/Vi j 3 3 et a!.: Establishing species-habitat associations for 4 eteline snappers 
303 
Opakapaka (90-210 m) 
o. 
CD 
Q 
300 
250 
200 
150 
100 -j 
50 
Kalekaie (180-270 m) 
VL. . ^=0.011 
0 20 40 60 80 100 
Fork length (cm) 
Q. 
<D 
a 
300 i 
250 
200 
150 - 
100 - 
50 
0 
Onaga (210-300 m) 
^= 0.002 
20 
40 60 
Fork length (cm) 
80 
100 
Q. 
<D 
Q 
Ehu (210-300 m) 
300 - 
250 - 
200 - 
150 - 
100 - 
50 - 
j j 
r^O.050 
20 
40 60 
Fork length (cm) 
80 
100 
Figure 5 
Comparison of fork length with depth for Opakapaka (. Pristipomoides filamentosus ) , Kalekaie (P. sieholdii), 
Onaga ( Etelis coruscans ), and Ehu (E. carbunculus) within the preferred depths of each species for our study 
of habitat associations of these species in the main Hawaiian Islands from May 2007 to June 2009. The coef- 
ficient of determination (r 2 ) is given for each species. 
(200x200 m) regardless of the visual area observed in 
the video. This assumption was made on the basis of 
the limited information available on the distance of 
bottomfish attraction to bait stations. Ellis and De Mar- 
tini (1995) estimated that the greatest distance of at- 
traction for juvenile Opakapaka to their baited cam- 
eras was between 48 and 90 m. Merritt et al. (2011), 
in their baited camera survey of Penguin Bank, used a 
200-m distance between deployment locations to avoid 
a cross influence of bait. 
The area of fish attraction (sampling area) has been 
quantified at abj^ssal depths by Priede and Merrett 
(1996) through the use of current velocity, fish swim- 
ming speed, and a bait dispersal model. Their deter- 
mination of the area of attraction, however, relied on 
assumptions (i.e., fish are evenly dispersed) that do not 
apply to the fish species and shallower depth ranges 
in this study. Furthermore, bottom current variabil- 
ity, habitat variability, and small-scale bathymetric 
features at mesophotic depths around Hawaii make 
the quantification of the area of attraction to bait ex- 
tremely challenging. In a comparison of baited and un- 
baited underwater video stations, Harvey et al. (2007) 
acknowledged that fish behavior and life history also 
may affect attraction to bait. All the species in this 
study are regularly attracted to bait and are taken on 
baited hooks, but other behavioral traits (e.g., mobility, 
schooling, and reproductive cycles) could affect species- 
specific responses to a bait-odor plume. Given the dif- 
ficulty involved in the determination of the actual area 
of bait influence, the appropriateness of the habitat- 
classification scale chosen for use in this study cannot 
be evaluated. Until an effective scale of attraction can 
be verified for deepwater snappers and other bottom- 
fishes, a fully quantitative assessment of species-habi- 
tat associations is not yet possible. 
Although previous studies have indicated that habi- 
tats with hard substrates and high slopes, such as 
headlands and promontories, are preferred by many 
bottomfish species (Ralston and Polovina, 1982; Ralston 
et al., 1986; Parrish, 1987; Kelley et al., 2006; Parke, 
2007), we determined that other habitat types, such as 
hard-low habitats, are important to eteline snappers 
and that species-specific differences in habitat pref- 
erence exist. On the basis of relative abundance, we 
found that the overall habitat preference of Opakapaka 
was for low-sloping hard substrates. Onaga was associ- 
ated with hard-high and hard-low habitats, and Ehu 
was seen mostly on hard-low habitats. The observed 
association of juvenile Opakapaka and Onaga with 
hard-low habitats may be driving their preference for 
this habitat type. In contrast, the finding for Ehu could 
