88 
Fishery Bulletin 1 12(1) 
Male 
Unknown 
0 200 400 600 
Home range (ha) 
Figure 2 
Distribution of home-range size (ha) of 109 Black Sea Bass (Centropris- 
tis striata ) on the basis of 85% kernel density estimates for 31 male 
and 78 fish of unknown sex, a group that included females, imma- 
ture males, and transitional males. Black Sea Bass were acoustically 
tagged and released between May and July 2003 at the Historic Area 
Remediation Site (Fig. 1) in the mid-Atlantic Bight. Mean home-range 
size is indicated by a filled circle; the vertical edges of the rectangle 
are the 25 th and 75 th percentiles; the vertical line inside the rectangle 
indicates the median; the whiskers extend to 1.5 times the difference 
between the 25 th and 75 th percentiles (the interquartile range); and 
the open circles are observations more extreme than those of the in- 
terquartile range. 
b 
• - 
I ooo o o 
b 
• 
t 1 1 r 
and fish of unknown sex. Duration of occupancy signifi- 
cantly affected the home-range size of fish of unknown 
sex (7^= 14. 15, P<0.01) but not the home-range size of 
males (F=0.01, P=0.91). Fish of unknown sex that occu- 
pied the site for short periods had significantly larger 
home ranges than those fish of unknown sex that oc- 
cupied the site for longer periods (J=-0.010, t=- 3.78, 
P<0.01; Fig. 3). Duration of occupancy was the only fac- 
tor that significantly affected home-range size of fish 
of unknown sex; neither fish size (F=0.81, P=0.37) nor 
release date (F=0.04, P=0.83) was significant in predic- 
tion of mean home-range size of these fish. As noted 
previously, estimates of the duration of occupancy may 
have been biased low, particularly for fish released lat- 
er in the tagging period. However, release date did not 
affect duration of occupancy (F=0.17, P=0.69), indicat- 
ing the lack of systematic bias in the estimated dura- 
tion of occupancy for these fish. 
In contrast to the results observed for fish of un- 
known sex, we found a significant effect of release date 
on home-range area of male fish (F= 5.81, P=0.02) such 
that, on average, males tagged and released in early 
June established home ranges that were at least twice 
the size of home ranges established by males tagged 
and released in July (predicted mean ear } y j un = 143.2 ha, 
n= 7; predicted meanj u i=63.0 ha, n=ll; 
Fig. 4). Furthermore, size of male fish 
did not explain the observed variation in 
mean home range (F=0.70, P=0.41). Re- 
gardless of when male fish were tagged, 
the average size of these fish was simi- 
lar (F= 0.55, P=0.46), implying that male 
body size did not account for the smaller 
home ranges observed among fish tagged 
in the latter part of the tagging period. 
Movement of Black Sea Bass 
Black Sea Bass were more likely to move 
in summer than in fall; fish were also 
more likely to move with increasing mean 
daily temperature differences at stations 
B1 and H7. Time period did not affect the 
probability of movement between adja- 
cent receivers, nor did differences across 
the study site in mean salinity at the bot- 
tom of the seafloor. The best model from 
the set of models considered was that 
with an m-dependent correlation struc- 
ture with m = 30 to model the correlations 
between observations; 30 time intervals 
corresponds to 7.5 days indicating that 
movement probabilities within a given 
week were significantly correlated, but 
not thereafter. The interactions retained 
in the model included cubic time interac- 
tions with release group and sex (and all 
the lower order 2-way interactions con- 
tained therein), as well as the interaction 
of time with mean temperature at station B1 (Table 1); 
these complex movement probabilities are therefore de- 
scribed separately for males and fish of unknown sex. 
Between June and October, the probability of mov- 
ing decreased for male fish and was lowest among 
males released in late June compared with males re- 
leased at other times (Fig. 5A); between October and 
mid-November, the probability of moving increased for 
males released in July but remained low for the males 
released in late June (Fig. 5A). Males released in early 
June were no longer present at the site by mid-Septem- 
ber; therefore, movement likelihoods could not be esti- 
mated for this group. In contrast, fish of unknown sex 
from all release groups exhibited similar probabilities 
of movement that decreased between June and Novem- 
ber (Fig. 5B); however, fish of unknown sex released in 
late June maintained higher probabilities of movement 
between October and mid-November than did fish of 
unknown sex from other release groups (Fig. 5B). The 
most pronounced difference among the sexes occurred 
for the July release group between October and mid- 
November: among males, those fish released in July 
exhibited the highest probability of moving during this 
time period, whereas fish of unknown sex released in 
July exhibited the lowest probability of movement. 
