200 
Fishery Bulletin 106(2) 
Residence times at the FSA were sex-specific, thus 
creating a potential to promote sexual selection and 
negatively impact reproductive output when fishing was 
concentrated on the FSA. From acoustic results, we 
determined that males remained at the KMS over lon- 
ger periods both within the reproductive season (2.2± 
0.3 SE mo., males; 1.6 ±0.2 mo., females) and during 
the 17-mo. survey period (3.0 ±0.7 mo., males; 1.9 ±0.4 
mo., females) (Table 3). Conversely, according to the 
acoustic data, males and females appeared to over- 
lap entirely within a reproductive month. This overlap 
was not reflected in fisheries-dependent data because 
no females were captured >5 days before a full moon, 
whereas males were taken as early as 8-9 days before 
a full moon, indicating sex-specific variations in feeding 
or catchability. 
Virtually all males acoustically tagged in January 
remained at the aggregation site until spawning was 
concluded in February, further enhancing the potential 
for sexual selection before actual spawning within the 
aggregation period. Few females appeared to remain at 
the FSA site after spawning during subsequent months. 
Only three individuals were present in all possible 
spawning months, and two males and three females 
returned to or resided at the KMS during nonreproduc- 
tive periods (Table 3). 
Discussion 
The vulnerability of FSAs to overfishing is widely recog- 
nized (Sadovy and Domeier, 2005) and the present study 
shows the conditions under which FSA loss and pos- 
sible population decline can occur. For example, under 
a hyperstable fishing scenario (no CPUE reduction in 
relation to abundance) and using CPUE estimates from 
this study (3.83 fish per hr per fisherman) and peak 
FSA abundance estimates for squaretail coralgrouper 
at KMS, only 250 fishing days at 6 h of fishing per day 
would be required to deplete the entire FSA, equivalent 
to only 36 fishermen fishing over 7 days. Such a sce- 
nario was borne out in 1998 when an estimated 4000 
reproductively active fish were taken from the adjacent 
camouflage grouper FSA in just over 7 days (Rhodes and 
Sadovy, 2002). During that event, up to 20 boats with 
2-3 fishermen/boat fished the aggregation daily with 
no observed change in CPUE, before the aggregation 
was included into the KMS by executive order in 1999. 
Underwater visual census (UVC) counts in later years 
showed that only several hundred camouflage grouper 
remained at the camouflage grouper FSA site and that 
only minor increases in abundance were apparent after 7 
years of monitoring (2001-07). Similarly heavy aggrega- 
tion fishing has resulted in either loss, near-extirpation, 
or a substantial reduction in FSA abundance elsewhere 
(e.g., Craig, 1969; Sadovy and Eklund, 1999; Sala et 
al„ 2001). 
Migratory corridors used by reproductive fish increase 
the vulnerability of P. areolatus (and other fishes) to 
overfishing by concentrating reproductively active or 
resting fish within confined areas similar to the FSA 
in our study. Although it is not clear whether all P. 
areolatus (at KMS or elsewhere) use these migratory 
corridors, several small groups of squaretail coralgrou- 
per (5-10 individuals, presumably males) were observed 
moving southward along the reef between Peleng Chan- 
nel and KMS and groups of up to 100 females (identi- 
fication based on size and color) were observed moving 
in deeper water (20-30 m) toward the FSA during the 
spawning season. Groups of roving female squaretail 
coralgrouper have also been reported near FSA in the 
Solomon Islands during reproductive periods. Other ser- 
ranids reported or suggested to use migratory corridors 
to reach FSAs include Epinephelus guttatus (red hind) 
in the U.S. Virgin Islands (Nemeth, 2005), P. areolatus 
in Palau, and Epinephelus striatus (Nassau grouper) 
in Belize (Starr et al., 2007). In each case, fishermen 
apparently target the unprotected areas where fish con- 
gregate, thereby reducing the effectiveness of existing 
MPAs to protect reproductive individuals, as observed 
in Pohnpei. The apparent targeting of these migratory 
corridors and the ease by which fishermen can remove 
individuals from these and other areas where fish con- 
gregate highlights the need to identify and incorporate 
those areas into MPAs. 
The observed sex-specific differences in the behavior 
of P. areolatus within the reproductive season have the 
potential to promote selective fishing and to impact 
the spawning sex ratio and reproductive output (e.g., 
Beets and Friedlander, 1998). In Pohnpei, sex-specific 
behavioral differences were manifested as 1) a greater 
abundance of males in FSA-catch both seasonal and 
monthly, particularly in January when no females were 
captured; 2) the persistence of males at the FSA over 
several weeks during the initial aggregation period 
(January-February); and 3) longer male residency times 
at the FSA seasonally. Similar to the current study, 
previous studies of FSA-forming serranids have also 
shown that males often precede females to the FSA 
and, therefore, have longer residency times (Samoilys, 
1997; Rhodes and Sadovy, 2002; Starr et al., 2007). For 
squaretail coralgrouper, longer residency times appear 
to increase the catchability of males. Alternatively, 
males may outnumber females at the FSA site in all 
months and be reflected in the catch sex ratio. Further 
fisheries-independent assessments of aggregation (and 
population) sex ratio are needed to confirm whether 
males indeed outnumber females. Acoustic data also 
indicate that in addition to taking bait over more days 
than females, male squaretail coralgrouper frequent 
the FSA over more months — a finding that is similar to 
those from acoustic surveys of Nassau grouper in Belize 
(Starr et al., 2007) and UVC monitoring of freeze-brand 
tagged leopard coralgrouper ( P . leopardus) along the 
Great Barrier Reef (Zeller, 1998). Fisheries-dependent 
size data from FSA catch have also indicated protracted 
residency times for male red hind in the U.S. Virgin 
Islands (Nemeth, 2005). Protracted male residency time 
may increase the potential for sexual selection, leading 
to adverse affects on reproduction, such as sperm limi- 
