Misa et al. : Establishing species-habitat associations for 4 eteline snappers 
305 
have been the result of a sampling artifact caused by 
the lack of habitat types other than the hard-low envi- 
ronments in Pailolo Channel, where many observations 
of this species were made. Regardless, the results of 
this study clearly show the importance of this habitat 
type for Ehu. Kaiekale were observed often in large 
schools in our video footage. For defense against preda- 
tors, this species may rely on its schooling behavior in- 
stead of associating with the bottom habitat. The lack 
of a significant habitat preference for Kaiekale could, 
consequently, be driven by this defense mechanism. As- 
sessment of species— habitat associations, therefore, re- 
quires an understanding of species behaviors and the 
changes in habitat use by life stage. 
Clear ontogenetic shifts in habitat associations were 
evident for 3 of the 4 species studied. For Opakapaka, 
there was a distinct ontogenetic progression in habitat 
association that expands what is known for this species. 
The known habitat for juveniles of this species at 7-25 
cm FL is shallow, low-sloping, soft substrates (Moffitt 
and Parrish, 1996). Juvenile Opakapaka have been ob- 
served at depths of 65-100 m offshore of Kaneohe Bay 
(Parrish, 1989; Moffitt and Parrish, 1996) and more re- 
cently off Waikiki, Oahu, at depths of 37-42 m (J. Bra- 
zen, unpubl. data). These juveniles move out of their 
nursery grounds and presumably merge with the adult 
schools in deeper waters after about 1 year (Parrish et 
al., 1997). Within the preferred depth range identified 
in our study for Opakapaka (90—210 m), the smallest 
mean lengths were found over hard-low habitats at 4 
of the 6 sampled locations. We recorded Opakapaka as 
small as 16 cm FL within our sampling depths over 
hard-low habitats. On the basis of growth curves from 
Be Mart i ni et al. (1994), the juvenile Opakapaka in our 
study were just under 1 year old and could be recent 
migrants from a surrounding nursery area. The results 
of this study show that these fish continue to stay in 
hard-low habitats until they reach 45 cm FL or about 5 
years of age and, thereafter, increasingly use hard-high 
habitats. It is possible that this species uses hard sub- 
strates with low slopes as a transitional habitat before 
a move into hard-high habitats. Opakapaka reaches 
sexual maturity at ~43 cm FL (Kikkawa, 1984). The 
shift in habitat from hard-low to hard-high could be a 
response to reproductive maturity, which is discussed 
later. 
Size-related habitat shifts also were evident for 
Kaiekale and Onaga but were observed without a 
change in their depth of occurrence. Previous stud- 
ies also showed a lack of depth change with size for 
these species (Kelley et al 4 ; Ikehara, 2006). The move 
into hard-high habitats with increasing size coincided 
roughly with the onset of sexual maturity in both spe- 
cies. The size (25-35 cm FL) at which Kaiekale shifted 
4 Kelley, C. D., B. C. Mundy, and E. G. Grau. 1997. The use 
of the Pisces V submersible to locate nursery grounds of com- 
mercially important deepwater snappers, family Lutjanidae, 
in Hawaii, 62 p. Paper presented at the 5 th Indo-Pacific Fish 
Conference; Noumea, New Caledonia, 10-16 November. 
to hard-high habitats from other types includes the size 
(29 cm FL) at which this species reaches sexual ma- 
turity (DeMartini and Lau, 1999). The onset of sexual 
maturity for Onaga occurs at 61 cm FL (Everson et al., 
1989) — a size larger than the size (55 cm FL) at which 
a shift in habitat use was observed in our study. On the 
basis of size-at-age curves, the onset of sexual maturity 
occurs between the ages of 3 and 6 years for Kaiekale 
(Williams and Lowe, 1997) and 5 to 6 years for Onaga 
(Everson et ah, 1989). 
In contrast to the other 3 species, no size-related hab- 
itat shifts were observed for Ehu, but very few juveniles 
of this species were measured (Fig. 6; n= 37). Juvenile 
Ehu, along with other smaller bottomfishes, are highly 
vulnerable to predation by demersal carnivores, such as 
the Greater Amberjack ( Seriola dumerili) (Humphreys 
and Kramer, 1984). A few instances where Greater 
Amberjack seemed to scare away Kaiekale and Ehu 
were observed in the BotCam video collected during 
our study. No aggressive behavior toward the target 
species by other predators was seen, but it is possible 
that carnivorous species could have affected our abil- 
ity to sample certain size ranges of bottomfishes, par- 
ticularly Ehu. Smaller snappers may have moved out 
of the BotCam’s field of view before predators entered. 
Even if they were possibly in the vicinity of the Bot- 
Cam, juveniles may have remained close to the bottom 
of the seafloor for protection and out of the unit’s field 
of view. Until very small Ehu (i.e., 5-15 cm FL) can be 
observed regularly, a complete ontogenetic assessment 
of habitat for this species will not be possible. However, 
it is important to note that the size range of Ehu har- 
vested by the fishery is represented in this study. 
The ontogenetic habitat shifts observed for Opak- 
apaka, Kaiekale, and Onaga could be related to shifts 
in diet, increases in reproductive output, and preda- 
tor avoidance at smaller sizes. Szedlmayer and Lee 
(2004) reported a shift in the diet of the shallow-water 
juvenile Red Snapper ( Lutjanus campechanus) from 
crustaceans to fishes and cephalopods with increasing 
size. This change in diet was associated with the mi- 
gration from nursery habitats to coral reefs. For deep- 
water snappers, diet shifts have yet to be documented. 
DeMartini et al. (1996) examined the diet of juvenile 
Opakapaka from the nursery in Kaneohe Bay and dis- 
covered that it was composed of crustaceans (shrimps 
and stomatopods), gelatinous organisms (salps and 
heteropods), nekton (fishes and squids), and benthic 
organisms (demersal octopods, echinoids, and micro- 
gastropods). With the exception of benthic prey, a simi- 
lar diet was found for Opakapaka caught at depths of 
100-300 m in Penguin Bank by Haight et al. (1993b). It 
is possible that smaller individuals (<43 cm FL) of this 
species associate with low-sloping, hardbottom habitats 
to feed on the benthos and then shift to a pelagic diet 
when they move into hard-high habitats where the pos- 
tulated increase in water flow increases prey availabil- 
ity (Ralston et ah, 1986; Haight et al., 1993a; Kelley et 
al., 2006). 
