h fish was measured for fork length and weight. L o c a t o r radio transmitters were implanted down the fish's esophagus into the stomach. Transmitters were first placed at the back o f the fish's mouth with the antenna end positioned anterior so the antenna trailed out the mouth. Transmitters were pushed down into the stomach, using the blunt end o f a pencil, until the transmitter could no longer be pulled out with a gentle tug on the antenna. antenna remains trailing out the side o f the mouth.  36  The  Behaviour Observations Behavioural observations were made on fish implanted with E M G transmitters, locator transmitters, and fish without transmitters.  Once a fish with  a transmitter was located for behaviour observations, all fish immediately adjacent to it without transmitters were selected for behaviour observation as well. Each specific behaviour observed is a single observation. A n observation period is the total time a fish was observed uninterrupted (i.e. 10 minutes).  The  maximum upstream movements for each fish with a transmitter were used to compare differences in stream migration between sexes and years. In 1994 and 1995 sockeye salmon were implanted with E M G radio transmitters, whereas in 1996 locator transmitters were used. In 1994 and 1995, when a fish was located, E M G data were collected with the radio receiver while each behaviour and time of each behaviour was verbally recorded on audio cassette for a 10 minute observation period. This duration o f observation was chosen because it allowed all tagged fish and additional fish to be observed within a day, and was the longest duration determined to be repeatable throughout the day without exhausting the observer.  In 1995 and 1996, when a  fish with a locator or E M G transmitter was located, it was observed for a 10 minute period and all the fish adjacent to it without transmitters were observed for 10 minute periods as well. Behaviours were identified visually and recorded verbally on audio cassette, providing additional records o f frequency o f each  37  behaviour in the 10 minute period. Behaviours were timed with a stop-watch in separate observation sets of behaviour duration.  38  The following behaviour types were observed and recorded:  attack, bite,  chase, dig, dorsal display, lateral display, quivering, schooling, false spawn, and true spawn. These behaviours were noted during preliminary observations of spawning early Stuart sockeye salmon and are described in Table 1.  Table 1. Behaviours observed and recorded during spawning.  1. Attack (charge): fish burst swimming towards and making physical contact with the side o f another  fish.  2. Bites: fish grasping and holding the body or fin o f another fish, or repeated ' b i t i n g ' motions on the side of another  fish.  3. Chase: one fish pursuing another fish at variable ranges o f speed.  The fish  rarely makes contact and the subject of the pursuit may accelerate away. 4. Digs: the fish turns on its side and uses undulations o f its tail to disturb the substrate used in redd building by females, and by dominant males as a display in their redd territory. 5. D o r s a l displays: a raising, or fanning out, o f the dorsal fin in males. This may occur alone or in combination with other aggressive behaviour, including lateral displays.  39  Table 1. (Continued) 6. Lateral displays: two male fish swimming side by side with their bodies tilted head up (posturing) at an angle with respect to the water  flow.  Often  accompanied by dorsal displays. 7. Quivering: high frequency, low amplitude lateral undulations o f a male courting a female. 8. Schooling: newly arrived and relatively inactive fish, often observed ' h o l d i n g ' in pools with other newly arrived fish. 9. False spawns: the female crouches in the redd as i f to deposit eggs, often causing several males to rush into the redd, but no eggs or sperm releases. This typically precedes a true spawn. 10. True spawn: the female crouching in the redd and releasing eggs with high frequency, l o w amplitude tail beats. Several males may rush into the redd and release sperm in the same manner.  40  Data Organization and Analysis E M G pulse interval values which were stored in the data logger, were matched with behaviour observations by associating the recorded time o f an E M G pulse interval to the behaviour start and stop time recorded on audio cassette. The clock in the receiver was checked multiple times per day and calibrated to the wrist watch used to observe the time of the behaviour. The E M G pulse interval may occur at any time during the behaviour, therefore the E M G pulse interval records should represent a random sampling o f the range o f E M G pulse interval values during the behaviour, from initiation o f the behaviour to its end. Although subjectivity is involved in associating E M G pulse interval values with behaviour time, these values can be considered random samples of E M G pulse interval values associated with that specific behaviour. E a c h fish with an E M G transmitter is considered as a single replicate. The same fish when sampled on a different day represents an additional replicate because fish behaviour changes throughout the spawning period. F o r each fish, the E M G pulse interval values, which were obtained about every 2 - 5 seconds, were converted to an estimate o f T B s" using an equation 1  developed by H i n c h and Rand (1998), who empirically assessed T B s" that were 1  associated with a range o f E M G values in swimming sockeye salmon. E M G values represent the duration between muscle contractions. Thus, the duration between muscle contractions decreases as the fish increases the rate o f muscle contraction, therefore the E M G values are inversely related to muscle contraction  41  and the tail beat index. T B s" is an index o f activity which may be positively 1  correlated with energy expenditure (Weatherley et a l . , 1996), so E M G values were converted to T B s" to show energy expenditure values in units that increase 1  with energy use.  A s with E M G values, the T B s" values were generated on a  time scale o f about 2 - 5  1  seconds.  Behaviour frequency for all fish was summarized by totaling the number o f behaviours occurring each observation period and expressing it as a specific behaviour per minute. This was further summarized by determining means and 95% confidence intervals for T T B , frequency of behaviours, and duration o f behaviours using individual fish as the replicates. A skewness test was done to evaluate the extent o f skewness. were L o g  1 0  If data were skewed greater than 2.000, they  transformed before conducting statistical analysis. A N O V A s were  used to assess differences among behaviour types in rate o f T B s" , frequency, 1  and duration. M e a n T T B required to complete a specific behaviour was determined by multiplying mean T B s" with the mean behaviour duration in 1  seconds.  M e a n T T B estimated during 10 minute observation periods was  determined by multiplying the mean T T B for each specific behaviour by the specific behaviour frequency per minute.  The muscle activity expressed in T T B  for each behaviour type was determined by determining the mean T T B expended in minutes among replicates for each category and multiplying by the mean longevity of spawning stage, sex, and year.  42  A n c i l l a r y observation was done to determine general spawning behaviour by sockeye salmon in Gluskie Creek, and to determine classifications o f fish activity in which to categorize data for more accurate analysis and estimates. The following is an overview of the preliminary observations:  Females were  fewer than males and would always establish a redd and engage in spawning activity. They would most often hold in a pool with hundreds o f other fish during the schooling stage which was used as a status designation. N e x t they would establish a redd territory and deposit eggs in nests in the stream substrate during the spawning stage. Females with remaining energy would guard their redd until death.  M a l e s would often hold in a school and this was used as a status  designation for males also. L i t t l e or no aggression was observed during the schooling stage among fish o f similar or different sex. M a l e s that subsequently established and defended a redd territory during the spawning stage were classified as dominant.  Smaller or senescing males that could not establish and  defend a redd territory were classified as subdominant. little or no access to females.  These males typically had  Hence, I classified female status into schooling  female, nesting female, and guarding female. I classified male status into schooling male, dominant male, and subdominant male. These behaviour types were used as the class variables for the A N O V A s . of fish used in this study is presented in Table 2.  43  A summary o f the description  Table 2. Summary o f fish w i t h transmitters used i n this study  YEAR 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996  SEX F F F F M M M M M M F F F F F F M M M M M M M M M F F F F F F F F M M M M M M M M M M M M M M M M  WEIGHT(ka) 2 85 2 45 2 30 1 90 4 10 3 35 3 20 2 65 2 50 1 15 3 50 2 30 2 25 2 25 2 15 2 15 4 15 3 75 2 80 2 75 2 65 2 60 2 10 2 00 1 45 3 15 2 60 2 60 2 40 2 35 2 25 2 15 1 95 4 15 4 15 3 80 3 75 3 50 3 50 3 35 3 05 3 00 3 00 2 90 2 85 2 65 2 45 2 40 2 35  LENGTH(cm) 61 5 59 2 63 1 56 8 69 7 64 5 63 5 58 8 58 46 6 68 1 61 7 58 5 58 4 59 1 57 8 70 6 66 7 60 3 60 5 62 5 60 1 59 6 58 4 53 5 63 5 61 5 60 60 60 5 58 2 58 56 5 69 71 5 68 5 68 5 66 67 66 5 64 61 5 62 62 61 63 5 60 60 5 60 5  44  METHOD EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG EMG LOCATOR EMG EMG EMG EMG LOCATOR EMG LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR LOCATOR  FREQUENCY 150 405 150 267 150 386 150 267 150 365 151 0 2 6 150 365 150 963 150 023 150 885 150 4 2 4 151 0 6 4 150 862 150 682 150 345 150 105 150 245 151 1 2 4 b 151 3 2 9 151 1 2 4 a 150 146 150 183 150 325 151 7 4 0 150 364 151 6 8 0 151 4 8 0 151 2 3 0 151 2 7 0 151 2 6 0 149 934 151 4 2 9 151 2 0 0 151 5 4 0 151 3 8 0 151 4 1 0 151 6 5 9 149 876 151 3 1 0 151 3 3 8 151 4 4 0 151 7 4 0 151 4 4 7 149 8 4 7 151 2 3 9 151 . 2 8 0 151 . 2 5 0 151 . 3 0 0 151 . 2 8 9  RESULTS  Some behaviours chosen for observation were eliminated due to small sample sizes (i.e. false spawns, gamete release), but mean T B s" was reported. 1  One false spawning sequence followed by a gamete release, and then a chase on an egg predating rainbow trout was observed and matched with corresponding T B s" values. Because the sample size was small for these types o f behaviours, I 1  classified them as "general activity". The sample size (n) and mean T B s" for the 1  false spawn, gamete release, and chasing the trout, were n = 10, mean = 3.3 T B s" n = 3, mean = 5.5 T B s" ; and n = 5, mean = 2.3 T B s" respectively. Gamete 1  1  release by this female had the highest mean T B s" of all behaviours. 1  Some behaviours observed were not included in the analysis. Bites were not often observed. What appeared as bites were a result o f l o c k i n g the jaws with the body o f the recipient fish during an attack.  N o repeated biting motions  occurred and the fish appeared to have difficulty separating because of the hooked kype. I observed biting in another Fraser River sockeye salmon population in the lower river. The biting motions were easily discernible and included repeated biting motions on the side o f the recipient fish. This behaviour appeared to be biting and added certainty that I did not observe biting behaviour in the early Stuart River population. C i r c l i n g is not a specific behaviour and can not be attributed to contesting or synchronizing gamete release.  C i r c l i n g may be  a single full turn or a half turn followed by a full turn. A s circling behaviour  45  appeared to be a non-specific behaviour, it was left out o f the specific behaviour analysis and included as general activity. Baseline T B s" were obtained from surgery records o f 5 fish under stage 4 1  anesthesia. tensed.  The fish were completely immobile and muscle fibers appeared non-  T B s" data was pooled because there were only 5 data sets. Out o f the 5 1  data sets, 4 were not significantly different. was not seen as an outlying data set. variant o f baseline data.  The fifth was examined visually and  Therefore I accepted it as a ' n o r m a l '  The T B s" was 0.694 with a 95% confidence interval o f 1  0.002 T B s" . 1  Sample sizes o f other behaviour types were quite large therefore, these other behaviours were not "pooled". Estimated number o f T B s" , frequency and 1  duration o f behaviour data sets were all positively skewed greater than 2.0 and were therefore logio transformed before doing A N O V A s .  In 1994 and 1995, I  collected 10,921 E M G pulse interval records converted to T B s" that were 1  matched with behaviour observations. Skewness o f T B s" data was 3.979 for all 1  behaviours combined. Skewness after log transformation was 1.591. Behaviours tested with one-way A N O V A s , showed significant effects o f behaviour on T B s"  1  (DF = 20, E r r o r D F = 10900, F = 418, P < 0.001). M e a n T B s" were lowest 1  during the schooling stage and increased when the fish left the school and began spawning (Figure 2). In males, quivers had the lowest mean T B s" and lateral 1  displays the largest. In females, nest digging had the largest mean T B s" . 1  46  Attacks had a greater mean T B s" with nesting females than fish o f other status. 1  N o t all behaviours were observed with females. Attempts to use the E M G receiver clock to estimate average behaviour duration was abandoned.  M o s t o f the behaviours occurred in such short duration  (e.g. less than 1 - 2 seconds) as to make estimates with the receiver data very uncertain. U s i n g a stopwatch provided accurate duration with 1/100 second precision. These data were collected in 1996 and applied to all T B s" data 1  (Figure 3). A p p l y i n g this method assumes the mean duration did not vary significantly between the years for behaviour. A s it applies only to specific behaviours, it may contribute little uncertainty to the estimate o f spawning ground T T B by sex and year. F o r some behaviour duration the sample sizes were small because the frequency of the behaviour was small, so the data were pooled by sex. Measurements o f behaviour duration provided n = 426 data records. Sample sizes for subdominant males and guarding females were small, but within the range o f data collected for dominant males and nesting females.  These  samples were combined and analyzed by sex, not by status. Skewness was 6.470 for all behaviour duration combined. Skewness after log transformation was . 2.606. A N O V A s revealed significant differences in duration o f behaviour ( D F = 8, E r r o r D F = 417, F = 55, P < 0.001). Dorsal displays and lateral displays had the longest mean duration (Figure 3). These were behaviours most often observed among dominant males o f similar size competing for territory and a female.  47  M u l t i p l y i n g the mean T B s" by duration generated an index o f mean 1  number o f T T B for each behaviour and by status (Figure 4). This describes the mean T T B used for the specific behaviours and allows comparison among specific behaviours using T T B as an index o f muscle activity. Observed frequency o f behaviours in 10 minute periods for male and female fish during different status provided n = 1952 frequency records for the 6 behaviours observed.  Sample sizes provided data for all behaviours during  different status. Skewness was 3.592 for all behaviour frequencies combined. Skewness after log transformation was 1.854. A one way A N O V A revealed significant effects o f fish status on behaviour frequency ( D F = 5, E r r o r D F = 1946, F = 79, P < 0.001). D i g g i n g was the most frequent behaviour within females, whereas attacks were the most frequent behaviour within males (Figure 5). D i g g i n g among subdominant males and guarding females, and quivering by subdominant males were the least frequent behaviours. In order to evaluate the T T B for each behaviour and status on an hourly time scale, I multiplied T T B (Figure 4) by the corresponding 10 minute behaviour frequency (Figure 5) and then by 6. In Figure 4, T T B values for behaviours o f dominant males and nesting females were used where T T B values for subdominant males and guarding females were absent. Because this occurred when frequency of the subdominant male and guarding female behaviours was l o w , it probably contributed little to the behaviour T T B hr" (Figure 6). Lateral displays among 1  dominant and subdominant males had the highest mean T T B hr" . This behaviour 1  48  occurred less frequently than many o f the other behaviours, but it had a long duration (Figure 3) which contributes to the relatively high value in Figure 6. I pooled the T T B hr" o f behaviour-specific activity and graphically 1  (visually) compared this to T B hr" of general activity (Figure 7). The T T B of all 1  specific behaviours among fish of each status contributed much less T T B than general activity during spawning. This was compared between fish o f four different status (dominant male, subdominant male, nesting female, and guarding female).  V i s u a l comparison of confidence intervals showed differences among the  T T B hr" (specific and general activity pooled) between fish o f different status 1  (Figure 7). M e a n T T B hr" expended during baseline activity, schooling activity, and 1  all spawning activity were compared (Figure 8). Baseline T T B are a large portion o f muscle activity in schooling and spawning fish. One male fish was observed and E M G s recorded during an upstream migration. It traversed pools, glides, and riffles, with riffles most common. Mean muscle activity was 2.13 T B s" . 1  The migration distance was 326 m.  requiring 18.21 minutes o f upstream migration; total time for migration was 180 minutes, but this included holding and contesting with other males, and was excluded from the migration time and T T B . Mean swim speed was 111 cm. s"  1  and ground speed was 30 cm. s" . 1  The difference is 81 cm. s" , attributable in 1  part to water velocity. When traversing shallow riffles, the lateral swim muscles were partly out o f the water and ventral contact with the substrate was observed.  49  Spawning ground duration provided n = 36 records o f total duration. A two way A N O V A was done to test the effect of year and sex on spawning ground duration. Year had no effect on spawning ground duration ( D F = 1, E r r o r D F = 32, F = 0.440, P < 0.512). 1, F = 13, P < 0.001). = 19, P < 0.001).  Sex had an effect on spawning ground duration ( D F =  A n interaction occurred between year and sex ( D F = 1, F  In 1995, male duration was lower and female duration higher  than 1996. In 1996, no difference occurred between male and female duration. This may contribute to the significance o f the interaction term and the effect o f sex but not year on spawning ground duration (Figure 9). T T B were summarized by each source o f T T B , and by sex and year (Figure 10). Longevity for each status was multiplied by the total equivalent tailbeats hr" 1  for each status. The constant baseline T B s" was first subtracted from each 1  general activity thereby demonstrating how each activity contributes an increase to the T T B . Baseline T T B were the largest component. M e a n and range o f migration upstream of the fence was summarized by sex and year (Figure 11). The data show that the distribution o f fish occurred within the lower reaches o f the stream which were observed to be high density spawning areas. The position and movement throughout the spawning ground duration o f each fish showed a pattern o f initial migration followed by association with a specific stream section. Night observations were difficult because fish appeared to alter their behaviour when exposed to artificial light. Some appeared attracted towards the  50  light, while others moved away. However, they appeared similarly active at night as during day and were observed to engage in nest digging and contests.  More  subtle behaviours such as quivering were not observable. A d d i t i o n a l E M G data was collected by allowing the radio receiver to run during various 24 hour periods to compare day and night activity levels. One fish was observed during one 24 hour period and another fish was observed during 5 separate 24 hour periods. M e a n day and night activity levels were summarized into 6 hour periods; 2 during day and 2 during night.  Some periods were missed  when the receiver battery charge was reduced and the receiver shut off. Skewness o f data was determined for each 24 hour data set, and the upper range was 3.338, so all data was log transformed.  Skewness after log transformation  was 1.439. One way A N O V A s revealed significant effects o f time periods on T B s" obtained within the same 24 hour period (P < 0.001). 1  Sample sizes were  large, preventing analysis o f all 6 days o f data as a single set.  M e a n activity  levels were compared visually between 24 hour periods and no differences were attributed to diel periodicity (Figure 12). Mean day activity levels and behaviour observations were used as an estimate o f mean night activity levels and behaviours. F o r a senescing male (day 1) and a spawning female (day 2 - 5 ) differences occurred between 6 hour time periods but did not reveal diel periodicity. On day 6, there was no significant difference ( D F = 1, E r r o r D F = 6663, F = 0.411, P < 0.521).  On this day the female had finished nesting and was  beginning redd guarding, thereby changing from nesting female status to guarding  51  female status. Significant differences occurred within the 24 hour periods in all other days, including the male in day 1. V i s u a l examination o f the graph (Figure 12) suggests that fluctuation in T B s" occurred independently o f diel periods. 1  N o conclusion can be drawn that night activity levels are different than day activity levels.  52  Figure 2: Mean level of activity measured as T B s " for specific behaviours compared by fish status. Clear bars, striped bars, spotted bars, and black bars represent dominant males, subdominant males, nesting females, and guarding females respectively. E r r o r bars represent 9 5 % confidence intervals. Baseline and schooling T B s" are included in this figure for comparison to specific behaviours and represent males and females respectively. Baseline and Schooling error bars are too small to be seen on this graph. This graph demonstrates the difference between means and 9 5 % confidence intervals between schooling and spawning fish, and between general "activity" and specific behaviours. 1  1  53  18 j 16  2  _  14  -  12--  i jo  10 -  £  6 4  8  U  f  -  1  •a  ,1  T  R a ,i i  [=1  B  ,1  — ' — —  1  1  i—i  1  Behaviour  Figure 3: D u r a t i o n o f specific behaviours for males and females. Data for different status within sex were pooled because sample sizes for subdominant males and guarding females were small. Clear bars and spotted bars represent males and females respectively. E r r o r bars represent 95% confidence intervals. Dorsal display, lateral display, and quiver are male displays only.  54  Behaviour  Figure 4: Total tail beats ( T T B ) for each behaviour by fish status. Clear bars, striped bars, spotted bars, and black bars represent dominant males, subdominant males, nesting females, and guarding females respectively. Confidence intervals represent 9 5 % confidence intervals. Where samples were very small (i.e. 0-1) for subdominant males or guarding females, data from dominant males and nesting females were used as an estimate.  55  Figure 5: M e a n frequency o f behaviour occurrence in 10 minute intervals by fish status. Clear bars, striped bars, spotted bars, and black bars represent dominant males, subdominant males, nesting females, and guarding females respectively. E r r o r bars represent 95% confidence intervals.  56  Behaviour  Figure 6: T T B hr" for each behaviour by fish status. Clear bars, striped bars, spotted bars, and black bars represent dominant males, subdominant males, nesting females, and guarding females respectively. E r r o r bars represent 95% confidence intervals. 1  57  6000 5000 ra ra t-  ra o  4000 3000 2000 1000 0  Status  Figure 7: M e a n T T B in one hour for spawning fish. Specific behaviours are pooled. B l a c k bars and clear bars represent specific behaviours and general activity respectively. The error bars represent 95% confidence intervals.  58  5 6000  |  4000-  | £  3000 2000  •  5000  • m  j •  •  •  •  ? :tLB_ll LI LB ioo  •  •  /  Stage  Figure 8: M e a n T T B for baseline, schooling, and spawning T T B hr" Clear bars, spotted bars, and black bars represent baseline, schooling, and spawning T T B respectively. E r r o r bars represent 95% confidence intervals. Baseline T T B are not deducted from the total schooling and spawning T T B .  59  Figure 9: M e a n spawning ground duration by sex and year. V e r t i c a l bars represent the mean schooling status in both sexes. With males, slanted bars and spotted bars represent the mean time spent as dominant and subdominant status respectively. W i t h females, slanted bars and spotted bars represent the mean time spent as nesting and guarding status respectively. E r r o r bars represent 95% confidence intervals. Females go through schooling, nesting, and guarding in sequential order. Males typically go through schooling, dominance, and subdominance in sequential order.  60  Figure 10: Mean T T B by sex and year. Clear bars, vertical striped bars, spotted bars and black bars represent T T B contribution by baseline, schooling, general activity, and specific behaviours respectively. In stream migration behaviour is included in the graph but is not visible because o f its relatively small contribution. Baseline T T B were first deducted from schooling, in stream migration, general activity, and specific behaviours before creating this graph.  61  2000  j  1800 - -  c E  9J O  E E «s  1600 • • 1400 - 1200 + 1000 - • 800 - • 600 - -  <D  400 - •  W  200  i_ *-»  a.  ID  4-  +  0  4? Year and sex  Figure 11: Upstream migration from release point at fish fence by sex an year, showing mean and range.  62  DAY 1  DAY 2  DAY 4  DAY 3  DAY 5  DAY 6  Sam p i e p e r i o d s  Figure 12: Mean activity levels in 6-hour periods during 6 days o f remote monitoring o f 2 fish. Nocturnal periods were from 00:00 to 06:00 and 18:00 to 24:00 (slanted and horizontal bars respectively). Daylight periods were from 06:00 to 12:00 and 12:00 to 18:00 (spotted bars and clear bars respectively). Day one represents a dominant male undergoing senescence. Days 2 through 6 represent one nesting female. Redd guarding started on Day 6. Some periods were not recorded i f the receiver was used for other purposes or i f the receiver malfunctioned.  63  DISCUSSION  M y first objective was to determine i f E M G technology could be used to discriminate behaviour-specific activity levels in free ranging sockeye salmon on their spawning grounds.  Significant differences were found between the mean  activity rate o f specific behaviours using E M G .  This result demonstrates that  E M G telemetry has the precision to discriminate among activity levels at a specific behaviour level. Duration and frequency data, when combined with E M G data, revealed differences in T T B for specific behaviours. E M G technology seems capable o f assessing T B s" and T T B among specific and non-specific 1  behaviours in free ranging sockeye salmon. M y second objective was to compare behaviour-specific activity and general activity among males and females of different status. Behaviour-specific activity was demonstrated to contribute fewer T T B than general activity for fish of all status and sex. Behaviour-specific analysis showed females expend the most T T B in digging nests and is their most frequent behaviour also. The effort required to develop a nest supports that energetically, digging may be the most important specific behaviour for reproductive success in females.  The 3  behaviours observed in females were for nest building (digging) and redd defense (attacks and chases). Territorial males engage in attacks.  Attacks have lower T B s" and 1  duration, but this behaviour has the most frequent occurrence in territorial males.  64  Attacks by males may not only be for contesting dominance but maintains territorial boundaries and protects access to females. L a t e r a l displays occurred with less frequency but had the highest T B s" and duration o f all behaviours, 1  making this a very intensively active behaviour. It occurs typically between males of similar size. Males o f similar size may have to contest more vigorously to establish dominance. Quivering, which is reported by Satou et al. (1991) to provide the male feedback on the females' reproductive state, used the least T T B of all male behaviours. In 5 out of the 6 specific behaviours, males expended the most T T B for territory defense and dominance contests. There was an effect of status on total mean T B s" . 1  Dominant males are  not significantly different from guarding females, and nesting females have the highest mean specific behaviour T B s" . 1  Females exhibited 3 out o f the 6  behaviours which have a higher mean T T B than the full 6 behaviours used by males. B o t h females and dominant males dig. Females have a higher T B s" for 1  digging than males. The higher mean T T B rate o f females may be to dislodge substrate in nest building, which male digging does not do. T B s" and T T B for specific spawning behaviours were shown to be 1  significantly different for some specific behaviours. T T B expended during specific behaviours demonstrated the important ecological interactions o f the fish with its environment from an activity perspective. Behaviour-specific activities may appear to be a large, or largest, source of muscle activity. Comparing hourly T T B from the specific activity to general activity demonstrates that specific  65  activity contributes relatively little to the T T B .  W i t h the exception o f baseline  metabolism, the majority o f T T B used by the fish were in general activity. In 1995, females had a longer duration on the spawning grounds than males. In 1996, females and males had similar duration on the spawning grounds. The difference in longevity for 1995 males and females may be an effect of total body energy or possibly the males may have engaged in some activity at the mouth of the creek before passing through the fence. Females passing through the fence had no redd territory. A l l males used for transmitter implants appeared to have arrived 'fresh' with the females. The schooling stage for 1995 males was similar to 1996 males which have significantly greater longevity than 1995 males. The occurrence o f schooling behaviour in the 1995 males suggests strongly that they were not already spawning, but freshly arrived. It was noted that schooling was typical o f newly arrived fish, but not for spawning males moving about the spawning stream. A l l females had their abdomens palpated and assessed for looseness of the eggs from the skeins. In all but one female the eggs appeared partially or entirely in the skeins. The schooling period was longest for females, which may indicate delayed release o f eggs from the skeins until arrival at the spawning grounds. The schooling period may allow the eggs to release from the skeins into the abdomen while expending the least energy via muscle activity before the female proceeds onto the riffles to establish a redd territory. Schooling uses fewer T T B than spawning, suggesting that when schooling occurs, reduction in T T B expenditure may be adaptive.  66  Females arriving with eggs still retained in the skeins may school to save energy while the eggs mature and release from the skeins. One female that had eggs already released from the skeins upon arrival at the fish fence proceeded immediately to establish a redd territory, bypassing schooling. This female was the shortest lived and just finished nesting, without time left for redd guarding, before death.  The observation o f this small female  arriving with eggs already released in the abdomen, suggests that females arriving late with little energy left may begin to release eggs from their skeins before they arrive. This would allow them to avoid expending available energy by not schooling. Partial spawns or pre-spawn mortalities may result i f available energy is insufficient to complete spawning. Delaying the release o f eggs from skeins until reaching the spawing grounds may be adaptive only i f available energy stores allow extra time schooling. The skeins may protect the eggs from infection, water hardening, and loss out the vent during a burst swim.  The  retention of the eggs in the skeins may also effect the centre o f gravity, and subsequently the swimming efficiency o f the fish. Baseline activity was determined from fish in stage 4 anesthesia.  This may  approximate basal metabolism for lateral swim muscles. This assumes that baseline T B s" are a measure o f involuntary random contraction o f muscle fibres. 1  Other metabolic processes in these anesthetized non-feeding fish may contribute little energy use.  The term 'baseline' metabolism rather than basal metabolism is  used to avoid making untested assumptions about the effects o f C 0  67  2  anesthetic on  metabolic processes of fish. Baseline activity contributed much o f the T T B . I f baseline T T B are subtracted from schooling and spawning activity, the contribution o f schooling T T B and spawning T T B to total muscle activity is shown to be far less than the absolute measurement during these activities. This suggests from an energetic perspective that remaining inactive for part o f the day is not a good spawning strategy unless the fish is not yet ready to spawn. T w o fish were observed for patterns of diel fluctuation in T T B expenditure.  The females T T B use may fluctuate more in association with cycles  of egg releases, and subsequent rapid nest backfilling.  H o w e v e r , data are  insufficient to determine the cause of any fluctuation. The lack o f a detectable pattern may indicate T T B expenditure is similar during day and night. Spawning, specifically the more obvious behaviours, were observable at night. B u t less obvious behaviour was not easily observed, so night observation was not able to be done accurately. Due to the similarity o f day and night T T B expenditure, I chose the assumption that diel activity was the same regardless o f light levels. Coho salmon spawn at low densities, and most often at night thereby reducing predation (Prince 1997).  Sockeye salmon spawn in high densities typical of a  'flocking strategy' (Roff 1992) and may thereby reduce their individual chance of being selected by a predator.  Early Stuart sockeye salmon have large energy  constraints due to their long migration. They may not have sufficient energy for diel periods o f spawning activity and must use an ' a l l out' spawning strategy because T B s" occur even when they-^are inactive. Schooling fish are inactive day 1  68  and night, but this may be adaptive for reducing T B s" while eggs mature, as 1  explained previously. M e a n temperatures varied little during the years 1995 and 1996. Brett (1995) found temperature increases the maximum sustainable activity rate more than increasing metabolic demand during activity, and this effect is small for temperatures between 5 to 10°C. A n y temperature effects may be corrected in part by the methodology used in this study (i.e. E M G radio telemetry)(Hinch and Rand 1998). I f a significant temperature effect occurred it would not alter comparisons o f total muscle activity in the same year. Therefore, a relative comparison between sexes is valid within each year. Comparing the muscle activity of the same sex between years may have additional error due to temperature effects.  H o w e v e r , this error may not be significant because the  differences in sexes between years is large and the differences between mean temperatures are small. M y third objective was to use the activity measures to derive an estimate of T T B , and compare the relative amount o f activity by sex and year to the results reported by Idler and Clemens (1959) and Gilhousen (1980). Gilhousen (1980) in a re-analysis o f Idler and Clemens (1959) data, reported male sockeye salmon utilize more energy than females in the early Stuart population. M y data shows two alternative results in comparing muscle activity. I found that males utilized less T T B and had shorter spawning ground duration in 1995, and males were equal to females in T T B and spawning ground duration in 1996. Idler and  69  Clemens (1959) used body constituent analysis to determine kilocalories utilized, whereas I used muscle activity ( T T B ) to compare energy use. Although the units of energy are different, comparative analysis is still appropriate.  Gilhousen  (1980) showed the males had more constituent energy upon death than females, but utilized more kcal/kg/day. M y data show the females used more T T B day"  1  during the spawning stage. H o w e v e r , i f the females hold in schools for a longer duration, this may lower their total mean spawning ground activity rate in years where spawning ground duration o f females is equal or greater than males. U n l i k e Idler and Clemens (1959), I assessed how energy is used during different stages o f spawning. Differences in male and female longevity and T T B expended by status may effect total energy used by sex and year. Gilhousen (1980) showed that male sockeye salmon used more energy than females in both years reported by Idler and Clemens (1959). T w o possible alternatives could have been found. The females could have used more energy, or both sexes could have used the same amount o f energy. M y data for two years showed both alternatives. In 1995 the females used significantly more T T B , and in 1996 both sexes used the same T T B .  This suggests that either alternative is possible and Idler and  Clemens' (1959) data do not define a ' r u l e ' o f energy use with male and female early Stuart sockeye salmon. Idler and Clemens (1959) sampled newly arrived fish on the spawning grounds and then 'fresh' dead fish. This method requires many assumptions about energy use by different fish and does not sample the same fish 'before and after'.  M y study had the improvement o f following specific  70  fish from arrival until death.  This provided detailed analysis o f effects o f  behaviour, status, and individual fish duration on the spawning grounds on T T B expenditure. In summary: following the procedures to test my first objective, I used E M G measurements o f lateral swim muscle contractions to discriminate muscle activity in T B s" and in T T B units for specific behaviours. This shows E M G 1  technology has the precision required to measure muscle activity associated with specific behaviours. Measures o f T B s" and T T B for specific behaviours may 1  allow a more insightful analysis o f the ecological significance o f the behaviour. F o l l o w i n g the procedures to test my second objective, mean T T B were compared for behaviour-specific behaviours and general activity (and by status, sex, and year).  Specific behaviours, which have a higher T B s" than general activity, were 1  shown to contribute the least to T T B o f the fish during i t ' s life on the spawning grounds. This suggest general activity is the largest source o f activity and subsequently energy expenditure in spawning sockeye salmon. M y third objective was difficult to test because I was comparing T T B to k i l o c a l o r i e s consumed. I used a relative activity approach to compare T T B , whereas Gilhousen (1980) used an absolute measure o f energy in kilocalories. H o w e v e r , I suggest that when comparing similar techniques first, comparisons may be made among the conclusions by the two different techniques.  Therefore I suggest the comparison  of my data to Gilhousen's (1980) data allows an inference to be made between the studies. T T B use by sex in both study years was different than kilocalorie  71  results reported by Gilhousen (1980). Gilhousen's (1980) results could be influenced by unstated assumptions about the duration o f males and females on the spawning grounds. Differences may occur between sexes and years in somatic energy storage and expenditure. In conclusion, E M G radio telemetry is an effective method o f measuring muscle activity and comparing energy expenditure. It may, with calibration, provide a direct estimate o f energy expenditure in spawning fish.  72  REFERENCES  Belding, D . L . 1934. The spawning habits o f the A t l a n t i c salmon. of the American Fisheries Society 64:211-218.  Transactions  Betts, J. E . 1976. Physics for Technology. Reston Publishing Company, I N C . Reston, V i r g i n i a . 723 pp. Brannon, E . L . 1987. Mechanisms stabilizing salmonid fry emergence timing. Pages 120-124, in H . D . Smith, L . M a r g o l i s , and C . C . W o o d s , editors. Sockeye salmon (Oncorhynchus nerka) population biology and future management. Canadian Special Publication o f Fisheries and A q u a t i c Sciences 96. Brett, J. R. 1964. The respiratory metabolism and swimming performance of young sockeye salmon. Journal o f the Fisheries Research B o a r d o f Canada 21 (5):1 183-1226. Brett, J.R. 1973. Energy expenditure of sockeye salmon, Oncorhynchus nerka, during sustained performance. Journal o f the Fisheries Research B o a r d of Canada 30 (12): 1799-1809. Brett, J.R. 1995. Energetics. Pages 3-68, in C . G r o o t , L . M a r g o l i s , and W . C. Clarke, editors. P h y s i o l o g i c a l ecology o f Pacific salmon. U B C Press. University o f B r i t i s h Columbia, Vancouver, B r i t i s h C o l u m b i a , Canada. Brett, J. R. and N . R. Glass. 1973. M e t a b o l i c rates and critical swimming speeds of sockeye salmon, Oncorhynchus nerka, in relation to size and temperature. Journal of the Fisheries Research B o a r d o f Canada 30 (3):379-387. Briggs, J. C . 1953. The behaviour and reproduction o f salmonid fishes in a small coastal stream. Fisheries Bulletin o f the California Department o f Fish and Game 94:62 pp. Boisclair, D . and M . Tang. 1993. E m p i r i c a l analysis o f the influence o f swimming pattern on the net energetic cost o f swimming in fishes. Journal of Fish B i o l o g y 42:169-183. Burgner, R. L . 1991. Life history o f sockeye salmon (Oncorhynchus nerka). Pages 1-117, in Groot, C . and L . M a r g o l i s , editors. Pacific salmon life histories. U B C Press, University o f B r i t i s h Columbia, Vancouver, Canada. 73  Calow, P. and A . S. Woolhead. 1977. The relationship between ration, reproductive effort and age-specific mortality in the evolution o f lifehistory strategies - some observations on freshwater triclads. Journal o f Animal E c o l o g y 46:765-781. Chebanov, N . A . , N . V . Varnavskaya, and V . S. V a r n a v s k i y i . 1983. Effectiveness o f spawning o f male sockeye salmon, Oncorhynchus (Salmonidae), o f differing hierarchical rank by means o f genetic biochemical markers. Journal of Ichthyology 23:51-55.  nerka  C l u t t o n - B r o c k , T. H . 1984. Reproductive effort and terminal investment in iteroparous animals. American Naturalist 123:212-229. Demers, E . , R. S. M c K i n l e y , A . H . Weatherley, and D . J. M c Q u e e n . 1996. A c t i v i t y patterns o f largemouth and smallmouth bass determined with electromyogram biotelemetry. Transactions o f the American Fisheries Society 125:434-439. Evans, D . M . 1994. Observations on the spawning behaviour o f male and female adult sea trout, Salmo trutta L . , using radio-telemetry. Fisheries Management and E c o l o g y 1994 1:91-105. Foote, C. J. 1989. Female mate preference in pacific salmon. Animal Behaviour 38:721-723. Foote, C . J. and P. A . L a r k i n . 1988. The role o f male choice in the assortative mating o f anadromous and non-anadromous sockeye salmon. Behaviour 106:43-62. Frank, T. H . 1968. Telemetering the electrocardiogram o f free swimming Salmo iridieus. I E E E Transactions in B i o m e d i c a l Engineering 15:111-114. Fry, F . E . J. 1971. The effect o f environmental factors on the physiology o f fish. Pages 1-98, in W . S. Hoar and D . J. Randall, editors. Fish physiology. Volume 6: Environmental relations and behaviour. Academic Press, N e w York, U S A Gilhousen, P. 1980. Energy sources and expenditures in Fraser R i v e r sockeye salmon during their spawning migration. International Pacific Salmon Fisheries Commission B u l l e t i n 22:51 pp. Greeley, J. R . 1932. The spawning habits o f brook, brown, and rainbow trout, and the problem o f egg predators. Transactions o f the American Fisheries Society 62:239-248.  74  Gross, M . R. 1984. Sunfish, salmon and the evolution o f alternative reproductive strategies and tactics in fishes. Pages 55-75, in R. Wootton and G . Potts, editors. Fish reproduction: strategies and tactics. Academic Press, L o n d o n . Gross, M . R. 1985. Disruptive selection for alternative life histories in salmon. Nature 313:47-48. Hanson, A . J. and H . D . Smith. 1967. Mate selection in a population o f sockeye salmon (Oncorhynchus nerka) o f mixed age-groups. Journal o f the Fisheries Research B o a r d of Canada 24(9): 1955-1957. Hartley, S. E . 1987. The chromosomes of salmonid fishes. 62:197-214.  Biological Review  Hasler, A . D . , E . S. Gardella, R. M . H o r r a l l , and H . F. Henderson. 1969. Open water orientation of white bass, Roccus chrysops, as determined by ultrasonic tracking methods. Journal of the Fisheries Research B o a r d of Canada 26:2173-2192. Hasler, A . D . , R. M . H o r r a l l , A . B . Stasko, and A . E . D i z o n i . 1970. Orientation cues and tracking o f migrating salmonid fishes. Proceedings o f the National Academy o f Science U S A 66:1374. Hazzard, A . S. 1962. Some phases o f the life history o f the eastern brook trout, Salvelinus fontinalis. Healey, M . C . 1987. The adaptive significance o f age and size at maturity in female sockeye salmon (Oncorhynchus nerka). Pages 110-117, in H . D . Smith, L . M a r g o l i s , and C . C. W o o d . Sockeye salmon (Oncorhynchus nerka) population biology and future management. Canadian Special Publication o f Fisheries and Aquatic Sciences 96. Heard, W . R. 1972. Spawning behaviour of pink salmon on an artificial redd. Transactions of the American Fisheries Society 2:276-283. Henderson, H . F . , A . D . Hasler, and G . C. Chipman. 1966. A n ultrasonic transmitter for use in studies o f movements o f fishes. Transactions of the American Fisheries Society 95:350-356. Hinch, S. G . and N . C . Collins. 1991. Importance o f diurnal and nocturnal nest defense in the energy budget of male smallmouth bass: insights from direct video observations. Transactions o f the American Fisheries Society 120:657-663.  75  H i n c h , S. G . , R. E . Diewert, T. J. Lissimore, A . M . J. Prince, M . C . Healey, and M . A . Henderson. 1996. Use o f electromyogram telemetry to assess difficult passage areas for river-migrating adult Pacific salmon. Transactions o f the American Fisheries Society 125:253-260. Hinch, S. G . and P. S. Rand. 1998. Swim speeds and energy use o f up-river migrating sockeye salmon: role of local environment and fish characteristics. Canadian Journal o f Fisheries and A q u a t i c Sciences. Hoopes, D . T. 1972. Selection of spawning sites by sockeye salmon in small streams. Fisheries B u l l e t i n 70(2):447-458. Idler, D . R. and W . A . Clemens. 1959. The energy expenditures o f Fraser River sockeye salmon during the spawning migration to Stuart and C h i l k o Lakes. International Pacific Salmon Fisheries Commission Progress Report 6:80 pp. Johnson, J. H . 1960. Sonic tracking of adult salmon at Bonneville D a m . Fisheries B u l l e t i n 176. Jonsson, N . , B . Jonsson, and L . P. Hansen. 1991. Energetic cost o f spawning in male and female Atlantic salmon. Journal o f Fish B i o l o g y 39:739-744. Kanwisher, J . , K . L a w s o n , and R. Strauss. 1974. A c o u s t i c telemetry from human divers. Undersea B i o m e d i c a l Research 1:99-109. Kaseloo, P. A . , A . H . Weatherley, J. Lotimer, and M . D . Farina. 1992. A biotelemetry system recording fish activity. Journal o f F i s h B i o l o g y 40:165-179. Kaseloo, P. A . , A . H . Weatherley, P: E . Ihssen, D . A . Anstey, and M . D . Gare. 1996. Electromyograms from radiotelemetry as indicators o f reproductive activity in lake trout. Journal o f Fish B i o l o g y 48:664-674. Keenleyside, M . H . and H . M . C. Dupuis. 1988. Courtship and spawning competition in pink salmon (Oncorhynchus gorbuscha). Canadian Journal of Z o o l o g y 66:262-265 >  K i l l i c k , S. R. 1955. The chronological order of Fraser R i v e r sockeye salmon during migration, spawning and death. International Pacific Salmon Fisheries Commission B u l l e t i n 7:95 pp. K r o h n , M . M . and D . B o i s c l a i r . 1994. Use o f a stereo-video system to estimate the energy expenditure o f free-swimming fish. Canadian Journal o f Fisheries and Aquatic Sciences 51:1119-1127.  76  Lucas, M . C , A . D . F . Johnstone, and I. G . Priede. 1993. Use o f physiological telemetry as a method o f estimating metabolism o f fish in the natural environment. Transactions o f the American Fisheries Society 122:822-833. Mathisen, O. A . 1962. The effect o f altered sex ratios on the spawning o f red salmon. Pages 137-245, in studies o f A l a s k a red salmon (1). University of Washington, Washington, U S A . M c C a r t , P.. 1969. D i g g i n g behaviour o f Oncorhynchus nerka spawning in streams at Babine L a k e , B r i t i s h Columbia. Pages 39-51, in T. G . Northcote, editor. H . R. M a c M i l l a n Lectures in Fisheries. Symposium on salmon and trout in streams. University o f B r i t i s h C o l u m b i a , Vancouver, B r i t i s h C o l u m b i a , Canada. M c C a r t , P. 1971. A Polymorphic population of Oncorhynchus nerka at Babine L a k e , B r i t i s h C o l u m b i a involving anadromous sockeye and non-anadromous (kokanee) forms. P h . D . thesis, Department o f Z o o l o g y , U n i v e r s i t y of B r i t i s h C o l u m b i a , Canada. 135 pp. M c K i n l e y , R. S. and G . Power. 1992. Measurement of activity and oxygen consumption for adult lake sturgeon (Acipenser fulvescens) in the wild using radio-transmitted E M G signals. Pages 307-318, in I. G . Priede and S. M . Swift, editors. Wildlife telemetry: remote monitoring and tracking o f animals. E l l i s H o r w o o d , West Sussex, United K i n g d o m . Miescher-Rusch, F . 1880. Contributions to the biology o f the Rhine salmon. United States Bureau o f Fisheries B u l l e t i n 8: 427-447. Needham, P. R. and A . C . Taft. 1934. Observations o f the spawning o f steelhead trout. Transactions o f the American Fisheries Society 64:332-338. Nomura, S and T. Ibaraki. 1969. Electrocardiogram o f the rainbow trout and its radio transmission. Japanese Journal o f Veterinary Science 31:135-147. Nomura, S., T. Ibaraki, H . Hirose, and S. Shirahata. 1972. Applications o f back pack cardiotelemetry for fishes. 1: Heart rate and cardiac reflex in fishes during unrestrained swimming. B u l l e t i n o f the Japanese Society o f Scientific Fisheries 38:1105-1117. Osborne, M . F . M . 1961. The hydrodynamic performance o f migratory salmon. Journal o f Experimental B i o l o g y 38:365-390. Poddubnyi, A . G . , Y . I. Spekor, and S. M . K i l i u n . 1966. The results o f preliminary experiments in tracking sturgeon carrying electronic tracer tags. V o p . Ikhtiol. 6:725-734.  77  Priede, I. G . and A . H . Y o u n g . 1977. The ultrasonic telemetry o f cardiac rhythms o f w i l d brown trout, Salmo trutta, as an indicator o f bioenergetics and behaviour. Journal of Fish B i o l o g y 10:299-318. Prince, A . M . J . , S. E . L o w , T. J. L i s s i m o r e , R. E . Diewert, and S. G . H i n c h . 1995.Sodium bicarbonate and acetic acid: an effective anesthetic for field use. N o r t h American Journal of Fisheries Management 15:170-172. Rand, P. S. and S. G . H i n c h . 1998. Swim speeds and energy use o f up-river migrating sockeye salmon: simulating metabolic power and assessing risk of energy depletion. Canadian Journal o f Fisheries and A q u a t i c Sciences. Randall, D . J. 1970. The circulatory system. Pages 133-172, in W . S. Hoar and D . J. R a n d a l l , editors. Fish physiology (4). Academic Press, N e w Y o r k . Roff, D . A . 1992. The evolution o f life histories. Chapman and H a l l , L o n d o n . 535 pp. Rogers, S. C , D . W . Church, A . H . Weatherley, and D . G . P i n c o c k . 1984. A n automated ultrasonic telemetry system for the assessment o f locomotor activity in free-ranging rainbow trout, Salmo gairdneri. Journal o f Fish B i o l o g y 25:697-710. Rogers, S. C , A . H . Weatherley, D . G . Pincock, and J. R. Patch. 1981. Telemetry, electromyograms, and oxygen demands o f fish activity. Pages 141-150, in Proceedings of the third international conference on wildlife biotelemetry. University o f W y o m i n g , Laramie, U S A . Ross, L . G . , W . Watts, and A . H . Y o u n g . 1981. A n ultrasonic biotelemetry system for the continuous monitoring o f tail-beat rate from free-swimming fish. Journal o f Fish B i o l o g y 18:479-490. Satou, M . , A . Shiraishi, T. Matsushima, and N . Okumoto. 1991. V i b r a t i o n a l communication during spawning behaviour in the hime salmon (landlocked red salmon, Oncorhynchus nerka). Journal o f Comparative Physiology A . 168:417-428. Semenchenko, N . N . , and V . I. O s t r o v s k i i . 1998. A method o f determining the oxygen consumption rate in the sockeye salmon, Oncorhynchus nerka, during the spawing period. Journal of Ichthyology 38 N o . 4:325-329. Sheridan, W . L . 1960. Frequency o f digging movements o f female pink salmon before and after egg deposition. Animal Behaviour 8(3-4):228-230.  78  Smith, J. N . M . 1981. Does high fecundity reduce survival in song sparrows. E v o l u t i o n 35:1142-1148. Stanley, B . V . 1983. Effect o f food supply on reproductive behaviour o f male Gasterosteus aculeatus. P h . D . thesis. University o f Wales. Page 249, in P. Tytler and P. Calow, editors. Fish energetics: new perspectives. C r o o m H e l m L t d . , Sydney, Australia. Stasko, A . B . and R. M . H o r a l l . 1976. M e t h o d o f counting tail-beats o f free swimming fish by ultrasonic telemetry techniques. Journal o f the Fisheries Research B o a r d o f Canada 33:2596-2598. Stearns, S. C . 1976. Life-history tactics: a review o f the ideas. R e v i e w of B i o l o g y 51:3-46.  Quarterly  Tautz, A . F . and C . G r o o t . 1975. Spawning behaviour o f chum salmon (Oncorhynchus keta) and rainbow trout (Salmo gairdneri). Journal o f the Fisheries Research B o a r d o f Canada 32:633-642. Tinbergen, N . 1952. " D e r i v e d " activities: their causation, biological significance, origin and emancipation during evolution. Quarterly R e v i e w of B i o l o g y 27:1-32. Trefethen, P. S. 1956. Some equipment for tracking individual fish. Special Scientific Report. United States Fish and Wildlife Service 179: 11 pp. Uematsu, K . , I. Hanyu, T. H i b i y a , and K . Y a m a m o r i . 1979. E M G recording from the spawning salmon. B u l l e t i n o f the Japanese Society o f Scientific Fisheries 45(3):409. V o e g e l i , F . A . and D . G . Pincock. 1980. Determination o f fish swimming speed by ultrasonic telemetry. Biotelemetry Patient M o n i t o r i n g 7:215-220. Wardle, C . S. and J. W . Kanwisher. 1974. The significance o f heart rate in free swimming cod, Gadus morhua: some observations w i t h ultrasonic tags. Marine Behavioural Physiology 2:311-324. Weatherley, A . H . 1976. Factors effecting maximization o f fish growth. of the Fisheries Research B o a r d of Canada 33:1046-1058.  Journal  Weatherley, A . H . , P. A . K a s e l o o , M . D . Gare, J. M . Gunn, and B . L i p i c n i k . 1996. F i e l d activity o f lake trout during the reproductive period monitored by electromyogram radiotelemetry. Journal o f Fish B i o l o g y 48:675-685.  79  Weatherley, A . H . , S. C . Rogers, and D . G . Pincock. 1980. Use o f telemetry in monitoring intensity and energetics o f activity in free-swimming fish with reference to zinc pollution. Canadian Technical Report o f Fisheries and Aquatic Sciences 975:162-170. Weatherley, A . Ff., S. C . Rogers, D . G . Pincock, and J. R . Patch. 1982. Oxygen consumption o f active rainbow trout, Salmo gairdneri, derived from electromyograms obtained by radio-telemetry. Journal o f Fish B i o l o g y 20:479-489. Weatherley, A . H . , P . A . K a s e l o o , M . D . Gare, J. M . Gunn, and B . L i p i c n i k . 1996. F i e l d activity o f lake trout during the reproductive period monitored by electromyogram radiotelemetry. Journal o f Fish B i o l o g y 48:675-685. Weintraub, M . J . and R . S. M a c K a y . 1975. Respiratory and heart beat synchrony studied by telemetry in the trout, Salmo gairdneri. Copea:78-85. White, H . C . 1942. Atlantic salmon redds and artificial spawning beds. Fisheries Research B o a r d o f Canada. P p . 37-44. Williams, I. V . , J. R . Brett, G . R . B e l l , G . S. Traxler, J. Bagshaw, J. R . M c B r i d e , U . H . M . Fagerlund, H . M . Dye, J. P. Sumpter, ,E. M . Donaldson, E . B i l i n s k i , H . Tsuyuki, M . D . Peters, E . M . Choromanski, J. H . Y . Cheng, and W . L . Coleridge. 1986. The 1983 early run Fraser and Thompson river pink salmon; morphology, energetics and fish health. International Pacific Salmon Fisheries Commission B u l l e t i n 23: 55 pp. Wootton, R . J. 1985. Energetics o f reproduction. Pages 231-256, in P . Tytler and P . C a l o w , editors. Fish energetics: new perspectives. C r o o m Helm L t d . , Sydney, Australia. Y o u n g , A . H . , P . Tyler, F . G . T. H o l l i d a y , and A . MacFarlane. 1972. A small sonic tag for measurement o f locomotor behaviour in fish. Journal o f Fish B i o l o g y 4:57-65. Y u e n , H . S. 1970. Behaviour o f skipjack tuna, Katsuwanus pelanus, as determined by tracking with ultrasonic devices. Journal o f the Fisheries Research B o a r d o f Canada 27:2071-2079.  80  APPENDICES  81  APPENDIX I D E F I N I T I O N OF T E R M S Pacific Salmon Names: Oncorhynchus  - Pacific salmon genus [Latin; hooked nose]  gorbuscha  - pink salmon [Russian; h i l l , hump salmon]  keta - chum salmon [Russian; China, Chinese salmon] kisutch - coho salmon [Russian; swarming, darting salmon] nerka - sockeye salmon [Russian; schooling, fearless salmon] tshawytscha  - chinook salmon [Russian; longest, largest salmon]  Definition o f Terms: charr - a salmonid [Scottish; from L a t i n 'char' - to change color] diel - a full 24 hour day [Latin; 'dies' - a completed cycle] diurnal - by day [Latin; ' d i u r n ' - day] esophageal - through the esophagus, placing a transmitter in the stomach. nocturnal - by night [Latin; ' n o c t u r n ' - night] redd - area o f nest digging [Scottish; - clear] salmo - trout genus [Latin; salmon, salmo, ' s a l i o ' - to leap] salvelinus  - charr genus [Latin; 'salve', Tinus' - healthful type]  skeins - 2 membranes containing egg follicles which mature into eggs. subcutaneous - under the skin. trout - a salmonid [Latin; trutta, from Greek troktes, ' t r o g o s ' - to gnaw]  82  A P P E N D I X II FISH E S C A P E M E N T TO G L U S K I E C R E E K Fish Escapement by Year and Stream.  YEAR  GLUSKIE  1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996  10370 966 4646 4295 10040 1049 10741 452 3781 6943 17381 3468 19973 17700 9062 12078 17365 3749 30052 3919 9500 8000  1  FORFAR  O'NE-  9818 1249 3628 9579 15805 2328 12228 17975 13073 2221 9922 5455 23395 24640 10824 14994 21885 12904 37700 4902  25124 6727 5893 10649 34228 10661 13452 1170 7822 16933 20347 7000 41424 53447 22528 21243 28489 11002 38046 4371  Gluskie Creek was the study stream. Two adjacent creeks are shown for comparison. 1  83  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items