Harding et al.: Regional and seasonal patterns of epipelagic fish assemblages from the central California Current 
269 
Table 3 
Species contributions to between-group dissimilarity for three pairs of communities with significantly different community 
structures (i.e., a<0.05 in pairwise PERMANOVA tests), determined by using the SIMPER (similarity percentages) routine. 
Species are listed in order of decreasing percent dissimilarity contribution with an 80% cumulative dissimilarity cutoff imposed. 
Abundance is the untransformed average number of fish/10 6 m 3 within each group: NC=north coast, GF=Gulf of the Farallones, 
S=summer, F=fall. Dominant (Dom.) region and dominant (Dom.) season indicate the region and season, respectively, where 
each species was more abundant in each of the three comparisons presented. For the species Chinook salmon: ad=adult (>250 
mm fork length); jv=juvenile (<250 mm fork length). 
NC 
GF 
Dom. 
Species Dissimilarity 
Contribution 
abundance 
abundance 
region 
Summer NC vs. summer GF, X(dissimilarity)=86.38 
Pacific herring, Clupea pallasii 18.10 
20.96 
127.46 
1644.41 
GF 
jacksmelt, Atherinopsis californiensis 15.48 
17.92 
99.12 
55.56 
NC 
Pacific tomcod, Microgadus proximus 7.23 
8.36 
0.08 
8.94 
GF 
Pacific sardine, Sardinops sagax 5.95 
6.88 
0.07 
184.68 
GF 
Chinook salmon, jv, Oncorhynchus tshawytscha 5.47 
6.33 
13.70 
9.48 
NC 
northern anchovy, Engraulis mordax 4.89 
5.66 
0.06 
2166.27 
GF 
medusafish, Icichthys lockingtoni 3.80 
4.40 
0.28 
1.65 
GF 
Chinook salmon, ad, Oncorhynchus tshawytscha 3.51 
4.06 
2.20 
2.07 
NC 
jack mackerel, Trachurus symmetricus 3.27 
3.79 
2.91 
0.00 
NC 
surf smelt, Hypomesus pretiosus 2.65 
3.07 
6.40 
6.82 
GF 
Fall NC vs. fall GF, S(dissimilarity)=81.04 
Pacific herring, Clupea pallasii 16.11 
19.88 
0.30 
756.19 
GF 
northern anchovy, Engraulis mordax 14.64 
18.07 
17.30 
204.83 
GF 
jacksmelt, Atherinopsis californiensis 13.91 
17.17 
24.42 
184.51 
GF 
Chinook salmon, jv, Oncorhynchus tshawytscha 5.75 
7.09 
2.90 
1.79 
NC 
medusafish, Icichthys lockingtoni 5.08 
6.27 
2.74 
2.80 
GF 
Pacific sardine, Sardinops sagax 4.85 
5.98 
3.22 
6.92 
GF 
Pacific saury, Cololabis saira 3.41 
4.21 
4.71 
0.82 
NC 
Pacific butterfish, Peprilus simillimus 3.16 
3.90 
0.42 
2.43 
GF 
Summer 
Fall 
Dom. 
Species Dissimilarity 
Contribution 
abundance 
abundance 
season 
Summer NC vs. fall NC, 2(dissimilarity)=75.35 
jacksmelt, Atherinopsis californiensis 17.24 
22.88 
99.12 
24.42 
s 
Pacific herring, Clupea pallasii 10.89 
14.46 
127.46 
0.30 
s 
northern anchovy, Engraulis mordax 7.16 
9.51 
0.06 
17.30 
F 
Chinook salmon, jv, Oncorhynchus tshawytscha 6.70 
8.88 
13.70 
2.90 
s 
Pacific sardine, Sardinops sagax 4.47 
5.93 
0.07 
3.22 
F 
jack mackerel, Trachurus symmetricus 4.23 
5.62 
2.91 
0.48 
S 
medusafish, Icichthys lockingtoni 3.82 
5.06 
0.28 
2.74 
F 
Chinook salmon, ad, Oncorhynchus tshawytscha 3.54 
4.70 
2.20 
0.39 
S 
Pacific tomcod, Microgadus proximus 2.26 
3.00 
0.08 
0.91 
F 
Water properties 
The analysis of environmental structure was designed 
to mirror the community analysis. Samples were aver- 
aged, tested, and plotted in the same arrangements 
in order to facilitate direct multivariate comparison 
of ocean conditions and community patterns. Interac- 
tion terms in the three-way PERMANOVA for differ- 
ences in environmental structure were consolidated 
with residuals before tests were run, and all main 
effects in the reduced model were found to be sig- 
nificant (region: pseudo-Fj 14 =8.08, P=0.0002; season: 
pseudo-Fj 14 = 2.58, P=0.0438; year: pseudo-F 5 14 = 1.88, 
P=0.0352). The two-way PERMANOVA for regional 
and seasonal differences was also highly significant 
for both main effects (region: pseudo-F 1 59 = 10.92, 
P=0.0001; season: pseudo-Fj 59 = 8.58, P=0.0001) and 
not significant for regionxseason interaction (pseudo- 
F x 59 = 1.47, P=0.20). All subsequent pairwise compari- 
sons were highly significant (Table 2). 
