Proceedings of the Third World Fisheries Congress: Feeding the World with Fish in the Next Millenium—The Balance between Production and Environment

Multilevel Beam Trawl Reveals Fish and Prawn Behavior: Potential to Reduce Bycatch in the Australian Northern Prawn Fishery

Steve Eayrs


In recent years, worldwide concern has increased about the impacts of fishing on the marine environment. In particular, the practice of capturing and discarding large volumes of unwanted bycatch has come under scrutiny, and tropical prawn trawl fisheries are generally perceived as leading offenders in this regard. In Australia’s Northern Prawn Fishery (NPF), 30,000–60,000 t of bycatch are caught each year (Harris and Ward 1997), and less than 5% of bycatch is retained for commercial purposes (Pen-der et al. 1992). The bycatch is typically made up of small fish, crustaceans, sponges, sharks, and turtles. In some locations of the fishery, the bycatch-to-prawn ratio (by weight) may be as high as 20:1 (Pender et al. 1992; Brewer et al. 1998).

Since the early 1990s, fishers in the NPF have been under considerable pressure to reduce the capture of bycatch, particularly sea turtles—all six species caught in this fishery are listed as either endangered or vulnerable under Australia’s Endangered Species Protection Act of 1992. In addition, a U.S. embargo on prawn exports from fisheries not using approved turtle-excluder devices (TEDs; which became mandatory in the NPF in 2000, and the embargo was removed) has also contributed to the pressure felt by fishers. More recently, pressure has begun to mount over the incidental capture and dumping of large volumes of fish and other bycatch, including elasmobranchs, sea snakes, and signathids, and fishers are stepping up efforts to exclude this bycatch.

Since 1993, most attempts to reduce the amount of bycatch caught in the NPF have focused on the development of bycatch reduction devices (BRDs). Brewer et al. (1998) assessed the performance of 16 BRDs in this fishery between 1993 and 1996, including several TEDs (see Eayrs et al. 1997 for a detailed description of these devices), and McGilvray et al. (1999) tested the flexible AusT-ED with promising results. Other authors reporting on more recent attempts to reduce bycatch in this fishery include Day (1998), Campbell (1998), Day and Campbell (1999), and Robins et al. (2000). Without exception, all these authors noted the success of TEDs in excluding turtles, sharks, and other large animals from the trawl; exclusion rates approached 100%. In contrast, attempts to exclude fish and other bycatch usually have been less successful; reductions typically are no more than 20% in comparison to a standard trawl (Robins et al. 2000). The range of BRDs tested in this fishery thus far includes square mesh cod ends, square mesh windows, fisheyes, bigeyes, and radial escape sections. All these BRDs have been located in or close to the cod end, and all have been designed to exclude bycatch only after they have entered the trawl. No attempts have been made to prevent the ingress of these animals into the trawl mouth or to develop BRDs for this region of the trawl.

All BRDs rely to some extent on the behavior of bycatch within the trawl to facilitate their exclusion. However, despite this reliance, the behavior of prawns and bycatch in the trawl and during capture is not well understood. Attempts to gain this knowledge have been limited, and visual observations of these animals responding to the trawl are difficult to obtain because prawn trawling is typically a nocturnal activity carried out in turbid waters. Under these conditions, low-light cameras struggle to record useable images, and the use of lights to illuminate the trawl during filming is also limited due to the absorption of some wavelengths, backscattering effects, and the potential to influence animal behavior. The ability to develop more selective prawn trawls for this fishery is therefore hampered by a lack of behavioral knowledge and an inability to observe the behavior of prawns and bycatch in response to a trawl. To overcome these limitations, a multilevel beam trawl (MBT) was tested in the NPF, and for the first time in a tropical prawn fishery, the vertical distribution and behavior of these animals were quantified.