238 
Fishery Bulletin 113(3) 
Discussion 
Time in the nursery 
It is unknown why the signals of cultured fish released 
in the first year of this study disappeared faster than 
wild fish in the second year. This release is the first one 
of a deepwater fish species raised in captivity. There 
are many potential factors, such as depth, light levels, 
and temperature, that likely influenced the behavior 
of the released cultured fish. Cultured fish reared in 
the warm surface water of a bay cage had none of the 
familiarity with the nursery habitat that the wild fish 
had. The released wild fish lingered because they knew 
the site, they had not yet matured enough to emigrate, 
or there were significant environmental differences be- 
tween years that influenced fish behavior. 
Environmental effects, including freshwater dis- 
charge, have been reported to influence the behavior 
of tagged marine species in coastal estuaries during 
their early life history (Manderson et al., 2014). The 
Kaneohe nursery differs from nurseries in estuarine 
habitats, because it is located deep, at the edge of an 
oceanic slope, where the substrate, temperature (Mof- 
fitt and Parrish, 1996), and nutrients (Parrish et ah, 
1997) have been found to be consistent between years 
in prior studies. The temperature records from both 
tagging years in this study were similar, and there 
is no clear association between temperature and be- 
havior of the crimson jobfish. The original plan was to 
conduct a simultaneous release of cultured and wild 
fish in 2007 to compare their behavior, but the funding 
for the culture program was unexpectedly cut. Future 
studies are needed to distinguish whether interannual 
environmental effects are indeed a concern or whether 
the observed fish behavior is due to the inability (or a 
lack of need) for cultured juvenile fish to integrate into 
an assemblage of wild juveniles. 
Potential tagging effects 
Surgical implantation of acoustic tags in small fish 
has the potential to add significant weight and affect 
the fish’s buoyancy and behavior. Prior studies (Mc- 
Cleave and Stred, 1975; Adams et al., 1998) recom- 
mended that tags should be no more than 2% of fish 
body weight. The V9 tags exceeded this percentage for 
8 of our fish; 1 tag in 1 fish accounted for as high as 
3.3% of estimated body weight. However, all the fish 
fell well within the percentage of tag weight identified 
by Brown et al. (1999), who showed success with the 
use of tags that were 6-12% of body mass of juvenile 
salmon (smolt). Our observations of fish actively swim- 
ming in the holding pen for days before release and 
documented movement between receivers after release 
indicate that there was little behavioral effect from the 
size of the tags used. 
Having endured the stress of tagging, release, and 
descent through the water column, most of the released 
fish successfully reached the bottom where they were 
repeatedly detected for a day or more by the nursery 
receivers. A single 20-cm fish (tag 1867) was not de- 
tected by either of the nursery receivers after it was 
released, but it appeared 9 days later at site 3 on the 
adjacent slope (Fig 1). It stayed there for 2 days be- 
fore disappearing from the study. The disappearance 
of tag signals is either the result of the fish emigrat- 
ing outside the receiver surveillance area or of it being 
consumed by a predator that leaves the area. The fea- 
tureless mud bottom of the nursery supports few other 
fish (DeMartini et al., 1996) that would be resident 
predators of the crimson jobfish; therefore, predation 
loss was likely from larger, transient fishes that passed 
through the nursery. Loss of tagged fish to predation 
is expected, although we have no way to evaluate the 
degree of this impact. 
Movement of wild fish 
There have been few studies of tagged deepwater snap- 
pers and most of them have been of adult fish. The 
most recent work (Weng, 2013) looked at movements of 
tagged adult eteline red snappers and found that ruby 
snapper ( Etelis carbunculus) showed more fidelity to 
the area of their release than did tagged flame snapper 
(Etelis coruscans ). A separate study of 18 tagged adult 
crimson jobfish (40-60 cm FL) showed that 75% of fish 
remained in the monitored release area (Ziemann and 
Kelley 5 ). Because those mature fish had already immi- 
grated to their adult habitat, we might expect greater 
fidelity for them than for the juvenile fish in our study. 
In contrast, we examined the temporary use of nurs- 
ery habitat by juveniles before they moved to the next 
stage in their life history. 
The time spent in the nursery varied from a few 
days to multiple weeks. For many of the fish tagged 
in our study, diurnal movements were detected in and 
out of the range of receivers in the nursery. The study 
on tagged adult crimson jobfish found that they aggre- 
gated during the day and ranged over a wider area 
at night (Ziemann and Kelley 5 ). The movements of the 
juveniles in our study were much more limited. The 
receivers deployed on the slope adjacent to the nurs- 
ery did not detect a greater number of juvenile crim- 
son jobfish during evening hours than during daytime 
hours; an increase would be expected if the fish were 
engaged in wide-ranging movements. It is possible 
that the juveniles could have moved inshore to shal- 
lower depths; however, data from a previous study that 
conducted boat-based tracking of 2 juveniles indicated 
that the fish would move deeper (Moffitt and Parrish, 
1996). The individual tracks of both fish showed cre- 
puscular movements, the fish stayed in shallower wa- 
5 Ziemann, D., and C. Kelley. 2004. Detection and documen- 
tation of bottomfish spillover from the Kaho'olawe Island Re- 
serve. Final report submitted to the Kaho'olawe Island Re- 
serve Commission for Study I. [Available from Kaho'olawe 
Island Reserve Commission, 811 Kolo St., Ste. 201, Wailuku, 
HI 96793.] 
