Parrish et al.: Nursery habitat in relation to production of Pristipomoides filamentosus 
143 
Table 3 
Statistical specifics associated with the regression for presence (>5) or absence (<5), of snapper aggregations at east Oahu. Model 
chi-square (x 2 )=54.11, P<0.0001, df=3. 
Estimated 
Name of variable 
coefficient 
Standard error 
Probability value P 
Cross product of clay-silt with proximity of drainage 
source 
1.45 x 10- 6 
3.47 x 10- 7 
<0.0001 
Distance to drainage 
-7.5 x 1(E 7 
1.58 x 10“ 7 
<0.0001 
Escarpment relief 
-1.586 
0.435 
0.0003 
Table 4 
Two by two table of presence (>5) or absence (<5) of snapper aggrega- 
tions predicted by the model versus presence or absence observed from 
baited video drops. Includes all 211 drops at east Oahu. 
Predicted by model 
Aggregations 
Aggregations 
Percent 
Observed 
absent (<5) 
present (>5) 
correct 
Aggregation absent 
73 
48 
60 
Aggregation present 
19 
71 
79 
68 overall 
provide an image of snapper distribution 
at east Oahu (Fig. 5). 
Abundance of juveniles in the 
archipelago 
Fishing surveys at insular slopes other 
than east Oahu (total 332 km) detected 
few juveniles (Table 5). Five of these sites 
(with snappers) were surveyed with 
video camera to compare with the east 
Oahu aggregations. Significant numbers 
of snappers were found only at a site off 
the southwest end of the island of 
Molokai (Sept 1993). A repeat video sur- 
vey indicated that the significant between-station 
differences in snapper abundance initially reported 
at South Molokai, persisted 7 months later (K-W, 
X 2 =50.8, P<0.05) (Fig. 6). 
The video index and CPUE of the conventional fish- 
ing gear were roughly consistent for all sites (Table 
6). Snapper abundance was found unrelated to sub- 
strate type (r s =0.59, P=0.40, n-1) or distance from 
the 15-m isobath (r s =-0.84, P=0.15, n-1). However, 
distance/depth of discharge at the four sites with 
known sources of coastal drainage (Kaneohe, Kailua, 
S. Molokai, and Hanalei) were associated with snap- 
per abundance (r s =-1.0, P<0.001, n- 4). 
Video- and catch-based production estimates 
Video abundance data from MHI sites at N. Molokai, 
Hanalei, and Kahului yielded a mean estimated den- 
iable 5 
Fishing effort, length of slope fished, and total snappers caught on surveys of Hawaiian insular slopes for juvenile snappers. 
Island 
Bottom trawls 
1990 
(no.) 
Bottom longlines 
1992 
(no. hooks) 
Fish traps 
1989-94 
(no.) 
Handlining 
1993-94 
(line-hr) 
Length of 
slope fished 
(km) 
Total snapper 
caught 
(no.) 
Oahu 7 
16 (51 2 
— (150) 
27 (53) 
60 (50) 
164 (14) 
16(828) 
Molokai 
26 
— 
101 
87 
60 
256 
Maui 
6 
150 
— 
— 
16 
4 
Lanai 
6 
— 
— 
— 
16 
0 
Kauai 
— 
150 
— 
— 
18 
3 
Necker 
— 
— 
25 
— 
16 
0 
FFS 3 
— 
— 
63 
— 
42 
5 
Total 
54 
300 
216 
147 
332 
284 
1 The east Oahu study site values are not included in any of the figured totals. 
2 Numbers in parentheses represent additional values from gear validation test done at the east Oahu site. 
3 FFS = French Frigate Shoals. 
