122 
Fishery Bulletin 11 6(2) 
Scotian Shelf could be benefiting Atlantic halibut and 
contributing to the stock rebound, both by providing 
habitat protection and by reducing halibut bycatch in 
other fisheries. 
NAFO divisions 5Y, 5Ze, and 5Zw fell short in ju¬ 
venile abundance despite adequate suitable habitat 
availability. Consistent with our findings, Shackell et 
al. (2016) detected low abundance in relation to suit¬ 
able-habitat availability in U.S. waters. They, and oth¬ 
ers, have proposed that this low abundance could be 
the product of finer-scale stock structure or a history 
of overfishing (or both) (Seitz et al., 2016; Shackell 
et al., 2016). Certainly historic fishing pressure is a 
plausible explanation for the observed low abundances 
in divisions 5Y, 5Ze, and 5Zw (Grasso, 2008). As bio¬ 
mass decreased during the 1800s in nearshore fishing 
grounds, exploitation progressed farther offshore and 
also began to include juvenile populations (Grasso, 
2008). The expansion of the fishery extended out from 
Massachusetts Bay (Fig. 1), which was the first re¬ 
gion to see populations decline and which endured the 
longest period of intense exploitation (Grasso, 2008). 
This area, once the home of the commercial fishery for 
Atlantic halibut, has yet to see a rebuilding of hali¬ 
but stocks (Hennen 4 ). The general consensus is that 
halibut were repeatedly depleted in the United States 
and more recently depleted throughout their range in 
the 1990s (Grasso, 2008; Trzcinski and Bowen, 2016). 
The varying rates of putative recovery throughout the 
range may be a reflection of smaller-scale stock struc¬ 
ture within this stock unit. Boudreau et al. (2017) 
observed the temporal persistence of several core ar¬ 
eas of high juvenile halibut abundance alongside the 
disappearance of others. A noteworthy core area that 
disappeared in the 1980s was the area along the shelf 
edge in NAFO divisions 30 and 3Ps (divisions that we 
identified as underperforming), while persistent core 
areas in NAFO divisions 4Vs, 4W, and 4X are associat¬ 
ed with our highest performing divisions. Boudreau et 
al.’s (2017) research supports the Shackell et al. (2016) 
hypothesis that subpopulations may exist within this 
Atlantic halibut stock. The erosion of subpopulations 
could explain why some divisions in our analysis fell 
below the 1:1 line of expected habitat productivity. 
Knutsen et al. (2007) observed genetic subpopula¬ 
tion structure among Greenland halibut (Reinhardtius 
hippoglossoid.es) and that differentiation increased 
with distance. They theorized that 2 genetically differ¬ 
ent populations exist and that ocean currents play a 
major role in determining population structure (during 
larval stages). Notably, for this species, dissimilarity 
between management units and stock structure was 
identified by genetic analysis (Knutsen et al., 2007). 
If subpopulations exist, assessing distinct stocks as a 
single large stock often runs the risk of overestimat¬ 
ing stock size; such a practice was likely responsible 
for the collapse of the stock of Atlantic cod (Gadus 
rnorhua) (e.g., Sterner, 2007). True spatial overlap be¬ 
tween management units and actual underlying popu¬ 
lation structures supports accurate stock assessments, 
and therefore a mismatch between the two is reason to 
revisit respective management regimes (Lundy et al., 
1999; Reiss et al., 2009). Here we do not show that 
subpopulation structures exist within the Scotian Shelf 
and Southern Grand Banks of Newfoundland Atlantic 
halibut management unit, but provide reasons to fur¬ 
ther investigate this possibility. 
It is also possible that deviation from the relation¬ 
ship of juvenile:adult abundance (Fig. 3C) may be in¬ 
fluenced by ontogenetic shifts in habitat and density- 
dependent migration. The mechanisms that influence 
Atlantic halibut movements are not fully understood, 
but in areas that have reached carrying capacity, it is 
more likely that animals will emigrate (Fretwell and 
Lucas 1970; MacCall, 1990). Here, we show a strong 
positive relationship between habitat suitable for ju¬ 
veniles and commercial landings, and an overlapping 
distribution of commercial landings and juvenile habi¬ 
tat. This finding is consistent with those from tagging 
studies that have repeatedly shown that the majority 
of halibut recaptures are in the area of initial tagging, 
despite the occasional far-distance movement by an 
individual (McCracken, 1958; Jensen and Wise, 1961; 
Neilson et al. 5 ; Stobo et al., 1988; Kanwit, 2007; den 
Heyer et al., 2012; Seitz et al., 2016). 
Several next steps can be taken to build on this 
research and further develop an understanding of At¬ 
lantic halibut distribution in the Northwest Atlantic. 
We limited our model to 5 predictor variables; however 
additional environmental and community variables, in¬ 
cluding fishing effort, methods and regulations, prey 
availability, predation, and interspecific competition, 
could influence (and potentially improve) the predictive 
ability of the model. Additionally, by changing existing 
values of environmental variables (such as tempera¬ 
ture or depth), the same model could be used to explore 
potential changes in suitable-habitat distributions in 
response to climate change: a valuable resource for 
planning adaptive measures for a fishery. Overall, our 
analysis supports the pursuit of future research on 
potential subpopulation structures and the impact of 
closed areas on productivity. 
Our analysis builds on current ecological knowledge 
of Atlantic halibut, and can be considered a contribu¬ 
tion for future international discussions and assess¬ 
ments of the stock. Our study has shown that suitable 
habitat can be used as a proxy for juvenile and adult 
halibut abundance, and that there are areas where 
population size is below its potential. The Canadian 
stocks are currently assessed as healthy, and rebuild¬ 
ing is not part of the management objective. However, 
the occurrence of regions in U.S. waters where abun¬ 
dances remain below expectations indicates that there 
is room for population growth, which could potentially 
be achieved through spatial management, more spe¬ 
cifically, by the protection of core juvenile habitat. The 
presence of noteworthy suitable-habitat availability 
outside the EEZ, off the Grand Banks of Newfoundland 
also highlights a need for international collaboration. 
These regions are not regularly surveyed and are sub- 
