Stock Identification and Billfish Management Robert A. Skillman While economic and social evaluations of the unit stock concept. with many authors pro- fisheries are today recognized as important posing new or modified definitions. Providing elements in managing fisheries. determining the a historical perspective might be appropriate at biological status of fish exposed to a fishery and this time. to a management regime remains a basic and Fishery biologists first defined a stock in important element in management. For the status terms of the various tools employed in conduct- of a stock to be assessed. boundaries must be ing an assessment. For example. Cushing (1968) set. If the boundaries actually encompass more said. '*. a stock may be defined as a popula- than one stock and management is carried out tion in which the vital parameters of recruitment, as if there were only one stock. the management growth, and mortality are homogeneous." In scheme may perform satisfactorily on average practice, fishery biologists frequently identified for the two stocks. However, the more that popu- stocks based on where the fish were found as lation parameters, such as growth and mortality, larvae, as adults in the fisheries. or while spawn- differ among individual stocks, the more likely ing, as well as by how the fish looked (meristics it is that the management regime will adversely and morphometrics). Then. the existence of impact one of the stocks. If the boundaries do these putative stocks was evaluated by using not completely encompass a stock, then events estimates of vital (or population) parameters. external to the management area can signifi- With the development of pobulation genetics, cantly affect the stock. the fishery biologists' definition of a stock was This chapter reviews stock identification viewed as vague and not quantitatively rigorous methods as they relate to fisheries management compared with the geneticists' definition of a and. in particular. to billfish management. The population or a subpopulation. Thus. fishery stock concept. itself. will be discussed before biologists began incorporating genetic concepts. reviewing the various methods used to study and new definitions were proposed. For example. stock structure. Then. what is known about ". a unit fish stock is one consisting of ran- billfish stock structure will be reviewed. followed domly interbreeding members whose genetic in- by recommendations regarding the most likely tegrity persists whether they remain spatially productive research strategies to follow for and temporally isolated as a group. or whether billfishes in the future. they alternately segregate for breeding and otherwise mix freely with members of other unit The Stock Concept stocks of the same species" (Kutkuhn 1981). Setting boundaries and identifying stocks is. Along with this definition came numerous in effect. determining what part of the total attempts to identify stocks by using various species population is going to be assessed and genetic techniques. managed. Thus. the term stock or unit stock is The debate on the utility of genetic techniques used. ". to connote as well as circumscribe in fisheries stock assessment and fisheries man- the basic grouping on which management needs agement continues today. One problem is whether to locus attention" (Kutkuhn 1981). This initial a population determined to be genetically htep in a5sessment has led to many reviews ot homogeneous meets the needs of fishery bicilo- 207 708 PLANNING THE FUTURE OF BILLFISHES gists to conduct stock assessments: namely, the exchange of individuals among the sites. is unit stock must be homogeneous with respect necessq for genetic separation and could contri- to recruitment. growth, and mortality. A second bute to differences in vital parameters. The dis- problem is whether genetic techniques can re- tribution of larvae and juveniles may also be used solve the structure of a large population inhabit- to identify stocks, particularly when the samples ing an extensive area having vague and changing are collected independently from any fishery. environmental boundaries when biological Thus, using juveniles collected from the stomachs (fishery) information suggests some degree of of commercially-landed predatory fishes should stock structuring. While all the evidence is cer- be done with care. tainly not in. the genetic and fishery biological Popularion Parameters. Estimates of popula- definitions apparently coincide reasonably well tion parameters, such as growth. recruitment. for demersal species and stocks with well-defined and mortality, are used to separate stocks. In physical or environmental boundaries. On the addition, they are commonly used to characterize other hand. they apparently do not coincide well putative stocks and evaluate whether their ex- for species inhabiting, as adults or larvae. the istence seems plausible; that is. whether the pelagic ecosystem with poorly defined physical parameters are homogeneous over the range of and environmental boundaries. the reputed stock. While demonstrating that parameter estimates are different among poten- Stock Identification Methods tially separate stocks does not necessarily prove As indicated above, the study of population the existence of separate stocks. stock structur- structure in fisheries consists of two basic ap- ing is supported when geographical clumping proaches. The traditional biological approach or clinal variation of parameters occurs. Such consists of a number of alternative techniques results are consistent with a lack of genetic mix- for identifying stocks. These stocks meet the ing (subpopulation formation) and a lack of ex- fishery biologists' stock asessment needs but change of fish among fishing grounds or mixing may not be the same as genetic subpopulations. of individuals in nursery areas (stock formation) The newer genetic approach consists of a few because separation. from either perspective, but growing number of techniques for identifying could lead to different population parameters. subpopulations that may or may not be helpful Given the poor ability to estimate mortality even for stock assessment and management purposes. for well-studied stocks, however. estimates of billfish mortality are not likely to provide as Bioloqical much information as growth estimates. On the Disrriburion. Probably the first method technical side. comparison of the growth esti- employed was to examine the geographical dis- mates is complicated because the commonly tribution of fishery catches. which often consist used von Bertalanffy growth parameters. of adults. Thus. this method has direct applica- maximum size (Lint) and rate of approaching bility to defining the boundaries of a fishery, that maximum (k). are correlated. Simple cluster and possibly the boundaries of management analysis conducted by plotting LInrversus k for action. but it may not be satisfactory for deter- samples from different areas has proven produc- mining the boundaries of a stock unless the dis- tive in the identification of stocks. Other popu- tribution of fishing effort is extensive. For ex- lation parameters (e.g., the age at first maturity ample. if the fishing effort is directed at the and gonadal index) are also used. Age at first species being examined. it is not likely to occur maturity is strongly affected by stock size. which beyond the region where the species is commer- is. in turn. affected by the size of the fishery. cially or recreationally important. In such cases. Gonadal development is seasonal but may differ stocks may be distributed more extensively than among stocks. indicated. Compiling the distribution of catches Parasires. Using the occurrence of fish para- by size classes may help substantiate the ex- sites as a means of studying stock structure is istence of putative stocks when recruitment can based on the concept that parasites are naturally he shown to vary. occurring "tags": that is. the incidence of para- Determining the distribution of spawning sites sitic species varies peographically and these may rewlt in more reliable identlfication of natural tags are not lost. The method is not \tacks. The cnistt'ncc of gcographically or applied frequently to large pelagic species tor a temporally separate spawning sites. wlth no number of reasons. Fishery biologists reco,'mze STOCK IDENTIFICATION & MANAGEMENT 209 that parasitology is not a tool in the sense that when released and retaken, recaptures may in- morphometric and multivariate statistical methods dicate movement but provide little information are. but is an additional level of biological com- on stock structuring. plexity requuing research itself. This complexity involves the identification of relatively bizarre animals: use of different sampling, preservation. and preparation techniques: and. most im- Morphomerrics and Meristics portantly, familiarity with a different set of life- Fish exhibit greater observable (phenotypic) history strategies involving host-parasite relation- variation than other vertebrates, apparently be- bhips. Parasitologists have tended to work more cause they are poikilothermic (cold-blooded) in freshwater and coastal marine habitats where and have a capacity for indeterminate growth fish stock sizes and parasite-to-host linkages are (e.g., Allendorfet al 1987). Thus, early workers more amenable to reseach. Consequently, the were able to recognize stocks on the basis of life histories of parasites occurring on large appearance. and later workers built on this by