Age and growth of sheepshead (Archosargus probatocephalus) in Tampa Bay, Florida.
The sheepshead (Archosargus probatocephalus) occurs from Nova Scotia (Gilhen et al., 1976) to Brazil (Caldwell, 1965) and is common in coastal waters from Chesapeake Bay to Texas in the United States (Bigelow and Schroeder, 1953; Collette and Klein-MacPhee, 2002). Two subspecies of sheepshead have been reported within its U.S. range: A. p. probatocephalus, found along the Atlantic coast and into the Gulf of Mexico as far north as Steinhatchee, Florida, and A. p. oviceps, which occurs in the Gulf of Mexico from St. Marks River, Florida, to Campeche Bank, Mexico (Caldwell, 1965). Subspecific distinction is based partly on pigmentation (size and number of vertical body bars) and meristic counts (lateral line scales, gill rakers, and dorsal fin spines and rays), both of which overlap considerably between the 2 subspecies (Caldwell, 1965). Results of recent genetic analyses in which mtDNA of sheepshead from the Gulf of Mexico and South Atlantic indicated that a single panmictic population of sheepshead exists within the range of this species from Texas through North Carolina (Anderson et al., 2008; Seyoum et al., in press). More detailed microsatellite analysis, however, has revealed a significant genetic break at the subspecies boundary in the Florida panhandle (Apalachee Bay), providing genetic support for the validity of 2 subspecies of sheepshead within its range in the United States (Seyoum et al., in press).
The combined recreational and commercial landings of sheepshead from the gulf coast of Florida between 1990 and 2009 made up 19-44% of the total annual landings of sheepshead for all U.S. states in the Gulf of Mexico (National Marine Fisheries Service, Fisheries Statistics and Economics Division commercial annual landings statistics, available from website, accessed June 2014, and Marine Recreational Information Program time-series data, available from website). The combined annual landings from the gulf coast of Florida peaked at 1755.6 metric tons in 1992; from 1996 to 2009, they averaged less than half that amount (841.3 metric tons/year) because of enactment in 1995 of a Florida constitutional amendment that limits the use of entangling nets and mandates the institution of minimum size and bag limits for recreational fishermen (Munyandorero et al. (1)). Historically, more sheepshead have been landed by recreational fishermen than commercial fishermen (70-95% of the combined annual landings during 1990-2009) along Florida's gulf coast (Munyandorero et al. (1)).
Growth has been described for larval and early juvenile sheepshead from Florida waters (Parsons and Peters, 1989). Elsewhere, age and growth studies of juvenile and adult sheepshead have been conducted in Georgia (Music and Pafford (2)), North Carolina (Schwartz, 1990), South Carolina (Wenner (3)), Louisiana (Beckman et al., 1991), and northwest Florida (Dutka-Gianelli and Murie, 2001). Validated (Music and Pafford (2)) and unvalidated (Schwartz, 1990) ages have been determined also from scales. However, Dutka-Gianelli and Murie (2001) reported that scales of sheepshead older than 3 years resulted in underestimated ages, and scales from sheepshead aged 2 or more years have been described as unreadable (Schwartz, 1990; Wenner (3)).
Validated ages determined from otolith sections have been used to estimate von Bertalanffy growth parameters for sheepshead from Louisiana (Beckman et al., 1991), South Carolina (Wenner (3)), and northwest Florida (Dutka-Gianelli and Murie, 2001). All 3 studies noted a high variability in size at age for sheepshead, reported von Bertalanffy growth parameters, and the predicted sizes at age varied considerably among the 3 studies (Dutka-Gianelli and Murie, 2001). Each study relied almost exclusively on the fishery (commercial or recreational) for its samples. Beckman et al. (1991) indicated that because the gear types used were more apt to catch certain sizes of fish than others and because fishermen occasionally sorted the catch before supplying the researchers with samples, the age and size structures of the sheepshead analyzed probably did not represent the overall population of sheepshead in Louisiana. Other researchers also have determined that reliance upon samples obtained only from the fishery can cause misrepresentation of the size distribution and age structure of a population (Miranda et al., 1987; Hilborn and Walters, 1992; Wilson et al., 2015).
By design, we used multiple gear types and fisheries-independent methods to provide a more representative sample across size and age classes of sheepshead, therefore generating estimates of growth pa rameters more representative of the true population. Otolith annuli (opaque zones) were validated to determine age and growth parameters for sheepshead, and these estimates were then compared with those previously reported for sheepshead from other geographical regions.
Materials and methods
Sheepshead were collected in Tampa Bay, Florida (Fig. 1), a large estuary on the west coast of Florida that has an average depth of approximately 3 m and a maximum depth of 13 m (Comp and Seaman, 1985). All sheepshead were captured from 1993 through 2009 by the Fish and Wildlife Research Institute's Fisheries-Independent Monitoring program during routine sampling with haul seines, trawls, gill nets, and trammel nets (Table 1). Haul seine and trawl samples were collected at both stratified-random and fixed sites; gill net collections were made at stratified-random sites. More detailed information about the sampling gears and protocols used by the Fisheries-Independent Monitoring program can be found in Tremain and Adams (1995), Nelson et al. (1997), Nelson (1998), and Winner et al. (2010). Sheepshead were also taken as bycatch from trammel nets, which had been set on visually detected schools of striped mullet (Mugil cephalus), red drum (Sciaenops ocellatus), or common snook (Centropomus undecimalis). For each fish, we recorded standard length (SL), fork length (FL), and total length (TL) to the nearest millimeter; sex; and total weight to the nearest 0.1 g before extraction of sagittal otoliths, which were then rinsed, cleaned, and stored dry for further examination.
Sex ratios of sheepshead were compared with a hypothetical 1:1 sex ratio by using the G-test (Sokal and Rohlf, 1981). Length distributions also were compared between sexes by using the Kolmogorov-Smirnov (KS) 2-sample test (Proc Npar1way procedure in SAS (4) software, vers. 5.1 (SAS Institute Inc., Cary, NC). Linear regression for all sheepshead collected was used to calculate sex-specific length-length and length-weight relationships (Proc Reg procedure in SAS software) for untransformed and transformed ([log.sub.10]) data, respectively, and these relationships were compared through analysis of covariance (ANCOVA; Snedecor and Cochran, 1967). Data from all fish collected were pooled when slopes and intercepts for sex-specific regressions were not significantly different. All significance testing was conducted at P [less than or equal to] 0.05.
Three or four thin (~0.5 mm) transverse sections were cut at or adjacent to the core of the left sagitta with a Buhler Isomet low-speed saw equipped with a diamond blade; a right sagitta was sectioned when the left sagitta was missing or had been damaged. Otolith sections were mounted on microscope slides by using Histomount solution (Thermo Fisher Scientific, Waltham, MA). With a dissecting microscope (8-25x magnification), 2 or 3 readers independently counted the opaque rings on each otolith twice under reflected light. Readers counted rings without knowing the sex, length, or capture date of specimens. Disagreements in annulus counts were resolved by at least 2 readers, without knowledge of previous counts. If an annulus count could not be agreed upon after reexamination, the otolith was rejected from the age and growth analysis.
Validation of annuli counts was completed through marginal-increment analysis, which provided indirect evidence at the otolith margin of the periodicity of annulus formation. Measurements from the core to the proximal edge of each annulus, along the ventral sulcal ridge, were completed with a digital image-processing system for all otoliths processed from 1995 through 1998. The marginal increment was calculated as a percentage by dividing the distance from the terminal annulus to the marginal edge by the distance between the last 2 annuli formed on the otolith and multiplying by 100. Monthly marginal-increment statistics (25th, 50th, and 75th percentiles) with all age classes pooled were calculated for February 1995-December 1998, the period during which monthly samples were collected consistently. Additionally, monthly marginal-increment statistics were plotted, with months pooled across all years (1995-1998), for individual age classes (ages 1-6 only). Fish age 7 and older were excluded from these age-class-specific analyses because of low sample size across sampled months.
Age of each sheepshead was calculated on the basis of annulus count, marginal increment, date of capture, and an assumed hatching date of 1 April (an assumption based on spawning and larval recruitment; Parsons and Peters, 1989; Tucker and Alshuth, 1997). Therefore, sheepshead collected in February and March that had recently formed an annulus, as determined by a low (<30%) marginal increment were assigned an age of one less than the ring count. Fish collected in April, May, or June that were about to deposit an annulus (at >80% marginal increment) were assigned an age of one more than the ring count. All other fish were assigned an age equal to the ring count. Daily age was calculated on the basis of the age and the number of days that had passed between 1 April and the date of collection:
(integer age + number of days)/365. (1)
The G-test was used to compare sex ratios for all fish collected and subsets of fish kept for or eliminated from the aging analysis. Lengths of retained and eliminated sheepshead were compared by using the KS 2-sample test (Proc Npar1way procedure; SAS, 2006). The KS test was also used to compare age-frequency distributions between the sexes.
The von Bertalanffy (1957) growth equation,
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)
where [L.sub.t] = the observed FL at time t;
[L.sub.[infinity]] = the asymptotic FL;
k = the growth coefficient;
t = the observed age; and
[t.sub.0] = the hypothetical age at size zero, was fit by nonlinear regression (Proc NLin procedure, Marquardt routine in SAS software) for sex-specific observed age and length data.
Growth models for males and females were compared with an approximate randomization test (Helser, 1996).
Size and sex composition
Sheepshead (n=2549) ranging in size from 107 to 524 mm FL were collected in Tampa Bay (Fig. 1) with a variety of gear types (Table 1). Although sampling was done throughout the estuary, most specimens were collected along the shoreline in the middle to lower portions of the estuary with a large haul seine (n=1931, 75.8%) or a trammel net (n=367, 14.4%). Together, the other gear types caught less than 10% of the specimens used in this study.
Sex was determined for 93% of the sheepshead collected. The majority of the specimens for which sex was not determined were immature ([less than or equal to] 2 years old) fish for which gonad samples were too small to allow sex determination. The sex ratio of males to females (1:1.75) in our samples was significantly different from 1:1 (G-test: 177.69, df=1, P [less than or equal to] 0.001). Mean length of females (308.5 mm FL) was slightly greater than that of males (302.0 mm FL), but length-frequency distributions did not differ significantly between sexes (Fig. 2; KS test: 0.058, P [greater than or equal to] 0.05). Neither the slopes nor the intercepts differed significantly in the sex-specific length-length regressions (ANCOVA: P>0.05); therefore, all sheepshead data were pooled to elucidate relationships among SL, FL, and TL (Table 2). All length-length regressions exhibited high coefficients of determination ([r.sup.2] [greater than or equal to] 0.988) (Table 2). Sex-specific length--weight regressions were necessary because regressions for males and females had significantly different intercepts (ANCOVA: F= 32.15; df=l, 2196; P [less than or equal to] 0.001), but [r.sup.2] was high for both males ([greater than or equal to] 0.978) and females ([greater than or equal to] 0.976) (Table 2).
Age determination and validation
Marginal-increment analysis of otoliths from sheepshead, with all age classes pooled, indicated that a single opaque ring formed annually between May and June (Fig. 3). Median marginal increment reached a consistent minimum from late spring to early summer (May 1995, June 1996, June 1997, June 1998) and a consistent maximum during winter (February 1995, January 1996, February 1997, January 1998). Large interquartile ranges in the months before and during opaque-ring deposition indicated that many individuals had either just deposited (and therefore had a low increment width) or were about to deposit an opaque ring (and had a high increment width). Pooling monthly marginal increments across all years for individual age classes (ages 1-6) also indicated that for each age class a single opaque ring was deposited during the late spring or summer (Fig. 4).
Otoliths of 2549 sheepshead were examined for age; 154 (6.0%) were excluded from the aging analysis (because there was no agreement among readers or because an otolith was damaged), and 169 (6.6%) were excluded from sex-specific age analyses (because no sex data were available). The male to female sex ratio for the sheepshead retained in the aging analysis (1:1.79) did not differ significantly from that of the overall sample (1:1.75; G-test: 0.340, df=1, P>0.05). But the male-to-female sex ratio for sheepshead excluded from the aging analysis (1:1.20) was significantly different from that of sheepshead retained in the aging analyses (G-test: 5.06, df=1, P<0.05). Sex-specific length-frequency distributions of fish excluded from the analysis did not differ significantly from those retained (Fig. 2; KS test: females, 0.084, P>0.05; males, 0.102, P>0.05).
Age and growth
Sheepshead ranged from <1 year to 15 years of age, and the mean ages of males (3.67 years) and females (3.73 years) were similar (Fig. 5). The overall age-frequency distributions of males and females (Fig. 5) did not differ significantly (KS test: 0.032, P>0.05). Sheepshead of ages 2-4 accounted for more than half of the individuals collected (62.9%), but sheepshead aged 7 or older were relatively rare (7.9%). The oldest fish (sex not determined: 524 mm FL, 14.7 years; male: 404 mm FL, 14.9 years; and female: 345 mm FL; 15.2 years) were collected in a large haul seine.
Observed length at age was variable for both sexes (Fig. 6). Growth was relatively rapid for both males and females. By age 1, sheepshead, regardless of sex, reached a size predicted to be more than 40% of [L.sub.[infinity]] and, by age 6, they had reached sizes greater than 80% of [L.sub.[infinity]]. Growth rates of both sexes slowed after age 6. Males achieved a slightly greater [L.sub.[infinity]] than females (3.4 mm FL greater), but females grew at a slightly higher rate (as measured by K; Table 3, Fig. 6) than males. The von Bertalanffy growth models for males and females (approximate randomization test: P<0.01) were significantly different. Although predicted size at age was greater for females than for males in all age classes from ages 1 through 10 (Table 4), the difference between predicted size at age between sexes was minimal, 7 mm FL or less (mean difference of 3.3 mm FL) across all age classes.
Age determination and validation
Sheepshead age and growth has been studied by using both scales and sagittal otoliths. Although scales have been used to age sheepshead (Music and Pafford (2); Schwartz, 1990; and Wenner (3)), validation of annuli on scales of sheepshead has indicated that scales are not as reliable as otoliths for aging this species. Music and Pafford (2) could validate scale annuli only in sheepshead younger than age 5, and annuli in scales of sheepshead older than age 2 have been reported to be unreadable (Schwartz, 1990; Wenner (3)). Age has been underestimated in sheepshead and other fish species when scales were used, and age estimates from the use of scales have been lower than those derived from otolith sections (Beamish and McFarlane, 1983; Carlander, 1987; Lowerre-Barbieri et al., 1994; Dutka-Gianelli, 1999). In our discussion, the only studies considered in growth comparisons are those in which ages were estimated on the basis of validated otolith annuli.
We used marginal-increment analysis, which has been used to validate annulus deposition in the sagittae of sheepshead from other areas. Studies from Louisiana (Beckman et al., 1991), northwest Florida (Dutka-Gianelli and Murie, 2001), and South Carolina (individual age classes, <age 5; Wenner (3)) used marginal-increment analysis to validate the deposition of a single annulus per year. Chemical marking with oxytetracycline validated the annual deposition of a single opaque ring in sheepshead of ages 2-3 (Dutka-Gianelli and Murie, 2001). Each of these studies reported that a single annulus was deposited from late winter to spring (March-May for sagittae)--a finding similar to that of our study (May-June).
In our study, fish older than age 7 were uncommon; therefore, we could not analyze marginal increments for those fish by age class to validate annulus deposition. Examination of fish aged 7-14, as a group, showed that annuli formed at the same time as fish in younger age classes. Although we presume that these older fish laid down a single opaque ring each year, ages of fish older than age 6 may sometimes have been misinterpreted (Beamish and McFarlane, 1983). It would be valuable in future studies in the Tampa Bay area that more age data be collected from larger and older sheepshead to further elucidate annulus deposition in these older fish.
Results of growth studies of sheepshead have shown marked geographic variation. Sheepshead in northern areas of the Gulf of Mexico and Atlantic are longer lived and grow larger than those in Florida (Beckman et al., 1991; Wenner (3); Dutka-Gianelli and Murie, 2001). We found that the growth of sheepshead in Tampa Bay was similar to that reported farther north along the gulf coast of Florida. Sheepshead from Tampa Bay were observed to reach at least 524 mm FL and to reach an estimated maximum age of 15 years. Dutka-Gianelli and Murie (2001) collected sheepshead in northwest Florida with a similar maximum age (15 years) and size (522 mm FL). In contrast, sheepshead from Louisiana (Beckman et al., 1991) lived longer (20 years) and grew larger (563 mm FL) than sheepshead collected in Florida (Dutka-Gianelli and Murie, 2001; our study). Specimens collected in South Carolina (Wenner (3)) included the greatest reported estimated age for sheepshead (26 years). Wenner (3) also reported a sheepshead maximum size (560 mm FL) similar to that reported for sheepshead from Louisiana.
Fish length is a poor measure for estimating age in sheepshead. Sheepshead of similar age can differ considerably in length (Schwartz, 1990; Beckman et al., 1991; Wenner (3); Dutka-Gianelli and Murie, 2001; our study). For example, in our study, age-5 sheephead ranged from 212 to 465 mm FL, and 350-mm-FL specimens ranged from age 2 to age 8. Length has also been seen as unreliable for estimating the age for other sparids, including red porgy (Pagrus pagrus; Hood and Johnson, 2000), black bream (Acanthopagrus butcheri; Sarre and Potter, 2000), pinfish (Lagodon rhomboids; Nelson, 2002), and littlehead porgy (Calamus proridens; Tyler-Jedlund and Torres, 2015).
Although growth models for male and female sheepshead from Tampa Bay differed statistically, the actual growth parameters were biologically very similar for the sexes. Sex-specific growth models had a significantly better fit for sheepshead from Louisiana waters than a model in which sexes were combined. Predicted sizes at age for Louisiana sheepshead were similar for both sexes through age 6, with a mean difference in size at age of 11.6 mm FL between the sexes (Table 4), but this difference in size was greater at older ages (25.8 mm difference in FL for ages 7-20). Despite finding no significant difference in sex-specific growth for sheepshead from northwest Florida, both sex-specific and combined-sex growth models were presented by Dutka-Gianelli and Murie (2001); mean differences in predicted size at age between sexes (11.2 mm FL) were larger than the differences we found (mean differences of only 3.3 mm FL; Table 4). Therefore, evidence of growth differences between males and females has often been varied among previous studies. For recent stock assessments of sheepshead in Florida, growth was assumed to be similar for males and females, but coast-specific growth parameters were used because growth varied significantly between the two regions (Munyandorero et al. (1)).
Our estimates of [L.sub.[infinity]] and the [t.sub.0] (Table 3) were within 2 standard errors of those estimated for fish from Louisiana waters (males: [L.sub.[infinity]]=419, [t.sub.0]= -0.901; females: [L.sub.[infinity]]=447, [t.sub.0]= -1.025; Beckman et al., 1991; Table 4). Estimated [L.sub.[infinity]] for sheepshead from South Carolina (505.0
mm FL) and northwest Florida (490.4 mm FL) were greater than those for sheepshead from either Tampa Bay or Louisiana. For sheepshead from South Carolina and northwest Florida and in our study, values of k were similar but smaller than those reported for sheepshead in Louisiana, indicating that Louisiana sheepshead reach [L.sub.[infinity]] more rapidly. Beckman et al. (1991) found predicted lengths for age-3 Louisiana male and female sheepshead were 80% and 77% of their [L.sub.[infinity]], respectively. We found that sheepshead in Tampa Bay are not predicted to reach 80% of [L.sub.[infinity]] until age 5 for females (81%) and age 6 for males (83.8%). Similarly, in studies conducted in South Carolina (Wenner (3)) and northwest Florida (Dutka-Gianelli and Murie, 2001), sheepshead were predicted to reach 80% of [L.sub.[infinity]] by age 5 and age 6, respectively. The differences in growth parameters between these latter 2 studies may be attributed to differences in sampling methods, ontogenetic habitat shifts, or estuarine-specific differences in growth or mortality. Furthermore, the variation in growth parameters between these studies may also be affected by variability in genetic composition (subspecies) among the regions where these studies were conducted.
Reliance on fishery-dependent samples can introduce size- and age-related biases that can result in misleading interpretations of fish growth and size distribution, and age structure of a population (Langler, 1978; Miranda et al., 1987; Hilborn and Walters, 1992). All the sheepshead analyzed in a Louisiana study were collected from commercial and recreational catches. Almost 60% of their fish came from catches with gill nets, which tend to be size selective, often resulting in a narrow size range of collected fish (Pope et al., 1975). Beckman et al. (1991) indicated that their sample of sheepshead was probably not representative of the Louisiana population because of gear selectivity and sorting of catches before sampling. In a South Carolina study, most sheepshead larger than 300 mm FL were caught in recreational fishing tournaments and probably represented a greater percentage of larger fish than that of the overall population (Wenner (3)). Consequently, the sampling methods of these studies could have introduced sufficient size-at-age bias to the effect that the sampled fish did not represent the population as a whole. Our fisheries-independent sampling design increased the likelihood that our data would approximately represent the size and age structure of the population of sheepshead in Tampa Bay. Specimens were collected by using a variety of gear types (a majority of specimens with the nonselective large haul seine [77%]), and at randomly selected sites (68%) that represented a variety of habitats.
Adult sheepshead have been reported to occur over hard structure (reefs, jetties, and piers) in both estuarine (Johnson, 1978; Ogburn, 1984) and offshore (Sedberry and van Dolah, 1984) waters. Sheepshead also have been reported to undergo an ontogenetic shift in habitat as juveniles (Hildebrand and Cable, 1938; Johnson, 1978), moving from shallow nursery habitats, which often include sea grasses, to hard-structure habitats of adults. Sheepshead in our study were collected from relatively shallow waters (mean depth: 1.2 m [standard error 0.02]) in the Tampa Bay estuary, whereas portions of sheepshead in the Louisiana and northwest Florida studies came from deeper, offshore areas. Render and Wilson (1992) found no significant differences in size or age between the sheepshead collected in offshore and inshore waters of Louisiana, but sample size and gear selectivity (i.e., similar among areas) may have obscured differences. If sheepshead undergo a habitat shift from shallow nearshore waters into deeper waters, the growth parameters we describe might not indicate the age and growth of sheepshead in the deeper habitats.
Regional differences in sheepshead growth parameters may be attributed partly to population genetics. Within the U.S. range of this species, 2 subspecies of sheepshead have been reported (Caldwell, 1965). We analyzed A. p. probatocephalus, which occurs along the Atlantic coast and into the Gulf of Mexico to Steinhatchee, Florida. A Louisiana study (Beckman et al., 1991) considered A. p. oviceps, which occurs in the Gulf of Mexico from St. Marks River, Florida, to Campeche Bank, Mexico. A study from northwest Florida (DutkaGianelli and Murie, 2001) looked at both subspecies (84% A. a. probatocephalus and 11% A. a. oviceps) and found no significant differences in growth. Anderson et al. (2008) concluded that molecular genetic data indicated a very limited genetic subdivision between the subspecies, despite considerable divergence in some morphological characters.
In contrast, recent analyses of 24 species-specific microsatellite DNA loci of both A. probatocephalus subspecies from the Atlantic (Florida to North Carolina) and the Gulf of Mexico (Florida and Texas), showed a genetic break at the site of the subspecies boundary at Apalachee Bay, Florida (Seyoum et al., in press). These recent genetics results, coupled with the known morphological differences between A. p. probatocaphalus and A. p. oviceps, support the validity of the 2 subspecies of sheepshead within its U.S. range (Caldwell, 1965), but further study is necessary to better understand processes that contribute to the genetic and morphological differences between these subspecies. A comparison of similarly collected age and growth data is necessary to determine the existence and extent of any subspecific differences in growth.
A fishery can modify the population size and age characteristics of a species by selectively removing younger, faster-growing fish (Ricker, 1975), possibly accounting for the larger fish collected in northwest Florida. Dutka-Gianelli and Murie (2001) suggested that because of the lower density of the human population of northwest Florida, sheepshead there may have experienced less long-term fishing mortality than those in Tampa Bay. Sheepshead enter the fishery in Florida waters at ~280 mm FL (~305 mm TL), at approximately the size predicted for age-3 sheepshead in Tampa Bay (Table 4). Predicted sizes at age are similar between sheepshead in Tampa Bay and those in northwest Florida through age 3 (mean difference of 13.2 mm FL), but after that age, sheepshead from northwest Florida consistently are predicted to attain larger sizes at age (mean difference of 52.4 mm FL).
Regional differences in sheepshead growth parameters are apparent, but within Florida waters it is unnecessary to manage sheepshead regionally. Several fishery management actions, including the ban on entangling gear, a minimum size limit, and recreational-bag (15 fish) and commercial-possession (50 fish) limits were enacted for sheepshead in Florida waters during the 1990s. These actions have brought about a decrease in combined landings of sheepshead and an increase in the size of sheepshead landed; transitional spawning potential ratios of sheepshead in Florida have increased since 1996 and, in 2009, were 37% and 29% for the Atlantic and gulf coasts of Florida, respectively (Munyandorero et al. (1)). Further studies, to better define the stock structure and to describe estuary- or stock-specific differences in growth, would be beneficial and help refine the management of sheepshead in Florida waters.
Manuscript submitted 28 April 2016.
Manuscript accepted 15 December 2016.
Fish. Bull. 115:155-166 (2017).
Online publication date: 31 January 2017.
The views and opinions expressed or implied in this article are those of the author (or authors) and do not necessarily reflect the position of the National Marine Fisheries Service, NOAA.
We thank staff of the Florida Fish and Wildlife Conservation Commission's Fisheries-Independent Monitoring program and its Age and Growth Lab for aiding with sample collection and processing and the preparation and reading of otoliths, D. Harshany for measuring marginal increments, and R. Crabtree and M. Murphy for scientific expertise. We also are grateful to D. Leffler, T. Tsou, A. Acosta, R. Taylor, M. Murphy, R. McMichael, J. Quinn, J. Leiby, and B. Crowder for their critical reviews that greatly improved this manuscript. This work was supported in part by funding from the U.S. Fish and Wildlife Service under Federal Aid for Sportfish Restoration Project Number F-43 as well as from Florida's saltwater fishing licenses. The statements, findings, views, conclusions, and recommendations contained in this document are those of the authors and do not necessarily reflect the views of the U.S. Department of the Interior and should not be interpreted as representing the opinions or policies of the U.S. government.
Anderson, J. D., W. J. Karel, K. A. Anderson, and P. A. Roper-Foo. 2008. Genetic assessment of sheepshead stock structure in the northern Gulf of Mexico: morphological divergence in the face of gene flow. North Am. J. Fish. Manage. 28:592-606.
Bagenal, T. B., and F. W. Tesch. 1978. Age and growth. In IBP handbook no. 3: methods for assessment of fish production in freshwater, 3rd ed. (T. B. Bagenal, ed.), p. 101-136. Blackwell Scientific Publications, Oxford, UK.
Beamish, R. J., and G. A. McFarlane. 1983. The forgotten requirement for age validation in fisheries biology. Trans. Am. Fish. Soc. 112:735-743.
Beckman, D. W., A. L. Stanley, J. H. Render, and C. A. Wilson. 1991. Age and growth-rate estimation of sheepshead Archosargus probatocephalus in Louisiana waters using otoliths. Fish. Bull. 89:1-8.
Bigelow, H. B., and W. C. Schroeder. 1953. Fishes of the Gulf of Maine. Fish. Bull. 53:1-577.
Caldwell, D. K. 1965. Systematics and variation in the sparid fish Archosargus probatocephalus. Bull. South. Calif. Acad. Sci. 64:89-100.
Carlander, K. D. 1987. A history of scale age and growth studies of North American freshwater fish. In Age and growth of fish (R. C. Summerfelt and G. E. Hall, eds.), p. 3-14. Iowa State Univ. Press, Ames, IA.
Collette, B. B., and G. Klein-MacPhee (eds.). 2002. Bigelow and Schroeder's fishes of the Gulf of Maine, 3rd ed., 748 p. Smithsonian Inst. Press, Washington, D.C.
Comp, G. S., and W. Seaman Jr. 1985. Estuarine habitat and fishery resources of Florida. In Florida aquatic habitat and fishery resources (W. Seaman Jr., ed.), p. 337-435. Fla. Chapter Am. Fish. Soc., Eustis, FL.
Dutka-Gianelli, J. 1999. Comparative age and growth of sheepshead, Archosargus probatocephalus (Walbaum 1972) (Pisces: Sparidae), from the northwestern coast of Florida. M.S. thesis, 68 p. Univ. Florida, Gainesville, FL.
Dutka-Gianelli, J., and D. J. Murie. 2001. Age and growth of sheepshead, Archosargus probatocephalus (Pisces: Sparidae), from the northwest coast of Florida. Bull. Mar. Sci. 68:69-83.
Gilhen, J., C., G. Grunchy, and D. E. McAllister. 1976. The sheepshead, Archosargus probatocephalus, and the feather blenny, Hypsoblennius hentzi, two additions to the Canadian Atlantic ichthyofauna. Can. Field-Nat. 90:42-46.
Helser, T. E. 1996. Growth of silver hake within the U.S. continental shelf ecosystem of the northwest Atlantic Ocean. J. Fish Biol. 48:1059-1073.
Hilborn, R., and C. J. Walters. 1992. Quantitative fisheries stock assessment: choice, dynamics and uncertainty, 570 p. Routledge, Chapman and Hall Inc., New York.
Hildebrand, S. F., and L. E. Cable. 1938. Further notes on the development and life history of some teleosts at Beaufort, N.C. Bull. Bur. Fish. 48:505-642.
Hood, P. B., and A. K. Johnson. 2000. Age, growth, mortality, and reproduction of red porgy, Pagrus pagrus, from the eastern Gulf of Mexico. Fish. Bull. 98:723-735.
Johnson, D. G. 1978. Development of fishes in the mid-Atlantic Bight IV: Carangidae through Ephippidae. U.S. Fish Wildl. Serv., FWS/OBS-78/12, 314 p.
Lowerre-Barbieri, S. K., M. E. Chittenden Jr., and C. M. Jones. 1994. A comparison of a validated otolith method to age weakfish, Cynoscion regalis, with the traditional scale method. Fish. Bull. 92:555-568.
Miranda, L. E., W. M. Wingo, R. J. Muncy, and T. D. Bates. 1987. Bias in growth estimates derived from fish collected by anglers. In Age and growth of fish (R. C. Summerfelt and G. E. Hall, eds.), p. 211-220. Iowa State Univ. Press, Ames, IA.
Nelson, G. A. 1998. Abundance, growth, and mortality of young-of-the year pinfish, Lagodon rhomboides, in three estuaries along the gulf coast of Florida. Fish. Bull. 96:315-328.
2002. Age, growth, mortality, and distribution of pinfish (Lagodon rhomboides) in Tampa Bay and adjacent Gulf of Mexico waters. Fish. Bull. 100:582-592.
Nelson, G. A., R. H. McMichael, T. C. MacDonald, and J. R. O'Hop. 1997. Fisheries monitoring and its uses in fisheries resources management. In Proceedings, Tampa Bay area scientific information symposium 3: applying our knowledge; Clearwater, FL; 21-23 October 1996 (S. F. Treat, ed.), p. 43-56. Tampa Bay Regional Planning Council, Clearwater, FL.
Ogburn, M. V. 1984. Feeding ecology and the role of algae in the diet of sheepshead Archosargus probatocephalus (Pisces: Sparidae) on two North Carolina jetties. M.S. thesis, 68 p. Univ. North Carolina, Wilmington, NC.
Parsons, G. R., and K. M. Peters. 1989. Age determination in larval and juvenile sheepshead, Archosargus probatocephalus. Fish. Bull. 87:985-988.
Pope, J. A., A. R. Margetts, J. M. Hamley, and E. F. Akyuz. 1975. Manual of methods for fish stock assessments. Part III. Selectivity of fishing gear. FAO Fish. Tech. Pap. 41, 65 p.
Render, J. H., and C. A. Wilson. 1992. Reproductive biology of sheepshead in the northern Gulf of Mexico. Trans. Am. Fish. Soc. 121:757-764.
Ricker, W. E. 1975. Computation and interpretation of biological statistics of fish populations. Bull. Fish. Res. Board Can. 191, 382 p.
Sarre, G. A., and I. C. Potter. 2000. Variation in age compositions and growth rates of Acanthopagrus butcheri (Sparidae) among estuaries: some possible contributing factors. Fish Bull. 98:785-799.
Schwartz, F. J. 1990. Length-weight, age and growth, and landings observations for sheepshead Archosargus probatocephalus from North Carolina. Fish. Bull. 88:829-832.
Sedberry, G. R., and R. F. van Dolah. 1984. Demersal fish assemblages associated with hard bottom habitat in the South Atlantic Bight of the U.S.A. Environ. Biol. Fish. 11:241-258.
Seyoum, S., R. S. McBride, C. Puchutulegui, J. Dutka-Gianelli, A. C. Alavarez, and K. Panzner. In press. Genetic population structure of a coastal marine fish (Sheepshead; Archosargus probatocephalus [Sparidae]) in the southeastern United States: multiple population clusters based on species-specific microsatellite markers. J. Mar. Sci.
Snedecor, G. W., and W. G. Cochran. 1967. Statistical methods, 593 p. Iowa State Univ. Press, Ames, IA.
Sokal, R. R., and F. J. Rohlf. 1981. Biometry: the principles and practice of statistics in biological research, 2nd ed., 859 p. W.H. Freeman and Company, New York.
Tremain, D. M., and D. H. Adams. 1995. Seasonal variations in species diversity, abundance, and composition of fish communities in the northern Indian River Lagoon, Florida. Bull. Mar. Sci. 57:171-192.
Tucker, J. W., Jr., and S. R. Alshuth. 1997. Development of laboratory-reared sheepshead, Archosargus probatocephalus (Pisces: Sparidae). Fish. Bull. 95:394-401.
Tyler-Jedlund, A. J., and J. J. Torres. 2015. Age, growth, and reproduction of the littlehead porgy, Calamus proridens, from the eastern Gulf of Mexico. Bull. Mar. Sci. 91:101-123.
von Bertalanffy, L. 1957. Quantitative laws in metabolism and growth. Q. Rev. Biol. 32:217-231.
Wilson, K. L., B. G. Matthias, A. B. Barbour, R. N. M. Ahrens, T. Tuten, and M. S. Allen. 2015. Combining samples from multiple gears helps to avoid fishy growth curves. North Am. J. Fish. Manage. 35:1121-1131.
Winner, B. L., D. A. Blewett, R. H. McMichael Jr., and C. B. Guenther. 2010. Relative abundance and distribution of common snook along shoreline habitats of Florida estuaries. Trans. Am. Fish. Soc. 139:62-79.
(1) Munyandorero, J., J. O'Hop, and C. Guenther. 2011. An assessment of the status of sheepshead in Florida waters through 2009. Florida Fish Wildl. Conserv. Comm., Fish Wildl. Res. Inst., IHR 2011-003, 137 p. Fish and Wildlife Research Institute, St. Petersburg, FL. [Available from website.]
(2) Music, J. L., Jr., and J. M. Pafford. 1984. Population dynamics and life history aspects of major marine sportfishes in Georgia's coastal waters, 382 p. Coast Res. Div., Georgia Dep. Nat. Resour., Atlanta GA.
(3) Wenner, C. 1996. Age and growth of sheepshead, Archo sargus probatocephalus, from South Carolina waters with some preliminary management concepts, 17 p. S. Carolina Dep. Nat. Resour., Charleston, SC.
(4) Mention of trade names or commercial companies is for identification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA.
Brent L. Winner (contact author) 
Timothy C. MacDonald 
Kimberly B. Amendola 
Email address for contact author: firstname.lastname@example.org
 Fish and Wildlife Research Institute Florida Fish and Wildlife Conservation Commission 100 Eighth Avenue Southeast St. Petersburg, Florida 33701-5020
 Southeast Regional Office
National Marine Fisheries Service, NOAA 263 13th Avenue South, Suite 113 St. Petersburg, Florida 33701
Caption: Figure 1
Sampling locations (indicated by black circles) in Tampa Bay, Florida, where sheepshead (Archosargus probatocephalus) were collected during 1993-2009 for age and growth analysis under the guidance of the Fisheries-Independent Monitoring program of the Florida Fish and Wildlife Conservation Commission.
Caption: Figure 2
Length-frequency distributions for female and male sheepshead (Archosargus probatocephalus) collected in Tampa Bay, Florida, 1993-2009. Sheepshead that were retained and excluded (otolith identified as unreadable) from the age and growth analysis are depicted. Sheepshead for which there was both age and sex information but which did not have a measured fork length were excluded from this plot
Caption: Figure 3
Median monthly percent marginal increment (indicated by black points) for otoliths of sheepshead (Archosargus probatocephalus), all ages combined, collected in Tampa Bay, Florida, 1995-1998. Vertical lines indicate the interquartile ranges. Numbers above monthly percent marginal increments indicate sample sizes. Months of minimal marginal increment (indicating annuli deposition) are highlighted in gray boxes.
Caption: Figure 5
Age-frequency distributions for sheepshead (Archosargus probatocephalus) collected in Tampa Bay, Florida, 1993-2009.
Caption: Figure 6
Observed fork lengths and estimated ages of female (indicated by black circles) and male (indicated by open triangles) sheepshead (Archosargus probatocephalus) collected in Tampa Bay, Florida, 1993-2009. The lines depict the predicted size at age from the von Bertalanffy growth models for males and females. Von Bertalanffy growth parameters for both sexes are presented in Table 3.
Table 1 Descriptive statistics for fork length and age (mean, minimum, maximum, and standard deviation [SD]) and total catch of sheepshead (Archosargus probatocephalus) collected in Tampa Bay, Florida, in 1993-2009, by gear type. Fork length (mm) Gear type Number of fish Mean Min Max SD Small haul seines 52 287.0 153 415 65.2 Large haul seines 1931 299.4 107 524 63.7 Purse seines 50 256.1 173 465 62.3 Gill nets 62 284.2 158 409 59.3 Otter trawls 36 254.6 159 383 55.7 Trammel nets 367 322.9 146 458 48.3 Unknown 51 318.2 190 433 52.8 Total catch 2549 Age (years) Gear type Mean Min Max SD Small haul seines 3.5 0.9 6.9 1.6 Large haul seines 4.1 0.5 15.2 2.1 Purse seines 3.1 1.2 7.9 1.6 Gill nets 3.7 1.0 9.6 2.0 Otter trawls 3.4 1.3 11.6 2.3 Trammel nets 4.5 0.6 11.5 1.8 Unknown 4.8 1.3 10.5 1.7 Total catch Table 2 Length-length and length-weight regressions for sheepshead (Archosargus pro-batocephalus) collected in Tampa Bay, Florida, 1993-2009. Measurements include standard length (SL) in millimeters, fork length (FL) in millimeters, total length (TL) in millimeters, and total weight (WT) in grams. Values in parentheses are standard errors. Sex-specific length-weight regressions were necessary because male and female regressions had significantly different intercepts. [r.sup.2]=coefficient of determination. Y = a + bX Y X n a TL FL 2218 0.853 (0.491) TL SL 2344 12.676 (0.742) FL TL 2218 0.621 (0.448) FL SL 2218 11.318 (0.629) SL FL 2218 -7.559 (0.582) SL TL 2344 -7.178 (0.627) [Log.sub.10] (WT), [Log.sub.10] (FL) 1406 -4.508 females (0.031) [Log.sub.10](WT), [Log.sub.10] (FL) 792 -4.367 males (0.038) Y = a + bX Y b [r.sup.2] TL 1.094 0.995 (0.002) TL 1.215 0.988 (0.003) FL 0.910 0.995 (0.001) FL 1.109 0.990 (0.002) SL 0.893 0.990 (0.002) SL 0.813 (0.002) 0.988 [Log.sub.10] (WT), 2.960 0.976 females (0.013) [Log.sub.10](WT), 2.899 0.978 males (0.015) Table 3 Estimates of the von Bertalanffy growth parameters, by sex, and with sexes combined: asymptotic length ([L.sub.[infinity]]), growth coefficient (k), and hypothetical age at size zero ([t.sub.0]) for sheepshead (Archosargus probatocephalus) collected in Tampa Bay, Florida, 1993-2009. Sample sizes (n) and asymptotic standard errors (in parentheses) are listed. Combined includes female and male sheepshead, as well as sheepshead for which sex was not determined (71=169). Females Males [L.sub.[infinity](mm) 419.1 (7.206) 422.5 (9.948) K 0.272 (0.019) 0.255 (0.023) [t.sub.0] -1.099 (0.162) -1.115 (0.205) n 1429 797 Combined [L.sub.[infinity](mm) 418.7 (5.309) K 0.273 (0.014) [t.sub.0] -0.981 (0.107) n 2395 Table 4 Estimates of the von Bertalanffy growth parameters, the asymptotic length ([L.sub.[infinity]]), growth coefficient (k), and hypothetical age at size zero ([t.sub.0]), and the predicted size at age, presented in fork length (FL) in millimeters, for sheepshead (Archosargus probatocephalus) collected from in Tampa Bay, Florida, 1993-2009 (this study); Louisiana, 1987-1988 (Beckman et al., 1991); northwest Florida, 1997-1998 (Dutka-Gianelli and Murie, 2001); and South Carolina, 1995-1996 (Wenner (3)). Tampa Bay Louisiana Growth parameters Females Males Females Males [L.sub.[infinity]] 419.1 422.5 447.0 419.0 k 0.272 0.255 0.367 0.417 [t.sub.0] -1.099 -1.115 -1.025 -0.901 Predicted size at ages (mm FL) 0 109 105 1 183 177 2 239 232 300 294 3 282 275 345 337 4 315 308 376 365 5 340 334 398 383 6 359 354 413 395 7 373 370 423 403 8 384 382 431 409 9 393 391 436 412 10 399 398 439 415 11 404 404 442 416 12 408 408 443 417 13 411 411 444 418 14 413 414 445 418 15 414 416 446 418 16 446 419 17 446 419 18 447 419 19 447 419 20 447 419 21 22 23 24 25 26 Northwest Florida South Carolina (1,2) Growth parameters Combined Females Males Combined [L.sub.[infinity]] 490.4 475.7 509.2 505.0 k 0.260 0.280 0.230 0.290 [t.sub.0] -0.420 -0.460 -0.520 -1.109 Predicted size at ages (mm FL) 0 139 1 151 160 150 231 2 229 237 224 299 3 289 295 283 351 4 335 339 329 389 5 371 373 366 418 6 398 398 396 440 7 419 417 419 456 8 435 431 437 469 9 448 442 452 478 10 458 450 464 485 11 465 456 473 490 12 471 461 481 494 13 475 486 497 14 479 491 499 15 482 495 500 16 502 17 503 18 503 19 504 20 504 21 504 22 505 23 505 24 505 25 505 26 505 (1) For the growth parameters, [L.sub.[infinity]] originally presented as total length (TL; 559 mm); converted to FL, for comparison, by using TL-FL relationship given by Wenner (3). (2) For predicted sizes at age calculated as TL ([L.sub.[infinity]]=559 mm) and converted to FL, for comparison, by using TL-FL relationship given by Wenner (3). Figure 4 Median monthly percent marginal increment (indicated by black points) for otoliths of sheepshead (Archosargus probatocephalus) collected in Tampa Bay, Florida, 1995-1998, by individual age classes: (A-F) ages 1-6. Vertical lines represent the interquartile ranges, and numbers above percent marginal increments represent the sample sizes. Months of minimal marginal increment (indicating annuli deposition) are highlighted in gray boxes. Month A J 17 F 14 M 21 A 19 M 13 J 6 J 4 A 6 S 10 O 17 N 11 D 9 B J 28 F 20 M 19 A 22 M 37 J 18 J 25 A 6 S 28 O 33 N 36 D 22 C J 22 F 18 M 11 A 37 M 23 J 22 J 18 A 11 S 20 O 52 N 46 D 23 D J 19 F 15 M 13 A 21 M 20 J 12 J 12 A 6 S 19 O 41 N 38 D 30 E J 30 F 12 M 5 A 11 M 9 J 2 J 14 A 7 S 9 O 24 N 42 D 11 F J 8 F 2 M 4 A 6 M 5 J 3 J 6 A 8 S 5 O 21 N 15 D 2 Note: Table made from bar graph.
|Printer friendly Cite/link Email Feedback|
|Author:||Winner, Brent L.; MacDonald, Timothy C.; Amendola, Kimberly B.|
|Date:||Apr 1, 2017|
|Previous Article:||Use of gill nets and telemetry in tracking movements and feeding of striped bass (Morone saxatilis), bluefish (Pomatomus saltatrix), and weakfish...|
|Next Article:||Feeding habits and dietary overlap of age-0 winter flounder (Pseudopleuronectes americanus) and summer flounder (Paralichthys dentatus) in southern...|