In what direction should the fishing mortality target change when natural mortality increases within an assessment?

Traditionally, the natural mortality rate (M) in a stock assessment is assumed to be constant. When M increases within an assessment, the question arises how to change the fishing mortality rate target (F-Target). Per recruit considerations lead to an increase in F-Target, while limiting total mortality leads to a decrease in F-Target. Application of either approach can result in nonsensical results. Short-term gains in yield associated with high F-Target values should be considered in light of potential losses in future yield if the high total mortality rate leads to a decrease in recruitment. Examples using yellowtail flounder (Limanda ferruginea) and Atlantic cod (Gadus morhua) are used to demonstrate that F-Target can change when M increases within an assessment and to illustrate the consequences of different F-Target values. When a change in M within an assessment is contemplated, first consider the amount and strength of empirical evidence to support the change. When the empirical evidence is not strong, we recommend using a constant M. If strong empirical evidence exists, we recommend estimating F-Target for a range of stock-recruitment relationships and evaluating the trade-offs between risk of overfishing and forgone yield.