University of New Hampshire University of New Hampshire Scholars' Repository
Doctoral Dissertations Student Scholarship
Spring 2002
Winter flounder Pseudopleuronectes americanus stock enhancement in New Hampshire: Developing optimal release strategies
Elizabeth Alden Fairchild University of New Hampshire, Durham
Follow this and additional works at: https://scholars.unh.edu/dissertation
Recommended Citation Fairchild, Elizabeth Alden, "Winter flounder Pseudopleuronectes americanus stock enhancement in New Hampshire: Developing optimal release strategies" (2002). Doctoral Dissertations. 62. https://scholars.unh.edu/dissertation/62
This Dissertation is brought to you for free and open access by the Student Scholarship at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact [email protected]. INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6* x 9a black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. WINTER FLOUNDER Pseudopleuronectes americanus STOCK ENHANCEMENT IN
NEW HAMPSHIRE: DEVELOPING OPTIMAL RELEASE STRATEGIES
BY
ELIZABETH ALDEN FAIRCHILD
B.A., University of New Hampshire, 1991
M.S., University of New Hampshire, 1998
DISSERTATION
Submitted to the University of New Hampshire
in Partial Fulfillment of
the Requirements for the Degree of
Doctor of Philosophy
in
Zoology
May, 2002
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number 3045322
UMI*
UMI Microform 3045322 Copyright 2002 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This dissertation has been examined and approved.
Dissertation Director, Dr. W. fiuntting Howell, Professor of Zoology, University of New Hampshire
Dr. Harr^ V. Daniels, Associate Professor of Zoology, North Carolina State University
Dr. Richard Langan, " Adjunct Associate Professor of Zoology, University of New Hampshire
Dr. Kenneth M. Leber, Director of Center for Fisheries Enhancement, Mote Marine Laboratory, Sarasota, Florida
Dr. Winsor H. Watson HI, Professor of Zoology University of New Hampshire
Date
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DEDICATION
This dissertation is dedicated to Glenn C. Walker, my husband and best friend.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS
There are many people who I am indebted to for their help in this research. First, I
would like to thank Captains Eric Anderson, Carl Bouchard, David Goethel, and Joe
Jurek who collected the winter flounder broodstock for me each spring. Many students
worked on this project and without their help, these studies never would have been
completed so rapidly. I am extraordinarily grateful to Sarah Abramson, Christopher
Benton, Elizabeth Carver, Jennifer Fleck, Katie Reynolds, Glen Rice, Paula Rodgers,
Garrison Smith, Kevin Sullivan, Marta Toran, and John Wiedenmann. Additionally, I
would like to thank all current and former graduate students in the Howell and Watson
labs who have lent a hand, especially Deborah Bidwell, Dr. Steve Jury, Nick King, Jennie
Mandeville, Mike Morin, Dan O’Grady, James Sulikowski, and Jenna Wanat. Thank you
to the Coastal Marine Laboratory and Jackson Estuarine Laboratory staff, in particular
Noel Carlson, Deb Lamson, and Dave Shay, for their assistance in the lab and with the
research vessels. Thank you to Dr. Ray Grizzle for help in implementing the benthic core
sampling protocol, and to Amy Harmon and Dr. Larry Ward for their tutelage and
assistance in analyzing sediment samples. Thank you to the Marine Program and Zoology
Department staff, especially Meriel Bunker, Becky Crawshaw, Diane Lavalliere, Tammy
McGlone, Barbara Millman, Nancy Richmond, and Nancy Wallingford, who made my
life easier by taking care of all the critical paperwork and financial details.
Various equipment and supplies used in these studies were generously loaned
from other agencies. Thank you to Drs. Tony Calabrese and Ronald Goldberg of the
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. National Marine Fisheries Service, Milford, Connecticut Laboratory who donated the fish
cages, and to Lee Blankenship who taught me how to use coded wire tags and then
loaned me the tagging equipment. Thank you to Charlie Sleeper who constructed the
beam trawl and to Glenn Walker who designed and built the bird pens. I am grateful to
Great Bay Aquafarms, who supplied me with many last minute microalgae, rotifer, and
artemia orders when my cultures were in peril and there were many hungry mouths to
feed.
I am fortunate to have a large, caring group of family and friends that have been
integrally involved in my life. I am eternally thankful for their encouragement, love, and
camaraderie during these years.
Finally, I would like to express my utmost gratitude to my mentor, Dr. Hunt
Howell, and to the rest of my doctoral committee, Drs. Harry Daniels, Rich Langan, Ken
Leber, and Win Watson. My doctoral program was enriched due to their excellent
guidance and advice. I thank them all for the time they spent on this project, and for
allowing me to set my own schedule and pace.
Funding for this research was provided by the UNH/UME College SeaGrant
program under grant number R/FMD-158. Additional support was provided by the Center
for Marine Biology, the Zoology Department, and the Graduate School.
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS
DEDICATION...... iii
ACKNOWLEDGEMENTS...... iv
LIST OF TABLES...... ix
LIST OF FIGURES...... xi
ABSTRACT...... xiv
CHAPTER PAGE
INTRODUCTION...... 1
I. PREDATOR-PREY SIZE RELATIONSHIP BETWEEN Pseudopleuronectes
americanus AND Carcinus maenas...... 11
Introduction ...... 11
Materials and Methods ...... 14
Results...... 17
Discussion ...... 24
Summary ...... 30
0. DETERMINING AN OPTIMAL RELEASE SITE FOR JUVENILE
Pseudopleuronectes americanus IN THE GREAT BAY ESTUARY, NEW
HAMPSHIRE...... 32
Introduction ...... 32
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Materials and Methods 34
Results...... 40
Discussion ...... 48
Summary ...... 54
III. DETERMINING AN OPTIMAL RELEASE SEASON FOR JUVENILE
Pseudopleuronectes americanus IN THE GREAT BAY ESTUARY, NEW
HAMPSHIRE...... 55
Introduction ...... 55
Materials and Methods ...... 57
Results...... 58
Discussion ...... 76
Summary ...... 82
IV. CONDITION OF CULTURED JUVENILE Pseudopleuronectes americanus FOR
RELEASE IN THE GREAT BAY ESTUARY, NEW HAMPSHIRE...... 84
Introduction ...... 84
Materials and Methods ...... 87
Results...... 93
Discussion ...... 104
Summary ...... 109
SYNTHESIS...... 110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX A. PILOT-SCALE RELEASES OF CULTURED JUVENILE
Pseudopleuronectes americanus IN THE GREAT BAY ESTUARY, NEW
HAMPSHIRE IN 1999-2001 ...... 113
APPENDIX B. ANIMAL CARE AND USE APPROVAL
DOCUMENTATION...... 127
LIST OF REFERENCES...... 128
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES
Table Title Page
1.1 Mean number of cultured winter flounder killed per day by green crabs ...... 18
1.2 Mean number of wild winter flounder killed per day by green crabs ...... 19
1.3 Effects of crab and fish size on amount of cultured and wild winter flounder
killed ...... 20
1.4 Mean number of winter flounder killed per day by green crabs in relation to
temperature...... 21
2.1 Biomass and numerical abundance of primary prey available to winter flounder
inside and outside of the pens ...... 46
3.1 Species caught by beam and otter trawls in the three sites in Great Bay Estuary,
New Hampshire...... 60
3.2 Mean CPUE from beam trawls and benthic cores for 1999-2001 ...... 61
4.1 Summary of predator cue experiment data ...... 94
4.2 Summary of substrate preference observations of light colored, cultured winter
flounder ...... 99
4.3 Results of chi-square analysis at each sampling interval for substrate preference of
light colored, cultured winter flounder ...... 100
4.4 Summary o f substrate preference observations of dark colored, cultured winter
flounder ...... 102
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.5 Results of chi-square analysis at each sampling interval for substrate preference of
dark colored, cultured winter flounder ...... 103
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES
Figure Title Page
1.1 Mean percentage of winter flounder killed by crabs per day, independent of
predator size ...... 22
1.2 Size-related predation rate of green crabs on cultured and wild winter
flounder...... 23
1.3 Mean percentage of winter flounder killed by crabs per day, independent of prey
size...... 25
1.4 Mean percentage of cultured and wild winter flounder killed by crabs per day,
independent of predator and prey size ...... 26
2.1 The three sites in Great Bay Estuary, New Hampshire...... 35
2.2 Flounder cages used for the experiment ...... 37
2.3 Mean growth in length of cultured winter flounder in pens ...... 41
2.4 Mean growth in weight of cultured winter flounder in pens ...... 42
2.5 Mean percent survival of cultured winter flounder in pens ...... 43
2.6 Mean bottom water temperature at the three sites during the experiment ...... 45
2.7 Sediment composition of benthic cores taken at each of the three sites ...... 47
3.1 Mean CPUE of araphipods, mysids, sand shrimp, green crabs, and winter flounder
caught by beam trawl in all three sites during 1999-2001 ...... 59
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.2 Bottom water temperature at the three sites during 1999-2001 sampling
seasons ...... 63
3.3 Mean CPUE of amphipods caught with beam trawl in Site 2 during 1999-
2001...... 64
3.4 Mean CPUE of mysids caught with beam trawl in Site 2 during 1999-
2001...... 65
3.5 Mean CPUE of sand shrimp caught with beam trawl in Site 2 during 1999-
2001...... 67
3.6 Monthly length-frequency histograms of sand shrimp caught by beam trawl in
Site 2 during 1999-2001 ...... 68
3.7 Mean CPUE of green crabs caught with beam trawl in Site 2 during 1999-
2001...... 69
3.8 Monthly length-frequency histograms of green crabs caught by beam trawl in Site
2 during 1999-2001 ...... 70
3.9 Mean CPUE of winter flounder caught with beam trawl in Site 2 during 1999-
2001...... 71
3.10 Monthly length-frequency histograms of winter flounder caught by beam trawl in
Site 2 during 1999-2001 ...... 72
3.11 Mean CPUE of sand shrimp caught with beam and otter trawls in Site 2 during
2001...... 73
3.12 Mean CPUE of green crabs caught with beam and otter trawls in Site 2 during
2001...... 74
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.13 Mean CPUE of winter flounder caught with beam and otter trawls in Site 2 during
2001...... 75
3.14 Monthly length-frequency histograms of sand shrimp caught by beam and otter
trawls in Site 2 during 2001 ...... 77
3.15 Monthly length-frequency histograms of green crabs caught by beam and otter
trawls in Site 2 during 2001 ...... 78
3.16 Monthly length-frequency histograms of winter flounder caught by beam and
otter trawls in Site 2 during 2001 ...... 79
4.1 Experimental design of predator cue experiment ...... 88
4.2 Experimental design of bird predation experiments ...... 91
4.3 Mean distance winter flounder moved when exposed to cues from a sea
raven ...... 96
4.4 Avian predation vulnerability of cultured and wild winter flounder ...... 97
4.5 Substrate preferences of light colored and dark colored, cultured winter
flounder ...... 101
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT
WINTER FLOUNDER, Pseudopleuronectes americanus, STOCK ENHANCEMENT IN
NEW HAMPSHIRE: DEVELOPING OPTIMAL RELEASE STRATEGIES
by
Elizabeth Alden Fairchild
University of New Hampshire, May, 2002
As part of a program to assess the feasibility of winter flounder,
Pseudopleuronectes americanus, stock enhancement, optimal release size, site, season,
and condition of cultured, juvenile winter flounder were evaluated.
To determine optimal release size, the predator-prey size relationship between
winter flounder and the green crab, Carcinus maenas, was examined. The number of
flounder killed per day was significantly higher (31%) in winter flounder < 20 mm
compared to all other larger fish size classes (4-8% killed/day). Additionally, these fish
were attacked at a faster rate than any other fish size class. These results suggest that only
flounder > 20 mm should be released.
Field studies were conducted in three potential release sites in the Great Bay
Estuary during 1999-2001 to determine optimal release site and season. Optimal site
selection was based on growth and survival of caged, cultured fish in relation to water
temperature, prey availability, and sediment composition. Optimal season was selected
based on the temporal distribution, abundance, and sizes of wild flounder, and their
primary prey and predators. Within the estuary, Broad Cove was chosen as the optimal
xiv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. release site due to high fish growth rates coupled with the high prey availability and
sandy substrate. Although predators were equally abundant throughout the summer
months, early summer was determined as the most appropriate time for winter flounder
releases because prey were most abundant and wild flounder sizes were similar to the
optimal release size for cultured fish.
The condition of the cultured flounder was studied through a series of
experiments to evaluate their vulnerability to predation based on behavior, color, and
substrate preference. Cultured winter flounder reacted differently than wild flounder
when exposed to cues from a potential predator and were significantly more vulnerable to
predation by birds, regardless of fish color. Additionally, cultured flounder selected
sediments consisting of small grains and of colors matching their own pigment.
Prior to any winter flounder enhancement effort, pilot-scale releases should be
conducted to test release strategies.
xv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION
Review of Stock Enhancement
Over two thirds of the world’s oceanic fisheries are either fully or over exploited
at this time, yet the world population and demand for seafood continue to grow (New
1997). The difference between supply and demand is being fulfilled primarily through
aquaculture. An additional option to solving this dilemma is stock enhancement. In this,
fish are raised in captivity, and then released to the wild to increase abundance, promote
or accelerate recovery, and/or to ensure the survival of stocks threatened by extinction
(Cowx 1994).
Stock enhancement can be an effective tool in many cases. Stocking is a good
option for “jumpstarting” populations that have been constrained by environmentally-
induced, recruitment-limiting bottlenecks, such as unfavorable temperatures during a
critical rearing period or detrimental off-shore currents resulting in larval transport to
nutrient-poor areas. Enhancement also is a technique that has the potential to help
mitigate fish loss due to man-induced interactions such as over-fishing and power plant
entrainment. However, stock enhancement is not a foolproof mechanism for use in
replenishing all depleted fish populations. There are many cases when stocking cultured
fish will not augment the natural population. For example, if the decline in the wild
population is due to loss of habitat, enhancement alone will not solve the problem.
Habitat restoration and revitalization need to be developed first before any enhancement
effort is considered. If the carrying capacity of the habitat is already fully utilized, then
stock enhancement will not expand the wild population either. Nonetheless, stock
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. enhancement can be a useful tool to boost many declining natural populations if used in
conjunction with proper resource management.
The first attempts to stock marine finfish began in the United States in the 1870's
with the establishment of hatcheries in Gloucester and Woods Hole, Massachusetts, and
in Boothbay Harbor, Maine (Ross 1997). From 1885 to 1952, billions of cod ( Gadus
morhua), haddock ( Melanogrammns aeglefinus), pollock (Pollachius virens), and winter
flounder ( Pseudopleuronectes americanus) embryos and larvae were released in an effort
to augment wild populations. Because there were no measurable increases in landings
during this time, the hatcheries were closed and the enhancement programs were
discontinued (Richards & Edwards 1986). The failure of these early programs is not
surprising. At the time, there were no methods for marking the released fish, so there was
no way to measure their survival and growth, or to determine whether the released fish
contributed to stock rebuilding. Further, the life cycles of the fish and their environmental
requirements were relatively unknown, and there was little knowledge of how the size of
the released fish, or the timing and location of release, affected survival.
Recent declines of many marine species have caused a renewed interest in marine
fish stock enhancement (Danielssen et al. 1994; Blankenship and Leber 1995), and the
likelihood of success is much greater than in those early, ill-fated attempts. Advances in
culture technology now allow large numbers of juvenile fish to be produced, there are a
wide array of marking techniques available, there is generally a much greater
understanding of the fish's ecological requirements, and there are more sophisticated
sampling techniques with which to measure survival and growth of the released
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. individuals. Over the last 25 years, enhancement programs have been developed for
dozens of marine finfish species, including cod (Svisand et al. 1990), Japanese flounder
(Paralichthys olivaceus)(Fvkasho 1993), lingcod ( Ophiodon e/ongafes)(Marliave et al.
1997), ocean perch ( Sebastes a/u/Ms)(Polovina 1991), Pacific threadfin ( Polydactylus
sexfilis){Leber et al. 1998), red drum ( Sciaenops oce//afws)(McEachron & Daniels 1995;
McEachron et al. 1995; Willis et al. 1995), red sea bream ( Pagrus mq/or)(Masuda &
Tsukamoto 1998), striped mullet ( Mugil cephalus){Leber et al. 1996), turbot
(Scopthalmus m (Atractoscion no6//w)(Drawbridge et al. 1995). In addition, research programs are investigating the feasibility of stocking California halibut ( Paralichthys califomicus){Jirsa & Drawbridge 2000), southern flounder {Paralichthys lethostigma)(Waters 1999), and summer flounder {Paralichthys 2000). Although many of these enhancement efforts are still in the developmental phase, there have been notable results with most of these programs. Why Winter Flounder Stocks Need Help The winter flounder is widely distributed in the Northwest Atlantic, ranging from Labrador to Georgia. While it supports important commercial and recreational fisheries throughout its range, it is most abundant from the Gulf of St. Lawrence to the Chesapeake Bay. The National Marine Fisheries Service recognizes and manages three separate stocks (Gulf of Maine, Georges Bank, and Southern New England - Middle Atlantic) of winter flounder, two of which (the northern stocks) have been classified as over-exploited (Clark 1998). As with most groundfish species, catches have declined 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. precipitously in recent years. The average for US winter flounder commercial landings dropped from 17,575 metric tons (mt) in 1981 to 5,313 mt in 1997 (NEFSC 1998; 2002). The cause(s) of the declines in winter flounder catch are not well understood, but the most likely is overfishing. Fisheries regulations designed to reverse the trend of declining catch include increased codend mesh sizes, reductions in fishing effort, minimum fish size, and area closures. While it is hoped that these more stringent fisheries regulations will allow winter flounder populations to rebuild to historic levels, recovery could take a decade or more. To compound this overfishing problem, there is evidence that nuclear power plants along the northeastern US coast entrain and kill millions of winter flounder larvae when seawater is pumped into the plants for cooling (NUSCO 2001). Other causes of the declines in winter flounder include unfavorable environmental conditions for recruitment and habitat degradation. Whv Enhancing Winter Flounder Could Work The concomitant decline in US flounder catch, and documented successes with marine fish stock enhancement, led a panel of fisheries and aquaculture experts to conclude that both aquaculture and stock enhancement should be explored as means of increasing the availability of flounder in the US, and they recommended further research on flounder aquaculture and stock enhancement (Waters 1996). Winter flounder were among the species strongly suggested for further study. Winter flounder have many attributes that make it an excellent candidate for stock enhancement and/or commercial aquaculture production. First, it supports significant commercial and recreational fisheries throughout its range, and thus represents an 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. important natural resource to New England. Second, techniques for culturing winter flounder on a small (laboratory) scale have been developed (Smigielski 1975; Rogers 1976; Laurence 1977; Klein-MacPhee et al. 1982,1993; Lee & Litvak 1996; Fairchild 1998; Howell & Litvak 2000; Fairchild & Howell 2001). Given these techniques, and the wealth o f information on commercial scale production techniques that have been developed for other flounder species including turbot (Person-LeRuyet et al. 1991, Dhert et al. 1994), Japanese flounder (Fukusho 1993), and summer flounder (Nardi et al. 1996), it is likely that a minimal amount of additional research would result in the ability to grow winter flounder on a scale sufficient for stock enhancement and/or commercial aquaculture. Third, the life history characteristics of winter flounder make them an ideal candidate for assessing stock enhancement in northern New England. Most spawn in esturarine locations, and the young fish spend their first two years in or near their natal waters (Perlmutter 1947; Topp 1968), living in shallow sand and silt areas (Buckley 1989), and making short tidal excursions (Tyler 1971b). Further, Saucerman and Deegan (1991), working in a Massachusetts estuary, found that there was very little movement of the young-of-the-year (YOY) fish, with 98% of released fish recaptured within 100 m of the release site. This early estuarine dependence and non-migratory nature make winter flounder an ideal candidate for studying marine stock enhancement. Their estuarine location guarantees that field study sites are numerous and easily accessible, and that sampling of young fish can be done using inexpensive techniques. Further, because New England estuaries are extremely well studied, there is a wealth of biological, physical, and chemical information that is available to support enhancement research. The non- 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. migratory nature of the young fish ensures their prolonged availability for discrete field studies. Another reason for the selection of winter flounder is that their biology and ecology are very well known relative to most fish species. Detailed reviews of these topics are provided in Bigelow and Schroeder (1953), Klein-MacPhee (1978), Buckley (1989), and Pereira et al. (1999). This wealth of background information is essential when considering stock enhancement, and winter flounder are nearly unique in this regard. Finally, there are well-established winter flounder sampling programs in many of the New England states, so there are very good records of abundance and distribution in time and space. These historical data are invaluable in assessing the success of any winter flounder enhancement program, and the sampling done by these state agencies can supplement the sampling done by the enhancement research program. Additional data also can be collected from the extensive commercial and recreational winter flounder fisheries. Scope of This Research The central issues associated with developing and evaluating stock enhancement programs have been presented by Folkvord et al. (1994) and Blankenship and Leber (1995). They include: a) selecting appropriate species, b) formulating stock rebuilding goals, c) developing techniques for raising sufficient numbers of fish, d) assessing reason(s) for the decline of the wild population, e) developing tagging techniques, 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. f) developing a management program to prevent spread of disease, g) achieving an understanding of all genetic issues, h) developing optimal release strategies, i) assessing the performance of released fish, j) studying the ecological impact of released fish, k) developing quantitative measures to evaluate success, 1) evaluating economic feasibility, and m) involving and educating those who will benefit from the program. For any given species, simultaneous consideration of all of these issues would require a large, well-financed, multidisciplinary research program. The scope of this research relates to the development of optimal release strategies that are critical to the success of any stock enhancement program. Included in this broad issue are the selection of fish size at release, the release location (habitat type), the timing of release, and the "fitness” of the cultured fish for survival in the wild. Each component part of the release strategy needs to be fine-tuned over time (Leber et al. 199S) and to test these, small-scale releases are valuable. Size at release is important because, in general, the larger the fish, the less vulnerable they are to predation (Sv&sand & Kristiansen 1990b). Conversely, the longer the fish are held in the hatchery, the more they cost. Several studies have examined the vulnerability o f YOY flounders to predators (van der Veer & Bergman 1987; Reichert & van der Veer 1991). In general, increasing predator size increases predation rate, and small flounder are more vulnerable than larger ones. The only data available specifically 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. for winter flounder are those of Witting and Able (1993, 1995) showing that YOY winter flounder—20 mm are vulnerable to predation by the sand shrimp (Crangon septemspinosa). Beyond this experimental work, there is little known about predation on juvenile winter flounders, except that they are eaten by several species of fish (Pearcy 1962; Poole 1964; Richards et al. 1979; Fairchild 1998; Manderson et al. 2000) and birds (Pearcy 1962; Tyler 1971a). Because other studies have found that crab species are important predators on YOY flounder species, the vulnerability of variously sized juvenile winter flounder to the green crab (Carcinus maenas) was examined in Chapter 1 to determine an optimal release size. Release location is critical because the site must have the parameters appropriate for the enhanced species. Temperature, abundance of food, and absence of predators all are important factors to consider when selecting optimal release habitats (Gibson 1994). Also necessary in the release site is appropriate habitat that offers protection and refuge from predators and thus, increases post-release survival (Olla et al. 1998). This is especially important for flatfish. The diets (Pearcy 1962), substrate type (Richards 1963; Oviatt & Nixon 1973), and temperature (Casterlin & Reynolds 1982) preferences of juvenile winter flounder have been well characterized. Based on this knowledge, three release sites were selected in the Great Bay Estuary, New Hampshire for a caged cultured fish study. Chapter 2 evaluates the differences in growth and survival of the caged flounder in relation to temperature, prey availability, and substrate composition as a means to select an optimal release site. 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The timing of release is also an important issue, since release season affects prey availability, the abundance and size of potential predators, and the abiotic environmental conditions (e.g., temperature) into which the fish are released (Parker 1986, Willis et al. 1995, Leber et al. 1997). The optimal time for a release is the period when prey are most prevalent but predators are less prevalent (Koshiishi et al. 1991; Folkvord et al. 1994). For juvenile winter flounder, this is a period of abundant amphipods and mysids, but few sand shrimp and green crabs. In Chapter 3, the results of an extensive benthic trawling and coring program in the three release areas are analyzed and discussed to predict an optimal release month for winter flounder stockings. The last component of optimizing any release strategy is ensuring that the cultured juveniles are "fit", meaning that they are morphologically, behaviorally, and physiologically similar to wild juvenile fish. Because intensive hatchery conditions are so unnatural, cultured fish, although genetically identical to their wild counterparts, can differ in many significant ways. Many studies have focused on the morphological (Blaxter 1970), behavioral (Howell 1994; Tsukamoto et al. 1997), and physiological (Sosiak et al. 1979; O’Grady 1983; Parker 1986; Stradmeyer & Thorpe 1987; Folkvord et al. 1994) differences between cultured and wild fish. If these abnormalities affect the fish's ability to avoid predators, their chances of survival are diminished. The studies in Chapter 4 build upon previous research (Fairchild 1988) by continuing the investigation into the fitness of cultured winter flounder. In this new work, the vulnerability of cultured, juvenile winter flounder to piscivorous and avian predators is assessed. Behavioral differences between cultured and wild fish in the presence of a predator are 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. evaluated; the relationship between fish coloration and predation vulnerability is explored; the mortality of cultured and wild flounder are compared; and the relationship between fish color and substrate choice is examined. Once an optimal release strategy has been determined, the ultimate way to measure its effectiveness and the performance of the cultured fish is by pilot-scale releases. These small, experimental stockings are conducted to simulate the events of a large-scale enhancement effort. If release methods are found to be inferior, there is still time for adaptation prior to the larger stocking. Appendix A reviews the pilot-scale releases of cultured, juvenile winter flounder in the Great Bay Estuary during 1999-2001. This body of research lays the framework to optimizing winter flounder releases. There are many details, however, that still need to be addressed pertaining to optimal release strategies. These range from improving hatchery protocols to raise larger numbers of fitter fish, to determining the best mode of transporting and stocking the fish to minimize stress, to establishing guidelines for the optimal time of day and tidal cycle for releases. 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER I PREDATOR-PREY SIZE RELATIONSHIP BETWEEN Pseudopleuronectes americanus AND Carcinus maenas Introduction It is generally accepted that mortality is density-dependent (Cushing 1974) and size-selective (van der Veer et al. 1997) for newly-settled flatfish. Most fish species are vulnerable to a series of predators during the early life history stages, and the suite of predators generally changes with fish size. The relative sizes of the predator and its prey are important to the dynamics of size-dependent mortality. If the prey is too small, it may yield insufficient energy to the predator and thus may be overlooked. If the prey is too big, it may be inaccessible to the predator based on physical limits, such as exceeding gape size of the predator or having better escape abilities. Additionally, smaller predators may spend a longer time handling their prey than larger predators, independent of prey size. Thus prey profitability increases with decreasing prey size and increasing predator size (Gibson et al. 1995). In general, however, smaller fish are more vulnerable and require a shorter handling time than larger fish to the same-sized predator (Gibson et al. 1995). In this negative size-selective predation, mortality risks decrease as individual prey sizes increase (van der Veer et al. 1997). Therefore, it is advantageous for the fish to grow II Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rapidly toward a "size refuge" in which they are no longer vulnerable to certain predators (Zijlstra et al. 1982). It follows that the inability to grow quickly, due to density-dependent mechanisms, prolongs the exposure to predation and can affect population dynamics. In addition to prey growth rates, the relative growth rates of the predators will impact the predator-prey interaction by altering the size-selective prey mortality (Rice et al. 1997). The behavior (i.e. activity level) of both the prey and predators is also important in determining the outcome of predator-prey interactions (van der Veer et al. 1997). All these mechanisms appear to be important in flatfish (Cushing 1974; Bergman et al. 1988; Rice et al. 1997; van der Veer et al. 1997). Decapod crustaceans are important predators of a number of species of flatfish in a number of areas. The most documented crustacean predator of YOY flatfish is the crangoid shrimp, Crangon spp. For example, in the Western Wadden Sea, plaice (P. platessa) otoliths have been found in Crangon crangon guts from March to May when YOY plaice are most abundant on the tidal flats (van der Veer and Bergman 1987), and high fish mortality is attributed to these shrimp (Zijlstra et al. 1982). In Sendai Bay, Japan, Yamashita et al. (1996) have found stone flounder ( Kareius bicoloratus) otoliths in C. afinis guts where the two species co-occur in the flounder nursery grounds. Similarly, along the western Atlantic coast, C. semptemspinosa have been observed feeding on fresh animal tissue and fish scales have been found in their guts (Wilcox and Jeffries 1974). These field observations are supported by extensive laboratory experiments which have investigated the size dependent nature of the predator-prey relationship. Not only do 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the studies provide evidence of crangoid shrimp predation on YOY flatfish, but of crab predation as well, van der Veer and Bergman (1987) were the first to show under experimental laboratory conditions the size relationship between YOY plaice as prey and C. crangon and green crabs ( Carcinus maenas) as predators. They found that plaice 4 30 mm TL are susceptible to shrimp predators, whereas 16-35 mm TL plaice are vulnerable to 26-50 mm carapace width green crabs. Using predator inclusion cages in the field, Reichert and van der Veer (1991) demonstrated that in southern estuaries in the United States, 40-70 mm blue crabs (Callinectes spp.) are an important predator of fringed flounder ( Etropus crossotus), bay whiff (Citharichthys spilopterus), and black cheek tongue fish ( Symphurus plagiusa) between 20 and 75 mm TL. In the laboratory, Seikai et al. (1993) demonstrated that Japanese flounder ( Paralichthys olivaceus) — 25 mm TL are prone to C. ajinis predation. Witting and Able (1993) showed that YOY winter and summer flounders, Pseudopleuronectes americanus and Paralichthys dentatus, respectively, are vulnerable to predation by the sevenspine bay shrimp (C. septemspinosa). Further work revealed that P. americanus — 20 mm are vulnerable to 47-74 mm C. septemspinosa (Witting and Able 1995). Because intensive hatchery conditions are so unnatural, cultured fish can differ in many significant ways from their wild counterparts. Many studies have focused on the morphological, behavioral, and physiological differences between cultured and wild fish (e.g., Blaxter 1975; Parker 1986; Svasand 1993; Howell 1994). Morphological differences can include abnormal pigmentation of cultured fish (Houde 1971), and a variety of other abnormalities (reviewed in Blaxter 1975). If these abnormalities affect the fish’s ability to feed or to avoid predators, their chances of survival are diminished in 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. an enhancement program. Studies of behavioral differences have focused on feeding and predator avoidance behaviors. It has been shown that cultured fish experience sensory deprivation, which can lead to the fish’s inability to recognize, react to, and avoid predators upon release. Deficiencies in anti-predator behavior have been shown for a number of species, including zebra danios (Brachydartio rerio){Dill 1974), European minnows {Phoxinus p/toxznus)(Magurran 1990), and coho salmon (Oncorhynchus kisutch)(Va.\Xen 1977). Physiological differences, induced by the very different rearing environments of cultured and wild fish, can occur as well (Stradmeyer and Thorpe 1987; Folkvord et al. 1994; Blankenship and Leber 1995). It is suspected that such differences may exist between cultured and wild juvenile winter flounder too. Along the western Atlantic coast where it is an introduced species (Miller 1996), it has been suspected that green crabs are predators of YOY winter flounder. Saucerman (1990) cited green crabs as a possible predator in her study on winter flounder habitats in a Massachusetts estuary and Witting and Able (1995) suggested an investigation of predation by crustaceans, including C. maenas, on winter flounder. However, to my knowledge, this predator-prey relationship has never been investigated. The intent of this study was fourfold: (1) to confirm that green crabs are predators of YOY winter flounder; (2) to examine if the predator-prey relationship is size- dependent under laboratory conditions; (3) to study whether potential mortality decreases with increasing fish size; and (4) to determine if cultured fish are more vulnerable to crab predation than wild fish. 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Materials and Methods To examine the predator-prey size relationship, a (6x6)+(4x6) factorial design experiment was conducted at the University of New Hampshire's Coastal Marine Laboratory (CML) in which two types of winter flounder (cultured and wild) were tested. Six flounder size class treatments (11-20,21-30,31-40,41-50, 51-60,61-70 mm TL) were tested against 6 green crab size class treatments (11-20,21-30, 31-40,41-50, 51-60, 61-70 mm carapace width). Two to nine replicate trials were conducted for each size combination and a control treatment of flounder only (no crabs) ensured all mortality was predator-related. Cultured winter flounder were reared at the CML from a wild broodstock brought into the laboratory in April 1998 (per methodology in Fairchild 1998). The cultured fish used in the trials ranged from 3 months (11-20 mm) to 7 months old (61-70 mm). They were fed a formulated diet daily. Wild winter flounder, ranging in size from 31-60 mm, were collected using a beach seine in September 1998 in Salem Harbor, Massachusetts. Wild flounder were brought back to the CML where they were allowed to acclimate for at least one week prior to use in the experiment. These fish were fed a diet of chopped Mytilus edulis daily while in the laboratory. All green crabs were collected locally either by capture in lobster traps or by searching in the intertidal zone at low tide. Upon capture, the crabs were brought to the CML, placed in individual containers, and starved for at least 48 hours prior to the trials to ensure that they would be hungry. 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. One crab of each size class (measured to the nearest millimeter) was placed into a white, plastic aquarium measuring 38x27x4 cm in the late afternoon. Each aquarium contained 1-2 cm of mud-silt substrate collected from an area where YOY winter flounder are known to occur (personal observation). The sediment was sieved to a diameter of 68-S00p so that even the smallest fish would be able to bury (Tanda 1990; Gibson and Robb 1992; Moles and Norcross 1993). Aquaria were covered with dark window screening to prevent the escapement of crabs. Light from a nearby window provided a natural photoperiod. Black plastic curtains were suspended around the tanks to prevent artificial laboratory light from penetrating the experimental area. All aquaria were placed in a water table and supplied with individual water lines that connected to a flow-through seawater system. Water temperature was recorded daily. After the crab acclimation period of 24 hours, ten YOY winter flounder from a single size class and fish type (cultured or wild) were released into each aquarium in the late afternoon. All fish were measured to the nearest millimeter prior to release to ensure proper size classification. After 24 hours, all crabs were removed and the surviving fish were counted and removed. The number of fish killed/day in each trial was tested against crab size and fish size in two-way analyses of variance followed by Bonferroni posterior tests. This measure of mortality was used in the analyses rather than the number of fish consumed/day since all fish killed were not completely eaten by the crabs in all trials. Data for cultured and wild fish were analyzed independently of each other because of the discrepancy in the levels of fish size classes used in the experiment (cultured fish = 6 fish size classes; wild fish = 3 fish size classes). Additionally, all cultured and wild fish killed 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were pooled together and tested against temperature (one-degree Celsius increments) in a one-way analysis of variance to determine if temperature had an effect on crab predation. Results A total of 269 trials were conducted: 183 with cultured fish and 86 using wild fish. For each predator-prey size combination tested, Tables 1.1 and 1.2 report the mean number of cultured and wild flounder killed/day (+/- one standard deviation) and the number of trials upon which the mean is based. For both cultured and wild fish, significant (p<0.05) interactions existed between the 2 variables tested (crab size and fish size)(Table 1.3). Temperature ranged from 9-17.5 °C over the four month experimental period, with a mean of 13.4 °C. Table 1.4 reports the mean number of flounder killed/day (+■/- one standard deviation) per degree Celsius temperature increment and the number of trials. The mean number of fish killed (both cultured and wild fish pooled) did not vary (ANOVA; p<0.05) between degree increments within this temperature range. Fish Size Cultured fish smaller than 20 mm were most vulnerable to crab predation (p<0.001)(Figure 1.1) and were killed at a faster rate than any other fish size class (Figure 1.2). From this smallest fish size class, 30.7% were killed/day by crabs (Figure 1.1). There were no significant differences in the numbers of cultured fish killed/day in any of the other fish size classes. Percentages ranged from 3.3 to 7.3%. Of the wild fish size classes, more 31-40 mm fish were killed (p=0.03) than the two larger fish size classes (Figure 1.2). Mortality decreased from 16.4 to 6.8% as fish size increased for the different wild fish/crab size combinations (Figure 1.1). 17 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 5 6 control Carcinus 4 5 5 4.40 0.000.60 0.00 (1.95) (0.00) (1.89) (0.00) killed per day by 5 5 5 0.40 (1-1) (0.55) (0.89) (0.00) (1.17) . . 4.80 5 6 Crab Size (mm) 3.80 0.80 31-40 41-50 51-60 (0.84) (1.00) (1-79) (2.12) (0.00) (0.45) (0.55) (0.55) (0.00) (0.55) (0.53) (1.73) (0.00) Pseudopleuronectes americanus 5 5 5 5 5 5 5 5 9 3 0.40 1.00 0.80 2.00 0.00 0.80 0.60 0.56 2.00 0.00 0.40 0.80 0.60 1.60 0.00 (0.55) (0.55) (0.45) 5 4 6 1.00 4.60 0.00 0.20 0.00 0.00 0.00 (0.82) (1.82) (1.92) (0.00) (0.48) (0.00) (0.00) (0.00) (0.55) (0.00) (0.84) 41-50 0.00 0.60 0.80 1.17 1.25 0.00 Fish Size (mm) 11-20 21-30 in each size class combination. Ten flounder were used in each 24-hour trial. Trials TrialsTrials Trials 5 5 5 5 5 5 5 5 Trials Trials 5 5 5 5 (+/• s.d.) (+/• s.d.) (+/• s.d.) (+/• s.d.) (+/• s.d.) (+/• s.d.) Mean No. Killed 21-30 Mean No. Killed 11-20 Mean No. Killed 31-40 Mean No. Killed Mean No. Killed 51-60 Mean No. Killed 61-70 Mean number (+/• one standard deviation) of cultured Table 1.1. maenas Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 0.00 0.00 0.00 (0.00) (0.00) 25 5 3 5 1.00 2.40 (2.61) (0.00) 5 6 killed per day by green crabs 1.33 1.00 3.57 2.00 (1.21) Crab Size (mm) 1.80 2.50 0.50 31-40 41-50 51-60 control (1-73) (2.37) (2.83) (1 4 8 ) 1.00 1.20 21-30 (0.45) (0.00) Pseudopleuronectes americanus 0.00 0.40 11-20 (0.00) 41-50 Fish Size (mm) Trials 5 5 5 Trials 5 5 4 7 Trials 5 6 4 (+/• s.d.) (+/• s.d.) (0.55) (+/• s.d.) (0.00) (0.84) (0.58) (1.22) (1.00) Mean No. Killed 31-40 Mean No. Killed Mean No. Killed 51-60 0.00 1.50 in in each size class combination. Ten flounder were used in each 24-hour trial. Mean number (+/• one standard deviation) of wild Table 1.2. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1.3. Effects of crab and fish size on number of cultured and wild Pseudopleuronectes americanus killed. Source of variation 2Way ANOVA df Sum of squares F-Value p-Value (a) Cultured fish Crab size 5 76.94 15.72 0.00 Fish size 5 158.57 32.40 0.00 Crab x fish size 25 57.79 2.36 0.00 Residual 149 145.85 (b) Wild fish Crab size 5 59.11 8.71 0.00 Fish size 2 13.00 4.79 0.01 Crab x fish size 10 34.03 2.51 0.01 Residual 69 93.67 11 16 10 43 killed/day 1.64 0.77 Pseudopleuronectes americanus 1.33 1.46 21 0.73 0.65 0.75 0.83 0.92 24 in predator-prey trials in relation to temperature. Control 9.0-9.9 Carcinus maenas 10.0-10.9 11.0-11.9 12.0-12.913.0-13.9 1.70 1.42 2.18 27 14.0-14.915.0-15.916.0-16.917.0-17.9 1.30 1.62 0.55 1.83 2.04 0.99 42 29 o Celsius Mean No. Fish Killed/Day Standard Deviation No. Trials Mean number (standard deviation) of by Table 1.4. trials (e.g. no crabs present) are excluded. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Pseudopleuronectes americanus (Pseudopleuronectes Figure 1.1. Mean percentage (+/- one standard deviation) of 11-70 mm cultured and 31-60 mm wild winter flounder flounder winter wild mm 31-60 and cultured 11-70 mm of deviation) standard one (+/- percentage 1.1. Mean Figure Percent Killed/Day 20 40 30 60 10 50 I 11-20 ) killed by green crabs crabs green by killed ) T 13 14 41-50 31-40 21-30 FishSize (mm) Wid Fish ild □W 0 Cultured Fish Fish Cultured 0 (Carcinus maenas (Carcinus ) per day, independent of predator size. predator of independent day, per ) 51-60 61-70 51-60 41-50 31-40 Wild Wild Fish 21-30 Carapace Width (mm) Width Carapace 11-20 on cultured and wild flounder 5 3 1 2 4 0 c o o c a> <0 o ). ). The numberof replicate trials varied between 2 and 9 for each combination. 41-50 31-40 21-30 Cultured Fish Carapace Width (mm) Width Carapace 11-20 5 3 1 4 2 0 • o c a» Reproduced with permission of the copyright owner. Further reproduction prohibited without permission Crab Size All sizes of crabs preyed on both cultured and wild winter flounder (Figure 1.3). Crabs > 20 nun killed significantly more cultured winter flounder (p ±0.15) than the smallest sized crabs and the control treatment. In addition, the largest crabs killed the greatest percentage of cultured flounder (20%). Significantly more wild fish were killed by crabs > 30 mm (p £ 0.04) with the two largest crab sizes killing the most fish (41-50 mm crabs killed 21.1% of wild fish; 51-60 mm crabs killed 19.0% of wild fish). Fish Type For the 31-60 mm fish size classes tested, more wild fish (11%) were killed by crabs than cultured fish (6.3%) (t-test, p=0.003, Figure 1.4). The two smallest and the largest fish size classes were not compared because wild flounder of these sizes were not available. In particular, for the 31-40 mm size class, crabs preyed on wild fish at a faster rate than cultured fish (Figure 1.2). Discussion Predator/Prev Size Relationship In this research on cultured P. americanus, smaller fish were more vulnerable than larger fish to predation by C. maenas (Figures 1.1 and 1.2). Although trials using large fish-large crab combinations increased the density in the aquaria, and therefore may have increased the encounter rate between predator and prey, the mortality rate was still much lower than in trials using small flounder (Figure 1.2). In studies with P. platessa and C. crangon, the flounder’s vulnerability to the predator was associated with the fish's size, rather than the predator’s; larger fish had better escape abilities than smaller fish (Gibson et al. 1995). This theory corresponds to the predation rate of crabs on cultured 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51-60 i (Pseudopleuronectesamericanus) perday, independent ofprey size. A control treatment containing no crabs Crab Carapace Width (mm) CarapaceCrab Width zk 11-20 21-30 31-40 41-50 r/ / / , , I (Carcinus maenas) □ Wild □Fish Wild 0 Cultured Fish control 20 40 0> O Q> 4) CL 10 CL -a Figure Mean 1.3. percentage (+/- one standard deviation) ofcultured and wild winter flounder killed by mm 11-60 green crabs ensured all mortality was crab-related. Is) Ut Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ) ) killed (Pseudopleuronectes americanus (Pseudopleuronectes Cultured Fish Type ^999999 per day, independent ofpredator and prey size. Carcinus maenas ) maenas Carcinus Figure Mean 1.4. percentage (+/- one standard deviation) ofcultured and wild winter flounder by green crabs ( N> O Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. flounder, however, the response of the wild fish is not as clear. Wild fish of the 31-40 mm size class were killed at a faster rate than cultured fish of the same size (Figure 1.2). It is unknown whether this rate is accurate or if the predation rate decreased due to a low sample size for the 31-40 mm fish/51-60 mm crab trial (Table 1.2). Alternatively, the mean of 3.57 31-40 mm fish killed by 41-50 mm crabs (Table 1.2) could be unnaturally high and thus, exaggerated the predation rate. If predator-prey trials had included wild fish < 30 mm, comparisons of both wild and cultured flounder predation rates would be more complete. However, the results from this experiment indicate that larger P. americanus can escape predation from a larger size range of crustacean predators than smaller P. americanus. In this study, larger crabs killed more flounder than smaller crabs (Figure 1.3). This higher proportion of fish killed may be an indication of the larger crab's higher metabolic demands (Waterman 1961), better ability to catch fish, and lower prey handling time. Crabs > 20 mm killed more cultured flounder than the smallest crab size class. Crabs > 30 mm killed more wild flounder than the smaller ones. Crabs < 20 mm seem less capable of catching, killing, and consuming live fish than the larger crabs. Gibson et al. (1995) found that P. platessa size was more important than the predatory C. crangon size in predator-prey interactions. It is possible that this is true of winter flounder-green crab interactions as well. Smaller crabs probably experienced a longer handling time of flounder prey than larger crabs did, and thus were restricted by time as well as digestion rates. Because crabs are poikilothermic, their metabolic rate varies with ambient temperature. C. maenas normally cease feeding at 6-7 °C (Ropes 1968; Cohen et al. 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1995) which is well below what they experienced in this study. Since the coolest water temperature in the trials was about 9 °C, the experimental temperature range did not significantly affect the rate of activity of the crabs. No differences were found between the mean numbers of flounder killed by crabs at any temperature degree increment. However, more than likely, photoperiod affected the crab's feeding rate. The predator- prey trials were run for 24 hours to cover complete, natural circadian and tidal cycles. Although data were not collected to analyze predation rates in relation to photoperiod, more fish should have been consumed during dark periods than light periods based on the behavior of the green crab. C. maenas exhibits diurnal and tidal rhythms of activity, with peaks occurring during night and at high tides (Naylor 1958; Lockwood 1967; Crothers 1968; Ropes 1968). Ansell and Gibson (1993) found in their laboratory experiments that predation on P. platessa by the portunid crab Liocarcinus holsatus was 2-3 times higher in dark than daylight. In the field, however, this is harder to confirm (van der Veer and Bergman 1987). Due to the cyclic behavior of both predators and prey, it can be inferred that predation by crustaceans in flatfish nursery grounds is probably highest at night. Sediment was used in the aquaria to mimic a natural environment. Ansell and Gibson (1993) found that the presence of sand had no effect on P. platessa predation rates by C. crangon and L. holsatus. While the sand provided some refuge, the shrimp were still able to detect the plaice. This too is probably true for C. maenas since they are not primarily visual hunters but rely on mechanoreceptors on their antennules and waiking-Ieg joints to sense water currents and vibrations, respectively (Waterman 1961). Green crabs also hunt using chemoreception (Waterman 1961), and readily detect 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. shallow buried infauna (Cohen et al. 1995). Therefore, the sediment in this research probably did not affect the crab's ability to locate the flounder. In addition to searching with receptors, the encounters between the flounder and crabs could have been a matter of random chance. Gibson et al. (1995) found that in predator-prey size trials with shrimp and plaice, the encounters were random by both organisms. One of the two animals would "blunder" into the other. It follows, then, that the encounters, and thus capture rates, were dependent upon the activity levels of the organisms. This may, in part, help to explain why wild P. americanus were unexpectedly more vulnerable to crab predation than cultured fish (Figure 1.4). It is possible that the wild fish were more active than the cultured fish, and therefore, encountered the crabs more frequently. Of course this difference in vulnerability to crab predation may also have been an artifact caused by the experimental design. The wild fish probably experienced some stress due to being captured and transported to the CML, as well as stress due to the unnatural laboratory environment. In addition, they experienced a rapid diet change from their natural diet to one of chopped Mytilus edulis which may have been suboptimal. For these reasons, it is difficult to interpret the increased vulnerability of the wild fish. This unexpected discrepancy between cultured and wild P. americanus warrants further studies. Differences in cortisol levels, metabolic rates, and other stress- related physiological indices should be quantified in future experiments. Implications for Stock Enhancement Winter flounder, a promising candidate for stock enhancement in the northeastern U.S., is being studied to assess the feasibility of an enhancement program in New Hampshire (Howell and Fairchild, SeaGrant R/FMD-158). One of many issues to 29 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. consider when planning an enhancement program is how to develop release strategies that reduce post-release mortality (Blankenhip and Leber 1995). Several causes of high post-release mortality have been identified including predation (Sv&sand 1998). Knowledge of potential predators and the sizes of newly stocked fish they are capable of consuming is, therefore, of paramount importance in the planning of a stock enhancement program. In general, the larger the fish, the less vulnerable they are to predation (Sv&sand and Kristiansen 1990b); conversely, the longer the fish are held in the hatchery, the more they cost to produce. Therefore, in determining the optimal release size for stocking the cultured fish, a balance must be reached between the increased cost of retaining the fish in the hatchery (rearing larger fish) and the advantage that larger, less vulnerable fish will have upon release (higher survival). The results of this laboratory study, along with those of Witting and Able (1995) suggest that the optimal size at release is >20 mm for cultured winter flounder. Until complimentary field studies are conducted, this optimal size at release is a preliminary estimate. Summary From this research, it is evident that predation is highest on the smallest fish by the larger crabs. The larger crabs killed more fish than the smaller crabs, and fish <21 mm suffered the highest predation. In this study, cultured flounder were less vulnerable to crab predation than wild flounder, however, this pattern may be due to increased stress experienced by the wild flounder. Based on this research and studies done on Crangon predation (Witting and Able 1995), the following preliminary guidelines are given to optimize the survival of released winter flounder in a stock enhancement program: (1) Release winter flounder during the 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. day at a tide other than high when crab activity is lowest; (2) Release fish > 20 nun since this size is more adept at escaping decapod crustacean predators and; (3) If possible, releases should be timed when smaller crabs are more prevalent than larger ones. While this research focused on decapod crustacean predation on winter flounder, it is important to note that with winter flounder, as well as all other flatfish species, the suite of predators does not end with crabs. From crustacean predator mortality, the P. americanus then are vulnerable to other larger predators, such as piscivorous fish (Tyler 1971c; Klein-MacPhee 1978; Buckley 1989) and birds (Klein-MacPhee 1978; Buckley 1989; Fairchild personal observation). These larger predators demand consideration for any future enhancement effort. 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER n DETERMINING AN OPTIMAL RELEASE SITE FOR JUVENILE Pseudopleuronectes americanus IN THE GREAT BAY ESTUARY, NEW HAMPSHIRE Introduction Identifying optimal release sites is one of the main criteria in developing an effective stocking strategy (Blankenship & Leber 1995; Yamashita & Yamada 1999). Temperature, abundance of food, and absence of predators all are critical parameters to consider when selecting optimal release habitats (Gibson 1994). Releases should be matched to areas with high prey availability and at a time when there is the increased opportunity of encountering prey (Folkvord et al. 1994; Masuda & Tsukamoto 1998; Yamashita & Yamada 1999). Appropriate habitat also is necessary in the release site to offer protection and refuge from predators and to increase post-release survival (Olla et al. 1998). Factors such as presence or absence of microalgae (Wennhage & Pihl 1994), sediment size and color (Yamashita & Yamada 1999), physical structures for refuge from predators or currents should be categorized. For flatfish cryptic behavior, substrate is. especially important. Selection of poor release sites could diminish the effectiveness of a stocking program by negatively affecting the survival and distribution of the released fish. 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dispersion paradigms often are linked to the quality of the release site. Generally, the greater the dispersion, the less favorable the release site is. For example, when stocked into coastal areas with high prey productivity in Denmark, juvenile turbot (Scophthalmus maximus) maintained high site fidelity (Stettrup et al. 1998). In Hawaii, cultured striped mullet ( Mugil cephalus) showed a strong affinity to the bays where they were released with little to no dispersion from their initial site (Leber & Lee 1997). Recovery rates exceeding 75% were recorded for Japanese flounder in the original release bay (Tominga & Watanabe 1998). Similarly, >96% of cultured cod ( Gadus morhua) were recaptured within 2 km of their release site (Svlsand & Kristiansen 1990a). However, Pacific threadfin ( Polydactylus sexfilis) were highly sensitive to release sites in Hawaiian bays and showed higher dispersion from sites depending on size and season of release (Leber et al. 1998). Likewise in Spain, the location of recaptured turbot varied drastically based on release sites (Iglesias & Rodriguez-Ojea 1994). Prior to a full-fledged stocking, experimental studies should be conducted to evaluate the effectiveness of the proposed release strategies (Blankenship & Leber 1995). In this manner, small-scale releases and studies can illuminate potential problems and then adaptive management (Leber 1999) can be employed to alter and improve the release techniques. For discrete, short-term flounder studies, cages can be useful tools. Sogard (1992) and Phelan et al. (2000) have employed cages to study juvenile winter flounder growth within a New Jersey estuary. Cages also have been used in a North Carolina estuary to examine critical nursery habitats (Guindon & Miller 1995), food availability, and growth (Kamermans et al. 1995) of southern flounder ( Paralichthys lethostigma). However, they must be used with caution as they can bias the results of a study as Kamermans et al. (1995) found. By definition, cages isolate the fish to one area 33 Reproduced with permission of the copyrightowner. Further reproduction prohibited without permission. which may contain sub-optimal properties and restrict their movements. For winter flounder that take advantage of tidal cycles to move in and out of the intertidal zone to forage (Tyler 1971c), caging may adversely affect their growth and survival. Cages also may disrupt the sediment, affecting benthic organism populations, and possibly, reduce the movement of prey into the cages. Although, juvenile winter flounder are ubiquitous throughout northeastern estuaries during the summer and fall, it is logical to choose a release site that has the optimal properties to ensure maximum fish growth and survival. Proposed release sites should be surveyed by collecting information on abundance of prey and predators, habitat type, and physical conditions (Tsukamoto et al. 1999). Following this rationale, I set out to determine if one of three areas in the Great Bay Estuary, New Hampshire was best for the release of juvenile winter flounder. Optimal site selection was based on growth and survival of caged, cultured flounder in relation to water temperature, prey availability, and sediment composition. Materials and Methods Three possible release locations (Figure 2.1) were selected in the Great Bay Estuary System, NH. Site 1, located on the west side of New Castle Island, was closest to the coastal terminus of the estuary. The other two sites at Broad Cove, Newington (Site 2) and Oyster River, Durham (Site 3) were located in the middle portion of the estuary. These sites were chosen due to their physical and biological characteristics. Preliminary sampling suggested that good juvenile winter flounder habitats existed at each site. All sites were contained by small coves (~ 100 m across) with some muddy-sand substrate present Precursory sampling showed adequate food items and the presence of wild 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43 10 Maine 1 Km New Hampshire Figure 2.1. Three potential release sites in Great Bay Estuary, New Hampshire evaluated for winter flounder stock enhancement. Site 1 is located to the west of Great Island, in New Castle, NH; Site 2 is Broad Cove in Newington, NH; Site 3 is a cove in the Oyster River, Durham, NH. 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. juvenile winter flounder. Further, the sites were bounded by deep channels (2-3 m) that tend to hold fish in the shallower coves (Saucerman & Deegan 1991). The field experiment was conducted from 7 July - 25 August 1999, using nine pens as the experimental units at each site, for a total of 27 pens. The l-m^ open-bottom pens, designed and built specifically for young-of-year (YOY) winter flounder by the National Marine Fisheries Service, Milford, Connecticut Laboratory, were constructed of a welded steel and wood frame with nylon mesh sides and a bolted, removable top (Figure 2.2). Along the bottom edge, 15 cm vertical steel plates extended into the substrate to prevent fish from escaping, while still allowing them to feed on the benthos, and prevented predators from entering. Pens were placed in the field one month prior to the onset of the experiment to ensure that any disturbances to the benthic community caused by positioning the pens in the substrate had time to recover. To verify they were securely set in the sediment, the bottom edges of the pens were examined in situ using SCUBA. Immediately prior to stocking, the contents of the pens were seined to remove any potential predators such as Crangon sp. and Carcinus maenas. A HOBO1 was attached to the base of one of the pens at each site to record bottom water temperature at the site every 36 min. Tagging All fish were differentially color tagged with Visible Implant Fluorescent Elastomer Tags2 injected subcutaneously on the ventral side of the fish. Eleven percent of the tagged fish (n=15) were held in the laboratory for the duration of the experiment to 1 Onset Computer Corporation, P.O. Box 3450, Pocasset, MA 02559-3450. 2 Northwest Marine Technology, Inc., P.O. Box 427, Shaw Island, WA 98286. 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2.2. Open-bottom lmJ winter flounder pens. The pens were constructed of welded 1 cm rebar and wood, covered by a 3 mm mesh size vexar. Removable wooden and vexar tops were secured to the tops of the frames with bolts. Fifteen centimeter steel plates extended into the substrate to anchor the pens and and to prevent fish from escaping and predators from entering. 37 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. examine tag retention rates and associated mortality. Fish used in the field experiment were held in the laboratory for 48 hours before placement into the experimental pens to monitor any tagging-induced stress and/or mortality. Growth On 7 July 1999 at the beginning of the experiment, all fish were measured to the nearest 0.1 cm and weighed to the nearest 0.1 g. Sampling continued weekly, at low tide when the pens were exposed, by seining the contents of each pen. After a total of 10 seine attempts, it was assumed that all surviving fish were captured. Captured fish were measured, weighed on a portable, digital balance, and returned to their respective pens. At the end of the 7-week experiment on 25 August 1999, all remaining fish were collected and final measurements were taken. A mean growth value for length and weight was calculated for each pen. Using these data, differences in mean growth rates between sites were examined by analyses of variance followed by Tukey's posterior test. Additionally, mean daily growth rates were calculated for fish at each site. Survival At the end of the 7-week experiment, all remaining fish were removed from the pens and counted. Percent survival data were arcsine transformed (Zar 1996) and survival was determined based on the final number of recovered fish. Differences in survival between the three sites were tested with analysis of variance. Prey Availability To track prey availability, initial and final core sediment samples were taken at the beginning and end of the study. On each occasion, one core from within and one from outside of each pen were taken with a coring tube (7 cm diameter) to a depth of 5 cm. 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sediment samples were sieved through a 500 pm screen. The residue was fixed with 10% buffered formalin and dyed with Rose Bengal for a minimum of 24 hours. Prey organisms were identified to lowest possible taxon, counted, wet weighed to the nearest 0.001 gram, and preserved in isopropyl alcohol. To examine if principal prey items of juvenile winter flounder (as determined by Fairchild 1998) changed over time, least squares regression analyses were employed. Both numerical abundance and the biomass of prey organisms in the sediment samples were used. Sediment Composition An additional core sample was taken from within each pen to determine the grain size composition of the substrate. Each core sample was processed in the following manner. The sample was homogenized and one tablespoon of the sediment was mixed with 50 ml HC1. After several weeks when the acid had broken down the organic contents of the sample, 250 ml deionized water were added to each solution. This in turn was wet sieved with dispersant solution through a 2 mm sieve suspended over a 63 pm sieve. The remaining filtrate was set aside in a measuring cylinder for future use. Any gravel present in the sample remained on the larger sieve, was dried overnight in an oven, and weighed. The smaller particles on the 63 pm sieve also were dried overnight in an oven and then weighed. These particles were further processed by dry sieving through the 2 mm sieve suspended over the 63 pm sieve. Any sand present in the sample remained on the smaller sieve and was weighed. The remaining particles < 63 pm were weighed and added back to the measuring cylinder containing filtrate and dispersant solution. After stirring for 15 min with a drill, the solution was left to settle for 20 s and the settled matter was removed by pipette, dried overnight, and weighed to determine the amount of 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. silt present. Once the silt was removed, the remaining solution was allowed to settle for 63 min and the settled matter was removed by pipette, dried overnight, and weighed to determine the amount of clay present. The mean percent gravel, sand, clay, and silt at each site were calculated. Percent sediment composition data were arcsine transformed (Zar 1996). Differences in particle sizes between the three sites were tested with analyses of variance followed by Tukey’s posterior tests. Results Tagging O f the 15 tagged fish retained in the laboratory, none lost their tags during the course of the experiment. Additionally, no tagged fish died during the seven weeks in the laboratory. All fish recovered from the pens were identified by their tags. Growth Over the course of the 7 weeks, all fish grew both in length and weight (Figures 2.3 and 2.4). Fish from the upper estuarine sites (Broad Cove and Wagon Hill) were significantly longer (p ^ 0.001) and heavier (p<0.001) than fish from the site at the mouth of the estuary (New Castle) from week 2 to week 7. Mean daily growth rates at Sites 1,2, and 3 were 0.37,0.54, and 0.56 mm/d, respectively. Survival There was a noticeable loss of fish from the pens. A total of 70 of 135 fish (52%) were not recovered by the end of the experiment. In Newcastle, only 44% of the fish were recovered; 49% at Broad Cove; and 53% at Wagon Hill (Figure 2.5). Some of the pens had holes in the mesh at the end of the experiment through which fish could have escaped or crabs could have entered and consumed the fish. Based on final numbers of 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pseudopleuronectes Site 3 11 • 11 if,- fij1 r'a-rf.’.H- . >•'. W eek **•... Site 2 . . . 1 ;.p ;.p - • ...... - ♦- Site 1 • ‘ o w » • : r 1 -'.c -'.c ' ' . i . “ ' ,l- held in pens atthe three sites. ' *• - 1 r . ' 0 1 2 3 4 5 Figure 2.3. Mean growth (+/- one standard errorof the mean) lengthin of cultured americanus Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. iO in i op S » e £ 002 • u s ■'T cm’ 2 • pens in at the three sites. 3 -Q 00 — U. .3O (3) WSjaM 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 held in 6 5 Pseudopleuronectes americanus 4 Week i 3 i i Site 2 Site 3 2 Site 1 1 0 20 60 80 100 u 120 > k. > 3 (0 8 Figure 2.5. Mean percent survival (+/- one standard errorof the mean) ofcultured pens at the three sites. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. recaptured fish, there were no differences in survival (p=0.80) between fish at the three sites. Temperature Mean bottom water temperatures were 17.8, 18.4, and 19.7 °C for site 1,2, and 3, respectively, (Figure 2.6), and were not statistically significant (p>0.05). Site 1 bottom water temperature ranged from 11.3 to 25.2 °C, Site 2 from 13.3 to 24.0 °C, and Site 3 from 10.6 to 26.7 °C. Prev Availability A total of 54 sediment samples were examined to determine if fish grazed down the available benthic food supply within the pens. The five primary prey items examined were oribiniidaes, nematodes, amphipods, cumaceans, and bivalves. Regression results show that the food items decreased over time inside the pens (Table 2.1). Within the pens at the Newcastle site, only cumacean abundance decreased during the experiment. At Broad Cove, both cumacean abundance and biomass decreased, nematode and amphipod biomass decreased, and oribiniidae abundance decreased over time. In the Wagon Hill pens, bivalve abundance and biomass decreased. No decreases in abundance or biomass o f these five prey items were observed in the cores collected outside of the pens. Sediment Composition Sediment composition analyses revealed that significant differences existed between sites (Figure 2.7). The Newcastle site, primarily composed of two elements (sand = 54.52% and gravel = 43.47%), contained more gravel (p<0.001) than the other two sites. Both Broad Cove and Wagon Hill were sandier (p=0.002) than the Newcastle site and consisted of 88.60% and 90.42% sand, respectively. All sites contained silt ( £ 3.3%) and clay ( £ 1.6%) to lesser degrees. 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Site 1 30 o 25 20 ¥3 2 15 & E 10 5 0 July 7-14 July 15-22 July 23-31 Aug 1-7 Aug 8-15 Aug 16-25 1999 Site 2 30 25 20 15 10 5 0 July 7-14 July 15-22 July 23-31 Aug 1-7 Aug 8-15 Aug 16-25 1999 Site 3 30 O 25 1- ¥20 | 15 s f■* | 10 . .* • •• —* ' ' JL •* - £ 5 •• •••• i'-w' - ,i V**-. : i L ' ■ : K---'*------H:------i r~--~ July 7-14 July 15-22 July 23-31 Aug 1-7 Aug 8-15 Aug 16-25 1999 Figure 2.6. Bottom water temperature at the three sites during 7 July - 25 August 1999. White boxes illustrate the mean (+/-1 sd) temperature and the black bars show the range. 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced Table 2.1. Biomass and numerical abundance of the five primary prey types available to Pseudopleuronectes americanus inside and outside of the pens at each of the three sites. Bolded p values denote significant decreases in biomass or abundance of prey items. CD E 0 o <75 <75 m CM 55 ph 4) 0 0 3? ‘ o '35 o '35 2 ■g 5 o 2 *35 o * c Vf c 0 0 3 © 3 0 3 0 c 0 "35 fS 3 "35 "35 "35 © O Q. © a > 0 0 0 o a PN © 0 o a 0 a 0 o a 0 a © 0 a d 3 t/» 3 > a 3 n i > tn 0 3 0 o > 0 > a 0 Q 0 > 0 o 3 o ’ o ' o o o o o o a> d s r s r pH o v o o o o 00 co o r s f CM CM o p h p CM n i s CM o CM o § d s r h p § \ c o o o * o d o 00 E ‘ d o * 0) ph ffv o CM * a ph 00 d s r o o £ 2 0 0 o o o o o fO PH n i o PH o o o t r CM s o d O CM s o p^ m o o o CM (/)cn S ite 3 o' o Figure 2.7. Sediment composition of benthic cores taken at each of the three sites. Mean percent particle type was calculated for gravel, sand, silt, and clay. Discussion Growth Individual juvenile winter flounder show variable growth rates (Pearcy 1962; Witting 1995; Sogard 1992; Sogard et al. 2001; Phelan et al. 2000) and this may be due to localized habitat (Sogard 1992; Sogard et al. 2001). In this study, growth was higher at Sites 2 and 3 indication trat local site conditions varied enough to affect growth. Winter flounder growth in a New Jersey estuary was correlated to habitat and site. Local site conditions affected juvenile winter flounder growth more than large-scale climatic forces such as temperature within the same estuary (Sogard et al. 2001). These results reinforce the importance of characterizing potential release sites for optimizing growth in cultured fish. In this study, fish growth was consistent with growth rates previously reported in other juvenile winter flounder field studies (Pearcy 1962; Witting 1995; Sogard et al. 2001) and other caged studies (Sogard 1992; Phelan et al. 2000). This was surprising considering two main differences exist between these previous studies and this one: 1) captured wild flounder were used rather than cultured fish; and 2) experiments were conducted in warmer mid-Atlantic (CT - NC) estuaries where fish growth is typically faster (Klein-MacPhee 1978). The growth rates in this study even surpass a previous caging experiment using cultured winter flounder in New Hampshire (Fairchild 1998). In that study, a mean growth rate of only 0.06 mm/d was observed at an area near Site 1. Little published information exists on estuarine Gulf of Maine populations or on the growth rates of juvenile cultured winter flounder older than 60 days post hatch (dph). Chambers et al. (1988) demonstrated that at a mean temperature of 4.92°C, cultured fish grew 0.145 mm/day from age 70 to 231 dph. In our laboratory, cultured winter flounder 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. growth rates have ranged from 0.13 to as high as 1.26 mm/d for fish between the ages of 41 and 172 dph (unpubl. data). Growth was not density-dependent inside the pens. Sogard (1992) found no effect between density and growth of caged juvenile winter flounder when 1-m2 cages were stocked to replicate natural densities with 3 fish each. Although fish density was higher (n=5) in this study, growth rates from a previous study using 20 fish/cage were similar (Fairchild 1998). Like Sogard’s (1992) study, growth trends for length and weight were synchronous during the 7 weeks with the exception of weight data from Site 1 at week 3. This decrease in weight was most likely due to sampling error rather than actual loss in fish biomass. The slow initial growth of all fish can be attributed to an acclimation period. Upon release into the wild, cultured fish require an acclimation period to recover from the stresses associated with transportation and stocking (Wallin & Van Den Avyle 1995); and to adjust to their new environment (Tsukamoto et al. 1999), hone their survival skills (Olla et al. 1992), and begin to forage efficiently (Nordeide & Salvanes 1991; Yamashita & Yamada 1999). In the case of winter flounder, the fish require a period of 2 days to sharpen their burial skills (Fairchild 1998) and at least 4 days to commence eating (unpubl. data). Survival The observed survival percentages are much lower than was expected considering the fish were in predator exclusion pens. These data do not accurately reflect the ability of the fish to survive at the site but rather the ineffectiveness of the cages at excluding predators. The large variance in the data illustrates that after 3 weeks, no fish were recovered from 2 cages and in 4 other cages, only 1 fish was recaptured. Holes were 49 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. found in the mesh sides of these cages and green crabs were discovered inside. No fish bodies were found which leads us to believe that the fish either escaped or were eaten by crabs. It is important to recognize that in this study, the flounder (theoretically) were separated from their predators. In order to judge the survival of released, cultured fish accurately, predators must be present. To choose a release site wisely, a survey of predator distribution and abundance should be incorporated into the site plan. Temperature Winter flounder are eurythermal and can tolerate a wide range of temperatures (Pearcy 1962; Casterlin & Reynolds 1982). Although several studies have shown that growth is temperature-dependent (Frame 1973; Laurence 1975), this relationship does not explain the growth differences between fish at Site 1 and at the upper estuarine sites. No temperature differences were seen between these three sites during the course of this experiment suggesting that other parameters affected fish growth in this study. Prev Availability Food was not a limiting factor at any of the 3 sites in this study. At Sites 1 and 3 there were hardly any changes in biomass and numerical abundance of prey items within the pens, yet all fish grew over the 7 weeks. This suggests that as the fish were consuming orbiniidae, nematodes, amphipods, cumaceans, and bivalves, recruitment to the benthos was occurring, resulting in a steady-state. At Site 2 where prey biomass declines were observed, the fish consumed the prey faster than it could recruit. However, based on good fish growth at this site, food was definitely not a limiting factor. In fact, amphipod surveys taken simultaneously in the same areas show that, on average, amphipods were more abundant at Site 2 than at either of the other sites (see Chapter 3). 50 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It could be that the Site 2 fish were in a superior physiological state and were better able to feed than their counterparts in the other sites. Two main factors may have influenced the prey analysis results. First, the core samples may not have been representative of actual prey abundances, due to the patchy distribution of benthic organisms (Ivlev 1961). Second, the pens themselves may have altered the initial prey densities as Sogard (1992) found in her caged winter flounder experiment. Because disturbances can cause rapid temporal changes to invertebrate communities in soft substratum (Choat 1982), the pens were placed in the field 1 month prior to the beginning of the experiment to allow time for the infauna to adjust. However, since the magnitude of the change to the benthos was unknown (as a result of both pen placement and core sampling), the effects on the fish’s diet can only be surmised. In addition, the pens may have altered the light and sediment, which in turn, may have modified the settlement of potential flounder prey (Choat 1982). Two schools of thought exist on the importance of food concentration and flatfish distribution. The first and older school of thought supports a high correlation between prey availability and juvenile fish distribution. Field studies have shown that plaice (Pleuronectes platessa) settle in food-rich areas (Zijlstra et al. 1982) and laboratory studies have shown that food supply affects the distribution of juvenile plaice on a small scale (Wennhage & Gibson 1998). The second, and more widely accepted, school of thought believes that substrate composition (which is indirectly related to prey availability) is the primary factor affecting flounder distribution. Through extensive field surveys, this has been shown for many flatfish including juvenile flathead sole (Hippoglossoides elassodon) and rock sole ( Pleuronectes bilineatus) in Alaska (Abookire & Norcross 1998), plaice in Sweden (Pihl et al. 2000), sole ( Solea solea) in Wales 51 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Rogers 1992) and, yellowtail flounder (Limanda ferruginea) in Newfoundland (Walsh 1992). Most recently, research by Wanat (unpubl. data) indicates that amphipod and juvenile winter flounder abundance are not correlated in Great Bay Estuary, NH and, therefore, amphipod abundance may not be a critical element to mapping suitable winter flounder habitat. In other cage studies, similar results were shown. Kamermans et al. (199S) showed that prey density was not critical to juvenile southern flounder growth in a similar cage experiment in a North Carolina estuary. Due to these results, they hypothesized that food abundance does not predict flounder distribution. Sogard (1992) found that winter flounder growth differences were not correlated to prey densities inside a caged experiment. On the contrary, she suggested a relationship existed between flounder growth and temperature and substrate. This study supports the hypothesis that flounder distribution may not necessarily be correlated to habitats with high prey availability. Winter flounder growth was highest at the upper estuarine Sites 2 and 3 despite low amphipod densities at Site 3. For an enhancement effort, however, it would be unwise to ignore the importance of prey abundance in choosing an optimal release site. Other stock enhancement programs have attributed low post-release survival to low prey availability. Since food availability is the primary factor influencing growth rate in juvenile Japanese flounder (Fujii & Noguchi 1996), the presence of suitable prey in the days immediately after a release may be critical to cultured fish’s survival, especially if their feeding ability is inferior to that of the wild flounder (Koshiishi et al. 1991). 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sediment Composition Obvious differences in sediment composition were seen between Site 1 and the other 2 sites. Site 1 was characterized as “gravely” whereas Sites 2 and 3 were characterized as “sandy.” This is an important difference. It is well known that juvenile flatfish prefer smaller sediment grain sizes (Tanda 1990; Gibson and Robb 1992; Moles and Norcross 1995) due to smaller body size, and this trend holds true for juvenile winter flounder too. Juvenile winter flounder prefer sandy rather than coarser or finer substrates. When given a choice, YOY always select fine-grained sediments (see Chapter 4; Phelan et al. 2001). Manderson et al. (2000) showed in laboratory studies that winter flounder <50 mm were not capable of complete burial in gravel. Since the mean size of the fish at the beginning of the study was only 36 mm, it is likely that the flounder could not bury at Site 1. Increased energy expenditure from unsuccessful burial attempts and position maintenance above the substrate during periods of high currents may have contributed to their lower growth rates. In this and Sogard’s (1992) winter flounder caging studies, higher growth rates were observed in sandier areas. Sediment composition is also a critical component for flounder crypsis. By burying into sandy substrates, flatfish become camouflaged to both prey and predators. However, burial is not the most effective survival skill in some predator-prey relationships when YOY winter flounder are prey (Manderson et al. 1999; Fairchild and Howell 2000; Manderson et al. 2000). In encounters with summer flounder (Paralichthys dentatus) as predators, sediment size did not affect survival (Manderson et al. 2000) though it was postulated that this effect may be transitory and that refuge by burial would augment survival as the winter flounder grew larger. 53 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Summary Based on the results of this study, juvenile winter flounder growth is linked more closely to substrate than prey availability. Due to the poor growth performance of the fish and the gravely substrate, Site 1 can be eliminated as a potential release site. No extreme differences in fish growth, water temperature, prey availability, or substrate composition could be discerned between Sites 2 and 3 from this experiment. Based on these findings, these sites are equally good for a pilot scale release o f winter flounder. Small-scale releases should be used to test this theory and fine tune release strategies. 54 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER in DETERMINING AN OPTIMAL RELEASE SEASON FOR JUVENILE Pseudopleuronectes americanus IN THE GREAT BAY ESTUARY, NEW HAMPSHIRE Introduction Once release size and site have been decided, the next logical step in developing optimal release strategies for an enhancement program is to determine release season (Blankenship & Leber 199S). Release season has proven to be a critical variable in the success of many stock enhancement programs. In order to select an appropriate time for releases, it is imperative that basic ecological interactions in the release area are understood. The optimal time for a release is the period when prey are most abundant and when predators are less prevalent (Koshiishi et al. 1991; Folkvord et al. 1994). In Japanese flounder ( Paralichthys olivaceus) stocking programs, recapture rates tripled when the release season was altered from June to May to correspond to elevated prey (mysid) levels and reduced predator (bartail da.the&d)(Piatycephalus indicus) and older Japanese flounder) levels (Yamashita & Yamada 1999). Release season also was found to be critical to white seabass (Atractoscion nobilis) post-release survival, in that 79% of all recaptured fish originated from spring releases (Drawbridge et al. 1995). Cultured lobster (Homarus gammarus) mortality was higher when releases occurred in summer rather than winter, presumably due to the increased presence of predators (van der Meeren 55 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2000). Similarly, Kristiansen et al. (2000) found that predator density, species, and size fluctuated by season which, in turn, affected the post-release survival of cod (Gadus morhua). Several enhancement efforts have achieved greater success when releases were timed to match wild populations of the same taxa temporally, spatially, and proportionately. In Hawaii, for example, striped mullet (Mugil cephalus) returns showed significant differences based on release season (Leber et al. 1997). Recapture rates were highest when cultured fish were stocked in spring and at a size (<60 mm) that coincided with the peak recruitment of similarly sized wild mullet populations. Similarly in Florida, higher returns of cultured red drum (Sciaenops ocellatus) occurred when fish were released in season (i.e. temporally) with wild red drum populations rather than out-of season (Willis et al. 1995). Less than 0.01% of red drum, raised in the hatchery out-of season and subsequently released, were recaptured. It is important to note that the optimal release season for a given species can be influenced by other release strategy components. Optimal release season can be altered by varying size-at-release. For example, in South Carolina, high recovery rates of cultured red drum occurred when larger fish were stocked in spring and fall (Smith et al. 1997). High Japanese flounder returns have been attributed to, in part, a high abundance of prey in the release site coupled with larger sized released fish (Masuda & Tsukamoto 1998). Similarly, for Pacific threadfin, optimal release season and size are interrelated (Leber et al. 1998). In one Hawaiian site, larger fish from winter releases and smaller fish from summer and fall releases led to higher recapture rates. 56 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Seasonal studies of prey, predator, and wild fish densities are critical to understanding the release site dynamics. Prior to any stock enhancement effort, the release site should be surveyed to determine temporal patterns of these species (Leber et al. 1998; Tsukamoto et al. 1999). For the purposes of understanding the ecological interactions in three sites in Great Bay Estuary, New Hampshire, and determining the optimal release season for juvenile winter flounder (Pseudopleuronectes americanus), I conducted an extensive trawling and benthic coring field study during 1999-2001. The abundance of amphipods and mysid shrimp (prey), sand shrimp (Crangon septemspinosa) and green crabs (Carcinus maenas)(predators), and wild winter flounder were tracked. Materials and Methods Three possible release locations (Figure 2.1) in the Great Bay Estuary System, NH were surveyed from 28 June - 6 October 1999 and I June - 5 October 2000. Sampling at each site occurred biweekly with the high tide. Using a 1-m beam trawl (2 m long net, 6 mm body, and 3 mm liner), three replicate tows were taken from one end of the cove to the other (~ 100 m). All catch from the beam trawls was identified, counted, and measured to the nearest mm. To survey amphipod abundance, six replicate benthic cores were taken randomly throughout the site with a 7-cm diameter corer to a depth of 5 cm (192 cm3). Only amphipods were identified and counted from these benthic cores. From 4 June - 2 October 2001, only Site 2 was surveyed. Sampling was conducted weekly and the fishing effort increased from 3 tows with the beam trawl to 6 tows. In addition, each week 6 tows were made with an otter trawl (4.9 m wide net with 2.5 cm body and 6 mm liner in the codend) for gear comparison. All fish, large 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. crustaceans, and sand shrimp from otter trawl catches were identified, counted, and measured to the nearest mm. Six benthic cores were taken weekly to measure amphipod abundance. Trawling and coring depths at all 3 sites ranged from 1 to 4 m. Every year at each site sampled, bottom water temperature was recorded every 36 min by a HOBO1 anchored with a cinder block. CPUE and size frequency distributions were calculated for all juvenile winter flounder caught, and for their predators, green crabs and sand shrimp. In addition, the numerical abundances of their primary prey, amphipods and mysid shrimp, were determined. ANOVA was used to compare CPUE data between sites and years, and between beam and otter trawl catches (Zar 1996). Results All Sites: CPUE in 1999-2001 Using Corer and Beam Trawl Initially, all three sites were considered as potential release sites. After the completion of release site studies in 1999 (see Chapter 2), Site 1 was eliminated. However, trawling continued there during 2000 to determine if the 1999 data were valid. Because there was consistency between the 1999 and 2000 data sets (Figure 3.1), sampling in Site 1 in 2001 was not continued. In addition, it was decided that Site 2 potentially was a better release area than Site 3, and so sampling was discontinued in Site 3 as well. This decision was based primarily on amphipod catches. Table 3.1 lists all species caught at the three sites. Figure 3.1 and Table 3.2 describe the mean CPUE of the primary prey (amphipods and mysids), predators (shrimp and green crabs) and winter flounder from 1999-2001 at all sampled sites. The most 1 Onset Computer Corporation, P.O. Box 34S0, Pocasset, MA 02559-3450. 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. M-02 \ ___ * * _j -/ - - '■ . .V1 ... 'f tL _ ■ ■ < * ♦ Myald Month-Yaar M u c . '. Month-Yaar Groan Crab Groan SltO'3 sits2 M-00 M-00 0-00 A-01 N-01 M-02 M-00 M-00 0-00 A-01 A SHa1 m - O 1 10000 ^ ^ g 8000 § . 6000 ui ? 4000 J-02 M-02 , Tzrer *'*? ' "J "J ' *'*? ’y . . ' ' ' ,,r 4 O F ® • f t Month-Yaar Month-Yaar Month-Yaar Amphipod SandShrimp M-00 D-00 Wintor Floundor Wintor ■-■:W..>A3SBg M-00 M-00 0-00 AOI ‘■I.-'. ■•■■■ ■•■■■ o v * * 'V -!!; ’ : p S F-99 A-99 M-00 0-00 A-01 N-01 M-02 100 120 100 250 | I ’2 ui 8 2 g 16 o S g 200 5 iso IU 2 ° 50 q — — §. 80 S 60 Figure 3.1. Mean CPUE ofamphipods, mysids, sand shrimp, green crabs, and winter floundercaught by beam trawl in all three sites during 1999-2001. Ui VO Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.1. Species caught by beam (B) and otter (O) trawls in the three sites in Great Bay Estuary, NH in 1999-2001. Species Site 1 Site 2 Site 3 jellyfish B green sea urchin ( Stronglyocentrosus droebachiensis) 0 common periwinkle Littorina( littorea) B B, O B Eastern mud snail (llyanassa obsoleta) B B, 0 B Atlantic dogwhelk (Nucella lapillus) B mysid shrimp (family Mysidacea) B B B sand shrimp ( Crangon septemspinosa) B B, 0 B jonah crab {Cancer borealis) B, 0 B green crab (Carcinus maenas) B B, 0 B hermit crab ( Pagurus sp.) B, 0 B horseshoe crab (Limulus polyphemus) B, 0 B lobster (Homarus americanus) B, 0 pipefish (Syngnathus fucus) B, 0 B Atlantic silverside ( Menidia menidia) B rainbow smelt (Osmerus mordax) B B, 0 B herring {Clupea harengus) B B rock gunnel (Pholis gunnellus) B cunner (Tautogolabrus adspersus) B B B 9-spine stickleback ( Pungitius pungitius) B B lumpfish (Cyclopterus lumpus) B.O butterfish (Poronotus triacanthus) 0 tomcod Microgadus( tomcod) B, 0 B winter flounder (Pseudopleuronectes americanus) B B, 0 B windowpane (Lophopsetta maculata) 0 sea raven ( Hemitripterus americanus) B shorthorn sculpin Myoxocephalus( scorpius) B B, 0 little skate (Raja erinacea) 0 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2. Mean CPUE (+/-1 sd) from beam trawls and benthic cores for 1999-2001. CPUE for amphipods is no./core. CPUE for all other species is no./100m2. Bolded values are significantly different between sites within years (p<0.001). Year 1999 2000 2001 Site 1 2 3 1 2 3 1 amphipods 0.3 (0.8) 19.5 (15.3) 1.4 (2.4) 1.1 (3.8) 30.8(31.7) 3.4 (7.1) 17.9(15.0) mysids 220 (715) 204 (474) 28 (83) 573 (669) 762 (2740) 513(656) 68 (229) sand shrimp 22 (14) 53 (100) 48(62) 29(29) 39(54) 31(47) 10(15) green crab 2.6 (2.7) 2.1 (3.9) 2.7 (4.6) 2.1 (2.2) 1.5 (1.1) 2.1 (1.7) 0.7 (1.0) winter flounder 0.1 (0.4) 0.2 (0.4) 0.25 (0.5) 0.6 (0.8) 0.7 (1.1) 1.4 (4.9) 0.1 (0.3) apparent trend from this series of data is the high abundance of amphipods at Site 2. In both 1999 and 2000, amphipods catches were higher at Site 2 than either Sites 1 (p<0.001) or 3 (p<0.001). This was one of the primary reasons for the selection of Site 2 for a potential release. Mysids were present at all 3 sites in 1999. Despite a high spike in Site 2, there was no difference in mysid catch between sites in 2000 (p=0.835). Both predatory shrimp and green crabs were found in all sites, and no differences were found in the amount of either predator at any site (p>0.05) or year (p>0.05). Winter flounder catches were very low at all 3 sites, especially in 1999. Average catches were only 0.1, 0.2, and 0.2S fish per tow for Sites 1,2, and 3, respectively. Catches were slightly higher in 2000 with 0.6,0.7, and 1.4 fish caught for Sites 1,2, and 3, respectively. There were no CPUE differences between sites in either year (p>0.05). Figure 3.2 illustrates the temperature at the three sites. Differences were seen between sites within years (p<0.001) and years within sites (p<0.001). Overall mean temperature (all years) was 16.6,17.7, and 19.0 °C at Sites 1,2, and 3, respectively. Temperature ranged from 8.2 to 2S.2 °C at Site 1,10.9 to 24.0 °C at Site 2, and 10.6 to 26.7 °C at Site 3. Site 2; CPUE and Length-Frequencv in 1999-2001 Using Corer and Beam Trawl Prev. Amphipods were consistently present at Site 2, and averaged 24 (+/- 24) individuals/core (Figure 3.3). Large spikes were observed in July/August 2000. Mysid catches were highly variable in Site 2 over 3 years (Figure 3.4). The mean number of mysids caught per 100 m2 was 286 (+/- 1502), however, it can be seen that there was a considerable amount of seasonal variation. Generally, mysids were present in catches during the first two months and absent after the first week in August. 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Site 1 _30 « 25 m I | 20 1 1 « 15 S. 10 # T - I 5 V 0 -- i—1— ; i— ------1------—I------— ------1------1——H------—1------— Jun- Jul- Aug- Sep- Jun- Jul- Aug- Sep- Jun- Jul- Aug- Sep- 99 99 99 99 00 00 00 00 01 01 01 01 1999 2000 2001 Site 2 P 15 Jun- Jul- Aug- Sep- Jun- Jul- Aug- Sep- Jun- Jul- Aug- Sep- 99 99 99 99 00 00 00 00 01 01 01 01 1999 2000 2001 Site 3 Jun- Jul- Aug- Sep- Jun- Jul- Aug- Sep- Jun- Jul- Aug- Sep- 99 99 99 99 00 00 00 00 01 01 01 01 iggg 2000 2001 Figure 3.2. Bottom water temperature at the three sites during 1999-2001 sampling seasons. White boxes illustrate the mean (+/- 1 sd) temperature and the black bars show the range. 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. % (ajoow) indO Figure Figure 3.3, Mean (+/-1 sd) CPUE of amphipods caught with a beam 1-m trawl in Site 2 during 1999-2001. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced 2000 (zui 001./#) 3fldO (zui 65 X X CM © CM a> 0> o> Figure 3.4. Mean (+/- 1 sd) CPUE of mysids caught with a 1-m beam trawl in Site 2 during 1999-2001. Predators. Sand shrimp catches fluctuated throughout the 3 years. Mean shrimp catch was 25/100 m2 (+/- 53), however, there were noticeable increases in the catch in early July of 1999 and 2000, and late summer 2000 (Figure 3.5). These data suggest a possible bimodal pattern with peaks late June/early July and again late August/early September. Shrimp ranged in size from 0.6 to 5.7 cm. Although representatives from each size category were caught in Site 2 during each month from June to September, the majority of shrimp ranged from 1.0 to 4.9 cm with monthly mean sizes of 2.6,2.3,2.4, and 2.3 cm for June, July, August, and September, respectively (Figure 3.6). Green crab catch and size frequency were relatively constant across 3 years in Site 2 (Figures 3.7 and 3.8). Average catch for 1999-2001 was 1.1 (+/-1.9) crabs/lOOm2. Green crab sizes ranged from 0.3 to 8.3 cm. Mean size of crabs was 4.6,3.1,4.6, and 3.3 cm for June, July, August, and September, respectively. No distinct monthly crab patterns were observed in Site 2 for either CPUE or size. Wild Flounder. Mean CPUE of wild winter flounder was low in Site 2 (0.1 +/- 0.3 fish/100 m2) (Figure 3.9). Flounder ranged in size from 1.1 to 28.0 cm with an overall mean length of 7.1 (+/-6.6) cm. Flounder < 6.0 cm were caught most frequently by beam trawls in June and July than later in the summer (Figure 3.10). Mean size of winter flounder was 3.5, 5.1,13.7, and 10.7 cm for June, July, August, and September, respectively. * Site 2; Gear Comparison of Beam and Otter Trawls in 2001 All species caught with otter trawls are included in Table 3.1. More shrimp (p=0.034), green crab (p<0.001), and winter flounder (p<0.001) were caught with otter trawls than beam trawls in Site 2 in 2001 (Figures 3.11-3.13). On average, otter trawls 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced 400 ( j u z ooi#) o d n a ) i/# o o 67 Figure 3.5. Mean (+/- 1 sd) CPUE of sand shrimp caught with a 1-m beam trawl in Site 2 during 1999-2001. June 200 150 mean = 2.6 (+/-1.3) cm 100 range (0.5-5.7 cm) ik ” ■*» ES9 _ , - _ <0.9 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 Length (cm) July 200 0 • , 150 • J"K- . • mean = 2.3 (+/-0.8) cm V* . 100 ■ 1 1 ■ range (0.6-5.2 cm) 1 50 ■ I 1 A 0 H 1 <0.9 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 Length (cm) Aug 200 150 mean = 2.4 (+/-0.8) cm 100 range (0.7-4.8 cm) 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 Length (cm) Sept 200 mean = 2.3 (+/-0.9) cm 150 range (0.3-4.6 cm) 100 ■k <0.9 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 Length (cm) Figure 3.6. Monthly Iength-frequency histograms of sand shrimp caught by beam trawl in Site 2 during 1999-2001. 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. feui 001./#) 3HdO Figure Figure 3.7. Mean (+/-1 sd) CPUE of green crabs caught with a beam 1-m trawl in Site 2 during 1999-2001. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. June 20 mean = 4.6 (+/-2.4) cm 15 10 range (0.6-7.3 cm) 5 0 <0.9 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 6.0-6.9 7.0-7.9 Length (cm) July "N mean = 3.1 (+/-2.5) cm . * * t e % •• • range (0.5-7.5 cm) T i - . £ r ^ i , a <0.9 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 6.0-6.9 7.0-7.9 Length (cm) Aug 20 mean = 4.6 (+/-2.5) cm £ 15 m m m •-••••■ . | 10 range (0.4-8.3 cm) “■ 5 .. . a . d , A a <0.9 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 6.0-6.9 7.0-7.9 Length (cm) Sept 20 15 mean = 3.3 (+/-2.4) cm 10 range (0.3-6.6 cm) 5 0 <0.9 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 6.0-6.9 7.0-7.9 Length (cm) Figure 3.8. Monthly length-frequency histograms of green crabs caught by beam trawl in Site 2 during 1999-2001. 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2001 2000 1999 a . O Figure 3.9. Mean (+/- I sd) CPUE ofwinter floundercaught with a beam 1-m trawl in Site 2 during 1999-2001. Jon prohibited without permission. June mean = 3.5 (+/-3.6) cm range (1.1-13.8 cm) 0-2.9 3-5.9 6-9.9 9-11.9 12-14.9 >15 Length (cm) July 10 - v .. .. ' .. r " ■■■ ■ .- " - ‘ :~v 5 8 P i l f e r ’&*%>• ■ mean = 5.1 (+/-3.9) cm aS 6 r • • ■ ? 4 range (2.0-13.7 cm) “■ 2 ■ —' . f p H v - ’ u ; , i , . . . A ' 0-2.9 3-5.9 6-8.9 9-11.9 12-14.9 >15 Length (cm) Aug 10 8 mean = 13.7 (+/-10.2) cm ■ :• I 6 -- -•> .> range (5.0-28.0 cm) ■kr ♦ .* *-•*-*'- .* V, 2 0 0-2.9 3-5.9 6-8.9 9-11.9 12-14.9 >15 Length (cm) Sept 10 • • A:’, i 8 .A,.. mean = 10.7 (+/-4.3) cm . v i 6 -- 5 4 . r^r> i'• • r , i* - . ik «***.«•> ••t-.-T* range (6.7-17.8 cm) 2 .jr.-*.—* -.r.-r •• _l . : 0 r] --r:' ■x. 0-2.9 3-5.9 6-8.9 9-11.9 12-14.9 >15 Length (cm) Figure 3.10. Monthly length-frequency histograms of winter flounder caught by beam trawl in Site 2 during 1999-2001. 72 Reproduced with permission of the copyrightowner. Further reproduction prohibited without permission. (ZUJ 00 m ) 3Hd3 Figure Figure 3.11. Mean (+/- sd) CPUE 1 of sand shrimp caught with beam and otter trawls in Site 2 during 2001. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. •'% % % H = -» - *• % KI % •3 % K = % % %3> r ^ ^o. r- ^8> O o (N I s A — % % •r» • •s-~v^w': r -,^s~ - :. .* ■ * . *f - 4&* •4. . - -:-■ - ._ •-: i-.i:; ■ - ■ >% .- : '. ;• • •n^SC'Svr % ->y» -j.'. • .«.;:• ,t.iv X '«4.• ^ ^ -■ '-: . ‘'~z: Mt -v~?f ' — "'■/ -fe \ ' *>- O Q O O o o o o o o o> oo r»* co in ■*■ co oj feu* ooi/#) ando Figure Figure 3.12. Mean (+/- sd) CPUE 1 of green crabs caught with beam and otter trawls in Site during 2 2001. 74 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. . % > % % % % %% % % 5 VL % ° *-r fig* ' * % % ,m m i - % 'eh % \ o CO CO CM o 00 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. captured 14.5 (+/-I4.9) shrimp/100 m2, 18.0 (+/-22.6) crabs/100 m2, and 3.6 (+/-4.7) flounder/100 m2, whereas beam trawls captured 9.5 (+/-14.6) shrimp/100 m2, 0.7 (+/-1.0) crabs/100 m2, and 0.1 (+/-0.3) flounder/100 m2. Overall for 2001, otter trawls captured larger shrimp (p<0.001), green crabs (p<0.001), and winter flounder (p=0.00l) than the beam trawls did. When analyzed by month, otter trawls always captured larger shrimp than beam trawls did (p<0.001)(Figure 3.14). With the exception of June (p=0.213), otter trawls caught larger green crabs than beam trawls (p 0.030) (Figure 3.15). Variation was found between the mean monthly sizes of winter flounder caught by beam and otter trawls (Figure 3.16). The trawls captured different sized fish only in July when larger flounder were caught by otter trawls (p=0.006). However, sample sizes for beam trawl catches were only 1,4,2, and 2 fish for June, July, August, and September, respectively. Discussion Despite the fact that beam trawls are the standard gear used for sampling flatfish populations and habitat (Kuipers et al. 1992), the otter trawl was more effective in Great Bay Estuary, NH. Because large discrepancies were seen between beam and otter trawl catches, all composite beam trawl data need to be re-evaluated. It is evident that all CPUE data from beam trawls are low. Using the otter trawl, catches for shrimp, green crabs, and winter flounder were 1.5,26, and 36 times higher, respectively. While this gear discrepancy changes the magnitude of the catches, it does not affect the seasonal patterns described for these species. Additionally, the overall mean sizes of shrimp, green crabs, and winter flounder caught with the beam trawl were low. All three species caught by otter trawls were 1.4 times larger than those caught with a beam trawl. 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jun# 2001 250 200 mean lengths: beam = 3.2 (+/-1.3) cm 2 ISO ■beam (1.0-5.7 cm) £100 □otter ib otter = 3.8 (+/-0.8) cm (1.2-7.0 cm) < 0.9 1.0- 1.9 2.0 2 .9 3 .0 -3.9 4 .0 4 .9 5 .0- 5.9 Length (cm) July 2001 250 200 mean lengths: ^ 150 beam = 2.5 (+/-1.1) cm ■beam (0.6-4.8 cm) y. "■*' □otter ? 100 otter = 3.5 (+/-0.9) cm ib ?>:f -V* v 1’| *’* " » ' '•’* ' * 50 (1.3-5.7 cm) 0 Iv r n < 0.9 1.0- 1.9 2.0-2.9 3 .0-3.9 4 .0 4 .9 5 .0 5 .9 Length (cm) August 2001 f c V = T P > -. ------V (• . •’ mean lengths: ■beam beam = 2.5 (+/-1.0) cm □otter (0.7-4.7 cm) r - a | •• • m m m w Si*? p . ' ' otter = 3.4 (+/-0.7) cm :| . & j S s j J r ’ ^ i: . (1.4-5.7 cm) < 0 .9 1.0 -1.9 2.0 2 .9 3.0 3 .9 4 .0 4 .9 5.0 5 .9 Length (cm) 8eptemb#r 2001 r-Sir1 ".. - 1 ; v>;. - .' mean lengths: ' ^ 7* ■beam beam = 2.6 (+/-1.0) cm HfevJ-vi: *&.— -. □otter (0.9-4.8 cm) . i ; §1 , .. * H^.. •- .-XfA r SJ otter = 3.2 (+/-0.7) cm ■L. i- ■ 1 ■ ■ ' - J l (1.1-5.0 cm) < 0.9 1.0 1 .9 2.0 2 .9 3.0-3.9 4 .0 4 .9 5.0 5 .9 Length (cm) Figure 3.14. Monthly length-frequency histograms of sand shrimp caught by beam and otter trawls in Site 2 during 2001. 77 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jun* 120 mean lengths: 100 '- ,, t** g 80 beam = 4.9 (+/-2.2) cm ■beam 3 60 (0.8-7.0 cm) ■ ■ □otter F 40 1 otter = 5.4 (+/-1.3) cm “ ■ 20 t (0.9-9.8 cm) 0 i ■ -■ , 1 1 , __ <0.9 1.0- 2.0- 3.0- 4.0- 5.0- 6.0- 7.0- 8.0- 9.0- 1.9 2.9 3.9 4.9 5.9 6.9 7.9 8.9 9.9 Length (cm) July 500 mean lengths: > 4 0 0 beam = 4.2 (+/-2.2) cm | 300 ■beam (0.5-7.5 cm) £200 □otter otter = 5.4 (+/-1.0) cm u. ioo n (0.5-7.8 cm) Length (cm) August mean lengths: beam = 2.1 (+/-2.3) cm ■beam • (0.4-6.0 cm) \ ~ V - v A ; - □otter . . j . s * ; - otter = 5.1 (+/-1.4) cm L U 'J . I h * 1 1 - —i n . (1.1-7.2 cm) <0.9 1.0- 2.0- 3.0- 4.0- 5.0- 6.0- 7.0- 8.0- 9.0- 1.9 2.9 3.9 4.9 5.9 6.9 7.9 8.9 9.9 Length (cm) mean lengths: ■beam beam = 3.0 (+/-2.2) cm r- - •- U’» -' '•! □otter (0.5-6.1 cm) fey r J •. otter = 4.4 (+/-1.6) cm <0.9 1.0- 2.0- 3.0- 4.0- 5.0- 6.0- 7.0- 8 .0- 9.0- (0J-6.7 cm) 1.9 2.9 3.9 4.9 5.9 6.9 7.9 8.9 9.9 Length (cm) Figure 3.15. Monthly length-frequency histograms of green crabs caught by beam and otter trawls in Site 2 during 2001. 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. June 2001 mean lengths: ; - / beam = 13.8 cm v:*~ (13.8 cm) '. • *'•' • □beam ■4 ■otter otter = 18.1 (+/-6.5)cm ' (2.1-32.4 cm) •*t .u, rs--- - ■ -'I L3 H 0-2.9 3-5.9 6-8.9 9-11.9 12-14.9 >15 Length (cm) July 2001 5?'. . * ..• ■ - f- t mean lengths: t ; beam = 9.3 (=/-5.1) cm - 1 □beam ■■■ _ 3 $ L _ (3.74-13.7 cm) ■otter otter = 20.6 (+/-6.5) cm i iKv, ■ • ■ i -M: (4.9-34.9 cm) 0-2.9 3-5.9 6-8.9 9-11.9 12-14.9 >15 Length (cm) August 2001 mean lengths: beam = 21.5 (+/-8.0) cm □beam ■otter (15.9-27.2 cm) ¥ io otter = 20.6 (+/-8.S) cm (4.1-34.3 cm) 0-2.9 3-5.9 6-8.9 9-11.9 12-14.9 >15 Length (cm) September 2001 5 » • ,t 4 - v -y rM . mean lengths: >4 r • 9p' beam = 14.5 (4.7) cm . s’* □beam i 3 (11.2-17.8 cm) ■otter u.i 2 otter = 14.2 (+Z-7.7) cm 1 (5.9-21.0 cm) 0 0-2.9 3-5.9 6-8.9 9-11.9 12-14.9 >15 Length (cm) Figure 3.16. Monthly length-frequency histograms of winter flounder caught by beam and otter trawls in Site 2 during 2001. 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. One of the most important criteria for choosing a release season is the period when prey are most abundant. Amphipods are one of the most important prey items for juvenile winter flounder (Linton 1921; Pearcy 1962) and are actively selected by the fish in the Great Bay Estuary (Fairchild 1998). Understanding the abundance of mysid shrimp, a primary prey item of Japanese flounder, has been a key factor to the success of enhancement programs in Japan (Yamashita & Yamada 1999). Because these shrimp can be important in the diets of winter flounder > 20 mm (Pearcy 1962) and they are present throughout the Great Bay Estuary, seasonal fluctuations of mysid populations should be tracked. Generally, amphipods were uniformly present throughout the summer months in Broad Cove but high spikes in abundance occurred in July-August 2000. Mysids were caught only during June and July, after which they were practically unavailable. Only a few benthic studies have been conducted in the Great Bay Estuary (Armstrong 1995; Grizzle unpubl. data). As a result of temperature and nutrient loading, generally, spring and fall peaks are observed for benthic organisms, and depletion occurs during the summer due to grazing (Grizzle, unpubl. data). However, these patterns may be offset by several months due to a whole host of variables including temperature and previous winter conditions. Therefore, in general, winter flounder releases should occur during these seasonal spring and/or fall peaks. But, when based on the Broad Cove prey data, winter flounder releases should occur in early summer, preferably July, when both amphipods and mysids coexist at the highest densities. In addition to the availability of prey, the release season should be timed when predators are absent or scarce. Sand shrimp and green crabs were present throughout 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. June-September, however, some seasonal patterns were observed. More shrimp were caught in late June/early July and in late August/early Sept. Additionally, shrimp were slightly larger during these periods. This seasonal abundance and size trend is consistent with studies of sand shrimp in various estuaries (Ansell et al. 1999; Modlin 1980). In predator-prey studies using 4.7- 7.4 cm sand shrimp, Witting and Able (1995) found that winter flounder £ 2.0 cm are vulnerable to predation. This predation vulnerability decreased as the flounder grew larger. While it is important to understand all aspects of the ecological interactions in the release site, shrimp abundance and size distribution may not be critical to a winter flounder stocking program if the fish are > 2.0 cm at release (see Chapter I). Green crabs are more of a predatory threat to released winter flounder juveniles than sand shrimp. As discussed in Chapter 1 and in Fairchild & Howell (2000), green crabs > 2.0 cm kill significantly more flounder than smaller crabs do. However, 11-70 mm size crabs are able to kill 11-70 mm size juvenile fish (Chapter 1; Fairchild & Howell 2000). Green crabs were uniformly present during June-September and averaged 4.6,4.3, 6.4, and 4.6 cm for June, July, August, and September, respectively (beam trawl data for July-September were extrapolated up by multiplying 1.4). Green crabs exhibit seasonal (Hunter and Naylor 1993; Ansell et al. 1999), diel, and tidal migrations (Naylor 1958; Lockwood 1967; Crothers 1968; Ropes 1968). Spring and fall abundance peaks are correlated with periods of increased foraging activity (Aagaard et al. 1995). Because these seasonal peaks in crab abundance were not detected during 1999-2001, the data suggest that no single month has fewer, or smaller sized, green crabs. 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In addition to examining the prey and predators in the release site, the abundance and size frequency of the wild counterparts also should be evaluated. While carrying capacity of the release site and season should be studied thoroughly to avoid any excessive competition for resources and to ensure that the site can support the planned release magnitude, it is prudent that the cultured fish are released in synchrony with their wild counterparts to maximize the potential to assimilate into the wild population. Juvenile winter flounder typically inhabit shallow coves and estuaries during the first 2 years of life (Perlmutter 1947; Topp 1968) making short tidal migrations to minimize interactions with predators and maximize foraging potential (Tyler 1971c; Gibson et al. 1998). In a study conducted in the Great Bay Estuary, Armstrong (1997) found that smaller winter flounder were present in the mid-estuarine sites in May-June and not at all present in the upper river estuarine sites. In addition, there was a strong seasonal migration pattern from the middle to the upper estuary. In this study, because catches were so low in Broad Cove (and the other two sites), it is difficult to gauge the distribution of wild winter flounder accurately. However, based on the available data, smaller flounder were caught in June and July than later in the summer, suggesting that these months are most suitable for a release. Summary In determining the optimal release season, a compromise must be reached between available food and potential predation. In Broad Cove, both amphipods and mysids are present only in June and July. Sand shrimp and green crabs, however, are present throughout the summer. Since enhancement program costs increase with the time fish are maintained in the hatchery, it is less expensive to release fish at a smaller size. 82 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. However, the survival rates of cultured fish decline as release size decreases. Therefore, a balance between cost and survival must be reached in an enhancement program. Because winter flounder > 2.0 cm are significantly less vulnerable to crustacean-related predation, released cultured fish should be no smaller than this size (Chapter 1; Fairchild and Howell 1998). Similarly sized wild juvenile fish are found in the release site in June and July. Therefore, based on available prey and equivalently sized wild conspecifics, early summer is the most appropriate release season for a winter flounder stock enhancement effort in Broad Cove, NH. Prior to any large-scale enhancement effort, small-scale releases should be conducted to test this release strategy. 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER IV CONDITION OF CULTURED JUVENILE Pseudopleuronectes americanus FOR RELEASE IN THE GREAT BAY ESTUARY, NEW HAMPSHIRE Introduction Goals of a successful stock enhancement program include the survival of the released cultured fish and their assimilation with their wild conspecifics. However, due to the unnatural hatchery environment, behavioral deficits in cultured fish may exist rendering them more vulnerable to predation. If the abilities of cultured fish to recognize and react appropriately to potential predators are different than that of wild fish, they are more vulnerable than their wild counterparts to predation, and any enhancement effort would be compromised significantly. Many studies have shown that cultured flatfish behave differently than wild flatfish. Cultured flounder such as sole (Solea solea), Japanese flounder (Paralichthys olivaceus), and summer flounder (.Paralichthys dentatus) tend to exhibit more off-bottom behavior than their wild counterparts, which increases their visibility to predators (Howell & Baynes 1993; Tsukamoto et al. 1997; Kellison et al. 2000). Not only did cultured Japanese flounder remain in the water column more during feeding, but they did not return to their initial spots like the wild flounder (Tanaka et al. 1998). In the laboratory, both Japanese flounder and sole were more active at night than wild fish, 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. which could lead to higher mortality in the wild (Howell & Baynes 1993; Miyazaki et al. 1997). Moreover, Kellison et al. (2000) found that predation rates were higher for cultured summer flounder than wild flounder when exposed to blue crabs (Callinectes sapidus). Despite these behavioral discrepancies between cultured and wild fish, there is evidence that cultured fish’s reactions can be improved by predator conditioning or training. Most predator conditioning studies have involved roundfish, and the majority used salmonids (reviewed in Suboski & Templeton 1989; Howell 1994; Olla et al. 1994; Maynard et al. 1995; Olla et al. 1998). In these studies, the cultured fish learned to avoid predators by negative associations with things such as an electrified model that shocked the fish if they approached it too closely (Kanayama 1968; Olla & Davis 1989) or by witnessing some of their conspecifics get eaten by a predator (Leshcheva 1968; Patten 1977; Olla & Davis 1989; Maynard & Flagg 2001). When these “experienced” fish were then compared to a new batch of predator “naive” fish, their survival rates were usually higher. Such has been the case with zebra danios(Brachydanio rerio){Dill 1974), European minnows (Phoxinus pAaxi>ius)(Magurran 1990), coho salmon (Oncorhynchus kisutch)(Patten 1977; Olla & Davis 1989), chinook salmon (Oncorhynchus tshawytscha)(Maynard & Flagg 2001), chum salmon (Onchorynchus £efa)(Kanayama 1968), Atlantic cod (Gadus morhua)(Nadtvedt et al. 1999), and roach larvae (Rutilus nrfi7us)(Leshcheva 1968). From these studies, it appears that cultured fish have the inherent ability to recognize predators but they need some actual experience to sharpen their avoidance behavior. Additionally, several studies have shown that fish leam rapidly. Naive fish groups of chum and coho salmon altered their behaviors to mimic the 85 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. experienced fish groups’ behavior (Kanayama 1968; Patten 1977). Therefore, the predator avoidance reaction may be transferred from one group of conditioned fish to another group of unconditioned fish. For flatfish, crypsis is the primary predator avoidance strategy. The ability to camouflage by burial in substrate and/or color change is paramount. Often, hatcheries promote an unnatural rearing environment by using colorful fiberglass tanks devoid of substrate. Due to lack of experience with substrate, cultured sole, winter (Pseudopleuronectes americanus), and summer flounders bury less, and require more time to bury and change their pigment for crypsis than wild fish do (Howell & Baynes 1993; Fairchild 1998; Kellison et al. 2000). Such conspicuousness could result in higher predation related mortality for the cultured fish. The purpose of this study was to assess the vulnerability of cultured, juvenile winter flounder to piscivorous and avian predation. To investigate if cultured and wild juvenile winter flounder behave differently in the presence of a predator, the reactions of the flounder when exposed to cues from a sea raven (Hemitripterus americanus) were studied. To determine if a relationship exists between flounder coloration and predation vulnerability, light colored and dark colored cultured flounder were made available to visually feeding avian predators. To discern whether predation vulnerability differences exist between cultured and wild juvenile winter flounder, the mortality of dark colored cultured and dark colored wild fish to avian predators was compared. To examine the relationship between fish color and substrate choice, the substrate preferences of light and dark colored cultured flounder were tested. 86 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Materials and Methods Cultured winter flounder were reared at the University of New Hampshire’s Coastal Marine Laboratory (CML) from a wild broodstock brought into the laboratory each spring (per methodology in Fairchild 1998). Wild winter flounder were collected using a beach seine each summer in Salem Harbor, Massachusetts. Wild flounder were brought back to the CML where they were allowed to acclimate for at least one week prior to use in the experiments. These wild fish were maintained in tanks containing a sandy substrate and fed a diet of chopped mussels (Mytilus edulis) daily while in the laboratory. Predation Vulnerability of Cultured Flounder Reaction to Predatory Cues. To examine behavioral responses of juvenile winter flounder when exposed to cues from a potential predator, a 2x3x2 factorial design experiment was conducted at the CML from 8 August to 12 October 2000. Two types of winter flounder (wild and cultured) alternately were exposed to three different cues (visual, olfactory, and a combination of visual and olfactory) from a fish predator (sea raven, Hemitripterus americanus). For each predatory cue, flounder were exposed to either an experimental (sea raven present) or a control (sea raven absent) treatment. The experiment was conducted using two aquaria (Figure 4.1). The smaller 40- liter tank (0.S x 0.25 x 0.3 m) at one end of the system contained an isolated predator. A 2-m long shallow raceway (2 x 0.6 x 0.08 m) was used as the flounder tank. To measure distances flounder traveled towards or away from the predator, the raceway bottom was scribed every 0.1 m. Water from a flow through seawater system was supplied to the predator tank and drained from the opposite end of the raceway. The 82-liter raceway 87 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fa n . t=a R) Figure 4.1. Experimental design of predator cue experiment to test cultured and wild winter flounder reactions to (a) olfactory, (b) visual, and (c) a combination of both olfactory and visual cues from a potential predator. 88 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was slightly angled to create a water gradient away from the predator tank towards the outflow pipe. To test the various predatory cues, the following methods were employed. For olfactory cues, the predator tank was covered with black plastic so that the flounder could not see into the tank (Figure 4. la). Water flowed into the raven tank, overflowed into the flounder tank, and continued down to the drain. The further the flounder moved away from the predator tank and its effluent, the more diluted the olfactory signal from the predator was. For visual cues, the clear predator tank was used in a static system; no water flowed from it into the raceway (Figure 4.1b). For both olfactory and visual cues combined, the predator tank was uncovered and its effluent water flowed into and down the raceway (Figure 4.1c). A control treatment was run for each predator cue. The same methods were used for the control except that the predator tank was empty. Multiple trials were run for each of the treatment levels (flounder type and predator cue) using a new flounder for each replicate. The same sea raven was reused in each trial for a given treatment so that the predator cue was constant. Prior to the start of the trials, the predator tank and raceway were flushed with new seawater to wash out any residual olfactory signal that a new flounder might detect. To start each trial, a single sea raven that had recently been fed flounder was placed into the predator tank. Next, a single flounder was introduced into the raceway at the midpoint between the raven tank and the outflow pipe through a 5-cm width PVC tube. Once the flounder settled to the bottom of the tank, the PVC tube was lifted, and the trial began. Each trial lasted 15 min during which an observer recorded the maximum distance (cm) the winter flounder traveled from the starting point. Movement towards the predator was recorded as a negative 89 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. direction whereas movement away from the predator (towards the drainpipe) was recorded as a positive direction. Distance moved was log transformed and analyzed using Mann-Whitney nonparametric t-tests (Zar 1996). Bird Predation Experiments. A field experiment designed to study avian predation on juvenile winter flounder was conducted during the summers of 1999 and 2001. Two floating, open-topped wooden pens (106 x 74 x 31 cm) containing a 2-cm layer of sediment on the bottom were used (Figure 4.2). These were attached to a floating dock away from any human and boat disturbances in the waters near the CML. A stationary motion detector camera photographed all bird activity. Two separate experiments were conducted using the same system. In the first experiment, the vulnerability of light and dark cultured, juvenile winter flounder to avian predation was compared. Light colored flounder used in the study were fish from the CML reared under normal grow-out conditions in light blue, fiberglass tanks devoid of substrate. Dark colored flounder used in the study also were cultured fish from the CML but these individuals had been acclimated for several weeks in tanks containing sand. In each trial, a total of ten flounder (3 light and S dark) were stocked out into each pen and made available to birds for a 5-hour period. At the end of the trial, all remaining fish were removed and the survivors were identified by color. Survival data were analyzed using Chi-square goodness of fit test. In the second experiment, the vulnerability of dark cultured and wild juvenile winter flounder to avian predation was compared. Cultured flounder used in this experiment were fish reared in the CML that had been acclimated for several weeks in tanks containing sand. Wild flounder used in this experiment were fish collected by beach 90 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.2. Design of the field experiment to study avian predation on juvenile winter flounder. 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. seining from Salem Harbor, MA. To distinguish between the cultured and wild fish, all fish were tagged with Visible Implant Flourescent Elastomer Tags1 injected subcutaneously on the ventral side of the fish. Cultured fish were tagged with a yellow mark whereas wild fish were tagged with a pink mark. In each trial, a total of ten flounder (5 cultured and 5 wild) were stocked out into each pen and made available to birds for a 5-hour period. At the end of the trial, all remaining fish were removed and the survivors were identified by tags. Survival data were analyzed using Chi-square goodness of fit test. Substrate Preference Two experiments were conducted to test substrate preferences of cultured, juvenile winter flounder. The first experiment was conducted from 20 June to 18 July 2000 with light colored fish from the CML reared under normal grow-out conditions in light blue, fiberglass tanks devoid of substrate. The second experiment was conducted from 26 June to 12 July 2001 with dark colored fish from the CML reared under normal grow-out conditions in black, fiberglass tanks devoid of substrate. Both experiments employed the same methods. Eight glass tanks, each with a bottom surface area of 248 cm^, were maintained in a flow through water table with natural illumination. Each tank bottom was partitioned into 4 sediment types, each covering approximately 62 cm^ (filled to a depth of 2 cm). The four sediment types tested consisted of 2 colors (natural and black) and 2 grain sizes (small = 250-1000 pm and large = 1000-2000 pm). 1 Northwest Marine Technology, Inc., P.O. Box 427, Shaw Island, WA 98286 USA. 92 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. At the start of each 24-hour trial, every tank was filled with 1-liter of seawater and the water temperature was recorded. A 5-cm PVC tube was placed in the center of the tank where all sediment types converged. A single winter flounder was introduced into the tank via the tube. Once the fish settled to the bottom, the PVC tube was removed, and the trial began. Observations were made 5 times over the course of a 24 hour period: observations were made hourly for the first 4 hours and a final observation was made at 24 hours. During these observation intervals, the sediment occupied by the greatest portion of the fish was recorded. After the final observation, the fish was removed, measured, and weighed. Final water temperature also was recorded. Before a new trial was initiated, the tank was flushed, new water was added, and the tank was rotated 90° to negate any light effects. A new flounder was used for each trial. To determine if the flounder exhibited a substrate preference, a Chi-square test of independence was used at each of the sampling intervals. In the case that there was no observed difference between 2 or more substrate types, these treatments were grouped and the Chi-square test was rerun (Zar 1996). Results Predation Vulnerability Reaction to Predatory Cues. A total of 381 trials were conducted (Table 4.1). The number of trials ranged from 19 to SO for the treatments tested with fewer control trials run than experimental trials. Within the same predatory cue (visual, olfactory, or combination), wild fish always moved further than the cultured fish (p — 0.02S), 93 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Visual Olfactory& 5.67 (35.85) 29 •0.85 (29.74) 24 Distance (cm) No. Trials 33.03(83.18) 25 38.67 (85.41) 26 Olfactory 9.03 (37.72) 50 4.56 (44.56) 23 10.00 (75.74) 50 45.67 (59.59) 21 ______Visual 9.44 (24.00) 19 17.38(44.21) 45 Distance (cm) No. Trials Distance (cm) No. Trials 29.33(72.10) 34 44.97 (82.28) 35 Summary of predatory cue experiment data. Maximum distance moved in cm (+/• one standard deviation) for each flounder type and treament is reported. Table 4.1. Wild Cultured Wild Wild control Cultured control £ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. regardless of the trial type (experimental or control)(Figure 4.3). Within the same predatory cue and fish type (cultured or wild), fish did not react differently to trial type (p>0.05). Because it was pertinent to know if the flounder moved towards or away from the predator cue, both negative and positive values were included in the means. Averaging positive and negative values resulted in large standard deviations. In all treatments, all fish types moved away from the predator cue except for the cultured control fish in the combination cue treatment (Figure 4.3). Bird Predation Experiments. Ten trials were run from 2 to 6 September 1999 in which more light-colored than dark-colored cultured flounder were eaten (Chi=2.79, p=0.018). One hundred percent of the light-colored flounder were eaten by birds whereas only 58% of the dark-colored flounder were consumed (Figure 4.4a). An overall total of 7.9 (+/- 1.4) fish were eaten/trial by birds. From 11 October to 12 November 1999,18 trials were run in which more cultured than wild dark-colored flounder were consumed by birds (Chi= 15.02, p<0.001). Seventy- nine percent of the cultured fish were eaten whereas only 21% of the wild fish were consumed by birds (Figure 4.4b). An overall total of 5.0 (+/- 2.6) fish were eaten/trial by birds. The main avian predators, documented photographically, were herring ( Larus argentatus) and great black-backed ( Larus marinus) gulls. Double-crested cormorants (Phalacrocorax auritus), green ( Butorides striatus), and black-crowned night ( Nycticorax nycticorax) herons also were observed near the pens. 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Predator Cue .20 Figure 4.3. Mean distance moved (+/- one standard deviation) by cultured and wild juvenile winter flounder when exposed to cues from a sea raven. sea a from cues to exposed when flounder winter wild juvenile and cultured by deviation) standard one (+/- moved distance Mean 4.3. Figure 0 20 40 Distance (cm) Distance 60 80 *,*• 100 <■ •. f. <-■■ •,. ■W ild C ild ■W ■W ild ild ■W C ultured C □ ultured C □ 120 $ w 140 •‘i • i t J . f f i* 1 i k 1 i \ t : . i P ■ :.'f■ • • <*5 • UM. V . P W * i <'■ v : ' ■ V ■ — '• .■*>'.. [■• irf. ’ [■• i..' . 5''i'.vS*^'>', £ -V »*■•) : 1 ., ■ <■««« ■ Dark Cultured Dark Wild Dark -\~• v . . ; > > < r. • ’ *• V Light Cultured Light TJ t IHT 100 120 Figure4.4. Predation vulnerability of(a) light and dark colored cultured flounderand (b) dark cultured and wild winter flounder. c (0 0> O 2 o IB Q. UJ vo Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. More bird predation trials were attempted from 19 August to 10 September 2001, however, these did not yield any results. Various methods were tried to lure birds to the pens and to calculate flounder mortality inflicted by birds. Prior to the start of a new trial, the pens were baited with chopped herring (Clupea harengns) to attract the birds. After 1- 2 days when the bait had been consumed by birds, flounder were stocked into the pens. The pens were monitored for varying amounts of time. Initially, a 2 hour lapse occurred before the flounder in the pens were evaluated. When it was discovered that all fish were still present after 2 hours, trials were extended 3 more hours. After several trials lasting a total of 5 hours, none of the fish were eaten by birds. Pens were rebaited to lure birds again and several 24-hour trials were conducted. Again, the fish were not preyed on by birds. Substrate Preference Light Colored Fish. A total of 101 substrate preference trials were conducted, however, not all 5 sampling intervals were observed in each trial (Table 4.2). At each of the sampling periods, flounder consistently selected the naturally colored, small grain size substrate (p ^ 0.004)(Table 4.3, Figure 4.5a). Mean fish size (+/-1 standard deviation) used in the trials was 31.93 (6.86) mm and 0.48 (0.31) g. Water temperature ranged from 11.0 to 18.5 °C during the experimental period, however, there was no difference between initial and final temperatures on any given day (ANOVA, p>0.05). Dark Colored Fish. A total of 104 substrate preference trials were conducted, however, again, not all 5 sampling intervals were observed in each trial (Table 4.4). At each of the sampling periods, flounder consistently selected the black colored, small grain size substrate (p<0.001)(Table 4.5, Figure 4.5b). Mean fish size (+/- 1 standard deviation) 98 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99 Totalblack, small black, large Totalblack, 2322 98 98 25 98 37 40 natural, small natural, large Substrate Type 13 45 1 20 18 3 23 14 38 26 101 2 21 15 4 23 11 40 25 24 15 Time (hr) Observation including the number of fish observed on each substrate type at each sampling interval. Summary of substrate preference observations of light colored, cultured winter flounder, Table 4.2. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.3. Results of chi-square analysis at each sampling interval for substrate preference of light colored cultured winter flounder. Observation Time (hr) Substrate Type Chi-value p value 1 natural, small 8.50 0.004 2 natural, small 13.07 < 0.001 3 natural, small 8.51 0.004 4 natural, small 12.53 < 0.001 24 natural, small 22.87 < 0.001 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Light Colored Flounder ■black small ■black large r'tt: □natural small □natural large Dark Colored Flounder ■black small ■black large □natural small □natural large Figure 4.5. Substrate preferences of (a) light colored and (b) dark colored cultured winter flounder. 101 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94 Total 14 89 natural, large 18 16 85 25 9 natural, small Substrate Type 6 11151210 22 19 21 20 15 90 85 black, large 54 36 42 blacK, blacK, small 1 37 2 3 4 41 24 Time (hr) Observation including the number of fish observed on each substrate type at each sampling interval. Summary of substrate preference observations ofcolored, darK cultured winter flounder, Table 4.4. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.5. Results of chi-square analysis at each sampling interval for substrate preference of dark colored cultured winter flounder. Observation Time (hr) Substrate Type Chi-value p value 1 black, small 12.46 < 0.001 2 black, small 13.65 < 0.001 3 black, small 22.18 < 0.001 4 black, small 24.47 < 0.001 24 black, small 52.78 < 0.001 103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. used in the trials was 19.74 (4.41) mm and 0.12 (0.08) g. Water temperature ranged from 12.3 to 15.5 °C during the experimental period, however, there was no difference between initial and final temperatures on any given day (ANOVA, p>0.05). Discussion Predation Vulnerability Reaction to Predatory Cues. When exposed to cues from a sea raven, the wild winter flounder consistently were more reactive (i.e. moved further) than the cultured flounder regardless of the type of predator cue or type of trial (experimental or control). This activity pattern could be a result of the cultured fish’s acclimation to the laboratory setting and consequent reduced activity relative to the easily startled wild fish. It also is possible that the experimental design may have been inadequate to test the hypothesis. While it appears that in the presence of a predator, cultured fish tended to move more than the control and wild fish moved less than the control, this trend was not significant. Neither type of fish showed a statistically significant reaction to the inclusion of the sea raven in the trials. The natural predator avoidance strategy of flatfish is to bury into the substrate and remain motionless. Therefore, distance moved may not be the best measurement for flatfish’s reactions to a predator. Rather, the rate of burial and/or minimal distance moved would be a more appropriate measure. Because the raceway was devoid of substrate, the higher activity level of all wild fish may have been due to their search for a place to bury. The results also may have been affected by the experimental system. There may have been something besides the decreased predatory cue at the far end of the raceway (positive end) that attracted the wild fish such as a difference in lighting, shadows, and/or interference from the laboratory. In hindsight, the aquaria 104 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. should have been reversed for half of the trials such that the predator tank was at the opposite end of the raceway. This may have negated any extraneous variables associated with the design. The idea of using predatory cues to compare cultured and wild fish behavior originated from a salmon conditioning experiment by Olla and Davis (1989). They were able to condition cultured coho salmon through exposure to visual and olfactory cues from a lingcod (Ophiodon elongatus). In addition, when the cultured fish approached the predator tank, a frozen lingcod was dunked into the salmon tank for a negative association to the predator. After 2 conditioning trials, the experienced salmon were less vulnerable to predation than naive salmon were. For flounder, predatory cues alone may not be strong enough stimuli to test behavioral differences. Perhaps predatory cues combined with a negative stimulus, such as an alarm pheromone or a model of the predator, are necessary to generate valid results. Olla and Davis (1988) commented that predator-prey experiments conducted in the laboratory are complex because the predators are unpredictable and can exhibit variable feeding responses. It becomes difficult to determine whether the cultured fish survive because the predator is not foraging normally or the cultured fish have become predator-conditioned. In a predator-conditioning study of summer flounder, Kellison et al. (2000), used blue crabs for predators. Although initial predation rates were higher for cultured fish than wild fish, effective predator conditioning was performed. Experienced cultured summer flounder (those that had been exposed to blue crabs for 24 hours while the crab preyed on spot (Leiostomus xanthurus)) were less vulnerable to crab predation in subsequent encounters than naive cultured flounder were. However, experienced fish 105 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were still more vulnerable to predation than wild flounder. It is possible that the cultured winter flounder exposed to cues from the sea raven became conditioned, and further testing in predation trials could have revealed this. Bird Predation Experiments. In comparison to wild fish, cultured winter flounder, regardless of color, were extremely vulnerable to predation by birds. This is a critical factor for an enhancement program. Maynard and Flagg (2001) recognized that bird predation could have an effect on chinook salmon stockings. Cages containing merganser ducks were suspended in the salmon raceways for predator conditioning. Salmon that either entered the cages or got too close could be eaten. When conditioned and naive salmon were released into test streams, the conditioned fish had a 26% higher survival rate than the naive fish did. The predation vulnerability of dark cultured fish differed by 21% between the two bird predation experiments (Figure 4.4). Although the birds consumed a variable number of fish/day, they tended to consume fish that were easiest to capture. In the first bird predation experiment, obvious color differences existed between the cultured fish. Given a choice, birds selected the most visible and easily captured fish, the light colored cultured flounder. The second prey choice in this experiment was the dark colored cultured fish. Although their color matched the substrate, it is likely that these cultured fish exhibited noncryptic behavior (i.e. increased movement, lack of burying) and as such, were still fairly conspicuous to birds, resulting in an overall total of 7.9 (+/- 1.4) fish eaten/trial. In the second bird predation experiment, 5.0 (+/- 2.6) fish/trial were eaten by birds. Neither fish type was more visible due to color variation, thus, increasing the difficulty to detect and capture the fish. However, it is likely that behavioral differences 106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. between the cultured and wild flounder increased the conspicuousness of the cultured fish to avian predators and, thus, increased their predation vulnerability. Wild flounder, probably were much more difficult for birds to capture so the cultured flounder in this experiment were preyed on more heavily, especially compared to the dark cultured flounder in the first bird predation experiment. Little research has been done on the extent of avian predation on flatfish populations. Although crustaceans and fish have been the focus of most predation studies for flatfish populations, birds probably are responsible for the greatest percentage of mortality for juvenile flatfish. In the Dutch Wadden Sea, studies have shown that great cormorants (Phalacrocorax carbo) prey heavily on flatfish populations (Leopold et al. 1998). Flatfish composed 73% of the species and 79% of the bulk of the cormorants’ diet. Young-of-the-year (0-group) fish accounted for the majority of the flatfish eaten by these birds. In the Great Bay Estuary, there are large populations of diving ducks (mergansers, (Mergus spp.)), double-crested cormorants, and stalking predators like great blue herons (Ardea herodias), green herons, black-crowned night herons, and snowy (Egretta thula) and great egrets (Ardea alba), most of which have been observed hunting and eating juvenile winter flounder (personal observation). While it has been acknowledged that great blue herons and cormorants are predators of young winter flounder (Pearcy 1962; Tyler 1971b; Pilon et al. 1982; Birt et al. 1987; Blackwell et al. 1995), the remainder of these piscivorous birds has not been documented as predators of winter flounder. Therefore, this avian predation pressure either has been radically underestimated or grossly overlooked. Both herring and great black-backed gulls are present in the Great 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bay Estuary and consume juvenile winter flounder as suggested by Howe et al. (1976) and determined by this study, but they probably are not the primary avian predators of flatfish. Because gulls were constantly around the docks of the CML, they were the main predators in this experiment. However, future studies should focus on more appropriate species like herons and cormorants. Despite the clear trend seen in 1999 that cultured fish are more vulnerable than wild fish to avian predation, the pattern was not reproducible in 2001 because of a lack of avian predators. This was a prime example of predator inconsistency (Olla and Davis 1988). To ensure that the predation pressure is a constant variable in this type of experiment, captive birds are needed to realize meaningful results. Future studies should incorporate captive herons and cormorants in an aviary-like enclosure around a tank or pond to achieve better results. Substrate Preference In the field there is a well-known relationship between flatfish distribution and grain size and sediment type (Rogers 1992; Walsh 1992; Gibson 1994; Abookire & Norcross 1998). This relationship has been corroborated further by sediment preference experiments in the laboratory (Tanda 1990; Gibson & Robb 1992; Moles & Norcross 1995). Therefore, it is not surprising that the winter flounder selected small grain sized sediments in both substrate preference experiments. Gibson and Robb (1992) showed that fish < 3.0 cm are incapable of burying effectively in sediment sizes > 1 mm, and Tanda (1990) showed that juvenile marbled sole (Limanda yokohamae) and Japanese flounder preferred grain sizes in which they could bury themselves easiest. Moles and Norcross 108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1995) found that several juvenile flatfish indigenous to Alaskan waters, actively selected sediments with grain sizes <500 pm. In addition to selecting the smaller grain size, winter flounder selected the color sediment that best matched their body color. This is a valuable point for rearing strategies for enhancement efforts. In order for the flounder to be inconspicuous in the release site, there must be a comparably colored, small-grained sediment available to the fish. Alternatively, the flounder could adjust their color to match the color of the release substrate while still in the laboratory by introducing sediment to the rearing tanks. This would give the cultured fish the added benefit of acclimation to sediment and time to hone their burial skills (Gibson & Robb 1992; Howell & Baynes 1993; Fairchild 1998; Kellison et al. 2000). Summary Cultured winter flounder react differently than wild flounder in the presence of predators and are significantly more vulnerable to predation by birds, regardless of fish color. However, prior to any winter flounder enhancement effort, additional studies need to be conducted for further assessment of the cultured fish’s vulnerability to predation. Cultured winter flounder select sediments consisting of small grains and colors matching their own pigment. It is recommended that the fish adjust to the release site substrate first in the laboratory and second in the field in acclimation pens immediately prior to their release. 109 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SYNTHESIS Results from this dissertation suggest some of the techniques that should be considered as any winter flounder stock enhancement program is designed. Planning to make the released fish as fit as possible should begin in the hatchery. For example, winter flounder should be conditioned to substrate from the release area so that they can adjust their coloration to match the substrate and hone their burial skills. Further, because the cultured flounder are more vulnerable to predation than their wild counterparts, pre release predator conditioning is strongly advised. Also, it is recommended that only fish > 2.0 cm should be released to reduce decapod crustacean-related mortality. Additionally, it seems important that fish should be acclimated in predator exclusion pens at the release site so that they can adjust to the abiotic and biotic elements of the release site. Selection of release site and season also is critical. In the Great Bay Estuary in New Hampshire, Broad Cove is a suitable release site for winter flounder because this area has appropriate sediment and prey types. Early summer is the optimal release season because prey, particularly amphipods and mysid shrimp, are abundant. The fact that similarly-sized wild conspecifics are most abundant at this time is an indication that natural selection has favored this season as well. Care should be taken, however, to avoid competition between stocked and wild fish. At present, too little is known about resource requirements and possible limits to make a thorough study of environmental carrying capacity, and this is an area that will need considerably more research before any stock enhancement program is undertaken. 110 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. There are many other details associated with optimal release strategies that need to be addressed before a winter flounder stock enhancement program can commence. These include more predator-conditioning research, the design of better acclimation pens, studies to determine the optimal acclimation period in these pens, and assessments of the optimal tidal stage and time of day for releases. Once an optimal release strategy has been determined, the ultimate way to measure its effectiveness and the performance of the cultured fish is by pilot-scale releases. As part of this study, small-scale releases of cultured, juvenile winter flounder were conducted in the Great Bay Estuary during 1999-2001 (see Appendix A). Unfortunately, not all optimal release strategies were used because the research that identified these strategies was being developed simultaneously. In hindsight, the releases should have been delayed until optimal release strategies were understood better. While the results of this study contribute to the understanding of a winter flounder stock enhancement program, more research is needed to determine if this endeavor is feasible. Besides the continuation of research needed to develop optimal release strategies, other stocking issues should be examined. Basic techniques for rearing winter flounder are established, but there are still some hatchery techniques that need to be refined. For example, little is known about winter flounder broodstock management and producing fish year-round. Additionally, the sex ratio of fish reared in the hatchery is unknown and this information is critical for any enhancement program. If the released fish are from a captive broodstock, a detailed genetics program needs to be established to ensure that the released fish are genotypically similar to the wild population. Health management procedures also will need to be established. I ll Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In this study both elastomer and coded wire tags were effective ways to mark juvenile winter flounder (see Appendix A). Retention rates for both tag types were high indicating that either would suffice for a stocking effort, however, a minimum size limit of 2.1 cm was observed for tagging juvenile winter flounder with coded wire tags. Different tagging techniques and types should be studied further. Extensive ecological studies are warranted to assess the impact of the released fish. Although difficult to measure, the release site carrying capacity should be calculated to determine the magnitude of the releases. This could be accomplished through a series of long-term studies on the growth, movements, and feeding rates of wild and cultured winter flounder populations in the release site combined with prey abundance surveys. Wild flounder should be tagged and monitored in addition to the released cultured flounder to determine if they are being displaced and to calculate the hatchery contribution to the wild population. In addition, the bioenergetics of the release site should be calculated to aide in selecting the release magnitude. More field studies would be helpful in understanding the dynamics of wild winter flounder populations. Of particular interest is the amount and rate of juvenile flounder mortality due to birds. At this point, it is not possible to determine if it is feasible to stock winter flounder. An educated and scientific decision can be made only after the completion of more research. 112 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX A PILOT-SCALE RELEASES OF CULTURED JUVENILE Pseudopleuronectes americanus IN THE GREAT BAY ESTUARY, NEW HAMPSHIRE IN 1999-2001 Introduction Pilot-scale releases are the ultimate way to measure the effectiveness of release methods and the performance of cultured fish. Through these small, experimental stockings, release strategies can be evaluated and altered if needed, without risking all of the cultured fish or damaging the release area. If release methods are found to be inferior, they can still be adapted prior to the larger enhancement effort. Caging experiments are highly effective tools for determining optimal release strategies (see Chapter 2), however, they are incomplete in that predators are excluded and cultured fish are spatially limited. With actual pilot scale releases, all release variables interact and the outcome is more meaningful. Prior to any enhancement effort, and regardless of the release magnitude, it is imperative that release protocols adhere to certain guidelines. First, released fish must be tagged so that they can be identified in subsequent surveys to assess their growth, survival, and movements. Tags must not be a source for mortality either by making them more visible to predators or by injuring them. Second, released fish must be healthy and represent the pathology of the wild population so that they neither spread nor succumb to 113 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. disease. Third, the cultured fish must be genotypically similar to the wild population so that they are representative of the genetic diversity of the wild fish. This is achieved, in part, by proper broodstock management. These guidelines are addressed by Blankenship and Leber (1995). To examine the performance of cultured, juvenile winter flounder {Pseudopleuronectes americanus), small, annual, pilot-scale releases were conducted during 1999-2001. It was intended that growth, survival, movements, and contributions to the wild populations would be measured through a field sampling program. Materials and Methods Ten thousand juvenile winter flounder, to be raised by Great Bay Aquafarms1 (GBA), were scheduled for release each year. Two weeks prior to release, all fish were tagged so their survival, growth, movements, and contribution to stock rebuilding could be evaluated. Tagging was done in two ways. In 1999, fish were differentially color- tagged according to release site and date with Visible Implant Florescent Elastomer Tags2 injected subcutaneously on the ventral side of the fish. In 2000 and 2001, sequential Coded Wire Tags2 were implanted into the muscular tissue. Different implantation sites (posterior dorsal, anterior dorsal, ventral) on the fish were used to identify different release dates. Immediately prior to release, all fish were inspected to ensure that tags were still in place. In addition, at least 2% of the tagged fish from 1999 and 2001 were randomly subsampled and held in the Coastal Marine Laboratory (CML) to quantify tag- retention rates (Leber et al. 1997). 1 Great Bay Aquafarms, Gosling Road, Portsmouth, NH 03801. 2 Northwest Marine Technology, Inc., P.O. Box 427, Shaw Island, WA 98286. 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Three release locations (Figure 2.1) in the Great Bay Estuary System, NH were selected for the pilot-scale releases. Fish were released into all three sites in 1999, however, only Site 2 was used for the 2000 and 2001 releases (Table A. I). Seven days prior to the 1999 release, sand was added to the holding tanks. This allowed the fish to enhance their cryptic abilities by adjusting their coloration to match the substrate, and to practice their burial skills. On release days, fish were transported in aerated coolers by boat to the release sites. The fish were brought in mesh dive bags to the bottom and released by divers. Most flounder immediately buried into the substrate, however many fish swam to the surface for a short period of time before finally returning to the bottom. For both 2000 and 2001 releases, all cultured fish were maintained in black fiberglass tanks in the CML so that they developed the appropriate dark pigmentation. On release days, fish were transported in aerated coolers by boat to Site 2. The flounder were stocked into three acclimation pens (43 x 91 x 15.5 cm) constructed of 4 cm plastic coated wire mesh with a 7 mm liner. These were lowered to the bottom by lines and anchored by two 8 kg weights. Forty-eight hours later, the pens were hauled to the surface, and the fish were released. Divers observed that all fish immediately swam to the bottom and buried in the sediment. In order to monitor the growth, survival, and movements of the released fish, a sampling program was established to recapture the cultured flounder. Using a 1-meter beam trawl and a 4.9 m wide otter trawl, daily sampling was conducted in the release area(s) for the first week afler each release. With the exception of the first release batch in 1999, sampling was conducted for at least three consecutive days. If no cultured fish 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A. 1. Summary of juvenile winter flounder releases in the Great Bay Estuary 1999-2001. Year Batch Date No. Released Site Mean (+/-1 sem) Size Tag Type No. Recaptured 1999 1 20-Sep 855 1,2,3 6.3 (+/-0.1)cm VIE 0 4.0 (+/- 0.2) g 2 27-Oct 1498 1 6.8 (+/-0.1) cm VIE 0 4.4 (+/-0.2) g Total 2353 0 2000 1 30-Aug 557 2 5.4 (+/-0.2) cm CWT 0 2.9 (+/-0.2) g 2 17-Sep 486 2 data not CWT 0 available 3 10-Oct 458 2 data not CWT 0 available Total 1501 0 2001 1 18-Aug 740 2 4.3 (+/-0.3) cm CWT 0 1.3 (+/-0.2) g Total 740 0 Grand Total 4594 0 were recaptured at this point, sampling returned to a 1-day/week schedule. For each sampling day, replicate 100 m2 tows were made in the release area and in the areas immediately surrounding the release site. In addition in 2001, tows were made in the near shore channels to determine if cross-channel dispersion was occurring. Because green crabs (Carcinus maenas) prey on winter flounder (Fairchild & Howell; see Chapter 1) and are ubiquitous in the Great Bay Estuary (see Chapter 3), all green crabs caught in trawls after releases were examined. In 1999, all green crabs captured the day after release were brought back to the CML and dissected to look for flounder otoliths. In 2000 and 2001 all crabs caught after fish releases were scanned with the handheld wand to detect if they had consumed any coded wire tagged fish. Results Releases Cultured fish production by GBA was disappointingly low. In 1999, approximately 900 fish > 30 mm were raised; in 2000, approximately 3000 fish were raised. GBA did not produce any winter flounder in 2001. For all years, cultured flounder production at the CML was very successful. Several thousand fish were reared to the juvenile stage according to the methodology of Fairchild (1998). These fish were used for various experiments (see Chapters 1,2, and 4) and the remainder was used to supplement the GBA flounder for the pilot-scale releases. A total of 2353, 1501, and 740 cultured juvenile winter flounder were released, in 1999,2000, and 2001, respectively. Table A.l describes the release season and location, the magnitude of the releases, and the mean size of the released fish. 117 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A.2 summarizes the sampling program results. None of the cultured winter flounder were recaptured from any of the releases in 1999*2001. Neither flounder otoliths nor coded wire tags were present in any of the green crabs dissected in 1999 or scanned in 2000 and 2001. After the releases, wild winter flounder were never caught in Site 1, but always caught in Sites 2 and 3 (Table A.2). Additionally, after the 2001 release, sampling effort was divided between the actual release site and the proximate areas including other coves, the channel, and coves on the opposite side of the channel. A total of 21 tows in the release site yielded 17 wild flounder, and 12 tows in the surrounding areas yielded 4 wild flounder. The acclimation pens used in the 2000 and 2001 releases were not completely secure. In 2001, a noticeable portion of the flounder had already escaped into the wild after the 2-day acclimation period. Additionally, two of the pens contained a total of 3 green crabs and there were obvious signs of predation, including a partially eaten flounder. Tagging and Mortality Cultured fish were monitored carefully during the period between tagging and release to ensure that the effects of tagging were not detrimental to the fish’s well-being. In 1999,47 tagged fish died between the tagging and release period. However, the tagging process was not fatal; all mortalities showed signs of Vibrio sp. infection. Mortality ceased once the fish were treated with antibiotics. In 2000,16 mortalities occurred that probably were due to the coded wire tagging process and the small fish size. In 2001,150 fish died after tagging. The mean (+/- 1 sd) sizes of these fish were 2.1 (+/- 0.4) cm and 0.2 (+/-0.1) g. 118 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. — - - -- (Range) Mean Size (cm) Cultured Flounder Cultured 0 0 0 0 0 0 0 0 0 0 0 Caught Number _ . _ 6.5 13.0 (Range) available (4.2 - 8.2) (4.9-26.1) (7.1 - 26.8) Mean Size (cm) Wild Flounder Wild 1 1 0 5 5.98 (+/-1.6) 16 0 39 21 Caught Number 3 3 3 13 7 9.2 (+/-7.5) 15 0 33 21 17.9 (+/-5.0) Total No. Tows 1 1 1 3 3 9 4 no data 3 9 Days Sampled No. Consecutive beam trawl beam trawl beam trawl beam trawl S beam trawl beamtrawl 2 2 3 2trawl otter 5 2 beam trawl 1 1 3 2 2 2 1 Total Total Total 2 1999 1 1 Year Batch Site GearType 2001 2000 Grand Total Grand Results theof sampling to program recapture released the in Greatflounder Bay winter Estuary 1999-2001. Table A.2. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Immediately prior to their release, all fish were inspected to verify that they still had tags. All fish in 1999 had elastomer tags prior to their releases. In 2000, some tag loss occurred for coded wire tagged fish between the tagging and release period. From the anterior dorsal position, 2.2% (n=13) of the flounder lost tags; 2.6% (n=12) of the fish lost tags from the posterior dorsal position; and 2.8% (n=13) of the fish lost tags from the ventral location. In addition to the evaluations between tagging and releasing the fish, long-term tag retention and mortality studies were conducted (Table A.3). In 1999,2% (n=45) of the tagged fish were held in the CML to monitor tag retention and any tag induced mortality. Only one tag related mortality occurred and the remainder did not lose their tags over a 6-month period. Although fish were not retained in the CML in 2000 for long term coded wire tag retention studies, 3% (n=22) of the coded wire tagged fish in 2001 were monitored. Two of these flounder shed their tags after 2 months. Discussion None of the cultured fish were recaptured from any of the releases, suggesting that they either emigrated from the release areas, were able to avoid recapture, or died. Since none of the released fish were recaptured and only a few wild flounder were caught, not enough data exist to draw any definite conclusions. Because of this, I can only speculate on the outcome of the cultured flounder. Cultured fish may have dispersed from the release sites but despite extensive trawling in the immediate days following most of the releases and, after the 2001 release, in areas around the release area, none of the cultured fish were recaptured. Thus, it is not likely that the flounder strayed from the release area, however, more sampling effort would have solidified this theory. It is also 120 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table A.3. Summary of winter flounder tag-retention rates and tag-related mortality study. Number Fish Number Fish Number Fish Number Months Tag Type Year Retained in Lab Lost Tags Died Monitored VIE 1999 45 0 1 6 CWT 2000 0 - - - CWT 2001 22 2 0 2 121 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. possible that the cultured fish were able to avoid the fishing gear, yet this theory is less plausible since wild flounder were caught in all areas except Site 1. Therefore, it is most likely that mortality was the primary reason why none of the released fish were recaptured. The reason(s) why the cultured fish perished is(are) mostly unknown, however, a small percentage of loss can be explained due to tagging. Because elastomer tags caused some mortality (Table A.3), 2% (n=47) of the released 1999 fish may have died due to the tags. A size-specific mortality was associated with coded wire tags such that fish ^ 2.1 cm were mortally wounded by the tagging process and /or the tags themselves. However, the loss of the released fish is not explained by this tag-related mortality since these small, tagged fish died in the immediate days after tagging while still in the laboratory. A 9% tag loss was observed for coded wire tagged fish (Table A.3) implying that 135 fish in 2000 and 66 fish in 2001 lost their tags after 2 months at large and, therefore, were not identifiable as cultured fish. It is unlikely that size-related predation of the cultured fish was a main source of mortality since the average sizes of the batches of released flounder ranged from 4.3 to 6.8 cm. These sizes were appropriate for increasing the fish’s survival chances after stocking because juvenile winter flounder £ 2.0 cm are highly vulnerable to predation by sand shrimp (Crangon septemspinosa)(Wi\ting Able & 1993,1995) and green crabs (Fairchild & Howell 2000; see Chapter 1), and because larger cultured fish have higher survival rates after stocking (Yamashita et al. 1994; Leber 1995; Tominaga & Watanabe 1998; Kristiansen et al. 2000). Therefore, size-related predation by decapod crustaceans should have been negligible. 122 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Inferior release sites may have contributed to the presumed mortality of the released cultured fish in 1999. In that year, releases occurred in all three sites. Two releases were scheduled for each site, however, because no fish were recovered after the first release, the release strategy was altered. Rather than split the second release between 3 areas, the remainder of the fish were released in only one site to increase recovery chances. Unfortunately, the flounder were released into the worst of the 3 areas, Site 1 (see Chapter 2). This, in part, explains why there were no recaptures for 1999. In contrast, all releases occurred at Site 2 in 2000 and 2001, yet still there were no recaptures in either year. Release season was not appropriate for any of the 3 release years. All stockings occurred in the late summer and fall with the earliest release occurring on 18 August 2001. Prior to surveys in the release areas, it was assumed that releases of fish should occur in the fall because by then, the cultured fish would have reached the desired size (>2.0 cm total length), and from past experience, similarly sized wild individuals have been abundant in the release locations. In hindsight, this was not the best release season strategy. Because prey are most abundant in early summer in Site 2 (Chapter 3), the flounder should have been released in June and July. Growth data from the CML indicate that cultured winter flounder achieve this size (2 cm) by late-June/early-July (unpubl. data). Additionally, catch data showed that similarly sized wild winter flounder are most abundant in the Great Bay Estuary in early summer (Chapter 3; Armstrong 1997). High post-release mortality is common in cultured fish that are not acclimated to the release site (Parker 1986; Olla et al. 1994; Wahl et al. 1995) or conditioned to avoid potential predators (see Chapter 4). Fish for all releases were conditioned to some aspects 123 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the release sites. All fish were conditioned to substrate, either in the CML or in pens at the release site, so that they could hone their burial skills and adapt their color for cyptic behavior. Additionally, flounder that were released first into acclimation pens, had the added benefit of adjusting to the abiotic conditions (temperature, salinity, dissolved oxygen, tides, currents, light) of the release site over 2 days prior to their release. These fish also may have had an opportunity to become cognizant of potential predators near the pens. Survival rates of the flounder released from the acclimation pens should have been higher than those released directly into the wild due to lowered stress levels. Many studies have shown that fish suffer from stress due to transporting, handling, and stocking (Parker 1986; Olla et al. 1994; Wallin & van den Avyle 1995) and this can reduce their survival in the wild (Jarvi 1990; Olla et al. 1994; Olla et al. 1995). By first stocking into acclimation pens, the fish are given a recovery period before their release into the wild. However, in order for the pens to provide an effective refuge for the cultured fish, they must be secure such that fish do not escape and predators cannot enter. This was not the case in the 2000 and 2001 releases. Therefore, it is difficult to measure the effectiveness of the acclimation pens used. Even if optimal release size, site, season, and conditioning strategies had been used for the 1999-2001 winter flounder releases, it is still likely that none of the cultured fish would have been recaptured because of the small magnitude of the releases and the low sampling effort. There have been many pilot release studies and despite the different species stocked, the varying release strategies, and the different times at large, most enhancement programs have recaptured some of the released, cultured fish (Leber & Lee 1997; McEachron et al. 1998; Kitada 1999; Sv&sand et al. 2000). In the majority of these 124 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. studies, tens of thousands of fish were released. However, despite the large release magnitudes, recapture rates have been highly variable ranging from 1.5 % for red drum (Sciaenops ocellatus)(Wi\\is et al. 1995) to ~ 30 % for cod (Gadus morhua)(Svhsand et al. 2000) and Japanese flounder {Paralichthys olivaceus)(Kitada 1999). This variation is due, in part, to the amount of effort invested in recapturing the released fish. Recapture rates are a function of sampling effort and rates increase as sampling intensity increases. The majority of enhancement programs have stocked species at sizes targeted by commercial and/or recreational fisheries, which substantially augments sampling effort. In addition, because sampling intensity is greater for commercial than recreational fisheries, recapture rates are generally higher for commercially targeted species. In this study, sampling intensity was much lower than in other enhancement programs. Because there is neither a commercial nor a recreational fishery for juvenile winter flounder, the research project provided the only sampling effort. Therefore, the low recapture rate for these pilot-scale releases is a function of the low numbers of fish released combined with the absence of a fishery to assist in recapturing them. The question then remains, if more flounder had been stocked using the known appropriate release strategies, and sampling effort had been increased, would the cultured fish have survived? Unfortunately, this cannot be determined until another release is conducted. An upper limit to the release magnitude of winter flounder may exist if resources are limited in the release site. Care must be taken to ensure that the carrying capacity of the release site is not exceeded. For example, if food resources are fully exploited by the wild fish, then it is likely that the cultured fish either will starve, compete with the wild fish and possibly displace them, or emigrate from the release site. Though carrying 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. capacity is difficult to measure, long-term studies evaluating the feeding, growth rates, and dispersion of wild fish, combined with prey abundance in the release site, may aid in solving this question (Yamashita & Yamada 1999; Svasand et al. 2000) Summary Although none of the released fish were recaptured, future winter flounder stock enhancement research should continue. Releases failed, in part, because optimal release methods were not used. Future stockings should incorporate the optimal release size (Chapter 1), site (Chapter 2), and especially season (Chapter 3) in a testable, statistically designed manner (Leber 1999). Additionally, better acclimation systems need to be designed and implemented at the release site. Further, the release magnitude and sampling effort need to be increased substantially to be able to detect the effects of stocking cultured fish. Lastly, more ecological research should be included in the enhancement program to assess the carrying capacity of the release site and any potential ne^tive-impacts to the wild population. 126 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX B University of New Hampshire Office of Sponsored RasMRh Strvfc* Buildtnf 51 College Iloea Durham. New Hampshire 03924-1555 (603) 842-3564 Bur Much 27,1997 M l. Elizabeth Fairchild c/o Hunt Howell Zoology Spauldiac Campus Mail Approval OaU: March 27, 1997 ACUC Protocol I: 970206 Deer M l. Fairchild: The Animal C m end Uia Committee hat reviewed i d approved the protocol submitted far the study "Feasibility of Winter Flounder for Stock Enhancement" under Category I on Page 3 of the "Application far Review of Animal Use or Instruction Protocol" - the research involves either no pain or potentially involves momentary, slight pain, discomfort or stress. Note: All cage, pen or other animal identification records must include your ACUC Protocol # ai lilted above. Please note: All approvals are issued contingent upon partidpahon in the UNH Health Services Occupational Health Program fa r Animal U ian and Caretakers. Participatioa is required for all principal investigators sod their project-affiliated personnel, employees of the University sod students alike. Project-related health services me provided at no coet to the employee or undent. To set up appointments with Health Services, please call 862- 1579. If you have difficulty reaching Health Services or in setting up an appointment, please feel ftee to contact Kara Eddy, Regulatory Compliance Officer, it 862-2003 for sasinance. Sincerely, C U s* j Rohan O. Mair, PhD. Chairperson Animal C an and Use Committee R G M tke 127 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. List of References Aagaard, A., C.G. Warman, and M.H. Depledge. 199S. Tidal and seasonal changes in the temporal and spatial distribution of foraging Carcinus maenas in the weakly tidal littoral zone of Kerteminde Fjord, Denmark. Mar. Ecol. Prog. Ser. 122: 165-172. Abookire, A.A. and B.L. Norcross. 1998. Depth and substrate as determinants of distribution of juvenile flathead sole ( Hippoglossoides elassodon) and rock sole (Pleuronectes bilineatus), in Kachemak Bay, Alaska. J. Sea Res. 39: 113-123. Ansell, A.D., C.A. Comely, and L. Robb. 1999. Distribution, movements and diet of macrocrustaceans on a Scottish sandy beach with particular reference to predation on juvenile fishes. Mar. Ecol. Prog. Ser. 176: 115-130. Ansell, A.D. and R.N. Gibson. 1993. The effect of sand and light on predation of juvenile plaice (Pleuronectes platessa) by fishes and crustaceans. J. Fish Biol. 43: 837- 845. Armstrong, M.P. 1997. Seasonal and ontogenetic changes in distribution and abundance of smooth flounder, Pleuronectes putnami, and winter flounder, Pleuronectes americanus, along estuarine depth and salinity gradients. Fish. Bull. 95:414-430. Armstrong, M.P. 1995. A comparative study of the ecology of smooth flounder (Liopsetta putnami) and winter flounder ( Pleuronectes americanus) from Great Bay Estuary, New Hampshire. PhD Thesis, University of New Hampshire, Durham, NH. Bergman, M.J.N., H.W. van der Veer, and J.J. Zijlstra. 1988. Plaice nurseries: effects on recruitment. J. Fish Biol. 33 (Suppl. A): 201-218. Bigelow, H.B. and W.C. Schroeder. 1953. Fishes of the Gulf of Maine. U.S. Fish Wildl. Serv., Fish. Bull. U.S. 53, 577p. Birt, V.L., T.P. Birt, D. Goulet, D.K. Cairns, and W.A. Montevecchi. 1987. Ashmole’s halo: direct evidence for prey depletion by a seabird. Mar. Ecol. Prog. Ser. 40: 205-208. Blackwell, B.F., W.B. Krohn, and R.B. Allen. 1995. Foods o f nestling double-crested cormorants in Penobscot Bay, Maine, USA: temporal and spatial comparisons. Colonial waterbirds 18(2): 199-208. 128 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Blankenship, H.L. and K.M. Leber. 1995. A responsible approach to marine stock enhancment. In Schramm, H.L. Jr. and Piper, R.G. (eds.), Uses and effects of cultured fish in aquatic ecosystems. Amer. Fish. Soc. Symp. 15: 167-175. Blaxter, J.H.S. 1975. Reared and wild fish - how do they compare? 10th European Symp. on Marine Biology, Ostend, Belgium. 1: 11-26. Blaxter, J.H.S. 1970. Sensory deprivation and sensory input in rearing experiments. Helgolander wiss. Meeresunters. 20: 642-654. Buckley, J. 1989. Species profile: life histories and environmental requirements of coastal fish and invertebrates (North Atlantic) - winter flounder. U.S. Fish Wildl. Biol. Rept. 82(11.87), 12p. Casterlin, M.E. and W.W. Reynolds. 1982. Thermoregulatory behavior and diel activity of yearling winter flounder, Pseudopleuronectes americanus (Walbaum). Env. Biol. Fish. 7(2): 177-180. Chambers, R.C., W.C. Leggett, and J.A. Brown. 1988. Variation in and among early life history traits of laboratory-reared winter flounder Pseudopleuroncectes americanus. Mar. Ecol. Prog. Ser. 47: 1-15. Choat, J.H. 1982. Fish feeding and the structure of benthic communities in temperate waters. Ann. Rev. Ecol. Syst. 13:423-429. Clark, S. (ed). 1998. Status of fishery resources off the northeastern United States for 1998. NOAA Tech. Memo. NMFS-NE-115. Cohen, A.N., J.T. Carlton, and M.C. Fountain. 1995. Introduction, dispersal and potential impacts of the green crab Carcinus maenas in San Francisco Bay, California. Mar. Bio. 122:225-237. Cowx, I.G. 1994. Stocking strategies. Fish. Manage, and Ecol. 1: 15-30. Crothers, J.H. 1968. The biology of the shore crab Carcinus maenas (L.). 2. The life of the adult crab. Field Studies 2: 579-614. Cushing, D.H. 1974. The possible density-dependence of larval mortality and adult mortality in fishes. In Blaxter, J.H.S. (ed.), The early life history of fish. Springer-Verlag, Berlin, pp. 103-111. 129 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Danielssen, D.S., B.R. Howell, and E. Moksness. (eds.) 1994. An international symposium on sea ranching of cod and other marine species, Arendal, Norway, 15-18 June 1993). Aquaculture and Fish. Manage. 25(1): 1-264. Dhert, P., P. Lavens, M. Dehasque, and P. Sorgeloos. 1994. Improvements in the larviculture of turbot Scophthalmus maximus: zootechnical and nutritional aspects, possibilities for disease control. Special Publ. No. 22, European Aquaculture Soc. Dill, D.M. 1974. The escape response of the zebra danio ( Brachydanio rerio). II. The effect of experience. Animal Behavior 22: 723-730. Drawbridge, M.A., D. Kent, M.A. Shane, and R.F. Ford. 1995. The assessment of marine stock enhancement in southern California: a case study involving the white seabass. Amer. Fish. Soc. Symp. 15: 568-569. Fairchild, E.A. and W.H. Howell. 2001.Optimal stocking density for juvenile winter flounder Pseudopleuronectes americanus. J. World. Aquacult. Soc. 32(3): 300- 308. Fairchild, E.A. and W.H. Howell. 2000. Predator-prey size relationship between Pseudopleuronectes americanus and Carcinus maenas. J. Sea Res. 44: 81-90. Fairchild, E.A. 1998. Winter flounder, Pseudopleuronectes americanus, stock enhancement: a preliminary investigation into the performance of cultured, juvenile fish. MS Thesis, University of New Hampshire, Durham, NH. Folkvord, A., G. Blom, O. Dragesund, A. Johannessen, O. Nakkenm, and G. Naevdal. 1994. A conceptual framework for enhancing and studying recruitment of marine fish stocks. Aquaculture and Fish. Manage. 25, Supplement 1:245-258. Frame, D.W. 1973. Conversion efficiency and survival of young winter flounder (Pseudopleuronectes americanus) under experimental conditions. Trans. Amer. Fish. Soc. 102(3): 614-617. Fujii, T. and M. Noguchi. 1996. Feeding and growth of Japanese flounder ( Paralichthys olivaceus) in the nursery ground. In: Watanabe, Y., Yamashita, Y., and Oozeki, Y. (eds.), Proceedings of an international workshop: survival strategies in early life stages of marine resources. A.A. Balkema, Brookfield, VT, pp. 141-151. Fukusho, K. 1993. Mass larval rearing of finfish for sea ranching in Japan. Intern. Symp. on Sea Ranching of Cod and Other Species, Arendal, Norway, 15-18 June 1993. 130 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gibson, R.N., L. Pihl, M.T. Burrows, J. Modin, H. Wennhage, and L.A. Nickell. 1998. Diel movements of juvenile plaice Pleuronectes platessa in relation to predators, competitors, food availability and abiotic factors on a microtidal nursery ground. Mar. Ecol. Prog. Ser. 165: 145-159. Gibson, R.N. 1994. Impact of habitat quality and quantity on the recruitment of juvenile flatfishes. Neth. J. Sea Res. 32(2): 191-206. Gibson, R.N. and L. Robb. 1992. The relationship between body size, sediment grain size and the burying ability of juvenile plaice, Pleuroncectes platessa L. J. Fish Biol. 40: 771-778. Gibson, R.N., M.C. Yin, and L. Robb. 1995. The behavioral basis of predator-prey size relationships between shrimp (Crangon crangon) and juvenile plaice (Pleuronectes platessa). J. mar. biol. Ass. U.K. 75: 337-349. Grizzle, R. Jackson Estuarine Laboratory, University of New Hampshire, Durham, NH, unpublished data. Guindon, K.Y. and J.M. Miller. 1995. Growth potential of juvenile southern flounder, Paralichthys lethostigma, in low salinity nursery areas of Pamlico Sound, North Carolina, USA. Neth. J. Sea Res. 34(1-3): 89-100. Houde, E.D. 1971. Developmental abnormalities of the flatfish Achirus lineatus reared in the laboratory. Fish. Bull. 69(3): 537-544. Howe, A.B., P.G. Coates, and D.E. Pierce. 1976. Winter flounder estuarine year-class abundance, mortality, and recruitment. Trans. Am. Fish. Soc. 105: 647-657. Howell, B.R. 1994. Fitness of hatchery-reared fish for survival in the sea. Aquaculture and Fish. Manage. 25(1): 3-17. Howell, B.R. and S.M. Baynes. 1993. Are hatchery-reared sole equipped for survival in the sea? ICES CM 1993/F:33 SESS R, 1 lp. Howell, W.H. and E. Fairchild. Assessing the feasibility of winter flounder (Pseudopleuronectes americanus) stock enhancement. SeaGrant Project R/FMD- 158. Howell, W.H. and M.K. Litvak. 2000. Winter flounder culture. In: Stickney, R.R. (ed.), Encyclopedia of aquaculture. John Wiley and Sons, Inc., New York, New York, pp. 998-1005. 131 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Iglesias, J. and G. Rodriguez-Ojea. 1994. Fitness of hatchery reared turbot, Scopthalmus maximus L., for survival in the sea: first year results on feeding, growth and distribution. Aquaculture and Fisheries Manage. 25(1): 179-188. Ivlev, V.S. 1961. Experimental ecology of the feeding of fishes. Yale University Press, New Haven, 302p. Jarvi, T. 1990. Cumulative acute physiological stress in Atlantic salmon smolts: the effect of osmotic imbalance and the presence of predators. Aquacult. 89: 337-350. Jirsa, D. and M. Drawbridge. 2000. Fingerling production of California halibut Paralichthys califomicus in a shallow water raceway system. Aquaculture America 2000, New Orleans, LA (abstract). Kamermans, P., K.Y. Guindon, and J.M. Miller. 1995. Importance of food availability for growth of juvenile southern flounder ( Paralichthys lethostigma) in the Pamlico River Estuary, North Carolina, USA. Neth. J. Sea Res. 34(1-3): 101-109. Kanayama, Y. 1968. Studies of the conditioned reflex in lower vertebrates. X. Defensive conditioned reflex of chum salmon fry in group. Marine Biol. 2: 77-87. Kellison, G.T., D.B. Eggleston, and J.S. Burke. 2000. Comparative behavior and survival of hatchery-reared versus wild summer flounder ( Paralichthys dentatus). Can. J. Fish. Aquat. Sci. 57: 1870-1877. Kitada, S. 1999. Effectiveness of Japan’s stock enhancement programmes: current perspectives. In: Howell, B.R., Moksness, E., and Sv&sand, T. (eds.), Stock enhancement and sea ranching. Fishing News Books, Oxford, pp. 103-131. Klein-MacPhee, G. 1978. Synopsis of biologica data for the winter flounder, Pseudopleuronectes americanus (Walbaum). NOAA Tech. Rept. NMFS Circular 414,43p. Klein-MacPhee, G., W.H. Howell, and A.D. Beck. 1982. Comparison of a reference strain and four geographical strains oiArtem ia as food for winter flounder, Pseudopleuronectes americanus, larvae. Aquaculture 29: 279-284. Klein-MacPhee, G., B.K. Sullivan, and A.A. Keller. 1993. Using mesocosms to assess the influence of food resources and toxic materials in larval fish growth and survival. Amer. Fish. Soc. Symp. 14:105-116. Koshiishi, Y., H. Itano, and Y. Hirota. 1991. Artificial stock size improvement of the flounder, Paralichthys olivaceous: present status of technological achievement. In: Marine Ranching: Proceedings of the 17th US-Japan meeting on aquaculture. NOAA Tech. Rep. NMFS 102, pp. 33-43. 132 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kristiansen, T.S., H. Ottera, and T. Sv&sand. 2000. Size-dependent mortality of juvenile Atlantic cod estimated from recaptures of released reared cod and tagged wild cod. J. Fish Bio. 56: 687-712. Kuipers, B.R., B. Maccurrin, J.M. Miller, H.W. van der Veer, and J.I.J. White. 1992. Small trawls in juvenile flatfish research: their development and efficiency. Neth. J. Sea Res. 29(1-3): 109-117. Laurence, G.C. 1977. A bioenergetic model for the analysis of feeding and survival potential of winter flounder, Pseudopleuronectes americanus, larvae during the period from hatching to metamorphosis. Fish. Bull. U.S. 75: 529-546. Laurence, G.C. 1975. Laboratory growth and metabolism of the winter flounder Pseudopleuronectes americanus from hatching through metamorphosis at three temperatures. Mar. Biol. (Berl.) 32: 223-229. Leber, K.M. 1999. Rationale for an experimental approach to stock enhancement. In: Howell, B.R., Moksness, E., and Sv&sand, T. (eds.), Stock enhancement and sea ranching. Fishing News Books, Oxford, pp.63-75. Leber, K.M. 1995. Significance of fish size-at-release on enhancement of striped mullet fisheries in Hawaii. J. World Aquacult. Soc. 26(2): 143-153. Leber, K.M., S.M. Arce, D.A. Sterritt, and N.P. Brennan. 1996. Marine stock- enhancement potential in nursery habitats of striped mullet, Mugil cephalus, in Hawaii. Fish. Bull. U.S. 94:452-471. Leber, K.M., N.P. Brennan, and S.M. Arce. 1998. Recruitment patterns of cultured juvenile Pacific threadfin, Polydactylus sexfilis (Polynemidae), released along sandy marine shores in Hawaii. Bull. Mar. Sci. 62(2): 389-408. Leber, K.M., N.P. Brennan, and S.M. Arce. 1995. Marine enhancement with striped mullet: Are hatchery releases replenishing or displacing wild stocks? Amer. Fish. Soc. Symp. 15: 376-387. Leber, K.M., H.L. Blankenship, S.M. Arce, and N.P. Brennan. 1997. Influence of release season on size-dependent survival of cultured striped mullet, Mugil cephalus, in a Hawaiian estuary. Fish. Bull. 95:267-279. Leber, K.M. and C.S. Lee. 1997. Marine stock enhancement potential with striped mullet, Mugil cephalus, in Hawaii. Bull. Natl. Res. Inst. Aquacult., Suppl. 3: 117-134. Lee, G.W. and M.K. Litvak. 1996. Weaning wild young-of-the-year winter flounder, Pleuronectes americanus (Walbaum) on a dry diet: effects on growth, survival and feed efficiency ratios. J. World Aqua. Soc. 27: 30-39. 133 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Leopold, M.F., C.J.G. van Damme, and H.W. van der Veer. 1998. Diet of cormorants and the impact of cormorant predation on juvenile flatfish in the Dutch Wadden Sea. J. Sea Res. 40:93-107. Leshcheva, T.S. 1968. Formation of defensive reflexes in roach (Rutilus rutilus L.) larvae through imitation. Prob. of Ichthyology 8(6): 838-841. Linton, E. 1921. Food of young winter flounders. Rep. U.S. Comm. Fish. Wash, for 1921, pp. 1-14. Lockwood, A.P.M. 1967. Aspects of the physiology of Crustacea. W.H. Freeman and Co., San Francisco. Magurran, A.E. 1990. The inheritance and development of minnow anti-predatory behavior. Animal Behavior 39: 834-842. Manderson, J.P., B.A. Phelan, A.W. Stoner, and J. Hilbert. 2000. Predator-prey relations between age-l+ summer flounder {Paralichthys dentatus, Linnaeus) and age-0 winter flounder {Pseudopleuronectes americanus, Walbaum): predator diets, prey selection, and effects of sediments and macrophytes. J. Exp. Mar. Biol. Ecol. 251: 17-39. Manderson, J.P, B.A. Phelan, A.J. Bejda, L.L. Stehlik, and A.W. Stoner. 1999. Predation by striped searobin ( Prionotus evolans, Triglidae) on young-of-the-year winter flounder {Pseudopleuronectes americanus, Walbaum): examining prey size selection and prey choice using field observations and laboratory experiments. J. Exp. Mar. Biol. Ecol. 242:211-231. Marliave, J.B., S.J.D. Martell, D.I. Kent, and H.L. Blankenship. 1997. Propogation and transplanting of lingcod {Ophiodon elongatus). World Aquaculture 97, Seattle, WA. (abstract). Masuda, R. and K. Tsukamoto. 1998. Stock enhancement in Japan: review and perspective. Bull. Mar. Sci. 62(2): 337-358. Maynard, D J . and T.A. Flagg. 2001. NATURES rearing as a tool for increasing ranched salmon survival. World Aquaculture 32(2): 56-69. Maynard, D.J., T.A. Flagg, and C.V.W. Mahnken. 1995. A review of seminatural culture strategies for enhancing the postrelease survival of anadromous salmonids. Amer. Fish. Soc. Symp. 15: 307-314. McEachron, L.W., R.L. Colura, B.W. Bumguardner, and R. Ward. 1998. Survival of stocked red drum in Texas. Bull. Mar. Sci. 62(2): 359-368. 134 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. McEachron, L.W. and K. Daniels. 1995. Red drum in Texas: a success story in partnership and commitment. Fisheries 20(3): 6-8. McEachron, L.W., C.E. McCarty, and R.R. Vega. 1995. Beneficial uses of marine fish hatcheries: enhancement of red drum in Texas coastal waters. American Fish. Soc. Symp. 15:161-166. Miller, T.W. 1996. First record of the green crab, Carcinus maenas, in Humboldt Bay, California. California Fish and Game 82:93-96. Miyazaki, T., R. Masuda, S. Furuta, and K. Tsukamoto. 1997. Laboratory observations on the nocturnal activity of hatchery-reared juvenile Japanese flounder Paralichthys olivaceus. Fish. Sci. 63(2): 205-210. Modlin, R.F. 1980. The life cycle and recruitment of the sand shrimp, Crangon septemspinosa, in the Mystic River Estuary, Connecticut. Estuaries 3(1): 1-10. Moles, A. and B.L. Norcross. 1995. Sediment preference in juvenile Pacific flatfishes. Neth. J. Sea Res. 34(1-3): 177-182. Nardi, G.C. 1996. Commercial tank culture of summer flounder. Final Report of Saltonstall-Kennedy Grant No. NA26FD0042-01. Naylor, E. 1958. Tidal and diurnal rhythms of locomotory activity in Carcinus maenas (L.). J. Exp. Bio. 35:602-610. NEFSC (Northeast Fisheries Science Center). 2002. Report of the 34th northeast regional stock assessment workshop (34th SAW) stock assessment review committee (SARC) consensus summary of assessments. Northeast fisheries science center. Reference document number unassigned as of 4/15/02. NEFSC. 1998. Report of the 28th northeast regional stock assessment workshop (28lh SAW) stock assessment review committee (SARC) consensus summary of assessments. Northeast fisheries science center. Ref. Doc. 99-08,304 p. New, M.B. 1997. Aquaculture and the capture fisheries - balancing the scales. World Aquaculture 28(2): 11-30. Nodtvedt, M., A. Femo, J. Gjesaeter, and P. Steingrund. 1999. Anti-predator behavior of hatchery-reared and wild juvenile cod (Gadus morhua L.), and the effect of predator training. In Howell, B.R., Moksness, E., and Sv&sand, T. (eds.), Stock enhancement and sea ranching. Fishing News Books, Oxford, pp. 350-362. 135 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Nordeide, J.T. and A.G.V. Salvanes. 1991. Observations on reared newly released and wild cod (Gadus morhua) and their potential predators. ICES mar. Sci. Symp. 192: 139-146. Northeast Utilities Service Company (NUSCO). 2001. The entrainment of fish eggs and larvae at Millstone Nuclear Power Station. In: Millstone Power Station: an evaluation of cooling water system alternatives, part 2, Northeast Utilities Environmental Laboratory, Waterford, CT, pp. 1-68. O’Grady, M.F. 1983. Observations on the dietary habits of wild and stocked brown trout, Salmo salar L. in Irish lakes. J. Fish Biol. 22: 393-601. Olla, B.L. and M.W. Davis. 1989. The role of learning and stress in predator avoidance of hatchery-reared coho salmon (Oncorhynchus kitsutch) juveniles. Aquaculture 76: 209-214. Olla, B.L. and M.W. Davis. 1988. To eat or be eaten: do hatchery-reared salmon need to leam survival skills? Underwater Naturalist 17(3): 16-18. Olla, B.L., M.W. Davis, and C.H. Ryer. 1998. Understanding how the hatchery environment represses or promotes the development of behavioral survival skills. Bull. Mar. Sci. 62(2): 531-550. Olla, B.L., M.W. Davis, and C.H. Ryer. 1994. Behavioral deficits in hatchery-reared fish: potential effects on survival following release. Aquaculture and Fish. Manage. 25(1): 19-34. Olla, B.L., M.W. Davis, and C.B. Schreck. 1995. Stress-induced impairment of predator evasion and non-predator mortality in Pacific salmon. Aquacult. Res. 26: 393- 398. Oviatt, C.A. and S.W. Nixon. 1973. The demersal fish of Narragansett Bay: An analysis of community structure, distribution and abundance. Estuarine Coastal Mar. Sci. 1:361-378. Parker, N.C. 1986. Physical and biological factors influencing growth and survival of stocked fish. In: R.H. Stroud (ed.), Fish Culture in Fisheries Management, Amer. Fish. Soc., Bethesda, MD, pp. 235-241. Patten, B.G. 1977. Body size and learned avoidance factors affecting predation on coho salmon, Oncorhynchus kisutch, fry by Torrent sculpin, Cottus rhotheus. Fish. Bull. U.S. 75:457-459. 136 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pearcy, W.G. 1962. Ecology of an estuarine population of winter flounder Pseudopleuronectes americanus Walbaum. Bull. Bingham Oceanogr. Collect., Yale Univ. 18(1), 78p. Pereira, J.J., R. Goldberg, J.J. Ziskowski, P.L. Berrien, W.W. Morse, and D.L. Johnson. 1999. Essential fish habitat source document: winter flounder, Pseudopleuronectes americanus, life history and habitat characteristics. NOAA Tech. Mem. NMFS-NE-138,39p. Perlmuter, A. 1947. The blackback flounder and its fishery in New England and New York. Bull. Bingham Oceanographic Collect., Yale Univ. 11(2), 92p. Person-LeRuyet, J., F. Baudin-Laurencin, N. Devauchelle, R. Metaliller, J. Nicholas, J. Robin, and J. Guillaume. 1991. Culture of turbot ( Scophthalmus maximus). In: J.P. McVey (ed.) Handbook of Mariculture, Vol. II, Finfish Aquaculture, CRC Press, Boca Raton, FL, pp. 21-41. Phelan, B.A., J.P. Manderson, A.W. Stoner, and A.J. Bejda. 2001. Size-related shifts in the habitat associations of young-of-the-year winter flounder (Pseudopleuronectes americanus): field observations and laboratory experiments with sediments and prey. J. Exp. Mar. Biol. Ecol. 257(2): 297-315. Phelan, B.A., J.P. Manderson, A.W. Stoner, and A.J. Bejda. 2000. Estuarine and habitat- related differences in growth rates of young-of-the-year winter flounder (Pseudopleuronectes americanus) and tautog (Tautoga onitis) in three northeastern U.S. estuaries. J. Exp. Mar. Biol. Ecol. 247(1): 1-28. Pihl, L., J. Modin, and H. Wennhage. 2000. Spatial distribution patterns of newly settled plaice (Pleuronectes platessa L.) along the Swedish Skagerrak archipelago. J. Sea Res. 44:65-80. Pilon, C., J. Burton, and R. McNeil. 1982. Summer food of the great and double-crested cormorants on the Magdalen Islands, Quebec. Can. J. Zool. 61: 2733-2739. Polovina, J.J. 1991. Evaluation of hatchery releases of juveniles to enhance rockfish stocks, with application to Pacific ocean perch Sebastes alutus. Fish. Bull. U.S. 89: 129-136. Poole, J.C. 1964. Feeding habits of the summer flounder in Great South Bay. N.Y. Fish Game J. 11:28-34. Reichert, M.J.M. and H.W. van der Veer. 1991. Settlement, abundance, growth and mortality of juvenile flatfishes in a subtropical tidal estuary (Georgia, U.S.A.). Neth. J. Sea Res. 27(3/4): 375-391. 137 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rice, J.A., L.B. Crowder, and E.A. Marschall. 1997. Predation on juvenile fishes: dynamic interactions between size-structured predators and prey. In Chamber, R.C. and Trippel, E.A. (eds.), Early life history and recruitment in fish populations. Chapman and Hall, London, pp. 333-356. Richards, S.W. 1963. The demersal fish population of Long Island Sound. Bull. Bingham Oceanogr. Collect., Yale Univ. 18(2): lOlp. Richards, S.W., J.M. Mann, and J.A. Walker. 1979. Comparison of spawning seasons, age, growth rates, and food of two sympatric species of searobins, Prionotus carolinus and Prionotus evolans, from Long Island Sound. Estuaries 2(4): 255- 268. Richards, W.J. and R.E. Edwards. 1986. Stocking to restore or enhance marine fisheries. In: R.H. Stroud (ed.), Fish Culture in Fisheries Management, Amer. Fish. Soc., Bethesda, MD. pp.75-80. Rogers, C. A. 1976. Effects of temperature and salinity on the survival of winter flounder embryos. Fish. Bull. U.S. 74:52-58. Rogers, S.1.1992. Environmental factors affecting the distribution of sole ( Solea solea (L.)) within a nursery area. Neth. J. Sea Res. 29(1-3): 153-161. Ropes, J.W. 1968. The feeding habits of the green crab, Carcinus maenas (L.). Fish. Bull. 67: 183-203. Ross, M.R. 1997. Fisheries Conservation and Management. Prentice Hall, NJ. 374p. Saucerman, S.E., 1990. Movement, distribution, and productivity of post metamorphic winter flounder ( Pseudopleuronectes americanus) in different habitat types in Waquoit Bay, Massachusetts. MS Thesis, University of Massachusetts, Amherst, MA. Saucerman, S.E. and L.A. Deegan. 1991. Lateral and cross-channel movement of young- of-the-year winter flounder ( Pseudopleuronectes americanus) in Waquoit Bay, Massachusetts. Estuaries 14(4): 440-446. Seikai, T., I. Kinoshita, and M. Tanaka. 1993. Predation by Crangoid shrimp on juvenile Japanese flounder under laboratory conditions. Nippon Suisan Gakkaishi 59(2): 321-326. Smigielski, A.S. 1975. Hormone induced ovulation of the winter flounder, Pseudopleuronectes americanus. Fish. Bull. U.S. 73:431-438. 138 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Smith, T.I.J., W.E. Jenkins, and M.R. Denson. 1997. Overview of an experimental stock enhancement program for red drum in South Carolina. Bull. Natl. Res. Inst. Aquacult., Suppl. 3: 109-115. Sogard, S.M. 1992. Variability in growth rates of juvenile fishes in different estuarine habitats. Mar. Ecol. Prog. Ser. 85: 35-53. Sogard, S.M., K.W. Able, and S.M. Hagan. 2001. Long-term assessment of settlement and growth of juvenile winter flounder (Pseudopleuronectes americanus) in New Jersey estuaries. J. Sea Res. 45(3/4): 189-204. Sosiak, A.J., R.G. Randall, and J.A. MacKenzie. 1979. Feeding by hatchery-reared and wild Atlantic salmon ( Salmo salar) parr in streams. J. Fish. Res. Bd. Can. 36: 1408-1412. Stettrup, J.G., K. Lehmann, and H. Nicolajsen. 1998. Turbot, Scophthalmus maximus, stocking in Danish coastal waters. In: Cowx, I.J. (ed.), Stocking and introduction of fish. Fishing News Books, Oxford, pp. 301-318. Stradmeyer, L. and J.E. Thorpe. 1987. The response of hatchery reared Atlantic salmon, Salmo salar L., parr to pelleted and wild prey. Aquaculture and Fish. Manage. 18: 51-61 Suboski, M.D. and J.J. Templeton. 1989. Life skills training for hatchery fish: social learning and survival. Fisheries Research 7: 343-352. Sv&sand, T. 1998. Cod enhancement studies in Norway - background and results with emphasis on releases in the period 1983-1990. Bull. Mar. Sci. 62(2): 313-324. Sv&sand, T. 1993. Are released juveniles fit for release into the wild? ICES Statutory Mtg. 1993. C.M. 1993/F:34,13p. Sv&sand, T. and T.S. Kristiansen. 1990a. Enhancement studies of coastal cod in western Norway. Q. Migration of reared coastal cod. J. Cons. int. Explor. Mer. 47: 13-22. Svisand, T. and T.S. Kristiansen. 1990b. Enhancement studies of coastal cod in western Norway. IV. Mortality of reared cod after release. J. Cons. int. Explor. Mer. 47: 30-39. Sv&sand, T., K.E. Jorstad, and T.S. Kristiansen. 1990. Enhancement studies of coastal cod in western Norway. I. Recruitment of wild and reared cod to a local spawning stock. J. Cons. int. Explor. Mer 47:5-12. 139 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sv&sand, T., T.S. Kristiansen, T. Pedersen, A.G.V. Salvanes, R. Engelsen, G. Naevdal, and M. Nedtvedt. 2000. The enhancement of cod stocks. Fish and Fisheries 1: 173-205. Tanaka, M., T. Seikai, E. Yamamoto, and S. Furuta. 1998. Significance of larval and juvenile ecophysiology for stock enhancement of the Japanese flounder, Paralichthys olixaceus. Bull. Mar. Sci. 62 (2): 551-571. Tanda, M. 1990. Studies on burying ability in sand and selection to the grain size for hatchery-reared marbled sole and Japanese flounder. Nippon Suisan Gakkaishi 56(10): 1543-1548. Tominaga, O. and Y. Watanabe. 1998. Geographical dispersal and optimum release size of hatchery-reared Japanese flounder Paralichthys olixaceus released in Ishikari Bay, Hokkaido, Japan. J. Sea Res. 40: 73-81. Topp, R.W. 1968. An estimate of fecundity of the winter flounder, Pseudopleuronectes americanus. J. Fish. Res. Board Can. 25: 1299-1302. Tsukamoto, K., H. Kuwada, K. Uchida, R. Masuda, and Y. Sakakura. 1999. Fish quality and stocking effectiveness: behavioral approach. In: Howell, B.R., Moksness, E., and Sv&sand, T. (eds.), Stock enhancement and sea ranching. Fishing News Books, Oxford, pp. 205-218. Tsukamoto, K., R. Masuda, H. Kuwada, and K. Uchida. 1997. Quality of fish for release: behavioral approach. Bull. Natl. Res. Inst. Aquacult., Suppl. 3:93-99. Tyler, A.V. 1971a. Monthly changes in stomach contents o f demersal fishes in Passamaquoddy Bay, N.B. Fish. Res. Board Can. Tech. Rept. 288, 114p. Tyler, A.V. 1971b. Periodic and resident components in communities of Atlantic fishes. J. Fish. Res. Board Can. 28:935-946. Tyler, A.V., 1971c. Surges of winter flounder Pseudopleuronectes americanus into the intertidal zone. J. Fish. Res. Board Can. 28(11), 1727-1732. van der Meeren, G.1.2000. Predation on hatchery-reared lobsters released in the wild. Can. J. Fish. Aquat. Sci. 57: 1794-1803. van der Veer, H.W., T. Ellis, J.M. Miller, L. Pihl, and A.D. Rijnsdorp.1997. Size- selective predation on juvenile North Sea flatfish and possible implications for recruitment. In Chamber, R.C. and Trippel, E.A. (eds.), Early life history and recruitment in fish populations. Chapman and Hall, London, pp. 279-303. 140 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. van der Veer, H.W. and M.J.N. Bergman. 1987. Predation by crustaceans on a newly settled 0-group plaice Pleuronectes platessa population in the western Wadden Sea. Mar. Ecol. Prog. Ser. 35:203-215. Wahl, D.H., RA. Stein, and D.R. DeVries. 1995. An ecological framework for evaluating the success and effects of stocked fishes. Amer. Fish. Soc. Symp. 15: 176-189. Wallin, J.E. and M.J. Van Den Avyle. 1995. Interactive effects of stocking site salinity and handling stress on survival of striped bass fingerlings. Trans. Amer. Fish. Soc. 124: 736-745. Walsh, S.J. 1992. Factors influencing distribution of juvenile yellowtaii flounder CLimanda ferruginea) on the Grand Bank of Newfoundland. Neth. J. Sea Res. 29(1-3): 193-203. Wanat, J. Zoology Department, University of New Hampshire, Durham, NH, unpublished data. Waterman, T.H. (ed.). 1961. The physiology of Crustacea Vol 2: sense organs, integration, and behavior. Academic Press, New York. Waters, E.B. 1999. Flounder aquaculture and stock enhancement in North Carolina. University of North Carolina Sea Grant College Program, Raleigh, N.C., Pub. No. UNC-SG-99-02,24p. Waters, E.B. 1996. Sustainable flounder culture and fisheries. University of North Carolina Sea Grant College Program, Raleigh, N.C., Pub. No. UNC-SG-96-14, 12p. Wennhage, H. and R.N. Gibson. 1998. Influence of food supply and a potential predator {Crangon crangon) on settling behaviour of plaice ( Pleuronectes platessa). J. Sea Res. 39: 103-112. Wennhage, H. and L. Pihl. 1994. Substratum selection by juvenile plaice ( Pleuronectes platessa L.): impact of benthic microalgae and filamentous macroalgae. Neth. J. Sea Res. 32(3/4): 343-351. Wilcox, J.R. and H.P. Jeffries. 1974. Feeding habits of the sand shrimp Crangon septemspinosa. Biol. Bull. 146: 424-434. Willis, S.A., W.W. Falls, C.W. Dennis, D.E. Roberts, and P.G. Whitchurch. 1995. Assessment of season of release and size at release on recapture rates of hatchery- reared red drum. Amer. Fish. Soc. Symp. 15: 354-365. 141 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Witting, D.A. 1995. Influence of invertebrate predators on survival and growth of juvenile winter flounder. PhD Thesis, Rutgers University, New Brunswick, NJ. Witting, D.A. and K.W. Able. 1995. Predation by sevenspine bay shrimp Crangon septemspinosa on winter flounder Pleuronectes americanus during settlement: laboratory observations. Mar. Ecol. Prog. Ser. 123:23-31. Witting, D.A. and K.W. Able. 1993. Effects of body size on probability of predation for juvenile summer and winter flounder based on laboratory experiments. Fish. Bull. 91:577-581. Yamashita, Y., S. Nagahora, H. Yamada, and D. Kitagawa. 1994. Effects of release size on survival and growth of Japanese flounder Paralichthys olivaceus in coastal waters off Iwate Prefecture, northeastern Japan. Mar. Ecol. Prog. Ser. 105: 269- 276. Yamashita, Y. and H. Yamada. 1999. Release strategy for Japanese flounder fly in stock enhancement programmes. In Howell, B.R., Moksness, E., and Sv&sand, T. (eds.), Stock enhancement and sea ranching. Fishing News Books, Oxford, pp. 191-204. Yamashita, Y., H. Yamada, K.D. Malloy, T.E. Targett, and Y. Tsuruta. 1996. Sand shrimp predation on settling and newly-settled stone flounder and its relationship to optimal nursery habitat selection in Sendai Bay, Japan. Proceedings of an international workshop: Survival strategies in early life stages of marine resources. A.A. Balkema Publishers, Brookfield, VT (USA), pp. 271-284. Zar, J.H. 1996. Biostatistical analysis. Prentice-Hall, Inc., Upper Saddle River, NJ, 662p. Zijlstra, J.J., R. Dapper, and J.IJ. Witte. 1982. Settlement, growth and mortality of post larval plaice (Pleuronectes platessa) in the Western Wadden Sea. Neth. J. Sea Res. 15(2): 250-272. 142 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.