282 



Fishery Bulletin 9 1(2), 1993 



Figure 1 



Location of vertical distribution studies, 1981-86, on Georges 

 Bank, designated by year and sampled sites. Sites (Roman 

 numerals) within a cruise have same symbols. 



Lough 1984, Smith & Morse 1985, Lough & Bolz 1989). 

 By late July-early August, the highest catches of juve- 

 nile gadids are usually found on eastern Georges Bank 

 (Lough etal. 1989). 



Haddock and cod undergo typical development and 

 morphological transformation at sizes >20mm stan- 

 dard length (SL, -2-3 mo post-hatching) (Fahay 1983, 

 Auditore et al. 1993). Their transition from pelagic to 

 demersal life occurs when they are >25mm in mid- 

 summer (Lough et al. 1989). 



Information on larval and juvenile vertical distribu- 

 tions and diel migrations are needed to (a) study their 

 transport by currents, (b) compare vertical distribu- 

 tions of their prey and predators, and (c) make accu- 

 rate abundance estimates of various stages caught by 

 different sampling gear. In this paper we summarize 

 data on the vertical distribution of 0-group haddock 

 and cod during spring and summer on Georges Bank 

 from 1981 to 1987. The following questions on the early 

 life of haddock and cod were addressed. ( 1 ) What is 

 the vertical distribution pattern of larvae and juve- 

 niles on Georges Bank? (2) What is the timing and 

 range of diel migratory behavior in relation to devel- 

 opment? (3) What is the transition period from pelagic 

 to demersal life for an individual fish? (4) How do 

 light, temperature, salinity, and currents influence ver- 

 tical distribution and migratory behavior? (5) How does 

 gear selectivity bias vertical distribution patterns? 



Methods 



Sampling protocol 



General cruise objectives and sampling strategy for 

 the larval-juvenile fish studies are described by Lough 



( 1984). A concentration of larvae on Georges Bank was 

 located by a fine-scale grid of stations using standard 

 bongo-net gear. Based upon real-time sample analyses 

 made during the grid survey, sites were selected within 

 well-mixed and stratified waters for the comparative 

 vertical time-series observations. Haddock and cod 

 were collected by three gear types on eight cruises 

 (Table 1). Pelagic larvae and juveniles were sampled 

 with an electronically-controlled, multiple opening/ 

 closing net and environmental sensing system 

 (MOCNESS), with a lnr or 10 m 2 mouth opening (see 

 Wiebe et al. 1976, 1982, 1985). A Yankee 36 bottom 

 trawl was used to sample the recently-settled juvenile 

 fish. The numbers of day, night, and twilight hauls at 

 each site are listed in Table 1. 



The MOCNESS used to sample the larval stage has 

 an effective mouth opening of lm 2 when towed at a 

 45° net angle near 3.7km/h (2kn). The unit consists of 

 nine 0.333 mm mesh nets that open and close sequen- 

 tially on command. Temperature (and salinity begin- 

 ning in 1985) was obtained simultaneously with stan- 

 dard net-sampling parameters such as depth and 

 volume of water filtered. When a concentration of lar- 

 vae was located, a drogue was deployed at a depth of 

 15 m to mark the water mass for further sampling. 

 Thereafter, the lm 2 MOCNESS was used to sample 

 every 3-6 h at the drogue for a period of 24-48 h. Dis- 

 crete depths were sampled at 10 m intervals from the 

 surface to within 5 m of the bottom within a 5min 

 period. Each net filtered -250 m 3 of water. 



The 10 m 2 MOCNESS, used to collect the pelagic 

 juveniles, has five 3.0 mm mesh nets. Because juve- 

 niles tend to remain localized for at least 2-4 d, sam- 

 pling was conducted at fixed sites that were deter- 

 mined to contain high concentrations of juveniles. The 

 vertical distribution of pelagic juveniles was sampled 

 every 6 h. Discrete depth strata ranged 10-30 m from 

 the surface to within 5 m of the bottom; each net filtered 

 between 15,000 and 20,000 m 3 of water during a 30 min 

 tow. 



A Yankee 36 bottom trawl was used to catch re- 

 cently-settled juveniles. Trawl specifications are found 

 in Azarovitz (1981). The opening height and width of 

 the trawl was 3.2 X 10.4 m. The trawl had graded stretch 

 mesh of 127-13 mm (mouth to cod-end). Standard roller 

 gear (40.6cm diameter) was used on the Albatross IV 

 86-03 June 1986 cruise; however, rollers were replaced 

 with a sweep chain of rubber disks (11.4cm diameter) 

 on the Delaware II cruises 85-05 of July 1985 and 85- 

 06 of August 1985, to collect fish close to bottom. The 

 standard bottom-trawl set was made at 6.5km/h 

 (3.5 kn) for 30 min on the bottom. Trawl sampling was 

 conducted for 24 h. 



All specimens were initially fixed in either 4% for- 

 maldehyde-seawater solution or 90% ethyl alcohol and 



