Citation
Comparative Age and Growth of Greater Amberjack (Seriola Dumerili) from Charterboat and Headboat Fisheries of West Florida and Alabama, Gulf of Mexico

Material Information

Title:
Comparative Age and Growth of Greater Amberjack (Seriola Dumerili) from Charterboat and Headboat Fisheries of West Florida and Alabama, Gulf of Mexico
Creator:
Leonard, Edward
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (54 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Fisheries and Aquatic Sciences
Forest Resources and Conservation
Committee Chair:
Murie, Debra J.
Committee Co-Chair:
Parkyn, Daryl C.
Committee Members:
Allen, Micheal S.
Fitzhugh, Gary
Graduation Date:
8/8/2009

Subjects

Subjects / Keywords:
Age structure ( jstor )
Coasts ( jstor )
Fish ( jstor )
Fisheries ( jstor )
Fishing ( jstor )
Frequency distribution ( jstor )
Gulfs ( jstor )
Otolith organs ( jstor )
Population growth ( jstor )
Population growth rate ( jstor )
Forest Resources and Conservation -- Dissertations, Academic -- UF
age, amberjack, dumerili, growth, seriola
Gulf of Mexico ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Fisheries and Aquatic Sciences thesis, M.S.

Notes

Abstract:
Recent stock assessments of greater amberjack Seriola dumerili in the Gulf of Mexico have had to use limited information regarding size-at-age, in part, due to the high degree of variability intrinsic to growth of greater amberjack. To identify possibly sources of this variability, size-at-age of greater amberjack was compared among fish landed in the charterboat and headboat fisheries of Florida, and the charterboat fishery of Alabama. Identification of sources of variability could lead to more accurate estimation of size-at-age by allowing stock assessment scientists to model within more similar source populations. Fish were collected from charterboats and headboats from the Gulf coasts of Florida and Alabama through collaboration with state and federal sampling programs, supplemented by scientific research sampling. Fish were aged using cross-sections of sagittal otoliths. Observed age was correlated with length data and compared between charterboats and headboats within Florida, and between charterboat catches from Florida and Alabama. Mean length-at-age was compared among fishery sectors for age classes 2, 3 and 4. Greater amberjack captured by charterboats in Florida were larger than those captured by charterboats in Alabama. Amberjack captured by headboats in Florida were larger at ages 3 and 4 years than those caught by Florida charterboats. Differences in size and size-at-age may be related to the distribution of fish or the distribution of specific fishing effort. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2009.
Local:
Adviser: Murie, Debra J.
Local:
Co-adviser: Parkyn, Daryl C.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31
Statement of Responsibility:
by Edward Leonard.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Leonard, Edward. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2010
Classification:
LD1780 2009 ( lcc )

Downloads

This item has the following downloads:


Full Text





CHAPTER 2
METHODS

2.1 Greater Amberjack Collections

Otoliths and otolith source data from greater amberjack landed in the Florida and Alabama

charterboat and headboat fisheries were supplied by the Gulf States Marine Fisheries

Commission under the Southeast Recreational Fisheries Information Network [RecFIN(SE)] and

by the NOAA Fisheries Laboratory in Panama City, Florida. These agencies collect amberjack

otoliths as part of their ongoing fishery-dependent monitoring programs. Additional samples

were collected by port sampling the headboats at Hubbard's Marina at Madeira Beach, Florida.

The data used in this study were a subset of a larger data set analyzed by Murie and Parkyn

(2008). Size frequencies of amberjack sampled from the Florida charterboat and headboat and

the Alabama charterboat fisheries were compared using a Kolmogorov-Smirnov D statistic

(Sokal and Rohlf 1969).

2.2 Aging

2.2.1 Otolith Measurements and Processing

Otoliths were cataloged in a database along with fish total length (TL, mm), fish fork

length (FL, mm), fish mass (M, kg; when possible), sex, date of capture, location of capture, type

of fishery (e.g., charterboat or headboat), and gear (e.g., hook-and-line or spear). All gear types

for charterboats were combined into one category because specific gear type was not available

for many samples but > 99% of samples with known gear type were caught by hook-and-line.

Prior to processing for aging, all whole otoliths were measured for otolith total length (OTL;

anterior tip of the rostrum in straight-line distance to the posterior edge), otolith antirostrum

length (OAL; anterior tip of the antirostrum in straight-line distance to the posterior edge),

otolith height (OH; maximum distance from the dorsal to ventral edge of the otolith), and













1400 -


1200


E 1000 -
E X

800 -



U-
O I
LL
400 --Thompson et al. 1999
A- Florida Headboat (this study)

200 ------ Florida Charterboat (this study)
200
-- -- Alabama Charterboat (this study)

0 1--
1 2 3 4 5 6 7 8 9 10

Age (years)



Figure 4-1. Comparison of mean observed length-at-age for greater amberjack from the Gulf of
Mexico sampled in this study compared to the von Bertalanffy growth curve of
Thompson et al. (1999).









Amendment 1 added greater amberjack to the list of species already being managed under that

plan. This amendment also established a 3 fish per person bag limit and a 28 inch fork length

minimum size limit for recreational fishers. Amendment 1 also established a 36 inch fork length

minimum size limit for the commercial fishery and implemented a requirement for a commercial

reef fish permit for all fish included in the RFFMP. Since that time, the most significant

regulatory changes to the fishery can be found in Amendments 12 and 15. Amendment 12

(December 1995) reduced the recreational bag limit for greater amberjack to 1 per person.

Amendment 15 (January 1998) closed the Gulf of Mexico greater amberjack commercial fishery

from March 1 through May 31. Subsequent amendments established harvest limits that were

expected to rebuild the stock of greater amberjack (SEDAR 9 Assessment Report 2, 2006). Gulf

of Mexico greater amberjack were declared "overfished" by NMFS on February 9, 2001, based

on the results of a stock assessment done by Turner et al. (2000). The Gulf of Mexico stock was

assessed as being overfished again in the most recent stock assessment review in 2006 (SEDAR

2006). Most recently in 2009, NOAA Fisheries Service published a new rule to limit

commercial harvest to 228 MT and limit recreational harvest to 621 MT. The rule also raises the

minimum size limit in the recreational fishery to a 30 inch fork length and establishes new

methods for adjusting annual catch limits in-season (NMFS 2008).

Assessments of stock condition depend greatly on accuracy of individual age and growth

information (Schirripa and Burns 1997; Cummings 1998; Turner et al. 2000). Burch (1979)

provided the earliest, comprehensive study of greater amberjack in southern Florida. In the Gulf

of Mexico, Beasley (1993), and later Thompson et al. (1999, which included the data from

Beasley's study), modeled age and growth of amberjack caught in several different fisheries,

including headboats (large, for-hire vessels carrying as many as 50 fishers), charterboats (smaller









LIST OF TABLES


Table page

3-1 Number of greater amberjack sampled from each fishery in the Gulf of Mexico
during 2000-2007. ...................................................... ................ 30

3-2 Relationship between greater amberjack forklength (FL, mm) and otolith total length
(OTL, mm), otolith antirostrum length (OAL, mm), and otolith height (OH, mm).
All regressions were significant at P 0.05. .......................................... .............. 30

3-3 Relationship between greater amberjack forklength (FL, mm) and otolith radius
(OTR, mm) for Florida charterboat (FL-CB), Florida headboat (FL-HB), and
Alabama charterboat (AL-CB) fisheries ...... ........ .............. .......................... 30

3-4 Mean length-at-age, standard error of the mean (SE), and sample size (n), for greater
amberjack from sampled fisheries and regions in the Gulf of Mexico. ......................... 31

3-5 Comparisons of mean length-at-age (ages 2, 3, and 4 yr old) for greater amberjack
from Florida Headboat (FLHB), Florida Charterboat (FLCH), and Alabama
Charterboat (ALCH) ............... ..... ................... ............ ......... 31









slower than charterboats and therefore may fish closer inshore on most days. Charterboats, in

turn, may fish closer to shore than commercial boats, since the latter can stay at sea for several

days prior to landing their catch. It is possible that these groups might fish in different locations,

use different techniques, and/or different gear, and these factors may select for fish of different

size and age, as well as different growth rates. Differential growth may occur due to differences

in habitat between locations. Differences in the presence or absence of benthic structure could

influence behavior and feeding success in greater amberjack. Additionally, Alabama is situated

closer to the deep water areas near the coasts of Mississippi and Louisiana. This deep water area

could provide resources not found in close proximity to Gulf of Mexico coastal Florida. The

purpose of this study was to compare the age and growth for greater amberjack caught in

headboat and charterboat fisheries on the west coast of Florida and the charterboat fishery in

coastal Alabama. These charterboat and headboat landings represent a significant portion of the

total recreational landings of greater amberjack in the Gulf of Mexico (Figs. 1-2, 1-3, andl-4).

While the overall goal of my research was to compare the age and growth of greater

amberjack caught in the charterboat and headboat fisheries off the west coast of Florida and the

charterboat fishery off Alabama, the specific objectives included: 1) to collaborate with private

fishers, state and federal fisheries agencies to collect and process otoliths to age greater

amberjack in the Gulf of Mexico, stratified by state (Florida west coast and Alabama) and by

fishery (charterboats and headboats); 2) establish aging criteria for greater amberjack in the Gulf

of Mexico based on sectioned otoliths, including validating the method using marginal-increment

analysis; and 3) model and compare age and growth of greater amberjack between headboat and

charterboat fisheries from the west coast of Florida, and between the charterboat fisheries on the

west coast of Florida and Alabama.









otoliths. Regressions between otolith measurements (total otolith length, otolith antirostrum

length, and otolith height) and amberjack fork length were all positively related (all _0.05) but

regression coefficients were all too low (i.e., r2 = 21-47%) to be predictive (Table 3-2).

3.2.2 Aging Criteria and Estimating Ages

When viewed using a stereomicroscope with transmitted light, cross-sections of amberjack

otoliths had alternating opaque and translucent zones, each pair comprising an annulus (Fig. 3-3).

Annuli were readily apparent in young fish on the dorsal, medial area of the sulcus. The core of

the otolith appeared as an opaque zone surrounded by a translucent zone at the base of the deep

sulcus (Fig. 3-3a). Annuli were visible on both the ventral and dorsal areas of the sulcus but

were much more apparent on the dorsal medial edge. In most cases, annuli that were visible on

the dorsal area of the sulcus were also visible on the ventral side (Fig. 3-3b). Otoliths from older

fish were distinctly different in shape from those of younger fish. While the core and inner area

of these otoliths was very similar to younger fish, otoliths of older fish had long processes on the

medial edges of the sulcus, as well as on the dorsal and ventral edges of the structure (Fig. 3-4).

These processes contained marks that resembled annuli, but it was very difficult to count them as

they tended to become very tightly stacked and less pronounced than the marks from earlier

years that were considered true annuli.

Qualitatively, staining otolith sections with Rapid Bone Stain did make it possible to

count annuli while using reflected light, since the core stained dark purple and opaque zones

showed as purple bands in an otherwise white background of translucent zones (Fig. 3-5a). This

lent no discernable advantage, however, over viewing unstained otoliths using transmitted light

(Fig. 3-5b). Stained otoliths were qualitatively not any easier to read than unstained otoliths, and

the staining process was time consuming. Therefore the staining method was not used in further

analysis.









aluminum oxide powder suspended in de-ionized water. All slides were rinsed in de-ionized

water and allowed to air-dry before aging.

2.2.2 Aging Criteria and Age Estimation

Aging criteria were established by viewing an initial subset of 100 sectioned otoliths. To

establish aging criteria, otoliths were viewed for clarity of the opaque and translucent zones over

the sectioned surface, for inconsistencies between the ventral and dorsal portions of the otolith,

and for distinguishing between true annuli and false annuli or checks. Checks are small lines or

marks within the translucent portion of the otolith that resemble annuli but do not continue

through the entire otolith. These marks are not considered true annuli as they only appear in a

limited area and at some point merge into complete annuli (Chilton and Beamish 1982).

A subsample of 24 pairs of otoliths were subjected to a staining regimen to evaluate the

feasibility and effectiveness of this procedure in improving otolith readability, based on the

method improving readability of pompano (Trachinotus carolinus) otoliths (pers. comm., Cathy

Guindon, Florida Fish and Wildlife Marine Research Institute, St. Petersburg, FL). Both otoliths

of each pair were sectioned and polished as described above. After polishing, one of each pair of

the prepared otolith slides was immersed in Sanderson's Rapid Bone Stain at 40 C for 8 hours.

Stained and unstained otoliths were then visually compared to evaluate readability.

For aging purposes, the number of pairs of opaque/translucent zones were enumerated

following Chilton and Beamish (1982) and specifically for greater amberjack (Harris 2004;

Harris et al. 2007). In addition, to be able to assign the fish into comparable age classes (based

on a 1 January birth date, Chilton and Beamish 1982), the amount of growth on the otolith's edge

after the deposition of the last opaque zone was semi-quantitatively characterized as 0 (opaque

zone was just visible on the edge of the otolith but with no translucent growth after it), 1 (the

amount of translucent growth after the ultimate opaque zone was more than zero but < 1/3 of the









width of the previously complete annulus), 2 (the amount of translucent growth past the ultimate

opaque zone was > 1/3 but < 2/3 of the width of the previously complete annulus), and 3 (the

amount of translucent growth past the ultimate opaque zone was > 2/3 of the width of the

previously complete annulus). In addition, a qualitative score of how easily the otolith was read

was recorded, ranging from '1' being "clear and distinct annuli, first annulus well defined, edge

well defined" to '4' being annulii diffuse and not distinct anywhere in section (i.e., unreadable)."

Previous experience by P. Harris (SCDNR, personal communication to D. Murie) has shown this

to be advantageous when comparing precision between readers.

Otoliths were read twice by the primary reader (EEL) independently, with at least 2

weeks between aging periods and with no knowledge of the size or date of collection of the fish.

For fish captured after January 1 and having significant translucent growth beyond the last

opaque zone, age class was determined by the annulus count plus one. When these two ages

agreed, this age was considered to be the resolved age. In cases where these two ages did not

agree, the otoliths were read a third time by the primary reader independent of the first two ages.

When two of the three assigned ages agreed, that age was considered to be the resolved age.

Precision was estimated by calculating percent agreement (Sikstrom 1983), index of precision

(D) (Chang 1982), average percent error (APE) (Beamish and Fournier 1981) and the coefficient

of variation (CV) (Kimura and Lyons 1991).

2.2.3 Determination of the First Annulus

Determining where the first annulus is deposited can be a problem when aging fish,

including greater amberjack. To attempt to address this specific aging criterion, two young-of-

the-year (YOY) fish collected in July in a fishery-independent trawl survey (SEAMAP, NOAA

Fisheries, Pascagoula Laboratory) were used to evaluate the location of the first annulus. The

otoliths of these fish were extracted and processed in the same manner as the other samples.










TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................... ...................3

L IST O F T A B L E S .................................................................. ....................................... . 5

L IST O F F IG U R E S .................................................................. .................................... . 6

A B S T R A C T ................... ................... ...................8..........

CHAPTER

1 IN T R O D U C T IO N ................................................................. ....................................10

2 M E T H O D S ............................................................................................................2 0

2.1 Greater Amberjack Collections ................................. ..............................20
2 .2 A going ...................... ......................... .... ... ........................... ............... ........ 20
2.2.1 Otolith Measurements and Processing ....................................................20
2.2.2 Aging Criteria and Age Estimation .................................................... 22
2.2.3 Determination of the First Annulus ....................................................... 23
2.3 Validation of Aging Method: Periodicity and Timing of Annulus Formation .............24
2 .4 A g e an d G ro w th ........................................................................................................ 2 4

3 R E S U L T S .............. .... ............. ................. ....................................................... 2 6

3.1 Greater Amberjack Collections .............. ..... ......... ................. 26
3.2 A going .............. .......... ... ............... .......... .............. 26
3.2.1 Otolith Measurements and Processing ........... ......... .................26
3.2.2 Aging Criteria and Estimating Ages ................ ..................27
3.2.3 Determination of the First Annulus ........................................ ....28
3.3 Validation of Aging Method: Marginal-Increment Analysis ........... ... ............... 28
3 .4 A g e an d G ro w th ........................................................................................................ 2 9

4 D IS C U S S IO N .............. .... ............. ................. ................................................... 4 2

R E F E R E N C E S ..................................................................................50

BIOGRAPHICAL SKETCH ............... ......... ......... ............ 54









ACKNOWLEDGMENTS

This thesis would not have been possible with the help and support of my advisor, Dr.

Debra Murie. Dr. Murie has guided me through this process with patience and understanding for

the circumstances in life that have made a quick finish impossible. I thoroughly enjoyed

working in the Murie Lab and I am grateful to have had the opportunity to collaborate with her.

Special thanks also go to Dr. Daryl Parkyn for his hard work, intelligent insight and constant

entertainment. I also thank the other members of my committee; Dr. Mike Allen and Dr. Gary

Fitzhugh, who provided valuable insight and advice throughout this process.

I would also like to thank the captains and crew of Hubbard's Marina in Madeira Beach,

FL, for providing access to their catch as well as the members of the Gulf States Marine

Fisheries Commission and NOAA Fisheries who also supplied samples. Many of the students

and staff of the Murie Lab and Lindberg Lab also provided invaluable assistance and made the

work a pleasure. These individuals include: Ivy Baremore, Liz Berens, Doug Colle, Jaclyn

Debicella Leonard, Rick Kline, Doug Marcinek, and Pat O'Day.

My family and friends provided continuing encouragement, support, and humor which has

helped me complete this process. I am especially thankful to my parents, Pat and Jerry Leonard,

for always supporting and encouraging me. A great deal of gratitude goes to my brothers and

sister, Jerry, Tony, and Tara for their love and support. I would also like to express thanks to my

wife, Jaclyn Debicella Leonard, who makes me better that I am. Financial support for this

project was provided by a NOAA Marine Fisheries Initiative (MARFIN) grant and the

University of Florida, College of Agricultural and Life Sciences, Program of Fisheries and

Aquatic Sciences.









BIOGRAPHICAL SKETCH

Edward Leonard was born and raised in Marietta, Georgia. After serving 4 years in the

United States Army, he returned home and graduated with a Bachelor of Science degree in

biology from Kennesaw State University in Kennesaw, GA. Eddie worked at the United States

Geological Survey in Gainesville, Florida before entering graduate school at the University of

Florida. He is currently employed as a Freshwater Fisheries Biologist with the Florida Fish and

Wildlife Conservation Commission.









LIST OF FIGURES


Figure page

1-1 Annual landings of greater amberjack in the Gulf of Mexico (NOAA Fisheries pers.
comm .; Cummings and M cClellan 2000) ............... ............................. .............. 16

1-2 Annual landings of greater amberjack in recreational fisheries of Alabama and the
west coast of Florida (NOAA Fisheries pers. comm.). ................................................. 17

1-3 Annual landings of greater amberjack in the charterboat fisheries of Alabama and the
west coast of Florida (NOAA Fisheries pers. comm.). ................................................. 18

1-4 Annual landings of greater amberjack in Florida's headboat fishery and the total
from the Gulf of Mexico headboat fishery (SEDAR 2006)..................................... 19

3-1 Length frequencies of greater amberjack from the Gulf of Mexico caught in: a)
headboats and b) charterboats off the west coast of Florida; and c) charterboats from
A lab am a. ............................................................. ............... 32

3-2 Comparative length frequency distributions for greater amberjack from the Gulf of
Mexico caught in: a) charterboat and headboat fisheries off the west coast of Florida;
and b) charterboats from Florida and Alabama......... ........... ..................... 33

3-3 Cross-sections of otoliths from greater amberjack: a) 3-year old fish, with the core
(C) clearly demarked; and b) 5-yr old fish.......... .. ....... .. ................................... 34

3-4 Otolith from an older greater amberjack showing the long processes on the medial
edges of the sulcus, as well as on the dorsal and ventral edges of the otolith. ................35

3-5 a) Otolith section stained with Rapid Bone Stain viewed under reflected light; and b)
otolith section from the same greater amberjack that was not stained as viewed with
transm itted light .............. .... ........................ ........... ............ 36

3-6 Otolith section from a) a young-of-the-year greater amberjack and b) a 1 year old
fish showing growth past the first annulus ........................................... .............. 37

3-7 Mean edge code of greater amberjack otoliths (n=467) as a function of month of
collection throughout the year. Vertical bars represent standard error of the mean.
Minimum mean edge code in June indicated the formation of the opaque zone .............38

3-8 Age frequency of greater amberjack from the Gulf of Mexico by location and fishery
(n = 4 6 8 ) ........................... ........... ...... ....................... ............... 3 9

3-9 Greater amberjack fork length as a function of otolith radius (n=455)..........................40

3-10 a) Length-at-age for greater amberjack caught in all categories (n=468); b) Mean
length-at-age for greater amberjack from age classes 2, 3, and 4.................. ................41



















3.0


2.5


i 2.0 -
0

1.5
LIU

1.0


0.5


0.0
0 1 2 3 4 5 6 7 8 9 10 11 12
Month


Figure 3-7. Mean edge code of greater amberjack otoliths (n=467) as a function of month of
collection throughout the year. Vertical bars represent standard error of the mean.
Minimum mean edge code in June indicated the formation of the opaque zone.









REFERENCES


Beamish, R.J. 1981. Use of fin ray sections to age walleye pollock, pacific cod, and albacore,
and the importance of this method. Trans. Am. Fish Soc. 110: 287-299.

Beamish, R.J. and H.H. Harvey. 1969. Age determination in the white sucker. J. Fish. Res. Bd.
Canada. 26: 633-638.

Beamish, R.J. and D. Chilton. 1977. Age determination oflingcod (Ophiodon elongates) using
dorsal fin rays and scales. J. Fish. Res. Bd. Canada. 34 (9): 1305-1313.

Beamish, R.J. and D.A. Fournier. 1981. A method for comparing the precision of a set of age
determinations. Canadian Journal of Fisheries and Aquatic Sciences 38: 982-983.

Beasely, M. 1993. Age and growth of greater amberjack, Seriola dumerili, from the northern
Gulf of Mexico. M.S. Thesis, Dept. of Oceanography and Coastal Sciences, Louisiana
State University. 85pp.

Bohlke, J.E. and C.C. Chaplin. 1993. Fishes of the Bahamas and adjacent waters. 2nd edition.
University of Texas Press, Austin. 773 pp.

Burch, R.K. 1979. The greater amberjack, Seriola dumerili: its biology and fishery off
Southeastern Florida. Unpublished M.S. Thesis. University of Miami. 112 pp.

Brennan, J.S. and G.M. Cailliet. 1989. Comparative age determination techniques for white
sturgeon in California. Trans. Am. Fish. Soc. 118: 296-310.

Brett, J.R. 1979. Environmental Factors and Fish Growth. In Fish Physiology, Vol. 8, ed. W.S.
Hoar, D.J. Randall, and J.R. Brett. London: Academic Press.

Cerrato, R.M. 1990. Interpretable statistical tests for growth comparisons using parameters in
the von Bertalanffy equation. Can. J. Fish. Aquatic Sci. 47: 1416-1426.

Chang, W.Y.B. 1982. A statistical method for evaluating reproducibility of age determination.
Can. J. Fish. Aquat. Sci. 39: 1208-1210.

Chilton, D.E., and R.J. Beamish. 1982. Age determination methods for fishes studied by the
Canadian Groundfish Program at the Pacific Biological Station. Can. Spec. Publ. Fish.
Aquat. Sci. 60. 102 pp.

Cummings, N.J. 1998. An analysis of the Gulf of Mexico greater amberjack, Seriola dumerili,
stock condition. Proc. Gulf and Carib. Fish Inst. 50: 206-227.
























s
I






'* /



*______________* ____-


Figure 3-4. Otolith from an older greater amberjack showing the long processes on the medial
edges of the sulcus, as well as on the dorsal and ventral edges of the otolith.









(ANOVA). If significant, the ANOVA for each age class was followed by Tukey's HSD (SAS,

Cary, NC) test to determine where the differences existed among the three fishing sectors.









difficulty of extracting whole otoliths from greater amberjack, and a need for improving aging

criteria.

In the first instance, few amberjack samples are collected by state and federal sampling

agencies when compared to other reef fishes in hook-and-line fisheries. One major reason for

this may be that the fishing regulations allow for only one fish per individual, and hence it may

not be cost-efficient to spend limited sampling time and dollars sampling only a few amberjack

per vessel in comparison to sampling other reef fishes. Sending port-samplers to collect data on

fishes landed in the various fisheries is expensive in both time and funds. Since researchers

require a robust sample size, this may also direct research effort toward species that are sampled

more frequently, such as snapper and grouper. These fish groups support important targeted

fisheries in the southeastern U.S., and intensive sampling is justified, but at the same time this

creates a shortage of data for other species that also must be managed effectively on the limited

funds available, especially when data are required over a period of time to be useful in stock

assessments (e.g., age of the catch).

The lack of relatively young, small fish in comparisons between headboats and

charterboats in this study was a result of the large (i.e. 28 inch FL prior to 2009) minimum size

regulation for GOM greater amberjack. Without the availability of smaller, and therefore

younger, amberjack, length-at-age comparisons were restricted to a truncated set of age classes

(2, 3, and 4 yr-old fish) with adequate sample size. To facilitate modeling a complete growth

curve over the full range of ages of amberjack, additional, onboard sampling of greater

amberjack slated for release in the fisheries, or fishery-independent sampling, would be

necessary (see Thompson et al. 1999; Murie and Parkyn 2008). The availability of younger fish

would also presumably improve predictability of regression equations for otolith radius as a









Since greater amberjack have a median spawning date of April 1 (Harris 2004), these

measurements were considered to be the minimum radius of a sectioned otolith from a fish -3-4

months old (April to July based on previous research by Wells and Rooker, 2004). This

measurement was then used as a reference distance on otolith sections for fish thought to be over

1 yr old and having one visible annulus (based on growth curve by Thompson et al. 1999). The

first opaque zone past this reference distance was considered to represent the end of the first year

of growth

2.3 Validation of Aging Method: Periodicity and Timing of Annulus Formation

Timing and periodicity of increment formation was indirectly validated by determining the

month that the edge code, and therefore the ultimate increment of growth, was at a minimum

(Harris et al. 2007) rather than a marginal increment ratio as used in Dutka-Gianelli and Murie

(2001). Edge code values for all samples were plotted by month over 12 months. The number of

minima present over the 12-month period indicates the number of increments deposited each

year (i.e., one, two, or multiple). If only one minimum is present, opaque zone formation occurs

annually (thus an annulus) with formation considered complete by the end of the month wherein

the minimum is detected.

2.4 Age and Growth

Sample sizes of aged amberjack otoliths for each fishery were not adequate enough to

model the data using a von Bertalanffy growth model (Ricker 1975); data were lacking for fish

below the minimum size regulation, as well as for very large individuals. Therefore, observed

mean length-at-age was determined for age classes 2, 3, and 4 for greater amberjack caught in

Florida headboat, Florida charterboat, and Alabama charterboat fisheries. Mean length-at-age

for each age class was compared among fishery sectors using one-way analysis of variance













a








b












Figure 3-3. Cross-sections of otoliths from greater amberjack: a) 3-year old fish, with the core
(C) clearly demarked; and b) 5-yr old fish.






















Alabama Charterboat
o Florida Charterboat
o Florida Headboat




il .


0 1 2 3 4 5
Age (years)


6 7 8


Figure 3-8. Age frequency of greater amberjack from the Gulf of Mexico by location and fishery
(n=468).


90%
80%
70%
60%
50%
40%
30%
20%
10%
0%


il



















800


600


400


1 1


I
A


o Florida CharterAll
o Florida Headboat All
A Alabama Charter All


200


01 2 3 4 5 6 7 8
0 1 2 3 4 5 6 7 8


1200


1000


800


600


400


--- Florida Charter Mean
n=164
-- Florida Headboat Mean
n=116
-A-Alabama Charter Mean
n=172


200


0
0 1 2 3 4 5 6 7 8
Age (years)

Figure 3-10. a) Length-at-age for greater amberjack caught in all categories (n=468); b) Mean
length-at-age for greater amberjack from age classes 2, 3, and 4.


1200


1000


3









variability of source populations of greater amberjack or whether different populations exist.

Gene flow across the northern GOM (between the Florida Middle Grounds and Port Aransas,

Texas) is thought to be continuous, but some evidence suggests a division between populations

of greater amberjack in the U.S. South Atlantic Ocean and the Gulf of Mexico (Gold and

Richardson 1998, SEDAR 2006). More information is needed to determine the degree of genetic

variation within the Gulf of Mexico, which is presently being addressed through a NOAA

Cooperative Research Program grant (D.J. Murie, D.C Parkyn, and J.D. Austin; pers. comm.).

Fish movement could influence genetic variability as well as the size distribution of fish

available to specific fisheries. For example, gag Mycteroperca microlepis are largest in those

portions of the fishery that exploit deeper waters farther from shore in the Gulf of Mexico

(Fitzhugh et al. 2003). In this case differences in growth rate relate more to the geographical

distribution of the specific type of fishing effort more than differences in the gear selectivity in

the specific fishery (Fitzhugh et al. 2003). Thompson et al. (1999) reported seasonal changes in

the size distributions of amberjack caught aboard charterboats in Louisiana, with larger fish

being caught from May to September. They further speculate that amberjack may move to

warmer water to avoid cooler water in winter, altering the size distribution (Thompson et al.

1999). These seasonal movements could explain the variation in the size distribution of the catch

between Florida and Alabama charterboats.

Water temperature also affects fish growth rates as well as scope for activities, such as

swimming and foraging (Brett 1979, Kline 2004). Average water temperature is higher in the

southern GOM (pers. comm., National Oceanographic Data Center web page

www.nodc.noaa.gov) and therefore amberjack in the southern GOM may grow faster than those

in the northern GOM, if sufficient food resources are available to support their potential for









otoliths in wild fish for validation is advantageous but can be problematic in open systems due to

difficulty in recapturing adequate numbers of the marked fish and regulatory prohibitions

(VanderKooy and Guindon-Tisdel 2003).

Once the aging method has been determined to be accurate and relatively precise, then

the estimated ages of the fish can be incorporated into a growth analysis, which commonly

involves comparing growth curves based on the length-at-age over the age range of the fish.

Information on fish length-at-age can be obtained through back-calculation of size at the

formation of the last annulus. Back-calculation assumes a relationship between the somatic

growth of the fish and the growth of the hard part being measured. Measurements offish length

are regressed against measurements of radii of their otoliths, and from this relationship the length

of each fish at the formation of its last annulus can be predicted by back-calculation (Francis

1990). These data can then be incorporated into the appropriate growth model. For greater

amberjack, Manooch and Potts (1997), Thompson (1999), and Burch (1979) determined that the

von Bertalanffy growth model was appropriate to describe growth. Growth models can then be

compared to determine if growth rates are similar between sexes, geographic regions, or sectors

of the fishery (Murphy and Taylor 1994; Dutka-Gianelli and Murie 2001; Murie and Parkyn

2005).

One stock assessment completed in 2000, stated that preliminary results indicated that the

length composition data for greater amberjack may not have been sufficient to accurately

estimate the degree of variability in their length at age, and that the variability in their growth

with age therefore needs to be better characterized by fishery and region (Cummings et al. 2000).

This variability continues to be problematic (SEDAR 2006). The nature of fishing activities

between fisheries that target amberjack can vary greatly. Headboats are typically larger and



































2009 Edward E. Leonard









CHAPTER 3
RESULTS

3.1 Greater Amberjack Collections

In total, 505 greater amberjack were sampled from Florida charterboat and headboat

fisheries, and the Alabama charterboat fishery, during 2000-2007 (Table 3-1). Florida

charterboat samples consisted of 201 fish in total (102 females, 46 males and 53 of unknown

sex) whereas 129 amberjack were sampled from Florida headboats (80 females, 39 males, and 10

of unknown sex). The Alabama charterboat samples consisted of 114 females, 53 males, and 8

of unknown sex for a total of 175 fish.

In general, greater amberjack caught by Florida headboats were larger than those caught by

Florida charterboats. Headboat caught fish ranged in size from 287 mm to 1245 mm, with a

mean of 880 mm (Fig. 3-1a), while charterboat caught fish ranged in size from 535 mm to 1278

mm, with a mean size of 835 mm (Fig. 3-1b). Comparative length frequency distributions of

amberjack from these two fisheries were significantly different (Kolmogorov-Smirnov D: d-max

= 0.3383, P=0.01), reflecting the larger size distribution of fish caught by headboats (Fig. 3-2a).

Fish caught in charterboats off Florida were, however, larger than fish caught in charterboats off

Alabama. Alabama charterboat fish ranged in size from 650 mm to 1190 mm with a mean of

782 mm (Fig. 3-1c). Comparative length frequency distributions of amberjack caught in the two

charterboat fisheries were significantly different (Kolmogorov-Smirnov D: d-max=0.3029,

P=0.01), further indicative of smaller Alabama fish (Fig. 3-2b).

3.2 Aging

3.2.1 Otolith Measurements and Processing

Of the 505 greater amberjack collected, 468 sagittae were available for aging and a subset

of these were intact for otolith measurements (Table 3-2) due to the fragility of amberjack


















1000 -
---All Recreational


S800



E 600



j 400



200



0
1996 1998 2000 2002 2004 2006 2008
Year





Figure 1-2. Annual landings of greater amberjack in recreational fisheries of Alabama and the
west coast of Florida (NOAA Fisheries pers. comm.).









3.4 Age and Growth

Greater amberjack caught in the charterboat fisheries from the west coast of Florida ranged

in age from 2-8 years, with the majority offish being 3 years old (Figure 3-8). The majority of

amberjack caught in the Alabama charterboat fishery were also 3 years old and ranged between 2

and 5 years of age. Greater amberjack caught in the headboat fishery from the same area on the

west coast of Florida also had a mean age of 3 years, but included fish from 0 to 6 yrs of age.

Otolith radius as a function of fish FL was significant overall for amberjack from

charterboats in Florida and Alabama and headboats in Florida (Table 3-3). The regressions all

had low r2 values (all <28%), however, indicative that otolith radius was a poor predictor of fish

FL (Fig 3-9). Based on the lack of a predictive relationship between otolith radius and fish

length, fish lengths could not be back-calculated to length-at-age. Therefore, comparisons of fish

size with age were restricted to observed length-at-age.

Mean observed length-at-age was significantly different among fisheries for amberjack in

age classes 2, 3, and 4 (Table 3-4) (P<0.01). Greater amberjack landed in the Florida headboat

fishery were smaller than either the Florida charterboat or Alabama charterboat fisheries at age 2,

but larger at ages 3 and 4 (Tukey's HSD: all P<0.05) (Table 3-4; Fig. 3-10). Amberjack landed

in the Florida charterboat fishery were not significantly different in size-at-age compared to fish

landed by Alabama charterboats (Table 3-4; Fig. 3-10) (all P>0.05), except for 3-year old

amberjack (all P<0.05).









































Figure 3-5. a) Otolith section stained with Rapid Bone Stain viewed under reflected light; and b)
otolith section from the same greater amberjack that was not stained as viewed with
transmitted light.















3000


S2500
0
5
2000

E
u, 1500
=
= 1000
-J


500


0
1970


3500


1980


1990
Year


2000


I Commerci
I o Recreatior

II
19


I II

I o
? N" R
* i '1 i


Annual landings of greater amberjack in the Gulf of Mexico (NOAA Fisheries
pers. comm.; Cummings and McClellan 2000).


al
lal
















2010


Figure 1-1.









increased growth. If some amberjack move southward during cooler months and fish remaining

in northern areas grow slower due to temperature, the interaction of these factors could produce

some of the disparity in growth rates between the sampled areas of Alabama and Florida. Even

if this is the case, however, it does not explain the higher growth rate of Florida headboat fish

when compared to Florida charterboat fish, since all of these fish are caught off the west coast of

Florida. Most likely this can be attributed to differences in gear and/or techniques between

charterboats and headboats.

Differences in fishing mortality and gear selectivity have also been shown to influence

observed growth parameters in fish populations (Ricker 1975). In many hook-and-line fisheries,

larger individuals are targeted, which can also be the faster growing individuals in a cohort.

Therefore, slower-growing individuals in a cohort are left to reproduce in the population. Over

decades, this selective pressure can lead to slower individual growth rates in a given population

(e.g., vermillion snapper Rhomboplites aurorubens; Zhao et al. 1997). When sampling

populations with methods that select for faster growing individuals, the average size at age may

appear higher when compared to methods that are less selective for higher growth rate. The

truncated age distribution apparent in this study indicates that only a very small range of fish

ages are being captured in these fisheries. The low number of age two and younger fish is likely

the result of the majority of greater amberjack not having recruited to the fishery at that age. The

truncation of the age distribution past age four could be the result of many possible factors,

including the type of fishing pressure described above, influence from other fishing sectors such

as commercial vessels, and fish movement patterns.

Perhaps the most anomalous finding of this project is in the larger size and larger size-at-

age of Florida headboat fish when compared to Florida charterboat fish. It is commonly believed














16000


14000 -- Florida

-A-- Total, Gulf of Mexico
12000


10000
0

E
5 8000


6000
J

4000


2000


0
1992 1994 1996 1998 2000 2002 2004
Year



Figure 1-4. Annual landings of greater amberjack in Florida's headboat fishery and the total from
the Gulf of Mexico headboat fishery (SEDAR 2006).









Table 3-4. Mean length-at-age, standard error of the mean (SE), and sample size (n), for greater
amberjack from sampled fisheries and regions in the Gulf of Mexico.
Age Florida Headboat Florida Charterboat Alabama Charterboat
Class Mean SE n Mean SE n Mean SE n
0 287 0 1
1 566 0 1
2 657.1 14.2 7 740.8 14.2 30 742.2 10.8 16
3 894.3 7.6 101 809.5 6.6 102 768.1 4,1 134
4 1,014.8 21.5 8 901.1 16.8 32 854.8 22.2 22
5 939 17 2 913.5 34.5 2 1,089 101 3
6 1,166 0 1 1,023.5 91.5 2
7 970.0 84 2
8 1,147 74 2
Total 121 172 175


Table 3-5. Comparisons of mean length-at-age (ages 2, 3, and 4 yr old) for greater amberjack
from Florida Headboat (FLHB), Florida Charterboat (FLCH), and Alabama
Charterboat (ALCH).
Source Comparison Difference Between Mean Tukey's HSD


FLHB-FLCH
FLHB-ALCH
FLCH-ALCH


FLHB-FLCH
FLHB-ALCH
FLCH-ALCH

FLHB-FLCH
FLHB-ALCH
FLCH-ALCH
'NS(P>.0.5), *(P


Length-at-age (mm)
Age Two
-83.7
-85.0
-1.4
Age Three
84.8
126.2
41.4
Age Four
113.7
159.9
46.3


Significance'


0.05)









captured by headboats in Florida were larger at ages 3 and 4 years than those caught by Florida

charterboats. Differences in size and size-at-age may be related to the distribution of fish or the

distribution of specific fishing effort.






















S --0.5mm











0.5 mm
b______ 0____"____


Figure 3-6. Otolith section from a) a young-of-the-year greater amberjack and b) a 1 year old fish
showing growth past the first annulus.













M 100
-- Florida Headboat
: n=129
3 80 --c- Florida Charterboat
( n=201 -

60


I 40


20



0 500 1000 1500




100 1
b) 10 ---Alabama Charterboat
n=175
80 -0- Florida Charterboat
.> n=201

60 -


w 40
Eu-
o 20


0
0 500 1000 1500
Fork Length (mm)



Figure 3-2. Comparative length frequency distributions for greater amberjack from the Gulf of
Mexico caught in: a) charterboat and headboat fisheries off the west coast of Florida;
and b) charterboats from Florida and Alabama.









that charterboats land larger fish, in general, than headboats. The larger size and faster growth of

headboat fish in this study may be explained by the nature of specific headboat trips. Most

headboats run half-day and full-day trips, but some run overnight trips as well. The differences

in trip duration relate more to targeted fishing area than to the amount of time that lines are in the

water. On full-day and overnight trips, boat captains target areas farther offshore than on half-

day trips. The difference in time on the boat is mostly spent in transit to fishing sites. This

targeting of areas farther offshore is more similar to charterboat fishing than it is to half-day

headboat fishing. Often, headboat captains will make a special effort to visit an amberjack

aggregation or "AJ hole". At these sites, most customers on the vessel (sometimes as many as

50) will catch at least one amberjack. By making these visits to known amberjack aggregation

sites, such as wrecks, headboats may indeed target populations of offshore amberjack that grow

faster than inshore amberjack. Charterboats may be less likely to visit such a site regularly,

especially if it is far offshore, because a charterboat usually carries a maximum of six people. It

may be less cost-efficient for a charterboat captain to visit an "AJ hole" because only six fish can

be kept (i.e., regulations of one amberjack per person). Customers and charterboat captains

might consider their time and fuel better used by targeting grouper-fishing sites (i.e., regulations

of five grouper per person), and only catch amberjack incidentally.

The relatively low sample size encountered in this project was indicative of the overall lack

of availability of greater amberjack in the sampling programs for fisheries in the GOM during

2000-2007, especially very young and very old fish. There are several reasons for this,

including: an overall lack of port sampling effort because greater amberjack may not be a

"priority" species for most agencies, a large minimum size regulation, an underlying lack of

older individuals in the population, a reduction in the likelihood of landing larger individuals,









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

COMPARATIVE AGE AND GROWTH OF GREATER AMBERJACK (SERIOLA
DUMERILI) FROM CHARTERBOAT AND HEADBOAT FISHERIES
OF WEST FLORIDA AND ALABAMA, GULF OF MEXICO

By

Edward E. Leonard

August 2009

Chair: Debra Murie
Co-Chair: Daryl Parkyn
Major: Fisheries and Aquatic Sciences

Recent stock assessments of greater amberjack Seriola dumerili in the Gulf of Mexico

have had to use limited information regarding size-at-age, in part, due to the high degree of

variability intrinsic to growth of greater amberjack. To identify possibly sources of this

variability, size-at-age of greater amberjack was compared among fish landed in the charterboat

and headboat fisheries of Florida, and the charterboat fishery of Alabama. Identification of

sources of variability could lead to more accurate estimation of size-at-age by allowing stock

assessment scientists to model within more similar source populations.

Fish were collected from charterboats and headboats from the Gulf coasts of Florida and

Alabama through collaboration with state and federal sampling programs, supplemented by

scientific research sampling. Fish were aged using cross-sections of sagittal otoliths. Observed

age was correlated with length data and compared between charterboats and headboats within

Florida, and between charterboat catches from Florida and Alabama. Mean length-at-age was

compared among fishery sectors for age classes 2, 3 and 4. Greater amberjack captured by

charterboats in Florida were larger than those captured by charterboats in Alabama. Amberjack









Manooch, C.S. 1984. Fisherman's Guide to the Fishes of the Southeastern United States. North
Carolina State Museum of Natural History. 362 p.

Manooch, C.S. and J.C. Potts. 1997. Age, growth, and mortality of greater amberjack, Seriola
dumerili, from the U.S. Gulf of Mexico headboat fishery. Bull. Mar. Sci. 61: 671-683.

Murie, D.J., and D.C. Parkyn. 2005. Age and growth of white grunt (Haemulonplumieri): a
comparison of two populations along the Florida west coast. Bulletin of Marine Science
76(1): 73-93.

Murie, D.J., and D.C. Parkyn. 2008. Age, growth and sexual maturity of greater amberjack
(Seriola dumerili) in the Gulf of Mexico. MARFIN Final Report (NA05NMF4331071).
34 p.

Murphy, B.R. and D. W. Willis. 1996. Fisheries Techniques. Second edition. American
Fisheries Society, Bethesda. 732 pp.

Murphy, M.D. and R.G. Taylor. 1994. Age, growth and mortality of spotted seatrout in Florida
waters. Trans. Am. Fish Soc. 123: 482-497.

NMFS (National Marine Fisheries Service). 2008. NOAA Fisheries Service announces the
publication of a new rule to end overfishing and rebuild greater amberjack and gray
triggerfish stocks. Southeast Fishery Bulletin FB08-040.

Ricker, W.E. 1975. Computation and Interperetation of Biological Statistics of Fish Populations.
Bulletin of the Fisheries Research Board of Cananda. Bulletin 191.

Schirripa, M.J. and K.M. Burns. 1997. Growth estimates for three species of reef fish in the
eastern Gulf of Mexico. Bull. Mar. Sci. 61 (3): 581-591.

Sokal, R. R., and F. J. Rolf. 1969. Biometry. W. H. Freeman and Company San Francisco, CA
776 pp.

SEDAR (Southeast Data, Assessment and Review). 2006. SEDAR9 Assessment Report 2.
Charleston, SC

Sikstrom, C.B. 1983. Otolith, pectoral fin ray, and scale age determination for arctic grayling.
Prog. Fish. Cult. 45(4): 220-223.

Tanaka, K., Y. Mugiya, and J. Yamada. 1981. Effects ofphotoperiod and feeding on the daily
growth patters in otoliths of juvenile Tilapia nilotica. Fish. Bull., U.S. 79: 459-465.

Thompson, B.A., C.A. Wilson, J.H. Render, M. Beasley, and C. Cauthron. 1992. Age, growth
and reproductive biology of greater amberjack and cobia from Louisiana waters. Final
report to Marine Fisheries Research Initiative (MARFIN) Program, NMFS, St.
Petersburg, FL. NA90AA-H-MF722, 77 pp.









COMPARATIVE AGE AND GROWTH OF GREATER AMBERJACK (SERIOLA
DUMERILI) FROM CHARTERBOAT AND HEADBOAT FISHERIES
OF WEST FLORIDA AND ALABAMA, GULF OF MEXICO




















By

EDWARD E. LEONARD


A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF
FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2009











a) Florida Headboat
0.12 -i


0.10
o 0.08 -
o 0.06
0 0.04
0.02
0.00
500


* Female
SMale
o Unknown


In. I n n hn il l I I I


600


700 800 900 1000 1100


nui I


1200 1300


b) Florida Charterboat


U. Iz -

0.10

0.08 -

0.06

0.04 -

0.02
S n n


vIv .-. .


500 600 700 800


c) Alabama Charterboat
0.12
0.10
o 0.08 -
o 0.06
0.04
0.02
0.00
500 600 700 800 900 1000

Fork Length (mm)


1000 1100 1200 1300


SFemale
o Male
O] Unknown


1100 1200 1300


Figure 3-1. Length frequencies of greater amberjack from the Gulf of Mexico caught in: a)
headboats and b) charterboats off the west coast of Florida; and c) charterboats from
Alabama.


* Female
O Male
o Unknown






* n ,


n i I_









function of fish length, possibly permitting back-calculated length-at-age. This could be

important for characterizing growth in amberjack because of their rapid increase in length in

their first 3 years, with observed length at age (as in this study) possibly influenced by the time

of year that sampling occurred.

In addition, older individuals are poorly represented in the data sets of previous studies of

greater amberjack age and growth (Manooch and Potts 1997, Thompson et al. 1999, Harris 2007,

Murie and Parkyn 2008), and maximum ages are reported from 10 to 15 years depending on the

specific study. While all of these studies have various limitations, it seems likely that the lack of

older individuals represents a real paucity of older fish. Alternatively, when older, and larger,

fish are encountered, there is likely a reduction in successful landing of the fish due to the brute

strength and endurance of this species. Greater amberjack are sometimes colloquially referred to

as "reef donkeys" for their stubborn resistance to angler success. Lastly, when older individuals

were sampled and aged for the present study, they were more likely to be excluded from the final

data analysis due to a lack of a resolved age classification. With further improvement in aging

criteria, these individuals were able to be included in fishery-specific growth models for greater

amberjack in the Gulf of Mexico (Murie and Parkyn 2008).

Reliability of otolith aging was comparable to previous studies on greater amberjack and

was not perceived as problematic for amberjack in ages 2 through 5, which covered the among-

fisheries comparisons in this study. Within-reader percent agreement for those otoliths was 89%

(APE of 1.9, CV of 2.7), indicating that otoliths were aged precisely. The age distribution of fish

in the samples from the charterboats and headboats should adequately reflect the relative

abundance of these age classes in the fisheries, as the samples were collected by state and federal

agencies using sampling designs meant to characterize the catch.









in the otoliths is not consistently reliable. Higher aging precision would indicate that the aging

criteria are sufficient to reliably assign an age.

The accuracy of the aging method must also be validated. Validation determines whether

one or more annuli are deposited each year throughout the life of a fish, and is therefore related

to whether the number of annuli enumerated using the aging criteria truly represents the actual

age of the fish. Validation is important in aging fish because the rates of deposition in otoliths

vary with growth of the fish, which can be affected by various factors (e.g., water temperature,

food availability). Layers of material added during periods of slower somatic growth are denser

than those layers formed in periods of faster growth, which results in alternating translucent and

opaque zones within the otolith. Two accepted methods for validation are marginal-increment

analysis and chemical marking of the otolith (Beamish 1981, Murie and Parkyn 2005).

Marginal-increment analysis requires the collection of fish at regular intervals, usually monthly,

for at least a period of 12 months. The margin or growth at the edge of the otolith is measured to

determine the amount of translucent material deposited after the ultimate opaque zone. If the

translucent marginal growth is at a minimum once in a 12-month period then deposition of

translucent and opaque zones occurs only once per year and are said to comprise an annulus. If

two translucent and two opaque zones are deposited in a year then they are referred to as

biannuli. In chemically marking an otolith in a fish, the fish is injected with a chemical, such as

oxytetracycline, calcein, or alizarin complexone, and the chemical is incorporated into the otolith

matrix to produce a visible mark in the otoliths under ultraviolet illumination. The fish is

captured, tagged, injected and released after size measurements are taken. After >1 year or more

at large the fish is recaptured and the otoliths extracted. The number of annuli deposited past the

visible mark should match the number of years the fish was at large. Chemical marking of









The most significant consequence of fishery-specific and region-specific differences in the

size of greater amberjack observed in this study, over the major age classes targeted in the Gulf

of Mexico, is that the use of a single age-length key or a single growth model to assign ages to

amberjack of a given length (SEDAR 2006) could be problematic in the stock assessment. Size-

at-age for fish caught by charterboats, in particular, appeared to be lower than the "average"

amberjack size-at-age used in the stock assessment based on Thompson et al. (1999) for

amberjack caught in Louisiana (Fig. 4-1). Ideally, growth rate differences of fish captured by

different fisheries or in different regions should be accounted for when applying ages to the catch

derived from the specific fisheries or regions. In the case of amberjack caught in charterboat

fisheries, whether from Florida or Alabama, assigning an age class by applying the "average"

growth model or age-length key would result in these fish being assigned to a younger age class

(e.g., 4 yr old amberjack from the Florida charterboat fishery were the same size as 3-yr old

amberjack based on the growth curve by Thompson et al. (1999) (Fig. 4-1). Although a 1-yr

difference in age class assignment may seem trivial, it is important to note that greater amberjack

fisheries in the Gulf of Mexico primarily harvest fish over four age classes (2, 3, 4 and 5)

(Cummings and McClellan 2000; this study), and a bias in determining the productivity in any

one of those age classes could therefore also potentially bias the stock assessment. In addition,

incorrectly assigning ages to the catch would make it difficult to correctly gauge the importance

of age class cohorts contributing to the fisheries. While it may not be feasible to assign ages of

amberjack based on fishery-specific and region-specific growth rates due to the paucity of data

addressing this issue, it should nevertheless be considered as a source of uncertainty in the

reliance on an age-structured stock assessment model for greater amberjack in the Gulf of

Mexico.









Within reader percent agreement for otolith age assignments was 89.1% for otoliths

assigned the same age, and 99.1% for ages in agreement by 1 year; 8% of otoliths were either

broken or contained no discernable annuli and were deemed unreadable. Otoliths with no

discernable annuli (unreadable) were not included in calculations of agreement and precision.

Indices of precision for aged otoliths returned an APE of 1.93% and a CV of 2.73%.

3.2.3 Determination of the First Annulus

Determination of first annulus was accomplished by examining the otoliths of two young-

of-year (YOY) amberjack captured in the northern Gulf of Mexico in July by a scientific trawl

survey (S. Nichols, pers. comm.). The sectioned otoliths of these two fish appeared to have

some opaque growth in the core area but had not yet completed their first annulus (Fig. 3-6a).

The growth pattern in these otoliths was then compared to that in otoliths of fish showing one

complete annulus and growth at the edge of the otolith (Fig. 3-6b). The radius of each YOY

otolith was 0.35 mm, whereas the first opaque zone completing an annulus in the 1-year old fish

had a radius of -0.6 mm.

Wells and Rooker's (2004) equation for YOY amberjack captured off Galveston, Texas, in

2000-2001 was: SL (mm) = 2.00-(age in days)-37.32. Using this equation, and solving for age

(days), then was: Age (days) = (SL + 37.32)/2.00. The two fish from the trawl were 162 mm and

174 mm SL, therefore these fish would be 3.3 and 3.5 months old. This would mean they were

spawned in April of 2005 which is considered the mean spawning date for amberjack in the Gulf

of Mexico (Harris et al. 2007).

3.3 Validation of Aging Method: Marginal-Increment Analysis

Edge code analysis resulted in a single minimum in average edge code over a 12-month

period. This single minimum indicated that opaque zone formation occurred once per year

between May and July (Figure 3-7).









4-1 Comparison of mean observed length-at-age for greater amberjack from the Gulf of
Mexico sampled in this study compared to the von Bertalanffy growth curve of
Thom pson et al. (1999).................................................................... .. .... ........ 49









rostrum height (RH; maximum dorsal to ventral distance of the rostrum). Otoliths were

measured using computerized digital calipers integrated into a stereomicroscope (MOTIC).

Not all measurements could be determined for all otoliths due to a large number of broken

otoliths. Otolith measurements were regressed against amberjack FL to determine if overall

otolith growth was correlated with fish length. These regression analyses, and all other statistical

analyses, were tested for significance at P< 0.05.

After measuring otoliths whole, the left otolith (right if left broken) was embedded in

Devcon Five-minute Epoxy using a 4 x 12 mm silicone bullet mold. Each mold was filled with

enough epoxy resin to cover the bottom and the otolith was then positioned in the resin, keeping

the otolith parallel to the long axis and perpendicular to the short axis of the mold. Once the first

layer of epoxy had cured, enough epoxy was added to the mold to just cover the otolith.

Embedded otoliths were removed from the mold and glued to frosted slides using cyanoacrylate

glue. Otoliths were then viewed under a stereomicroscope and the core of the otolith was

marked with a fine felt-tip marker. A Buehler Isomet 1000 digital sectioning saw was then

used to obtain two transverse sections of the embedded otolith taken through the core. To do

this, three diamond-edged watering blades (7.6 cm diameter x 0.15 mm width) (Norton

Company, Worcester, MA) were separated by two stainless steel shims (0.5 mm thick), with the

entire apparatus mounted together on the saw. This assembly allowed for removal of two 0.5

mm sections with a single pass through the core of the otolith. The sections were then rinsed in

de-ionized water, patted dry, and mounted flat on a microscope slide using Loctite Crystalbond

509 clear thermoplastic adhesive (Henkel, Rocky Hills, CT). Mounted sections were sanded by

hand using 600 grit wet sandpaper and polished with a Buehler polishing cloth and 0.3 micron










Table 3-1. Number of greater amberjack sampled from each fishery in the Gulf of Mexico
during 2000-2007.
Source 2000-2001 2002-2003 2004-2005 2006-2007
Florida Headboat 11 31 7 80
Florida Charterboat 5 145 51 0
Alabama Charterboat 0 148 27 0


Table 3-2. Relationship between greater amberjack forklength (FL, mm) and otolith total length
(OTL, mm), otolith antirostrum length (OAL, mm), and otolith height (OH, mm). All
regressions were significant at Pc0.05.
Regression r2 n
FL = 259.8 + 55.2*OTL 0.31 84
FL = 47.8 + 107.2*OAL 0.47 198
FL = 214.1 + 189.0*OH 0.21 298


Table 3-3. Relationship between greater amberjack forklength (FL, mm) and otolith radius
(OTR, mm) for Florida charterboat (FL-CB), Florida headboat (FL-HB), and
Alabama charterboat (AL-CB) fisheries.


Regression
=546.3+321.6*OTR
=488.4+406.7*OTR
=469.2+414.3*OTR
=656.7+288.9*OTR
=584.3+396.8*OTR
=609.9+354.2*OTR
=617.9+212.6*OTR
=430.0+456.0*OTR
=361.4+551.0*OTR


r
0.23
0.16
0.27
0.05
0.16
0.11
0.05
0.14
0.19


Significance
**
***
***
NS
***
***
NS
***
***


Area
FL
FL
FL
FL
FL
FL
AL
AL
AL
NS (I


Fishery
CB
CB
CB
HB
HB
HB
CB
CB
CB
>0.05),


*(P<0.05), **(P<0.01), ***(P 0.001)









for-hire vessels usually limited to 6 fishers), commercial, and several recreational sources, but

was limited to fish collected from Louisiana waters only. Manooch and Potts (1997) established

another growth curve for greater amberjack in the Gulf of Mexico derived from fish collected

only from headboats, with samples from Texas (53%), northwest Florida and Alabama (46%),

and Louisiana (1%). Recently, Murie and Parkyn (2008) modeled age, growth and sexual

maturity of greater amberjack from major fisheries in the Gulf of Mexico, primarily in Florida,

Alabama, and Louisiana. All of these previous studies noted that greater amberjack had a high

degree of variation in their size-at-age depending on the area and fishery sampled. This

variability, combined with inadequate samples of aged fish over past decades, resulted in the

recommendation that a production model be used in the stock assessment for greater amberjack

(SEDAR 2006).

Aging fish is important to understanding the dynamics of fish populations. Age is used to

estimate critical population parameters, such as mortality and growth, and in combination with

reproductive data can be used to estimate fecundity at age and age of maturity (Brennan and

Cailliet 1989; Schirripa and Burns 1997). Greater amberjack are typically aged using their

sagittal otoliths, which are very thin and fragile, making them relatively difficult to process. In

addition, both Manooch and Potts (1997) and Thompson et al. (1999) reported that the annuli of

amberjack otoliths are troublesome to identify and count. Therefore, a set of criteria must be

developed where the reader decides what features of the otolith to count or "read" as true annuli

in order to estimate ages precisely. Precision, or reproducibility, of the ages is estimated by

having the structures aged by a primary reader and then compared to the age estimates obtained

by a second, experienced (or trained) reader using the same aging criteria. Low aging precision

would indicate that the aging criteria are either poorly established, or that the deposition pattern









CHAPTER 1
INTRODUCTION

The greater amberjack (Seriola dumerili) is a widely distributed pelagic reef fish inhabiting

most of the world's oceans including the Mediterranean Sea, and the Atlantic, Pacific, and Indian

Oceans (Hoose and Moore 1998). Greater amberjack are associated with reef environments and

can be found in the western Atlantic from Nova Scotia to Brazil, including the Gulf of Mexico

and the Caribbean Sea (Manooch 1984). As it is the largest member of the Carangidae, it can

reach sizes in excess of 81.6 kg and can be caught at depths in excess of 91.4 m (Bohlke and

Chaplin 1993; Hoese and Moore 1998). Greater amberjack like other members of the genus

Seriola are thought to spawn in the spring and early summer, as their gonadosomatic index

reaches a maximum between April and June (Burch 1979; Beasely 1993; Gillanders et al. 1999).

The commercial importance of greater amberjack has changed dramatically in the past 40

years. The commercial catch peaked in 1988 at 1,043 metric tons (MT) but was around 500 MT

from 1993 to 2003 (Fig. 1-1), with the largest declines occurring off of Florida's west coast and

the coast of Louisiana (Cummings and McClellan 1998). Recreational targeting of amberjack

remains popular, despite some concerns about ciguatera poisoning and parasitic worm

infestation, owing to the strong fighting ability of the fish when hooked and the recent popularity

of smoked amberjack (Manooch 1997; Cummings and McClellan 2000). Recreational catches

exceed commercial catches in most years (Fig. 1-1). Recreational catch peaked in 1986 at 3,420

MT, but also rapidly declined, with landings dropping to 458 MT in 2000. A small peak in

recreational landings occurred in 2003 at 1,096 MT but have since declined to 449 MT in 2007

(NOAA Fisheries: www.nmfs.noaa.gov).

Federal management of Gulf of Mexico greater amberjack stocks originated with the

implementation of Amendment 1 to the Reef Fish Fishery Management Plan (RFFMP) in 1990.









CHAPTER 4
DISCUSSION

For greater amberjack in the Gulf of Mexico (GOM), fishery-specific size and age of the

catch were different among Florida headboat, Florida charterboat and Alabama charterboat

fisheries. Surprisingly, for greater amberjack landed in the Florida headboat fishery, the size

frequency distribution was skewed toward larger fish compared to fish landed in the Florida

charterboat fishery. In addition, the observed length-at-age for fish in age classes 3 and 4, which

represented 30% of the landed catch in 2004 (SEDAR 2006), was greater in the headboat fishery

compared to fish landed in the charterboat fishery of Florida. We also found that fish landed by

charterboats in Alabama were smaller than those landed by charterboats in Florida. Charterboat

caught fish from Florida and Alabama differed in size-at-age for age 3 fish only. This difference

was much less dramatic in scale than the charterboat/headboat difference (differences of 41 mm

and 84 mm respectively at age 3). These differences in size-at-age indicate that growth rates for

greater amberjack were not equal between all regions and among fishing sectors.

Fish are known to exhibit regionally distinct rates of growth (Murphy and Taylor 1994,

Dutka-Gianelli and Murie 2001, Murie and Parkyn 2005). Murphy and Taylor (1994) found that

spotted seatrout Cynoscion nebulosus exhibited different growth rates between two Florida

estuary systems. Dutka-Gianelli and Murie (2001) found qualitatively that regional differences

in growth rates may exist for sheepshead Archosargusprobatocephalus. White grunt Haemulon

plumieri have also been show to exhibit differences in growth rates between regions on the west

coast of Florida, with even more pronounced differences being observed between the Gulf of

Mexico and some Atlantic coast populations (Murie and Parkyn 2005). Regional differences in

growth rates could occur due to several reasons, for example, differences in genetics, habitat,

temperature, migratory patterns and year-class effects. Little is known about the genetic










Thompson, B.A., Beasley, and C.A. Wilson. 1999. Age distribution and growth of greater
amberjack, Seriola dumerili, from the north-central Gulf of Mexico. Fishery Bulletin 97:
362-371.

Turner, S.C. 2000. Catch rates of greater amberjack caught in the headboat fisheries in the Gulf
of Mexico in 1986-1998. NMFS/SEFSC, Miami Laboratory. Document SFD 99/00-107.
17 pp.

Turner, S.C., N.J. Cummings, and C.E. Porch. 2000. Stock assessment of Gulf of Mexico
greater amberjack using data through 1998. NMFS/SEFSC, Miami Laboratory.
Document SFD 99/00-100. 27 pp.

VanderKooy, S., and K. Guindon-Tisdel (Editors). 2003. A practical handbook for determining
the ages of Gulf of Mexico fishes. Gulf States Marine Fisheries Commission, Ocean
Springs, MS. Publication No. 111. 114p.

Victor, B.C., and E.B. 1982. Age and growth of the fallfish Semotilus corporalis with daily
otolith increments as a method of annulus verification. Can. J. Zool. 60: 2453-2550.

Wells, R. J. and J. R. Rooker. 2004. Distribution, Age, and Growth of Young-of-the-year Greater
Amberjack (Seriola dumerili) Associated with Pelagic Sargassum. Fish. Bull. 102:545-
554

Zhao, B. and J.C. McGovern. 1997. Temporal variation in sexual maturity and gear-specific sex
ratio of the vermilion snapper, Rhomboplites aurorubens, in the South Atlantic Bight.
Fishery Bulletin 95: 837-848.


















1300
1200 -
1100
E $
E 1000 o

8900

700 1

-L 600
500
400
0.4 0.6 0.8 1 1.2 1.4
Otolith Radius (mm)

Figure 3-9. Greater amberjack fork length as a function of otolith radius (n=455).









Cummings, N.J. and D.B. McClellan. 2000. Trends in the Gulf of Mexico greater amberjack
fishery through 1998: Commercial landings, recreational catches, observed length
frequencies, estimates of landed and discarded catch at age, and selectivity at age. U.S.
Dept of Commerce, National Oceanographic and Atmospheric Administration, National
Marine Fisheries Service, Sustainable Fisheries Division.

Dutka-Gianelli, J., and D.J. Murie. 2001. Age and growth of sheephead, Archosargus
probatocephalus (Pisces: Spaidae), from the northwest coast of Florida. Bull. Mar. Sci.
68: 69-83.

Fitzhugh, G.R., L.A. Lombardi-Carlson and N.M. Evou. 2003. Age structure of gag
(Mycteroperca microlepis) in the eastern Gulf of Mexico by year, fishing mode, and
region. Proceedings of the Gulf and Caribbean Fisheries Institue 54: 538-549.

Francis, R.I.C. 1990. Back-calculation of fish lengths: A critical review. J. Fish. Biol. 36: 883-
902.

Gillanders, B. M., D. J. Ferrell and N. L. Andrew. 1999. Size at maturity and seasonal changes in
gonad activity of yellowtail kingfish (Seriola lalandi; Carangidae) in New South Wales,
Australia. New Zealand Journal of Marine and Freshwater Research, 1999, 33: 457-468.

Gold, J. R. and L. R. Richardson. 1998. Mitochondrial DNA Diversification and Population
Structure in Fishes from the Gulf of Mexico and Western Atlantic. J. Heredity 89:404-
414

Harris, P. J, D.M. Wyanski, D.B. White, and P.P. Mikell. 2007. Age, Growth, and Reproduction
of Greater Amberjack off the Southeastern Atlantic Coast. Trans. Of the American
Fisheries Society 136: 1534-1545.

Harris, P. J, 2004. Age, Growth, and Reproduction of Greater Amberjack, Seriola dumerili, in
the southwestern North Atlantic. Analytical Report of MARMAP Program. South
Carolina Department of Natural Resources, Charleston, SC.

Hoese, H. Dickinson and R.H. Moore. 1998. Fishes of the Gulf of Mexico, Texas, Louisiana,
and adjacent waters. University of Texas Press, Austin. 422 pp.

Kimura, D.K. 1980. Likelihood methods for the von Bertalanffy growth curve. Fish. Bull. 77:
765-776.

Kimura, D.K., and J.J. Lyons. 1991. Between-reader bias and variability in the age determination
process. Fish. Bull., U.S. 89: 53-60.

Kline, R.J. 2004. Metabolic rate of gag grouper, (Mycteroperca microlepis) in relation to
swimming speed, body size and temperature. Unpublished M.S. thesis. Univeristy of
Florida. http://etd.fcla.edu/UF/UFE0008925/kliner.pdf














-- Alabama Charter
-- Florida Charter


400


350


n 300

U
u
S250
E

o 200


- 150


100-


50


0
1996


Annual landings of greater amberjack in the charterboat fisheries of Alabama and the
west coast of Florida (NOAA Fisheries pers. comm.).


450


1998 2000 2002 2004 2006
Year


Figure 1-3.




Full Text

PAGE 1

1 COMPARATIVE AGE AND GROWTH OF GREATER AMBERJACK ( SERIOLA DUMERILI ) FROM CHARTERBOAT AND HEADBOAT FISHERIES OF WEST FLORIDA AND ALABAMA, GULF OF MEXICO By EDWARD E. LEONARD A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Edward E. Leonard

PAGE 3

3 ACKNOWLEDGMENTS This thesis would not have been possible with the help and support of my advisor, Dr. Debra Murie. Dr. Murie has guided me through this process with patience and understanding for the circumstances in life that have made a quick finish impossible. I thoroughly enjoyed working in the Murie Lab and I am grate ful to have had the opportunity to collaborate with her. Special thanks also go to Dr. Daryl Parkyn for his hard work, intelligent insight and constant entertainment. I also thank the other members of my committee; Dr. Mike Allen and Dr. Gary Fitzhugh wh o provided valuable insight and advice throughout this process. I would also like to thank the captains and crew of Hubbards Marina in Madeira Beach, FL, for providing access to their catch as well as the members of the Gulf States Marine Fisheries Commis sion and NOAA Fisheries who also supplied samples. Many of the students and staff of the Murie Lab and Lindberg Lab also provided invaluable assistance and made the work a pleasure. These individuals include: Ivy Baremore, Liz Berens, Doug Colle, Jaclyn Debicella Leonard, Rick Kline, Doug Marcinek, and Pat ODay My family and friends provided continuing encouragement, support, and humor which has helped me complete this process. I am especially thankful to my parents, Pat and Jerry Leonard, for always s upporting and encouraging me. A great deal of gratitude goes to my brothers and sister, Jerry, Tony, and Tara for their love and support. I would also like to express thanks to my wife, Jaclyn Debicella Leonard, who makes me better that I am. Financial support for this project was provided by a NOAA Marine Fisheries Initiative (MARFIN) grant and the University of Florida, College of Agricultural and Life Sciences, Program of Fisheries and Aquatic Sciences.

PAGE 4

4 TABLE OF CONTENTS ACKNOWLEDGMENTS ........................................................................................................... 3 page LIST OF TABLES ...................................................................................................................... 5 LIST OF FIGURES .................................................................................................................... 6 ABSTRACT ............................................................................................................................... 8 CHAPTER 1 INTRODUCTION ............................................................................................................. 10 2 METHODS ........................................................................................................................ 20 2.1 Greater Amberjack Collections ................................................................................. 20 2.2 Aging ........................................................................................................................ 20 2.2.1 Otolith Measurements and Processing ....................................................... 20 2.2.2 Aging Criteria and Age Estimation ........................................................... 22 2.2.3 Determination of the First Annulus ........................................................... 23 2.3 Validation of Agi ng Method: Periodicity and Timing of Annulus Formation............. 24 2.4 Age and Growth ........................................................................................................ 24 3 RESULTS .......................................................................................................................... 26 3.1 Greater Amberjack Collections ................................................................................. 26 3.2 Aging ........................................................................................................................ 26 3.2.1 Otolith Measurements and Processing ....................................................... 26 3.2.2 Aging Criteria and Estimating Ages .......................................................... 27 3.2.3 Determination of the First Annulus ........................................................... 28 3.3 Validation of Aging Method: Marginal Increment Analysis ...................................... 28 3.4 Age and Growth ........................................................................................................ 29 4 DISCUSSION .................................................................................................................... 42 REFERENCES ......................................................................................................................... 50 BIOGRAPHICAL SKETCH ..................................................................................................... 54

PAGE 5

5 LIST OF TABLES Table page 31 Number of greater amberjack sampled from each fishery in the Gulf of Mexico during 20002007. ......................................................................................................... 30 32 Relationship between greater amberjack forklength ( FL, mm) and otolith total length (OTL, mm), otolith antirostrum length (OAL, mm), and otolith height (OH, mm). All regressions were significant at P .................................................................... 30 33 Relationship between gr eater amberjack forklength (FL, mm) and otolith radius (OTR, mm) for Florida charterboat (FL CB), Florida headboat (FL HB), and Alabama charterboat (AL CB) fisheries. ........................................................................ 30 34 Mean lengthat age standard error of the mean (SE), and sample size (n), for greater amberjack from sampled fisheries and regions in the Gulf of Mexico. ........................... 31 35 Comparisons of mean length at age (ages 2, 3, and 4 yr old) for greater amberjack from Florida Headboat (FLHB), Florida Charterboat (FLCH), and Alabama Charterboat (ALCH). ..................................................................................................... 31

PAGE 6

6 LIST OF FIGURES Figure page 11 Annual landings of greater amberjack in the Gulf of Mexico (NOAA Fisheries pers. comm.; Cummings and McClellan 2000). ...................................................................... 16 12 Annual landings of greater amberjack in recreational fis heries of Alabama and the west coast of Florida (NOAA Fisheries pers. comm.). ................................................... 17 13 Annual landings of greater amberjack in the charterboat fisheries of Alabama and the west coast of Florida (NOAA Fisheries pers. comm.). ................................................... 18 14 Annual landings of greater amberjack in Floridas headboat fishery and the total from the Gulf of Mexico headboat fishery (SEDAR 2006). ............................................ 19 31 Length frequencies of greater amberjack from the Gulf of Mexico caught in: a) headboats and b) charterboats off the west coast of Florida; and c) charterboats from Alabama. ....................................................................................................................... 32 32 Comparative length frequency distributions for greater amberjack from the Gulf of Mexico caught in: a) charterboat and headboat fisheries off the west coast of Florida; and b) charterboats from Florida and Alabama. .............................................................. 33 33 Crosssections of otoliths from greater amberjack: a) 3 year old fish, with the core (C) clearly demarked; and b) 5 yr old fish. ..................................................................... 34 34 Otolith from an older greater amberjack showing the long processes on the medial edges of the sulcus, as well as on the dorsal and ventral edges of the otolith. ................. 35 35 a) Otolith section stained with Rapid Bone Stain viewed under reflected light; and b) otolith section from the same greater amberjack that was not stained as viewed with transmitted light. ............................................................................................................ 36 36 Otolit h section from a) a young of the year greater amberjack and b) a 1 year old fish showing growth past the first annulus ..................................................................... 37 37 Mean edge code of greater amberjack otoliths (n=467) as a function of month of collection throughout the year. Vertical bars represent standard error of the mean. Minimum mean edge code in June indicated the formation of the opaque zone. ............. 38 38 Age frequency of greater amberjack from the Gulf of Mexico by location and fishery (n=468). ......................................................................................................................... 39 39 Greater amberjack fork length as a function of otolith radius (n=455). ........................... 40 310 a) Length at age for greater amberjack caught in all categories (n=468); b) Mean lengthat age for greater amberjack from age classes 2, 3, and 4. ................................... 41

PAGE 7

7 41 Comparison of mean observed lengthat age for greater amberjack from the Gulf of Mexico sampled in this study compared to the von Bertalanffy growth curve of Thompson et al. (1999). ................................................................................................. 49

PAGE 8

8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science COMPARATIVE AGE AND GROWTH OF GREATER AMBERJACK ( SERIOLA DUMERILI ) FROM CHARTERBOAT AND HEADBOAT FISHERIES OF WES T FLORIDA AND ALABAMA, GULF OF MEXICO By Edward E. Leonard August 2009 Chair: Debra Murie Co Chair: Daryl Parkyn Major: Fisheries and Aquatic Sciences Recent stock assessments of greater amberjack Seriola dumerili in the Gulf of Mexico have had to use limited in formation regarding sizeat age, in part, due to the high degree of variability intrinsic to growth of greater amberjack. To identify possibly sources of this variability, size at age of greater amberjack was compared among fish landed in the charterboat and headboat fisheries of Florida, and the charterboat fishery of Alabama. Identification of sources of variability could lead to more accurate estimation of sizeat age by allowing stock assessment scientists to model within more similar source populatio ns. Fish were collected from charterboats and headboats from the Gulf coasts of Florida and Alabama through collaboration with state and federal sampling programs, supplemented by scientific research sampling. Fish were aged using cross sections of sagit tal otoliths. Observed age was correlated with length data and compared between charterboats and headboats within Florida, and between charterboat catches from Florida and Alabama. Mean length at age was compared among fishery sectors for age classes 2, 3 and 4. Greater amberjack captured by charterboats in Florida were larger than those captured by charterboats in Alabama. Amberjack

PAGE 9

9 captured by headboats in Florida were larger at ages 3 and 4 years than those caught by Florida charterboats. Difference s in size and size at age may be related to the distribution of fish or the distribution of specific fishing effort.

PAGE 10

10 CHAPTER 1 INTRODUCTION The greater amberjack ( Seriola dumerili ) is a widely distributed pelagic reef fish inhabiting most of the worlds oceans including the Mediterranean Sea, and the Atlantic, Pacific, and Indian Oceans (Hoose and Moore 1998). Greater amberjack are associated with reef environments and can be found in the western Atlantic from Nova Scotia to Brazil, including the Gulf of Mexico and the Caribbean Sea (Manooch 1984). As it is the largest member of the Carangidae, it can reach sizes in excess of 81.6 kg and can be caught at depths in excess of 91.4 m (Bhlke and Chaplin 1993; Hoese and Moore 1998). Greater amberjack like oth er members of the genus Seriola are thought to spawn in the spring and early summer, as their gonadosomatic index reaches a maximum between April and June (Burch 1979; Beasely 1993; Gillanders et al. 1999). The commercial importance of greater amberjack ha s changed dramatically in the past 40 years. The commercial catch peaked in 1988 at 1,043 metric tons (MT) but was around 500 MT from 1993 to 2003 (Fig. 11), with the largest declines occurring off of Floridas west coast and the coast of Louisiana (Cumm ings and McClellan 1998). Recreational targeting of amberjack remains popular, despite some concerns about ciguatera poisoning and parasitic worm infestation, owing to the strong fighting ability of the fish when hooked and the recent popularity of smoked amberjack (Manooch 1997; Cummings and McClellan 2000). Recreational catches exceed commercial catches in most years (Fig. 1 1). Recreational catch peaked in 1986 at 3,420 MT, but also rapidly declined, with landings dropping to 458 MT in 2000. A small peak in recreational landings occurred in 2003 at 1,096 MT but have since declined to 449 MT in 2007 (NOAA Fisheries: www.nmfs.noaa.gov ). Federal management of Gulf of Mexico greater amberjack stocks originated wit h the implementation of Amendment 1 to the Reef Fish Fishery Management Plan (RFFMP) in 1990.

PAGE 11

11 Amendment 1 added greater amberjack to the list of species already being managed under that plan. This amendment also established a 3 fish per person bag limit and a 28 inch fork length minimum size limit for recreational fishers. Amendment 1 also established a 36 inch fork length minimum size limit for the commercial fishery and implemented a requirement for a commercial reef fish permit for all fish included in the RFFMP. Since that time, the most significant regulatory changes to the fishery can be found in Amendments 12 and 15. Amendment 12 (December 1995) reduced the recreational bag limit for greater amberjack to 1 per person. Amendment 15 (January 1998) closed the Gulf of Mexico greater amberjack commercial fishery from March 1 through May 31. Subsequent amendments established harvest limits that were expected to rebuild the stock of greater amberjack (SEDAR 9 Assessment Report 2, 2006) Gulf of Mexico greater amberjack were declared overfished by NMFS on February 9, 2001, based on the results of a stock assessment done by Turner et al. (2000). The Gulf of Mexico stock was assessed as being overfished again in the most recent stock assessment review in 2006 (SEDAR 2006). Most recently in 2009, NOAA Fisheries Service published a new rule to limit commercial harvest to 228 MT and limit recreational harvest to 621 MT. The rule also raises the minimum size limit in the recreational fishery to a 30 inch fork length and establishes new methods for adjusting annual catch limits in season (NMFS 2008). Assessments of stock condition depend greatly on accuracy of individual age and growth information (Schirripa and Burns 1997; Cummings 1998; Turner et al. 2000). Burch (1979) provided the earliest, comprehensive study of greater amberjack in southern Florida. In the Gulf of Mexico, Beasley (1993), and later Thompson et al. (1999, which included the data from Beasleys study), modeled age and growth of amberjack caught in several different fisheries, including headboats (large, for hire vessels carrying as many as 50 fishers), charterboats (smaller

PAGE 12

12 for hire vessels usually limited to 6 fishers), commercial, and several recreational sources, but was limited to fi sh collected from Louisiana waters only. Manooch and Potts (1997) established another growth curve for greater amberjack in the Gulf of Mexico derived from fish collected only from headboats, with samples from Texas (53%), northwest Florida and Alabama (4 6%), and Louisiana (1%). Recently, Murie and Parkyn (2008) modeled age, growth and sexual maturity of greater amberjack from major fisheries in the Gulf of Mexico, primarily in Florida, Alabama, and Louisiana. All of these previous studies noted that grea ter amberjack had a high degree of variation in their sizeat age depending on the area and fishery sampled. This variability, combined with inadequate samples of aged fish over past decades, resulted in the recommendation that a production model be used in the stock assessment for greater amberjack (SEDAR 2006). Aging fish is important to understanding the dynamics of fish populations. Age is used to estimate critical population parameters, such as mortality and growth, and in combination with reproduct ive data can be used to estimate fecundity at age and age of maturity (Brennan and Cailliet 1989; Schirripa and Burns 1997). Greater amberjack are typically aged using their sagittal otoliths, which are very thin and fragile, making them relatively diffic ult to process. In addition, both Manooch and Potts (1997) and Thompson et al. (1999) reported that the annuli of amberjack otoliths are troublesome to identify and count. Therefore, a set of criteria must be developed where the reader decides what feat ures of the otolith to count or read as true annuli in order to estimate ages precisely. Precision, or reproducibility, of the ages is estimated by having the structures aged by a primary reader and then compared to the age estimates obtained by a secon d, experienced (or trained) reader using the same aging criteria. Low aging precision would indicate that the aging criteria are either poorly established, or that the deposition pattern

PAGE 13

13 in the otoliths is not consistently reliable. Higher aging precisio n would indicate that the aging criteria are sufficient to reliably assign an age. The accuracy of the aging method must also be validated. Validation determines whether one or more annuli are deposited each year throughout the life of a fish, and is t herefore related to whether the number of annuli enumerated using the aging criteria truly represents the actual age of the fish. Validation is important in aging fish because the rates of deposition in otoliths vary with growth of the fish, which can be affected by various factors (e.g., water temperature, food availability). Layers of material added during periods of slower somatic growth are denser than those layers formed in periods of faster growth, which results in alternating translucent and opaque zones within the otolith. Two accepted methods for validation are marginal increment analysis and chemical marking of the otolith (Beamish 1981, Murie and Parkyn 2005). Marginal increment analysis requires the collection of fish at regular intervals, us ually monthly, for at least a period of 12 months. The margin or growth at the edge of the otolith is measured to determine the amount of translucent material deposited after the ultimate opaque zone. If the translucent marginal growth is at a minimum on ce in a 12 month period then deposition of translucent and opaque zones occurs only once per year and are said to comprise an annulus. If two translucent and two opaque zones are deposited in a year then they are referred to as biannuli. In chemically ma rking an otolith in a fish, the fish is injected with a chemical, such as oxytetracycline, calcein, or alizarin complexone, and the chemical is incorporated into the otolith matrix to produce a visible mark in the otoliths under ultraviolet illumination. The fish is captured, tagged, injected and released after size measurements are taken. After >1 year or more at large the fish is recaptured and the otoliths extracted. The number of annuli deposited past the visible mark should match the number of years the fish was at large. Chemical marking of

PAGE 14

14 otoliths in wild fish for validation is advantageous but can be problematic in open systems due to difficulty in recapturing adequate numbers of the marked fish and regulatory prohibitions (VanderKooy and Guindo n Tisdel 2003). Once the aging method has been determined to be accurate and relatively precise, then the estimated ages of the fish can be incorporated into a growth analysis, which commonly involves comparing growth curves based on the length at age over the age range of the fish. Information on fish lengthat age can be obtained through back calculation of size at the formation of the last annulus. Back calculation assumes a relationship between the somatic growth of the fish and the growth of the hard part being measured. Measurements of fish length are regressed against measurements of radii of their otoliths, and from this relationship the length of each fish at the formation of its last annulus can be predicted by back calculation (Francis 1990). These data can then be incorporated into the appropriate growth model. For greater amberjack, Manooch and Potts (1997), Thompson (1999), and Burch (1979) determined that the von Bertalanffy growth model was appropriate to describe growth. Growth models can then be compared to determine if growth rates are similar between sexes, geographic regions, or sectors of the fishery (Murphy and Taylor 1994; Dutka Gianelli and Murie 2001; Murie and Parkyn 2005). One stock assessment completed in 2000, stated t hat preliminary results indicated that the length composition data for greater amberjack may not have been sufficient to accurately estimate the degree of variability in their length at age, and that the variability in their growth with age therefore needs to be better characterized by fishery and region (Cummings et al. 2000). This variability continues to be problematic (SEDAR 2006). The nature of fishing activities between fisheries that target amberjack can vary greatly. Headboats are typically lar ger and

PAGE 15

15 slower than charterboats and therefore may fish closer inshore on most days. Charterboats, in turn, may fish closer to shore than commercial boats, since the latter can stay at sea for several days prior to landing their catch. It is possible tha t these groups might fish in different locations, use different techniques, and/or different gear, and these factors may select for fish of different size and age, as well as different growth rates. Differential growth may occur due to differences in habit at between locations. Differences in the presence or absence of benthic structure could influence behavior and feeding success in greater amberjack. Additionally, Alabama is situated closer to the deep water areas near the coasts of Mississippi and Louis iana. This deep water area could provide resources not found in close proximity to Gulf of Mexico coastal Florida. The purpose of this study was to compare the age and growth for greater amberjack caught in headboat and charterboat fisheries on the west c oast of Florida and the charterboat fishery in coastal Alabama. These charterboat and headboat landings represent a significant portion of the total recreational landings of greater amberjack in the Gulf of Mexico (Figs. 1 2, 1 3, and14). While the overall goal of my research was to compare the age and growth of greater amberjack caught in the charterboat and headboat fisheries off the west coast of Florida and the charterboat fishery off Alabama, the specific objectives included: 1) to collaborate with private fishers, state and federal fisheries agencies to collect and process otoliths to age greater amberjack in the Gulf of Mexico, stratified by state (Florida west coast and Alabama) and by fishery (charterboats and headboats); 2) establish aging crit eria for greater amberjack in the Gulf of Mexico based on sectioned otoliths, including validating the method using marginal increment analysis; and 3) model and compare age and growth of greater amberjack between headboat and charterboat fisheries from th e west coast of Florida, and between the charterboat fisheries on the west coast of Florida and Alabama.

PAGE 16

16 0 500 1000 1500 2000 2500 3000 3500 1970 1980 1990 2000 2010 Year Landings (metric tons) Commercial Recreational Figure 11. Annual landings of greater amberjack in the Gulf of Mexico (NOAA Fisheries pers. comm.; Cummings and McClellan 2000).

PAGE 17

17 0 200 400 600 800 1000 1200 1996 1998 2000 2002 2004 2006 2008 Year Landings (metric tons) Florida Alabama All Recreational Figure 12. Annual landings of greater amberjack in recreational fisheries of Alabama and the west coast of Florida (NOAA Fisheries pers. comm.).

PAGE 18

18 0 50 100 150 200 250 300 350 400 450 1996 1998 2000 2002 2004 2006 2008 Year Landings (metric tons) Alabama Charter Florida Charter Figure 13. Annual landings of greater amberjack in the charterboat fisheries of Ala bama and the west coast of Florida (NOAA Fisheries pers. comm.).

PAGE 19

19 0 2000 4000 6000 8000 10000 12000 14000 16000 1992 1994 1996 1998 2000 2002 2004 Year Landings (number of fish) Florida Total, Gulf of Mexico Figure 14. Annual landings of greater amberjack in Floridas headboat fishery and the total from the Gulf of Mexico headboat fishery (SEDAR 2006).

PAGE 20

20 CHAPTER 2 METHODS 2.1 Greater Amberja ck Collections Otoliths and otolith source data from greater amberjack landed in the Florida and Alabama charterboat and headboat fisheries were supplied by the Gulf States Marine Fisheries Commission under the Southeast Recreational Fisheries Information Network [RecFIN(SE)] and by the NOAA Fisheries Laboratory in Panama City, Florida. These agencies collect amberjack otoliths as part of their ongoing fishery dependent monitoring programs. Additional samples were collected by port sampling the headboats at Hubbards Marina at Madeira Beach, Florida. The data used in this study were a subset of a larger data set analyzed by Murie and Parkyn (2008). Size frequencies of amberjack sampled from the Florida charterboat and headboat and the Alabama charterboa t fisheries were compared using a Kolmogorov Smirnov D statistic (Sokal and Rohlf 1969). 2.2 Aging 2.2.1 Otolith Measurements and Processing Otoliths were cataloged in a database along with fish total length (TL, mm), fish fork length (FL, mm), fish ma ss (M, kg; when possible), sex, date of capture, location of capture, type of fishery (e.g., charterboat or headboat), and gear (e.g., hook and line or spear). All gear types for charterboats were combined into one category because specific gear type was not available for many samples but > 99% of samples with known gear type were caught by hook and line. Prior to processing for aging, all whole otoliths were measured for otolith total length (OTL; anterior tip of the rostrum in straightline distance to the posterior edge), otolith antirostrum length (OAL; anterior tip of the antirostrum in straight line distance to the posterior edge), otolith height (OH; maximum distance from the dorsal to ventral edge of the otolith), and

PAGE 21

21 rostrum height (RH; maximum do rsal to ventral distance of the rostrum). Otoliths were measured using computerized digital calipers integrated into a stereomicroscope (MOTIC). Not all measurements could be determined for all otoliths due to a large number of broken otoliths. Otolith measurements were regressed against amberjack FL to determine if overall otolith growth was correlated with fish length. These regression analyses, and all other sta tistical analyses, were tested for significance at P After measuring otoliths whole, the left otolith (right if left broken) was embedded in Devcon Five minute Epoxy using a 4 x 12 mm silicone bullet mold. Each mold was filled with enough epoxy resin to cover the bottom and the otolith was then positioned in the resin, keeping the otolith parallel to the long axis and perpendicular to the short axis of the mold. Once the first layer of epoxy had cured, enough epoxy was added to the mold to just cover the otolith. Embedded otoliths were removed from the mold and glued to frosted slides using cyanoacrylate glue. Otoliths were then viewed under a stereomicroscope and the core of the otolith was marked with a fine felt tip marker. A Buehler Isome t 1000 digital sectioning saw was then used to obtain two transverse sections of the embedded otolith taken through the core. To do this, three diamond edged wafering blades (7.6 cm diameter x 0.15 mm width) (Norton Company, Worcester, MA) were separated by two stainless steel shims (0.5 mm thick), with the entire apparatus mounted together on the saw. This assembly allowed for removal of two 0.5 mm sections with a single pass through the core of the otolith. The sections were then rinsed in de ionized w ater, patted dry, and mounted flat on a microscope slide using Loctite Crystalbond 509 clear thermoplastic adhesive (Henkel, Rocky Hills, CT). Mounted sections were sanded by hand using 600 grit wet sandpaper and polished with a Buehler polishing cloth a nd 0.3 micron

PAGE 22

22 aluminum oxide powder suspended in de ionized water. All slides were rinsed in de ionized water and allowed to air dry before aging. 2.2.2 Aging Criteria and Age Estimation Aging criteria were established by viewing an initial subset of 100 sectioned otoliths. To establish aging criteria, otoliths were viewed for clarity of the opaque and translucent zones over the sectioned surface, for inconsistencies between the ventral and dorsal portions of the otolith, and for distinguishing between tr ue annuli and false annuli or checks. Checks are small lines or marks within the translucent portion of the otolith that resemble annuli but do not continue through the entire otolith. These marks are not considered true annuli as they only appear in a l imited area and at some point merge into complete annuli (Chilton and Beamish 1982) A subsample of 24 pairs of otoliths were subjected to a staining regimen to evaluate the feasibility and effectiveness of this procedure in improving otolith readability, based on the method improving readability of pompano ( Trachinotus carolinus ) otoliths (pers. comm. Cathy Guindon Flor i da Fish and Wildlife Marine Research Institute, St. Petersburg, FL). Both otoliths of each pair were sectioned and polished as describ ed above. After polishing, one of each pair of the prepared otolith slides was immersed in Sandersons Rapid Bone Stain at 40 C for 8 hours. Stained and unstained otoliths were then visually compared to evaluate readability. For aging purposes, the n umber of pairs of opaque/translucent zones were enumerated following Chilton and Beamish (1982) and specifically for greater amberjack (Harris 2004; Harris et al. 2007). In addition, to be able to assign the fish into comparable age classes (based on a 1 January birth date, Chilton and Beamish 1982), the amount of growth on the otoliths edge after the deposition of the last opaque zone was semi quantitatively characterized as 0 (opaque zone was just visible on the edge of the otolith but with no transluce nt growth after it), 1 (the amount of translucent growth after the ultimate opaque zone was more than zero but

PAGE 23

23 width of the previously complete annulus), 2 (the amount of translucent growth past the ultimate opaque zone was > 1/3 but the width of the previously complete annulus), and 3 (the amount of translucent growth past the ultimate opaque zone was > 2/3 of the width of the previously complete annulus). In addition, a qualitative score of how easily the otolith was read was record ed, ranging from 1 being clear and distinct annuli, first annulus well defined, edge well defined to 4 being annuli diffuse and not distinct anywhere in section (i.e., unreadable). Previous experience by P. Harris (SCDNR, personal communication to D. Murie) has shown this to be advantageous when comparing precision between readers. Otoliths were read twice by the primary reader (EEL) independently, with at least 2 weeks between aging periods and with no knowledge of the size or date of collectio n of the fish. For fish captured after January 1 and having significant translucent growth beyond the last opaque zone, age class was determined by the annulus count plus one. When these two ages agreed, this age was considered to be the resolved age. I n cases where these two ages did not agree, the otoliths were read a third time by the primary reader independent of the first two ages. When two of the three assigned ages agreed, that age was considered to be the resolved age. Precision was estimated b y calculating percent agreement (Sikstrom 1983), index of precision (D) (Chang 1982), average percent error (APE) (Beamish and Fournier 1981) and the coefficient of variation (CV) (Kimura and Lyons 1991). 2.2.3 Determination of the F irst A nnulus Determi ning where the first annulus is deposited can be a problem when aging fish, including greater amberjack. To attempt to address this specific aging criterion, two youngof the year (YOY) fish collected in July in a fisheryindependent trawl survey (SEAMAP, NOAA Fisheries, Pascagoula Laboratory) were used to evaluate the location of the first annulus. The otoliths of these fish were extracted and processed in the same manner as the other samples.

PAGE 24

24 Since greater amberjack have a median spawning date of April 1 (Harris 2004), these measurements were considered to be the minimum radius of a sectioned otolith from a fish ~3 4 months old (April to July based on previous research by Wells and Rooker, 2004). This measurement was then used as a reference distance o n otolith sections for fish thought to be over 1 yr old and having one visible annulus (based on growth curve by Thompson et al. 1999). The first opaque zone past this reference distance was considered to represent the end of the first year of growth 2.3 Validation of Aging Method : Periodicity and Timing of Annulus Formation Timing and perio dicity of increment formation was indirectly validated by determining the month that the edge code, and therefore the ultimate increment of growth, was at a minimum (Ha rris et al. 200 7) rather than a marginal increment ratio as used in Dutka Gianelli and Murie (2001) Edge code values for all samples were plotted by month over 12 months. The number of minima present over the 12 month period indicate s the number of incr ements deposited each year (i.e., one, two, or multiple). If only one minimum is present, opaque zone formation occurs annually (thus an annulus) with formation considered complete by the end of the month wherein the minimum is detected. 2.4 Age and Grow th Sample sizes of aged amberjack otoliths for each fishery were not adequate enough to model the data using a von Bertalanffy growth model (Ricker 1975); data were lacking for fish below the minimum size regulation, as well as for very large individuals. Therefore, observed mean lengthat age was determined for age classes 2, 3, and 4 for greater amberjack caught in Florida headboat, Florida charterboat, and Alabama charterboat fisheries. Mean length at age for each age class was compared among fishery s ectors using one way analysis of variance

PAGE 25

25 (ANOVA). If significant, the ANOVA for each age class was followed by Tukeys HSD (SAS, Cary, NC) test to determine where the differences existed among the three fishing sectors.

PAGE 26

26 CHAPTER 3 RESULTS 3.1 Greater Am berjack Collections In total, 505 greater amberjack were sampled from Florida charterboat and headboat fisheries, and the Alabama charterboat fishery, during 20002007 (Table 31). Florida charterboat samples consisted of 201 fish in total (102 females, 46 males and 53 of unknown sex) whereas 129 amberjack were sampled from Florida headboats (80 females, 39 males, and 10 of unknown sex). The Alabama charterboat samples consisted of 114 females, 53 males, and 8 of unknown sex for a total of 175 fish. In general, greater amberjack caught by Florida headboats were larger than those caught by Florida charterboats. Headboat caught fish ranged in size from 287 mm to 1245 mm, with a mean of 880 mm (Fig. 3 1a), while charterboat caught fish ranged in size from 535 mm to 1278 mm, with a mean size of 835 mm (Fig. 3 1b). Comparative length frequency distributions of amberjack from these two fisheries were significantly different (Kolmogorov Smirnov D: dmax = 0.3383, P=0.01), reflecting the larger size distributio n of fish caught by headboats (Fig. 3 2a). Fish caught in charterboats off Florida were, however, larger than fish caught in charterboats off Alabama. Alabama charterboat fish ranged in size from 650 mm to 1190 mm with a mean of 782 mm (Fig. 31c). Comp arative length frequency distributions of amberjack caught in the two charterboat fisheries were significantly different (Kolmogorov Smirnov D: dmax=0.3029, P=0.01), further indicative of smaller Alabama fish (Fig. 3 2b). 3.2 Aging 3.2.1 Otolith Measure ments and Processing Of the 505 greater amberjack collected, 468 sagittae were available for aging and a subset of these were intact for otolith measurements (Table 3 2) due to the fragility of amberjack

PAGE 27

27 otoliths. Regressions between otolith measurements (total otolith length, otolith antirostrum length, and otolith height) and amberjack fork length were all positively related (all P but regression coefficients were all too low (i.e., r2 = 21 47%) to be predictive (Table 3 2). 3.2.2 Aging Criteria and Estimating Ages When viewed using a stereomicrosco pe with transmitted light, cross sections of amberjack otoliths had alternating opaque and translucent zones, each pair comprising an annulus (Fig. 3 3). Annuli were readily apparent in young fish on the dorsal, medial area of the sulcus. The core of the otolith appeared as an opaque zone surrounded by a translucent zone at the base of the deep sulcus (Fig. 3 3a). Annuli were visible on both the ventral and dorsal areas of the sulcus but were much more apparent on the dorsal medial edge. In most cases, annuli that were visible on the dorsal area of the sulcus were also visible on the ventral side (Fig. 3 3b). Otoliths from older fish were distinctly different in shape from those of younger fish. While the core and inner area of these otoliths was very similar to younger fish, otoliths of older fish had long processes on the medial edges of the sulcus, as well as on the dorsal and ventral edges of the structure (Fig. 3 4). These processes contained marks that resembled annuli, but it was very difficult to count them as they tended to become very tightly stacked and less pronounced than the marks from earlier years that were considered true annuli. Qualitatively, staining otolith sections with Rapid Bone Stain did make it possible to count annuli whil e using reflected light, since the core stained dark purple and opaque zones showed as purple bands in an otherwise white background of translucent zones (Fig. 3 5a). This lent no discernable advantage, however, over viewing unstained otoliths using trans mitted light (Fig. 3 5b). Stained otoliths were qualitatively not any easier to read than unstained otoliths, and the staining process was time consuming. Therefore the staining method was not used in further analysis.

PAGE 28

28 Within reader percent agreement for otolith age assignments was 89.1% for otoliths assigned the same age, and 99.1% for ages in agreement by 1 year; 8% of otoliths were either broken or contained no discernable annuli and were deemed unreadable. Otoliths with no discernable annuli (unreadable) were not included in calculations of agreement and precision. Indices of precision for aged otoliths returned an APE of 1.93% and a CV of 2.73%. 3.2.3 Determination of the F irst A nnulus Determination of first annulus was accomplished by examini ng the otoliths of two youngof year (YOY) amberjack captured in the northern Gulf of Mexico in July by a scientific trawl survey (S. Nichols, pers. comm.). The sectioned otoliths of these two fish appeared to have some opaque growth in the core area but had not yet completed their first annulus (Fig. 3 6a). The growth pattern in these otoliths was then compared to that in otoliths of fish showing one complete annulus and growth at the edge of the otolith (Fig. 3 6b). The radius of each YOY otolith was 0 .35 mm, whereas the first opaque zone completing an annulus in the 1 year old fish had a radius of ~0.6 mm. Wells and Rookers (2004) equation for YOY amberjack captured off Galveston, Texas, in 20002001 was: SL (mm) = 2.00 (age in days) 37.32. Using this equation, and solving for age (days), then was: Age (days) = (SL + 37.32)/2.00. The two fish from the trawl were 162 mm and 174 mm SL, therefore these fish would be 3.3 and 3.5 months old. This w ould mean they were spawned in April of 2005 which is considered the mean spawning date for amberjack in the Gulf of Mexico (Harris et al. 2007). 3.3 Validation of Aging Method: Marginal I ncrement Analysis Edge code analysis resulted in a single minimum in average edge code over a 12 month period. This single minimum indicated that opaque zone formation occurred once per year between May and July (Figure 3 7).

PAGE 29

29 3.4 Age and Growth Greater amberjack caught in the charterboat fisheries from the west coast of F lorida ranged in age from 2 8 years, with the majority of fish being 3 years old (Figure 3 8). The majority of amberjack caught in the Alabama charterboat fishery were also 3 years old and ranged between 2 and 5 years of age. Greater amberjack caught in the headboat fishery from the same area on the west coast of Florida also had a mean age of 3 years, but included fish from 0 to 6 yrs of age. Otolith radius as a function of fish FL was significant overall for amberjack from charterboats in Florida and Alabama and headboats in Florida (Table 3 3). The regressions all had low r2 values (all <28%), however, indicative that otolith radius was a poor predictor of fish FL (Fig 3 9). Based on the lack of a predictive relationship between otolith radius and f ish length, fish lengths could not be back calculated to length at age. Therefore, comparisons of fish size with age were restricted to observed length at age. Mean observed lengthat age was significantly different among fisheries for amberjack in age classes 2, 3, and 4 (Table 3 4) (P<0.01). Greater amberjack landed in the Florida headboat fishery were smaller than either the Florida charterboat or Alabama charterboat fisheries at age 2, but larger at ages 3 and 4 (Tukeys HSD: all P<0.05) (Table 34; Fig. 3 10). Amberjack landed in the Florida charterboat fishery were not significantly different in sizeat age compared to fish landed by Alabama charterboats (Table 3 4; Fig. 3 10) (all P>0.05), except for 3 year old amberjack (all P<0.05).

PAGE 30

30 Table 3 1. Number of greater amberjack sampled from each fishery in the Gulf of Mexico during 20002007. Source 2000 2001 2002 2003 2004 2005 2006 2007 Florida Headboat 11 31 7 80 Florida Charterboat 5 145 51 0 Alabama Charterboat 0 148 27 0 Table 3 2. Rel ationship between greater amberjack forklength (FL, mm) and otolith total length (OTL, mm), otolith antirostrum length (OAL, mm), and otolith height (OH, mm). All regressions were significant at P Regression r2 n FL = 259.8 + 55.2*O T L 0.31 84 FL = 47.8 + 107.2 O A L 0.47 198 FL = 214.1 + 189.0 OH 0.21 298 Table 3 3. Relationship between greater amberjack forklength (FL, mm) and otolith radius (OT R, mm) for Florida charterboat (FL CB), Florida headboat (FL HB), and Alabama charterboat (AL CB) fisheries. Area Fishery Sex Regression r 2 N Significanc e 1 FL CB M FL=546.3+321.6*OTR 0.23 37 ** FL CB F FL=488.4+406.7*OTR 0.16 81 *** FL CB All FL =469.2+414.3*OTR 0.27 163 *** FL HB M FL=656.7+288.9*OTR 0.05 36 NS FL HB F FL=584.3+396.8*OTR 0.16 74 *** FL HB All FL=609.9+354.2*OTR 0.11 117 *** AL CB M FL=617.9+212.6*OTR 0.05 53 NS AL CB F FL=430.0+456.0*OTR 0.14 114 *** AL CB All F L=361.4+551.0*OTR 0.19 175 *** 1 NS (P>0.05), *(P<0.05), **(P<0.01), ***(P<0.001)

PAGE 31

31 Table 3 4. Mean lengthat age, standard error of the mean (SE), and sample size (n), for greater amberjack from sampled fisheries and regions in the Gulf of Mexico. Age C lass Florida Headboat Florida Charterboat Alabama Charterboat Mean SE n Mean SE n Mean SE n 0 287 0 1 1 566 0 1 2 657.1 14.2 7 740.8 14.2 30 742.2 10.8 16 3 894.3 7.6 101 809.5 6.6 102 768.1 4,1 134 4 1,014.8 21.5 8 901.1 16.8 32 854.8 2 2.2 22 5 939 17 2 913.5 34.5 2 1,089 101 3 6 1,166 0 1 1,023.5 91.5 2 7 970.0 84 2 8 1,147 74 2 Total 121 172 175 Table 3 5. Comparisons of mean lengthat age (ages 2, 3, and 4 yr old) for greater amberjack from Florida Headboat (FLHB), Florida Charterboat (FLCH), and Alabama Charterboat (ALCH). Source Comparison Difference Between Mean Length at age (mm) Tukeys HSD Significance 1 Age Two FLHB FLCH 83.7 FLHB ALCH 85.0 FLCH ALCH 1.4 NS Age Three FLHB FLCH 84 .8 FLHB ALCH 126.2 FLCH ALCH 41.4 Age Four FLHB FLCH 113.7 FLHB ALCH 159.9 FLCH ALCH 46.3 NS 1 NS (P>0.05), *(P<0.05)

PAGE 32

32 Figure 31. Length frequencies of greater amberjack from the Gulf of Mexico caught in: a) he ad boats and b) charter boats off the west coast of Florida; and c) charterboats from Alabama.

PAGE 33

33 b) 0 20 40 60 80 100 0 500 1000 1500 Fork Length (mm) Cumulative Relative Frequencies Alabama Charterboat n=175 Florida Charterboat n=201 Figure 32. Comparative length frequency distributions for greater amberjack from the Gulf of Mexico caught in: a) charterboat and headbo at fisheries off the west coast of Florida; and b ) charterboats from Florida and Alabama.

PAGE 34

34 Figure 33. Cross sections of otolith s from greater amberjack: a) 3 year old fish, with the core (C ) clearly demarked; and b) 5 yr old fish. b a C

PAGE 35

35 Figure 34. Otolith from an older greater amberjack showing the long processes on the medial edges of the sulcus, as well as on the dorsal and ventral edges of the otolith.

PAGE 36

36 Figure 35. a) Otolith section stained with Rapid Bone Stain viewed under reflected light; and b) otolith section from the same greater amberjack that was not stained as viewed with transmitted light. a b

PAGE 37

37 Figure 36. Otolith section from a) a young of the year greater amberjack and b) a 1 year old fish showing growth past the first annulus a b 0.5 mm 0.5 mm

PAGE 38

38 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0 1 2 3 4 5 6 7 8 9 10 11 12 Month Edge code Figure 37. Mean edge code of greater amberjack otoliths (n=467) as a function of month of collection throughout the year. Vertical bars represent standard error of the mean. Minimum mean edge code in June indicated the formation of the op aque zone.

PAGE 39

39 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 0 1 2 3 4 5 6 7 8 Age (years) Frequency Alabama Charterboat Florida Charterboat Florida Headboat Figure 38. Age frequency of greater amberjack from the Gulf of Mexico by location and fishery (n=468).

PAGE 40

40 400 500 600 700 800 900 1000 1100 1200 1300 0.4 0.6 0.8 1 1.2 1.4 Otolith Radius (mm) Fork Length (mm) Figure 39. Greater amberjack fork length as a function of otolith radius (n=455).

PAGE 41

41 a) b) 0 200 400 600 800 1000 1200 0 1 2 3 4 5 6 7 8 Fork Length (mm) Florida Charter All Florida Headboat All Alabama Charter All 0 200 400 600 800 1000 1200 0 1 2 3 4 5 6 7 8 Age (years) Fork Length (mm) Florida Charter Mean n=164 Florida Headboat Mean n=116 Alabama Charter Mean n=172 Figure 310. a) Length at age for greater amberjack caught in all categories (n=468); b) Mean lengthat age for greater amberjack from age classes 2, 3, and 4.

PAGE 42

42 CHAPTER 4 DISCUSSION For greater amberjack in the Gulf of Mexico (GOM) fisheryspecific size and age of the catch were different among Florida headboat, Florida charterboat and Alabama charterboat fisheries. Surprisingly, for greater amberjack landed in the Florida headboat fishery, the size frequency distribution was skew ed toward larger fish compared to fish landed in the Florida charterboat fishery. In addition, the observed lengthat age for fish in age classes 3 and 4, which represented 30% of the landed catch in 2004 (SEDAR 2006), was greater in the headboat fishery compared to fish landed in the charterboat fishery of Florida. We also found that fish landed by charterboats in Alabama were smaller than those landed by charterboats in Florida. Charterboat caught fish from Florida and Alabama differed in sizeat age f or age 3 fish only. This difference was much less dramatic in scale than the charterboat/headboat difference (differences of 41 mm and 84 mm respectively at age 3). These differences in sizeat age indicate that growth rates for greater amberjack were not equal between all regions and among fishing sectors. Fish are known to exhibit regionally distinct rates of growth (Murphy and Taylor 1994, Dutka Gianelli and Murie 2001, Murie and Parkyn 2005). Murphy and Taylor (1994) found that spotted seatrout Cynosc ion nebulosus exhibited different growth rates between two Florida estuary systems. Dutka Gianelli and Murie (2001) found qualitatively that regional differences in growth rates may exist for sheepshead Archosargus probatocephalus White grunt Haemulon plumieri have also been show to exhibit differences in growth rates between regions on the west coast of Florida, with even more pronounced differences being observed between the Gulf of Mexico and some Atlantic coast populations (Murie and Parkyn 2005). R egional differences in growth rates could occur due to several reasons, for example, differences in genetics, habitat, temperature, migratory patterns and year class effects. Little is known about the genetic

PAGE 43

43 variability of source populations of greater amberjack or whether different populations exist. Gene flow across the northern GOM (between the Florida Middle Grounds and Port Aransas, Texas) is thought to be continuous, but some evidence suggests a division between populations of greater amberjack in the U.S. South Atlantic Ocean and the Gulf of Mexico (Gold and Richardson 1998, SEDAR 2006). More information is needed to determine the degree of genetic variation within the Gulf of Mexico, which is presently being addressed through a NOAA Cooperative R esearch Program grant (D.J. Murie, D.C Parkyn, and J.D. Austin; pers. comm.). Fish movement could influence genetic variability as well as the size distribution of fish available to specific fisheries. For example, gag Mycteroperca microlepis are largest in those portions of the fishery that exploit deeper waters farther from shore in the Gulf of Mexico (Fitzhugh et al. 2003). In this case differences in growth rate relate more to the geographical distribution of the specific type of fishing effort more than differences in the gear selectivity in the specific fishery (Fitzhugh et al. 2003). Thompson et al. (1999) reported seasonal changes in the size distributions of amberjack caught aboard charterboats in Louisiana, with larger fish being caught from Ma y to September. They further speculate that amberjack may move to warmer water to avoid cooler water in winter, altering the size distribution (Thompson et al. 1999). These seasonal movements could explain the variation in the size distribution of the ca tch between Florida and Alabama charterboats. Water temperature also affects fish growth rates as well as scope for activities, such as swimming and foraging (Brett 1979, Kline 2004). Average water temperature is higher in the southern GOM (pers. comm., National Oceanographic Data Center web page www.nodc.noaa.gov) and therefore amberjack in the southern GOM may grow faster than those in the northern GOM, if sufficient food resources are available to support their potential for

PAGE 44

44 increased growth. If some amberjack move southward during cooler months and fish remaining in northern areas grow slower due to temperature, the interaction of these factors could produce some of the disparity in growth rates between the sa mpled areas of Alabama and Florida. Even if this is the case, however, it does not explain the higher growth rate of Florida headboat fish when compared to Florida charterboat fish, since all of these fish are caught off the west coast of Florida. Most l ikely this can be attributed to differences in gear and/or techniques between charterboats and headboats. Differences in fishing mortality and gear selectivity have also been shown to influence observed growth parameters in fish populations (Ricker 1975). In many hook and line fisheries, larger individuals are targeted, which can also be the faster growing individuals in a cohort. Therefore, slower growing individuals in a cohort are left to reproduce in the population. Over decades, this selective press ure can lead to slower individual growth rates in a given population (e.g., vermillion snapper Rhomboplites aurorubens ; Zhao et al 1997). When sampling populations with methods that select for faster growing individuals, the average size at age may appear higher when compared to methods that are less selective for higher growth rate. The truncated age distribution apparent in this study indicates that only a very small range of fish ages are being captured in these fisheries. The low number of age two and younger fish is likely the result of the majority of greater amberjack not having recruited to the fishery at that age. The truncation of the age distribution past age four could be the result of many possible factors, including the type of fishing pre ssure described above, influence from other fishing sectors such as commercial vessels, and fish movement patterns. Perhaps the most anomalous finding of this project is in the larger size and larger sizeat age of Florida headboat fish when compared to Florida charterboat fish. It is commonly believed

PAGE 45

45 that charterboats land larger fish, in general, than headboats. The larger size and faster growth of headboat fish in this study may be explained by the nature of specific headboat trips. Most headboats run half day and full day trips, but some run overnight trips as well. The differences in trip duration relate more to targeted fishing area than to the amount of time that lines are in the water. On full day and overnight trips, boat captains target areas farther offshore than on half day trips. The difference in time on the boat is mostly spent in transit to fishing sites. This targeting of areas farther offshore is more similar to charterboat fishing than it is to half day headboat fishing. Often, h eadboat captains will make a special effort to visit an amberjack aggregation or AJ hole. At these sites, most customers on the vessel (sometimes as many as 50) will catch at least one amberjack. By making these visits to known amberjack aggregation si tes, such as wrecks, headboats may indeed target populations of offshore amberjack that grow faster than inshore amberjack. Charterboats may be less likely to visit such a site regularly, especially if it is far offshore, because a charterboat usually car ries a maximum of six people. It may be less cost efficient for a charterboat captain to visit an AJ hole because only six fish can be kept (i.e., regulations of one amberjack per person). Customers and charterboat captains might consider their time an d fuel better used by targeting grouper fishing sites (i.e., regulations of five grouper per person), and only catch amberjack incidentally. The relatively low sample size encountered in this project was indicative of the overall lack of availability of greater amberjack in the sampling programs for fisheries in the GOM during 20002007, especially very young and very old fish. There are several reasons for this, including: an overall lack of port sampling effort because greater amberjack may not be a p riority species for most agencies, a large minimum size regulation, an underlying lack of older individuals in the population, a reduction in the likelihood of landing larger individuals,

PAGE 46

46 difficulty of extracting whole otoliths from greater amberjack, and a need for improving aging criteria. In the first instance, few amberjack samples are collected by state and federal sampling agencies when compared to other reef fishes in hook and line fisheries. One major reason for this may be that the fishing regu lations allow for only one fish per individual, and hence it may not be cost efficient to spend limited sampling time and dollars sampling only a few amberjack per vessel in comparison to sampling other reef fishes. Sending port samplers to collect data o n fishes landed in the various fisheries is expensive in both time and funds. Since researchers require a robust sample size, this may also direct research effort toward species that are sampled more frequently, such as snapper and grouper. These fish gr oups support important targeted fisheries in the southeastern U.S., and intensive sampling is justified, but at the same time this creates a shortage of data for other species that also must be managed effectively on the limited funds available, especially when data are required over a period of time to be useful in stock assessments (e.g., age of the catch). The lack of relatively young, small fish in comparisons between headboats and charterboats in this study was a result of the large (i.e. 28 inch FL prior to 2009) minimum size regulation for GOM greater amberjack. Without the availability of smaller, and therefore younger, amberjack, length at age comparisons were restricted to a truncated set of age classes (2, 3, and 4 yr old fish) with adequate s ample size. To facilitate modeling a complete growth curve over the full range of ages of amberjack, additional, onboard sampling of greater amberjack slated for release in the fisheries, or fishery independent sampling, would be necessary (see Thompson et al. 1999; Murie and Parkyn 2008). The availability of younger fish would also presumably improve predictability of regression equations for otolith radius as a

PAGE 47

47 function of fish length, possibly permitting back calculated length at age. This could be im portant for characterizing growth in amberjack because of their rapid increase in length in their first 3 years, with observed length at age (as in this study) possibly influenced by the time of year that sampling occurred. In addition, older individuals are poorly represented in the data sets of previous studies of greater amberjack age and growth (Manooch and Potts 1997, Thompson et al. 1999, Harris 2007, Murie and Parkyn 2008), and maximum ages are reported from 10 to 15 years depending on the specific study. While all of these studies have various limitations, it seems likely that the lack of older individuals represents a real paucity of older fish. Alternatively, when older, and larger, fish are encountered, there is likely a reduction in successfu l landing of the fish due to the brute strength and endurance of this species. Greater amberjack are sometimes colloquially referred to as reef donkeys for their stubborn resistance to angler success. Lastly, when older individuals were sampled and age d for the present study, they were more likely to be excluded from the final data analysis due to a lack of a resolved age classification. With further improvement in aging criteria, these individuals were able to be included in fishery specific growth mo dels for greater amberjack in the Gulf of Mexico (Murie and Parkyn 2008). Reliability of otolith aging was comparable to previous studies on greater amberjack and was not perceived as problematic for amberjack in ages 2 through 5, which covered the among fisheries comparisons in this study. Within reader percent agreement for those otoliths was 89% (APE of 1.9, CV of 2.7), indicating that otoliths were aged precisely. The age distribution of fish in the samples from the charterboats and headboats should adequately reflect the relative abundance of these age classes in the fisheries, as the samples were collected by state and federal agencies using sampling designs meant to characterize the catch.

PAGE 48

48 The most significant consequence of fishery specific and r egion specific differences in the size of greater amberjack observed in this study, over the major age classes targeted in the Gulf of Mexico, is that the use of a single age length key or a single growth model to assign ages to amberjack of a given length (SEDAR 2006) could be problematic in the stock assessment. Sizeat age for fish caught by charterboats, in particular, appeared to be lower than the average amberjack sizeat age used in the stock assessment based on Thompson et al. (1999) for amberjac k caught in Louisiana (Fig. 4 1). Ideally, growth rate differences of fish captured by different fisheries or in different regions should be accounted for when applying ages to the catch derived from the specific fisheries or regions. In the case of ambe rjack caught in charterboat fisheries, whether from Florida or Alabama, assigning an age class by applying the average growth model or age length key would result in these fish being assigned to a younger age class (e.g., 4 yr old amberjack from the Flor ida charterboat fishery were the same size as 3 yr old amberjack based on the growth curve by Thompson et al. (1999) (Fig. 41). Although a 1yr difference in age class assignment may seem trivial, it is important to note that greater amberjack fisheries in the Gulf of Mexico primarily harvest fish over four age classes (2, 3, 4 and 5) (Cummings and McClellan 2000; this study), and a bias in determining the productivity in any one of those age classes could therefore also potentially bias the stock assess ment. In addition, incorrectly assigning ages to the catch would make it difficult to correctly gauge the importance of age class cohorts contributing to the fisheries. While it may not be feasible to assign ages of amberjack based on fishery specific an d region specific growth rates due to the paucity of data addressing this issue, it should nevertheless be considered as a source of uncertainty in the reliance on an age structured stock assessment model for greater amberjack in the Gulf of Mexico.

PAGE 49

49 F igure 41. Comparison of mean observed lengthat age for greater amberjack from the Gulf of Mexico sampled in this study compared to the von Bertalanffy growth curve of Thompson et al. (1999).

PAGE 50

50 REFERENCES Beamish, R.J. 1981. Use of fin ray sections to a ge walleye pollock, pacific cod, and albacore, and the importance of this method. Trans. Am. Fish Soc. 110: 287299. Beamish, R.J. and H.H. Harvey. 1969. Age determination in the white sucker. J. Fish. Res. Bd. Canada. 26: 633 638. Beamish, R. J. and D. Chilton. 1977. Age determination of lingcod ( Ophiodon elongates ) using dorsal fin rays and scales. J. Fish. Res. Bd. Canada. 34 (9): 1305 1313. Beamish, R.J. and D.A. Fournier. 1981. A method for comparing the precision of a set of age determinations. Canadian Journal of Fisheries and Aquatic Sciences 38: 982 983. Beasely, M. 1993. Age and growth of greater amberjack, Seriola dumerili from the northern Gulf of Mexico. M.S. Thesis, Dept. of Oceanography and Coastal Sciences, Louisiana State University. 85pp. Bhlke, J.E. and C.C. Chaplin. 1993. Fishes of the Bahamas and adjacent waters. 2nd edition. University of Texas Press, Austin. 773 pp. Burch, R.K. 1979. The greater amberjack, Seriola dumerili : its biology and fishery off Southeastern Florida. Unpublished M.S. Thesis. University of Miami. 112 pp. Brennan, J.S. and G.M. Cailliet. 1989. Comparative age determination techniques for white sturgeon in California. Trans. Am. Fish. Soc. 118: 296310. Brett, J.R. 1979. Envi ronmental Factors and Fish Growth. In Fish Physiology Vol. 8, ed. W.S. Hoar, D.J. Randall, and J.R. Brett. London: Academic Press. Cerrato, R.M. 1990. Interpretable statistical tests for growth comparisons using parameters in the von Bertalanffy equati on. Can. J. Fish. Aquatic Sci. 47: 14161426. Chang, W.Y.B. 1982. A statistical method for evaluating reproducibility of age determination. Can. J. Fish. Aquat. Sci. 39: 12081210. Chilton, D.E., and R.J. Beamish. 1982. Age determination methods for fishes studied by the Canadian Groundfish Program at the Pacific Biological Station. Can. Spec. Publ. Fish. Aquat. Sci. 60. 102 pp. Cummings, N.J. 1998. An analysis of the Gulf of Mexico greater amberjack, Seriola dumerili stock condition. Proc Gulf and Carib. Fish Inst. 50: 206227.

PAGE 51

51 Cummings, N.J. and D.B. McClellan. 2000. Trends in the Gulf of Mexico greater amberjack fishery through 1998: Commercial landings, recreational catches, observed length frequencies, estimates of landed and dis carded catch at age, and selectivity at age. U.S. Dept of Commerce, National Oceanographic and Atmospheric Administration, National Marine Fisheries Service, Sustainable Fisheries Division. Dutka Gianelli, J., and D.J. Murie. 2001. Age and growth of she ephead, Archosargus probatocephalus (Pisces: Spaidae), from the northwest coast of Florida. Bull. Mar. Sci. 68: 6983. Fitzhugh, G.R., L.A. Lombardi Carlson and N.M. Evou. 2003. Age structure of gag ( Mycteroperca microlepis ) in the eastern Gulf of Mexi co by year, fishing mode, and region. Proceedings of the Gulf and Caribbean Fisheries Institue 54: 538 549. Francis, R.I.C. 1990. Back calculation of fish lengths: A critical review. J. Fish. Biol. 36: 883 902. Gillanders, B. M., D. J. Ferrell and N. L. Andrew. 1999. Size at maturity and seasonal changes in gonad activity of yellowtail kingfish (Seriola lalandi; Carangidae) in New South Wales, Australia New Zealand Journal of Marine and Freshwater Research, 1999, 33: 457 468. Gold, J. R. and L. R. Ric hardson. 1998. Mitochondrial DNA Diversification and Population Structure in Fishes from the Gulf of Mexico and Western Atlantic. J. Heredity 89:404 414 Harris, P. J, D.M. Wyanski, D.B. White, and P.P. Mikell. 2007. Age, Growth, and Reproduction of Greater Amberjack off the Southeastern Atlantic Coast. Trans. Of the American Fisheries Society 136: 15341545. Harris, P. J, 2004. Age, Growth, and Reproduction of Greater Amberjack, Seriola dumerili, in the southwestern North Atlantic. Analytical Report of MARMAP Program. South Carolina Department of Natural Resources, Charleston, SC. Hoese, H. Dickinson and R.H. Moore. 1998. Fishes of the Gulf of Mexico, Texas, Louisiana, and adjacent waters. University of Texas Press, Austin. 422 pp. Kimura, D.K. 19 80. Likelihood methods for the von Bertalanffy growth curve. Fish. Bull. 77: 765776. Kimura, D.K., and J.J. Lyons. 1991. Between reader bias and variability in the age determination process. Fish. Bull., U.S. 89: 53 60. Kline, R.J. 2004. Metabolic rate of gag grouper, ( Mycteroperca microlepi s) in relation to swimming speed, body size and temperature. Unpublished M.S. thesis. Univeristy of Florida. http://etd.fcla.edu/UF/UFE0008925/kline_r.pdf

PAGE 52

52 Manooch, C.S. 1984. Fishermans Guide to the Fishes of the Southeastern United States. North Carolina State Museum of Natural History. 362 p. Manooch, C.S. and J.C. Potts. 1997. Age, growth, and mortality of greater amberjack, Seriola dumerili from the U.S. Gulf of Mexico headboat fishery. Bull. Mar. Sci. 61: 671683. Murie, D.J., and D.C. Parkyn. 2005. Age and growth of white grunt ( Haemulon plumieri ): a comparison of two populations along the Florida west coast. Bull etin of Mar ine Sci ence 76(1): 73 93. Murie, D.J., and D.C. Parkyn. 2008. Age, grow th and sexual maturity of greater amberjack ( Seriola dumerili ) in the Gulf of Mexico. MARFIN Final Report ( NA05NMF4331071). 34 p. Murphy, B.R. and D. W. Willis. 1996. Fisheries Techniques. Second edition. American Fisheries Society, Bethesda. 732 pp Murphy, M.D. and R.G. Taylor. 1994. Age, growth and mortality of spotted seatrout in Florida waters. Trans. Am. Fish Soc. 123: 482497. NMFS (National Marine Fisheries Service). 2008. NOAA Fisheries Service announces the publication of a new rule to end overfishing and rebuild greater amberjack and gray triggerfish stocks. Southeast Fishery Bulletin FB08 040. Ricker, W.E. 1975. Computation and Interperetation of Biological Statistics of Fish Populations. Bulletin of the Fisheries Research Board o f Cananda. Bulletin 191. Schirripa, M.J. and K.M. Burns. 1997. Growth estimates for three species of reef fish in the eastern Gulf of Mexico. Bull. Mar. Sci. 61 (3): 581591. Sokal, R. R., and F. J. Rolf. 1969. Biometry. W. H. Freeman and Company San Francisco, CA 776 pp. SEDAR (Southeast Data, Assessment and Review). 2006. SEDAR9 Assessment Report 2. Charleston, SC Sikstrom, C.B. 1983. Otolith, pectoral fin ray, and scale age determination for arctic grayling. Prog. Fish. Cult. 45(4): 220223. Tanaka, K., Y. Mugiya, and J. Yamada. 1981. Effects of photoperiod and feeding on the daily growth patters in otoliths of juvenile Tilapia nilotica Fish. Bull., U.S. 79: 459 465. Thompson, B.A., C.A. Wilson, J.H. Render, M. Beasley, and C. Cauthr on. 1992. Age, growth and reproductive biology of greater amberjack and cobia from Louisiana waters. Final report to Marine Fisheries Research Initiative (MARFIN) Program, NMFS, St. Petersburg, FL. NA90AA H MF722, 77 pp.

PAGE 53

53 Thompson, B.A., Beasley, and C.A. Wilson. 1999. Age distribution and growth of greater amberjack, Seriola dumerili from the northcentral Gulf of Mexico. Fishery Bulletin 97: 362371. Turner, S.C. 2000. Catch rates of greater amberjack caught in the headboat fisheries in the Gulf of Mexico in 19861998. NMFS/SEFSC, Miami Laboratory. Document SFD 99/00107. 17 pp. Turner, S.C., N.J. Cummings, and C.E. Porch. 2000. Stock assessment of Gulf of Mexico greater amberjack using data through 1998. NMFS/ SEFSC, Miami Laboratory. Document SFD 99/00100. 27 pp. VanderKooy, S., and K. Guindon Tisdel (Editors). 2003. A practical handbook for determining the ages of Gulf of Mexico fishes. Gulf States Marine Fisheries Commission, Ocean Springs, MS. Publicati on No. 111. 114p. Victor, B.C., and E.B. 1982. Age and growth of the fallfish Semotilus corporalis with daily otolith increments as a method of annulus verification. Can. J. Zool. 60: 2453 2550. Wells, R. J. and J. R. Rooker. 2004. Distribution, Age, a nd Growth of Young of the year Greater Amberjack ( Seriola dumerili ) Associated with Pelagic Sargassum. Fish. Bull. 102:545554 Zhao, B. and J.C. McGovern. 1997. Temporal variation in sexual maturity and gear specific sex ratio of the vermilion snapper, Rh omboplites aurorubens in the South Atlantic Bight. Fishery Bulletin 95: 837848.

PAGE 54

54 BIOGRAPHICAL SKETCH Edward Leonard was born and raised in Marietta, Georgia. After serving 4 years in the United States Army, he returned home and graduated with a B achelor of Science degree in biology from Kennesaw State University in Kennesaw, GA. Eddie worked a t the United States Geological Survey in Gainesville, Florida before entering graduate school at the University of Florida. He is currently employed as a Freshwater Fisheries Biologist with the Florida Fish and Wildlife Conservation Commission.