Assessment and Management of Atlantic and Gulf Menhaden Stocks

Item Type article

Authors Vaughan, Douglas S.; Merriner, John V.

Download date 07/10/2021 16:15:09

Link to Item http://hdl.handle.net/1834/26534 Assessment and Management of Atlantic and Gulf Menhaden Stocks

D. S. VAUGHAN and J. V. MERRINER

Introduction offbetween catching greater numbers of age-l are obtained from virtual popula­ younger and smaller fish or catching tion analysis (VPA) conducted separately Stock assessment refers to the process fewer numbers ofolder and larger fish. for both Atlantic and Gulf menhaden. ofcollecting and analyzing biological and The trade-offdepends both on the rate of Equilibrium spawning stock ratios are statistical information to determine growth of individual fish and on their computed to compare the level ofspawn­ changes in abundance of fish stocks in natural mortality rate. The concept of ing stock expected based on estimated response to fishing, and, to the extent "recruitment overfishing, " on the other fishing mortality for a fishing year to that possible, predict future trends of stock hand, refers to a concern that insufficient level with no fishing mortality. Estimates abundance. Relevant questions arising numbers of fish are reaching spawning of potential yield from varying age at during such an assessment include: age and subsequently reducing the entry to the fishery and fishing mortality number of future recruits below levels rates are investigated using yield-per­ 1) Do current landings exceed max­ that will maintain the stock at fishable recruit analysis. Recent landings are imum sustainable yield or some other levels. compared to estimates ofmaximum sus­ measure of optimal yield? Landings and fishing effort data have tainable yield (MSY) obtained from 2) Is recruitment sufficient to support been collected for the Atlantic menhaden surplus production, spawner-recruit, current levels of landings? purse-seine fishery since 1940 and for the and population simulation models. Final­ 3) Does recruitment depend more on Gulfmenhaden purse-seine fishery since ly, management implications drawn from spawning stock size or more on envi­ 1946 (Smith, 1991). Port sampling data this information are discussed in light ronmental conditions? for size and age composition from scales ofthe pastdecade ofmanagementactions 4) Should age-at-entry to the fishery (procedures described by Chester, 1984) and inactions by the coastal states. (minimum size or age landed) be raised have been conducted by NMFS port to increase yield to the fishery? samplers since 1955 for the Atlantic Organization of Coastwide menhaden purse-seine reduction fishery Management Programs Fishery management, or the system and since 1964 for the Gulf menhaden Authority for menhaden fishery man­ used to conserve and allocate fishery re­ purse-seine reduction fishery. agement resides with individual states. sources, must address these questions. This paper describes the organization The formation of the Atlantic and Gulf Two "overfishing" concepts are re­ ofcoastwide management programs for States Marine Fisheries Commissions ferred to in this paper. The concept of the Atlantic and Gulf menhaden stocks, (ASMFC and GSMFC, respectively) in "growth overfishing" refers to the trade- summarizes current information about the 1940's provided a forum for discus­ the effects of purse-seine fishing for re­ sion and resolution of common marine duction on these stocks, and discusses resource issues and a vehicle for develop­ management implications drawn from ment ofcooperative multistate fisheries this information. Regular stock assess­ programs such as for menhaden. The ABSTRACT-The organization of coast­ ments are conducted on Atlantic and Gulf NMFS serves as the primary research wide management programs for Atlantic menhaden and have been presented at agency dealing with menhaden for the menhaden, Brevoortia tyrannus, and Gulf stock assessment workshops held by the commissions. Through the years this menhaden, B. patronus, are described. Recent NMFS Southeast Fisheries Science state-federal research and fishery man­ assessments ofthe status ofthe Atlantic and Center (Powers, 1983; Vaughan et al. , agement system drew heavily upon in­ Gulfmenhaden stocks are summarized. Esti­ mates ofpopulation size and fishing mor­ 1986). Estimates of age-specific popu­ put and voluntary participation of the talities are obtained from virtual popula­ lation numbers and fishing mortality menhaden industry. This informal in­ tion analysis, and are used in determining rates, spawning stock, and recruits to stitutional arrangement provided guid­ spawner-recruit relationships, spawning ance and fishery oversight through the stock ratios, yield-per-recruit, and surplus D. S. Vaughan and J. V. Merriner are with the 1960's and into the 1970's. Interstate production. Management issues are ad­ Beaufort Laboratory, Southeast Fisheries Science dressed in the framework of assessment Center, National Marine Fisheries Service, NOAA, research and monitoring coordination results. Beaufort, NC 28516. was through the ASMFC Advisory Com­

53(4),1991 49 mittee and the GSMFC Technical Coor­ For various reasons, several states and Federal representation and serve as dinating Committee (TCC). recently have passed severe limits on a clearinghouse for proposed regulations menhaden purse-seine fishing (New or laws affecting the menhaden fishery. Atlantic Menhaden Jersey, Delaware, and South Carolina Preparation of a major rewrite of the Development ofan Atlantic Menhaden have closed all or part of their waters). Fishery Management Plan by the Atlan­ Management Plan under the auspices of Continued human population growth in tic Menhaden Advisory Committee the ASMFC and the State-Federal Fish­ the Atlantic coastal zone and conditions began in 1989 with expected completion eries Management Program was begun in the U.S. economy have resulted in in 1992. by the Statistical and Scientific Commit­ pressures for fishery restrictions based tee (SSC) in fall 1976. The SSC had a very on political reasons rather than the bio­ Gulf Menhaden general charge and prepared an interim logical needs of the fishery resource as The GulfMenhaden Regional Fishery report (plan) in February 1977. In July called for in the plan. This is tending Management Plan (Christmas and Et­ 1977 the SSC was asked to prepare a plan toward a reduced number of operating zold, 1977) was developed in the 1970's for the utilization of Atlantic menhaden plants and reduced fleet size in the Atlan­ as a product ofthe Gulf Menhaden Sub­ that' 'is biologically, economically, and tic menhaden fishery (Smith, 1991). By committee ofthe TCC; it was composed socially sound and which protects the 1988 reduction plants for Atlantic of state, industry, and Federal repre­ resource and its users" (AMMB, 1981). menhaden were reduced to four U.S. sentatives. The charge to the subcom­ Over the next 3 years the SSC and At­ shore-based facilities (one in Maine, mittee was to consider the need for and lantic Menhaden Management Board two in Virginia, and one in North Caro­ possible procedures for establishing a (AMMB) developed a plan, with support­ lina). The 1989 fishing year was the uniform menhaden fishing season in ing analyses by staffofthe NMFS Beau­ third during which menhaden caught the GulfCoast states. In spring 1976 the fort Laboratory. In 1981 the AMMB, and off Maine were transported to a land­ Menhaden Subcommittee, the TCC, later that year the full ASMFC, adopted based facility in New Brunswick, Can., and the GulfState-FederalFishery Man­ the menhaden plan (AMMB, 1981). This for reduction, and the second year ofan agement Board endorsed a proposal to plan recommended adjustment of time Internal Waters Processing venture be­ develop a fishery management plan for and area closures if necessary (based tween a Maine company and the Soviet Gulf menhaden. With support ofNMFS, upon the best data available) to achieve Union; Maine fisherman caught Atlantic the states, and industry, the plan was a short-term objective ofattaining an age menhaden and offloaded the fish onto developed and issued in May 1977 composition in the population which a Soviet processing vessel (FlY Riga). (Christmas and Etzold, 1977). Revisions included 10% age-3 or older and estab­ The land-based reduction plant in Maine of the plan have been issued by the lished an organizational framework for ceased operation following the 1988 GSMFC in about 5-year intervals management. The contents of the plan fishing year. (Christmas et al. ,1983,1988), providing and the management structure were a Furthermore, in 1987 the ASMFC updates of stock assessments of the combined state, industry, and Federal in its Interstate Fishery Management resource, descriptions ofindustry, fish­ effort of data gathering, analysis, and Program reorganized the management ery, biology, and identification and fishery management decision-making. structure for territorial sea fish species. ranking of research needs. The major Later, fishery statistical information in That action eliminated the AMMB and management feature of the Gulf men­ the fishery managementplan was updated AMAC; it also removed most of the in­ haden plan is a uniform season (third (AMMB,1986). dustry's participation in the manage­ Monday of April through the Friday In May 1982 the AMMB considered ment process. The new plan review pro­ following the second Tuesday in Octo­ several management options which cedure calls for annual status reports ber, equals 26 weeks) adopted by all Gulf would reduce pressure on age-O and on the fishery and resource to be done of Mexico states, with the exception of age-l fish and increase yield-per-recruit: by the ASMFC Advisory Committee or Florida, and a formal organizational Season options, closed corridor, and a designated special plan review sub­ structure for interjurisdictional manage­ mesh size. The AMMB approved a vari­ committee (which for menhaden con­ ment. Individual Gulfcoast states vary in able seasonal closure by geographic area sisted of the former membership of other measures (such as licenses, sanc­ (also known as option 7) which reduced AMAC). tuaries, and penalties) enforced upon the the duration offishing activity by 4 weeks As a result of the changes in ISMFP menhaden fishery. The program has been in each offour geographicareas (AMAC, structure and in the reduction fishery, very effective and GMAC continues to 1982). Recent analyses have been con­ the ASMFC reconstituted the Atlantic meet biannually to review stock status ducted comparing the shortened season Menhaden Management Board in 1988 and management measures. option to the closed corridor option with three state members (Maine, Vir­ (Blomo, 1987; Vaughan and Smith, ginia, and North Carolina). Two industry Atlantic Menhaden 1991). That recommendation has yet members were added to the Board in to be implemented in all coastal states 1990. The Atlantic Menhaden Advisory Summary of Recent having an active purse-seine fishery, Committee also was reconstituted to Stock Assessments most notably North Carolina. maintain the vital mix ofstate, industry, Growth overfishing has been of pri-

50 Marine Fisheries Review Table 1.-Annualestimatesol Atlantic menhaden popula­ time (usually 1 year). Population ex­ tion size (age 1-8 at start ollishlng year) and numbers land- a ed (age 1 to maximum age observed), exploitation rates ploitation rates (based on age-l and older (u, age 1-8), and weighted mean F (1/year), lor lishing .§ 0.8 ----­ Atlantic menhaden) averaged 38 % ofthe years, 1955-87. '0 ~ 0.6 population removed by fishing for 1955 Population Numbers ~ Fishing size in landed in through 1987 (Table 1). From 1955 year billions billions u F through 1961 when population size and 1955 6.97 2.36 0.339 0.501 landings were high, population exploita­ 1956 8.30 3.53 0.425 0.825 tion rate averaged 35 %. During the 1957 983 3.21 0.327 0.819 1958 7.12 2.61 0.367 0.666 period of low population size and land­ 1959 17.63 5.34 0.303 0.552 ings from 1962 through 1974, population 1.2 b 1960 9.31 2.70 0.290 0.491 exploitation rate averaged 43 % (initial­ 1961 6.84 2.60 0.380 0.706 c 1 1962 4.59 2.05 0.447 1.162 .~ ly high during the mid-1960's and lower 1963 3.60 1.67 0.464 1.092 E 0.8 f-----,i, '0 during the late 1960's and early 1970's). 1964 2.77 1.43 0.516 1.013 a. ~ 0.6 Since 1975 when population size and 1965 2.59 1.26 0.486 1.040 landings have improved significantly, 1966 2.07 0.99 0.478 0.758 ~ 0.4 1967 2.50 0.98 0.392 0.787 '0 the population exploitation rate has 1968 2.02 0.99 0.490 1.211 Q:: 0.2 ­ 1969 2.22 0.71 0.320 0.676 averaged 35 %. For fishing years 1955 a f--'--'-~>--,-,-~>--,-,-~>--,-,--'-"-->--,-,--'-Y>--'-'--'-Y~ 1955 1960 1965 1970 1975 1980 1985 through 1987, an average of24 %ofage-l 1970 3.44 1.38 0.401 0.778 1971 2.46 0.90 0.366 0.807 Fishing year menhaden and 65 % of age-2 and older 1972 4.32 1.66 0.384 1.140 menhaden were taken by the fishery an­ 1973 4.18 1.79 0.428 1.755 Figure I.-Estimates of annual rates 1974 4.33 1.68 0.388 1.302 of exploitation of Atlantic menhaden nually, with 30 % and 20 %, respective­ (ages 0 through 5 and total population 1975 5.24 1.86 0.355 1.235 Iy, being lost to natural causes annually 1976 8.74 3.01 0.344 0.975 (ages 0-8». for fishing years 1955-87. (compared to 36 % lost to natural causes 1977 8.40 3.19 0.380 1.011 1978 7.62 2.63 0.345 1.041 annually in the absence of fishing mor­ 1979 7.08 2.38 0.336 0.728 tality). Age-specific exploitation rates 1980 9.35 3.24 0.347 0925 for age-O menhaden range from essen­ 1981 8.15 2.80 0.344 0.763 tially 0% to 26%. 1982 9.42 3.06 0.325 1.224 1983 6.15 2.98 0.485 1.082 t), and averaged 3.2 billion menhaden The number of Atlantic menhaden 1984 5.48 2.25 0.411 0.888 between 1962 and 1974 when landings spawners (age 3 and older at the start 1985 6.85 2.39 0.349 0.990 were low (averaging 289,000 t). How­ of the fishing year) ranged between 1986 7.26 1.82 0.251 1.152 1987 8.19 2.37 0.289 0.902 ever, since 1975 population size has 0.03 billion in 1973 and 1.3 billion in averaged 7.5 billion menhaden, compar­ 1961 (Table 2). High spawning stock ing favorably to population sizes be­ size (averaging 0.6 billion menhaden) tween 1955 and 1961, but landings have was the rule between 1955 and 1961, low improved only slightly to an average of spawning stock size predominated be­ ..... 341,000 1. The inability of the modern tween 1962 and 1974 (averaging 0.1 mary concern with the Atlantic men­ fishery to regain former high levels of billion menhaden), and some improve­ haden stock. Information presented in landings is due primarily to declining ment in spawning stock size has oc­ this section is drawn primarily from mean weight at age occurring since 1970 curred since 1975 (averaging 0.2 billion Ahrenholz et al. (1987), Vaughan and (Reish et aI., 1985; Ahrenholz et aI., menhaden). Between 1955 and 1961 Smith (1988), and some recent analytical 1987; Vaughan and Smith, 1988), caused high spawning stock size resulted in results prepared for AMAC. The Atlan­ in part by changes in fishing patterns both excellent recruitment ofage-l menhaden tic menhaden fishery is believed to exploit geographically and seasonally. Part of (averaging 5.5 billion) entering the fish­ a single stock or population offish based this decline is due to the shift ofthe center able stock. Low spawning stock size on tagging studies (Dryfoos et al., 1973; of the fishing activity southward and present from 1962 through 1974 pro­ Nicholson, 1978a). The Atlantic men­ subsequent seining on smaller fish at age duced poor recruitment (averaging 2.2 haden fishing season runs from March 1 and part can also be explained by the in­ billion menhaden). However, the some­ through the end of February of the fol­ verse relationship noted between first what improved spawning stock size lowing calendar year for the reduction year growth of Atlantic menhaden and present since 1975 has produced excel­ fishery. year class strength (Reish et aI., 1985; lent recruitment (averaging 5.0 billion Population size (age-l and older at Ahrenholz et al., 1987). menhaden), comparable to that produced the start of the fishing season) ranged Short-term losses to the Atlantic men­ during the high spawning stock years from 2.0 billion Atlantic menhaden in haden stock due to the fishery can be (1955-61). 1968 to 17.6 billion fish in 1959 (Table assessed by considering the exploitation Since 1955 the contribution of late 1). Population size averaged 9.2 billion rate (Fig. I), which is the fraction of age-2 spawners to the spawning stock menhaden between 1955 and 1961 when the remaining stock removed by the has averaged about 76 % in numbers and landings were high (averaging 604,000 fishery during some specified period of 66% in index ofegg production (Fig. 2).

53(4). 1991 51 Table 2.-Estimated number of spawning Atlantic 100,------, 80- menhaden (age 3-8females) thatproduced theyearclass, ~ u ~ estimated egg production from the spawning stock, and ~ 80 estimated numbers of recruits to age-1 by year class, .... 0> S 1955-86. C a: 'c 60 () "'a: Spawners Number of ~ a: 40- recruits -80 "' Q. Year Number in Number of eggs to age-1 0 class billions in trillions in billions ..J "'>: 60 1955 0.795 152.6 5.68 1956 0.587 124.1 7.25 1960 1965 1970 1975 1980 1985 1957 0.282 65.9 3.32 40 1958 0.216 42.5 15.10 fishing yeor 1959 0.538 77.2 2.22 ------20 1960 0.307 57.5 3.01 Figure 2.-Contribution of age-3 1961 1.321 152.9 2.23 spawners to total spawning stock (num­ 91.1 2.23 I I I I , 1962 0.545 bers) and to total egg production (eggs) 0.2 0.6 1.0 1.4 1.8 1963 0.176 30.3 1.74 1964 0.084 14.1 1.91 ofAtlantic menhaden for fishing years F-MULTIPLE 1955-87. 1965 0.059 93 1.37 Figure 5. - Yield-per-recruit isopleth 1966 0.028 4.0 1.93 diagram for Atlantic menhaden using 1967 0.056 9.4 1.18 average growth and fishing mortality 1968 0.050 7.4 1.68 8000 '--;(".)"'-;-5,"",0""0-;'T;-2"",76.------, values by quarter for the 1981 fishing 1969 0.039 6.3 2.57 .c ,2 year. := 6000 1970 0.045 7.3 1.33 .§. 1971 0.084 12.6 3.44 1972 0.120 22.1 2.69 R = 67.B S np(-O.0038 S) ~ 4000 1973 0.031 5.9 2.99 o0> 1974 0.038 5.3 3.75 o which it should not be allowed to decline ~ 2000 6.80 1975 0.044 5.9 .~ to protect the stock from recruitment 1976 0.057 6.6 5.12 4.69 oL----~--~--~-~--~ overfishing. These ratios, as presented 1977 0099 10.7 '" 1978 0.226 18.1 4.21 o 300 600 900 1200 1500 here, are calculated under the assumption 0.199 16.3 6.65 1979 Spowner5 (million5) of equilibrium; that is, annual age-spe­ 1980 0.253 23.8 4.67 cific estimates ofF are used to project a 1981 0.198 17.5 6.36 Figure 3. - Numbers ofAtlantic men­ 1982 0.316 19.6 2.45 haden recruits (R) plotted against num­ fixed number ofrecruitsthroughouttheir 1983 0.178 14.4 3.81 bers ofspawners (S) for year classes, 1984 0.254 22.4 5.07 lifespan and sum the spawning stock in 1955-86. Curve represents the fitted weight or index of egg production. The 1985 0.076 8.1 4.70 Ricker function R = as exp (- (3S). 1986 0.083 7.2 4.95 index of egg production for Atlantic menhaden is based on the egg-length relation provided in Lewis et al. (1987). 301------;:======~ Since 1962 the spawning stock ratio has

2S • ~SR: _ remained below 10% (Fig. 4). Because These values were exceptionally high values of20%-40% have been used by during the 1970's (87% and 78%, re­ 20 the Gulf of Mexico and South Atlantic spectively), but havedeclined somewhat Fishery Management Councils in their during the 1980's (77% and 65%, re­ definitions ofoverfishing for a numberof spectively), lessening the concern that fish stocks, these low values for Atlantic recruitment failure in a single year class menhaden raise concern. However, of-'--'~_+"'_~"'_+_~"_'_t~~+_'__~ ...... ~_'_+-'-' could have significant consequences on 1955 1960 1965 1970 1975 1980 1985 periods of both poor recruitment and future year classes. When spawner and Fishing yeor excellent recruitment have occurred recruit data are fit to the Ricker model since 1962, reinforcing the concept that Figure 4.-Equilibrium spawning (Ricker, 1975), a statistically significant stock ratio in weight and index ofegg environmental conditions are most im­ relationship is obtained (Fig. 3). How­ production for Atlantic menhaden for portant in determining recruitment ever, considerable unexplained variabil­ fishing years, 1955-86. success. ity about the estimated spawner-recruit Yield-per-recruit models are used to curve suggests that recruitment variabil­ determine whether Atlantic menhaden ity depends little on spawning stock size, are being removed at too young an age and that environmental factors are pro­ spawning stock size calculated when Fis (growth overfishing). Ayield-per-recruit bably more important in controlling equal to 0 (unfished stock). This ratio isopleth shows gains and losses ofyield­ recruitment success or failure. does not consider such compensatory per-recruit as a function offishing mor­ Gabriel et al. (1984) suggested that a mechanisms as increased growth rate or tality rate or age at entry to the fishery ratio of spawning stock size calculated earlier maturity when a fish stock is (Fig. 5). Overall yield-per-recruit forthe when fishing mortality is equal to that reduced from fishing. Itwas thought that age at entry of 0.5 year and F-multiple estimated for the present divided by the this ratio would provide values below of 1.0 has been decreasing since 1971

52 Marine Fisheries Review Table 3.-Estimates 01 yield-per-recruit (g) lor Atlantic with an average of 58.7 g for the period menhaden lor each lishing yearlrom 1970 through 1981, recruit, except for small F-multiples; 1970-84 (Table 3). The proportional and lor mean conditionslorthe period 1970-84. Estimates e.g., F-multiple = 0.2 (Table 3). On are presented lor three ages at entry (0.5, 1.0, and 2.0 contribution of younger age groups to years) and three F-multiples (0.4,1.0, and 1.6)(Vaughan the other hand, decreasing the F-multiple the landings has been increasing, and and Smith, 1988). to F-multiple = 0.6 generally causes a the average size at age (as noted earlier) Age at entry F-multiple .decrease in yield-per-recruit, except for Fishing the 1979 fishing year. Greaterdeclines or is decreasing. Reduced growth and re­ year 0.5 1.0 2.0 0.4 1.0 1.6 direction of effort toward younger fish any increases in the F-multiple general­ 1970 93.6 94.3 93.8 77.1 93.6 93.5 are contributing to the reduced levels in 1971 107.5 108.8 118.0 97.9 107.5 104.6 ly causes a decrease in yield-per-recruit yield-per-recruit. 1972 102.1 103.4 118.9 108.6 102.1 94.1 at the current age at entry. These results Recent landings ofAtlantic menhaden 1973 92.2 93.3 101.7 96.5 922 865 suggest that the fishery is harvesting have been increasingly dependent on 1974 87.2 91.1 101.6 91.8 87.2 800 the Atlantic menhaden stock at too young 1975 786 80.5 86.8 83.4 78.6 73.4 age-Omenhaden(e.g., 1979, 1981, 1983, 1976 667 68.9 78.7 65.8 66.7 60.9 an age, and that the age at entry should 1984, and 1985 as noted in Vaughan 1977 54.4 57.5 63.5 59.5 54.4 48.7 be raised to increase potential yield from and Smith, 1988). Gain in yield-per­ 1978 54.1 57.0 59.8 56.4 54.1 50.3 the stock. 1979 53.0 58.9 62.0 48.9 53.0 49.2 recruit from increasing age-at-entry to 1980 53.8 54.3 61.6 54.5 53.8 49.5 In summary, the modern purse-seine age-l would have ranged from 0.7 % in 1981 45.9 49.9 51.8 51.1 45.9 41.3 fishery for the Atlantic menhaden has 1970 to 11.1 % in 1979and8.7% in 1981. a high dependency on prespawners Even greater gain in yield-per-recruit ------Mean conditions ------•• -­ (age-2 and younger fish), so large fluc­ could be obtained by raising the age­ 1970-84 58.7 60.6 64.4 57.3 58.7 55.7 tuations in year-to-year availability and at-entry to age-2 (17.0% in 1979 and catches are to be expected. To increase 12.9% in 1981). However, management yield and enhance the stability of the options that protect only age-O (or age-I) resource, it is desirable that the number menhaden have been difficult to devise ofage classes significantly contributing (Vaughan and Smith, 1991). ent structure of the fishery. However, to the fishery be increased. The intent is Historical estimates ofmaximum sus­ during the 1980's landings averaged to harvest more of the stock at an older tainable yield (MSY) range from 370,000 341 ,000 t with a mean F averaging age, while decreasing fishing mortality to 560,000 t (Schaaf and Huntsman, 0.94/year suggesting greater landings on the younger immature menhaden. 1972; Schaaf, 1975, 1979; Ahrenholz would be available with less effort. This would create a buffer in the spawn­ etal., 1987; Vaughan and Smith, 1988). Historical MSY estimates since the ing stock against future years of poor The most recent application of surplus early 1970's have shown no trends, recruitment and lessen the year-to-year production models conducted for ranging between 370,000 and 570,000t. fluctuations in landings by increasing the AMAC, which relate landings and Sufficient recruitment to attain MSY proportion of Atlantic menhaden that fishing effort, suggest estimates ofMSY has been available since 1975, but with survive to spawning age. Furthermore, of484,000 t (±87,000 t) based on land­ considerable variation about the fitted greater yields would be obtained from the ings and adjusted fishing effort data spawner-recruit curves (Fig. 3). It ap­ stock. Whether landings near the MSY through 1986. High recruitment from pears that managing the fishery to main­ estimates of 450,000 to 490,000 tare 1975-1981 indicates potential yields of tain large numbers of spawners may attainable is questioned because of 416,000 to 481,000 t based on yield­ prove fruitless since environmental changes in plant locations and fishing per-recruit analysis. In general, esti­ conditions appear to outweigh the avail­ patterns. However, gains in yield-per­ mates ofMSY exceed recent landings of ability ofspawners (as numbers oreggs) recruit are possible by adjusting the age Atlantic menhaden which range from in controlling subsequent recruitment. at entry to the fishery (Vaughan and 238,000 to 418,600 t since 1980 with This is suggested by a poor, but sta­ Smith,1991). landings in 1989 at about 322,000 t. tistically significant, spawner-recruit relation and high recruitment both Gulf Menhaden Management Implications following and concurrent with low Although landings have recovered spawning-stock ratio. However, the Summary of Recent somewhat from the depressed levels Ricker spawner-recruit relationships Stock Assessments of the 1970's, they have not returned to are marginally significant, and age-3 Recruitment overfishing has been of the levels attained during the late 1950's spawners are ofgreat importance to the primary concern with the Gulf menha­ when they averaged 625,000 t during spawning stock (Fig. 2). Thus, further den stock. The summary ofinformation the 1955-59 fishing years (Smith et al., increasing the number of older (age-3 presented in this section draws heavily 1987b). Recent estimates of MSY of and older) spawners would guard against on Nelson and Ahrenholz (1986) and 484,000 t ± 87,000 t at a mean F of a possible stock collapse brought on by Vaughan (1987). The Gulf menhaden 0.54/year were obtained from a gen­ heavy fishing during a period of poor fishery is also believed to exploit a single eralized production model (1955-86); recruitment. stock or population offish based on tag­ those levels are unlikely to be attained In general, increasing the age at entry ging studies (Ahrenholz, 1981). The over an extended period given the pres­ causes an increase in the yield-per­ Gulfmenhaden fishing season runs from

53(4), 1991 53 Table 4.-Annual estimates of Gulf menhaden population Table 5.-Estimated numberof spawning Gulfmenhaden size (age 1-4 at start offlshlng year) and numbers landed a (age 2·4 females) that produced a year class, estimated (ages 1-4), population exploitation rate (u), and population egg production from the spawning stock, and estimated c F (lIyear), for fishing years, 1964-83 (Vaughan, 1987). o numbers of recruits to age-1 by year classes, 1964·82 :£ 0.4 ------(Vaughan, 1987). Population Numbers Population % All oges 1-4 Fishing size in landed in ~ 0.3 ------~--- Spawners' Number of year billions billions u F recruits o ~ 0.2 ------..- ..-----­ Year Numbers in Number of eggs to age-l 1964 11.32 4.95 0.437 1.071 class billions in trillions in billions 1965 11.76 6.19 0.526 1.459 0.1 ------1 1966 7.48 4.21 0.563 1.658 1964 3.24 38.6 13.36 1967 10.71 4.62 0.431 1.048 o e-.-~'_+_~~_I_'~--'--

54 Marine Fisheries Review 120r------, 60r------, Table 6.-Estimates of yield-per-recruit (g) for Gulf u menhaden for each fishing yearfrom 1964thraugh 1983, '" 50 "* 100~'l~;?"~___=_~ and fortwo setsofmeanconditions (1978-85and 1964·85 g­ fishing years). Estimatesare presented fortwo agesat en­ try (0.5 and 1.25years)andthreeF.multiples(0.4, 1.0, and ." c 1.6) (Vaughan, 1987). •~ . 60 ------~ )0 Age at entry F-multiple (; 40 ------.,------, Fishing ~ -'ge-2 eonlribullon: - Weight year 0.5 1.25 0.4 1.0 1.6 ~ 20 ------·="Numbe~---[QQi-·------­ 10 -- .-.------­ Q. 17.2 19.4 o~~'-+~'-+-4--'-+~+--,-~-,-+~~--1 1964 17.2 17.1 11.8 1964 1968 1972 1976 1980 1984 1968 1972 1976 1980 1984 1965 18.0 18.0 13.2 18.0 19.7 Fishing yeer Fishing yeor 1966 18.6 18.6 14.1 18.6 20.3 1967 17.0 17.0 11.1 17.0 19.3 1968 17.3 17.3 12.5 17.3 19.3 Figure 7. -Contribution of age-2 Figure 9.-Equilibrium spawning spawners to total spawning stock stock ratio in weight and index of egg 1969 10.4 10.4 5.8 10.4 13.4 1970 15.3 153 96 153 18.1 (numbers) and to total egg production production for Gulf menhaden for 1971 17.5 17.5 12.4 17.5 19.5 (eggs) of Gulf menhaden for fishing fishing years, 1955-83. 1972 15.7 15.7 11.1 15.7 17.3 years, 1964-84. 1973 15.1 15.1 10.2 15.1 17.5 1974 16.3 16.3 9.6 16.3 19.8 1975 16.5 16.5 10.9 16.5 19.2 1976 18.6 18.6 11.9 18.6 21.1 50r------, 1977 19.3 19.3 13.2 19.3 21.4 c 1978 14.0 14.0 8.3 14.0 16.8 o • 20.0,------______== 40 ---.------._.------­ 1979 13.4 13.4 8.3 13.4 16.2 15.0 1980 15.4 15.4 9.3 15.4 18.3 ~ 30 1981 10.2 10.2 5.8 10.2 12.4 '"o 1982 6.9 6.9 39 6.9 9.2 .2 20 20.0 1983 13.6 13.6 8.5 13.6 16.1

~ .~ 10 15.0 ------... - Mean conditions· .. - .. -- .. ­ '" \0 15 20 25 30 1978-85 12.2 12.2 7.9 12.2 14.2 Spawners (billions) 1964-85 15.8 15.8 10.6 15.8 18.1

Figure 8.-Numbers ofGulfmenha­ den recruits (R) plotted against num­ '>!--,---t--r--+O.O bers of spawners (S) for year classes, 1.0 1.5 2.0 gains and losses of yield-per-recruit as 1964-82. Curve represents the fitted F-MUlTIPlE Ricker function R = as exp (- (3S). a function of fishing mortality rate or Figure 10.-Yield-per-recruit iso­ pleth diagram for Gulf menhaden age-at-entry to the fishery (Fig. 10). using average conditions of growth Overall yield-per-recruit of Gulf men­ and fishing mortality for the period haden, for the age at entry of 1.25 years and 98% in numbers (Fig. 7). Since the 1978-85. (or 0.5 year) and F-multiple of 1.0 has mid-1970's, the contribution of age-2 shown no trend since 1964 with an spawners to the spawning stock has average of 15.8 g for the period 1964-85 averaged about 88 % in numbers. As and an average of 12.2 g for the period with Atlantic menhaden, after fitting an upward trend (Fig. 9). The index of 1978-85 (Table 6). Yield-per-recruit the Ricker curve to Gulf menhaden egg production for Gulf menhaden is could actually be increased with higher spawner and recruit data, considerable based on the egg-length relationship rates of fishing, as maximum biomass unexplained variability remains due to provided in Lewis and Roithmayr (1981). is obtained at age 1.5 years and the rate environmental conditions or measure­ The lowest values were obtained from of natural mortality is quite high. How­ ment error (Fig. 8). Again, these relation­ estimated fishing mortality rates from ever, attempts to increase yield-per­ ships are statistically significant, so that the 1965 and 1966 fishing years, the recruit could have severe consequences: future recruits do depend to some extent highest values were obtained from Results from population simulation on the size of the spawning stock which estimated fishing mortality rates from studies by Nelson and Ahrenholz (1986) produced them, but the large scatter the 1981 fishing year. These results are and Vaughan (1987) indicate that recruit­ suggests that environmental factors are not surprising given the decline in F ment overfishing is likely to occur at F­ probably more important in controlling shown in Figure 5, both for all ages 1-4 multiples greater than 1.5 for 1978-85 recruitment success or failure. and especially for age-I. However, with mean conditions. Contrary to the equilibrium spawning the high natural mortality (M = 1.1) for Estimates of MSY from surplus pro­ stock ratios obtained annually for Atlan­ Gulfmenhaden and short lifespan (about duction models continue the upward tic menhaden, those obtained annually 4 years), a sudden turnaround in recruit­ trend noted in Vaughan et al. (1986). for Gulf menhaden (both biomass and ment could still present problems to the Chapoton (1972) obtained an estimate index of egg production) have been fishery. of MSY of 430,000 t for the 1946-70 generally much larger (20-50%) and with A yield-per-recruit isopleth shows period, Schaaf (1975) obtained an

53(4), 1991 55 estimate of 478,000 t for the 1946-72 effective fishing effort has actually nual production, fishing effort, and fleet period, Nelson and Ahrenholz (1986) declined since the 1960's (Vaughan, size appear reasonably balanced. Al­ obtained estimates ranging from 540,000 1987). The implication is that the in­ though recent harvests have declined to 640,000 t forthe 1946-79 period, and creased landings since 1978 are the result rapidly (from 894,000 t in 1987 to more recently Vaughan (1987) obtained ofexceptionally good recruitment (i.e., 570,000t in 1989, it was not considered estimates ranging from 620,000 to year classes) and not increased effective likely that the high landings (above MSY) 825,000 t. The primary biological con­ fishing effort. Increased geographic obtained during the 1980's could be cern raised by stock assessment scientists availability of Gulf menhaden to the maintained indefinitely. Landings on the is that the nature ofthe descending limb fishery does not seem likely given the order of 600,000 t are probably more on the surplus production model can closure of reduction plants at the realistic as a long-term average. only be determined accurately if land­ geographic extremes (Nicholson, ings exceed the current MSY for sev­ 1978b), and area and seasonal closures Summary eral years. Ifthe descending limb were that have been implemented (Christmas In conclusion, management programs steep, heavy fishing could put the stock and Etzold, 1977; Christmasetal., 1983, are in place for both Atlantic and Gulf at greater risk. The Pella and Tomlinson 1988). menhaden stocks through the Atlantic (1969) model, estimated from Gulfmen­ Historical estimates ofmaximum sus­ and Gulf States Marine Fisheries Com­ haden catch and effective effort data, tainable yield (MSY) range from 430,000 missions. Cooperation is ongoing and has a flat descending limb (Vaughan, to 585,000 t (Chapoton, 1972; Schaaf, seems to work on the Gulf of Mexico 1987). 1975; Nelson and Ahrenholz, 1986). Coast, but the Atlantic menhaden plan is Estimates ofMSY range from 620,000 With data through the 1984 fishing year, not fully implemented. Reevaluation of to 700,000 t based on surplus production surplus production models produced the management options for the Atlantic models (landings and fishing effort data estimates of MSY from 620,000 to menhaden reduction fishery is underway from 1946-85 fishing years), and from 700,000 t (Vaughan, 1987). Recent re­ and may replace the variable season 705,000 to 825,000 t based on popula­ cruitment (1976-82) has been excellent closure management recommendation. tion simulation models (using Ricker and indicates potential yields of about The expansion of fishing on the spawn­ spawner-recruit relationships based on 718,000 t based on a yield-per-recruit ing stock in New England waters concur­ 1964-82 yearclasses) (Vaughan, 1987). analysis. Employing population simula­ rently with increasing fishing pressure on The latter range ofestimates for MSY is tion models with a spawner-recruit rela­ prespawning menhaden offVirginia and probably more indicative ofthe average tion produced estimates of MSY ranging North Carolina in the fall prompts con­ landings that could have been removed from 705,000 to 825,000 t. In general, cern for maintenance of the Atlantic from the Gulf menhaden stock during MSY has been exceeded by recent Gulf menhaden resource. the period from which data were ob­ menhaden landings ranging from tained (1964-85 fishing years), given the 552,600t0982,8ootduring the 1980's. Acknowledgments limitations in adjusting the fishing ef­ Landings since 1987 have dropped below fort (restricted to 1964-83 fishing years) 800,000 t to 570,000 t in 1989. Thanks are due to the many past and used in surplus production models and The Gulfmenhaden is short-lived and present employees of the Beaufort Lab­ the firmer biological basis for the pop­ has a higher natural mortality than the oratory, Southeast Fisheries Science ulation simulation approach. Our major Atlantic menhaden. Both estimated Center, NMFS, NOAA, who were in­ concern was that the high landings greater spawning stock size (Table 5) and spawn­ volved in the many aspects of the men­ than 800,000 t during most ofthe 1980's ing stock ratio (Fig. 10) appear healthy, haden program, especially Dean W. may have been too high. but a rapid change in favorable conditions Ahrenholz, Donnie L. Dudley, Ethel A. could alter this picture rapidly, so caution Hall, Eldon J. Levi, and Joseph W. Management Implications is advised relative to the high F's found Smith. We are grateful to Charles W. Both landings and fishing effort have and the dependency of the fishery upon Krouse for his assistance in preparing the increased dramatically since 1946 (Smith very few age groups. Hence, expansion computerized data sets for analysis and et aI., 1987a), with record landings dur­ of this fishery by effort or area is not for other programming help. We would ing the 1984 fishing season (982,000 t) recommended. Concern is therefore also like to thank Dean W. Ahrenholz and and record high nominal fishing effort raised with the operation ofa new reduc­ Joseph W. Smith, for their many helpful during the 1983 fishing season (655,800 tion plant in 1989 adjacent to a defunct comments, and Herbert R. Gordy and vessel-ton-weeks). Recruitment to age-1 site near Morgan City, La., and expan­ Curtis W. Lewis for drafting offigures. (on 1 January) varied between 8.3 and sion ofthe fishing season for bait fisheries Literature Cited 41.8 billion fish (Table 5). Many of the in Florida and Louisiana. higher values have occurred since 1977 , In summary, the Gulf menhaden fish­ Ahrenholz, D. W. 1981. Recruitment and exploita­ tion of Gulf menhaden, Brevoortia patronus. producing greater values for population ery is currently fully exploited and ap­ Fish. Bull. 79:325-335. size and numbers ofspawners (Tables 4, pears reasonably stable biologically in ___ , W. R. Nelson, and S. P. Epperly. 1987. Population and fishery characteristics of 5) and smaller values for exploitation rate view of the age composition, life span, Atlantic menhaden, Brevoortia tyrannus. Fish. (Table 4) in the last decade. Therefore, and effects ofenvironmental factors. An­ Bull. 85:569-600.

56 Marine Fisheries Review AMAC. 1982. Status and management recommen­ Bull. 71:719-734. Schaaf, W. E. 1975. Status ofthe Gulfand Atlantic dations for the purse-seine fishery, report to the Gabriel, W. L., W. J. Overholtz, S. A. Murawski, menhaden fisheries and implications for re­ Atlantic Menhaden Implementation Subcommit­ andR. K. Mayo. 1984. Spawning stock biomass source management. Mar. Fish. Rev. 37(9): 1-9. tee (AMIS). Atl. Menhaden Advis. Committee, per recruit analysis for seven Northwest Atlan­ 1979. An analysis of the dynamic Atl. States Mar. Fish. Comm., Wash., D.C., tic demersal finfish species. NMFS Woods Hole response ofAtlantic menhaden, Brevoortia tyran­ 15 p. Lab. Ref. Doc. 84-23,36 p. nus, to an intensive fishery. Rapp. P.-v. Reun. AMMB. 1981. Fishery management plan for Atlan­ Lewis,R. M.,D. W. Ahrenholz,andS. P. Epper­ Oms. Int. Explor. Mer 177:243-251. tic menhaden, Brevoortia tyrannus (Latrobe). ly. 1987. Fecundity of the Atlantic menhaden, _,.,.-,-...,.---_ andG. R. Huntman. 1972. Effects of Atl. Menhaden Manage. Board, Atl. States Mar. Brevoortia tyrannus. Estuaries 10:347-350. fishing on the Atlantic menhaden stock: Fish. Comm., Wash., D.C., Fish. Manage. Rep. ____ and C. M. Roithmayr. 1981. Spawn­ 1955-1969. Trans. Am. Fish. Soc. 101:290-297. 2,134 p. ing and sexual maturity of Gulf menhaden, Smith,J. W. 1991. The Atlantic and Gulfmenhaden 1986. 1986 supplement to Atlantic Brevoortiapatronus. Fish. Bull. 78:947-951. purse seine fisheries: Origins, harvesting menhaden fishery management plan. Atl. Nelson, W. R., and D. W. Ahrenholz. 1986. technologies, biostatistical monitoring, recent Menhaden Manage. Board,Atl. States Mar. Fish. Population and fishery characteristics of Gulf trends in fisheries statistics, and forecasting. Mar. Comm., Wash., D.C., Fish. Manage. Rep. 8, menhaden, Brevoortia patronus. Fish. Bull. Fish. Rev. 53(4):28-39. 61 p. 84:311-325. ____ ,E.J. Levi,D. S. Vaughan,andE. A. Blomo, V. J. 1987. Distribution ofeconomic im­ ____ , M. C. Ingham, and W. E. Schaaf. Hall. 1987a. The Gulf menhaden, Brevoortia pacts from proposed conservation measures in the 1977. Larval transport and year-class strength of patronus, purse seine fishery, 1974-1985, with U.S. Atlantic menhaden fishery. Fish. Res. Atlantic menhaden, Brevoortia tyrannus. Fish. a briefdiscussion ofthe age and size composition 5:23-38. Bull. 75:23-41. of the landings. U.S. Dep. Commer., NOAA Chapoton, R. B. 1972. The future of the Gulf Nicholson, W. R. 1978a. Movementsandpopula­ Tech. Rep. NMFS 60, 8 p. menhaden, the United States' largest fishery. tion structureofAtlantic menhaden indicated by ____ , W. R. Nicholson, D. S. Vaughan, Proc. GulfCaribb. Fish. Inst. 24: 134-143. tag returns. Estuaries I: 141-150. D. L. Dudley, and E. A. Ha11.1987b. The Atlan­ Chester, A. J. 1984. Sampling statistics in the At­ ____ . 1978b. Gulf menhaden, Brevoortia tic menhaden, Brevoortia tyrannus, purse-seine lantic menhaden fishery. U.S. Dep. Commer., patronus, purse seine fishery: Catch, fishing ac­ fishery, 1972-I984, with a briefdiscussion ofthe NOAA Tech. Rep. NMFS 9,16 p. tivity, and age and size composition, 1964-73. age and size composition oflandings. U.S. Dep. Christmas, J. Y., and D. J. Etzold. 1977. The U.S. Dep. Commer., NOAA Tech. Rep. NMFS Commer., NOAA Tech. Rep. NMFS59, 23 p. menhaden fishery ofthe GulfofMexico, United SSRF 722,9 p. Vaughan, D. S. 1987. A stock assessment of the States: A regional management plan. GulfCoast Pella, J. J., andP. K. Tomlinson. 1969. A general­ Gulf menhaden, Brevoortia patronus, fishery. Res. Lab. Tech. Rep. Ser. 1,53 p. ized stock production model. Int.-Am. Trop. U.S. Dep. Commer. , NOAATech. Rep. NMFS --:-c:c:-:::-,-- , , and L. B. Simpson. Tuna Comm. Bull. 14:420-496. 58,18 p. 1983. The menhaden fishery ofthe GulfofMex­ Powers, J. E. 1983. Report of the Southeast ____ ,J. V. Merriner, D. W. Ahrenholz, and ico' United States: A regional management plan. Fisheries Center Stock Assessment Workshop, R. B. Chapoton. 1986. Stock assessment of Gulf States Mar. Fish. Comm., Rep. 8,107 p. August3-6, 1982. U.S. Dep. Commer., NOAA menhaden and coastal herrings. U.S. Dep. Com­ ____ , , , and S. Tech. Memo. NMFS-SEFC-127, 229 p. mer., NOAA Tech. Memo. NMFS-SEFC-178, Meyers. 1988. The menhaden fishery ofthe Gulf Reish, R. L., R. B. Deriso, D. Ruppert, and R. J. 46p. of Mexico, United States: A regional manage­ Carroll. 1985. An investigation ofthe population ___.,....,- andJ. W. Smith. 1988. A stock assess­ ment plan. GulfStates Mar. Fish. Comm., Rep. dynamics of Atlantic menhaden (Brevoortia ment ofthe Atlantic menhaden, Brevoortia tyran­ 18, 135 p. tyrannus). Can. J. Fish. Aquat. Sci. 42: 147-157. nus, fishery. U.S. Dep. Commer. , NOAA Tech. Dryfoos, R. L., R. P. Cheek, and R. L. Kroger. Ricker, W. E. 1954. Stock and recruitment.1. Fish. Rep. NMFS 63, 18 p. 1973. Preliminary analysis ofAtlantic menhaden, Res. Board Can. II:559-623. ____ and . 1991. Biological Brevoortia tyrannus, migrations, population ____ . 1975. Computation and interpretation analysis oftwo management options for the Atlan­ structure, survival and exploitation rates, and ofbiological statistics offish populations. Bull. tic menhaden fishery. Mar. Fish. Rev. 53(4): availability as indicated from tag returns. Fish. Fish. Res. Board Can. 191, 382 p. 56-64.

53(4),1991 57