Separation of pollock larvae from those of similar species in field samples was first accomplished in 1981 [ Dunn and Matarese, 1987). At the same time, the strong year-classes in the mid-1970's resulted in a large fishery [ Megrey, 1990], and hydroacoustic surveys were conducted to estimate the adult population for input to assessment and prediction models. The surveys showed that large concentrations of pre-spawning walleye pollock migrate from the southwest end of Shelikof Strait (taken to include the sea valley that extends from the Barren Islands southwestward to the slope off Chirikof Island, Figure 1) primarily to the area near Cape Kekurnoi. Spawning takes place mainly in the deep sea valley in early April [ Kim and Nunnallee, 1991], at depths between 150-250 m. Each female produces about 1/2 million free floating planktonic eggs in a series of about 10 batches over a period of a few weeks. The eggs reside mainly below 150 m and hatch in about 2 weeks. The short localized spawning pattern creates a large ``patch'' of eggs observed in plankton surveys [ Kendall and Picquelle, 1990]. Mortality rates of the eggs decrease through the spawning season from 0.4/day to 0.1/day [ Kim and Gunderson, 1989].
The larvae are about 3-4 mm Standard Length (SL) at hatching and are
relatively undeveloped, without functioning mouths or eyes. They
quickly rise from their deep hatching depth to the upper 50 m of the
water column, where they drift in the prevailing currents for the next
several weeks (late April through May). They generally remain in
identifiable patches, growing about 0.2 mm/day with a daily mortality
rate of 
8.7% during this time. Their diet consists mainly of
early life stages of copepods [ Canino et al., 1991], and the
size range of prey increases as the larvae grow. Larger larvae undergo
diel migrations (deepest during the day) between 15-50 m. Larvae
appear to stay just below the turbulent wind mixed upper layer of the
water column [ Kendall et al., 1994]. They are visual feeders,
eating mostly during the day [ Kendall et al., 1987].
By late May considerable year-to-year variation exists in abundance
and location of larvae [ Kendall and Picquelle, 1990] as well as
the zooplankton that produce their prey [ Incze and Ainaire,
1994]. Two primary tracks characterize westward larval drift: along
the Alaska Peninsula in the relatively slow moving flow (<10 cm
s
) over the shoreward edge of the sea valley, and offshore in
the rapidly (>25 cm s
) moving Alaska Coastal Current [ Kim
and Kendall, 1989]. By midsummer the larvae have transformed into
juveniles, which by late summer are schooled and become concentrated
in nearshore areas along the Alaska Peninsula [ Hinckley et al.,
1991]. After this period, we know little about their life until they
enter the fishery and become sexually mature at age 
3 years. The
adults live for 
11 years.
We have established the life history pattern between spawning migration and young of the year (Figure 1). Using results from egg and larval surveys, together with estimates of the spawning adult population from hydroacoustic observations and indices of juvenile abundance [ Bailey and Spring, 1992], the stages when year-class strength is established have been determined. A low abundance of larvae results in weak recruitment. A large abundance of larvae, however, does not imply strong recruitment. This suggests that processes during the juvenile stage can also be critical [e.g. Bailey et al., 1994b]. Recently, we have focussed more research on understanding young of the year juveniles since interannual variability in their survival also affects year class strength [ Bailey and Spring, 1992].