Johansson et al.: Seasonal distribution and environmental associations of larval and juvenile Sebastes spp. 
277 
2002; Auth et al., 2011). Our work off the coast of Or¬ 
egon, like previous research in and around Southern 
California (Taylor et ah, 2004; Thompson et ah, 2016; 
Thompson et ah, 2017), has helped illuminate the di¬ 
versity of the largest of these previously unidentifiable 
groups of fishes. 
Because of the distributional patterns we observed, 
we may not have sampled the offshore extent of a num¬ 
ber of our late-larval rockfishes sufficiently because the 
highest catches often were found at the most offshore 
station along our transects. Many species spawn well 
offshore and move progressively shoreward as they age 
before finally settling in most cases in demersal habi¬ 
tat over the continental shelf (Laroche and Richardson, 
1980; 1981). Time constraints precluded us from sam¬ 
pling beyond our normal sampling grid during most 
cruises. However, on one occasion in June 2008, we 
were able to trawl at a station 185 km offshore along 
the Newport transect (Fig. 1). This single trawl tow 
sampled more than 60 km farther offshore than our 
target stations and captured 20,756 rockfish, which 
were mostly late-larval individuals and too small to 
be visually identified. We were able to run genetic 
analysis on a subsample from that haul of 105 indi¬ 
viduals that were identified as mostly widow and yel- 
lowtail rockfish. In a study of larval fish distribution 
out to 364 km from shore along the Newport transect 
and from another transect off Northern California, 
Auth (2009) found that the majority of Sebastes larvae 
were seaward of 120 km during the spring of 2007, but 
were closer to the shore during the summer months. 
However, as acknowledged by the author, because the 
larvae could not be identified to species, these shifts 
in regional observations could be confounded by differ¬ 
ences in cryptic species not observed in that study. Sta¬ 
ble isotope analysis of juvenile rockfish caught on the 
continental shelf also revealed that they had derived 
much of their previous nutrition from taxa found off 
the shelf, indicating potential onshore advection before 
settlement (Bosley et al., 2014). However, as postulated 
by Kamin et al. (2014) for a Gulf of Alaska population 
of Pacific ocean perch (S. alutus), these offshore popu¬ 
lations may be subjected to substantial mortality and 
only a small fraction of the larvae may subsequently 
return to shelf nursery areas. 
Because some closely related groups of rockfishes 
are still undergoing lineage sorting (Hyde and Vetter, 
2007) or may experience mtDNA introgression (Pearse 
et al., 2007), it was not possible to confidently iden¬ 
tify some individuals to species. In our sampling, the 
WEVZ complex represented a substantial proportion 
of the larval catch, and is a still-unresolved diversity 
of species. Although microsatellites represent one ap¬ 
proach to separate this (and potentially other multi¬ 
species complexes) (Pearse et al., 2007), microsatel¬ 
lite peak sizes can vary between sequencers and PCR 
chemistries, and therefore would require standardiza¬ 
tion between different laboratories to produce useful 
data (Weeks et al., 2002; Moran et al., 2006). This need 
for standardization makes it more difficult to compare 
and use microsatellites in databases of known samples 
than to use DNA sequences, as used here. Ultimately, 
the development of a suite of linked diagnostic single¬ 
nucleotide polymorphisms from genomes of all the 
members of the genus Sebastes (microhaplotypes) may 
prove most effective for confidently identifying any spe¬ 
cies by a molecular approach. 
Our study combines the relative speed and ease of 
visually identifying more distinctive species of rock¬ 
fish with the power of current sequence-based genetic 
identification methods. This combined approach pro¬ 
vides new information on the distribution and sea¬ 
sonal timing of pelagic and larval juvenile stages of 
several rockfish species which have been challenging 
to identify on the basis of traditional morphologi¬ 
cal metrics and pigmentation. However, there remain 
some species (i.e., the WEVZ complex) that cannot be 
resolved to species level with the cytochrome b data¬ 
base of known samples. Analysis down to the species 
level for collections that contain a high proportion of 
rockfish (e.g., Thompson et al., 2016) can provide valu¬ 
able information for ecosystem-based fishery manage¬ 
ment. Identifying Sebastes larvae to species may allow 
estimates of spawning biomass of the more abundant 
species (Ralston et al., 2003; Ralston and MacFarlane, 
2010) and enable managers to track recruitment over 
time (Ralston et al., 2013). As the availability of high 
throughput sequencers bring the cost per sample down 
substantially, molecular identification of rockfish spe¬ 
cies should become more commonplace in the future. 
Acknowledgments 
The authors particularly thank the captain and crew 
of the FV Piky for their assistance with sampling. This 
research was funded by the NWFSC Stock Assessment 
Improvement Program; the Oregon State University 
Cooperative Institute for Marine Resources Studies; 
the National Science Foundation Research Experi¬ 
ences for Undergraduates program; and the NWFSC 
Conservation Biology Division. We especially thank A. 
Thompson, G. DiNardo, T. Auth, and 4 anonymous re¬ 
viewers, whose helpful comments greatly improved the 
manuscript. 
Literature cited 
Auth, T. D. 
2009. Importance of far-offshore sampling in evaluating 
the ichthyoplankton community in the northern Califor¬ 
nia Current. CalCOFI Rep. 50:107-117. 
2011. Analysis of the spring-fall epipelagic ichthyoplank¬ 
ton community in the northern California Current 
in 2004-2009 and its relation to environmental fac¬ 
tors. CalCOFI Rep. 52:148-167. 
Auth, T. D., and R. D. Brodeur. 
2006. Distribution and community structure of ichthyo¬ 
plankton off the coast of Oregon, USA, in 2000 and 
2002. Mar. Ecol. Prog. Ser. 319:199-213. 
