Watson et al: Experimental analyses of the mechanisms leading to lobster ventless trap saturation 
213 
and after, a soak time of 24 h to determine if ventless traps 
captured enough lobsters to reduce the local population, 
resulting in fewer entries on the second day of fishing; 
2) either pre-stocking ventless traps with lobsters before 
deploying them, or removing the lobsters captured after 
the first 24 h, to determine if lobsters in a ventless trap 
reduce the rate of entry of additional lobsters; 3) fish¬ 
ing ventless traps with new bait, versus 1-day-old “pre¬ 
soaked” bait, to determine if bait loses attractiveness after 
24 h; and 4) measuring the rate of leaching of putative 
lobster attractants from bait to determine if this might 
explain the reduced entry of lobsters into traps after 24 h 
of fishing. Some of these experiments involved deploying 
traps for 24 h, and in others we soaked traps for 48 h. The 
methods below, and the results, are organized according to 
both the type and duration of each experiment. 
Study area 
All of the experiments were carried out just off Wallis 
Sands State Beach in New Hampshire at depths of 7-12 m. 
This location was chosen based upon previous, comparable 
studies of lobster trap dynamics that also occurred at this 
site (Jury et al., 2001; Watson and Jury, 2013; Clark et al., 
2015, 2018). Furthermore, the bottom substrate in this 
location is mostly sand, which makes it easier to conduct 
replicated, large-scale, transect surveys without having to 
disturb either the habitat or the lobsters. This habitat type 
also facilitates analysis of time-lapse videos of lobsters in 
and around traps. Finally, most commercial lobstermen 
do not fish in this area, so there was less chance of traps 
that were not part of our experiment being unexpectedly 
deployed and influencing the results. 
Most of the lobsters captured at this site were below 
the minimum legal size limit (83 mm in carapace length 
[CL]), with an average CL of 64.6 mm (standard devi¬ 
ation [SD] 10.3; range: 34-117 mm; n=3005; see Clark 
et al., 2015, for the size-frequency distribution of lobsters 
captured by ventless traps and observed during scuba 
surveys at this same study site). In the 12-week period 
(mid-July-early October, 2013) during which we con¬ 
ducted this study, the catch per unit of effort remained 
relatively consistent (Fig. 1). 
Traps 
All of the traps used in this study were identical to the 
single-parlor ventless traps used by the Massachusetts 
Division of Marine Fisheries and the New Hampshire 
Fish and Game Department for their ventless trap stock 
assessment surveys, and they were deployed in the same 
manner as during their surveys (for details, see Clark 
et al., 2015). Typically, we used single-parlor traps that 
were set ~30 m apart, baited with 3-4 newly thawed 
Atlantic herring (Clupea harengus) (with a combined 
weight of approximately 1 kg), unless otherwise stated. 
For a subset of experiments, we used a ventless trap 
equipped with a lobster trap video (LTV) system (see Jury 
et al., 2001; Clark et al., 2015, 2018) to observe lobster 
10 
0 -■-■-----.---T-,-,-,---■ 
1 2 3 4 5 6 7 8 9 10 11 12 
Week 
Figure 1 
Mean catch per unit of effort of American lobsters 
(Homarus americanus) in ventless traps deployed off Wal¬ 
lis Sands State Beach in New Hampshire for a 12-week 
period in 2013 (15 July-9 October). All data presented are 
based on 48-h soak times (n= 12, with 8-13 traps deployed 
per week). Error bars denote standard errors of the mean. 
behaviors in and around traps over the course of the soak 
period. The LTV system in these experiments consisted 
of a time-lapse digital video camera and recorder (Brinno 
BCC200 3 , Brinno, Inc., Taipei City, Taiwan) secured inside 
an Ikelite clear plexiglass case (Ikelite, Indianapolis, IN) 
that was mounted above the trap on a PVC frame. Video 
frames were captured every 5 s, and video recordings were 
played back at 15 frames/s during analysis. The videos 
were taken only during the daylight hours of a typical 
48-h soak because no supplemental light source was used 
in order to avoid the potential influence of artificial lights 
on lobster behavior. 
Density of lobsters around traps before and after a 24-h soak 
This experiment was designed to determine if ventless 
traps temporarily reduce the number of American lob¬ 
sters in the vicinity of a trap after deployment for 24 h. 
On day 1 of the experiment, a single trap without bait 
was deployed, and then a pair of scuba divers conducted 
a transect survey of lobster abundance around the trap 
as described below. During each survey, divers swam 4 
transects of 30 m each, extending out from the trap in the 
4 cardinal directions. This distance was chosen because 
the fishable area of a trap is an estimated 2600 m 2 , or 
a circle with a radius of 28.8 m, based upon data from a 
study by Watson et al. (2009). The width of each transect 
was typically 2-4 m, depending on visibility, yielding a sur¬ 
vey area of 240-480 m 2 for each experiment. Immediately 
after completing the survey, the divers opened a water¬ 
tight container of bait and placed it in a previously empty 
bait bag, in the trap. Lobsters were not handled during 
3 The mention of trade names or commercial companies is for 
identification purposes only and does not imply endorsement by 
the National Marine Fisheries Service, NQAA. 
