Asch and Collie: Changes in a benthic megafaunal community due to disturbance from bottom fishing 
441 
Table 1 
Classifications of disturbance and mean scallop dredging effort at study sites based on data from the National Marine Fisheries 
Service (NMFS) and the Canadian Department of Fisheries and Oceans (DFO). Dashes indicate years when no photographic 
samples were collected. D and U indicate shallow sites that were classified as “disturbed” or “undisturbed,” respectively, before 
the development of the NMFS satellite vessel monitoring program in 1998. Classifications made at shallow sites before 1998 were 
based on trawl and dredge marks seen in side can sonar images. No photographic data were gathered in 1995. Dredging effort 
from U.S. and Canadian sites cannot be easily compared because data were reported in different units and the type of fishing 
gear used by each country may have varied. 
Site 
Classification of disturbance 
1994 
1996 
1997 
1998 
1999 
2000 
Shallow sites: Mean dredging effort in hours fished per year 
within two miles (3.2 km) of site 
16 
Disturbed 
— 
D 
D 
— 
— 
— 
17 
Disturbed before establishment of a fishery closure 
(i.e., 1994) and undisturbed afterwards (i.e., 1996-2000) 
D 
U 
U 
0 
0 
0 
17W 
Disturbed 
— 
— 
D 
172.6 
58.1 
— 
18 
Undisturbed before establishment of a closed area 
(i.e., 1994) and disturbed afterwards (i.e., 1996-2000) 
U 
D 
D 
97.4 
1.0 
1.2 
Deep sites: Mean dredging effort in days fished per year within 
a square minute latitude of a transect 
10 
Mildly to moderately disturbed 
2.5 
— 
— 
— 
— 
— 
13 
Disturbed, except for two transects surveyed in 1998 
3.5 
5.0 
4.0 
2.0 
— 
— 
20 
Undisturbed 
0 
0.2 
0.1 
0 
0 
— 
sification of bottom-fishing-induced disturbance at sites 
17 and 18 was reversed in 1995. In order to evaluate 
the effectiveness of the fishery closure, other disturbed 
sites were sampled surrounding CA-II (i.e., sites 16 and 
17W) in later years. 
Fieldwork and laboratory procedures 
Photographs of the seafloor of northeastern Georges 
Bank were collected during seven research cruises, 
which were conducted on a nearly annual basis from 
November 1994 through November 2000. During all 
cruises, except for one occurring in July 1995, the 
SEABed Observation and Sampling System (SEABOSS; 
Woods Hole Science Center, U.S. Geological Survey, 
Woods Hole, MA) was deployed to take 35-mm pho- 
tographs of the seafloor. The SEABOSS consists of a 
tethered van Veen grab sampler that is equipped with 
two Hi-8 video cameras, one still camera, accompany- 
ing lights, and a pressure-depth sensor (Blackwood 
and Parolski, 2001). The SEABOSS also contains two 
parallel lasers placed 20 cm apart, which are used to 
gauge the size of objects on the seabed. The frame of 
the SEABOSS instrument was attached to a winch and 
allowed to drift under the research vessel as it traveled 
at a speed of 1-2 knots (0. 5-1.0 m/s). At each study 
site surveyed during a research cruise, three transects 
were generally conducted with the SEABOSS. However, 
depending on weather conditions, research priorities, 
and available ship time, anywhere between one and 
ten transects were filmed. Because of changes in tidal 
speed, transect length varied between 183 and 4316 m 
(mean distance of 921 m, standard deviation [SD] = 527 
m). Along transects, photographs were taken at 30-60 
s intervals. In addition to the SEABOSS, during the 
1996 and 1999 research cruises, videos and photographs 
of the seafloor were also taken with a MaxRover MK1 
remotely operated vehicle (ROV; Deep Sea Systems 
International, Inc., Falmouth, MA) provided by the 
National Undersea Research Center at the University 
of Connecticut. This ROV system included a set of 
two 35-mm still cameras and two Hi-8 video cameras. 
As with the SEABOSS, two parallel lasers placed 20 
cm apart were used to estimate the size of objects 
on the seafloor. During the 30-minute ROV transects 
(approximately 1000 m in length), photographs of the 
seabed were snapped at 30-second intervals. Each pho- 
tograph taken with the SEABOSS covered an area of 
approximately 0.25-0.27 m 2 of the seafloor, whereas 
approximately 0.31 m 2 of the sea bottom was examined 
in each ROV photo. 
Because of the impracticality of analyzing the hun- 
dreds of photographs taken during each cruise, we ran- 
domly selected a subsample of 12-30 photographs per 
site (average = 16) to be examined in detail each year. 
As a result of the scarcity of quality photos (i.e., not 
blurry or sandy), a reduced number of photographs were 
chosen from site 13 in 1996 and site 20 in 1999. A total 
of 454 benthic photographs from 94 transects surveyed 
from 1994 and 2000 were analyzed. These photographs 
covered an area of 120.8 m 2 of the seafloor. Following 
the method described in Collie et al. (2000a), we over- 
