Laman et al Correlating environmental and biogenic factors with abundance and distribution of Sebastes alutus in Alaska 
283 
eratively against excluded years, we showed that the 
conditional abundance GAMs did a fair job of predicting 
juvenile Pacific ocean perch abundance and a better job 
for adults than for juveniles. The deviance explained in 
each of the jack-knifed cross-validations ranged from 
23.2% to 28.5% for juveniles and from 41.4% to 44.8% 
for adults. Examination of diagnostic plots and residu- 
als indicated that model assumptions were met. 
Discussion 
Depth, temperature, and proximity to major Aleu- 
tian passes (D,T and Long., respectively) were domi- 
nant influences in our GAMs ( i.e. , accounted for the 
greatest proportion of the deviance explained) and op- 
erated over broad spatial scales. These effects compared 
favorably with similar findings of other researchers by 
confirming the depth distributions and temperature af- 
filiations of Pacific ocean perch life stages (Carlson and 
Haight, 1976; Brodeur, 2001; Love et al., 2002; Rooper 
and Boldt, 2005; Rooper 2008). The GAMs predicted in- 
creases of occurrence and abundance of Pacific ocean 
perch in proximity to most major Aleutian passes (e.g., 
Amukta, Seguam, and Buldir) that were similar to 
findings by Logerwell et al. (2005). We also determined 
that the inclusion of predictors that act over more local 
scales (e.g., biogenic structures and slope of the bottom 
at a sampling station) can improve the GAM fit and 
enhance our understanding of Pacific ocean perch dis- 
tribution and abundance patterns within the context of 
broader scale oceanographic processes. 
Logerwell et al. (2005) concluded that patterns of 
demersal fish distribution and abundance could reflect 
proximity to biological “hot spots,” where prey aggre- 
gations form and productivity increases because of 
the convergence of bathymetric and hydrographic pro- 
cesses, such as those observed near Aleutian passes. 
There is substantial tidal mixing in the passes (Ladd 
et al., 2005), and the dominant flow of water through 
them is from the south to the north (Stabeno et ah, 
1999). The tidal mixing, the persistent northward wa- 
ter movement, and the interaction of mixed water with 
the euphotic zone near these passes often create local 
