Table of Contents Introduction Methods Results Discussion Conclusions and Recommendations References Cited
Northeast Fisheries Science Center Reference Document 11-16
Proceedings of the Pollock Ageing Workshop 13 - 14 July 2010, Boothbay Harbor, Maineby William J. Duffy1, W. Eric Gross2, Cecil Nelson2, and Sarah Emery1
1NOAA National Marine Fisheries Service, Northeast Fisheries Science Center, 166 Water Street, Woods Hole, MA 02543
2 St. Andrews Biological Field Station, Fisheries and Oceans Canada, 531 Brandy Cove Road, St. Andrews, New Brunswick, Canada ESB 219
Web version posted September xx, 2011Citation: Duffy WJ, Gross WE, Nelson C, Emery S. 2011. Proceedings of the pollock ageing workshop 13 – 14 July 2010,Boothbay Harbor, Maine. US Dept Commer, Northeast Fish Sci Cent Ref Doc. 11-16; 16 p. Available from: National Marine Fisheries Service, 166 Water Street, Woods Hole, MA 02543-1026, or online at http://nefsc.noaa.gov/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.
Pollock (Pollachius virens) is a highly migratory, schooling, amphiboreal gadoid ranging from Labrador to Georges Bank.However, it is most commonly found on the southwestern Scotian shelf and the Gulf of Maine. Pollock can reach lengths up to 130 cm and weigh up to 16 kg (Penttila et al. 1988). Pollock start maturing at 4 years old and can reach ages in excess of 20; there are no differences in growth rates between sexes (Penttila et al. 1988).
Initial ageing studies for pollock began in the middle 1950's and early 1960's using scales (Hoberman and Jensen 1962). Currently, pollock are aged by viewing sectioned otoliths, using a dissecting microscope, under reflected light. Age determinations are made by counting translucent rings from nucleus to the edge of the otolith (Penttila et al. 1988). Using the otolith as an ageing structure was validated by Steele (1963) who used otolith marginal increment analysis and later was confirmed by Neilson (2003) using mark-recapture.
The most recent pollock stock assessment (SAW 50) was conducted in June, 2010 (NEFSC 2010).Historically both Canadian and US survey age data was used because the stock overlapped between countries. The stock assessment model was updated for this assessment which included U.S. commercial age data. For this latest assessment, only US age data was used because the stock definition was "refined" to coincide with the Canadian / US boundary (NEFSC 2010). However, to account for the possibility of a combined U.S./Canada stock assessment in the future it became necessary to review the ageing practices between the two labs.
Prior to the assessment (Early May 2010), ageing material was exchanged between the Canadian Department of Fisheries and Oceans (DFO) Biological Station in St Andrews, New Brunswick, Canada and the Northeast Fisheries Science Center (NEFSC) laboratory in Woods Hole, MA, U.S.A.One motivation for holding the exchange was that ageing responsibilities for pollock at both the DFO and NEFSC has recently changed hands, and it was important to develop a working relationship between the new age readers in the event that a joint assessment is performed in the future (The current approach has separate assessments for Canadian and U.S. management units).
The results of this Canada/U.S. ageing exchange suggested that there were systematic differences in the methods used to age pollock between the two laboratories. Therefore, an ageing workshop was conducted to discuss the results of this exchange. This workshop took place in Boothbay Harbor, ME, U.S.A., at Maine's Department of Marine Resources (DMR) during July 13-14, 2010. Attendees included representatives from DFO, DMR, and NEFSC (Table 1). The objectives of the workshop were to:
1. Reconcile the differences between age readers in the exchange.
2. Review standard ageing methods between the two laboratories.
3. Demonstrate improvements in within and between reader agreement as a result of the workshop.
This report describes the results of both the exchange and the workshop, and offers recommendations for future collaboration.
Between May and June 2010, the ageing laboratories in St. Andrews and Woods Hole exchanged otolith samples from pollock. Each laboratory supplied otolith samples from both their survey and commercial fisheries: the United States supplied 50 spring commercial samples and 50 autumn survey samples; Canada supplied 85 commercial samples and 58 summer survey samples. The primary pollock age reader at each laboratory viewed the exchanged samples.
From the resulting ages, the precision between the two age readers was assessed with percent agreement and age-bias plots (Campana et al. 1995). Also, a Bowker's test of symmetry (Bowker 1948) was used to test for systematic differences between readers.
During the workshop, both agreements and disagreements from the exchange were discussed. Otolith images were taken from both the US and Canadian survey and commercial samples using a digital camera and viewed using imaging software, loaded into Power Point® and viewed using a projector. This enabled discussion between the primary age readers from each laboratory on these disagreements. After an initial discussion between primary age readers, comments were opened to all workshop participants.
A total of 243 otoliths were available for the ageing exchange, although 16 were later evaluated as low quality and not used to calculate agreement (Tables 2-5). Levels of agreement ranged between 40 and 71 percent, with systematic differences in the ages (Tables 6-9). Bowker's tests for all four sample sets were significant (P < 0.05; Tables 6-9). The U.S. age reader usually had higher ages than the Canadian age reader (Tables 6-9), with the exception of the U.S. survey samples, where the Canadian age reader had a tendency to have higher ages (Table 7).
Ten samples from the Canadian survey where disagreements occurred were discussed. Of these, five were changed to the U.S. age, four remain unresolved, and one was omitted (Table 10). A total of five otoliths where disagreements occurred were discussed from the U.S. commercial and survey samples. Two of these ages were changed to the Canadian age, two were unresolved, and one remained the same (Table 10).
During the workshop, we identified a number of issues that caused the differences in age determination between the two primary age readers, including whether to count the edge, microscope technique, and the location of the first annulus.
First, there were some issues as to whether counting the edge of the otolith was appropriate. One of the concerns with respect to counting edge material is that the Canadian Scotian Shelf survey takes place in July, and most pollock from that area already have a completely formed opaque edge (Steele 1963), where as pollock in the Gulf of Maine start laying down their annulus between March and April (which coincides with NEFSC's spring bottom trawl survey) and in the summer months have very little opaque area on the edge of their otoliths (Penttila et al. 1988). Such a difference in biological timing and dates of collection between the two areas could result in differences in the interpretation of age, based on each reader's notion of the amount of opaque material on the edge expected before another year should be counted.
For example, Figure 1 shows an otolith from a 36 cm pollock caught on the Canadian survey. The Canadian age reader assigned this fish an age of 2, while the U.S. reader assigned it an age of 3. The age was unresolved during the workshop. The U.S. age reader noted enough opaque area after the last visible annulus to count it as an annual ring, whereas the Canadian reader did not count the edge.
Interpreting the difference between a false annulus and true annulus was also an area of concern. During the workshop there were a number of occasions were the Canadian age reader interpreted the fourth annulus as a false annulus. A sample where this issue affected age interpretation is shown in Figure 2. The Canadian age reader originally aged this otolith from a 69 cm pollock, as a 7-year-old, while the U.S. age reader aged it as an 8-year-old. After some discussion, it was agreed that the fourth annulus was a true annulus and that the fish was 8 years old. The intensity of the opaque area before and after the translucent area is a good indicator when trying to determine if a translucent zone is a true annulus.
Optic standardization was another issue raised in the workshop. It is standard practice for the Canadian age reader to view an otolith under a single magnification without modification. During the workshop, this issue was raised while interpreting the edge of some otoliths. This was also an issue which effected the Canadian age reader's interpretation of the fourth annulus. The U.S. age reader is not limited to one magnification, but will adjust the magnification, especially when there is difficulty viewing the edge of an otolith. This issue was a factor in the discussion of the sample depicted in Figure 1. Raising the magnification allowed the age readers to see more opaque material after the formation of the annulus, thus allowing it to be counted as a full year.
Finally, the location of the first annulus was another issue where disagreements arose. Figure 3 shows an otolith from a 70 cm pollock where, the Canadian age reader had an age of 6, whereas the U.S. age reader had an age of 7. The disagreement lay in interpretation of the location of the first annulus. After discussion, there was agreement that the fish was 7 years old. It was agreed that the first annulus can often be identified by its irregular ("cauliflower") shape and by the presence of a thin settling check near the nucleus (Figure 3).
This pollock ageing workshop helped to examine and reconcile differences in ageing between the NEFSC and DFO. If the NEFSC is going to use both U.S. and Canadian age data in future stock assessments for pollock, consistency in age interpretation for both laboratories would be required for any analysis (i.e. for combined Canada/U.S stock assessment).
It was recommended that future ageing exchanges take place on an annual basis between the two laboratories in order to prevent discrepancies. It was also recommended that ageing workshops be conducted every few years to discuss the results of these exchanges.
Furthermore, a reference collection should be created from a combination of Canadian and U.S. age samples. These ages should be corroborated by both the Canadian and U.S. age reader, perhaps from exchange samples. The reference collection could be used to train new age readers, and it would allow age readers to test themselves before starting to age new age samples as a measure of ageing accuracy.
Bowker, A. H. 1948. A test for symmetry in contingency tables. Journal of the American Statistical Association. 43: 572-574.
Campana, S. E., M. C. Annand, and J. I. McMillan. 1995. Graphical and statistical methods for determining the consistency of age determinations. Trans.Am.Fish. Soc. 124:131-138.
Hoberman, J.M., and A.C. Jensen. 1962. The growth rate of New England pollock. Trans. Am. Fish. Soc. 91: 227-228.
Neilson, J. D., W. T. Stobo, and P. Perley. 2003. Age and growth of east coast pollock: Comparison of results from otolith examination and mark-recapture studies. Trans. Am. Fish. Soc. 132: 536-545.
Northeast Fisheries Science Center (NEFSC). 2010. 50th Northeast Regional Stock Assessment Workshop (50th SAW) Assessment Report. US Dept Commer, Northeast Fish Sci Cent Ref Doc. 10-09; 57 p.
Penttila, J., F. Nichy, L. M. Dery, and A. Jerald. 1988. Methods and equipment. Pages 7-16 in J. Pentilla, and L. M. Dery, editors.Age determination methods for northwest Atlantic species. NOAA Technical Report NMFS 72.
Steele, D.H. 1963. Pollock (Pollachius virens (L.)) in the Bay of Fundy. J. Fish. Res. Bd. Can. 20(5):1267-1314.