Our State-of-the-Art Otolith Sectioning System
Scientific evaluations of fish populations in the northeast United States rely heavily on age data, and the Fisheries Biology Program provides ages for several fish stock assessments each year. In the last few decades, there has been a trend for our group to provide increased numbers of ages in the same amount of time, without the addition of more staff for processing or ageing fish. Instead, we have succeeded in making several improvements to our efficiency, including the 2003 purchase of a high-speed otolith sectioning (OS) system from Benetec Ltd.* in the United Kingdom. At the time, similar systems were being used in ageing laboratories in Europe, but this was the first of its kind in North America and only the eighth in the world.
Our traditional sectioning method had been to align one otolith on a cardboard tag and mount it in paraffin wax. This tag was then inserted into a custom-made chuck on an Isomet* low-speed saw. Two closely spaced diamond-impregnated blades, separated by a spacer, were used to cut a thin section. This section was removed, placed in paper, and then returned to the envelope from which it came. More information on this method can be found in Penttila and Dery (1988) or in our Ageing Manual online.
The high-speed OS system allows us to mount and section many otoliths at once; this reduces processing time, and the layout of otolith sections allows the samples to be aged more efficiently. With this system, rows of otoliths are embedded between two layers of polyester resin in aluminum molds. In advance of mounting, the otoliths are marked at the nucleus, and the bottom layer of resin is prepared. The marks on the otoliths are then aligned with notches on the molds with the use of a video camera and monitor. Eight to fifteen otoliths can fit into each row, with six rows per block and two blocks per mold. The top layer of resin is later poured over the otoliths. After it hardens, lines are etched onto each block to indicate the position of the nuclei.
The blocks of resin are then cut using a high-speed saw with one thin metal-bonded diamond blade. Two cuts are made for each row of otoliths, and one extra cut is made to reduce the size of the last piece with embedded otoliths. This yields six thin strips of sections and seven thick chunks with otolith halves on one or both sides. The six strips are then glued onto a Plexiglas slide for viewing, and the chunks are placed into a small plastic bag for use if needed.
A great amount of time is saved by this system, merely because so many otoliths can be sectioned simultaneously. It takes about 8 minutes to make the thirteen cuts in one block of otoliths. With the traditional method, mounting and sectioning could take up to 5 minutes per otolith. Admittedly, three days are needed to complete the entire process (from preparing molds to gluing the sections onto Plexiglas), and the mounting and cutting phases are most effective with two people. However, these steps only take a few hours in a given workday. The resin needs time to harden, which means there are long periods of inactivity while other projects can be accomplished. It is possible to prepare, mount, and section as many as 20 blocks in one week. This accounts for 1,000 to 1,800 fish, which would have taken a few weeks of dedicated work on the traditional system.
Furthermore, the OS system eliminates the need for fine-tuning of the saw, as was required by the traditional method. The dual blades on the Isomet saw regularly require extensive adjustment to attain the proper thickness for each species and to avoid breaking sections. With the OS system, otoliths of all species are mounted in the same way, and the sectioning machine has only one blade.
Another advantage is the ease of ageing samples mounted on Plexiglas slides. With traditional individually-sectioned samples, age readers would have to remove each section from its envelope, but now they can view many sections at once. This is both more efficient and more ergonomic. Long-term archival of sections is also more effective: there is less breakage after the samples have been sectioned, because the resin mounting medium stabilizes the sections better than the traditional wax, and there is no handling of samples individually.
Comparable Age Data
Age determinations are comparable between sections prepared with the high-speed OS system and the traditional sectioning method. To verify this, a study was undertaken using haddock from the 2002 fall survey. Both otoliths from 219 haddock were sectioned, one otolith with each method, and one person aged all the samples using standard criteria. Precision of ages obtained with the two sectioning methods was measured by the total coefficient of variation (CV), which is the ratio of the standard deviation to the mean averaged over all the fish (Chang 1982). Of the 219 fish, 198 were successfully aged with both sections. Ages ranged from 0 to 9 years, with a high level of agreement (90.4%). The total CV for all samples was 2.15%. These estimates of ageing precision passed our standards for precision (over 80% agreement and under 5.0% CV). The age-bias plot reveals a slight underageing bias at age 4 for the Isomet samples (average OS age 3.78± 0.14 years).
Upon re-examination of the 19 fish where the ages differed, most fish (13) were changed to the age seen on the Isomet section, while four fish were changed to the age seen with the OS system. The source of the disagreements was largely because the OS section was harder to read: either it appeared checkier, or resolution of annuli near the edge was more difficult. There were also issues with lost or broken sections, but this has become less of a problem as we have gained experience with the OS system.
Overall, the high agreement level and low CV seen in the comparison study indicate a high level of precision (i.e., repeatability) between the two sectioning systems. No conversion is necessary to incorporate ages obtained with the OS system into the age database.
A Few Caveats
Some species are not suitable for sectioning on the OS system, including Acadian redfish and silver hake. In the case of redfish, the first annulus is quite small, requiring greater care to be taken in aligning the nucleus for sectioning. In silver hake, extra-thin sections are needed to improve clarity. For these species, we continue preparing sections with the Isomet system, despite the decreased efficiency.
Sections cut with the OS system may differ in appearance from traditional sections, as observed in the comparison study. Ageing errors due to this may be avoided with experience, but it is also good practice for the age reader to confirm age determinations by viewing the resin-mounted otolith halves as well as the sections. These halves can resolve issues with edge type and general checkiness, as well as acting as a backup if the section is lost or broken.
Problems with broken and lost sections have largely been resolved, thanks to greater experience in avoiding trapped air bubbles near the mounted otoliths. After a few years, we are seeing that breakage of sections mounted with the OS system is less frequent than was the custom with the Isomet system.
We regard the high-speed OS system as a step forward in modernizing and streamlining age determinations for several of the fish species that we routinely age. This system enables us to prepare otolith sections much more efficiently, and the Plexiglas-mounted sections are easier to handle and store. In addition, these sections reveal ages nearly identical to our traditional processing method. All of these advantages allow our laboratory to provide more age data to fish stock assessments and other groups using our data.
Campana S.E., Annand M.C., and McMillan, J.I. (1994) Graphical and statistical methods for determining the consistency of age determinations. Transactions of the American Fisheries Society 124:131-138.
Chang, W.Y.B. (1982) A statistical method for evaluating the reproducibility of age determination. Canadian Journal of Fisheries and Aquatic Sciences 39:1208-1210.
Penttila J. and Dery, L.M. (1988) Age determination methods for Northwest Atlantic species.
Report No. NOAA-TR-NMFS-72, National Marine Fisheries Service, Woods Hole, MA (USA).