Charis Y. Burridge

An assessment of Bycatch Reduction Devices in a tropical Australian prawn trawl fishery

Our study assessed 16 different Bycatch Reduction Devices (BRDs) for use in Australias Northern Prawn Fishery (NPF), during three cruises in the Gulf of Carpentaria. The study was a response to pressure from conservation and community groups to make harvesting of the sea more selective and less damaging to the non-target communities. The approach was to incrementally improve the performance of BRDs after each stage: underwater observations in a flume tank and engineering trials at sea; rigorous and controlled scientific trials on the RV Southern Surveyor; testing under normal commercial fishing conditions. Bycatch Reduction Devices were improved throughout these procedures; only the best-performing devices were tested in the next stage. All four inclined grids (Super Shooter, Nordmore grid, AusTED and NAFTED) were extremely effective at excluding large sharks, stingrays and sea turtles. They were also effective at excluding some of the unwanted small-fish catch, especially when used in combination with other BRDs, such as a fisheye or square-mesh window. Fish exclusion ranged between 0 and 39%. Exclusion of sea snakes was substantially increased by using BRDs, such as the AusTED or square-mesh windows. The ability of BRDs to maintain the catch of the target species was part of the assessment. The Super Shooter had the best prawn retention rate, losing only between 2 and 12% of prawns in scientific trials and gaining 3.4% during the commercial trials. The Super Shooter also performs well in areas where the other inclined grids tend to clog with sponges or other large objects. The NAFTED, which was used only during the commercial trials, lost only 3.3% of the prawn catch of a standard trawl. There is evidence that the fisheye may increase prawn catches. Other devices lost substantial numbers of prawns, but modifications could greatly improve their performance. Some of these devices could be used successfully in their present form. The performance of most BRDs were also strongly affected by the weather and fishing procedures, such as delay in haulback of the trawl. This study has shown that there are BRDs that can be of major benefit to prawn trawl fishers in the NPF, by excluding much of the unwanted bycatch while maintaining catches of commercially valuable prawns. The main benefits to the NPF fishery of using BRDs are: • fewer or no sea turtles caught or killed; • fewer small fish to be sorted from catches; • fewer large animals (stingrays, sharks and sea turtles) that can damage valuable prawns, resulting in a higher catch value; and • decreased fishing impacts on bycatch species, which helps to maintain ecological biodiversity and resilience of these demersal communities. Failure of the NPF and other Australian prawn trawl fisheries to voluntarily adopt BRDs may result in an involuntary adoption brought on by pressures from community, conservation and trade bodies.

The effectiveness of fish and shrimp trawls for sampling fish communities in tropical Australia

Catches of fish in a Frank and Bryce fish trawl during the day and in a Florida Flyer shrimp trawl at night were compared over 2 years (1992, 1993). The effectiveness of a Frank and Bryce demersal trawl as a sampling tool for demersal fish communities was assessed from 122 stations in the far northern section of the Great Barrier Reef. Day-time and night-time catches of the Frank and Bryce net were compared to identify diurnal and nocturnal species. Of a total of 259 species caught, 138 were caught both during the day and at night. An analysis of variance was used to identify species that were caught exclusively or in greater biomass during the day or night; 17 species (mainly small apogonids and mullids) were significantly nocturnal and one (a leiognathid) was significantly diurnal. The overall catch rate of the fish trawl during the day was 395.8 kg h−1 (s.e. 141.3) and that of the shrimp trawl at night was 28.0 kg h−1 (s.e. 1.7). A total of 340 species was caught by both types of net—the fish trawler caught 236 species, the shrimp trawler 243, and 141 were common to both trawl types. An analysis of variance was used to help identify species that were caught in greater biomass by either trawl. Fifty-eight fish species each had a greater biomass in the fish trawl than in the shrimp trawls, for 60 species it was the reverse and for 32 species there was no difference. The fish caught only by the fish trawl were mainly pelagic species (mackerels, carangids) or specimens of large species (lutjanids, sharks and rays), while the fish caught only by the shrimp trawl were small benthic species (apogonids, platycephalids, scorpaenids and flatfish). Over 80% of the fish caught by both nets were small (< 30 cm SL). The length-frequency distributions of only seven species differed significantly between fish and shrimp trawl catches, with the shrimp trawl generally catching the smaller component. No large commercially or recreationally important species of reef fish were caught by the shrimp trawl. The results of this study indicate that each type of trawl net will catch different species and, where they catch the same species, they catch different numbers, sizes and weights. Neither gear type can be used on its own for an adequate description of the fish community.

A validation of short-duration scientific tows as a representation of long commercial-length tows: comparing the catch rates, size composition and species composition of prawn trawler bycatch in the far northern Great Barrier Reef, Australia

The duration of tows for scientific trawl surveys in northern Australia is generally 30 min. Commercial trawls may be up to 200 min long. In this study we compare the catch rates, size of fish and species composition for 10 short (30 min) tows paired with 10 commercial length (165 min—not in a straight line) tows at both an inshore and an offshore location in the far northern Great Barrier Reef. Evaluations of this kind have not been reported in tropical Australian multi-species fisheries. Overall, catch rates of fish and invertebrates differed between short and long tows both inshore and offshore. Inshore, the mean catch rate for the fish and the invertebrates was significantly greater in short duration tows (P<0.05). Offshore, the mean catch rate for all fish, and all invertebrates was not significantly different between short and long tows, although higher catch rates of invertebrates were recorded in long tows. Inshore, 12 fish species out of 49 had significantly different catch rates, most being greater in short tows. Only 2 invertebrates out of 16 had significantly different catch rates (P<0.05). The combined weight of species showing significant differences in catch rates represented less than 10% of the total weight in both inshore and offshore samples. Principal component analysis indicated that catch composition was similar between short and long tows both inshore and offshore. The results from our sampling are that the species composition of 30-min trawls is similar to that of commercial length tows, and that size frequencies are equivalent. The implications are that short tows (30 min) can be used in scientific surveys to give a true representation of both size and species composition of commercial length tows, but may overestimate catch rates.

Benefits of statistical blocking techniques in the design of gear evaluation trials: introducing the Latin Square design

Abstract We evaluated the use of a Latin Square design for assessing the performance of bycatch reduction devices (BRDs). We wanted to ensure the most precise result possible given limited resources, high variability and the need to detect small differences. The Latin Square design has two features to reduce variation: removing the tow-to-tow variation by deploying the experimental and standard nets simultaneously; and removing the different efficiencies of the nets by swapping codends and averaging consecutive pairs of tows on the same net. The design, which was relatively easy to implement, was used for evaluating the AusTED II in three locations in northern Australia. We reviewed a number of similar studies to see what efficiency they achieved with simpler statistical designs and found that trials based on less than 100 tows had low power (at most 31%) to detect a 5% fall in prawn catch. The Latin Square has the potential to produce more powerful results with some extra effort in swapping codends between nets. The best of our trials (60% power to detect a 5% fall in prawn catch) out-performed similar-sized trials. We recommend this design for gear trials in variable environments, if the codends can be swapped quickly and especially if the trial involves hundreds of tows. We also demonstrate the importance of assessing the statistical power of such trials to avoid over-interpretation of non-significant differences between fishing gear.