Lough and Potter Vertical distribution of Melanogrammus aeglefinus and Oadus morhua 



289 



Figure 1 1 



Time-series of weighted-mean depths of (A) larval haddock Melanogrammus 

 aeglefinus and (B) cod Gadus morhua size-classes on weakly-stratified Site 83-1 

 during the period 13-14 May 1983. In the haddock time-series, boundary tem- 

 peratures are recorded beside the brackets; for cod, the rate of temperature 

 change (8T/8ZI over the thermocline region is specified. 



cially after reaching 9-13 mm. Day 

 abundances of haddock are greater 

 than night for the 6-8 mm and 9- 

 13 mm size-classes, which may indi- 

 cate a different avoidance behavior at 

 this size. 



The 10 m 2 MOCNESS night-day 

 abundance ratios (zz/lOOOm 2 ) are 

 shown in Table 9 for size-classes of 

 cod and haddock from study sites in 

 June 1984, 1986, and July 1985. A 

 greater size-range of cod is available 

 than for haddock. For cod, the small- 

 est (9-3 mm) and the largest (50- 

 59 mm) size-classes are not fully vul- 

 nerable to the net, due to extrusion 

 of the smallest larvae through the 

 net mesh and distribution of the larg- 

 est juveniles below the sampling 

 depth. The total night-day ratios for 

 cod are -2-3 (range 1.49-3.05) for 

 size-classes 14-19 mm through 40- 

 49 mm. Haddock juveniles have 

 night-day ratios of -3-4 for the 14- 

 19 mm and 20-29 mm size-classes. 



Sampler retention and avoidance 



The 1 m 2 MOCNESS sampled effectively on a hori- 

 zontal scale of 0.2-0.3 km; the 10 m 2 MOCNESS, 1- 

 2 km; and the bottom trawl, 3-4 km. Minimum length 

 of fish fully retained and not extruded by the vari- 

 ous nets was estimated using a model based on the 

 stretch mesh size and a body height-to-length re- 

 gression (Potter et al. 1990). These estimates indi- 

 cated that almost all haddock and cod larvae 

 >~4mmSL are retained by the 1 m 2 MOCNESS 

 0.333 mm mesh nets. For the 10 m 2 MOCNESS, 

 3.0 mm mesh nets, haddock <17mm and cod <15- 

 16 mm probably are not fully retained. Minimum size 

 retention of juvenile fish collected by the research 

 bottom trawl is 31-37 mm. 



The lm- MOCNESS night-day abundance ratios 

 (n/10m 2 ) are shown in Table 8, page 307 for size- 

 classes of cod and haddock collected in April and May 

 1981 and May 1983. The total night-day ratios for 

 cod are -1.2 for the 2-5 mm and 6-8 mm size-classes, 

 near 2.0 for the 9-13 mm size-class, and to -4.0-6.0 

 for the 14-19 mm and 20-29 mm size-classes (Fig. 19). 

 An exponential curve fit appears to adequately de- 

 scribe the increase of night-to-day catch ratios with 

 increasing size of fish. Cod appear to visually avoid 

 the 1 m 2 MOCNESS net by day at all lengths, espe- 



Discussion 



The difficulty of detecting vertical migration in plank- 

 tonic organisms, given the high sampling variability 

 and limited resources for replicate time-series of ob- 

 servations as well as gear limitations, has been dis- 

 cussed by Pearre (1979), Shulenberger (1978), Denman 

 & Piatt (1978), and others. We are forced to look for 

 repeatable patterns, often in the absence of statistical 

 proof. Because of these difficulties, diverse observa- 

 tions have been reported in the literature. In this study, 

 the vertical distribution patterns and diel migrations 

 of haddock and cod larvae and juveniles have been 

 documented in stratified and well-mixed waters of 

 Georges Bank in greater detail than previously seen 

 because of electronic MOCNESS samplers and the use 

 of submersibles. 



Sampler avoidance 



Many investigators have reported night-day catch ra- 

 tios >1. Net avoidance usually is greater for day than 

 night tows because of visual detection of the net (Morse 

 1989). In another Georges Bank study using the 10 m 2 

 MOCNESS, Potter et al. (1990) reported night catches 

 of pelagic cod juveniles averaging 30 mm to be signifi- 



