Abstract.— Two experiments were 

 done in a flume tank to quantify the 

 effects of codend mesh circumference 

 and weight of catch on water flow. The 

 mesh circumference of the posterior 

 section of prawn-trawl codends (with 

 and without bycatch reducing square- 

 mesh panels) were assessed with three 

 weights of catch (30 kg, 50 kg, and 70 

 kg). Compared with a codend with a cir- 

 cumference of 100 meshes throughout 

 its entire length, a codend with an an- 

 terior section of 100 meshes and a pos- 

 terior section of 200 meshes in circum- 

 ference significantly increased the dis- 

 placement of water forwards (up to 

 1120 mm from the end of the codend). 

 This result varied with the weight of 

 catch in the codend. The second experi- 

 ment involved placing composite pan- 

 els of square-shaped mesh (bycatch re- 

 ducing devices) into the tops of the an- 

 terior sections of two codends with the 

 same configurations as those above 

 (termed the "100 panel codend" and 

 "200 panel codend," respectively). There 

 was a displacement of water forwards 

 immediately under the square-mesh 

 panel in the 200 panel codend (by up to 

 2200 mm from the end of the codend). 

 The results are discussed in terms of 

 1) the probable effects that codend 

 mesh circumference and water dis- 

 placement in codends have on fish be- 

 havior, and 2) implications for the fu- 

 ture development of bycatch reducing 

 devices like square-mesh panels in 

 prawn-trawls. 



Flow-related effects In prawn-trawl codends: 

 potential for increasing the escape of 

 unwanted fish through square-mesh panels 



Matt K. Broadhurst 



N.S.W, Fisheries Research Institute 



P.O. Box 21 Cronulla, New South Wales 2230, Australia 



Present address: Universidade Federal Rural de Pernambuco-UFRPE 



Departamento de Pesca, Laboratono de Oceanografia Pesqueira 



Av. Dom Manuel de Medeiros, s/n 



Dols Irmaos, Recife-PE, Brasil, CEP; 52 171-900 

 E-mail address: fhvhazin@truenet-Com.br 



Steven J. Kennelly 



N.S.W. Fisheries Research Institute 



PO. Box 21 Cronulla, New South Wales 2230, Australia 



Steve Eayrs 



Australian Maritime College 



PO. Box 21 Beaconsfield, Tasmania, 7270, Australia 



Manuscript accepted 30 April 1998. 

 Fish. Bull. 97:1-8(1999). 



Oceanic prawn-trawling occurs 

 from nine major ports in New South 

 Wales (NSW) Australia and has a 

 total value of approximately A$17 

 million per annum. The principal tar- 

 get species is the eastern king prawn 

 iPenaeus plebejus), although a signifi- 

 cant proportion of the total value is 

 derived from the sale of legally re- 

 tained bycatch comprising individu- 

 als of several species of fish, crusta- 

 ceans, and cephalopods (see Kennelly, 

 1995). In addition to this landed 

 catch, however, significant numbers 

 of nontarget organisms of no commer- 

 cial value are also captured and dis- 

 carded, including juvenile fish of spe- 

 cies which, when larger, are targeted 

 in other commercial and recreational 

 fisheries (Kennelly, 1995). 



Concerns over the incidental cap- 

 ture and mortality of large numbers 

 of juvenile fish have led to the de- 

 velopment of various bycatch reduc- 

 ing devices (BRDs), designed to 

 minimize undesirable bycatch while 

 maintaining catches of prawns and 

 other commercially valuable indi- 

 viduals (Broadhurst et al., 1996; 



Broadhurst and Kennelly, 1996; 

 1997). In particular, a new design 

 comprising composite panels of 

 square-shaped mesh (referred to as 

 the composite square-mesh panel) 

 was shown to increase catches of 

 prawns (by up to 14% ) (Broadhurst 

 and Kennelly, 1997) while signifi- 

 cantly reducing up to 40% of total 

 unwanted bycatch and up to 70% of 

 the numbers of small individuals of 

 commercially important species 

 such as whiting iSillago spp.) 

 (Broadhurst and Kennelly, 1996). 



The results from these experi- 

 ments are attributed primarily to 

 differences in the behavior of fish 

 and prawns in the trawl. Fish are 

 believed to be herded close together 

 in the anterior section of the codend, 

 upsetting the normal balance of the 

 school and initiating an escape re- 

 sponse towards the sides and top of 

 the net and out through the open 

 square-shaped meshes (Broadhurst 

 and Kennelly, 1996). A contributing 

 factor towards this escape is the dis- 

 placement and change in direction 

 of water flow due to the circumfer- 



