Vasconcelos et al.: Population identification of Trachurus picturatus in the Northeast Atlantic 
89 
Discriminant function 1 
Figure 8 
Plots of function scores determined with discriminant analysis, 
with sexes combined, for blue jack mackerel (Trachurus pictura¬ 
tus) sampled in 2015 from off the Madeira arhcipelago (black 
crosses), off mainland Portugal (open circles), and off the Canary 
Islands (black triangles). Scores were based on 57 normalized el¬ 
liptical Fourier descriptor coefficients of the first 15 harmonics. 
Stars indicate class centroids per area sampled (each individual 
was allocated to the group with the nearest centroid*. 
uals at higher-latitudes did not only grow faster than 
lower-latitude individuals but also at all temperatures 
that they examined and suggested that populations 
moving to higher latitudes have developed a capacity 
for growth by vertically shifting the thermal reaction 
norms for ascendant growth rates. These thermal re¬ 
actions results in an increase in the growth rates of 
individuals at a certain temperature (Yamahira and 
Takeshi, 2008). Local adaptation in its two forms (tem¬ 
perature adaptation and countergradient variation) 
may also be expected to vary within and among species 
(Yamahira and Conover, 2002). 
The results of this study reveal a clear differentia¬ 
tion between Peniche and the Macaronesian archipel¬ 
agos, especially between Peniche and the Canary Is¬ 
lands. This differentiation may be due to an adaptation 
of the population off mainland Portugal to high-latitude 
environments, and therefore to a faster growth rate 
than that of the Canary Islands population. Consider¬ 
ing that the northeast Atlantic system, the Canary Is¬ 
lands, and the region of the Iberian Peninsula form two 
quite distinct subsystems (Dias, 2015), their separation 
is not only geographic but also a result of the 
unique oceanographic features of the northeast 
Atlantic region; the discontinuity of currents 
imposed by the flow of the Mediterranean Sea 
through the Strait of Gibraltar to the Gulf of 
Cadiz (Dias, 2015). These subsystems may be 
seen as biogeographic breaks that often result 
in abrupt changes in phenotypic traits among 
resident populations (Barria et al., 2014). This 
separation may explain our results that show 
low levels of population mixture between the 
Canary Islands and Peniche. 
Both archipelagos, i.e. Madeira and the Ca¬ 
nary Islands, are under the influence of the 
subtropical gyre of the eastern central Atlan¬ 
tic, which would facilitate the transport of 
planktonic larvae to these archipelagos from 
American, European, and northwest African 
coasts. This gyre may influence the mixing 
levels found between the two archipelagos in 
our study. In the case of the Canary Islands 
(relatively close to Africa—104 km from Cape 
Juby, Morocco), a strong mesoscale distribution 
of the larval community has been recently de¬ 
scribed from filaments of the upwelling system 
from Africa reaching the southeast of this ar¬ 
chipelago (Rodriguez et al., 2004; Becognee et 
al., 2009). This pattern in the larval commu¬ 
nity corroborates our findings and supports the 
distinction of the Madeira and Canary Islands 
stocks, although some mixing of fish from these 
two archipelagos exists. 
The final determination of the existence of 
stocks of blue jack mackerel in the northeast 
Atlantic for assessment and ultimately the 
management of the fisheries must obviously 
be based on more than one method or single 
result. Clearly, the most successful way of de¬ 
fining stock limits is through a holistic approach, in¬ 
volving a combination of a broad spectrum of comple¬ 
mentary techniques (Begg and Waldman, 1999) which 
will, in combination, provide considerable insight into 
the practical identification and delineation of indi¬ 
vidual population subunits or stocks (Elliott et al., 
1995; Cadrin et al., 2005; Waldman, 2005; Waples et 
al., 2008). The different methods (e.g., life history pa¬ 
rameters, tagging, otolith elemental composition, fatty 
acid profiles, parasites as biological tags, morphometric 
landmarks, morphometric outlines, and genetic analy¬ 
sis) that can be used to identify stocks were exhaus¬ 
tively compiled by Cadrin et al. (2005) and updated by 
Cadrin et al. (2014). Considering the existence of three 
possible T. picturatus populations as suggested by the 
results of the present work and in order to maintain 
the sustainability and genetic biodiversity (Begg and 
Waldman, 1999) of this resource in Madeira, the man¬ 
agement strategy should be one in which a precaution¬ 
ary approach is taken (e.g., to regard the fish in the 
three areas studied as separate stocks) while other 
avenues of research are being pursued. A failure to ac- 
