Reliable measures of fish length-at-age are needed to estimate the biological reference points used in fish stock assessment. These data are usually obtained by counting and measuring seasonal growth marks in otoliths or other calcified structures for a representative sample of the population. For anglerfish (Lophius spp.) and hake (Merluccius merluccius) in the Northeast Atlantic, traditional age-estimation methods are unreliable. The assessments for these stocks are based on survey indices or are length-based and require accurate information on length-at-age. Without this information, accurate assessments and the estimation of MSY reference points are problematic.

For some other species, objective methods of age-determination have been developed using seasonal patterns in chemical composition across otolith transects. This project (Age validation of Hake and Anglerfish: AHA project) investigated the feasibility of using this approach in age determination for selected stocks of anglerfish and hake with a view to improving the growth models used in stock assessment. Validation of the annual periodicity of chemical signals was conducted using existing collections of chemically tagged hake (from ICES divisions 7 and 8a, and from 8c and 9a), and through analysis of edge chemistry in otoliths and illicia of white anglerfish (Lophius piscatorius) (from ICES Divisions 7b-k and 8a,b,d). A comprehensive assessment of a range of approaches was carried out to establish the optimal methods for sample preparation, microchemical analysis and data analysis.

The most effective approach to detecting annual signals was high resolution profiling of trace elements (elements present in minute amounts) across otolith growth transects using laser ablation inductively coupled plasma mass spectrometry (LA ICPMS). This technique collects data by ablating the otolith surface with a laser and passing it to a mass spectrometer where the quantity of elements in the ablated material is determined. As the laser moves across the otolith from the core to the edge, changes in otolith elemental concentrations from hatching until capture are recorded. These patterns reflect fluctuations in the environment or the fish’s internal state and may display annual cycles. By collecting data continuously (within the limits of machine speed), the LA ICPMS approach ensures that peaks are recorded at a sufficiently high resolution to be detected.

Concentrations of strontium and sodium at the edge of anglerfish otoliths showed small but significant levels of seasonal variation that could provide an indicator of age. Variability was also detected between size groups and areas, which could make annual signals more difficult to detect. Here, statistical treatment of the data (detrending using a general additive model smoother) helped to elucidate age related patterns. Microchemistry analysis of chemically tagged hake otoliths revealed pronounced fluctuations in phosphorous concentrations. When these patterns were examined in relation to the chemical tag in the otolith (known time point on the otolith growth transect) they appeared to have an annual periodicity in some fish. However, there was a lot of variability between fish and sub-annual fluctuations also occurred in the profiles.

A novel method was developed for detecting annual cycles in microchemistry data, which incorporates properties of Von Bertalanffy growth. This approach performed well relative to other methods when applied to simulated and real microchemistry data. The technique was used to estimate age of anglerfish and hake from strontium and phosphorous profiles respectively.

The mean estimated growth rate of 14 chemically tagged hake was 0.38mm-1.day compared to an actual mean growth rates of 0.42mm-1.day. The absolute mean percentage error rate was 23%. Estimated back calculated lengths at age across the growth transect were reasonably consistent with the current growth model used in the stock assessment and corroborate existing evidence that the pervious international ageing criteria underestimated growth rates. The microchemistry based age estimates of length at age 3 were lower than the growth model predictions, however this comparison is based on a small sample size (N=13). Although the accuracy of Draft Final report EASME/EMFF/2016/1.3.2.7/SI2.762036 11 microchemistry based age estimation of hake is moderate, the technique could be useful for investigating growth rate variability between sexes and across years, thus improving the accuracy of the stock estimates and reference points. Future work should focus on applying the approach to a larger sample size that is more representative of the population size structure to enable robust comparison of the current growth model and microchemistry based age estimates.

For anglerfish, age estimates derived from the analysis of otolith microchemistry profiles tended to be higher than ages estimated from counts of growth marks on illicia. The microchemistry method also produced lower estimates of growth than the method that is used to split survey length frequency data into pseudo age classes prior to stock assessment; the mean difference in the age assignments was 0.42 years (+ 0.09 95% confidence interval). The microchemistry based method for age determination of anglerfish may overestimate age; further work is needed to assess accuracy. This could be achieved using daily increment analysis (to confirm the timing of the first microchemistry cycle) and chemical tagging (which could be incorporated into planned international tagging programmes for anglerfish). The length splitting approach supports an age structured assessment for white anglerfish and is more robust than direct ageing methods. Nonetheless, given the discrepancy between the microchemistry based age determination method and the cohort splitting approach, it is recommended that implications of growth overestimation are evaluated within a management strategy evaluation framework.

For black anglerfish (Lophius budegassa) and for other stocks of white anglerfish which do not have an age structured assessment, microchemistry based age estimation, in combination with other direct ageing methods could be used to estimate growth parameters and to evaluate the accuracy of stock estimates and reference points. Given their biological similarity, it is likely that the microchemistry patterns observed in white anglerfish will be detectable in otoliths from black anglerfish and other stocks of white anglerfish. Microchemistry based age estimation could be applied to these stocks using the methodologies developed here.

The AHA project has demonstrated for the first time that otolith composition of anglerfish and hake varies seasonally and has identified the optimal methods for measuring age related microchemistry patterns in these stocks. A novel statistical technique has been developed to process these time series while incorporating their biological properties. The patterns identified in the otolith elemental profiles show promise as indicators of age and provide some insight that can inform the assessment of the stocks. The application of microchemistry techniques to age determination is a relatively new technique; as these methods develop, rigorous testing and refinement are required and sources of error should be quantified. The microchemistry based age determination methods presented here are resource intensive and are unlikely to provide a routine method of age estimation. Nonetheless, they have practical application for corroborating other sources of data describing age and growth and for investigating variability in growth rates. It is recommended that the methods presented here be applied within a large sampling program that ensures full representation of size structure, maturity stages and sexes in the stocks of interest and that incorporates tag recapture. This would facilitate a comprehensive evaluation of growth rate variability, information that is currently lacking from the stock assessment.

Keywords: age validation, illicium, otolith microchemistry, seasonality, anglerfish, hake, Lophius piscatorius, Lophius budegassa