Age determination of Baltic herring from whole otoliths and from neutral red stained otolith cross sections

Heikki Peltonen 0 Jari Raitaniemi 0 Raimo Parmanne 0 Jan Eklund 0 Kari Nyberg 0 Folke Halling 0 0 H. Peltonen: Finnish Environment Institute , PO Box 140, FIN-00251 Helsinki , Finland. R. Parmanne: Finnish Game and Fisheries Research Institute , PO Box 6, FIN-00721 Helsinki , Finland. J. Raitaniemi: Finnish Game and Fisheries Research Institute , Ita inen Pitka katu 3, FIN-20520 Turku , Finland. K. Nyberg: Department of Limnology and Environmental Protection , PO Box 27 , FIN-00014 University of Helsinki , Finland. J. Eklund: Vaasa Administrative Court, PO Box 204, FIN-65101 Vaasa , Finland. F. Halling: Finnish Game and Fisheries Research Institute , Parkgatan 6, FIN-22100 fax: Two methods of determining the age of herring (Clupea harengus L.) from the Bothnian Sea, in the northern Baltic Sea, were compared. Ages were estimated from whole otoliths, which is a routine method for herring age determination, and from neutral red stained otolith cross sections. There was disagreement between the methods, especially in specimens with more than five annual rings. The agreement between the determinations of different age readers was better with the cross section method and, besides, it is clear that with the whole otolith method there is considerable negative bias in old fish. Thus, more herring were classified as more than 15 years old with the cross section method. The study indicates that conventional methodology may generate considerable errors in age distributions, especially in samples which mainly consist of old fish. However, the differences in age reading produced relatively small changes in growth rate estimates, because of the slow growth of the old fish, that are most prone to age determination errors. 1054-3139/02/040323+10 $35.00/0 - Herring (Clupea harengus L.) is one of the most important species in the fishery in the Baltic Sea (Parmanne et al., 1994; ICES, 2000). Herring in the northern Baltic Sea are slow growing and there was a substantial decrease in growth in the 1990s (ICES, 2000). Factors affecting herring growth rate include top-down processes, i.e. decreased predation on herring due to collapse of the cod stock (Beyer and Lassen, 1994), and bottom-up processes due to changes in zooplankton community accompanying declining salinity (Flinkman et al., 1998). The biomass of herring in the Baltic Sea has been decreasing concomitantly to the decrease in growth rates. More restrictive catch quotas have been set to avoid further reduction in stock biomass (International Baltic Sea Fishery Commission, 2000). Management of diminishing populations demands reliable estimates of stock dynamics. With the presently applied age structured assessment models such estimates are possible only if the age determination is reliable (Rivard, 1989; Hilborn and Walters, 1992). However, variability in Baltic herring age determination is known to be high and several attempts to evaluate the quality and to reduce variability in age determination have been undertaken (ICES, 1975, 1986, 1997, 1998; Eklund et al., 2000). Herring is a key species in the ecosystem in the Baltic Sea both as a prey for predatory fish and as a consumer of planktonic organisms (Rudstam et al., 1994). The Baltic Sea is severely polluted with harmful chemicals. These substances accumulate in the tissues of herring and move further to predatory fish, birds, and mammals, and to people consuming fish (Parmanne et al., 1994; Kiviranta et al., 2000). The concentrations of several toxic substances in herring are age-dependent (P. Vuorinen, R.P., T. Vartiainen, M. Keinanen, H. KiViranta, O. Kotovuori, and F. Halling, unpublished results). Accurate age determination of herring is needed in order to analyze the bioaccumulation processes of these toxic substances in the food web. Herring was first aged from the otolith by Jenkins (1902), who demonstrated that the whole otolith method should be considered valid for young Baltic herring. At present, Baltic herring is routinely aged by counting the growth zones in whole sagittal otoliths in surface light under low magnification (Fetter et al., 1991). Hyaline layer is assumed to represent the yearly period of slow growth. Despite the recognized importance of the need to validate ageing methods (Francis, 1995; Beamish and McFarlane, 2000), Baltic herring otolith ages have not been validated for old fish because of the practical problems involved, e.g. difficulties in obtaining known-age material for age determination. In otolith growth, accretion of new material initially takes place over the entire surface of the otolith and growth zones are formed around the entire margin. Because the otolith is flat and elongated, this stage of otolith growth will result in distinct growth zones on the otolith edge. After the initial growth stage, growth on the otolith edges gradually slows down and in old fish, the otolith grows only in thickness (Power, 1978; Hamrin et al., 1998; Raitaniemi et al., 2000). In old individuals of herring and several other species, growth zones are deposited mainly around the sulcus acusticus (Blacker, 1974; Zhang and Moksness, 1992). Because of this growth pattern, the later yearly growth zones of old fish tend to be difficult to discern in whole otoliths. Also, increasing thickness of the otolith tends to obscure the early growth zones. Observations by the authors that for old and slowgrowing individuals of several fish species, otolith cross sections gave higher age estimates than whole otoliths (Raitaniemi et al., 2000) pointed to the possibility that Baltic herrings might be older than believed. The observation by Eklund et al. (2000) that between-reader variation of Baltic herring age estimates increased with the size of the fish also supported this hypothesis. Preliminary tests with Baltic herring otoliths indicated that even in fish less than ten years old, one or two additional annual rings may be distinguished on the cross section, compared with the whole otolith age. We compare the standard method of herring age determination from whole otoliths with age determination from neutral red stained otolith cross sections. Fish ages, sample age distributions, and lengths-at-age obtained by different readers and methods are com61 14.2 N 20 34.0 E 61 14.8 N 19 32.2 E 61 57.9 N 20 34.2 E Number of sampled fish pared. The relevance of the findings of this study on herring stock assessment and management is briefly discussed. Material and methods A random sample of 111 herring was taken from three pelagic trawl hauls performed during a hydro acoustic survey in the Bothnian Sea, in the northern Baltic Sea in October 1999 (Table 1). The fish were measured (total length, mm) and weighed (total weight, g) and the sex and gonad maturity stage were identified. From each fish, both sagittal otoliths were removed. One of each pair of otoliths was studied whole while the other one was cut and stained. The whole otoliths were mounted on transparent, (...truncated)