94 
Fishery Bulletin 11 7(1-2) 
to control for species-specific differences in attraction 
to or avoidance of divers (Kulbicki, 1998; Watson and 
Harvey, 2007), 2) superior detection capacity (McCau¬ 
ley et al., 2012), 3) easy deployment at much greater 
depths and in environments where diver safety may 
be a concern (De Vos et al., 2015), and 4) longer soak 
times and sampling intervals because BRUVS are not 
subject to decompression limits (Watson et al., 2010). 
Additionally, the lower relative costs and fewer person¬ 
nel required for deployment of BRUVS, compared with 
those of large-scale UVC or longline sampling, allow 
larger sample sizes (Willis et al., 2000; Brooks et al., 
2011) and facilitate the deployment of BRUVS at large 
spatial scales (White et al., 2013; Espinoza et al., 2014; 
Santana-Garcon et al., 2014). Studies of temperate, 
subtropical, and coral reefs in Western Australia deter¬ 
mined that analysis of videos from BRUVS results in 
greater species richness and a greater number of tar¬ 
geted, carnivorous large-bodied species, such as those 
in the families Lethrinidae and Serranidae, whereas 
UVC conducted by divers results in observations of 
more site-attached, cryptic small-bodied species, such 
as some members of the families Pomacentridae, Labri- 
dae, and Scaridae (Langlois et al., 2010; Watson et al., 
2010). On the other hand, observations in videos from 
BRUVS may be affected by currents and fish traits, 
such as appetite and search behavior, as well as by 
inter- or intraspecific interactions (Stoner et al., 2008; 
Barley et al., 2017). 
In this study, sharks were observed at more sites 
and with a greater species diversity through analysis 
of video from BRUVS than with UVC conducted by div¬ 
ers. By using UVC, 3 species of sharks were recorded 
at 5 sites, but 7 species were observed in videos from 
BRUVS at 25 sites. Site-specific comparisons of data 
from the use of BRUVS and UVC revealed differences 
in the presence and absence of sharks and in the spe¬ 
cies of sharks observed. These differences may reflect 
species-specific responses (attraction or avoidance) to 
the presence of divers versus traps (Kulbicki, 1998; 
Meekan and Cappo, 2004; Watson and Harvey, 2007). 
For example, in our study, the tiger shark was not ob¬ 
served by divers but was the second-most abundant 
shark observed in videos from BRUVS. Meekan and 
Cappo (2004) also reported that divers rarely encoun¬ 
tered hammerhead sharks (Sphyrnidae) and the tiger 
shark but that those taxa were recorded often in videos 
from BRUVS. Finally, differences in results between 
the use of UVC and analysis of videos from BRUVS 
in our study may reflect the difference in the spatial 
footprint of habitat sampled with the 2 methods: the 
area sampled with video recordings was not necessar¬ 
ily a subset of the habitat sampled with UVC, and in 
some cases there was no overlap at all between the 
areas covered by the 2 methods (Bacheler et al., 2017). 
Sharks fill a key ecosystem role because of their 
predatory activity. Although some species are capable 
of feeding at the apex predator level, this group also 
includes planktivores, scavengers, and mesocarnivores. 
Sharks have been implicated in control of the abun¬ 
dance and biomass of prey in lower trophic levels and 
in maintenance of biological diversity (Heupel et al., 
2014; Osgood and Baum, 2015; Terborgh, 2015; Frisch 
et al., 2016; Barley et al., 2017). Given that the SERFS 
annually samples -1500 sites across the continental 
shelf from North Carolina to Florida, our observations 
of 7 species of sharks in videos from BRUVS deployed 
at a relatively small number of geographically restrict¬ 
ed sites indicate that a targeted analysis, for sharks, 
of videos collected as part of the SERFS could provide 
critical information, without requiring extraction, for 
some of the 42 species 3 of sharks federally managed in 
the southeastern United States. Such an effort has the 
potential to provide estimates of relative abundance 
for assessments of shark stocks, improve predictions 
of spatiotemporal distribution, and aid in the identi¬ 
fication of essential fish habitat for these important 
predators. 
Acknowledgments 
We thank N. Geraldi, D. Grenda, J. Hackney, M. 
Judge, G. Kellison, D. Meyer, and B. Teer for field as¬ 
sistance and the captain and crew of the MW Spree. 
We acknowledge N. Bacheler, J. Carlson, A. Chester, 
A. Hohn, G. Kellison, K. Siegfried, and 3 anonymous 
reviewers for their comments on previous versions of 
this manuscript. Funding was provided by the Marine 
Fisheries Initiative of the National Marine Fisheries 
Service (grant no. 13MFIH0006). 
Literature cited 
Bacheler, N. M., and K. W. Shertzer. 
2015. Estimating relative abundance and species richness 
from video surveys of reef fishes. Fish. Bull. 113:15-26. 
Bacheler, N. M., D. J. Berrane, W. A. Mitchell, C. M. Schober- 
nd, Z. H. Schobernd, B. Z. Teer, and J. C. Ballenger. 
2014. Environmental conditions and habitat characteris¬ 
tics influence trap and video detection probabilities for 
reef fish species. Mar. Ecol. Prog. Ser. 517:1-14. 
Bacheler, N. M., Z. H. Schobernd, D. J. Berrane, C. M. 
Schobernd, W. A. Mitchell, B. Z. Teer, K. C. Gregalis, and 
D. M. Glasgow. 
2016. Spatial distribution of reef fish species along the 
Southeast US Atlantic Coast inferred from underwater 
video survey data. PLoS ONE ll(9):e0162653. 
Bacheler, N. M., N. R. Geraldi, M. L. Burton, R. C. Munoz, 
and G. T. Kellison. 
2017. Comparing relative abundance, lengths, and habi¬ 
tat of temperate reef fishes using simultaneous underwa¬ 
ter visual census, video, and trap sampling. Mar. Ecol. 
Prog. Ser. 574:141-155. 
Barley, S. C., M. G. Meekan, and J. J. Meeuwig. 
2017. Species diversity, abundance, biomass, size and tro- 
3 Atlantic highly migratory species, Table 1 of Appendix A to 
part 635—oceanic sharks, 54 C.F.R. 635 (2018). [Available 
from website]. 
