Cullen and Stevens: Underwater video recordings of Centropristis striata in waters off Maryland 
411 
Figure 2 
Diagram of the camera system used to collect video of black sea bass ( Centro- 
pristis striata ) from 14 June to 4 August 2011 at 6 sampling sites in waters 
off the coast of Maryland. The system consisted of a rectangular fish trap as 
a base, with a steel frame equipped with 5 GoPro HD Hero 1 cameras mount- 
ed over the top. Four cameras were faced outward at 45° angles to obtain a 
standardized view of fish and bottom habitat near the trap. A fifth camera 
was faced downward at a 45° angle over the top of the trap to capture behav- 
ioral responses of black sea bass to the trap that may have occurred during 
each deployment of the fish trap and camera system. 
tions. Approximately 20 min were allowed to elapse 
between each deployment. 
With the intention of obtaining observations of re- 
sponses of black sea bass to the trap, we baited the 
trap with -230 g of northern shortfin squid ( Illex il- 
lecebrosus ) during the first 2 deployments per day. For 
these deployments, whole frozen squid were thawed, 
cut into strips, placed in a plastic mesh bait bag, and 
hung inside the trap kitchen. The bait bag was re- 
moved for the last 2 deployments. 
Video analysis 
Nearly 160 h of video were collected during the study 
period. Videos used for analysis were selected ran- 
domly. The 4 outward facing cameras were assigned a 
number 1 to 4 and a random number generator was 
used to select one camera for each sampling date. Al- 
though 4 cameras were used, a single video was chosen 
from each of the 4 deployments made each day to help 
reduce recounts of fish moving in and out of multiple 
camera views. The 40 selected videos were viewed on a 
wide-screen monitor with standard video editing soft- 
ware (Adobe Premiere Pro CSS; Adobe Systems Inc., 
San Jose, CA); no videos were excluded from analysis 
or substituted with others from another camera be- 
cause all displayed a clear view of the bottom habitat 
and fish when present. Video processing began ~1 min 
after the camera view was clear of silt or debris sus- 
pended when the camera system landed on the bottom. 
Habitat appearing on videos was classified into 3 types: 
1) sand — smooth or coarse sand and gravel adjacent 
to structure, bivalve shells were often present but no 
rocks or boulders (i.e., small cobble to large rocks; sedi- 
ment types were identified based on definitions from 
Wentworth [1922]), 2) sand+rock — sand with scattered 
rocks and boulders but no rocky outcroppings or cor- 
al species present, and 3) live bottom — complex reef 
habitats with boulders, rocky outcroppings, and possi- 
bly other structures colonized by gorgonian sea whips 
(. Leptogorgia spp.) and stony corals. Bottom habitats 
were classified as live bottom because they were fre- 
quently occupied by other species in addition to black 
sea bass and corals, including cunner ( Tautogolabrus 
adspersus), American lobster ( Homarus americanus), 
and crabs ( Cancer spp.). In video frames where more 
than sand habitat was present, the habitat was clas- 
sified as sand+rock only when rocks or boulders were 
present but no rocky outcroppings or coral species and 
as live bottom only when corals (sea whips, stony cor- 
als) were present in addition to other species and habi- 
tat features. 
On the basis of the pattern of counts of black sea 
bass over time (i.e., fish counts generally increased to 
