292 
Fishery Bulletin 119(4) 
Table 10 
Comparison of mean, standard deviation (SD), and median values and quartiles of model estimates of dredge efficiency between 
all depletion experiments and subsets of experiments flagged by all error terms or by error terms Err2, Err3, and Err4. Exper- 
iments were flagged because estimates were at or above the 80th percentile for an error term. Data used in the model are from 
depletion experiments conducted during 1997-2011 for populations of ocean quahogs (Arctica islandica) and Atlantic surfclams 
(Spisula solidissima) off the mid-Atlantic coast of the United States. 
Efficiency estimate 
Species 
Ocean All experiments 
quahog 
Atlantic 
surfclam 
All experiments 
Interestingly, the mean and median efficiency estimates 
for the hydraulic dredges that targeted Atlantic surf- 
clams and ocean quahogs are nearly identical. Neither 
the species nor the presence of one generally in deeper 
water than the other significantly influences the overall 
efficiency that stands at approximately 70% regardless of 
mean or median determination. 
The analyses of this study permit the evaluation of the 
factors involved in defining a standard operating protocol 
for experiments of this kind that are inherently extremely 
expensive to conduct. Several characteristics that define 
the success of the experimental design, such as the disper- 
sion of clams on the bottom, will likely be unknown. As a 
consequence of the uncertainty behind these experiments, 
a sufficient number of replicates will always be required in 
order to provide a useful recommendation for catchability. 
Effectively the same conclusion was made in other studies 
of dredge calibration with a significant number of experi- 
ments (Hennen et al., 2012; Morson et al., 2018). 
Estimates from the Patch model are useful in develop- 
ment of stock assessment models in the future; for exam- 
ple, capture efficiency estimates can be used to form 
prior distributions for catchability parameters (NEFSC, 
2010a, 2010b). However, these estimates are only as use- 
ful as the data from the depletion experiments used to 
inform the Patch model. Three groups of experiments 
have significantly different efficiency estimates and 
CVs from the remainder, as indicated by results of the 
Wilcoxon rank sums tests conducted on efficiency esti- 
mates flagged by Err2, Err3, and Err4. Although these 
error estimates can be used only to infer experimental 
quality, they identify experiments with a range of ques- 
tionable attributes that strongly implicate them as out- 
liers biasing the efficiency estimates for the entire data 
set. Removing these questionable experiments from the 
NMFS depletion data set provided the best estimates of 
efficiency for these commercial hydraulic dredges, and 
the results of the removal indicate that these dredges 
are highly efficient and minimize the degree of bottom 
contact relative to the catch. 
1st 3rd 
Mean SD quartile Median quartile 
0.586 0.260 0.381 0.629 0.779 
9 experiments flagged by error terms 2, 3, and 4 removed 0.711 0.195 0.629 0.667 0.795 
10 experiments flagged by all error terms removed 
0.758 0.169 0.641 0.716 0.898 
0.635 0.229 0.533 0.590 0.779 
7 experiments flagged by error terms 2, 3, and 4 removed 0.694 0.196 0.570 0.647 0.852 
13 experiments flagged by all error terms removed 
0.738 0.172 0.584 0.733 0.889 
Acknowledgments 
This research was supported by the Science Center for 
Marine Fisheries, an Industry—University Cooperative 
Research Center of the National Science Foundation 
(NSF), through membership fees under the direction of the 
Industry Advisory Board provided through NSF award no. 
1266057 and no. 1841112. We thank the NOAA Northeast 
Fisheries Science Center for making the data sets used in 
this study available, and we acknowledge the scientists and 
crew who participated in the depletion experiments at sea. 
Literature cited 
Chai, A.-L., M. Homer, C.-F. Tsai, and P. Goulletquer. 
1992. Evaluation of oyster sampling efficiency of patent 
tongs and an oyster dredge. North Am. J. Fish. Manage. 
12:825-832. 
Clausen, S. E. 
1998. Applied correspondence analysis: an introduction, 80 p. 
Sage Publ. Inc., Thousand Oaks, CA. 
Da Ros, L., N. Nesto, C. Nasci, V. Moschino, D. Pampanin, and 
M. G. Marin. 
2003. Biochemical and behavioural effects of hydraulic 
dredging on the target species Chamelea gallina. Fish. Res. 
64:71-78. 
Gedamke, T., W. D. DuPaul, and J. M. Hoenig. 
2005. Index-removal estimates of dredge efficiency for sea 
scallops on Georges Bank. North Am. J. Fish. Manage. 
25:1122-1129. 
Gilkinson, K. D., G. B. J. Fader, D. C. Gordon, R. Charron, 
D. McKeown, D. Roddick, and Q. Liu. 
2003. Immediate and longer-term impacts of hydraulic clam 
dredging on an offshore sandy seabed: effects on phys- 
ical habitat and processes of recovery. Cont. Shelf Res. 
23:1315-1336. 
Gilkinson, K. D., D. C. Gordon, K. G. MaclIsaac, D. L McKEown, 
E. L. R. Kenchington, C. Bourbonnais, and W. P. Vass. 
2005. Immediate impacts and recovery trajectories of mac- 
rofaunal communities following hydraulic clam dredg- 
ing on Banquereau, eastern Canada. ICES J. Mar. Sci. 
62:925-947. 
