102 
Fishery Bulletin 113(2) 
Zone in 1977, domestic fishery landings averaged 208 
t (SD 185) (Rago and Sosebee, 2009). The majority 
of spiny dogfish landed during the 1970s and 1980s 
(mean: 4500 t [SD 1280]) were indirectly caught by 
otter trawls (OT) as bycatch in groundfish and other 
fisheries (Rago and Sosebee, 2009). After the depletion 
of commercially important groundfish stocks, including 
the Atlantic cod ( Gadus morhua) in the 1980s, a direct- 
ed domestic spiny dogfish fishery commenced in 1990 
(Rago et al., 1998), which predominantly used sink 
gillnets (SGN) to target spiny dogfish (Rago and Sos- 
ebee, 2009). Domestic landings increased substantially 
throughout the 1990s (mean: 18,800 t [SD 4230]) and 
peaked at 27,200 t in 1996 (Rago and Sosebee, 2009). 
In 1998, the National Marine Fisheries Service de- 
clared the stock overfished as a result of rapid expan- 
sion of the fishery and overharvest of large fecund fe- 
males (ASMFC 2 ). In 1999, nearly twice as many spiny 
dogfish (9.3 million versus 4.6 million) were needed to 
match the same weight (17,000 t) landed in 1992 (Rago 
and Sosebee, 2009). This substantial exploitation drove 
recruitment to record lows between 1997 and 2003 
(Rago and Sosebee, 2009), leading to implementation 
of regulatory measures, such as trip quotas and strict 
regulations, in the early 2000s to reduce commercial 
harvest (ASMFC 2 ). The population was expected to be 
rebuilt by 2020 (ASMFC 2 ); however, target reference 
points for spawning stock biomass (i.e., mature female 
abundance) were met in 2008 (Rago and Sosebee 1 ), 
partly the result of an abnormally large spawning 
stock estimate from the bottom trawl survey conducted 
in spring 2006. This high estimate was the outcome of 
above average catches in 5 separate survey strata that 
may have been caused in part by dogfish schooling be- 
havior and shifts in concentration from Southern New 
England and Georges Bank to the Gulf of Maine and 
Mid-Atlantic Bight (NEFSC 3 ). This nearly 5-fold in- 
crease in estimated stock size between 2005 and 2006, 
however, appears biologically unrealistic given the slow 
life-history characteristics of the species (NEFSC 3 ). 
Collection of fishery-dependent data offers an in- 
expensive opportunity to obtain highly detailed data 
on commercially exploited species over large tempo- 
ral and spatial scales (Bertrand et al., 2004; Hilborn, 
2007). Catch per unit of effort (CPUE), an important 
metric derived from either fishery-dependent or fish- 
ery-independent data, is often assumed proportional to 
abundance (Hilborn and Walters, 1992). In nature, this 
assumption rarely holds because nonlinear relation- 
ships often arise between CPUE and stock abundance 
2 ASMFC (Atlantic States Marine Fisheries Commission). 
2002. Interstate fishery management plan for spiny dogfish. 
Fishery Management Report No. 40, 107 p. ASMFC, Wash- 
ington, D.C. [Available at http://www.asmfc.org/uploads/file/ 
spinyDogfishFMP.pdf.] 
3 NEFSC (Northeast Fisheries Science Center). 2006. 43rd 
Northeast Regional Stock Assessment Workshop (43rd SAW): 
43rd SAW assessment report. Northeast Fish. Sci. Cent. 
Ref. Doc. 06-25, 400 p. [Available from http://www.nefsc. 
noaa.gov/publications/crd/crd0625/.] 
because of changes in catchability (Arreguin-Sanchez, 
1996; Harley and Myers, 2001; Salthaug and Aanes, 
2003). 
Defined as the proportion of the population biomass 
caught by 1 unit of effort (Hilborn and Walters, 1992), 
catchability incorporates both the proportion of the 
stock accessible (i.e., available) and the proportion of 
fish in the swept volume that are caught by the gear 
(i.e., vulnerable) (Michalsen et al., 1996; Francis et al., 
2003; Trenkel et al., 2004). The probability that an in- 
dividual encountered by the gear is captured is called 
gear efficiency (Trenkel et al., 2004). Catchability is 
often parsimoniously assumed constant in both space 
and time (Godo, 1994; Pennington and Godo, 1995; 
Aglen et al., 1999) but can vary with the environment 
(Swain et al., 2000), fish behavior (Frisk et al., 2011), 
or fleet dynamics (Bertrand et al., 2004), among other 
factors. In particular, CPUE estimates obtained from 
fishery-dependent data are often discouraged as a mea- 
sure for relative stock abundance. However, examina- 
tion of trends from fishery-dependent data can enhance 
understanding of fishery distribution, fishery behavior, 
and fishery-fish interactions over time. For spiny dog- 
fish, simultaneous investigation of fishery-dependent 
and fishery-independent information may provide criti- 
cal insight into whether commercial fisheries and spiny 
dogfish are becoming more coincident in time and space, 
a potential explanation for unacceptably high catches 
of spiny dogfish (Tallack and Mandelman, 2009). 
Survey-derived estimates of abundance provide the 
best available science for quantifying distribution of 
spiny dogfish and are used in U.S. stock assessment 
models to describe population dynamics. However, con- 
cerns regarding spiny dogfish availability to the NEF- 
SC bottom trawl survey are well documented (Carlson 
et al., 2014; Sagarese et al., 2014a; Sagarese et al., 
2014b). Complex seasonal and transboundary migrato- 
ry behaviors may displace spiny dogfish outside of the 
survey domain (Nye et al., 2009). Further, the avail- 
ability of spiny dogfish to the survey varies with envi- 
ronmental conditions (e.g., bottom temperature) at the 
time of the survey (Sagarese et al., 2014a). Relatively 
warmer water temperatures early in the season may 
cue migration earlier from southern wintering grounds 
(North Carolina) to northern feeding grounds (north- 
ern U.S. and Canadian waters), resulting in a greater 
portion of the population at the northernmost extent 
of the survey domain in the United States (and poten- 
tial emigration into Canadian waters) (Sagarese et al., 
2014a). Additionally, spiny dogfish occur south of Cape 
Hatteras, the southernmost point sampled by the sur- 
vey, where densities may depend upon the position of 
the Gulf Stream (Rulifson and Moore, 2009). 
This migratory behavior of spiny dogfish has result- 
ed in a highly seasonal fishery, with fisheries gener- 
ally operating in New England during the summer and 
off North Carolina during the winter (Camhi, 1998). 
In the late 1990s, the implementation of semiannual 
quota periods (period 1: May 1-Oct. 31; period 2: Nov. 
1-Apr. 30) prevented southern fishermen from harvest- 
