176 
Fishery Bulletin 99(1) 
Table 4 
Summary of simulations showing ending biomasses (t) and average fishing mortalities from 20-year future projections. Three 
refuge sizes and two fishing intensity schedules are compared under current and refuge management systems. F values under 
refuge management pertain only to areas outside of refugia. (SSR=small-size refugia; MSR=middle-size refugia; LSR=large- 
size refugia) 
Ending biomass (year=2016) Average F (during 1997-2016) 
F for ABC Actual F F for ABC Actual F 
(F= 0.023) (F=0.063) (F=0.023) (F=0.063) 
Species 
Current 
Refuge 
Current 
Refuge 
Current 
Refuge 
Current 
Refuge 
Shortraker rockfish 
SSR 
23,442 
18,270 
0.018 
0.043 
MSR 
20,847 
23,437 
12,623 
18,236 
0.023 
0.022 
0.063 
0.055 
LSR 
23,425 
18,153 
0.030 
0.090 
Rougheye rockfish 
SSR 
50,187 
54,764 
0.019 
0.012 
MSR 
43,351 
50,176 
50,444 
54,760 
0.025 
0.023 
0.015 
0.014 
LSR 
50,160 
54,753 
0.030 
0.018 
Table 5 
Percentage of Gulf-wide catches in middle-size refugia ( MSR), by species and depth, from domestic observer data ( 1987-1996 ). POP = 
Pacific Ocean perch. 
Depth (m) 
Shortraker 
rockfish 
Rougheye 
rockfish 
POP 
Thornyhead 
Northern 
rockfish 
Dusky 
rockfish 
Sablefish 
Rex sole 
100-713 
28 
29 
15 
12 
12 
17 
6 
12 
200-713 
27 
28 
12 
12 
1 
1 
5 
6 
300-713 
23 
25 
2 
9 
0 
0 
3 
1 
Incorporation of harvest refugia into current manage- 
ment for shortraker and rougheye rockfish resulted in in- 
creased ending biomass for both levels of F (Table 4). Un- 
der the “F for ABC” scenario, ending biomass increased 
about 12% for shortraker rockfish and about 16% for 
rougheye rockfish with refuge management. Under the 
“actual F” scenario, ending biomass increased about 44% 
for shortraker and about 8% for rougheye rockfish when 
refugia were employed. Figure 5 shows the future biomass 
trajectories for the two management systems. For both 
fishing intensities, higher Gulf-wide biomass levels were 
projected in refuge management for both species. 
Size differences for harvest refugia had relatively little 
impact on the ending biomass estimates. Table 4 shows 
similar ending biomass in refuge management irrespec- 
tive of refuge sizes for both “F for ABC” and “actual F” 
scenarios. This is because the same catch removal was ap- 
plied, irrespective of refuge size. 
Impacts of harvest refugia on other target fisheries 
Table 5 summarizes the proportion of catches within refu- 
gia compared with those from the whole Gulf for major 
species in the slope region. About 30% of the Gulf-wide 
catches of shortraker and rougheye rockfish occurred at 
depths greater than 100 m in MSR. Out of this, over 80% 
still remained in MSR in depths greater than 300 m. 
Dusky rockfish had the next highest proportion in MSR but 
their distribution was primarily confined to the 100-200 m 
depth interval. Shortspine thornyhead remained at 9% of 
the Gulf-wide catch in refugia deeper than 300 m. The 
catch proportions of other major fishes in refugia deeper 
than 300 m declined to 3% or less. Clearly, if harvest refugia 
can be established with depth limits greater than 300 m, 
then the impacts of no-take zones for shortraker and rough- 
eye rockfish on other fisheries seem minor. 
Discussion 
Understanding the changes in fishing effort under refuge 
management is important for its practical application. In 
refuge management, fishing is allowed only in harvestable 
areas where fish density is assumed to be less than in 
refuge areas. However, high levels of fishing intensity out- 
side of refugia are to be avoided because high fishing effort 
