Scientific basis


Data and methods


The assessment is based on long-term catch data, fishery-independent data from two scientific surveys, and three indices of commercial catch per unit effort. The longest survey series has an apparent break in 2001 when the survey design was altered. This year’s assessment used the 4th Quarter BITS for the first time. The introduction of commercial cpue data in the 2005 assessment has made the assessment much less dependent upon the survey indices.

There is information on substantial underreporting of catch in 1993–1996, and this has also been the case since 2000. In such a situation, ICES could i) conduct stock assessments based on catch information, including information on mis- and non-reporting, or ii) use official data only, which would result in very poor or heavily biased assessments. In this situation ICES has chosen to include mis- and non-reportings in the assessment.

Estimates on the amount of misreporting are available from the national industries and control agencies. These indicate that recent catches have been around 35–45% higher than the reported figures. These estimates have been incorporated in the assessment. By nature this information is highly uncertain, and also incomplete, with no information available for some of the nations where misreporting is suspected to occur. Although the corrected landings values derived by ICES are the best possible estimates, they are likely to be only minimum values.
The eastern Baltic cod stock has been below or close to B_lim in the last 15 years and recruitment has stabilized at a low level. The reduction in recruitment might be primarily environmentally driven, and the spawning stock is likely to have decreased following the decline in recruitment rather than vice versa. As a result it is difficult to justify the existing B_lim, although for similar reasons it is equally difficult to suggest a more appropriate value. An ICES expert group recently tasked to update reference points concluded that there are inconsistencies in the definition of the fishing mortality and biomass reference points presently established. Work is ongoing to evaluate the reference points.

Discard data have been available since 1996 and are applied in the assessment as yearly proportions discarded per age-group. Before 1996, an average proportion discarded per age-group estimated for 1996–2003 is applied. The season and area coverage of discard sampling requires improvement. A relationship between year-class strength and discard rates cannot be estimated from the available data. Due to recent changes in technical regulations, e.g. increase of minimum landing size, the introduction of the BACOMA 110 mm codend, and varying closures, discard rates may have additionally varied.

Information from the fishing industry


Some of the information on misreporting came from industry sources. Misreporting could occur between the western and eastern area after the introduction of a separate management in 2005.

The 110-mm ‘BACOMA’ codend has been much more widely accepted than its 120-mm predecessor.

Uncertainties in assessment and forecast


Problems with misreporting, age-reading, and a new survey design result in a very uncertain assessment. Adding yet another year’s data with a large proportion of the catch being non-reported means that all year classes now occurring in this stock are subject to large uncertainties.

There are large inconsistencies in age determination for this stock as a result of the lack of clear growth rings in the otoliths. This results in poor quality catch-at-age and survey data and a likely underestimation of the fishing mortality (Reeves, 2003). However, these ageing problems are thought to have no major effect on the perception of the state of the stock with respect to the precautionary reference points. ICES has attempted to resolve the inconsistencies in age reading for this stock, but no consensus on the interpretation of age readings could be reached. An EU-funded study starting in 2007 (DECODE) will take a different approach tto deliver validated data for the assessment.

There are some indications from surveys that the 2003 year class is stronger than any other year class in the past 15 years. The results of the assessments point to stronger year classes in 2003 and 2004. Problems with the catch and survey data as well as inconsistent age determination make it difficult to determine how strong these year classes are. The 2003 year class could make a major contribution to the catch in 2007 and spawning stock in 2008, so estimates of these quantities are sensitive to the estimated strength of this year class.

Environmental conditions


Cod distribution in the Baltic is affected by environmental conditions, specifically a lack of oxygen. This is taken into account in the way the surveys results are raised.