Kotwicki et al : Effect of autotrawl systems on ttie performance of a survey trawl 



43 



been shown to be influenced by a variety of factors, 

 such as net speed through the water (Somerton 

 and Weinberg, 2001) that can increase footrope 

 distances off-bottom. Weinberg et al. (2002), using 

 a different survey bottom trawl, found capture 

 probability of some benthic species to decrease as 

 a result of footrope separations with the bottom. 

 These findings likely apply to the 83-112 Eastern 

 trawl as well. Because capture probability is gener- 

 ally size dependent (Munro and Somerton, 2002), 

 an unstable footrope may not only increase the 

 variance of overall biomass estimates, but may bias 

 the size distribution data generated by a survey 

 as well. If autotrawl systems can help maintain 

 footrope contact with the bottom, then the use of 

 an autotrawl during surveys may be warranted. 



In our experiment, the tension-controlled auto- 

 trawl system provided the best footrope contact 

 overall and the lowest standard deviation (Fig 3). 

 High heave and moderate to strong crabbing were 

 largely responsible for the differences observed in 

 the bottom-tending performance of the footrope 

 among treatments (Figs. 5-7). The tension treat- 

 ment was most effective in counteracting the ef- 

 fects of environmental conditions, whereas the 

 locked-winches treatment was the least stable of 

 the three treatments given the changing environ- 

 mental conditions. 



We found both autotrawl systems have a po- 

 tential to improve bottom trawl survey biomass 

 estimates by increasing the dynamic stability of 

 the trawl, thereby reducing the variance in its 

 catchability. The equal-tension system proved bet- 

 ter than the symmetry system in improving the 

 overall stability of the footrope bottom-tending 

 performance given the low bottom current ve- 

 locities observed. Symmetry autotrawl systems, 

 designed to react to crosscurrent conditions at 

 towing depth, could prove to be more effective 

 under stronger current conditions, but during our 

 experiment, current velocities may have been too 

 low (<35 cm/s) for us to be able to detect a differ- 

 ence between the symmetry and tension towing 

 modes. 



Bridles 



Flatfish are stimulated to herd by close proximity 

 to or actual contact with the lower bridle (Main 

 and Sangster, 1981a), whereas semipelagic species 

 such as Atlantic cod (Gadus morhua) are probably 

 stimulated to herd by the sight of the otter doors 

 and mud clouds created by the doors and lower bri- 

 dles (Main and Sangster, 1981b). For this reason, 

 the length of the lower bridle in contact with the 

 bottom and the frequency of this contact affect the 

 herding efficiency of the trawl (Somerton, 2003). If 

 the among- and within-tow variability of the bridle 

 bottom-tending distance could be reduced by using 

 an autotrawl system, then standardizing survey 



16 



12 



AL dS ot 



A. S 







10 



40 



60 



80 



100 



Heave (cm/s) 



Figure 8 



Comparison of regression lines illustrating the relationship 

 between mean bridle distance off-bottom and heave for locked 

 winches (L), symmetry (Si, and tension (T) treatments. Treat- 

 ments in which the heave had a statistically significant effect 

 on the mean are identified with an asterisk (*). 



E 



u 

 Q 



20 



40 60 



Heave (cm) 



80 



100 



1,5 2 



Crabbing (cm) 



Figure 9 



Comparison of regression lines illustrating the relationship 

 between standard deviation (SD) in bridle distance off-bottom 

 and environmental variables for locked winches (L), symme- 

 try (S), and tension (T) treatments. Treatments in which the 

 environmental variable had a statistically significant effect 

 on the SD are identified with an asterisk (*). 



