Tagging along: association among individual Pacific herring (Clupea pallasi) revealed by tagging1

More than 570 000 Pacific herring (Clupea pallasi) were tagged with external anchor tags during 429 tag release sessions between 1979 and 1992. Individually numbered tags were released in quantities of 1000–2000 at a time and recovered from commercial fisheries. Often several tags were recovered at the same time and place, and some recoveries occurred as “matches”, where two or more tags from a single release session were recovered together. We tested the hypothesis that the frequency of matching tag recoveries occurred by chance through random mixing of tagged herring before their recapture during fishing operations. The alternative is nonrandom, positive association among tagged individuals that persisted through time and during migrations. We used a statistical method developed to address a similar question in steelhead trout (Oncorhynchus mykiss). In separate tests, we examined tag recovery data from migratory stocks in five major regions of the British Columbia. The results indicate nonrandom association of herring for periods of 6 months to several years and through migrations over considerable distances. Résumé : Plus de 570 000 harengs du Pacifique (Clupea pallasi) ont été marqués à l’aide d’étiquettes-ancres externes au cours de 429 sessions de marquage de 1979 à 1992. Les poissons marqués individuellement ont été libérés en groupes de 1000–2000 à la fois et recapturés au cours de la pêche commerciale. Souvent, plusieurs étiquettes étaient récupérées simultanément au même endroit et quelquefois il y avait des “appariements” lorsque deux étiquettes ou plus provenant de la même session de libération étaient récupérées en même temps. Nous avons éprouvé l’hypothèse selon laquelle la fréquence de récupération de ces étiquettes appariées se fait au hasard, le résultat d’un mélange aléatoire des poissons marqués avant leur capture durant la pêche. L’hypothèse de rechange est qu’il existe une association positive non aléatoire entre les individus marqués qui persiste dans le temps et se maintient durant les migrations. Nous avons utilisé une méthode statistique mise au point pour résoudre un problème semblable chez la truite arc-en-ciel anadrome (Oncorhynchus mykiss). Des tests séparés nous ont permis d’examiner les données de recapture des stocks migrateurs provenant de cinq régions majeures de la Colombie-Britannique. Nos résultats confirment l’existence d’une association non aléatoire des harengs qui dure pour des périodes de 6 mois à plusieurs années et qui persiste au cours de migrations sur de grandes distances. [Traduit par la Rédaction] 1968 Hay and McKinnell Introduction There is a substantial scientific literature on the subject of the biology of schooling and aggregations of herring (Clupea sp.) and other fishes. This literature covers an array of subjects such as observations and measurements of school density or aggregations, polarization of schools and shoals, stratification with depth and effect of fish size, and ecological significance of schools (e.g., general review of clupeid biology by Blaxter and Hunter 1982). In this paper, we present another perspective on herring schooling that has implications for understanding stock structure. This perspective is based on evidence from tagging data from Pacific herring (Clupea pallasi). First, we emphasize that a fundamental characteristic of herring is that they are schooling fish and spend most of their lives in schools, “shoals”, or aggregations. There may be some important biological distinctions between these terms but such distinctions are not required for this paper (see Blaxter and Hunter (1982) or Pitcher (1986) for a discussion of terminology). Instead, we present evidence that herring, tagged together and released from specific locations, have an affinity for each other and that they do not mix randomly with tagged herring released from other locations. From cursory analysis of the spatial and temporal patterns of tag recoveries, we observed that most tagged herring were recovered in groups of two or more, often originating from the same tagging release date and site. Sometimes two or more tagged fish were recovered at the same date and location, after having been tagged and released in the same operation some months or years earlier. This raises the question of whether the event was simply a coincidence or whether there might be some aspect of fish Can. J. Fish. Aquat. Sci. 59: 1960–1968 (2002) DOI: 10.1139/F02-141 © 2002 NRC Canada 1960 Received 18 January 2002. Accepted 13 September 2002. Published on the NRC Research Press Web site at http://cjfas.nrc.ca on 15 January 2003. J16708 D.E. Hay.2 Pacific Biological Station, Fisheries and Oceans Canada, 3190 Hammond Bay Road, Nanaimo, BC V9R 6N7, Canada. S.M. McKinnell. PICES, c/o Institute of Ocean Sciences, P.O. Box 6000, Sidney, BC V8L 4B2, Canada. 1Presented at the North Pacific Marine Science Organization (PICES) 10th Annual Meeting in Victoria, British Columbia, Canada, in 2001. 2Corresponding author (e-mail: [email protected]). J:\cjfas\cjfas59\cjfas5912\F02-141.vp Friday, January 10, 2003 12:24:45 PM Color profile: Disabled Composite Default screen