232 
Fishery Bulletin 119(4) 
will be captured if the gear encounters it. Several factors 
can affect selectivity, including fishing gear characteristics 
and fish behavior (Crone et al.”). Selectivity influences the 
relationship between the size composition of the fish caught 
in the survey and the size composition of the population 
(Crone et al.”; Maunder et al., 2014). Therefore, understand- 
ing a fishery-independent survey’s selectivity, along with 
its underlying assumptions, is fundamental to the interpre- 
tation of the resulting index of relative abundance and of 
its ability to track the population trends and cohorts. This 
selectivity can be modeled as curves describing the proba- 
bility of capture at length and may have 4 general shapes: 
increasing, asymptotic, dome-shaped, and saddle-shaped 
(Sampson and Scott, 2012). The most common selectivity 
curve used in stock assessments is flat-topped selectivity, a 
form of asymptotic selectivity in which the proportion of a 
demographic group that is vulnerable to the gear increases 
over the lifespan of the group until all members are fully 
vulnerable (Cadrin et al., 2016). 
The Atlantic cod (Gadus morhua), hereafter primarily 
referred to as cod, has historically been the focal species of 
the New England groundfish fishery. In the Gulf of Maine 
(GOM), commercial landings, mostly caught with trawl 
and gill nets, exceeded 10,000 metric tons (t) in the early 
1990s. However, this level of exploitation has not been 
sustainable because population abundance has been in 
a downward trajectory since the 1980s (NEFSC, 2018a). 
According to the most recent assessment (NEFSC°), the 
GOM cod stock is overfished and experiencing overfishing 
with spawning stock biomass at 6—9% of the biomass tar- 
get. In response to the stock condition, annual quotas have 
been reduced to <1000 t in recent years (GARFO"). 
One of the consequences of continued overfishing has 
been the truncation of the size and age compositions of 
the GOM cod stock. Data from the NEFSC BTS, which 
has involved sampling in the area since 1963 (Grosslein’; 
Azarovitz, 1981), indicate that large cod (=100 cm in total 
length [TL]) were consistently present from the onset of 
the survey until the mid-1980s, when they represented 
on average about 8% of the annual catches (Fig. 1) (data 
from the NEFSC 2019 GOM Atlantic cod stock assess- 
ment [NEFSC°], available from website). Since then, their 
encounter rate has gradually declined and they have 
become rare in BTS catches (<1% on average) over the 
past 10 years. The commercial landings indicate a consis- 
tent but low presence of cod >100 cm TL, typically com- 
posing 1—2% of the annual landings since the early 1980s 
(Fig. 1). In addition to a reduction in abundance of fish 
at this size range, there has been a truncation of the age 
structure of cod caught in the BTS and in the commercial 
fishery beyond age 8, despite a potential maximum age 
greater than 20 years (NEFSC°). 
3 NEFSC (Northeast Fisheries Science Center). In preparation. 
Operational assessment of 14 Northeast groundfish stocks, 
updated through 2018. Northeast Fish. Sci. Cent. Ref. Doc. 
[Available from website.] 
4 GARFO (Greater Atlantic Regional Fisheries Office). 2019. 
Northeast multispecies (groundfish) fishing year 2019 commer- 
cial regulations. 18 July 2019. [Available from website.] 
The white hake (Urophycis tenuis) is another important 
groundfish species in the GOM with recent annual com- 
mercial landings around 2000 t, predominantly caught 
with trawl gear. The stock was considered overexploited 
in 2008 (NEFSC, 2008) and is currently overfished (with 
biomass at less than half of the biomass at maximum 
sustainable yield), although fishing mortality has been 
reduced to a level below the threshold for overfishing 
(NEFSC®). The commercial fishery routinely encounters 
white hake older than 9 years (NEFSC, 2013b; NEFSC°), 
indicating that mature individuals are present in the pop- 
ulation. However, large white hake (=90 cm TL) have been 
uncommon in the size composition of the BTS since the 
beginning of the time series in 1963 (Fig. 1) (data from 
the NEFSC 2019 white hake stock assessment [NEFSC"], 
available from website). 
Consequently, the recent stock assessment applied a 
dome-shaped selectivity pattern at age for indices of abun- 
dance and biomass derived from data collected during 
the spring and fall BTS (NEFSC, 2013b, 2017; NEFSC°); 
this selectivity pattern assumes that the largest or oldest 
members of a demographic group are not fully vulnera- 
ble to survey gear because of gear avoidance (Sampson 
and Scott, 2012; Cadrin et al., 2016). In contrast, white 
hake >90 cm TL have composed a significant portion of 
the commercial landings since the mid-1990s, with at least 
6% of landings exceeding 90 cm TL in most years since 
2000 (Fig. 1). The difference in commercial versus survey 
selectivity is presumably a result of commercial fleets tar- 
geting fishing locations and of different tow speeds and 
durations (K. Sosebee, personal commun.). 
The truncated size (and age) distribution of adult cod as 
well as a decline in weight at age for fish at ages 3 and 
above have been a concern for fishery managers for some 
time (NEFSC, 2008, 2017). Previous discussions and anal- 
yses considered several possible reasons for the trunca- 
tion, such as the presence of fixed gears (e.g., lobster gear) 
in inshore areas that would make potential cod habitat 
unavailable for sampling during the BTS or decreased 
selectivity of large fish in the fishery; however, no support- 
ing evidence was found (NEFSC, 2013a). An alternative 
explanation may be linked to the history of the GOM cod 
stock. A significant decline in abundance, as experienced 
by this stock, can affect the spatial distribution of a stock. 
Range contraction and density-dependent habitat selec- 
tion can occur in depleted stocks (Holt, 1987; Lawton, 
1993), and both processes have been documented for cod 
(Swain and Wade, 1993; Hutchings and Reynolds, 2004; 
Tamdrari et al., 2010). Other factors, such as prey distri- 
bution and climate change, may also influence the spatial 
distribution of groundfish in general and of cod in particu- 
lar (Nye et al., 2009; Ames and Lichter, 2013; Orio et al., 
2019). Given the combined circumstances of overfishing 
and changes in environmental conditions in the ecosystem 
of the GOM (NEFSC®), it is possible that the spatial 
5 NEFSC (Northeast Fisheries Science Center). 2019. State of the 
ecosystem 2019: New England, 31 p. Report to the New England 
Fisheries Management Council. [Available from website.] 
