The European Union, Norway, and the United Kingdom jointly requested ICES to identify appropriate combinations of Ftarget and Btrigger in a harvest control rule that, together with possible TAC constraints, could form part of a long-term management strategy for North Sea autumn spawning herring (Clupea harengus) in Subarea 4 and divisions 3.a and 7.d, (North Sea, Skagerrak and Kattegat, eastern English Channel). A Management Strategy Evaluation (MSE) process was set up to evaluate such combinations, check that the management strategy was robust to a 10% banking and borrowing mechanism, and to investigate sensitivity to a number of exploitation pattern scenarios. Key sources of uncertainty (related to recruitment and natural mortality) were captured by a reference set of operating models, and results were integrated over these.
Key findings are that several management strategies are possible for a similar level of precaution (less than 5% risk), with similar levels of catch (less than 5% difference in average yield between the possible rules). These different rules involve either a high Ftarget and high Btrigger, or a lower Ftarget combined with a lower Btrigger, and provide a trade-off between maximising catch and minimising interannual variability in catch. However, a high Ftarget-high Btrigger combination (e.g. Ftarget=0.34, Btrigger=1.7 million tonnes) results in a marginally higher yield (average annual catch of 0.37 million tonnes) but with a realised fishing mortality well below Ftarget (0.23 against Ftarget=0.34) because such high theoretical Ftarget rules result in the stock being frequently below the Btrigger value. This combination of control points are also associated with more unstable catches (IAV=18.5%), lower SSB (1.3 million tonnes), and the more frequent suspension of TAC constraints (as a result of SSB being below Btrigger). In contrast, a low Ftarget-low Btrigger combination (e.g. Ftarget=0.21, Btrigger=0.8 million tonnes) results in marginally lower yield (average annual catch of 0.36 million tonnes) but with realised fishing mortality close to Ftarget (0.2 against Ftarget=0.21), substantially more stable catches (IAV=9.9%), higher SSB (1.5 million tonnes), and less frequent suspension of any stability mechanisms in place.
Assuming current fishing conditions (2022-3), as opposed to longer-term recent (2013-2021) or historical (1998-2003) conditions, leads to higher risk and more variable fishing mortality on ages 0-1, providing assurance that assuming current conditions (as was adopted for the reference set of operating models) is both reasonable and errs on the side of precaution with respect to risk. Increasing fishing mortality on ages 0-1 has a clear negative impact on risk, SSB and catch in the long-term. Furthermore, shifting the selection pattern towards younger ages is generally negative across all performance metrics. Management strategies appear to be robust to implementing a 10% banking and borrowing scheme, even under an extreme version that deliberately forces unrealistic annual fluctuations in catch (the opposite of its intended purpose). It should be stressed that Ftarget and Btrigger selected from the MSE should not be confused with reference points set by ICES and should not be used to indicate stock status.