Scientific advice v Fishing Ministers – December 2019

This morning I learned the Fisheries Committee received the list of the scientific advice back in December 2019.

From this, you can start to work out how often fisheries ministers and the Commission followed the advice of scientists.

 

Source EP link and Council text

Name Area ICES Advice

Catches corresp.

to advice (tonnes)

Commission Proposal Council Agreement 2020 Variation from Scientific

Advice

%

Eastern Baltic cod

 

0 2 000 (by-catch only) 2 000

(by-catch only)

infinity
Western Baltic cod

 

3065 3 065 3 806   23
North Sea cod (North Sea*, English channel, Skagerrat) 13,686 17,679 17,679 29
Common Sole 3a < 539 533 533 0
Common Sole 7a 561 457 457 18 (-)
Common Sole 7e < 1 488 1 478 1478 0
Common Sole 7f & g < 1488 1 652 1652 0
Haddock 7b etc <16 671 10 859 10 859 0
Herring 7a 0 8 064 8 064 infinity
Plaice 7fg <2 295 2003 2003 0
Great silver smelt Union <10 270 1 324 1234 0
Ling 3a etc <18 516 24 601 24803 33
Norway Lobster 8c 0 2.7 2.7 infinity
Norway Lobster 3a <19 904 13 733 13 733 0
Tusk 1,2 14 <11 077 21 21 0

 

The comparison is not straightforward. In a few cases, the area and stock for the advice do not correspond with the final TAC agreed to.

The record of following the advice is varied.  Often, the advice is followed, but in enough cases to be noticed, the gaps are several hundred to thousand % difference.  In those cases, I just put affinity.

What’s interesting is the ICES catch data.

Where stocks are in a good place, scientific advice is followed. When it is in a poor or dire place, – and the catches are depleted from recent historical level – officials and ministers are more comfortable to second guess nature.

 If healthy stocks are re-built following advice and poor stocks don’t, maybe someone will find out that there is obvious to re-build the stocks.