308 
Fishery Bulletin 115(3) 
64 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
CT 
CO 
£ 
E, 80 
E 
CO 
o 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
es 
E 
m 
o 93 
c 
.2 
c 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
0 
O 
104 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
9 simulations 
(3 degrees of patchiness 
x 3 captain type) 
3 
5 
7 
Closure duration (years) 
Figure 4 
Matrix design for the sets of simulations used in pairwise comparisons of perfor- 
mance metrics for this study of the management strategy evaluation for the Atlan- 
tic surfclam ( Spisula solidissima ) in the Mid-Atlantic Bight. The matrix is repeated 
for each of the 2 closure location rules (rule 1: ratio of the number of small dams to 
the number of market-size clams; Rule 2: number of small clams per square meter). 
The increase in degree of patchiness approximately doubles between each level 
(i.e., the most patchy distribution is twice as patchy as the intermediate degree 
of patchiness). The 3 captain types are standard (does not search or use survey 
data), survey (uses survey data but does not search fishing grounds), and confident 
(searches but does not use survey data). Definitions of a small clam are given as of 
clams is given as shell length (SL) in millimeters. 
gies that result in a large proportion of simulations 
that showed improvement in comparison with the base 
case (even if the proportional increase is small) are 
preferable because the scenario would be more likely 
to result in improvements if implemented than a sce- 
nario with few simulations showing improvement; that 
is, improvement can be expected over a wider range 
of contingencies influenced by differential recruit- 
ment patterns and captains’ behaviors. It is possible 
that management decisions could be based on a large 
amount of increase even though the possibility of that 
outcome is low. For this reason, investigation of the 
possibility of the outcomes and the magnitude of the 
changes seen are included in this study. In addition 
to a comparison of the sets of alternative management 
and base cases, a second series of comparisons was con- 
ducted between alternative area management strate- 
gies; these offer additional insight as to which manage- 
ment options offer the most benefit. 
In certain scenarios, a 4-year closure duration was 
examined in addition to the 3-, 5-, and 7-year closures. 
The performance metric values for the 4-year closure 
duration routinely fell between the 3- and 5-year clo- 
sure durations performance metric values as seen in 
the number of clams per bushel and LPUE included 
in Table 2 as examples for comparison with simulation 
results discussed subsequently. For this reason, results 
of simulations with the use of the 4-year closure dura- 
tion will not be presented subsequently. 
Results 
Closure location based on rule 1 : the ratio of small clams 
to market-size clams 
Stock density A greater proportion of simulations show 
a significant increase in stock density when the defini- 
tion of a small clam was 93-120 mm SL or 80-120 mm 
SL (Table 3), which is representative of clams expected 
to reach market size (120 mm SL) in <3 and <4 years 
respectively. As the duration of the closure increased 
