Early ocean survival and marine movements of hatchery ... - CiteSeerX

4 downloads 10 Views 268KB Size Report
Charlotte Strait, a large (20 Â 100 km2) marine area separating Vancouver Island from the mainland. The results provide the first detailed data on the ocean biology of hatchery and wild ... from release point to recovery site is unclear. Although ...

ARTICLE IN PRESS

Deep-Sea Research II 51 (2004) 897–909

Early ocean survival and marine movements of hatchery and wild steelhead trout (Oncorhynchus mykiss) determined by an acoustic array: Queen Charlotte Strait, British Columbia David W. Welcha,b,*, Bruce R. Wardc, Sonia D. Battena a

Kintama Research Corporation, 4737 Vista View Center, Nanaimo, BC, Canada V9V 1N8 Pacific Biological Station, Fisheries and Oceans Canada, Nanaimo, BC, Canada V9T 6N7 c BC Ministry of Water, Land and Air Protection, Aquatic Sciences Section, 2204 Main Mall, University of British Columbia, Vancouver, BC, Canada V6T 1Z4 b

Received 23 September 2002; accepted 3 May 2004 Available online 3 August 2004

Abstract Early ocean movements, residency, and survival of steelhead (Oncorhynchus mykiss) were examined in Queen Charlotte Strait, a large (20  100 km2) marine area separating Vancouver Island from the mainland. The results provide the first detailed data on the ocean biology of hatchery and wild steelhead smolts. Initial ocean movements were not strongly directed, with most smolts swimming in the range of 0.2–0.5 body length (BL) s1. The majority (78%) vacated Queen Charlotte Strait within 1 week of release in freshwater. Relative marine survival of hatchery smolts surgically implanted 1 month prior to release was identical to that of wild smolts implanted on the day of release; survival of hatchery smolts transported to the study site, implanted, and released all on the same day was significantly lower. The results suggest that the early marine survival of hatchery and wild smolts may be fundamentally similar, but that the cumulative stress of transportation and surgery may reduce post-surgery survival. Hatchery smolts moved at higher average swimming speeds than wild smolts, but the difference was not statistically significant. Early marine survival within the study region appears to be relatively high (X55%), contradicting assumptions that the early marine phase is the critical period for determining salmon recruitment. r 2004 Elsevier Ltd. All rights reserved.

1. Introduction The construction of a continental-scale marine tracking array has been proposed for the west coast of North America (Welch et al., 2003). If *Corresponding author. Kintama Research Corporation, 4737 Vista View Center, Nanaimo, BC, Canada V9V 1N8. E-mail address: [email protected] (D.W. Welch).

built, such a tracking array would for the first time provide detailed information on the ocean movements of fish as small as 10–11 cm body length (BL) over periods of many months or years. As part of this project, a large-scale trial array was deployed for 2 months off the east coast of Vancouver Island in 2002 to assess the feasibility and potential scientific value that a permanent array could provide.

0967-0645/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr2.2004.05.010

ARTICLE IN PRESS 898

D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

The early ocean life history of Pacific salmon smolts is the least understood phase of the life history, chiefly because of the difficulty of studying salmon in the ocean. Although there are now several research programs that have started to study the ocean biology of Pacific salmon based on newly developed large epipelagic trawls (see review by Brodeur et al., 2000), there are significant limitations on the interpretation of data collected from traditional research programs that capture and kill fish at sea. Because capture at sea almost invariably involves the death of the animal, the detailed behaviour determining the movement of animals from release point to recovery site is unclear. Although sampling of fish caught at sea provides important information on the collected animals as a group, interpretation of the data is limited by the lack of information on how the animals arrived at the capture location. A common assumption is that the stock composition of a sample of fish does not change over time, so that temporal changes can be interpreted as variation in the characteristics of the same group of fish (e.g., changes in size), rather than changes brought about by movement of different populations through the study area. Such assumptions may be unwarranted. Pacific salmon demonstrate a remarkable ability to return precisely to the vicinity of their own birthplace from the sea (e.g., Quinn, 1993; Quinn et al., 1999), so it seems reasonable that their marine feeding grounds and the ocean migration pathways taken to get to them there may be just as population specific as their freshwater spawning grounds (Welch et al., 2003). For example, Weitkamp and Neely (2002) provide evidence suggesting that the marine migratory pathways of coho salmon (O. kisutch) show considerable population-specific diversity, although the detailed migratory movements necessary to yield these patterns are uncertain. There is, as yet, little direct evidence for sharply defined and population-specific movements in the ocean because of the technical difficulties involved in studying salmon movements within the ocean. However, the evidence for remarkable populationspecific seasonal migration behaviours is well

known in both birds (e.g., Hobson, 2002; Rubenstein et al., 2002) and insects (e.g., Denlinger, 1994; Wassenaar and Hobson, 1998; Anonymous, 1999). It seems unlikely that marine fish, with their evolutionarily ancient lineage and the vast size and complexity of the oceans, would not have developed migratory abilities at least as sophisticated as those of birds and insects. The marine survival of Keogh River steelhead (Oncorhynchus mykiss) in recent years has been very low (4%), in common with all east coast Vancouver Island and southern mainland steelhead populations (Ward, 2000). Curiously, West Coast Vancouver Island steelhead stocks continued to do well during the 1990s, raising the question of what differences in marine experience led to such radically different marine survival histories (Welch et al., 2000). Despite the near proximity of the Keogh River steelhead population to the West Coast Waukwaas River—the river mouths (RMs) lie only 11 km apart by land and 150 km apart by sea around the northern tip of Vancouver Island-marine survival of all steelhead stocks on the east coast of Vancouver Island dropped dramatically in the 1990s, reducing the number of adults returning to these systems to a few dozen adults, as compared to several thousand adults in the late 1980s (Ward and Slaney, 1988; Ward, 2000; Welch et al., 2000). In recent years, acoustic technology suitable for use in the marine environment has advanced to the point where it has been successfully used to tag and track Atlantic salmon smolts (Voegeli et al., 1998; Lacroix and Voegeli, 2000), and is suitable for long-term implantation in fish as small as 11 cm (Welch et al., 2004). Because of the relatively low acoustic power output of the tags necessary to ensure long-life, the detection range of these tags by a given receiver is only ca. 500 m. The primary objective of the Queen Charlotte Strait work described here was to test the concept that a large acoustic array can be put in place in the ocean, and provide important information on movement patterns and (potentially) survival of juvenile steelhead. It has not been possible to contemplate gathering such information in the ocean until the recent convergence of relatively low-cost and low-power-consumption electronics;

ARTICLE IN PRESS D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

it is on the basis of the rapid developments in electronics that the concept of a large-scale acoustic tracking array is feasible. Because acoustic tags with large numbers of unique codes and multi-month lifespans can be surgically implanted into fish as small as 11 cm (Welch et al., 2004), it is possible to imagine the development of a single continental-scale acoustic tracking array sitting permanently on the seabed, tracking the movements of many thousands of individual fish over almost their entire life history. The continental shelf off the west coast of North and South America is in many places only 20– 30 km wide, yet stretches for more than 10,000 km. Thus, for shelf or slope resident marine animals, a series of seabed acoustic ‘‘curtains’’ stretching from land out to the edge of the continental shelf and down the slope would provide detailed information on the movements of marine animals at relatively low cost. Because the shelf is narrow relative to its length, relatively few seabed nodes will be required to form each line, thus making the development of an extensive multi-line array economically feasible. The primary goal of this study was to demonstrate that large-scale acoustic curtains formed of multiple passive acoustic detectors sitting permanently on the seabed can provide useful information on the movements and survival of marine animals. The need for permanent seabed observatories has recently been highlighted (National Research Council, 2000); however, the use of such observatories for constructing a fish-tracking system has not been investigated. Our work was intended to provide data on how well such a system might perform if constructed. The second objective of this study was to compare the initial marine movements and survival of wild and artificially reared juvenile steelhead from the same genetic population. Major efforts have been made in the past century to artificially enhance many Pacific salmon populations by supplementing the wild stock with hatchery-reared juveniles. Because survival during the part of the life cycle completed in a hatchery is higher than in the wild, hatchery-reared salmon are generally expected to have higher productivity. However, hatchery supplementation has come under increas-

899

ing question partly because it is unclear whether survival subsequent to release is sufficiently low to negate this initial benefit and because of concerns that artificial rearing may alter behaviour (National Research Council, 1996). The 2002 trial provided an opportunity to examine this question by comparing the movements and survival of hatchery and wild steelhead from the same stock.

2. Materials and methods 2.1. Study location Queen Charlotte Strait (50.9 N; 127.5 W) is a large (20  100 km2) marine area lying between the Northeast corner of Vancouver Island and mainland British Columbia (Fig. 1). To the north it expands into Queen Charlotte Sound—the open continental shelf bounded 250 km to the north by the Queen Charlotte Islands—and to the south it funnels down to Johnstone Strait, a narrow (2– 3 km), deep, and tidally active passage leading southwards to the Strait of Georgia and then eventually out to the open Pacific via the Strait of Juan de Fuca, which separates Vancouver Island and Washington State. At the western side of Queen Charlotte Strait, the Keogh River drains directly into the Strait, running 33 km north-east from the headwaters at Keogh Lake down to the sea. A weir situated about 300 m away from the ocean blocks the movement of steelhead smolts and kelts (adults which survive spawning), providing an easy collection point for tagging. 2.2. Acoustic array Vemco Inc., VR-2 acoustic receivers were used to form the detection array. The VR-2 is a single frequency autonomous battery powered underwater unit capable of long-term placement in the ocean (up to 15 months using a single lithium ‘‘D’’-cell). Receivers detect acoustic codes transmitted by the tags and then compare them with check-sums transmitted as part of the code train. The receivers internally store the detected tag code (along with the date and time of the detection)

ARTICLE IN PRESS 900

D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

Fig. 1. Location of the Queen Charlotte Strait acoustic array, British Columbia. The array was formed of two lines: (AB) 19 km long to the north and (C and D) to the south (2 and 6 km long, respectively). One receiver (RM) was placed in the ocean offshore from the mouth of the Keogh River. Triangles indicate the location of each acoustic receiver.

only in the event that the tag code and the subsequently transmitted check-sums agree. False detections, where a decoded serial number does not agree with the check-sum, are not logged, but a record of the total number of such occurrences is stored internally in the receiver. Each VR-2 receiver used in this study was programmed with a specially chosen code map that was unique to our study. Because of the code map we designed for the study, acoustic tags from studies other than our own would not be recorded by our receivers; conversely, our acoustic tags

would not be recorded by any other investigation’s receivers. We chose this approach because the continental-scale array planned for future deployment can potentially monitor 256,000 unique codes at the same time. As modern acoustic tags can have lifespans of years, it is critical for the integrity of future scientific studies that the uniqueness of a specific acoustic code be protected. Our study was designed to adhere to this principle, by using a unique code map which we could be certain was not compromised by being used elsewhere.

ARTICLE IN PRESS D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

Two major acoustic detection lines were placed in Queen Charlotte Strait on 5 and 6 May 2002, prior to the planned release of steelhead smolts with implanted acoustic tags (Fig. 1). Each line consisted of multiple VR-2 acoustic receivers placed about 850 m apart on the seabed and stretching across the breadth of the Strait in order to detect any tagged smolts present. Receivers were deployed by attaching them to a groundline, using technology similar to that used in commercial fishing. The northern AB Line extended 19 km from Duval Point, Vancouver Island (50.8 N, 127.5 W) to the British Columbia mainland just south of Browning Island (50.9 N, 127.3 W). This line was intended to detect the movement of any tagged steelhead smolts moving north towards Queen Charlotte Sound and the open Pacific. The southern CD Line was laid in two lines, with line C (1.7 km) covering Broughton Strait, the body of water lying between Vancouver Island and Malcolm Island, and Line D (6 km) extending from the southern tip of Malcolm Island to Swanson Island (50.3 N, 126.74 W). These lines covered the two possible approaches to Johnstone Strait and the southern exit from Queen Charlotte Strait for tagged smolts. The receiver placed off the RM was positioned 12 m below the surface and 320 m from shore. This location was the middle of a broad channel cut into the plate rock forming the foreshore, and which was probably the river course during periods of lower sea level. At high tide the ocean flooded a broad area of plate rock surrounding the river, and it was recognised that tagged animals that did not swim directly offshore from the RM would likely not be heard by the acoustic receiver because the receiver was below the top of the channel. This was a compromise intended to provide some information on the time of entry of tagged steelhead into the ocean while ensuring the safety of the receiver. Acoustic receivers were held about 0.5 m above the seabed by attaching them to a vertical line, attached on the bottom to the groundline and with pressure-resistant floatation attached to the upper end to maintain the receiver with the hydrophone pointed towards the surface. The flotation was located approximately 4 m above the hydrophone

901

in order to form only a small acoustic shadow, except where water depths exceeded 200 m. In depths exceeding 200 m (the rated depth of the equipment), the length of the vertical risers was increased. Spacing between adjacent receivers was 850 m or less to provide an overlap of the detection ranges of adjacent receivers, with the greatest separation from a receiver at the end of an acoustic listening line to the shoreline being 400 m. The entire array was placed on the seabed, with no floats or equipment visible from the surface. Individual lines were recovered by triggering an acoustic release located on one or both ends of the line. This was done partly to ensure that equipment would not be molested during the nearly 2month deployment, but also because an important design principle for the continental-scale array is that is must be placed on the seabed, where it will not constitute a hazard to navigation or impediment to commercial fishing operations. Although the eventual design of the full array will use different deployment methodologies, which will be much less labour intensive, the Census of Marine Life field project was intended to adhere to this design philosophy. 2.3. Tagging Steelhead smolts were individually implanted using either Vemco V8SC-2L or -6L acoustic tags (20 and 28 mm long, by 9 mm diameter). Each tag was programmed to broadcast its code at 69 kHz with an average transmission interval of 60 s, with actual intervals distributed with uniform probability in the 30–90 s range. Projected lifespans for this programming choice were 4.5 and 15.5 months for the -6L and -2L tags, respectively. The larger V8SC-6L tags with their 15.5 lifespans were superfluous for the purposes of this 2-month study, but were used to illustrate that very longlived tags could also be used with the size of wild smolts encountered in the field study (16–18 cm). Tagging protocols are described in more detail elsewhere (Welch et al., 2004), but involved surgically implanting tags into the body cavity of the wild and hatchery smolts under general anaesthesia using clove oil. Surgery was done at

ARTICLE IN PRESS 902

D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

the river bank using a specially built portable battery powered surgical table. The main deviation from our previous protocol was the use of oxytetracycline, a broad spectrum antibiotic, which was injected into the body cavity through the incision prior to tag insertion at a dosage of 20 mg g1 body weight. Wild and hatchery smolts tagged at the weir were released back into the river after holding them in a darkened holding tank until dusk. They were held until dusk to reduce predation because substantial numbers of river otters, mink, kingfishers, and eagles were observed to congregate at the lower river each year during the smolt outmigration and feed on the migrating animals. Tagging was designed to allow a comparison of the movements of several different groups of Keogh River steelhead smolts. Because of very low marine survival of steelhead to the river, the British Columbia government instituted a ‘‘Living Gene Bank’’ program in the late 1990s to supplement the wild population while trying to minimise genetic impacts. Wild adult steelhead were collected from the river after their return from the sea and spawned in captivity, with the offspring held first in hatchery raceways and then outplanted to net pens located in O’Connor Lake at the headwater of the Keogh River. We refer to these animals as hatchery smolts. A total of 86 steelhead smolts and five adult kelts were surgically implanted with acoustic tags without respect for size. Smolt tagging was conducted to assess both potential differences in behaviour between (a) wild and hatchery smolts and (b) the effect of the surgical procedure on subsequent movement. Three groups are used in the analysis: 2.3.1. Implanted and held hatchery smolts Twenty hatchery smolts (16.4–21.1 cm fork length, FL) were taken from a net pen at O’Connor Lake and tagged on 25 April 2002, 1 month prior to release, and then returned to the same net pen. One smolt was subsequently found floating dead at the surface of the net pen 2–3 days later by hatchery staff, leaving a tagged group of 19 smolts. All smolts in the net pen were transported by tanker truck on 22 and 23 May

to the mouth of the Keogh River approximately 50 km away by logging road and released via a ca. 20 cm pipe into a pool just below the weir. The exact date of release for each tagged animal is uncertain because the animals from the net pen were transported over 2 days; we take 22 May as the time of release in subsequent calculations. We refer to this group as ‘‘implanted and held’’ hatchery smolts, as they were allowed to recover from surgery prior to release. 2.3.2. Implanted and immediately released hatchery smolts Eighteen hatchery smolts (17.4–24.1 cm FL) from the same net pen group were surgically implanted 1 month later, on 23 May, immediately following tanker transport from the headwaters. These smolts were then held until dusk in a darkened box with flow-through river water at the weir, and then released below the weir. We refer to this group as ‘‘immediately implanted and released’’ hatchery smolts. 2.3.3. Implanted and immediately released wild smolts Forty-nine wild smolts (15.6–21.5 cm FL) collected at the weir on 23, 25 and 26 May were surgically implanted, held until dusk, and then immediately released. Their handling was identical to the tagged and immediately released hatchery smolts, except that they had not been transported prior to surgery. We refer to this group as ‘‘implanted and released’’ wild smolts. 2.3.4. Wild kelts In addition to the smolt tagging, one 3-yr-old female kelt (80.5 cm FL) was acoustically tagged and released on 25 May. This kelt formed the single adult steelhead detected by the acoustic array. Three additional 2-yr-old kelts surgically implanted with internally recording archival tags (which involves a more complicated and time consuming surgical procedure) were also internally double-tagged with an acoustic tag at the time of surgery. All three kelts implanted with an archival tag appear to have died after release when they were caught in the river and eaten by eagles (three carcasses of animals with implanted archival tags

ARTICLE IN PRESS D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

were recovered near the weir the morning after release). One of these archival tags was subsequently re-implanted in a fourth kelt along with an acoustic tag. The carcass was not subsequently found nor was the acoustic tag detected by the array. All tagged animals were released at the peak of the wild smolt emigration from the Keogh River. Timing of the migration was delayed by several weeks from normal, probably because of dry weather and low water levels in the river (Ward and McCubbing, 1998).

3. Results All of the receivers on the AB Line and at the RM were recovered on 5 July, almost 6 weeks after release of the smolts. One receiver placed on the northern end of line A was not initialised and one receiver on the southern end of line B was not placed on the array at the time of deployment. Thus, a small gap of ca. 1 km occurred in the northern detection line AB centred around Doyle Island. Most of the two southern lines were not recovered. Line C was present when retrieval began on 4 July, but the acoustic release had jammed. Upon a subsequent recovery effort in August, no evidence of the release or groundline was found. Line D was located in an area of high currents. One receiver was recovered after triggering the release, but the groundline subsequently fouled and snapped. The remaining VR-2s were not initially recovered. On 8 December one of the lost VR-2s was returned after being found washed up on the beach off the southern end of Malcolm Island. This receiver was originally placed as 2nd from the western end of Line D, nearest to Malcolm Island. The single (eastern-most) receiver initially recovered from Line D recorded no fish detections up until recovery on 4 July. The second receiver which washed up on shore recorded no detections prior to 4 July, at the end of the experiment, but then recorded one detection of an implanted and held hatchery smolt on 12 July, and then detected the same smolt a total of 61 times over an 8-h period on 20 July. Because this smolt

903

was detected on the partially recovered southern line after all of the other equipments were removed, we excluded it from the analysis reported below, but comment on the observations in Section 4. 3.1. Initial movements Of the 87 tagged and released smolts, 45 were subsequently detected at the RM receiver and 25 of these were also heard on the AB Line. Fifteen smolts also were recorded only on the AB Line, giving a total of 40 fish recorded on the AB Line. The number of fish heard on each receiver is shown in Fig. 2 (note that one fish may be heard on more than one receiver). Smolts generally seem to be evenly spread along most of the detection line, but with few fish detected on the western side of Queen Charlotte Strait, nearest to the release site. None of the steelhead tagged in a parallel and simultaneous Quatsino Sound study on the west coast of Vancouver Island (Welch et al., in preparation) were detected entering Queen Charlotte Strait. One of the two acoustically tagged kelts was detected by the acoustic array. This kelt was first detected entering the ocean at RM 3 days after release, on 28 May at 1:25 AM, and then detected on the AB Line 5.7 km offshore from Vancouver Island and 15.6 km distant from RM on the following day at 9:30 PM, 42 h later. It was last detected 3 days after ocean entry by the same receiver at 7:28 AM on 30 May, 34 h after the first detection, providing 122 separate detections on that receiver. Two single detections, separated by almost 7 h, were also detected by the next farthest offshore receiver, but were closely matched with detections recorded on the first receiver; thus, the kelt spent most of this 34-h time period within a radius of about 500 m of the receiver. 3.2. Habitat use 3.2.1. Time to ocean entry (river residence) Most fish left the river relatively quickly after release at the weir, with the wild smolts being the quickest to reach the ocean (Fig. 3). For the implanted and held hatchery-reared smolts we

ARTICLE IN PRESS D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

904

Frequency

19.2

17.4

15.6

16 14 12 10 8 6 4 2 0

Wild Immediate release

1

3

4

5

6

7

9 10 11 12 13 14 15

8

4

13.8 Frequency

Hatchery implanted & held

Line B

12.0

3 2 1 0

10.2

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15

4

8.4

Hatchery Immediate release Frequency

Distance From Vancouver Is. (km)

2

6.6

2 1 0

No Data

4.8

3

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15

Number of Days

2.1

Fig. 3. Time to ocean entry for fish detected on the acoustic receiver situated off the RM. The panels break the timing of ocean entry into the three smolt treatment groups. The kelt time of 3 days to reach RM is not shown.

Line A

0.4

RM 0

10

20

30

40

Number of fish detected Fig. 2. Number of tagged steelhead detected on the AB Line relative to distance from Vancouver Island.

have used the first possible date of release, 22 May (although some or all could have been released on the 23 May). There is a statistically significant difference between the mean time that the wild smolts took to reach the ocean (2.3 days) and that of either of the hatchery groups (Mann–Whitney non-directional U-test; Pp0:01), but not between the two hatchery groups (5.7 and 4.2 days; P ¼ 0:48). Even if we assume that all the

implanted and held smolts were released on the second possible date (23 May), therefore reducing the time to the ocean by a day, this group still took substantially longer to reach the ocean than the wild smolts. 3.2.2. Duration of residence—Queen Charlotte Strait Of the 25 steelhead that were first heard at RM and then on the AB Line, none returned to be detected in the vicinity of the RM. Similarly, no fish first heard on the AB Line were subsequently heard at RM. This suggests that movement is unidirectional away from the Keogh River and out of the Strait. It is thus possible to examine how long the fish occupied Queen Charlotte Strait by calculating the time taken to detection on the AB Line. Since

ARTICLE IN PRESS D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

some fish were heard on the AB Line but not at RM we have calculated the residence time as the time from release to time of last detection on the AB Line (Fig. 4). Although it is possible that some

Wild-Immediate Release

Number of Fish

12 10 8 6 4 2 0 1

4

7

10 13 16 19 22 25 28 31 34 37 40

Number of Fish

12 Hatchery-Implanted & Held

10 8 6 4 2 0 1

4

7

10 13 16 19 22 25 28 31 34 37 40

12

Number of Fish

10 Hatchery-Immediate Release 8 6 4 2 0 1

4

7

10 13 16 19 22 25 28 31 34 37 40

905

smolts were still present in Queen Charlotte Strait south of the AB Line, only one smolt was detected on the AB Line between 14 June and the recovery of the AB Line on 7 July (3 July), 42 days after release. It seems likely that the majority of the smolts thus moved over the line from south to north and did not return. Overall residence times in the strait were short, with 78% of the smolts leaving the Strait within 1 week of release, with 2–5 days of this time spent in the river. Wild and hatchery-reared implanted and held smolts were not significantly different in the average length of time that they remained in Queen Charlotte Strait during the duration of the experiment (Mann–Whitney non-directional U-test; P ¼ 0:09), with wild smolts staying for an average of 5.5 days and the hatchery-reared smolts an average of 11.9 days. However, the detection of one additional implanted and held hatchery smolt in the vicinity of the southern line on 12 and 20 July, 52 and 60 days after release and some time after the main listening lines were recovered, increases the mean residency time for hatchery smolts to significantly longer than for the wild smolts (P ¼ 0:04). This suggests that hatchery smolts may on average have somewhat greater residence times within Queen Charlotte Strait as well as in the river, but the evidence for an extended ocean residency needs to be treated with caution. As only three immediately released hatchery-reared smolts were detected on the AB Line, it is not possible to meaningfully compare the mean length of their stay with the other groups.

12

3.3. Speed of ocean movement

Number of Fish

10

All Groups Combined 8 6 4 2 0 1

4

7

10 13 16 19 22 25 28 31 34 37 40

Days Since Release

Fig. 4. Time to departure from Queen Charlotte Strait for fish detected on the AB Line. Note that this calculation includes time spent in freshwater after release.

It is possible to calculate the speed of ocean movement of steelhead in body lengths per second from RM to time of first detection on the AB Line assuming straight line movement from receiver to receiver. None of the immediate release group of hatchery-reared smolts were heard at both RM and the AB Line. For the remaining two groups the mean speeds of movement between RM and the AB Line are shown in Fig. 5. Mean speed (7SD) according to group (wild vs. implanted and held hatchery-reared

ARTICLE IN PRESS D.W. Welch et al. / Deep-Sea Research II 51 (2004) 897–909

906 16

2.0

14

Body lengths sec-1

12

Frequency

Mean Speed of Ocean Movement

All Groups Combined

10 8 6 4 2

1.5

1.0 0.5

0

Suggest Documents