Report of the Workshop on Sexual Maturity Staging of Herring and Sprat (WKMSHS)

The ICES Workshop on Sexual Maturity Staging of Herring and Sprat (WKMSHS) was held 20 to 23 June 2011 at DTU Aqua in Copenhagen, Denmark. A total of 40 persons from 15 countries participated. The workshop was chaired by Jonna Tomkiewicz, Rikke Hagstrøm Bucholtz, and Maria Krüger-Johnsen, DTU Aqua.  
The purpose of the workshop was to elaborate standardised maturity scales for herring and sprat for common use among laboratories and evaluate sampling strategies and timing for accurate classification of maturity. The establishment of standardised maturity scales included identification of reliable maturity determination criteria for fe-males and males and comparison of interpretation of stages according to existing scales.  
The background for the workshop included doubts and difficulties in the maturity staging of herring, which have been subject for discussion since the end of the 1990s. Particularly in the North Sea, different measures have been taken to improve classifi-cation and consistency. However, inaccuracy in herring maturity staging prevails in spite of these previous efforts and its relatively simple reproductive strategy, being a determinate, total spawner. For sprat, being an indeterminate batch spawner, in particular stages relevant to estimation of spawning frequency for application in, e.g. the Daily Egg Production Method (DEPM), needed attention. Besides inconsistency in staging criteria, a difficulty in accurate maturity staging of herring and sprat appears to be that classification is often performed using frozen samples in poor condition.      
In order to identify and validate criteria for maturity determination of herring and sprat, preparation for the workshop included sampling of gonads from herring and sprat at different times of the year to cover the reproductive cycle of different stocks. Sampling was conducted by the participating institutes on board research vessels and supplemented by harbour sampling. It included visual judgement of maturity stage and photography of fresh gonads and preservation of each sample for histological examination. The gonads sampled were processed histologically at DTU Aqua, IMR, and SLU, in order to reveal microscopic characteristics that could be applied to accurately assess gonad development and validate maturity stages.
Standardised maturity scales for herring and sprat were established, using the histo-logical validation of samples and macroscopic photos of the sampled female and male fish to define criteria and describe histological and macroscopic maturity stages. For herring, the interpretation was challenged by the fact that female herring in the devel-opment stage that usually characterise onset of the seasonal reproductive cycle (eviddenced histologically by cortical alveoli, CA) occurred year round. Thus, this characteristic (CA) does not appear valid as an indicator of the onset of the reproductive cycle in herring and the classification was revised accordingly. For sprat, small gonad size and the batch spawning strategy with several cohorts of eggs being developed simultaneously and batches spawned successively over a long period of time were the main challenges for standardising a maturity scale. In particular, the inability to distinguish early developing specimens (characterised by CA) from immature specimens, and maturing from re-maturing specimens was considered, and a simplified scale elaborated based on visible gamete development. In addition, spawning active specimens were characterised to allow identification of specimens with hydrated eggs for use in estimation of the spawning fraction. 

The established maturity classification scales for both species and sexes are compatible with all existing maturity scales applied by the participating countries, however, the interpretation of stages in relation to sexual maturity changed in many cases. Emphasis was on the interpretation of maturity stages of herring in a mixed stock context. Par-ticularly in the North Sea, the mixing of stocks represents an additional problem, which is independent of the accuracy of the staging. As a consequence, the interpretation of sampled records in relation to their use in the estimation of specific spawning stocks depends on the survey and time of the year. Maturing specimens caught in June (acous-tic survey) are likely to represent the autumn spawning herring stock, while recovering specimens belong to the spring spawning herring stock. Similarly, some stocks are as-sessed after the spawning period and in these cases, is equally important to distinguish between maturing, spent and regenerating stages, but with only the latter relevant for assessing the proportion of adult fishes participating in spawning.
The established standardised maturity scales of herring and sprat includes 6-stages for both sexes, which are comparable to the existing 6-stage IBTS and BITS scales, although the interpretation of stages may differ.  
For herring, the new scale includes:  I Immature, II Maturing, III Spawning, IV Spent, V Regeneration (including regressing and skipped spawning), and VI Abnormal. In particular, the division between I Immature and II Maturing has changed making the absence/presence of visible vitellogenic oocytes an unambiguous character to distin-guish the stages. Stage V Regeneration renames previous Resting/Skipped spawning of the two previous scales, and two sub-stages Va) Regressing and Vb) Regenera-tion/Skipped spawning are distinguished. This allows recognition of post-spawning specimens, i.e. Va) Regressing that is applied on some cruises to assess the proportion of spawners, while Vb) during the pre-spawning and spawning season identifies skipped spawning. The latter together with VI Abnormal may serve as ecosystem state indicators.  
For sprat, the new scale includes:  I Immature, II Preparation, III Spawning (active and inactive including maturing and re-maturing), IV Cessation, V Regeneration (includ-ing skipped spawning), and VI Abnormal. In particular, specimens without visible development have been combined into Immature and Preparation, whereas the spawning stage has been sub-divided into a non-active spawning stage (maturing and re-maturing characterised by visible development of gametes) and an active spawning stage indicated by hydrated eggs/running milt. The integration of maturing and re-maturing into the spawning stage allows an accurate determination of maturing and spawning specimens and reliable assessment of the spawning fraction of the population.  
Conversion schemes aligning old and new scales were established for maturity scales presently applied by the participants to convert existing stage records and interpret former time series.
Illustrated manuals were elaborated for both species on a preliminary basis including the available stages. Macroscopic maturity staging of herring and sprat is best made on fresh material as dead specimens quickly deteriorate. Use of frozen material is prone to error not only in maturity determination, but also sex determination may be difficult. Hence it is recommended to avoid frozen material if possible. Histological validation proved to be a useful tool, being fast to learn and interpret for both sexes. The two staining methods H&E and Toluidine blue were both applicable. Quality assurance using histological validation would be useful on a regular basis. The same criteria can be applied to frozen tissue for both sexes, using mainly the structure of the tissues that remains intact, while the appearance of the cells changes. A preliminary manual for interpretation of histologically processed frozen gonad tissues was developed.   
For both species, emphasis was on accurate identification on the proportion of the population that belongs to a particular spawning stock through sampling of specific stages before, during or after the spawning period. In general, separation among immature specimens and adult reproducing specimens is problematic to apply in a mixed stock, and estimation of the spawning proportion in relation to stock and season was recom-mended. Similarly, optimal sampling strategies and sampling times for accurate classification of maturity can be established, but they are difficult to apply as sampling relates to specific surveys and data series that require consistency. Focus was therefore on the interpretation of maturity data primarily using existing sampling strategies, but with recommendation for exclusion of sampling during periods with no or little development of gonads. In general, revising and focusing sampling strategies to enhance data for specific purposes is recommended.