Sagarese et al.: Spatiotemporal overlap of Squalus acanthias and commercial fisheries within the northeast U.S. marine system 
113 
Table 3 
Ordinary kriging model performance for prediction of catches of spiny dogfish ( Squalus acanthias) from the sink gillnet 
(SGN) and otter trawl (OT) fisheries and the Northeast Fisheries Science Center bottom trawl survey (Survey) during 
autumn and spring between 1989 and 2009 in the northeast U.S. shelf large marine ecosystem. RMSE=root mean square 
error of prediction; CV=cross-validation; Var=variance. An en dash indicates no data were available. Additional metrics can 
be found in Sagarese. 1 
Year 
Mean prediction RMSE 
100-fold CVVar 
Autumn 
Spring 
Autumn 
Spring 
SGN 
OT 
Survey 
SGN 
OT 
Survey 
SGN 
OT 
Survey 
SGN 
OT 
Survey 
1989 
0.35 
0.99 
29.13 
_ 
1.37 
139.76 
4.71 
8.57 
2.40 
_ 
7.23 
2.90 
1990 
0.32 
0.81 
58.83 
2.67 
1.36 
309.12 
3.82 
12.26 
4.35 
0.07 
6.70 
2.03 
1991 
0.53 
1.43 
108.86 
2.17 
1.31 
90.94 
4.27 
8.22 
2.87 
0.36 
7.02 
1.84 
1992 
0.89 
1.00 
88.06 
1.69 
1.40 
89.23 
4.40 
4.56 
1.77 
1.04 
7.59 
2.57 
1993 
0.98 
0.69 
53.23 
2.27 
1.73 
63.03 
5.08 
4.79 
2.31 
0.92 
4.28 
1.70 
1994 
1.90 
1.78 
43.44 
2.42 
1.71 
180.14 
0.34 
4.37 
2.31 
0.61 
7.37 
2.09 
1995 
1.53 
1.69 
107.64 
2.25 
1.65 
39.02 
3.34 
3.91 
2.62 
1.18 
4.24 
3.81 
1996 
1.12 
1.68 
101.36 
2.35 
1.91 
99.61 
3.46 
5.07 
2.58 
0.58 
2.55 
2.16 
1997 
1.34 
1.89 
58.86 
1.97 
1.61 
43.53 
4.27 
3.63 
2.47 
1.40 
3.36 
3.75 
1998 
1.23 
1.53 
92.84 
1.84 
1.99 
34.09 
3.99 
2.18 
3.00 
2.73 
4.51 
3.47 
1999 
1.17 
1.61 
51.07 
1.67 
2.02 
48.72 
3.28 
4.22 
2.22 
0.60 
3.05 
4.76 
2000 
2.63 
2.23 
36.73 
2.16 
1.58 
28.50 
0.50 
1.87 
1.42 
2.72 
5.03 
3.78 
2001 
2.51 
1.61 
72.45 
2.76 
1.44 
42.29 
0.58 
3.84 
2.71 
0.05 
4.57 
2.72 
2002 
2.13 
1.57 
49.65 
2.41 
0.81 
56.69 
0.27 
3.41 
3.13 
0.60 
4.22 
3.34 
2003 
1.56 
1.61 
78.64 
2.73 
1.63 
84.51 
4.47 
4.69 
4.03 
1.38 
4.03 
3.08 
2004 
1.00 
1.38 
55.32 
2.45 
1.58 
46.27 
3.25 
4.67 
3.00 
2.59 
4.51 
2.25 
2005 
1.20 
1.71 
94.89 
2.66 
2.06 
68.57 
4.06 
3.29 
2.54 
0.10 
4.01 
3.79 
2006 
1.97 
1.25 
72.39 
2.67 
1.80 
85.31 
2.24 
5.88 
3.11 
1.17 
3.88 
3.03 
2007 
1.88 
1.40 
71.29 
2.48 
1.86 
45.62 
0.62 
8.25 
2.08 
1.16 
4.02 
2.56 
2008 
1.54 
1.97 
53.40 
2.44 
1.81 
46.04 
0.60 
3.71 
3.22 
0.01 
4.65 
2.73 
2009 
0.71 
1.80 
85.35 
2.27 
2.10 
143.12 
0.61 
6.60 
4.54 
0.05 
3.70 
3.31 
1 Sagarese, S. R. 2013. The population ecology of the spiny dogfish in the Northeast (US) shelf large marine ecosystem: im- 
plications for the status of the stock. Ph.D. diss., 501 p. Stony Brook Univ., Stony Brook, NY. 
LME. Traditionally, mature female spiny dogfish were 
preferentially harvested for maximal profit and be- 
cause they were close to shore (Shepherd et al., 2002; 
Sagarese et al., 2014b). Other factors, such as environ- 
mental conditions, also influence catchability observed 
in analyses of data from both surveys and fisheries, 
for example, by concentrating fish in preferred habitats 
(e.g., Smith and Page, 1996). Environmental conditions 
at the time of the NEFSC trawl survey influenced the 
probability of occurrence of life-history stages of spiny 
dogfish throughout the NES LME (Sagarese et al., 
2014a). Changes in availability to monitoring surveys 
could affect abundance estimates used in stock assess- 
ment models (Methot and Wetzel, 2013). 
Changes in the spatial distribution of spiny dogfish 
may also affect their availability to both commercial 
fisheries and monitoring surveys. Any change in spa- 
tial structure of the stock (Pennington and Godp, 1995), 
such as the documented shoreward shifts of spiny dog- 
fish (NEFSC 3 ), can increase availability and catch- 
ability if fish density increases on commercial fishing 
grounds. Increased availability to commercial fisheries 
can increase bycatch, which corresponds to higher lev- 
els of mortality from OTs (Mandelman and Farrington, 
2007). 
For spiny dogfish, a relationship was identified be- 
tween abundance of mature females in the bottom trawl 
survey and their availability to each fishery. During au- 
tumn, availability to the OT fishery increased with ma- 
ture female abundance. The plausibility of this result is 
supported by recent field studies that examined devices 
or repellents aimed at deterring spiny dogfish from mul- 
tiple gears, including trawls, in the NES LME (Chosid 
et al., 2012; O’Connell et al., 2012). For the SGN fishery, 
an opposite trend of reduced availability was observed 
when there was a higher abundance of mature female 
spiny dogfish during spring. A potential explanation, 
based on MacCall (1990), is that as abundance increases 
mature females spread into offshore areas or seasonal 
areas less frequented by SGN fishing crews. It also 
is possible that the SGN fishery actively avoids large 
aggregations of mature female spiny dogfish because 
this nuisance species can overwhelm SGNs and either 
saturate some nets or destroy them (Tallack and Man- 
