Report of the Workshop on estimation of maturity ogive in Norwegian spring spawning herring (WKHERMAT)

This report presents the result from a Workshop on estimation of maturity ogive in Norwegian spring spawning herring (WKHERMAT), held in Bergen (Norway), 1-3 March 2010. The workshop was chaired by Erling Kåre Stenevik (Norway) and 10 participants from 5 countries attended the workshop.

The workshop met its Terms of References which were to

a) compare back-calculated maturity at age estimates in Norwegian spring spawning herring with annual estimates derived directly from the ecosys-tem survey in May.

b) develop guidelines for sampling data on maturity at age in Norwegian spring spawning herring, for use in assessment, in accordance with the PGCCDBS guidelines for collecting maturity data and maturity ogive es-timation for stock assessment purpose (see PGCCDBS Report 2009, Annex 10).

Data on maturity ogive from three different sources were presented and discussions on strengths and weaknesses of the different data sources took up most of the time. The three different data sources were: a) maturity ogive used in assessment, b) survey data on maturity ogive from the Ecosystem survey in May and c) back-calculated ma-turity ogive using Gulland’s method. In addition, data on maturity cycle in Norwe-gian spring spawning herring were presented and guidelines for sampling of maturity data were discussed in accordance with PGCCDBS.

The maturity matrix used by ICES goes back to 1907. Documentation on the source of information and the justification of changes is almost absent and the lack of docu-mentation is a general problem in this data set. The data cannot be reproduced be-cause the sources are unknown and most changes which have been made in the past are not explained.

The May survey may potentially provide data to construct updated maturity ogives. The survey indicate that most (but not all) herring in the Norwegian Sea are mature and most (but not all) herring in the Barents Sea are immature. However, the time series is short and some potential problems were discovered. The deviation of a ma-turity ogive from the survey data compared to back-calculations appeared to be prob-lematic. There appear to be differences in the catchability in the survey between the Norwegian Sea (where most of the mature fish is distributed) and the Barents Sea (where most of the immature fish is found). This needs to be addressed further before data from the survey can be used for maturity ogive estimations.

The back calculation data set indicates that maturation of large year classes is slower than for others. This applies to a lesser extend to the 2002 year class. However, the estimates for this year class are suggesting that at least a correction needs to be con-sidered in the maturation assumed for this year class in the assessment by ICES. WKHERMAT considered the data set derived by back calculation as a suitable poten-tial candidate for use in the assessment because it is conceived in a consistent way over the whole time period and can meet standards required in a quality controlled process. However, the back calculation estimates cannot be used for recent years. Since the surveys do not provide suitable data at the moment, assumptions have to be made for recent year classes.