1

Movements and Mortality of Recently-Stocked Rainbow Trout in Tennessee

Reservoirs

Fisheries Report 10-07

Tennessee Technological University

and

Tennessee Cooperative Fishery Research Unit

Tomas Ivasauskas, M.S.

Tennessee Cooperative Fishery Research Unit

Department of Biology, Tennessee Technological University,

Cookeville, TN 38505 USA

Phillip W. Bettoli, Ph.D. U.S. Geological Survey, Tennessee Cooperative Fishery Research Unit

Department of Biology, Tennessee Technological University,

Cookeville, TN 38505 USA

September 2010

2

Movements and Mortality of Recently-Stocked Rainbow Trout in Tennessee

Reservoirs Fisheries Report 10-07

A Final Report Submitted To

Tennessee Wildlife Resources Agency

By

Tennessee Technological University

and

Tennessee Cooperative Fishery Research Unit

Tomas Ivasauskas, M.S.

Tennessee Cooperative Fishery Research Unit

Department of Biology, Tennessee Technological University,

Cookeville, TN 38505 USA

Phillip W. Bettoli, Ph.D. U.S. Geological Survey, Tennessee Cooperative Fishery Research Unit

Department of Biology, Tennessee Technological University,

Cookeville, TN 38505 USA

September 2010 ________________________________________________________________

Development of this report was financed in part by funds from Federal Aid in Fish and Wildlife

Restoration (Public Law 91-503) as documented in Federal Aid Project FW- 6 (TWRA Project

7307).

This program receives Federal Aid in Fish and Wildlife Restoration. Under Title VI of the Civil

Rights Act of 1964 and Section 504 of the Rehabilitation Act of 1973, the U.S. Department of

the Interior prohibits discrimination on the basis of race, color, national origin, or handicap. If

you believe you have been discriminated against in any program, activity, or facility as described

above, or if you desire further information, please write to: Office of Equal Opportunity, U.S.

Department of the Interior, Washington, D.C. 20240.

3

TABLE OF CONTENTS

Part I. Effects of Suture Material and Ultrasonic Transmitter Size on

Survival, Growth, Wound Healing, and Expulsion in Rainbow Trout

Abstract............................................................................................ 6

Introduction……………………………………………………….. 8

Methods ………………………………………………………….. 10

Results…………………………………………………………… 13

Discussion……………………………………………………….. 15

References……………………………………………………… 18

Tables and Figures ……………………………………………. 21

Part II. Movements by and Predation on Recently-stocked Rainbow Trout in

Tennessee Reservoirs

Abstract........................................................................................ 26

Introduction…………………………………………………… 28

Study Area…………………………………………………… 31

Methods ……………………………………………………… 32

Results………………………………………………………… 38

Discussion……………………………………………………. 41

References………………………………………………….. .. 44

Tables and Figures……………………………………………… 49

4

FOREWORD

This final report is based on an M.S. thesis prepared by the first author and it is presented

in two parts. Part I describes a laboratory study under controlled conditions that investigated the

best approach to implant telemetry tags into hatchery-reared rainbow trout. Part II describes the

field study component of this project, which was based in part on the results presented in Part I.

Rainbow trout transferred from a TWRA hatchery truck to a floating net pen were towed

offshore and stocked in an attempt to reduce mortality due to inshore predators.

5

ACKNOWLEDGEMENTS

Funding for this research was provided by the Tennessee Wildlife Resources Agency, the

Center for the Management, Utilization, and Protection of Water Resources at Tennessee

Technological University, and the USGS Tennessee Cooperative Fishery Research Unit. This

manuscript benefited from constructive comments on earlier drafts by J. M. Redding, T. H.

Roberts, R. S. Brown, K. Hoffman, D. Dreves, and C. Caldwell. Many thanks are extended to T.

Campbell and the staff at Dale Hollow National Fish Hatchery, S. Surgenor at Flintville State

Fish Hatchery, and M. Smith at Eagle Bend State Fish Hatchery for allowing us free-reign in

their respective hatcheries. We are especially indebted to Dr. Tom Holt, DVM, for the assistance

he provided during the necropsies we performed in Part I of this project.

6

Part 1

Effects of Suture Material and Ultrasonic Transmitter Size on Survival,

Growth, Wound Healing, and Expulsion in Rainbow Trout

Abstract.- We examined the effects of suture material (braided silk versus Monocryl) and

relative transmitter size on healing, growth, mortality, and tag retention in rainbow trout

Oncorhynchus mykiss. Fish in experiment 1 (205 - 281 mm total length [TL]; 106 - 264 g) were

implanted with Sonotronics® IBT-96-2 or IBT-96-2-E ultrasonic telemetry tags. Fish in

experiment 2 were larger (342 -405 mm TL; 520 -844 g) and were implanted with Sonotronics®

IBT-96-5 ultrasonic tags. Tag burdens for all implanted fish ranged from 1.1 - 3.4% and fish in

both studies were held at 10-15oC. At the conclusion of both experiments (65-d post-surgery),

no mortalities were observed in any of the 60 tagged fish, most incisions were completely

healed, and all fish in both experiments grew in length, although tagged fish grew slower than

control fish in experiment 1. In both experiments, fish sutured with silk expelled tags more

frequently than those sutured with Monocryl; tag expulsion was not observed until 25 - 35 d

post-surgery. Fish sutured with silk exhibited a more severe inflammatory response three weeks

post-surgery than those sutured with Monocryl. In experiment 1, the rate of expulsion was

linked to the severity of inflammation. Although braided silk sutures were applied faster than

Moncryl sutures in both experiments, knots tied with either material were equally reliable and

fish sutured with Moncryl experienced less inflammation and lower rates of tag expulsion.

7

A rainbow trout is sutured following implantation of an ultrasonic tag; note the tube

supplying water over the gills which is treated with a maintenance dose of anesthetic.

A healed incision with no inflammation 60-days post-surgery.

8

INTRODUCTION

The validity of any biotelemetry study depends on the assumptions that the attachment of

a transmitter tag does not affect behavior, growth, or survival, and that tag loss is minimal.

Summerfelt and Smith (1990) provided early technical guidelines for surgically implanting

telemetry tags; however, there are many variations on the basic technique and there is little

consensus regarding the efficacy of common modifications. Because there is substantial

interspecific variation in the anatomy, morphology, and physiology of fishes, the efficacy of

modifications to surgery techniques probably differs on a species-specific basis. The findings

and recommendations of methodological studies are outlined in several more recent reviews of

surgical implantation techniques (Jepsen et al. 2002; Bridger and Booth 2003; Mulcahy 2003;

Brown et al. 2010).

It is generally accepted that a sterile surgical field is important for deterring post-

operative infections (Wagner and Cooke 2005). The use of surgical gloves, sterile drapes, and

disinfected or sterilized tools are easily followed means to this end. Transmitter tags should also

be disinfected prior to implantation; sterilization using ethylene oxide gas or glutaraldehyde is

preferable, but both agents are extremely toxic and specific safe-handling instructions must be

followed (Burger 1994). Wildgoose et al. (2000) advised that the wound location should be

prepared by simply swabbing it to remove most of the mucus.

Sutures are utilized in tag implantation surgeries to close the wound that the transmitter

was inserted through, thereby inhibiting passive ejection of the implanted tag and maintaining

tissue contact, which aids in healing. Sutures are made of a variety of materials that may be

absorbable or nonabsorbable. Absorbable suture materials that have been used in fish surgeries

include plain or chromic surgical gut, PDS (polydioxanone monofilament), Vicryl (Ethicon®),

and Monocryl (Ethicon®). Non-absorbable suture materials include silk, nylon, polyester, and

steel; surgical staples and adhesives have also been used. Absorbable sutures tend to break down

over time (i.e., after the wound is healed; Walsh et al. 2000). Suture materials are further

classified as either monofilament or braided multifilament. Multifilament sutures are generally

regarded as easier to work with (Wagner and Cooke 2005), which can decrease the time needed

to apply sutures (Cooke et al. 2003), but multifilament sutures produce tissue drag and facilitate

bacterial wicking (Smeak 1998). Wagner et al. (2000) reported greater inflammation of the

9

suture site in rainbow trout when braided silk suture was used, as compared with absorbable or

non-absorbable monofilament sutures. Similar findings have been reported from studies utilizing

Chinook salmon Oncorhynchus tshawytscha (Deters et al. 2010), koi Cyprinus Carpio (Hurty et

al. 2002), and largemouth bass Micropterus salmoides (Cooke et al. 2003). Most (74%) fish

surgeons typically use a monofilament suture to close incisions and absorbable monofilament is

used more often than nonabsorbable monofilament (Wagner and Cooke 2005). Simple

continuous, simple interrupted, horizontal mattress, and continuous Ford interlocking sutures are

several patterns that have been used to apply sutures to fish. Wagner et al. (2000) had best

results using a simple interrupted suture pattern to close the wound on rainbow trout

Onchyrynchus mykiss.

Once the incision wound is healed, transmitters are often encapsulated by the body and

can be expelled through the site of the incision (Bunnell and Isely 1999), through another part of

the body wall (Helm and Tyus 1992; Lucas 1989), through the site of the antenna exit wound

(i.e., for radio transmitters; Bunnell and Isely 1999) or through the intestine (Chisholm and

Hubert 1985; Helm and Tyus 1992). Lucas (1989) and Jepsen et al. (2008) described and

photographed the biological processes involved with encapsulation and expulsion of a telemetry

transmitter tag. This process of encapsulation and expulsion does not seem to affect fish health

(Lucas 1989). The use of heavier or larger tags has been shown to increase the likelihood of

transmitter expulsion in trout (Chisholm and Hubert 1985).

Discrepancies in methods used for disinfection and sterilization of surgical tools, the

suture type, and relative transmitter size can inadvertently bias telemetry field studies. Also,

rates of tag expulsion have been extremely variable among studies that implanted rainbow trout,

cutthroat trout O. clarkii, brown trout Salmo trutta, and brook trout Salvelinus fontinalis (Table

1). Before beginning a telemetry study of movements and survival of rainbow trout stocked into

Tennessee reservoirs, we undertook the present study to evaluate the effects of two common

suture materials (braided silk and Monocryl1) and relative transmitter size on healing, growth,

mortality, and transmitter retention in rainbow trout.

1 The use of trade, product, industry or firm names or products is for informative purposes only and does not

constitute an endorsement by the U.S. Government or the US Geological Survey

10

METHODS

Experiment 1

Forty rainbow trout were implanted with transmitters; 2/0 Monocryl suture (Ethicon®)

and 2/0 braided silk (Ethicon®) with FS-1 reverse cutting needles were used on equal numbers

of fish. Ten fish received a sham treatment (fish were operated on but no transmitter was

implanted); 5 were sutured with each material. Ten fish served as controls. Mean total length

(TL) of all fish was 247 mm (SE=2.02) and mean weight was 187 g (SE=3.99). Fish ranged in

size from 205 mm TL and 106 g to 281 mm TL and 264 g. Nonfunctioning OEM replicas

(dummy tags) of Sonotronics® IBT-96-2 ultrasonic transmitter tags (23 mm x 7 mm, weight in

air 4.4 g, weight in water 2.4 g) and IBT 96-2E tags (30 mm x 7 mm, weight in air 4.9 g, weight

in water 2.4 g), were used in this experiment. Transmitter weight-to-body weight ratios

(hereafter referred to as ‗burdens‘) ranged from 1.7% to 3.4%. A PIT tag was injected into the

dorsal musculature of each fish and dummy transmitters were individually numbered for

identification purposes.

Treatments were applied in a rotational order to avoid bias due to possible serial

correlation (Cooke et al. 2003). The two surgeons who performed the surgeries classified

themselves as competent in skill (Wagner and Cooke 2005) but novice in experience (Cooke et

al. 2003). Both surgeons had prior experience implanting transmitters and received informal

training on proper surgical practices. Surgeon 1 had 15 years of experience conducting

implantation surgeries and had operated on more than 120 fish; whereas, Surgeon 2 had less than

one year of experience and had operated on 20 fish.

Generally accepted surgical practices were followed (Wagner and Cooke 2005).

Surgeons wore sterile gloves, and the work area and fish were covered in a sterile drape. Tags

were disinfected by soaking for at least 30 min in a bath of chlorhexadine gluconate solution and

then rinsed with sterile saline (Burger et al. 1994). Tools were sterilized using a damp-heat

autoclave before each experiment and were disinfected between surgeries by soaking for at least

30 min in the chlorhexadine gluconate solution and rinsed with sterile saline. Needles and

excess suture material were discarded after use on a single fish.

Trout receiving an implanted transmitter were anesthetized in a 75 ppm solution of tricaine

methanesulfonate (MS-222) at a temperature of 10oC for approximately 3 min until ventilation

11

became slow and the fish began to exhibit a total loss of equilibrium (Summerfelt and Smith

1990). They were then measured (TL, mm), weighed (g), and injected with the PIT tag. During

surgery, trout were restrained ventral side up in a U-shaped trough lined with a water-repellant

non-abrasive sterile drape and were continuously provided with anesthesia by pumping a 35 ppm

MS-222 solution over the gills. The incision site was prepared by gently swabbing the epithelial

mucous with an untreated sterile cotton swab (Wildgoose 2000). The tag was inserted through a

small ventral incision along the linea alba. The incision was closed using two interrupted sutures

(surgeon‘s knots with 2x1x1 throws for braided silk or 2x1x1x1 throws for Monocryl). Each

step of the surgical procedure was timed.

Fish receiving a sham surgery treatment were handled and treated in the same manner as

those that received an internal tag; they were anesthetized, measured, weighed, injected with a

PIT tag, and underwent surgery, but no transmitter was implanted. Control fish were

anesthetized, weighed, measured, and injected with a PIT tag.

Trout were held in an indoor concrete flow-through raceway maintained at 10-15oC at

Dale Hollow National Fish Hatchery. Fish were fed once daily. The raceway was examined

daily for dead fish and shed tags. Twenty-one d following surgery, all fish were netted and

sedated in a 75 ppm solution of MS-222. They were scanned for the PIT tag, weighed,

measured, and the incision site was assigned a score that reflected the severity of

macroinflamation (Wagner 1999; Table 2).

All trout were removed 65 d post-surgery and euthanized in a lethal dose of MS-222.

Fish were individually weighed and measured and the overall condition of each fish was noted

(Cooke et al. 2003). Necropsy was conducted and the degree of tag encapsulation, internal

attachment points of the fibrous capsule, presence of hemorrhaging, and any other macroscopic

physiological responses to the treatment were noted. For fish that expelled tags, the expulsion

location was determined from external scars or tunneling within the fibrous capsule.

Experiment 2

Larger IBT-96-5 ultrasonic transmitters (36 mm x 11 mm, weight in air 9.1 g, weight in

water 4.1 g) were used in this experiment and the fish were larger than those in Experiment 1.

Mean TL of fish in Experiment 2 was 372 mm (SE=3.25) and mean weight was 660 g

(SE=16.07). Fish ranged in size from 342 mm and 520 g to 405 mm and 844 g. Transmitters

represented burdens of 1.1% to 1.7% of the weight of each fish. Twenty fish were implanted

12

with transmitters; 10 received Monocryl and 10 received silk sutures. Five fish were sham

tagged; two received Monocryl and three received silk sutures. Five fish served as controls;

however, one control fish jumped out of the tank and died during the first 21 d. Surgical

procedures in this experiment were nearly identical to those described in Experiment 1. The only

exception was that a larger incision was necessary for transmitter insertion; accordingly, three

interrupted sutures were used to close the incision.

Fish in this experiment were held at a facility at Tennessee Technological University in a

round 1,000-L tank that was part of a 2,250-L recirculating aquaculture system maintained at 10-

15oC. Water quality was checked daily and partial water changes were conducted at least three

times per week. Fish were fed once daily. As with Experiment 1, the tank was checked daily for

expelled transmitters and fish were sedated and removed for observation after 21 d and were

necropsied after 65 d.

Statistical Analysis

The results of each experiment were analyzed separately to avoid the possibly

confounding effects of holding different sizes of fish in different holding facilities. All statistical

analyses were performed using SAS programs. The times required to apply sutures were

compared between surgeons and suture materials using t-tests. Because fish were identified with

PIT tags, it was possible to estimate individual growth rates (G; %·d-1

) between days 0 and 21

and over the duration of the study using the equations:

t

)TL(log)(TLlog 100G initialfinale

Lengthe and

t

)Weight(log)(Weightlog 100G initialfinale

Weighte .

Mean growth rates were compared among implanted, sham, and control fish using

ANOVA and Tukey‘s HSD test. Growth was compared between implanted fish sutured using

Monocryl and braided silk using a t-test. Incision scores for Monocryl and silk were compared

using the Kruskall-Wallace test. Simple linear regression was used to test the null hypothesis

that growth was not affected by transmitter burden. Logistic regression was used to test whether

transmitter expulsion over 65 d was related to burden. Logistic regression was also used to

determine if expulsion was related to the 21-d incision score.

13

RESULTS

Experiment 1

Silk sutures were applied faster than Monocryl (t-test; P=0.0002); mean suturing time

was 1:42 (SE=0:04) for silk and 2:38 (SE= 0:11) for Monocryl. Suturing time with both

materials did not vary with surgeon (t-test; P=0.0783). The time to complete surgery (i.e.,

incision, transmitter insertion, and suturing) averaged 2:56 (SE = 0:04) when silk was used and

3:54 (SE=0:11) when Monocryl was used. No mortality or morbidity was observed during this

experiment.

After 21 d, there was no significant difference in growth in weight (ANOVA; df=2, 57;

F=2.08; P=0.1346) or in length (ANOVA; df=2, 56; F=0.38; P=0.6826) as a result of transmitter

implantation or the surgical process. Among implanted fish, there was no difference in growth

in weight (t-test; P=0.1043) or in length (t-test; P=0.1553) due to suture material (Table 3). After

65 d, growth differences manifested themselves among treatment groups in terms of length

(ANOVA; df=2, 56; F=5.47; P=0.0067) and to a lesser extent, weight (ANOVA; df=2,56;

F=2.86; P=0.0659). Implanted fish grew slower than control fish and there was no difference in

growth between sham and control fish. The mean growth rate (G) for implanted fish was 77%

(by weight) and 70% (by length) of that of non-implanted fish (Table 3). Among implanted fish,

suture material had no effect on growth in weight (t-test; P=0.1386) or length (t-test; P=0.6435)

and growth in weight and length was independent of transmitter burden (linear regression; P ≥

0.4875).

The 21-d incision score was better for fish sutured with Monocryl than with silk

(Kruskall-Wallace; χ2=10.94; df=1; P=0.0009) and all wounds were granulated to some extent.

No sutures were lost during the 21 d post-surgery.

During the 65-d course of the study, transmitter expulsion occurred in 45% of fish

sutured using silk and 25% of fish sutured using Monocryl. Expulsion was first observed after

28 d and 35 d for fish sutured with silk and Monocryl, respectively (Figure 1). The likelihood of

transmitter expulsion was linked to the 21-d incision score (logistic regression; df=1; χ2=4.7845;

P=0.0287); fish with better scores (i.e., less inflammation) tended to expel transmitters less

frequently. The likelihood of transmitter expulsion was directly related to transmitter burden for

fish sutured with silk (logistic regression; df=1; χ2=3.9017; P =0.0482); however, this trend was

14

not observed for fish sutured with Monocryl (logistic regression; df=1; χ2=0.3872; P =0.5338).

Transmitters were expelled most frequently through the lateral body wall (39%) or the

granulated incision site (31%). Trans-intestinal passage was the most probable cause of

expulsion in 15% of the cases of transmitter expulsion; the expulsion location was

indeterminable for the other 15% of expulsions.

At 65 d, all transmitters that were not expelled were encapsulated. Fibrous capsules were

held in place via connective tissue to one (74% of fish) or more points of attachment.

Attachment points included the parietal peritoneum (65%), granulation tissue at the incision site

(15%), viscera (15%), intestine (15%), stomach (5%), and the testes (5%). Tunneling at one or

both ends of the fibrous capsules was apparent in most fish. Severe transmigration through the

lateral abdominal wall was observed in 12% of fish, suggesting imminent transmitter expulsion.

Macroscopic petechial or diffuse hemorrhaging on the capsule, connective tissue, or adjacent

musculature occurred in 39% of fish. Hyperaemia was also common near the capsule. Fungus

was visible on the braided silk sutures applied to one fish.

Experiment 2

As in Experiment 1, silk sutures were applied faster than Monocryl (t-test; P=0.0013);

mean suturing time was 1:52 (SE=0:06) for silk and 2:35 (SE=0:11) for Monocryl and did not

differ by surgeon (t-test; P=0.0539). Surgery times averaged 2:42 (SE=0:14) for silk and 3:48

(SE=0:10) for Monocryl. No mortality or morbidity resulted from the surgical procedure.

After 21 d, there was no difference in growth in weight (t-test; P=0.9038) or in length (t-

test; P=0.1685) due to transmitter implantation or the surgical process. Among implanted fish,

there was no difference in growth in weight (t-test; P=0.8648) or in length (ANOVA; df=1, 17;

F=0.03; P=0.8576) due to suture material (Table 4). After 65 d, fish in all treatment groups grew

in length but all fish lost weight; however, there was no difference in growth in weight

(ANOVA; df=2, 25; F=1.53; P=0.2364) or in length (ANOVA; df=2, 25; F=0.65; P=0.5330)

among implanted, control, and sham fish. Among implanted fish, suture material had no effect

on growth in weight (t-test; P=0.5008; Table 4) or length (t-test; P=0.8004). Growth in weight

was negatively, but weakly, related to transmitter burden (linear regression; P=0.0696); no

relationship was evident between tag burden and growth in length (linear regression; P=0.5958).

Incision scores after 21-d were generally lower (i.e., better) for fish sutured with

Monocryl than for those sutured with silk (Kruskall-Wallace; df=1; χ2=2.8397; P=0.0920) and all

15

wounds were granulated to some extent. At 21 d post-surgery, there was no difference in the

number of sutures that came undone, were ripped out, or were expelled (t-test; P=0.4073); two

fish sutured with Monocryl and four sutured with silk exhibited at least one missing suture.

During the 65 d course of experiment 2, transmitter expulsion occurred in 50% of fish

sutured with silk but was not observed for any fish sutured using Monocryl. Expulsion was first

observed after 25 d (Figure 2). The probability of a fish expelling a transmitter was not

influenced by the 21-d incision score (logistic regression; df=1; χ2=1.16; P = 0.2807) or

transmitter burden (logistic regression; df=1; χ2=0.2782; P=0.5979). All expulsions in this

experiment occurred through the granulated tissue at the incision location. Dehiscence (i.e., a

rupture of the incision site) was evident in one fish that expelled its transmitter.

At 65 d, all transmitters except one were encapsulated. The non-encapsulated transmitter

was lodged between the intestine and lateral wall, posterior to the anus. All encapsulated

transmitters were held in place by connective tissue from the fibrous capsule to one (72% of

observations) or more points of attachment. Attachment points included the parietal peritoneum

(47%), granulation tissue at the incision site (14%), viscera (34%), intestine (7%), and the spleen

(7%). Tunneling at one or both ends of the capsule was apparent. Macroscopic petechial or

diffuse hemorrhaging on the capsule, connective tissue, or adjacent musculature occurred in 47%

of fish. Hyperaemia was also common. One fish exhibited an enterocutaneous fistula—likely

resulting from the accidental trapping of omentum in the wound during the surgical process.

Fungus was visible on the braided silk sutures applied to 2 fish.

DISCUSSION

Although some transmitters were shed, the results reported herein indicate that rainbow

trout can be successfully implanted with ultrasonic transmitters, especially when the tag burden

is low and Monocryl sutures are used. At the conclusion of both experiments, no mortality was

observed, most incisions were completely healed, and all implanted fish grew in length, although

tagged fish grew slower than control fish in experiment 1.

These experiments underscored the importance of good surgical technique. The presence

of competent assistants facilitated rapid implantation of transmitters. No mortality or morbidity

resulted from the surgical procedure or post-operative infections. Mortality in other studies is

usually attributed to internal injury caused during the surgical process (Cooke et al. 2003) or

16

post-operative fungal (Adams et al. 1998; Brown et al. 2006) or bacterial infection (Chisholm

and Hubert 1985). The pre-operative condition of the fish should also be considered before

performing surgeries because stress is known to compromise a fish‘s ability to heal and fight

infection (Walters and Plumb 1980; Wendelaar Bonga 1997). All of the hatchery fish used in

these experiments were in good health and were handled with care in order to avoid injury and

minimize stress.

Fish did not begin to shed transmitters until 25-35 d post-surgery. Because all incision

wounds were granulated by that time, it was presumed that all shed transmitters had been

encapsulated and actively expelled. These findings suggest that in field studies, tagged hatchery

rainbow trout should be released into the wild after a short recovery period of no more than a few

hours or days. Doing so would increase the number of observations for the telemetered cohort

and minimize the confounding effects of transmitter expulsion.

Experiment 1 provided evidence that tagging slowed the growth of rainbow trout in a

hatchery setting. Because growth rates never differed significantly between control and sham-

surgery fish, it is unlikely that the surgical procedure alone affected growth and this finding is

comparable to other studies (Adams et al. 1998; Brown et al. 2006). One possible explanation

for slower growth is the reduced expandability of the stomach when a tag is present in the caecal

cavity, which may limit the amount of food a trout is able to consume in one feeding bout.

Differential growth of implanted fish versus non-implanted fish was not observed in Experiment

2, when transmitter burdens were lower. Rainbow trout were fed to satiation once per day in

both experiments, whereas trout in natural settings theoretically can feed on smaller rations ad

libitum throughout the day; thus, it is possible that no differential growth occurs in the wild.

Both experiments demonstrated that Monocryl performed better than braided silk in

terms of surgical outcomes. In both experiments, fewer (if any) expulsions were observed for

fish sutured with Monocryl. The encapsulation-expulsion process was accelerated in fish

exhibiting a more severe inflammatory response and this and other studies (Wagner et al. 2000;

Hurty et al. 2002; Deters et al. 2010) have shown that braided silk elicits a more severe

inflammatory response than monofilament suture materials. Additionally, braided silk sutures

provided a superior medium for fungal growth. Although silk suture material is easier to handle

and hastens the surgical process, no transmitter losses for either experimental group were

attributed to passive ejection through an open incision.

17

The commonly cited ‗2% rule‘ cautions against implanting transmitters that represent

burdens greater than 2% (Winter 1983). Results from this and other studies indicate that strict

adherence to the 2% rule may be unnecessary. Critical swimming speeds (Brown et al. 1999)

and swimming endurances (Thorstad et al. 2001) of salmonids implanted with transmitter

burdens greater than 2% did not differ from those of control fish. In the present study, the

negative consequences of imposing tag burdens near or exceeding 2% were limited to slower

growth in length after 65 d among the smaller fish (i.e., with higher burdens) in experiment 1 and

a higher probability of tag expulsion with increasing burden in experiment 1, but only when silk

sutures were used (which we do not advocate).

Despite maintaining a sanitary surgical field, post-operative fish in this and other

experiments are released into unsterile environments. The open wound provides a pathway for

water to enter, bringing with it contamination and possible infection. Topical antiseptics are

routinely implemented in mammalian surgeries to deter infection; however, due to the unique

structure of fish epithelium, antiseptics typically used in mammalian surgeries may not be

appropriate for fish. Hart and Summerfelt (1975) noted that the topical use of benzalkonium

chloride irritated the skin of flathead catfish Pylodictis olivaris. Briggs (1995) found that alcohol

used as a cutaneous disinfectant disrupted the mucus layer of rainbow trout and Wagner et al.

(1999) noted that preparation of the incision sites with a povidone–iodine antiseptic did not

improve wound healing in rainbow trout. Future research for improving the efficacy of

telemetry transmitter implantation should focus on reducing inflammation at the wound site and

minimizing expulsion rates.

18

REFERENCES

Adams, N. S., D. W. Rondorf, S. D. Evans, and J. E. Kelly. 1998. Effects of surgically and

gastrically implanted radio transmitters on growth and feeding behaviour of juvenile

chinook salmon. Transactions of the American Fisheries Society 127:128–136.

Baldwin, C. M., D. A. Beauchamp, and C. P. Gubala. 2002. Seasonal and diel distribution and

movement of cutthroat trout from ultrasonic telemetry. Transactions of the American

Fisheries Society 131: 143–158.

Bridger, C. J., and R. K. Booth. 2003. The effects of biotelemetry transmitter presence and

attachment procedures on fish physiology and behavior. Reviews in Fisheries Science 11:

13-34.

Briggs, C. T. 1995. The metabolic costs of activity of free swimming, adult rainbow trout

(Oncorhynchus mykiss) estimated by electromyogram telemetry. Master‘s thesis.

University of Calgary, Calgary, Alberta.

Brown, R. S., S. J. Cooke, W. G. Anderson, and R. S. McKinley. 1999. Evidence to challenge

the ―2% rule‖ for biotelemetry. North American Journal of Fish Management 19:867-

871.

Brown, R. S., D. R. Geist, K. A. Deters, and A. Grassell. 2006. Effects of surgically implanted

acoustic transmitters >2% of body mass on the swimming performance, survival and

growth of juvenile sockeye and Chinook salmon. Journal of Fish Biology 69:1626–1638.

Brown, R. S., S. J Cooke, G. N. Wagner, M. B. Eppard. 2010. Methods for surgical

implantation of acoustic transmitters in juvenile salmonids: a review of literature and

guidelines for technique. Report to U.S. Army Corps of Engineers, Portland District.

Bunnell, D. B., J. J. Isely, K. H. Burrell, and D. H. VanLear. 1998. Diel movement of brown

trout in a southern Appalachian river. Transactions of the American Fisheries Society

127: 630–636.

Bunnell, D. B., and J. J. Isely. 1999. Influence of temperature on mortality and retention of

simulated transmitters in rainbow trout. North American Journal of Fisheries

Management 19: 152–154.

Burger, B., D. DeYoung, and D. Hunter. 1994. Sterilization of implantable transmitters.

Telonics Quarterly 7: 3-4.

Burrell, K. H., J. J. Isely, D. B. Bunnell, D. H. Van Lear, and C. A. Dolloff. 2000. Seasonal

movement of brown trout in a Southern Appalachian River. Transactions of the American

Fisheries Society 129: 1373–1379.

Chisholm, I. M., and W. A. Hubert. 1985. Expulsion of dummy transmitters by rainbow trout.

Transactions of the American Fisheries Society 114: 766–767.

Cooke, S. J., B. D. S. Graeb, C. D. Suski, and K. G. Ostrand. 2003. Effects of suture material on

incision healing, growth and survival of juvenile largemouth bass implanted with

miniature radio transmitters: case study of a novice and experienced fish surgeon. Journal

of Fish Biology 62: 1366-1380.

19

Deters K. A., R. S. Brown, K. M. Carter, J. W. Boyd, M. B. Eppard, and AS. G. Seaburg. 2010.

Performance assessment of suture type, water temperature, and surgeon skill in juvenile

Chinook salmon surgically implanted with acoustic transmitters. Transactions of the

American Fisheries Society 139: 888-899.

Garrett, J. W., and D. H. Bennett. 1995. Seasonal movements of adult brown trout relative to

temperature in a coolwater reservoir. North American Journal of Fisheries Management

15: 480–487.

Hansbarger, J. L. 2005. Trout movement and habitat use in the Upper Shavers Fork of the Cheat

River. M. S. Thesis. Wildlife and Fisheries Department, West Virginia University,

Morgantown, West Virginia.

Hart, L. G., and R. C. Summerfelt. 1975. Surgical procedures for implanting ultrasonic

transmitters into flathead catfish (Pylodictis olivaris). Transactions of the American

Fisheries Society 104: 56-59.

Helm, W. T., and H.M. Tyus. 1992. Influence of coating type on retention of dummy

transmitters implanted in rainbow trout. North American Journal of Fisheries

Management 12: 257–259.

Hurty, C. A., D. C. Brazik, J. M. Law, K. Sakamoto, and G. A. Lewbart. 2002. Evaluation of

the tissue reactions in the skin and body wall of koi (Cyprinus carpio) to five suture

materials. The Veterinary Record 151: 324–328.

Jepsen, N., A. Koed, E. B. Thorstad, and E. Baras. 2002. Surgical implantation of telemetry

transmitters in fish: how much have we learned? Hydrobiologia 483: 239-248.

Jepsen, N., J. S. Mikkelson, and A. Koed. 2008. Effects of tag and suture type on survival and

growth of brown trout with surgically implanted telemetry tags in the wild. Journal of

Fish Biology 72: 594-602.

Lucas, M.C. 1989. Effects of implanted dummy transmitters on mortality, growth and tissue

reaction in rainbow trout, Salmo gairdneri Richardson. Journal of Fish Biology 35: 577-

587.

Martin, S. W., J. A. Long, and T. N. Pearsons. 1995. Comparison of survival, gonad

development, and growth between rainbow trout with and without surgically implanted

dummy radio transmitters. North American Journal of Fisheries Management 15: 494–

498.

Mulcahy, D. M. 2003. Surgical implantation of transmitters into fish. ILAR Journal 44: 295-306.

Smeak, D. D. 1998. Selection and use of currently available suture materials and needles. Pages

19–26 in M. J. Bojrab, S. W. Crane, and S. P. Arnoczky, editors. Current techniques in

small animal surgery, 4th edition. Williams and Wilkins, Baltimore.

Summerfelt, R. C., and L. S. Smith. 1990. Pages 213-272 in C. B. Schreck and P. B. Moyle,

editors. Methods for fish biology. American Fisheries Society, Bethesda, Maryland.

Thorstad, E. B., F. Okland, and T.G. Heggberget. 2001. Are long term negative effects from

external tags underestimated? - Fouling of an externally attached telemetry transmitter.

Journal of Fish Biology 59: 1092-1094.

20

Wagner, G. N. 1999. Investigation of effective fish surgery techniques. M. S. Thesis, Zoology

Department, University of Guelph, Ontario, Canada.

Wagner, G. N., and S. J. Cooke. 2005. Methodological approaches and opinions of researchers

involved in the surgical implantation of telemetry transmitters in fish. Journal of Aquatic

Animal Health 17: 160-169.

Wagner, G. N., E. D. Stevens, and C. Harvey-Clark. 1999. Wound healing in rainbow trout

Oncorhynchus mykiss following surgical site preparation with a povidone-iodine

antiseptic. Journal of Aquatic Animal Health 11: 373-382.

Wagner G. N., E. D. Stevens, and P. Byrne. 2000. Effects of suture type and patterns on surgical

wound healing in rainbow trout. Transactions of the American Fisheries Society 129:

1196–1205.

Walsh, M. G., K. A. Bjorgo, and J.J. Isely. 2000. Effects of implantation method and temperature

on mortality and loss of simulated transmitters in hybrid striped bass. Transactions of the

American Fisheries Society 129: 539-544.

Walters G. R. and J. A. Plumb. 1980. Environmental stress and bacterial infection in channel

catfish, Ictalurus punctatus Rafinesque. Journal of Fish Biology 17: 177-185.

Wendelaar Bonga, S.E. 1997. The stress response in fish. Physiological Reviews 77:

591–625.

Wildgoose, W. H. 2000. Fish surgery: an overview. Fish Veterinary Journal 5: 22–36.

Winter. J. D. 1983. Underwater biotelemetry. Pages 371–395 in L. A. Nielsen and D. L. Johnson,

editors. Fisheries techniques. American Fisheries Society, Bethesda, Maryland.

Zale, A. V., C. Brooke, and W. C. Fraser. 2005. Effects of surgically implanted transmitter

weights on growth and swimming stamina of small adult westslope cutthroat trout.

Transactions of the American Fisheries Society 134: 653-660.

21

Table 1. Mortality and expulsion rates reported for cutthroat trout, rainbow trout, and brown trout implanted with telemetry tags and 1

sutured using various materials (BS = braided silk, AM = absorbable monofilament, NAM = non-absorbable monofilament). An 2

asterisk denotes estimates where mortality and expulsion rates were combined. 3

Source Expulsion (%) Mortality (%) Species Study type Suture type

Baldwin et al. 20021 11 66 Cutthroat Field Not reported

Baldwin et al. 20022 20 20 Cutthroat Field Not reported

Bunnell & Isely 19993 13 20 Rainbow Laboratory BS

Bunnell & Isely 19994 19 20 Rainbow Laboratory BS

Bunnell & Isely 19995 27 7 Rainbow Laboratory BS

Bunnell et al. 1998 * 43* Brown Field AM, NAM

Burrell et. al 2000 * 50* Brown Field AM

Chisholm & Hubert 1985 59 27 Rainbow Laboratory BS

Garrett & Bennett 1995 * 20* Brown Field Not reported

Hansbarger 2005 * 3* All three Field Not reported

Helm & Tyus 19926 3 0 Rainbow Laboratory NAM

Helm & Tyus 19927 13 0 Rainbow Laboratory NAM

Helm & Tyus 19928 40 0 Rainbow Laboratory NAM

Jepsen et al. 2008 14 n/a Brown Field AM

Jepsen et al. 2008 31 n/a Brown Field NAM

Lucas 1989 14 5 Rainbow Laboratory BS

Martin 1995 0 0 Rainbow Laboratory AM

Zale et al. 2005 4 10% Cutthroat Laboratory Staples

1 Observed during summer;

2 Observed during fall;

3 Fish held at 10

oC;

4 Fish held at 15

oC;

5 Fish held at 20

oC;

6 Tag coated with 4

beeswax; 7 Tag coated with paraffin;

8 Tag coated with silicon.

5

22

Table 2. Scores used to describe the extent of macroinflammation occurring at the wound and 6

sutures, as presented by Wagner (1999). Lower scores represent a less extreme inflammatory 7

response. 8

Score Description

0 incision completely closed;, no inflammation

1 incision closed; some inflammation along incision site

2 incision held in proximity, but part not completely closed as edges still

slide; little to moderate inflammation

3 incision held in proximity, but edges slide if fish moves; moderate

inflammation

4 incision partially opened at one end or middle; moderate to high

inflammation

5 More than 50% of wound open; moderate to high inflammation along

wound edges

6 completely open wound; moderate to high inflammation along edges

23

Table 3. Growth rates (G; %·d-1

) and associated variances for rainbow trout implanted with IBT 96-2 transmitters calculated over the 9

first 21 d and 65 d post-surgery. 10

21 day 65 day

N weightG Variance lengthG Variance weightG Variance lengthG Variance

Implanted, Monocryl 20 0.422 0.417 0.218 0.064 0.701 0.112 0.173 0.007

Implanted, Silk 20 0.719 0.220 0.131 0.008 0.774 0.040 0.183 0.004

Sham 10 0.712 0.573 0.212 0.022 0.857 0.133 0.243 0.004

Control 10 0.994 0.206 0.212 0.004 0.873 0.039 0.282 0.006

Table 4. Growth rates (G; %·d-1

) and associated variances for rainbow trout implanted with IBT 96-5 transmitters calculated over the 11

first 21 d and 65 d post-surgery. 12

21 day 65 day

N weightG Variance lengthG Variance weightG Variance lengthG Variance

Implanted, Monocryl 10 0.019 0.010 0.037 0.002 -0.121 0.003 0.027 0.0004

Implanted, Silk 10 0.027 0.012 0.034 0.001 -0.136 0.002 0.025 0.0003

Sham 5 -0.01 0.006 0.054 0.00001 -0.108 0.0001 0.037 0.00003

Control 4 0.051 0.003 0.013 0.003 -0.088 0.002 0.027 0.0002

24

Days post-surgery

0 10 20 30 40 50 60

Pro

port

ion R

eta

inin

g T

ag

0.0

0.2

0.4

0.6

0.8

1.0

day vs DHH (silk, IBT96-2)

day vs DHH (MC, IBT96-2)

Figure 1. Retention of Sonotronics IBT 96-2 ultrasonic transmitter tags for 205-281 mm TL

rainbow trout sutured using braided silk and Monocryl suture materials.

Monocryl

Silk

25

Days post-surgery

0 10 20 30 40 50 60

Pro

port

ion R

eta

inin

g T

ag

0.0

0.2

0.4

0.6

0.8

1.0

day vs % retained Ptrailer (silk; IBT96-5)

day vs Col 5

Figure 2. Retention of Sonotronics IBT 96-5 ultrasonic transmitter tags for 342-405 mm TL

rainbow trout sutured using braided silk and Monocryl suture materials.

Monocryl

Silk

26

Part II

Movements by and Predation on Recently-stocked Rainbow Trout in

Tennessee Reservoirs

Abstract.-Hatchery-raised rainbow trout (n = 44; mean total length = 235 mm) were implanted

with Sonotronics® IBT-96-2 ultrasonic transmitter tags and stocked into Dale Hollow Lake on 7

January and 6 March 2009. Fish were tracked at least once per week for eight weeks to describe

post-stocking dispersal rates, movements, and habitat use (cove versus main-channel habitat).

Dispersal and movement followed a 3-stanza pattern characterized by rapid movement away

from each stocking site during the first week, relatively little movement during the following

three weeks, and a resumption of large movements during the final four weeks that fish were

tracked. Rainbow trout exhibited a strong affinity for embayments. Telemetered trout stocked in

March exhibited lower mortality (instantaneous weekly mortality = 0.027) than those stocked in

January (instantaneous weekly mortality = 0.062) during the first eight weeks post-stocking.

Diets of potential predators in Dale Hollow, Watauga, and South Holston Lakes were examined.

Walleye Sander vitreus , smallmouth bass Micropterus dolomieu, largemouth bass M. salmoides,

spotted bass M. punctulatus and holdover rainbow trout all preyed on recently stocked rainbow

trout. Smallmouth bass and walleye differentially consumed smaller rainbow trout and the

probability of consuming a trout was positively related to predator size. Walleye were more

likely to feed during January than March. Mortality of trout due to walleye predation over the

first eight weeks post-stocking was estimated to be 2.2 to 2.9 times higher for fish stocked in

January than for fish stocked in March.

27

A hydrophone is used to detect the ultrasonic signal of tagged rainbow trout

recently stocked into Dale Hollow Lake.

Walleye and other suspected predators of stocked rainbow trout were sampled in three

Tennessee reservoirs to describe their diets and frequency of trout consumption.

28

INTRODUCTION

In an effort to cater to the variety of angler-types that pursue rainbow trout

(Onychorhynchus mykiss, Irwin strain) in the state of Tennessee, the Tennessee Wildlife

Resources Agency (TWRA) routinely stocks rainbow trout in tailwaters, free-flowing streams,

urban ponds, and large reservoirs. The management goals for each of these programs are distinct

and are largely dictated by the biology of the species and the associated challenges presented by

the receiving habitats (Fiss and Habera 2006).

Dispersal and movement patterns of rainbow trout vary greatly according to habitat.

Rainbow trout in small lotic systems often exhibit little movement (Cargill 1980; Mellina et al.

2005; Simpson 2006), whereas trout in larger lotic systems (Bettinger and Bettoli 2000; Runge et

al. 2008) and lentic systems (James and Kelso 1995; Warner and Quinn 1995; Lindberg et al.

2009) tend to move greater distances and at a more rapid rate. Post-stocking dispersal rates have

been associated with the catchability of recently stocked trout (Cresswell and Williams 1982;

Helfrich and Kendall 1982; Baird et al. 2006). Movement rates have also been linked to

energetic demands (Briggs and Post 1997; Cooke et al 2004).

In lentic systems, rainbow trout are most commonly observed in littoral areas when

temperatures in the epilimnion do not exceed their thermal preferences (James and Kelso 1995;

Warner and Quinn 1995; Baldwin et al. 2000). Tabor and Wartsbaugh (1991) observed that

rainbow trout in lentic ecosystems are cover-oriented. Higher electrofishing catch rates of

rainbow trout in cove habitats compared to main-channel habitats indicated differential use of

these macrohabitats in Tennessee reservoirs (Bergthold and Bettoli 2009).

The TWRA stocks approximately 190,000 rainbow trout in seven large reservoirs in

Tennessee: Dale Hollow, South Holston, Watauga, Fort Patrick Henry, Calderwood, Chilhowee,

and Tellico (about 24,300 ha total). These reservoirs provide a year-round supply of well-

oxygenated cold water, thereby permitting over-summer survival. Natural reproduction is

negligible and populations must be maintained through stocking. Stocking is conducted during

the winter when reservoirs are isothermal. Despite such efforts, the return to the creel is low

(e.g., 7%, by number in Dale Hollow Lake; Bettoli 1996), and average daytime catch rates are

also generally low (e.g., 0.13 fish per hour in South Holston Lake in 2006; Malvestuto and Black

2007). Rainbow trout mortality due to stocking stress reportedly is minimal (Barwick 1985);

29

however, predation has been implicated as an important source of mortality for stocked trout in

reservoirs (Talley 1976; McMahon and Bennett 1996; Hyvarinen and Vehanen 2004). In

Tennessee reservoirs, potential predators include walleye Sander vitreus, black bass Micropterus

spp., and catfish F. Ictaluridae.

Rainbow trout are an ideal prey item for larger reservoir predators because of their soft

rays and fusiform body shape, physical features that result in prey-selectivity in some

piscivorous fishes (Wahl and Stein 1988). Also, hatchery-reared trout may be at a behavioral

disadvantage relative to wild trout in terms of surviving the challenges of the natural

environment (Bachman 1984). The lack of natural selection in hatchery-reared trout reduces

fitness (Miller 2004) and causes an overall loss of genetic variability (Vuorinin 1984). The

rigors of the hatchery environment may result in physical anomalies such as missing paired fins,

which are used in swimming to regulate body pitch and brake forward movement (Alexander

1974). Although experimental data indicated that fin loss did not affect trout survival in a low-

gradient stream, fin loss could affect survival in more perilous or demanding habitats (Heimer et

al. 1985). Furthermore, rainbow trout have not evolved specific innate avoidance or defense

mechanisms for most predators in reservoir ecosystems because they do not naturally coexist.

Predation on stocked fish is most intense soon after stocking (Buckmeier and Betsill

2002). Talley (1976) found that predation on stocked fingerling trout in a reservoir was greatest

during the first 10 d after stocking. Upon introduction to a structurally complex reservoir habitat

(compared with hatchery raceways or rearing ponds), stocked fish are naive of its intricacies and

are immediately at a higher risk to predation (Schlechte et al. 2005). Fish are usually stocked in

high densities from relatively few points and the resulting aggregation may entice a feeding

frenzy among predators (Johnson and Ringler 1998; Jepsen et al. 1998). Additionally, the

abundance of other prey species may be low during the winter stocking period; the availability of

other prey has been found to buffer predation on trout (Talley 1976; Johnson and Rakoczy 2004).

Stocked rainbow trout can represent a notable portion of walleye diets in systems where

they coexist (Baldwin et al. 2003). Walleye in Tennessee reservoirs stocked with trout generally

exhibited higher relative weights than walleye in systems without trout (Vandergoot 2001). In

Wyoming reservoirs, there was a positive correlation between walleye relative weights and

fingerling trout stocking densities (Marwitz and Hubert 1997). During the 1970s, the

introduction of nonnative walleye to impoundments on the North Platte River system, Wyoming,

30

had a detrimental effect on what was once considered a blue-ribbon trout fishery supported

through fingerling stocking. In fact, the resulting trout population declines in Pathfinder

Reservoir were so drastic that the prospect of maintaining a viable fishery based on stocked trout

was uncertain (McMillan 1984).

Black bass also prey on trout stocked in reservoirs. Talley (1976) found evidence that

largemouth bass Micropterus salmoides, smallmouth bass M. dolomieu, and spotted bass M.

punctulatus in Dale Hollow Lake, Tennessee, fed opportunistically on stocked fingerling trout.

In one Wisconsin reservoir, small (165-277 mm total length [TL]) smallmouth bass preyed

extensively on stocked fingerling trout, often to satiation. After six weeks, however, predation

by smallmouth bass was deemed negligible (Avery 1975). The author postulated that lower

consumption of trout could be attributed to the growth of the trout (which rendered them less

vulnerable to predation) or to their decline in availability.

Managers can increase survival of hatchery fish by stocking larger individuals, which

hold refuge from predators by virtue of their size (Lundvall et al. 1999). For many piscivores,

capture success decreases monotonically with increasing prey size (Fuiman and Magurran 1994).

Several studies identified a strong negative relationship between trout size at stocking in

reservoirs and mortality as a result of predation (Flinders and Bonar 2008; Cunningham and

Anderson 1992). In a controlled experiment, adult walleye (483-533 mm TL) were able to ingest

trout as large as 229 mm TL, but preferred smaller prey (Yule et al. 2000). In Pathfinder

Reservoir, Wyoming, a successful put-grow-take stocking program was established only by

ceasing the practice of stocking fingerling trout and instead stocking sub-catchable trout at a size

exceeding the walleye‘s physical capability to ingest them (McMillan 1984). Despite the

escalating cost of rearing larger fish, when mortality is taken into account the actual cost of

adding trout to the creel can be drastically reduced by stocking larger fish (Wiley et al. 1993;

Walters et al. 1997). To fulfill its goal of optimizing the use of trout hatchery resources, the

TWRA should pursue cost-effective management activities that will reduce predation on stocked

trout in reservoirs

The objectives of this study were to (1) describe post-stocking movements and dispersal

by rainbow trout in Dale Hollow Lake; (2) describe their use of habitat (cove versus main-

channel habitat) during the first two months post-stocking; (3) utilize telemetry data to estimate

mortality of recently stocked rainbow trout in Dale Hollow Lake; (4) assess temporal and size-

31

based trends in predation on rainbow trout in Dale Hollow Lake and two other Tennessee

reservoirs; and (5) quantify consumption of recently-stocked rainbow trout by walleyes in Dale

Hollow Lake.

STUDY AREAS

Dale Hollow Lake is situated on the Obey River and lies mainly in northern Tennessee,

where it covers portions of Clay, Pickett, Overton, and Fentress Counties; parts of the reservoir

also extend northward into two Kentucky counties. Constructed in 1943, this 11,210 ha tributary

storage impoundment is operated for hydroelectric power and flood control by the Army Corps

of Engineers. It has a maximum depth of 42 m. Dale Hollow Lake is stocked annually with

about 76,500 rainbow trout, representing about 7 fish/ha. It was stocked twice during the winters

of both 2007-2008 and 2008-2009. During 2007, some fish were stocked inshore at boat ramps

and some were transported offshore using a barge; during 2008, trout were only stocked from

boat ramps. The study area for the movement and mortality study encompassed the main-

channel of Dale Hollow Lake and all embayments and creeks (except Mitchell Creek) ranging

from the dam up-reservoir about 15 km (measured along the Obey River channel) to First Island

and encompassed about 1550 ha; 1085 ha were deemed main-channel habitat and 465 ha were

deemed cove habitat (Figure 1). Mitchell Creek is a large sinuous embayment encompassing

about 820 ha and was omitted from this study along with areas upstream of First Island due to

personnel limitations. Horse Creek Marina, a stocking site, is located at the head of Horse

Creek, which is a 2.1 km long 69-ha embayment near the dam. The marina is comprised of a

number of floating docks and boat slips. Pleasant Grove Recreation Area, a second stocking site,

is located on the main channel, approximately 3.75 km upstream of the dam.

South Holston Lake is situated on the Holston River and is located in Sullivan County in

northeastern Tennessee and extends into Washington County, Virginia. Construction of this

3,068-ha impoundment was completed in 1950. Retention time averages 340 d and water levels

fluctuate an average of 12 m annually. The trophic state of this tributary storage impoundment is

mesotrophic. Maximum depth in the forebay is 75 m. Approximately 40,000 rainbow trout are

stocked annually, representing about 13 fish/ha. During 2007, some fish were stocked inshore at

32

boat ramps and some were transported offshore using barges or by towing net-pens; during 2008,

trout were only stocked from boat ramps.

Watauga Lake is a 2,602-ha tributary storage impoundment situated on the Watauga

River and covers parts of Johnson and Carter Counties in northeastern Tennessee. The

maximum depth is 95 m and the mean annual fluctuation in water level is 13.4 m. It is

mesotrophic and has an average retention time of 400 d. About 43,000 trout are stocked

annually, or about 17 fish/ha. During 2007, fish were stocked both inshore and offshore; during

2008, trout were only stocked inshore from boat ramps.

All three reservoirs support healthy fish communities. Predatory fishes include holdover

rainbow trout, walleye (maintained through stocking), largemouth bass, smallmouth bass, spotted

bass, channel catfish Ictalurus punctatus, and flathead catfish Pylodictis olivaris. Additionally,

Dale Hollow Lake supports a population of muskellunge Esox Masquinongy and South Holston

Lake and Watauga Lake are stocked annually with lake trout Salvelinus namaycush. Avian

predators including common loons Gavia immer, great blue herons Ardea herodias, and bald

eagles Haliaeetus leucocephalus frequent all three study sites.

METHODS

Movements and Habitat Use

Hatchery-raised rainbow trout were implanted with Sonotronics® IBT-96-2 ultrasonic

transmitter tags2 (23 mm x 7 mm, weight in air 4.4 g) at Dale Hollow National Fish Hatchery,

Celina, Tennessee. Surgical procedures were identical to those used in a preliminary tagging

study (Ivasauskas and Bettoli, In Review). Monocryl sutures were used on all fish. Tagged fish

ranged in length from 210 to 280 mm and averaged 235 mm (SE=2.57). Tag burdens ranged

from 1.9 - 4.4% and averaged 2.9%. Immediately following surgery, trout were transferred to

one of two 580 L transport tanks equipped with aerators; water in the tanks was treated with a

commercial ―bait saver‖ formulation and ice was added as needed. Trout spent no longer than

20 h in the tanks and transport to stocking sites took less than 15 min.

2 The use of trade, product, industry or firm names or products is for informative purposes only and does not

constitute an endorsement by the U.S. Government or the US Geological Survey

33

Tags were implanted in 24 trout on 6 January 2009. On 7 January 2009, 12 tagged trout

were stocked at Horse Creek Marina in Dale Hollow Lake and 12 were stocked at Pleasant

Grove Recreation Area. Tags were implanted in an additional 20 trout on 5 March 2009. On 6

March 2009, 10 tagged fish were stocked at Horse Creek Marina and 10 were stocked at Pleasant

Grove Recreation Area. All tagged fish were released alongside at least 1,500 untagged rainbow

trout used in normal stocking operations.

Tagged trout were manually tracked from a boat during daylight hours by following a

transect around the perimeter of the study area and stopping at 143 listening points. A

Sonotronics® directional hydrophone (DH-4) and Sonotronics® receivers (USR-5W and USR-

96) were used to detect tag transmissions. When a fish was located, it was identified, its position

was triangulated, and its geographic coordinates were recorded as a waypoint using a Garmin®

ETrex Venture HC Global Positioning System (GPS). Fish stocked in January and March were

tracked 5 and 7 times, respectively, during the first 7-d post-stocking; trout from each stocking

were then tracked once every 6-8 d thereafter for an additional 7 weeks. Dale Hollow Lake was

isothermal at the times of both stockings and exhibited weak stratification at the end of April

2009, two months after the March 2009 stocking event. However, surface water temperature

(measured at a depth of 1.5m) was suitable for rainbow trout (i.e., < 21oC; Rowe and Chisnall

1995) through the duration of the study.

Tracking data were analyzed using ESRI ® Arcsoft Geographic Information System™

(ArcGIS) software. The Dale Hollow Lake polygon was acquired from the Tennessee Federal

GIS Data Server (www.tngis.org 2009). All distances were measured as the minimum within-

polygon distance between two points. Tags that were relocated at the same location for at least

two weeks and exhibited no further movement were considered ―stationary.‖ Tags were declared

stationary during the first week that they ceased movement. Observations of stationary tags and

movement during the one preceding week were excluded from analyses.

Dispersal distances (km) were determined by measuring the distances from stocking sites

to a fish‘s respective location. If a fish was not located during a given tracking event, but was

located during the immediately preceding and subsequent tracking events, the dispersal distance

for the week that the fish was not found was inferred as the average of the two distances. If the

fish was not located during two or more subsequent events, no inference was made. Fish that

were last observed within 2 km of Mitchell Creek or First Island and were not observed on any

34

subsequent tracking events were assumed to have moved out of the study range; those fish were

assigned a dispersal distance equivalent to the distance from the stocking site to the appropriate

transect boundary (i.e., 11 km for fish that dispersed from Horse Creek to Mitchell Creek, 13.5

km for fish that dispersed from Horse Creek to First Island, 6 km for fish that dispersed from

Pleasant Grove to Mitchell Creek, and 9 km for fish that dispersed from Pleasant Grove to First

Island). The aforementioned rules were separately applied for analyses of daily dispersal during

the first week and weekly dispersal during the first two months post-stocking.

Minimum traversed distances (km/wk) were determined by measuring the distances

between individual fish locations during subsequent weeks. If a fish was not located during a

given week, but was located during the immediately preceding and subsequent weeks, the

traversed distance was equally divided between the two one-week-timeframes. If the fish was

not located on two or more subsequent weeks, no inference was made for movement during the

interim timeframe.

Mean dispersal distances at one week post-stocking and total distances traversed during

the first week post-stocking were compared between fish stocked at Pleasant Grove and Horse

Creek Marina using t-tests. Total distances traversed were determined by summing daily

distances traversed during the first week. If differences were not apparent during the first week

post-stocking, behavior of fish stocked from the two locations was deemed similar, and data

were pooled for further analyses.

Habitat preference was assessed by conducting a chi-square test comparing the number of

observations of fish located in coves and in the main channel of the reservoir with the number

expected in each habitat-type, based on an equal distribution according to availability (i.e.,

surface area) within the study area. Designations of habitat-type are depicted in Figure 1.

Estimates of mortality from telemetry data

Estimates of rainbow trout mortality in Dale Hollow Lake during the first two months

post-stocking were derived from observations of movement for the 44 telemetered rainbow trout

stocked in 2009 and described above. Stationary tags are indicative of mortality (Bettoli and

Osborne 1998; Hightower et al. 2001) or expulsion. Tags that were relocated at the same

location for at least two weeks and exhibited no further movement were considered ―stationary.‖

35

Any tag that was not detected during an individual tracking event but was later relocated in a

different location was assumed to be non-stationary during that interim period. No inference was

made for any tag that was not later relocated (i.e., data were right-censored). Some of the

transmitters implanted in trout that were stocked on 7 January 2009 functioned longer than the

guaranteed 2-month battery life and were therefore encountered during efforts to locate fish

stocked on 6 March 2009; those observations were considered when determining the fate of

individual fish.

There is a delay in detecting mortality resulting from predation because telemetry tags

implanted into ingested prey must pass through the predator‘s gastrointestinal tract and be

excreted before assuming a stationary position. Hyvarinen and Vehanen (2004) reported that

when pike Esox lucius consumed telemetered brown trout Salmo trutta (N=8), the implanted tags

were passed within 1-9 d. Similarly, Jepsen et al. (1998) reported that when pike ate telemetered

salmonid smolts (N=27), tags were passed within 3–6 d.

We conducted a simple experiment in March 2008 to examine tag excretion utilizing 15

walleye brood fish. Walleyes at Eagle Bend State Fish Hatchery (Clinton, TN) were

anesthetized using MS-222 and were force-fed a semi-frozen rainbow trout bolus implanted with

a replica telemetry transmitter. After holding in a circular tank for 10-12 h to monitor for

regurgitation, walleyes were transferred to a lined hatchery pond and food was withheld. Twelve

walleyes were held for 6-8 d and three were held for 13 d. This experiment confirmed that

walleye are able to excrete consumed transmitters: one out of 12 fish excreted the transmitter

within 6-8 d and 2 out of 3 excreted it within 13 d. However, ingestion of subsequent meals is

known to increase the rate of gastric emptying (Bromley 1994); because of this, it is likely that

tags consumed by walleye in the wild will be excreted faster than those observed in this

controlled experiment. Empirical evidence subsequently obtained from monitoring the tagged

trout stocked in Dale Hollow Lake suggested that tags can be excreted by some predators in less

than a week. Thus, tags in the present study were declared stationary during the first week that

they ceased movement.

A stationary transmitter may also be indicative of a tag that has been expelled. Expulsion

is a well-documented phenomenon in several fish species including rainbow trout (Ivasauskas

and Bettoli, In Review). Tag expulsion that is mistakenly interpreted as mortality causes

survival to be underestimated. To account for this possible bias, the Kaplan-Meier survival

36

estimate for those weeks when a stationary transmitter was detected was adjusted by adding the

expected cumulative probability of transmitter expulsion for that week. Expected cumulative

probabilities of transmitter expulsion were adopted from laboratory findings described by

Ivasauskas and Bettoli (In Review) and are reported in Table 1.

Survival curves for the first eight weeks post-stocking were computed using the Kaplan-

Meier product limit estimator (Kaplan and Meier 1958). The Kaplan-Meier estimator is

preferred to other estimators because it does not require assumptions of constant survival or

constant sample size over intervals (White and Garrott 1990). Instantaneous weekly mortality

rates over the first eight-weeks post-stocking were calculated for each of the two tagged cohorts

stocked early- and late-winter using the equation:

Z=t

NN ))(log)((log 8e0e

where 0N is the proportion of trout surviving at week 0 (1.00), 8N is the estimated proportion of

trout alive at eight weeks, and t is the interval in weeks (8).

Trends in predation on recently-stocked trout

South Holston Lake was stocked with rainbow trout 3–11 December 2007 and 3-12

December 2008 and potential predators were sampled 12 December 2007 and 11-12 December

2008. Watauga Lake was stocked 12-18 December 2007 and 11-15 December 2008 and

predators were sampled 18 December 2007 and 19 December 2008. Stocking in Dale Hollow

Lake took place in two phases (early- and late-winter) during both years. During early-winter,

trout were stocked on 2 January 2008 and 9 January 2009 and predators were sampled 8 January

2008 and on nine dates between 9 January and 6 February 2009. During late-winter, trout were

stocked on 2-3 April 2008 and 6 March 2009 and sampled 3 April and 8 April 2008 and 9-10

March 2009.

Experimental sinking gillnets measuring 30 m x 2.4 m with mesh sizes ranging from 38

mm to 76 mm bar measure were fished perpendicular to the shore. Nets were deployed during

the afternoon and recovered the following morning. All potential trout predators (i.e., adult

walleye, largemouth bass, smallmouth bass, spotted bass, channel catfish, flathead catfish, and

rainbow trout) were weighed (g) and measured (TL, mm), euthanized with an overdose of MS-

37

222, and their stomachs were removed. Stomach contents were categorized to the lowest

identifiable taxonomic order, quantitated, and any trout remains were measured (standard length

[SL], mm). For trout remains that were too digested to accurately measure, the predigested SL

was visually estimated; when this was not possible, no length observation was recorded.

Measurements taken from a random sample of 210 Dale Hollow National Fish Hatchery rainbow

trout destined for stocking into reservoirs were used to create a simple SL-TL linear regression

model and to create a length-frequency distribution of stocked fish.

Analyses were conducted separately for each predator species for which there was an

adequate number of observations of fish that had consumed trout (N ≥ 10; walleye and

smallmouth bass). The proportions of diets containing trout were compared among the three

study sites using chi-square analysis; where differences were not discernible (P>0.10), data were

pooled. Logistic regression models were constructed using data from fish with non-empty

stomachs to relate the probability of a trout being in a predator‘s diet to that predator‘s size. The

Hosmer-Lemeshow goodness-of-fit test was used to assess the fit of logistic models to the data.

The Kolmogorov-Smirnov test was used to compare the length-frequencies of ingested trout and

all trout stocked. Simple linear regression was used to relate the TL of consumed trout to the TL

of the fish that ingested it (Poe et al. 1991; Zimmerman 1999).

Data collected from Dale Hollow Lake permitted an analysis of temporal trends in

consumption of trout during the first week and first month post-stocking. Multiple logistic

regression models with forward selection were constructed to identify possible factors that may

have influenced the proportion of walleye that consumed a trout. The predictor variables were (in

the order that they were entered into the model) predator size and days since trout stocking. To

assess feeding activity, the proportions of non-empty stomachs following early- and late-winter

stocking events in Dale Hollow Lake were compared using a chi-square test. The proportions of

diets containing trout were also compared between early- and late-winter using a chi-square test.

Quantify predation on rainbow trout by walleyes in Dale Hollow Lake

Daily predation on stocked rainbow trout by walleyes in Dale Hollow Lake was

quantified using a method similar to those used by Rieman et al. (1991), Zimmerman (1999), and

Fritts and Pearsons (2004). We used the equation:

38

agemax

1age

troutincludes age @diet nonempty

timedigestion gastric

PPC

tN,

where maximum age is the maximum age of walleye in the system (16 yrs; Vandergoot 2001);

tN is the number of walleyes at a given age and is estimated by applying the annual mortality

estimate to the number of fish recruited to age-1; nonemptyP is a constant that describes the

proportion of walleye with non-empty stomachs and is estimated from the gillnet data for early-

and late-winter; troutincludes age @diet P is the probability that the diet of a walleye at a given age is

comprised of at least one trout and was estimated by inputting the average length-at-age of

walleyes in Dale Hollow to the corresponding logistic regression model; the average duration of

gastric digestion was estimated to be about 1 day, using the equation proposed by Swenson and

Smith (1973). Estimates of annual mortality (28%) and length-at-age were derived from annual

gillnet samples collected by TWRA between the years 2001 and 2008. Because walleye

recruitment is generally low and variable (Forney 1976), conservative and liberal estimates of the

number of walleye surviving to age-1 were established at 1% and 5% of the mean number of

walleye stocked annually between 2001 and 2008 (i.e., 258,000).

RESULTS

Movements and Habitat Use

The mean distances dispersed after one week did not differ for fish stocked at Horse

Creek Marina and Pleasant Grove Recreation Area (t-test; df=1, 34; F=0.05; P=0.8305). The

mean total distance traversed during the first week post-stocking also did not differ between

stocking sites (t-test; df=1, 33; F = 0.01; P=0.9789). Behavior of fish stocked at each site was

therefore considered similar and data were pooled for further analyses.

Dispersal took place in three stanzas (Figure 2; Table 1). The first stanza (approximately

0-7 d post-stocking) was categorized by rapid dispersal of trout away from each stocking site.

During the second stanza (1 – 4 weeks), dispersal was minimal. During the third stanza (4 – 8

weeks), fish once again steadily dispersed away from their respective stocking sites.

39

Minimum distances traversed also followed a three-stanza pattern (Figure 3; Table 2).

During the first stanza (0-7 d post-stocking), trout exhibited high levels of movement away from

each stocking site. During the second stanza (1 – 4 weeks), trout exhibited low levels of mainly

localized movement. During the third stanza (4 – 8 weeks), trout exhibited relatively high

sustained levels of movement.

Rainbow trout exhibited a strong affinity for coves and embayments during the first 8-

weeks post-stocking (χ2=176.3; df=1; P < 0.0001). Cove habitat represented 30% of the study

site‘s surface area, whereas 63% of trout observations were recorded in coves. Rainbow trout

were infrequently encountered in pelagic, offshore habitats; most (69%) observations of rainbow

trout in the main-channel were along the shoreline.

Estimates of mortality from telemetry data

Trout stocked in March 2009 exhibited better survival than those stocked in January 2009

(Figure 4). Weekly sample sizes used for construction of the survival curve are given in Table 2.

Weekly instantaneous mortality (Z) during the first eight weeks post-stocking for trout stocked in

January 2009 was 0.062; whereas, weekly instantaneous mortality for trout stocked in March

2009 was 0.027. These estimates of Z correspond to weekly survival rates of 94% and 97%,

respectively for trout stocked in January and March, or 61% and 81% survival over eight weeks.

Trends in predation on recently-stocked trout

Walleye, smallmouth bass, largemouth bass, spotted bass, and holdover rainbow trout all

preyed on recently stocked rainbow trout (Table 3). Sample sizes of walleye and smallmouth

bass were large enough to justify statistical analyses. There was no strong evidence indicating

that the proportions of diets containing trout differed among the three study reservoirs for

walleye (2 = 4.4099, df=2, P=0.1103) or for smallmouth bass (

2=2.3730, df=2, P=0.3053).

Therefore, data from the three reservoirs were pooled for each of these species.

The probability that a walleye recently preyed on at least one rainbow trout was

positively related to walleye size (Figure 5; logistic regression; 2=32.46; df=1; P < 0.0001;

Hosmer-Lemeshow Goodness-of-fit test, P = 0.8897). The smallest walleye that consumed a

40

trout was 521 mm TL. Diets containing rainbow trout were most often comprised of only a

single trout (56%); some stomachs contained other fish prey (e.g., clupeids) and some larger

walleyes consumed up to four rainbow trout. The probability of consuming more than one

rainbow trout was also significantly related to the size of the walleye (Figure 6; logistic

regression; 2=14.93; df=1; P=0.0001; Hosmer-Lemeshow G.O.F. = 0.3748).

The length-frequency distribution for consumed trout was negatively shifted and left-

skewed as compared to that of fish measured at the hatchery (Figure 7; Kolmogorov-Smirnov

test; P < 0.0001). There was no discernible linear trend between the total lengths of walleye and

consumed trout (Figure 5; R2 = 0.025; P = 0.3495).

A larger proportion of walleyes collected from Dale Hollow Lake had non-empty

stomachs during early-winter (41%) than during late-winter (14%) in 2008 and 2009 (2=26.93;

df=1; P < 0.0001), but the proportion of non-empty stomachs that contained trout remained

constant between the two time periods (2=0.4953; df=1; P= 0.4816). Following the early-

winter stocking, no temporal trend in trout consumption was detected during the first week

(multiple logistic regression; df=1; P=0.2280) or the first month post-stocking (multiple logistic

regression; df=1; P=0.3254).

The probability that a smallmouth bass recently preyed on a rainbow trout was positively

related to its size (Figure 8; logistic regression; df=1; 2=2.972; P=0.0847; Hosmer-Lemeshow

G.O.F. = 0.4289). Diets containing rainbow trout were most often comprised of only a single

trout (55%); two smallmouth bass (446 and 470 mm TL) had each consumed two trout. Other

prey items included other fish prey (i.e., clupeids and centrarchids) and crayfish. The length-

frequency distribution of consumed trout was negatively shifted compared to that of fish

measured at the hatchery (Figure 4; Kolmogorov-Smirnov test; P < 0.0001). There was no linear

relationship between the total lengths of smallmouth bass and consumed trout (Figure 7;

P=0.8120). In Dale Hollow Lake, the proportion of smallmouth bass with non-empty stomachs

did not differ between early- and late-winter (2=0.40; df=1; P=0.5262).

Quantify predation on rainbow trout by walleyes in Dale Hollow Lake

Average overall consumption of rainbow trout by walleyes in Dale Hollow Lake during

the first week post-trout stocking was 195 – 977 trout per day for trout stocked in early-winter.

41

Average overall consumption was 67 – 335 trout per day for those stocked in late winter.

Assuming that the average number of rainbow trout stocked annually (76,500) were stocked

during January and March in two equal-sized batches (38,250 per stocking), instantaneous

weekly mortality (Z) due only to walleye predation was 0.032 – 0.18 and 0.011 – 0.053,

respectively, depending on first-year walleye survival. Using these estimates, walleye would

consume 23-75% of trout stocked in early-winter and 8-34% of trout stocked during late winter

during the first eight weeks post-stocking.

DISCUSSION

When rainbow trout are stocked in systems where they tend to disperse slowly, catch

rates near the stocking sites, especially during the first several weeks post-stocking, are high;

resultantly, fishing mortality and return to creel are high (Helfrich and Kendall 1982; Fay and

Pardue 1986; Shetter and Hazzard 1941). The rapid rate of dispersal from both main-channel

(Pleasant Grove Recreation Area) and cove stocking sites (Horse Creek Marina) indicated that

rainbow trout stocked in Dale Hollow Lake quickly become less vulnerable to anglers fishing

near stocking sites.

Bergthold and Bettoli (2009) reported that electrofishing catch rates in coves were up to

237% greater than in the main-channel. In this study, trout were observed 241% more frequently

in coves than main-channel (taking into account the total availabilities of the two habitat types).

During tracking events, most trout encountered in the main channel, especially those in the

pelagic zone; were moving, whereas tagged trout in coves were typically stationary under docks,

amongst large woody debris, or associated with benthic morphometric features (e.g., drop offs,

channel points, pockets). Together, these two studies provide strong evidence that recently

stocked rainbow trout seek cove habitats. Coves in Dale Hollow Lake tend to have greater

shoreline development indices than the main channel. As such, coves provide a more

structurally complex habitat and receive more allochthonous inputs. Habitat complexity and

structure provide protection from predation (Walters et al. 1991, Tabor and Wurtsbaugh 1991,

Kahler et al. 2000). The influx of allochthonous materials (coupled with the more complex

habitat structure) may also promote greater abundances of invertebrates (the primary food of

recently stocked trout; Bergthold and Bettoli 2009) in cove habitat.

42

The observed behaviors of recently stocked rainbow trout in Dale Hollow Lake reflect

their vulnerability to a variety of predators inhabiting a variety of habitats. This study confirmed

that stocked rainbow trout are readily consumed by predatory fishes in Dale Hollow Lake,

Watauga Lake, and South Holston Lake. Effective strategies for promoting the survival of

recently stocked trout should therefore focus on biologic aspects of predators.

It was demonstrated for walleye and smallmouth bass that the sizes of individual

predators predicted whether they would consume a trout. Such a relationship was expected, as

physical constraints of the predator, such as mouth gape (Hambright 1991) and esophagus length

(Carothers et al. in press), limit the maximum size of prey it can consume. Although predator

size predicted trout consumption, the size of trout consumed was not linked to predator size.

Nevertheless, the shift for consumed trout toward smaller size classes, relative to the sizes of fish

stocked, indicated that the vulnerability of trout to predation by walleyes and smallmouth bass

(and presumably other predator species) decreased with size. This finding is a logical extension

to the results of the laboratory experiment presented by Yule et al. (2000), as larger stocked trout

were invulnerable to being predated on by a larger proportion of predators inhabiting the

systems. Although stocking larger fish would reduce losses to predation, the economic cost of

rearing larger fish may not justify the return, in terms of survival, and additional research must

be conducted in order to evaluate the cost-effectiveness of doing so.

Survival can also be increased by stocking trout later in the winter. Tracking and

gillnetting data both indicated that the weekly mortality rate over two-months is higher for

rainbow trout stocked during January than during March or early-April. Additionally, rainbow

trout exhibit very low growth rates (≤ 0.2 %-d-1

) in Tennessee reservoirs between December and

mid-March (Bergthold and Bettoli 2009). They are therefore subject to high levels of predation

for a longer period of time (i.e., until they assume a faster growth rate) when stocked earlier

during the winter.

A larger proportion of walleyes collected from Dale Hollow Lake had non-empty

stomachs during January and February (41%) than during March and April (14%). An almost

identical pattern was described by Libbey (1969): the proportion of walleye with non-empty

stomachs in that earlier study in Dale Hollow Lake decreased abruptly from 48% during January

and February to 17% in March and 7% in April. Muench (1966) observed a total cessation of

walleye feeding activity in March in nearby Center Hill Lake, Tennessee. This decline in feeding

43

activity could be attributed to walleye spawning behavior. In Dale Hollow Lake, spawning has

been observed between mid-March and late-April (Libbey 1969). Many of the walleye collected

during April 2008 and March 2009 for purposes of the present study released milt or eggs upon

removal from the net indicating, that spawning had commenced.

Most (78%) of the transmitters implanted into trout stocked in January and subsequently

designated as stationary were designated as such within 3 weeks post-implantation. In a

laboratory experiment (Ivasauskas and Betoli, In Review), no rainbow trout sutured with

Monocryl expelled tags during the first three weeks. It was therefore unlikely that these

stationary transmitters resulted from expulsion as opposed to mortality. The rate of tag

encapsulation and expulsion has been previously shown to be positively related to water

temperature (Bunnell and Isely 1999). However, this relationship is of minor importance when

considering differences in expulsion likelihood of telemetered fish stocked in January and

March, as the water temperature (taken at a depth of 1.5 m) differed by only 1.3oC between 22

January and 19 March, 2009 (unpublished data).

It is important to note that conclusions drawn in this study are only applicable to recently

stocked small-catchable-sized rainbow trout during the winter and early-spring. Epilimnetic

water temperatures greater than 21oC (which occurred by the end of May 2009) force rainbow

trout to move deeper in the water column (Horak and Tanner 1964). Ontogenetic shifts in

habitat utilization (Landry et al. 1999) and prey selectivity (Bergthold and Bettoli 2009) have

also been noted for rainbow trout in lentic systems; therefore, behavior and survival of older

trout should not correspond with behaviors observed in this study.

44

REFERENCES

Alexander, R. 1974. Functional design in fishes. Hutchinson and Company, London, England.

Avery, E. L. 1975. An evaluation of stocking fingerling trout in a "two story" trout lake.

Wisconsin Department of Natural Resources, Madison, Wisconsin.

Bachman, R. A. 1984. Foraging behavior of free-ranging wild and hatchery brown trout in a

stream. Transactions of the American Fisheries Society 113:1-32.

Baird, O. E., Krueger, C. C. & Josephson, D. C. 2006. Growth, movement, and catch of brook,

rainbow, and brown trout after stocking into a large marginally suitable Adirondack river.

North American Journal of Fisheries Management 26: 180–189.

Baldwin, C. M., J. G. McLellan, M. C. Polacek, and K. Underwood. 2003. Walleye predation on

hatchery releases of kokanees and rainbow trout in Lake Roosevelt, Washington. North

American Journal of Fisheries Management 23:660-676.

Barwick, D. H. 1985. Stocking and hooking mortality of planted rainbow trout in Jocassee

Reservoir, South Carolina. North American Journal of Fisheries Management 5:580-583.

Bergthold, C. L., and P.W. Bettoli. 2009. Growth, Diet, and Sampling of Rainbow Trout in

Three Tennessee Reservoirs. Tennessee Wildlife Resources Agency, Fisheries Report 09-

04, Nashville.

Bettinger, J. M., and P. W. Bettoli. 2002. Fate, dispersal and persistence of recently stocked and

resident rainbow trout in a Tennessee tailwater. North American Journal of Fisheries

Management 22:425-432.

Bettoli, P. W. 1996. Survey of nighttime trout angling in Dale Hollow Lake. Tennessee Wildilfe

Resources Agency, Fisheries Report 96-17, Nashville.

Bettoli, P.W. and R.S. Osborne. 1998. Hooking mortality and behavior of striped bass following

catch and release angling. North American Journal of Fisheries Management 18, 609–

615.

Briggs, C. T. & Post, J. R. 1997. Field metabolic rates of rainbow trout estimated using

electromyogram telemetry. Journal of Fish Biology 51, 807-823.

Bromley, P. 1994. The role of gastric evacuation experiments in quantifying the feeding rates of

predatory fish. Reviews in Fish Biology 4:36–66.

Buckmeier, D. L., and R. K. Betsill. 2002. Mortality and dispersal of stocked fingerling

largemouth bass and effects on cohort abundance. Pages 667–676 in D. P. Phillip and M.

S. Ridgway, editors. Black bass: ecology, conservation, and management. American

Fisheries Society, Symposium 31, Bethesda, Maryland.

Bunnell, D. B., and J.J. Isely. 1999. Influence of temperature on mortality and retention of

simulated transmitters in rainbow trout. North American Journal of Fisheries

Management 19: 152–154.

Cargill, A. S. II. 1980. Lack of rainbow trout movement in a small stream. Transactions of the

American Fisheries Society 109:484–490.

45

Carothers, C., C. Kraft, and D. C. Josephson. In press. The role of gape limitation and prey shape

on smallmouth bass feeding success. Journal of Fish Biology.

Cooke, S. J., Thorstad, E. B. & Hinch S. G. 2004. Activity and energetics of free-swimming fish:

insights from electromyogram telemetry. Fish and Fisheries 5, 21–52.

Cresswell, R. C, and R. Williams. 1982. Post-stocking movements and recapture of hatchery-

reared trout released into flowing waters: effect of time and method of stocking. Fisheries

Management 13:97-103.

Cunningham, P. K., and C.S. Anderson. 1992. Effect of size at stocking on harvest of rainbow

trout in Bad Medicine Lake. Minnesota Department of Natural Resources, St. Paul.

Fay C.W., and G.B. Pardue. 1986. Harvest, Survival, Growth, and Movement of Five Strains of

Hatchery-Reared Rainbow Trout in Virginia Streams. North American Journal of

Fisheries Management; 6: 569-579.

Fiss, F. C., and J. W. Habera. 2006. Trout management plan for Tennessee 2006-2016.

Tennessee Wildlife Resources Agency, Nashville.

Flinders, J.M., and S.A. Bonar. 2008. Growth, condition, diet, and consumption rates of

northern pike in three Arizona reservoirs. Lake and Reservoir Management 24: 99-111.

Forney, J. L. 1976. Year-class formation in the walleye (Stizostedion vitreum vitreum) population

of Oneida Lake, New York, 1966–1973. Journal of the Fisheries Research Board of

Canada 33:783-792.

Fritts, A. L., and T. N. Pearsons. 2004. Smallmouth bass predation on hatchery and wild

salmonids in the Yakima River, Washington. Transactions of the American Fisheries

Society 133:880–895.

Fuiman, L. A., and A. E. Magurran. 1994. Development of predator defenses in fishes. Reviews

in Fish Biology and Fisheries 4:145-183.

Hambright, K. D. 1991. Experimental analysis of prey selection by largemouth bass: Role of

predator mouth width and prey body depth. Transactions of the American Fisheries

Society 120:500-508.

Heimer, J. T., W. M. Frazier, and J. S. Griffith. 1985. Post-Stocking Performance of catchable-

size hatchery rainbow trout with and without pectoral fins. North American Journal of

Fisheries Management 5:21-25.

Helfrich, L. A. and W. T. Kendall. 1982. Movements of hatchery-reared rainbow, brook, and

brown trout stocked in a Virginia mountain stream. Progressive Fish Culturist 44 : 76-80.

Hightower, J. E., J. R. Jackson, and K. H. Pollock. 2001. Use of telemetry methods to estimate

natural and fishing mortality of striped bass in Lake Gaston, North Carolina. Transactions

of the American Fisheries Society 130:557-567.

Horak, D. L., and H. A. Tanner. 1964. The use of vertical gill nets in studying fish depth

distribution, Horsetooth Reservoir, Colorado. Transactions of the American Fisheries

Society 93:137-145.

Hyvarinen, P., and T. Vehanen. 2004. Effect of brown trout body size on post-stocking survival

and pike predation. Ecology of Freshwater Fish 13:77-84.

46

Ivasauskas, T. J., and P. B. Bettoli. In Review. Effects of Suture Material and Ultrasonic

Transmitter Size on Survival, Growth, Wound Healing, and Expulsion in Rainbow Trout.

Transactions of the American Fisheries Society.

James, G. D., and J. R. M. Kelso. 1995. Movements and habitat preference of adult rainbow trout

(Oncorhynchus mykiss) in a New Zealand montane lake. New Zealand Journal of Marine

and Freshwater Research 29:493–503.

Jepsen, N., A. Kim , F. Økland, and G. Rasmussen. 1998. Survival of radiotagged Atlantic

salmon (Salmo salar L.) and trout (Salmo trutta L.) smolts passing a reservoir during

seaward migration. Hydrobiologia 371-372:347-353.

Johnson J.E., and G.P. Rakoczy. 2004. Investigations into recent declines in survival of brown

trout stocked in Lake Charlevoix and Thunder Bay, Lake Huron. Michigan Department

of Natural Resources Reasearch Report 2075, Lansing.

Johnson, J. H., and N. H. Ringler. 1998. Predator response to releases of American shad larvae in

the Susquehanna River basin. Ecology of Freshwater Fish 7:192-199.

Kahler, T., M. Grassley, and D. Beauchamp. 2000. A summary of the effects of bulkheads, piers,

and other artificial structures and shorezone development on ESA-listed salmonids in

lakes. Report to the City of Bellevue, Bellevue, Washington.

Kaplan, E.L. & Meier, P. 1958. Nonparametric estimation from incomplete observations. Journal

of the American Statistical Association 53: 457–481.

Landry, F., J. R. Post, and E. A. Parkinson. 1999. Spatial ontogeny of lentic age-0 rainbow trout,

Oncorhynchus mykiss: whole-lake manipulations of population size structure Canadian

Journal of Fisheries and Aquatic Sciences 56:1916–1928.

Libbey, J. E. 1969. Certain aspects of the life history of the walleye, Stizostedion vitreum vitreum

(Mitchell), in Dale Hollow Reservoir, Tennessee–Kentucky, with emphasis on spawning.

Master‘s thesis. Tennessee Technological University, Cookeville.

Lindberg, M., P. Rivinoja, L.-O. Eriksson, A. AlanÄrÄ. 2009. Post-release and pre-spawning

behaviour of simulated escaped adult rainbow trout Oncorhynchus mykiss in Lake

Oumlvre Fryken, Sweden. Journal of Fish Biology 74: 691-698.

Lundvall, D. S., L. Persson, and P. Byström. 1999. Size-dependent predation in piscivores:

interactions between predator foraging and prey avoidance abilities. Canadian Journal of

Fisheries and Aquatic Science 56:1285-1292.

Malvestuto, S., and P. W. Black. 2007. Tennessee Reservoir Creel Survey: 2006 Results.

Marwitz, T. D., and W. A. Hubert. 1997. Trends in relative weight of walleye stocks in

Wyoming reservoirs. North American Journal of Fisheries Management 17:44-53.

McMahon, T. E., and D. H. Bennett. 1996. Walleye and northern pike. Fisheries 21(8):6-13.

McMillan, J. 1984. Evaluation and enhancement of the trout and walleye fisheries in the North

Platte River system of Wyoming with emphasis on Seminoe Reservoir. Wyoming Game

and Fish Dept., Laramie.

47

Mellina, E., Hinch, S.G., and MacKenzie, K.D. 2005. Seasonal movement patterns in stream-

dwelling rainbow trout in North-central Britsh Columbia, Canada. Transactions of the

American Fisheries Society 134: 1021-1037.

Miller, L. M., T. Close, and A. R. Kapuscinski. 2004. Lower fitness of hatchery and hybrid

rainbow trout compared to naturalized populations in Lake Superior tributaries.

Molecular Ecology 13:3379-3388.

Muench, K. A. 1966. Certain aspects of the life history of the walleye, Stizostedion vitreum

vitreum, in Center Hill Reservoir, Tennessee. Master‘s thesis. Tennessee Technological

University, Cookeville.

Poe, T. P., H.C. Hansel, S. Vigg, D.E. Palmer, and L. A. Prendergast. 1991. Feeding of

predaceous fishes on out-migrating juvenile salmonids in John Day Reservoir, Columbia

River. Transactions of the American Fisheries Society 120: 405-420.

Rieman, B.E., R.C. Beamesderfer, S. Vigg, and T.P. Poe. 1991. Estimated loss of juvenile

salmonids to predation by northern squawfish, walleyes, and smallmouth bass in John

Day Reservoir, Columbia River. Transactions of the American Fisheries Society

120:448-458.

Rowe, D. K., and B. L. Chisnall. 1995. Effects of oxygen, temperature and light gradients on the

vertical distribution of rainbow trout, Oncorhynchus mykiss, in two North Island, New

Zealand, lakes differing in trophic status. New Zealand Journal of Marine and Freshwater

Research, 29(3):421-434.

Runge, J.P., J.T. Peterson, C.R. Martin. 2008. Survival and dispersal of hatchery-raised rainbow

trout in a river basin undergoing urbanization. North American Journal of Fisheries

Management 28: 745-757.

Schlechte, J. W., R. K. Betsill, and D. L. Buckmeier. 2005. A laboratory evaluation of

poststocking predatory losses for cultured largemouth bass. Transactions of the American

Fisheries Society 134:141-148.

Shetter, D.S., and A.S. Hazzard. 1941. Results from plantings of marked trout of legal size in

streams and lakes of Michigan. Transactions of the American Fisheries Society 70: 446-

468.

Simpson, G. 2006. Movement of catchable rainbow trout after stocking in Rapid Creek. South

Dakota Department of Game, Fish and Parks. Completion Report 06-05, Pierre.

Swenson, W. A., Smith, L. L. 1973.Gastric digestion, food consumption, feeding periodicity and

food conversion efficiency in walleye, Stizostedion vitreum vitreum. Journal of Fisheries

Research Board Canada 30: 1327-1336.

Tabor, R. A., and W. A. Wurtsbaugh. 1991. Predation risk and the importance of cover for

juvenile rainbow trout in lentic systems. Transactions of the American Fisheries Society

120:728–738.

Talley, T. S. 1976. Predation on rainbow trout, Salmo gairdneri richardson, stocked in a two-

story reservoir. Master‘s thesis. Tennessee Technological University, Cookeville.

48

Vandergoot, C. J. 2001. Evaluation of current management practices and assessment of

recruitment, growth and condition of walleyes in Tennessee reservoirs. Tennessee

Technological Universtiy Cookeville.

Vuorinen, J. 1984. Reduction of genetic variability in a hatchery stock of brown trout, Salmo

trutta. Journal of Fish Biology 24:339-348.

Wahl, D. H., and R. A. Stein. 1988. Selective predation by three esocids: the role of prey

behavior and morphology. Transactions of the American Fisheries Society 117:142-151.

Walters, D. A., W. E. Lynch, Jr., and D. L. Johnson. 1991. How depth and interstice size of

artificial structures influence fish attraction. North American Journal of Fisheries

Management 11:319-329.

Walters, J. P., T. D. Fresques, and S. D. Bryan. 1997. Comparison of creel returns from rainbow

trout stocked at two sizes. North American Journal of Fisheries Management 17:474-476.

Warner, E. J., and T. P. Quinn. 1995. Horizontal and vertical movements of telemetered rainbow

trout (Oncorhynchus mykiss) in Lake Washington. Canadian Journal of Zoology 73:146-

153.

White G.C., Garrott R.A. 1990. Analysis of Wildlife Radio-tracking Data. Academic Press, San

Diego.

Wiley, R. W. 1993. Assessment of stocking hatchery trout: a Wyoming perspective. North

American Journal of Fisheries Management 13:160-170.

Yule, D. L., R. A. Whaley, and P. H. Mavrakis. 2000. Use of strain, season of stocking, and size

at stocking to improve fisheries for rainbow trout in reservoirs with walleyes. North

American Journal of Fisheries Management 20:10-18.

Zimmerman, M. P. 1999. Food habits of smallmouth bass, walleyes, and northern pikeminnow in

the lower Columbia River Basin during outmigration of Juvenile anadromous salmonids.

Transactions of the American Fisheries Society 128: 1036-1054.

49

Table 1. Dispersal distances of telemetered rainbow trout stocked in Dale Hollow Lake on 7

January and 6 March 2009. An asterisk denotes the maximum observable dispersion distance

(Horse Creek Marina to First Island). N = number of trout.

Post-stocking

period N

Mean Distance

Dispersed (km) SE Minimum Median Maximum

Day 1 42 0.70 0.117 0.07 0.45 3.90

Day 2 41 1.04 0.185 0.08 0.60 6.11

Day 3 38 1.07 0.195 0.06 0.54 6.36

Day 4 35 1.17 0.235 0.05 0.51 6.60

Day 5 37 1.53 0.281 0.06 0.97 8.35

Day 6 34 1.56 0.332 0.07 1.03 10.06

Day 7 34 1.77 0.381 0.08 1.24 11.76

Week 2 31 2.09 0.459 0.08 1.60 13.50*

Week 3 30 2.23 0.475 0.04 1.65 13.50*

Week 4 26 2.24 0.521 0.09 1.65 13.50*

Week 5 18 3.44 0.713 0.14 3.00 13.50*

Week 6 17 3.57 0.741 0.09 3.10 13.50*

Week 7 17 4.50 0.866 0.07 4.00 13.50*

Week 8 11 4.81 1.420 0.09 3.05 13.50*

Table 2. Weekly minimum distances traversed by telemetered rainbow trout stocked in Dale

Hollow Lake on 7 January and 6 March 2009. N = number of trout.

Post-stocking

period N

Mean Distance

Traversed

(km/wk) SE Minimum Median Maximum

Week 1 36 1.73 0.362 0.08 1.24 11.76

Week 2 31 0.95 0.189 0.00 0.45 4.20

Week 3 27 0.62 0.171 0.00 0.16 3.10

Week 4 24 1.30 0.307 0.00 0.35 4.60

Week 5 16 1.59 0.379 0.05 1.20 5.00

Week 6 14 1.63 0.463 0.05 0.93 5.70

Week 7 13 2.10 0.748 0.00 0.90 9.70

Week 8 7 1.13 0.217 0.20 1.20 1.90

50

Table 3: Correction factors applied to observations of stationary transmitters observed during a

given week. Values are adopted from laboratory findings reported by Ivasauskas and Bettoli

n(In review) and are the cumulative proportions of trout that expelled transmitters through eight

weeks.

Week 1 2 3 4 5 6 7 8

Correction factor 0.0 0.0 0.0 0.05 0.05 0.10 0.15 0.2

Table 4: Weekly sample sizes based on observations of 44 telemetered trout stocked in Dale

Hollow Lake on 7 January and 6 March 2009. N is the total number of alive and dead/expelled

fish within the study site.

Week 0 1 2 3 4 5 6 7 8

January stocking

Alive 24 20 15 13 11 9 7 7 7

Dead/expelled 0 4 7 7 8 9 9 9 9

N 24 24 22 20 19 18 16 16 16

March stocking

Alive 20 20 19 18 16 13 9 8 6

Dead/expelled 0 0 0 0 1 2 3 4 4

N 20 20 19 18 17 15 12 12 10

51

Table 5: Summary of sample sizes for the diet analysis of predatory species caught in gillnets

following rainbow trout stocking events occurring December 2007 through March 2009. A

―holdover‖ rainbow trout refers to a trout stocked during previous years.

Species Ntotal Nnon-empty Nwith trout

Dale Hollow Lake (January and February)

Walleye 155 66 7

Smallmouth bass 20 13 3

Largemouth bass 1 0 0

Rainbow trout (holdover) 6 4 1

Dale Hollow Lake (March and April)

Walleye 127 18 3

Smallmouth bass 17 12 1

Largemouth bass 4 3 1

Spotted bass 3 2 0

South Holston Lake (December)

Walleye 91 42 7

Smallmouth bass 23 10 4

Largemouth bass 4 4 0

Flathead catfish 15 2 0

Watauga Lake (December)

Walleye 146 60 16

Smallmouth bass 12 7 2

Largemouth bass 2 2 2

Spotted bass 5 5 1

Flathead catfish 3 0 0

52

Figure 1. The study area (shaded region) encompassed 1,550 ha of Dale Hollow Lake Crosshatching indicates cove habitat.

Telemetered rainbow trout were stocked at Horse Creek Marina and Pleasant Grove Recreation Area.

53

Weeks Post-stocking

0 2 4 6 8

Dis

tance F

rom

Sto

ckin

g S

ite (

km

)

0

1

2

3

4

5

6

7

Figure 2. Dispersal of tagged rainbow trout in Dale Hollow Lake after stocking. Points

represent mean distances from stocking sites and vertical bars denote standard error.

54

Weeks Post-stocking

0 2 4 6 8

Dis

tan

ce

Tra

ve

rse

d (

km

/wk)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Figure 3. Minimum weekly distances traversed (km/wk) by tagged rainbow trout in Dale

Hollow Lake after stocking. Points represent mean distances traversed and vertical bars denote

standard errors.

55

Weeks Post-stocking

0 2 4 6 8

Surv

ival

0.0

0.2

0.4

0.6

0.8

1.0

Figure 4. Kaplan-Meier survivorship curves constructed from observations of telemetered trout

stocked in 2009 and corrected for expulsion. The solid line represents fish stocked in January

and the dashed line represents fish stocked in March.

56

Walleye Total Length (mm)

200 300 400 500 600 700 800

Pro

ba

bili

ty d

iet in

clu

des tro

ut

0.0

0.2

0.4

0.6

0.8

1.0

Figure 5. Probability of walleye diet including at least one trout as a function of walleye total

length.

P < 0.0001

57

Walleye Total Length (mm)

200 300 400 500 600 700 800

Pro

ba

bili

ty d

iet in

clu

des 2

+ tro

ut

0.0

0.2

0.4

0.6

0.8

1.0

Figure 6. Probability of walleye diet including at least two trout, as a function of walleye total

length.

P = 0.0001

58

1

Fre

quency

2

Fre

qu

en

cy

3

Total Length (mm)

80 120 160 200 240 280

Freq

uenc

y

4

Figure 7. Length-frequency distributions of rainbow trout that were stocked, consumed by 5

walleyes, and consumed by smallmouth bass. N is the number of measured trout 6

Stocked

N = 210

Consumed by walleyes

N=36

Consumed by smallmouth bass

N=9

59

Smallmouth Bass Total Length (mm)

250 300 350 400 450 500 550

Pro

ba

bili

ty d

iet in

clu

de

s tro

ut

0.0

0.2

0.4

0.6

0.8

1.0

7 8

Figure 8. Probability of smallmouth bass diet including at least two trout, as a function of 9

smallmouth bass total length. 10

11

12

P = 0.0847