Munoz and Burton: Video observations of sharks in reef habitats off east-central Florida 
93 
Table 2 
Rank relative abundance 1 of shark species from the use of 3 different sampling methods: analysis of entire video, 
analysis of a 20-min subsample of video (41 frames), and underwater visual census (UVC). Scuba divers conducted 
UVC surveys simultaneously with deployments of baited remote underwater video stations in July 2014 on the conti¬ 
nental shelf off east-central Florida. Also provided are frequency of occurrence (FO) from analysis of entire videos and 
the mean number of individuals observed in video subsamples or by divers over an area of 300 m 2 . A dash indicates 
that a species was not observed with this method. 
Species 
Sampling method 
Entire 
video analyzed 
Subsample of 
video analyzed 
UVC 
Rank 
relative 
abundance 
FO 
Rank 
relative 
abundance 
Mean 
number of 
individuals 
Rank 
relative 
abundance 
Mean 
number of 
individuals 
Ginglymostoma cirratum 
1 
14% 
3 
0.01 
1 
0.013 
Galeocerdo cuvier 
2 
8% 
- 
- 
- 
- 
Carcharhinus b rev ip i n n a 
3 
7% 
1 
0.30 
- 
- 
C. plumbeus 
4 
5% 
3 
0.01 
2 
0.004 
Rhizoprionodon terraenovae 
5 
4% 
3 
0.01 
2 
0.004 
C. leucas 
6 
3% 
2 
0.03 
- 
- 
Negaprion brevirostris 
6 
3% 
- 
- 
- 
- 
1 For each method, the smaller the number, the greater the relative abundance. See table SI in Bacheler et al. (2017) 
for details about abundance determined with analysis of video subsamples and UVC for all fish species. 
(Wirsing et al., 2007; Nadon et al., 2012; Heupel and 
Simpfendorfer, 2014). In this study, chevron traps as¬ 
sociated with videos that recorded sharks contained 
significantly higher total numbers of individual fish 
than traps associated with videos in which sharks were 
absent, a result that may relate to prey availability at 
sites from which the former videos were collected. 
Depending on the relative abundance of shark spe¬ 
cies encountered at the regional scale, these data could 
be used to generate species-specific abundance indices 
for shark stock assessments. Data gathered from video 
analysis can provide, without the need for extraction 
of sharks, estimates of relative abundance for a vari¬ 
ety of coastal shark species on the continental shelf 
of the southeastern United States. A standardized, 
non-destructive method of video analysis can supple¬ 
ment traditional sampling methods (e.g., those that 
use longlines or gillnets). Brooks et al. (2011) com¬ 
pared estimates of shark relative abundance from the 
use of BRUVS and longline surveys, found significant 
positive correlations between methods for the more 
abundant species, and concluded that the use of videos 
from BRUVS represented a non-invasive, cost-effective 
alternative to the use of longlines for monitoring broad 
trends in the relative abundance of sharks. A shark- 
dedicated addition to the SERFS would complement 
the 3 large-scale regional longline surveys of sharks 
that currently operate in the southeastern United 
States: Cooperative Atlantic States Shark Pupping and 
Nursery Survey, Apex Predators Program Large Coast¬ 
al Shark Survey, and Shark and Red Snapper Bottom 
Longline Survey. These surveys tend to have annual 
sample sizes or areas of coverage that are smaller than 
and mostly inshore of those of the SERFS. 
In our study, 7 species of sharks were observed in 
videos from 25 of 72 sites (35% of sites). This propor¬ 
tion of sites with sharks is high relative to a regional 
survey of the entire Great Barrier Reef, off northeast¬ 
ern Australia, that detected sharks in videos collected 
during 25% of 2438 BRUVS deployments (Espinoza et 
al., 2014), likely a result of the limited sample size and 
spatial coverage of our study. The most common finding 
in both studies is that a single species of shark was 
observed at each site. In our study, the nurse shark 
was observed in greater abundance than other species, 
followed in descending order by the tiger, spinner, and 
sandbar sharks. An assessment of population status for 
the sandbar shark found the stock to be overfished but 
that overfishing was not occurring (SEDAR, 2017), and 
the results of recent studies indicate potential recovery 
since the early 1990s for the tiger, spinner, sandbar, 
bull, and lemon sharks (Carlson et al., 2012; Peterson 
et al., 2017). The only other species observed in our 
study that has been assessed by the National Marine 
Fisheries Service, the Atlantic sharpnose shark, was 
not found to be overfished (SEDAR, 2013). 
The results of our study support the idea that anal¬ 
ysis of videos from BRUVS is superior to the use of 
UVC by divers for detecting sharks, in agreement with 
reports from other studies (Meekan and Cappo, 2004; 
McCauley et al., 2012; Osgood and Baum, 2015; Bar¬ 
ley et al., 2017). The use of BRUVS provides 1) a way 
