318 
Fishery Bulletin 109(3) 
Table 1 
Juvenile walleye pollock (Theragra chalcogramma ) (1732 fish) were collected in 124 trawl catches (i.e., samples) from three 
regions of the western Gulf of Alaska during four seasons from August 2000 to September 2001. Fish size is indicated by mean 
standard length (standard error) and range (minimum-maximum). 
Standard length (mm) 
Region 
Season 
Collection 
date 
No. of 
samples 
No. of 
fish 
Mean (SE) 
Range 
Kodiak 
Late summer 2000 
15-19 Aug 2000 
4 
63 
71 (0.7) 
57-88 
Winter 2001 
1 Feb-21 Mar 2001 
9 
78 
122(1.6) 
95-158 
Summer 2001 
26 Jun-8 Jul 2001 
2 
20 
165 (2.7) 
144-182 
Late summer 2001 
3 Sep 2001 
1 
7 
200(2.3) 
192-207 
Semidi 
Late summer 2000 
8-19 Sep 2000 
45 
784 
71 (0.4) 
37-109 
Winter 2001 
2 Feb-23 Mar 2001 
4 
38 
107 (2.1) 
83-127 
Summer 2001 
9-15 Jun 2001 
4 
40 
118(1.9) 
96-145 
Late summer 2001 
8-18 Sep 2001 
20 
161 
198(2.1) 
136-250 
Shumagin 
Late summer 2000 
4-7 Sep 2000 
16 
374 
60(0.4) 
43-86 
Winter 2001 
14 Feb-9 Mar 2001 
6 
58 
103 (1.8) 
82-153 
Summer 2001 
2-7 Jun 2001 
5 
46 
118(1.7) 
95-149 
Late summer 2001 
4-7 Sep 2001 
8 
63 
173 (3.3) 
127-232 
of the shelf according to prominent bathymetric features 
(Fig. 1) and covered the area alongshore where there is 
a shift in body length of age-0 walleye pollock (Wilson, 
2000). The use of this scheme facilitated integration of 
our results with those of Buchheister et al. (2006) in our 
modeling exercise. 
Two of the nine cruises were regularly scheduled by 
NMFS to study juvenile walleye pollock. Samples were 
collected at predetermined sites between Shelikof Strait 
and the Shumagin Islands during 3-19 September 2000 
and 3-18 September 2001. Samples were collected with- 
out regard to time of day with a Stauffer (anchovy) 
midwater trawl (Wilson et al., 1996). The trawl was 
equipped with a 3 -mm mesh codend liner, and it was 
fished over double-oblique tows to a depth of 200 m or 
to 10 m off bottom, whichever was shallowest. These 
two cruises produced 90 samples. Originally, the data 
from these samples were obtained for a multiyear study 
of food habits of walleye pollock (Wilson et al., 2009). 
The remaining seven cruises provided all the samples 
from WinOl and SumOl, and most of the samples from 
the Kodiak region. Collections of juvenile walleye pol- 
lock on these cruises were ancillary to primary cruise 
objectives (i.e., opportunistic); consequently, we had 
little control over sampling effort, date, collection site 
location, and method of sampling (e.g., gear, time of 
day). On four of these cruises, samples were collected 
without regard to time of day with midwater and bot- 
tom trawls equipped with 3 2 -mm mesh codend liners. 
On the remaining three cruises, samples were collected 
during daylight with a poly-nor’eastern bottom trawl 
equipped with a 3 2 -mm mesh codend liner. These seven 
cruises produced 34 samples. On all nine cruises, wall- 
eye pollock were sorted from the catch and frozen for 
subsequent examination of their stomach contents in 
the laboratory. 
Stomach content analysis 
Fish were selected from thawed samples by standard 
length (SL) to narrow the focus of our study to mem- 
bers of the 2000 year class. Age-0 individuals were 
easily identified from historical distinctions in size that 
separated them from the next older cohort (Brodeur 
and Wilson, 1996). Age-0 individuals were obtained 
from samples collected in LSumOO and WinOl. Age-1 
individuals were obtained from samples collected later 
in WinOl, and in SumOl and LSumOl. The upper SL 
limit of age-1 fish was estimated from NMFS length- 
at-age data according to the method of Buchheister et 
al. (2006). 
Fish food habits were characterized by total stomach 
content weight (SCW) and taxonomic composition (by 
prey weight and number). Fish were thawed in sea- 
water, measured to the nearest mm SL, blotted dry, 
and weighed to the nearest 1 mg (as in Wilson et al. 
[2009]). Stomachs were excised and preserved in 10% 
formalin. Later, stomachs were dissected and the con- 
tents were blotted dry, weighed to the nearest 0.01 mg, 
and sorted by taxonomic group. Although most cope- 
pods were calanoids, some harpacticoids were detected 
and quantified separately. The calanoids, hereafter 
referred to as copepods, were further divided by size: 1) 
small copepods (<2 mm prosomal length [PL]); and 2) 
large copepods (>2 mm PL). Euphausiids were divided 
into the following developmental stages: 1) furciliae (<5 
mm length, Siegel [2000]); and 2) juveniles and adults. 
For each taxonomic group, the total weight and total 
count of largely intact (ca. <50% digested) individu- 
als were used to estimate mean weight per individual 
prey item. Items in each group were enumerated and 
weighed collectively to the nearest 0.01 mg after being 
blotted dry. 
