Soh et a!.: Role of marine reserves in the management of Sebastes borealis and S. aleutianus 
169 
Table 1 
Gulf-wide discard rates (%) for the four slope rockfish management subgroups in the commercial fishery during 1991-97. Data are 
from weekly production and observer reports. 
Management subgroup 
1991 
1992 
1993 
1994 
1995 
1996 
1997 
Pacific ocean perch 
15.7 
21.5 
79.2 
60.3 
19.8 
17.2 
14.3 
Shortraker and rougheye rockfiish 
42.0 
10.4 
26.8 
44.8 
30.7 
22.2 
22.0 
Northern rockfish 
— 
— 
26.5 
17.7 
12.7 
16.5 
27.8 
Other slope rockfish 
20.0 
29.7 
48.9 
65.6 
72.5 
75.6 
52.1 
considers ABC recommendations, social and economic 
factors, and determines total allowable catches (TACs). 
Shortraker and rougheye rockfish are considered as a 
management unit in the setting of TACs. 
In fishery management, fishing seasons are imposed to 
regulate fishing effort based on three catch levels: total al- 
lowable catch (TAC), acceptable biological catch (ABC), and 
overfishing level (OFL). There is an open, directed fishery 
(with allowances for bycatch) until the catches reach the 
TAC level. The TAC level is often lower than the ABC but 
frequently they have been the same for the shortraker- 
rougheye rockfish subgroup. Once the ABC is reached, these 
species are prohibited to be caught and must be returned 
to the sea. Finally, the fishery is closed when it reaches the 
overfishing level. A directed fishery for shortraker- rougheye 
rockfish occurred only in 1991 and 1992; the fishery has 
since been designated as a bycatch-only fishery. 
A constant exploitation-rate strategy has been used for 
the shortraker and rougheye rockfish subgroup. Fishing 
mortality for determining ABC was set equal to natural 
mortality ( F=M rule), which, like F 0 p was generally con- 
sidered to be conservative (Deriso, 1987). This has been 
used to determine ABC since 1991, when these stocks were 
established as an independent management subgroup of 
the slope rockfish assemblage. The main reason for using 
this simple management procedure compared with the 
more sophisticated one used for Pacific ocean perch and 
other species is a lack of historical catch data and bio- 
logical information. In 1997, revised definitions of ABC 
and OFL replaced the Plan Team’s previous procedures 
(NMFS, 1996): ABC is a preliminary description of the ac- 
ceptable harvest (or range of harvests) for a given stock or 
stock complex and overfishing is defined as any amount of 
fishing in excess of a prescribed maximum allowable rate. 
For rougheye rockfish, fishing mortality rates for OFL and 
ABC are determined by 
^OFL - ^30% ar *d F.ABC - ^40% ’ 
where ^X7< refers to the fishing mortality rate associated 
with an equilibrium level of spawning per recruit equal to 
X% of the equilibrium level of spawning per recruit in the 
absence of any fishing (Clark, 1991; 1993). 
For shortraker rockfish, the equations are Fofl - M and 
^abc - 0-75 x M because there is less information for this 
species. 
Fish discards are a serious problem under the current 
management system because fish are discarded when the 
catch of a management category attains ABC. Discard 
rates for the four slope rockfish subgroups in the Gulf of 
Alaska were estimated as shown in Table 1 (Heifetz et al., 
1997). There are two situations which result in discards at 
sea. One case happens under "bycatch only" status, where 
theoretically, only a fixed percentage of the haul-by-haul 
catch can be retained and the rest must be discarded to 
protect the stocks from a directed fishery. This “retained” 
catch should be available only when the bycatch status is 
still open. The other case occurs under prohibited status. 
If the total harvest exceeds ABC levels, all catch must be 
discarded. 
Shortraker and rougheye rockfish are a slow-growing, 
long-lived species. Maximum age for rougheye rockfish, 
for example, is about 140 years (Chilton and Beamish, 
1982) and the von Bertalanffy growth parameter, K, for 
this species is about 0.05. Both shortraker and rougheye 
rockfish are very large fish; L x s 55 cm for rougheye rock- 
fish and L^= 72 cm for shortraker rockfish. McDermott 
(1994) estimated natural mortality of rougheye and short- 
raker as 0.030-0.039 and 0.027-0.042, respectively, using 
the gonadosomatic index method (Gunderson, 1997a). Ma- 
turity analysis showed that about 50 percent are mature 
at 43.87 cm FL (about 20 years) for rougheye rockfish, 
whereas about 50 percent are mature at 44.90 cm FL for 
shortraker rockfish (McDermott, 1994). 
Rougheye rockfish undertake only limited migrations 
once they have recruited to the fishing ground (Gunder- 
son, 1997b) and this is generally considered to be true 
for shortraker rockfish as well. Historical catch locations 
show specific habitat preferences and a patchy distribu- 
tion for shortraker and rougheye rockfish in the Gulf of 
Alaska. Shortraker and rougheye rockfish present in “hot 
spots” can be readily overexploited by the fishing industry 
by targeting these spots and harvesting the fish through 
specialized harvesting skills. 
Refuge management 
Harvest refugia are areas protected from some or all fish- 
ing activities. The general objectives of creating harvest 
refugia can be discussed with respect to the fish commu- 
nity, fishery practices, and ecological aspects of fish habitat 
