CONTENTS Life History Fisheries Biological Reference Points Fisheries Dependent Data Fishery Independent Data SCALE Model Input Model Results Projections Conclusion Literature Cited
Northeast Fisheries Science Center Reference Document 10-13
Black Sea Bass 2010 Stock Assessment UpdateGary R. Shepherd and Julie Nieland
NOAA Fisheries, Northeast Fisheries Science Center, Population Dynamics Branch, 166 Water Street, Woods Hole MA 02543
Web version posted August 2, 2010Citation: Shepherd GR, Nieland J. 2010. Black sea bass 2010 stock assessment update. US Dept Commer, Northeast Fish Sci Cent Ref Doc. 10-13; 25 p. Available from: National Marine Fisheries Service, 166 Water Street, Woods Hole, MA 02543-1026, or online at http://nefsc.noaa.gov/nefsc/publications/
Information Quality Act Compliance: In accordance with section 515 of Public Law 106-554, the Northeast Fisheries Science Center completed both technical and policy reviews for this report. These predissemination reviews are on file at the NEFSC Editorial Office.
The northern stock of black sea bass (Centropristis striata) was evaluated using length-based population models. This new assessment approach was accepted by the Data Poor Workshop review panel (NEFSC 2009) and involved estimates of fishing mortality and population size determined from changes in size composition of the population. The panel accepted the results using a constant M=0.4 as the preferred model. The resulting F40% , as a proxy for FMSY, was equal to 0.42 with an associated SSB equal to 12,537 mt and MSY at 3,903 mt. The 2009 average F, resulting from a catch of 2,086 mt, was estimated at 0.29 and SSB at 12,978 mt. Therefore, the conclusion is that fishing mortality is below FMSY and the stock is above the optimal level of spawning biomass. The reference points and current stock status should be used with caution due to the uncertainty in the natural mortality estimate and the model input parameters, as well the uncertainty associated with management of a protogynous species.
Black sea bass (Centropristis striata) are distributed from the Gulf of Maine to the Gulf of Mexico, however, fish north of Cape Hatteras, NC are considered part of a single fishery management unit. Sea bass are generally considered structure oriented, preferring live-bottom and reef habitats. Within the stock area, distribution changes on a seasonal basis and the extent of the seasonal change varies by location. In the northern end of the range (New York to Massachusetts), sea bass move offshore crossing the continental shelf, then south along the edge of the shelf (Musick and Mercer 1977, Moser and Shepherd 2009). By late winter, northern fish may travel as far south as Virginia, however most return to the northern inshore areas by May. Sea bass originating inshore along the Mid-Atlantic (New Jersey to Maryland) head offshore to the shelf edge during late autumn, travelling in a southeasterly direction. They also return inshore in spring to the general area from which they originated. Black sea bass in the southern end of the stock (Virginia and North Carolina) move offshore in late autumn/early winter. Given the proximity of the shelf edge, they transit a relatively short distance, due east, to reach over-wintering areas. Fisheries also change seasonally with changes in distribution. Inshore commercial fisheries are prosecuted primarily with fish pots (baited and unbaited) and handlines. Recreational fisheries generally occur during the period that sea bass are inshore. Once fish move offshore in the winter, they are caught in a trawl fishery targeting summer flounder, scup and Loligo squid (Shepherd and Terceiro, 1994). Handline and pot fisheries in the southern areas may still operate during this offshore period.
Black sea bass are protogynous hermaphrodites and can be categorized as temperate reef fishes (Steimle et al. 1999, Drohan et al. 2007). Transition from female to male generally occurs between the ages of two and five (Lavenda 1949, Mercer 1978). Based on sex ratio at length from NMFS surveys, males constitute approximately 30% of the population by 20 cm, with increasing proportions of males with size (Figure 1). Following transition from female to male, sea bass can follow one of two behavioral pathways; either becoming a dominant male, characterized by a larger size and a bright blue nuccal hump during spawning season, or subordinate males which have few distinguishing features. The initiation of a transition appears to be based on visual rather than chemical cues (Dr. David Berlinsky, UNH, Personal communication). In studies of protogny among several coral reef fishes, transition of the largest female to male may occur quickly if the dominate male is removed from the reef; however, similar studies have not been published for black sea bass.
Spawning in the Middle Atlantic peaks during spring (May and June) when the fish reside in coastal waters (Drohan et al. 2007). The social structure of the spawning aggregations is poorly known although some observations suggest that large dominant males gather a harem of females and aggressively defend territory during spawning season (Nelson et al. 2003). The bright coloration of males during spawning season suggests that visual cues may be important in structuring of the social hierarchy.
Black sea bass attain a maximum size around 60 cm and 4 kg. Although age information is limited for the northern stock of black sea bass, growth curves are available from one published study as well as several unpublished studies. Lavenda (1949) suggests a maximum age for females of 8 and age 12 for males. However it was noted that large males (>45 cm) were present in deeper water and may have been older. Size at age information and Von Bertalanffy curve parameters used as input to model are shown in Table 1. Although growth information is available for use in models, annual age length keys are not, therefore sea bass modeling efforts are length based rather than age based.
Maturity data is routinely collected on Northeast Fisheries Science Center survey cruises. Proportion mature for all years and sexes combined (n=10,318) was fitted to a logistic model. The model estimate for length at 50% maturity was 20.4 cm and 95% maturity is attained by 28 cm (Figure 2).
In the Northwest Atlantic, black sea bass support commercial and recreational fisheries. Prior to WWII in 1939 and 1940, 46-48% of the landings were in New England, primarily in Massachusetts. After 1940 the center of the fishery shifted south to New York, New Jersey and Virginia. Landings increased to a peak in 1952 at 9,883 mt with the bulk of the landings from otter trawls, then declined steadily reaching a low point in 1971 of 566 mt (Table 2). Historically, trawl fisheries for sea bass have focused on the over-wintering areas near the shelf edge. Inshore pot fisheries, which were primarily in New Jersey, showed a similar downward trend in landings between the peak in 1952 and the late 60s. The large increase in landings during the 1950’s appears to be the result of increased landings from otter trawlers, particularly from New York, New Jersey and Virginia. During the same period, a large increase in fish pot effort, and subsequent landings, occurred in New Jersey. In recent years, fish pots and otter trawls account for the majority of commercial landings with increasing contributions from handline fisheries. Landings since 1974 have remained relatively steady around 1,400 mt. (Table 2). Recreational landings, available from MFRSS data since 1981, average about 1,600 mt annually (Table 2). Estimates for recreational sea bass landings in1982 and 1986 (4,500 mt and 5,621 mt, respectively) are unusually high as they are for other species for those years. Similarly, recreational landings for 1998 and 1999 are lower than expected. Although the estimates have been confirmed by MRFSS, they remain suspect.
The species affinity for bottom structure during its seasonal period of inshore residency increases the availability to hook and line or trap fisheries compared to the decreasing susceptibility to bottom trawl gear commonly used for scientific surveys. In autumn when water temperatures decline, black sea bass migrate offshore to areas along the edge of the continental shelf. During this offshore period, sea bass are vulnerable to otter trawl gear as part of a multispecies fishery (Shepherd and Terceiro 1994).
Development of biological reference points for black sea bass is hampered not only by a lack of annual age data but also by limited understanding of how black sea bass productivity responds to exploitation. Traditional fisheries models, generally developed for gonochoristic species, may not apply to a protogynous hermaphrodite (Hamilton et al 2007). Simulation studies of populations exhibiting protogny suggest that conservation of large terminal males is critical for sustainability (Alzono et al. 2008, Brooks et al. 2008, Hamilton et al. 2007, Heppell et al. 2006, Huntsman and Schaaf 1994). The implication is that removal of the terminal male will not only hamper male fertilization success but will induce transitioning of the larger females into males. The consequence is not only removal of male biomass but removal of potential egg production in the larger females. Reduction of dominant males in a population may, in effect, have a similar effect as increasing natural mortality on females.
Development of updated biological reference points for black sea bass was addressed at the NEFSC Data Poor Workshop held in December 2008. The panel recommended abandoning the previous index-based reference points in favor of the metrics based on the results of a statistical catch at length model. As part of the revised model, the natural mortality of the stock was assumed equal to 0.4 rather than the previously used 0.2. The Panel recommended an FMSY proxy of F40%, estimated with the new model to be F=0.42, and a related SSBMSY proxy of 12,537 mt.
Since annual age information was unavailable, a length based model (SCALE developed by Paul Nitschke of the NEFSC) was used as the method for evaluating sea bass status (NEFSC 2009). SCALE data input includes catch history, survey indices, recruitment indices, growth information, survey length frequencies and catch length frequencies (Figure 3). The updated model covers the period 1968 to 2009 based on the times series of NEFSC spring offshore surveys.
Commercial length frequencies were compiled beginning with samples in 1984. Sampling was done randomly by market categories and expanded as the ratio of sample weight to total landings, by calendar quarter. Black sea bass were culled as small, medium, large, jumbo or unclassified. In the rare cases where fish were categorized as extra small and extra large, they were combined with small and large, respectively. Total annual length measurements have ranged from 300 to 7,768 with an average of 3,334 per year (Table 3). During 2009 5,650 measurements were used for the commercial length frequency.
Annual commercial landings were determined from the NEFSC and state data. In recent years, the predominant gear type for back sea bass has been fish pots, otter trawls and handline. Landings between 1996 and 2006 averaged 1,326 mt, however following quota reductions landings in 2008 declined to 854 mt and continued in 2009 to 536 mt.
Commercial discards for otter trawls were estimated between 1989 and 2008 using a standard approach developed for national standardized by-catch reporting (Wigley et al., 2008). Discard estimates were developed from the ratio of discarded black sea bass in mt to total landings (mt) of all fish species in the comparable statistical area. Final data was not available for 2009, therefore the series average was assumed for trawl discards. Pot and handline discards were estimated using the ratio of reported discards to landings in vessel trip reports, expanded to total annual landings. Since a component of the pot fishery is solely in state waters and not required to submit VTR logs, they are not included in the total. A 75% survival rate was applied across all pot and handline commercial gears and 50% for otter trawl. Total discards averaged 113 mt annually and represented 17% of reported commercial landings (Table 2). Discards in 1993 and 2004 were well above average at 18 and 31% of landings, respectively. The total commercial discards estimated for 2009 were 113 mt for all gear types based on the time series average.
Complete recreational landings north of Cape Hatteras were available since 1981. Landings in 2009 were 1,049 mt, an increase from 713 mt in 2008. Landings were hindcast to 1968 using the relationship between commercial pot and handline landings with recreational landings between 1981 and 1997 (Table 2). In 1998 management regulations were imposed which controlled landings based on quota. The two abnormally large recreational landings in 1982 and 1986 were excluded. The ratio between average recreational landings and pot/handline landings was 2.63. This ratio applied to the commercial pot landings produced the recreational landings for 1968 to 1980. Length frequencies of sea bass were based on dockside sampling by MRFSS staff.
Recreational discards were from MRFSS estimates of discards using 25% discard mortality as in previous assessments (Table 2). Discard number until 2004 was converted to weight assuming comparable mean weight as landings. Discard estimates for 2005-2009 were based on mean weight of sea bass discards measured on for-hire vessels as part of the MRFSS program. Between 1981 and 1998 the ratio of discard losses to landings was relatively constant with an average of 13%. Since 1999 however, the proportion discarded has increased, averaging 41% of landed sea bass by weight. Since 2000, total weight of discard losses of sea bass averaged 486 mt. Black sea bass discard losses in 2008 equaled 426 mt but decreased to 388 mt in 2009.
The NEFSC spring bottom trawl survey conducted since 1968 provides indices of relative abundance in number and weight. Beginning in 2009 the survey was conducted by the FSV Henry B. Bigelow, so indices of abundance have been adjusted by calibration coefficients to provide units of measure equivalent to the FV Albatross IV (Miller et al. 2010). A coefficient for black sea bass of 3.416 was assumed applicable to all size sea bass. During autumn, sea bass are generally inshore on structured bottom that is not conducive to sampling with an otter trawl. Consequently those survey results are not considered indicative of adult sea bass abundance. However, commercial trawl fisheries since the 1930’s have had significant landings of sea bass caught during the winter and early spring on the continental shelf. The spring offshore bottom trawl survey takes place in the same areas suggesting that the use of trawl gear for sampling sea bass at this time of year is not hampered by habitat. Comparison of survey length frequencies and length frequencies of commercial landings in previous assessments (NEFSC 2009) suggest the selectivity at length is comparable. Additionally, the winter survey relative abundance time series from 1992 to 2007, which was included in the model as an index of abundance, is correlated to the spring abundance. Although the catch per tow in the spring survey was low, the correlation to the winter survey as well as the comparable length frequency to the commercial fishery suggests that the survey is able to sample sea bass. Finally, the index of abundance from the spring survey also closely resembles the time series of recreational catch per angler trip estimated from MRFSS dockside sampling (NEFSC 2009).
The ln re-transformed spring offshore index of black sea bass abundance and biomass have shown a steady decline since achieving the time-series maximum in 2003 (Figure 4a and 4b). The 2009 values of stratified mean number and weight per tow were both below the time series averages. The index of exploitable biomass (defined as fish > 22 cm presented as the loge re-transformed stratified mean weight per tow) began in 1968 increased to a peak value in 1976 followed by a decline to the series low in 1982 (Figure 5). A slight rise in abundance was evident in the late 1980s but followed by a decade of fluctuations around low levels of abundance. Between 1999 and 2002 the index increased again peaking in the series high in 2002 (0.8 kg per tow), followed once again by a steady decline through 2009 when the index dropped to 0.08 kg per tow. The latest value is below the long-term average of 0.25 kg per tow. The NEFSC winter survey, initiated in 1992, follows a similar pattern with a peak in the loge re-transformed index value for 2003 (1.83 kg/tow) followed by declining indices to 0.40 kg/tow in 2007 (Figure 5).
Juvenile indices of black sea bass from the winter and spring surveys provide some insight into cohort strength. The juveniles appear as clearly defined modes at sizes < 14 cm in the autumn surveys (Figure 6). There appears to be little growth during the winter, as the same distinct size mode appears in the winter and spring survey length frequencies. In the spring, fish < 14 cm would be considered one year old. Indices were calculated as the sum of log re-transformed mean #/tow at length for sea bass less than or equal to 14 cm. The indices in both the winter and spring surveys suggest large 1999 and 2001 cohorts (peaks in the 2000 and 2002 surveys) (Figure 7). Both of these modes in the length frequency appear the following year as increases in a mode above 20 cm, which is consistent with known growth rates. The winter and spring surveys show an above average 2002 year class and the spring survey shows a strong 1998 cohort that was below average in the winter survey. The 2009 spring juvenile index (0.17 fish /tow) was above average.
Included as input to the SCALE model were NEFSC spring and winter offshore indices of abundance. The spring series of stratified ln re-transformed mean number per tow included 1968 to 2009 while the comparable indices from the winter survey were 1992 to 2007 (Figure 8). Juvenile indices in the spring and winter surveys were computed as the sum of re-transformed indices at length for fish less than or equal to 14 cm. Mean lengths at age were predicted from an average growth curve among available studies and length-weight equation parameters were from fitted length weight data collected on NMFS surveys. Total catch was commercial landings since 1968, recreational landings since 1981 estimated in MRFSS and 1968 to 1980 estimates derived from commercial inshore fishery landings, recreational discard losses since 1981 and commercial discard estimates since 1989. The model was allowed to fit survey length frequencies greater than 30 cm to counter the lack of discard length data in the fishery length frequencies. Selectivity periods were chosen based on regulatory changes in the fisheries. The three periods were 1968 to 1997, 1998 to 2000 and 2001 to 2009. The model was allowed to fit the initial fishing mortality in phase two. The model was fit with a constant natural mortality of 0.4. A variety of model configurations were examined. Initial runs suggested that the winter juvenile index for 2007 heavily influenced the results in the terminal year of 2008. Several runs were made removing the 2007 winter recruitment index as well as two runs which included an autumn recruitment index (fish < 14 cm). Additional variations included changes in the weighting for catch, recruitment indices and spring adult indices. The final results were based on a model averaging approach (Burnham and Anderson 2002) with the averages weighted by the inverse of the value of the likelihoods normalized to the mean. Confidence intervals (95%) for the average were based on the variation among the 10 models, not from estimates of within model variation. Model output included total biomass, fishing mortality and population number at length. The population numbers at length were used to calculate annual estimates of spawning stock biomass (based on a length–weight equation and maturity at length from NEFSC survey results).
Estimated average fishing mortality has declined steadily since a peak in 2001 of 1.17 (Figure 9). The 2009 value was below FMSY (0.42) with the average among models equal to 0.29 (among model std. dev = 0.035) decreasing from the 2008 estimate of 0.35.
Total biomass exceeded BMSY (13,977 mt) in 2002 through 2003 then declined slightly (Figure 10). Estimates for 2009 (model average of 14,379 mt (among model std. dev = 1,302) increased to above BMSY, although the confidence intervals encompassed BMSY. SSB followed a similar pattern, increasing above SSBMSY (12,537 mt) in 2002-2005 then declining slightly in 2007 but increasing again in 2009 to 12,978 mt (among model std. dev=1,245) (Figure 11). The bounds of the 2009 estimates include the value of the reference point.
Recruitment estimated by the model was relatively constant through the time series with the exception of the 1999 and 2001 year classes (Figure 12). These cohorts appeared to be the driving force behind the increase in biomass and SSB. The estimated average recruitment (age one) in 2009 (2008 cohort) of 27.4 million fish (among model std. dev=1613.6) is equivalent to the long term average of 27.2 million recruits.
Retrospective analysis was done by sequentially removing the terminal year in the analysis back to 2002. The analysis was done for each model configuration. Results (Figure 13) suggest a retrospective bias in fishing mortality and recruitment, with under-estimates of F and over-estimates of recruit abundance. Total biomass shows the potential for overestimation in the terminal year. The average Mohn’s rho values for SSB and recruitment was 0.25 while rho for fishing mortality was -0.29 (Table 4). No adjustments for retrospective bias were made in the projections.
Deterministic projections for black sea bass were made using the SCALE model. Final values from the model averages for catch, recruitment, etc. were input as fixed values in the SCALE input. Run number 7, which closely approximated the average results, was used as the basis for weighting, CVs, etc. Catch estimate for 2010 was assumed equal to the sum of commercial landings quota (798.3 mt), the 3 year average of recreational landings, commercial and recreational discards (totaling 2,223 mt). Length composition was assumed constant since 2008. Fishing mortality in 2010 was calculated by iterating across possible F values until the model predicted catch was equivalent to the assumed 2010 catch. Projections for 2011 were completed by inputting a value for fishing mortality, such as FMSY, and allowing the model to predict the associated catch. Additionally, an iterative process with F was used to produce catch equivalent to MSY. The ratio of discards to landings was assumed equal to the 2000 to 2009 average (20%).
Under the current 2010 quota, fishing mortality was calculated as 0.26. The subsequent catch in 2011 for a fishing mortality equal to F2009 (0.29) projected to be 2,467 mt (Table 5), with landings equal to 1,974 mt. Similarly, a 2011 F equal to FMSY (0.42) would result in an expected total catch of 3,465 mt and landings of 2,772 mt, while an F of 0.315, which is 75% of FMSY, would project to a 2011 catch of 2,708 mt or landings of 2,166 mt. In all cases, the expected total biomass would exceed BMSY beginning in 2011. A similar projection fixing the catch at the 2010 quota would project to a 2011 fishing mortality of 0.23. Total biomass and SSB associated with each F are presented in Table 5. In all cases Jan 1, 2011 spawning biomass remains above SSBMSY.
The conclusion of the assessment update is that black sea bass are not overfished and overfishing is not occurring. Projections through 2011 suggest that an increase in fishing mortality up to FMSY will not result in a decrease in biomass below BMSY. However, underlying these conclusions is the uncertainty associated with an assessment of a data poor stock as noted in the Northeast Data Poor Stocks Working Group report:
“These new reference points and stock status determinations should be used with caution due to the uncertainty in the natural mortality estimate, the model input parameters, residuals patterns in model fit, and significant uncertainty associated with managing a protogynous species (i.e., individuals change sex from female to male).”
In addition, tagging results suggest spatial partitioning along the coast that is not yet accounted for in the assessment model. Consequently the results may not reflect the stock condition in all local groups of black sea bass.
Alonzo, S.H., T. Ish, M. Key, A.D. MacCall, and M. Mangel. 2008. The importance of incorporating protogynous sex change into stock assessments. Bulletin of Marine Science 83(1): 163-179.
Brooks, L., K.W. Shertzer, T. Gedamke, and D.S. Vaughan. 2008. Stock assessment of protogynous fish: evaluating measures of spawning biomass used to estimate biological reference points. Fish Bulletin 106(1):12-23.
Burnham, K. P., and D. R. Anderson. 2002. Model selection and inference: a practical information-theoretical approach. Springer-Verlag, New York.
Drohan, A. F., J. P. Manderson, and D. B. Packer. 2007. Essential Fish Habitat Source Document: Black Sea Bass, Centropristis striata, Life History and Habitat Characteristics. Second Edition. NOAA Technical Memorandum NMFS-NE-200.
Hamilton, S.L. J.E. Caselle, J.D. Standish, D.M. Schroeder, M.S. Love, J.A. Rosales-Casian and O.Sosa-Nishizaki. 2007. Size-selective harvesting alters life histories of a temperate sex-changing fish. Ecological Applications. 17(8): 2268-2280.
Heppell, S.S, S.A. Heppell, F.C. Coleman and C.C. Koenig. 2006. Models to compare management options for a protogynous fish. Ecological Applications. 16(1):238-249.
Huntsman, G.R. and Schaaf W.E.1994. Simulation of the impact of fishing on reproduction of a protogynous grouper, the graysby. N. Am. J.Fish. Management. 14:41-52.
Lavenda, N. 1949. Sexual difference and normal protogynous hermaphroditism in the Atlantic sea bass, Centropristes striatus. Copeia 3:185-194.
Mercer, L.P. 1978. The reproductive biology and population dynamics of black sea bass, Centropristis striata. Ph.D. thesis, Va. Inst. Mar. Sci.,
College of William and Mary, Gloucester Point, VA 23062.
Miller, T.J., C. Das, P.J. Politis, A.S. Miller, S.M. Lucey, C.M. Legault, R.W. Brown, and P.J. Rago. 2010. Estimation of Albatross IV to Henry B. Bigelow calibration factors. NEFSC Ref. Doc. CRD 10-05.
Moser, J. and G.R. Shepherd. 2009. Seasonal Distribution and Movement of Black Sea Bass (Centropristis striata) in the Northwest Atlantic as Determined from a Mark-Recapture Experiment. Journal of Northwest Fisheries Science. 40:17-28.
Musick, J.A., and L.P. Mercer. 1977. Seasonal distribution of black sea bass, Centropristis striata, in the Mid-Atlantic Bight with comments on the ecology and fisheries of the species. Transactions of the American Fisheries Society. 106(1):12-25.Nelson, D.A., D. Perry, and E. Baker. 2003. Natural spawning of black sea bass, Centropristis striata, at the NMFS Milford Laboratory and the UMASS Dartmouth Laboratory with observations on spawning behavior. Journal of Shellfish Research. Vol. 22, no. 1, pp. 297-298.
NEFSC Northeast Data Poor Stocks Working Group. 2009. The Northeast Data Poor Stocks Working Group Report, December 8-12, 2008 Meeting. Part A. Skate species complex, Deep sea red crab, Atlantic wolfish, Scup, and
Black sea bass. US Dept Commer, Northeast Fish Sci Cent Ref Doc. 09-02; 496 p. http://nefsc.noaa.gov/publications/crd/crd0902/
Shepherd, G.R. and M. Terceiro. 1994. The summer flounder, scup and black sea bass fishery of the Middle Atlantic Bight and Southern New England waters. NOAA Tech. Report NMFS 122.
F.W., C.A.Zetlin, P.L. Berrien, and S. Chang. 1999. Essential Fish Habitat source document: Black Sea Bass, Centropristis striata, Life History and Habitat Characteristics. NOAA Technical Memo.NMFS-NE-143.
Wigley, S.E., M.E. Palmer, J.Blaylock, and P.J. Rago. 2008. A brief description of the discard estimation for the National Bycatch Report. NEFSC Ref. Doc. 08-02.