Shertzer et at.: Probability-based approach to setting annual catch levels 
229 
D 
2006 
2010 
2014 
B 
E 
Figure 4 
Example projection with managed risk of overfishing set at P* = 0.1 and 
uncertainty in management implementation described by its coefficient 
of variation CV=0.2. (A) Probability of overfishing, (B) annual fishing 
mortality rate, (C) spawning biomass (100 metric tons It]), (D) landings 
(100 t), (E) dead discards (100 t), < F ) annual catch level (100 t). In B-F, 
results are presented as medians (thick lines, circles), along with 10 th and 
90 th percentiles (thin lines), from ,/V=10,000 projection replicates. 
approaches zero quickly on the right because F MSY is 
connected tightly with steepness, a bounded parameter 
with positive probability at its upper bound. 
To allow uncertainty in management implementation, 
the annual catch level was assumed to follow a normal 
distribution with the mean equal to the target annual 
catch and coefficient of variation (CV) equal to a preset 
value. In some applications, the CV of implementa- 
tion might be estimated from data on performance of 
the fishery; in this application we considered values 
of 0.1, 0.2, and 0.3, with a focus on the assumption of 
CV=0.2. 
The final requirement of PASCL is to specify the al- 
lowable risk of overfishing. This analysis considered six 
different levels: P*e[0.05, 0.10, 0.15, 0.20, 0.25, 0.30], 
Results 
During the premanagement period (2005-07), overfish- 
ing was projected to occur in at least half of the projec- 
tions, and thereafter, at the acceptable P* (Fig. 4A). 
Because the probability of overfishing in the premanage- 
ment period was higher than P*, the fishing mortality 
rate was reduced when annual catch targets took effect 
(Fig. 4B). This allowed spawning biomass to increase 
(Fig. 4C). With nearly constant F, the increase in bio- 
mass provided for an increase in catch, composed mostly 
of live landings but also dead discards (Fig. 4, D-F). 
Catch within a year varied by replicate, reflecting uncer- 
tainty in management implementation, but by design 
was centered on the annual catch level (Fig. 4F). 
