To raise or not to raise? Raising precision in discard monitoring data

Het verzamelen van discard data wordt gereguleerd via de EC DCF. Deze gegevens worden jaarlijks opgevraagd door diverse eindgebruikers (bijv. ICES, STECF). De afgelopen jaren is het aantal bestanden waarvoor discard gegevens zijn opgevraagd en verstrekt toegenomen. Vooral voor de minder voorkomende en/of niet-commerciële soorten is duidelijk geworden dat er zorgen bestaan over de vraag of de verzamelde gegevens voldoende zijn om voor deze soorten te worden opgewerkt. Dit project zal onderzoek doen naar de nauwkeurigheid van discardgegevens om de validatie van verstrekte totale discard data te verbeteren. Daarnaast zal dit project richtlijnen geven voor het omgaan met discard data van minder voorkomende en/of niet-commerciële soorten zoals haaien en roggen.

The discard monitoring programme uses a statistical sound principle to annually randomly divide 160 trips among a reference fleet, consisting of around 20-25 vessels. All sampled trips are assigned to their respective metiers after a trip is executed. Within a trip, two boxes of discards (one box equals approx. 40 kg) are collected during two separate hauls, thus collecting a total of approximately 160 kg of discards per trip. The ratio between total fleet effort and the sampling effort (from the reference fleet) is used as an auxiliary variable to estimate total discards by species for the Dutch demersal fleet by metier. These estimated total discards are requested through data calls by different end-users every year. The main end-users for the discards data are ICES and STECF. Where ICES has used the provided data in the assessment of a number of fish stocks, STECF incorporates the provided data into the FDI database, which is a public database that has become increasingly important the last few years. In recent years, there has been an increase in the number of stocks for which discard data has been requested and provided. This has led to an internal discussion on when the collected data is sufficient enough to be raised (e.g. sample size, variation) to fleet level and consequently provided to the end-user. This may especially play a role for the less abundant and/or non-commercial species such as sharks and rays. Therefore, this project will have an in depth look into the raising precision of discard data by researching the sample size, (seasonal) variation and accuracy within species. Rays and skates will be taken as a case study. This project will create better guidelines (i.e. based on the estimated accuracy) on when discard data are considered accurate enough to be delivered to the end-user. The aim of this project is to create a general script that can be used as a yearly routine to check the accuracy of the total raised discard data per species and metier.