580 



Fishery Bulletin 92(3). 1994 



distribution and abundance data to evaluate the role 

 of this northern Gulf of Alaska fjord in the early life 

 history of walleye pollock. Specifically, the objectives 

 of this study were 1) to determine the distribution 

 and abundance of walleye pollock larvae in a glaci- 

 ated fjord, 2) to quantify growth rates of larvae within 

 this fjord and compare growth rates to literature 

 values from other areas, and 3) to estimate hatch 

 dates of the observed population. 



Materials and methods 



Resurrection Bay is a fjord approximately 32 km long 

 and 4-8 km wide, located within the coastal moun- 

 tain range on the Kenai Peninsula on the south- 

 central coast of Alaska (Fig. 1). The fjord's bathym- 

 etry shows an inner basin with a maximum depth of 

 300 m, separated by a sill from the outer basin. The 

 sill is located about 15 km from the fjord's mouth at 

 the narrowest point, between our sampling stations 

 RES 2.5 and RES 3 (Fig. 1), and rises to a depth of 

 approximately 185 m. The outer basin is slightly 

 shallower (265 m) than the inner basin and has an 

 open connection with the shelf. 



Six stations were sampled along the fjord axis ( Fig. 

 1) during two cruises, 1-4 May 1989 and 7-9 June 

 1989. Ichthyoplankton samples for this study were 

 collected from the RV Little Dipper, a 9-m aluminum 

 boat. Horizontal plankton tows were taken at dis- 

 crete depths by using a 1-m 2 Tucker trawl, rigged 

 with two 505-u mesh nets. Because no previous data 



Figure 1 



Map of Alaska and Resurrection Bay including stations sampled 



were available we took samples throughout the wa- 

 ter column. We tried to obtain at least one sample 

 from each of the following depth strata per station: 

 0-15 m, 15-30 m, 30-50 m, 50-80 m, 80-150 m, and 

 150-m to the bottom. Because of weather and time 

 constraints, fewer samples were taken at some sta- 

 tions. Sample depths were initially estimated from 

 wire angle and length of extended wire. Actual depths 

 were recorded with an attached Seabird Seacat con- 

 ducting-temperature-depth (CTD) (SBE 19) profiler 

 and retrieved after completion of the cruise. The nets 

 were rigged to a double tripper which allowed the 

 second net to be opened and closed via a messenger 

 from the surface. The net was towed for five minutes 

 in the direction of tidal flow at a towing speed of 1.5 

 to 2.5 knots. Only daytime tows were made. Volume 

 filtered during each tow was calculated from a TSK 

 or General Oceanics flowmeter that was attached in 

 a central position to the mouth of the net. Samples 

 used for this analysis were immediately preserved 

 in 509r isopropyl alcohol or 959f ethyl alcohol. The 

 alcohol was renewed for each sample after 24 hours 

 and after 2-3 days. Because differential shrinkage was 

 observed between preservatives, only larvae preserved 

 in isopropyl alcohol were used in size comparisons. 



A Seabird CTD Profiler was attached to the net 

 during most tows to record conductivity, tempera- 

 ture, and pressure throughout the tow. When no CTD 

 data were recorded, depth was estimated from the 

 wire angle and the length of extended wire. In addi- 

 tion, CTD data were taken at each station and along 

 cross-fjord transects through 

 each station. Because of equip- 

 ment failure, no temperature 

 and salinity data were obtained 

 during the June cruise. Addi- 

 tional CTD-profiles for RES 2.5 

 and GAK 1 were obtained from 

 a cruise on 6 April of the same 

 year. 



Samples were sorted in the 

 laboratory to isolate finfish lar- 

 vae. Walleye pollock larvae 

 were identified and measured to 

 standard length (SL). Densities 

 in larvaem ,! were calculated 

 and abundance in larvaem ' 2 at 

 each station was estimated by 

 integrating larval densities over 

 the water column by using ver- 

 tical distribution profiles. Den- 

 sity was set to zero at the sur- 

 face and was assumed to change 

 in a linear fashion between suc- 

 cessive sampling depths. Be- 



