Action

Modify the design or configuration of trawl gear (mixed measures)

How is the evidence assessed?
  • Effectiveness
    not assessed
  • Certainty
    not assessed
  • Harms
    not assessed

Study locations

Key messages

COMMUNITY RESPONSE (0 STUDIES)

POPULATION RESPONSE (0 STUDIES)

BEHAVIOUR (0 STUDIES)

OTHER (19 STUDIES)

About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

  1. A replicated study in 1989 in an area of seabed in the South Pacific Ocean off New South Wales, Australia (Andrew et al. 1991) found that modifying the configuration (wire length and single or triple trawls) of four prawn trawl nets resulted in mixed effects on the reduction of non-target catches of red spot whiting Sillago bassensis and sand flathead Platycephalus caeruleopuncta. Overall catch rates of red spot whiting and sand flathead were similar for three of the four configurations (whiting, single/7 m: 2.4, single/40 m: 3.9, triple: 4.3 fish/ha; flathead, single/7 m: 1.7, single/40 m: 3.0, triple: 2.0 fish/ha) and higher in single trawls with 140 m wires/bridles (whiting: 13.8 fish/ha, flathead: 6.1 fish/ha). However, average fish length was greater in single trawls with 140 m bridles for both species (red spot whiting: 17.9 cm, sand flathead: 33.5 cm) compared to any other single trawls (whiting: 17.3–17.4 cm, flathead: 31.9–32.5 cm) and the triple trawl (whiting: 16.6 cm, flathead: 32.4 cm). In addition, the total weight of non-target catch (fish and invertebrates combined) was not statistically different between single- and triple net trawls (single, 7 m: 15 kg/ha, 40 m: 19 kg/ha, 140 m: 23 kg/ha; triple: 13 kg/ha), as were catches of target prawns Penaidae and shovelnose lobster Ibacus spp. (see paper for data). Replicate trawl deployments (30 min, 2.8 knots) were conducted with four trawl designs (single trawls with 7 m, 40 m or 140 m bridles, or triple rigged trawls) in December 1989 in 35–40 m depth. Full details of trawl designs are provided in the original study.

    Study and other actions tested
  2. A replicated, controlled study in 1986–1988 on bottom fishing grounds in the North Sea, UK (Reeves et al. 1992) found that modifying the design of bottom trawls and seine nets (changing length of extension piece) resulted in an increase in size-selectivity with shorter extension pieces for haddock Melanogrammus aeglefinus, whiting Merlangius merlangus and cod Gadus morhua. For both gear types, extension length affected size-selectivity and increased with decreasing length (data reported as statistical results). Overall, the estimated lengths at which fish had a 50% chance of escape with the shortest extension piece (0 m) ranged from 15–32 cm for haddock, 20–35 cm for whiting and 16–36 cm for cod. For the longest extension length (13.7 m) these were decreased to 11–28 cm for haddock, 15–30 cm for whiting and 12–32 cm for cod. In addition, gear size-selectivity increased with increasing mesh size and narrower diameters of the codend, and their effect was greater than the effect of extension length. Data were collected during three surveys on two commercial vessels between 1986–1988 in the northern or central North Sea. Fishing was conducted with two seine nets and one trawl net configured with three different extension lengths (0, 9.1 and 13.7 m), mesh sizes (80, 90 and 100 mm) and codend diameters (2.2, 3.2 and 4 m). For seine nets, 114 hauls were completed with a 9.1 m extension, 104 hauls with a 13.7 m extension and 110 hauls with a 0 m extension. For the trawls, 37, 35 and 35 hauls were completed for the 9.1, 13.7 and 0 m extensions, respectively. Small mesh covers attached over the codends sampled fish escaping through the meshes. All fish in the codends and covers were identified, and length recorded. No cod were caught in trawl net deployments.

    Study and other actions tested
  3. A replicated, controlled study in 1994 of an area of seabed in the Mediterranean Sea, off Turkey (Lök et al. 1997) found that two different trawl codend designs increased the size selectivity of red mullet Mullus barbatus compared to a standard codend, but there were no differences between codend types for annular seabream Diplodus annularis. The length at which red mullet had a 50% chance of escape was higher in both a ‘shortened lastridge rope’ codend (15.1cm) and a narrowed circumference codend (14.3 cm) compared to the standard (13.7 cm). For annular seabream, the 50% selection length was similar between all codends (short rope: 9.8 cm, narrow: 10.1 cm, standard: 9.9 cm). Data were collected in June and September 1994 in the Aegean Sea, from 40 trawl deployments (40–100 m depth, 50–60 min) of three different codend types: a roped codend rigged onto shortened ropes along each seam (14 hauls), a codend with the circumference reduced to 120 from 150 meshes (12 hauls), and a standard codend of 44 mm diamond mesh and 150 mesh circumference (see original paper for gear specifications). Small mesh covers attached over the codends sampled escaped fish. Codend and cover catches were sampled, and fish (fork) length recorded.

    Study and other actions tested
  4. A replicated, paired, controlled study of an area of seabed in the Pacific Ocean off the coast of Oregon, USA (Hannah & Jones 2003) found that modifications to the configuration of a shrimp trawl (to alter the height of the fishing line/footrope) resulted in reduced catches of unwanted small flatfish (Pleuronectiformes) and immature rockfish Sebastes spp. with increasing height of the fishing line. Catches of flatfish and rockfish between trawls with the same or higher fishing line height relative to the standard, were lower for two of two modified configurations (higher: 2–64 fish, standard: 4–81 fish). Conversely, catch numbers of both flatfish and rockfish increased in two of three trawl configurations with lower fishing line heights relative to the standard height (lower: 31–147 fish, standard: 2–50 fish), but were not significantly different in the other (lower height: 4–285 fish, standard height: 5–156 fish). Data were collected from 26 paired trawl deployments on a twin-rigged (dual net) commercial shrimp vessel fishing out of Newport. One side of the trawl was fished with a ‘standard’ configuration in which the central ‘drop’ chains between the fishing line and groundline were shortened to 51 cm. Four different configurations of drop chains were tested against the standard, each either increasing or decreasing the height of the fishing line (see original paper for full gear specifications). Five sets of comparative hauls were carried out (4–6 of each comparison). Codend catches were weighed and counted by species. The study does not report when the sampling took place.

    Study and other actions tested
  5. A replicated, controlled study in 2002 of seabed in a coastal bay in the Mediterranean Sea, off Turkey (Özbilgin & Tosunoğlu 2003) found that modifying the design of a bottom trawl net (single or double layer codends) resulted in improved size-selectivity and reduced catches of smaller fish in single codends for red mullet Mullus barbatus, annular sea bream Diplodus annularis, and common pandora Pagellus erythrinus, compared to a double layer codend. The length at which fish had a 50% chance of escape was greater in single layer codends than double codends for red mullet (single: 10 cm, double: 9 cm), annular sea bream (single: 9 cm, double: 8 cm) and common pandora (single: 11, double: 8 cm). The total number of fish that escaped capture was higher with a single codend for all three species (red mullet, single: 1,928, double: 599 fish; annular sea bream, single: 304, double: 53 fish; common pandora, single: 381, double: 82 fish). In April 2002, bottom trawl deployments were carried out in Izmir Bay in the eastern Aegean Sea; nine with codends of a single layer of netting (one codend) and nine with double layer codends (one codend mounted around another, see original paper for gear specifications). Gear was towed for 45 minutes at 2.2–2.6 knots and 25–30 m depth. Codends had 200 mesh circumferences and 40 mm mesh size. Covers attached over each codend type collected escaped fish. Both codend and cover catches were sampled, and fish lengths recorded.

    Study and other actions tested
  6. A replicated, controlled study in 2001 in an area of seabed in the North Sea, off the Orkney Islands, UK (Kynoch et al. 2004) found that modifying a bottom trawl by removing the strengthening bag (a large-mesh cover to prevent the codend from splitting when catch is heavy) improved the size-selectivity of haddock Melanogrammus aeglefinus at two mesh sizes. The length at which haddock had a 50% chance of escape was higher without a strengthening bag at both 110 mm codend mesh size (without: 31.4 cm, with: 31.4 cm) and 120 mm mesh codend (without: 34.3 cm, with: 32.4 cm). A total of 26 trawl deployments were completed on a commercial fishing vessel in June 2001: seven each with a 110 mm codend, with and without a strengthening bag, and six each with a 120 mm codend, with and without a strengthening bag (all diamond mesh). Hauls were 120–198 minutes at 68–87 m depth. The strengthening bag used 265 mm diamond mesh and 6 mm diameter twine (see original paper for full gear specifications). A cover attached over the codends sampled fish that escaped through the meshes. Codend and cover catches were sampled, and fish lengths recorded.

    Study and other actions tested
  7. A replicated, controlled study in 1996–1997 in two areas of seabed in the Skagerrak and Baltic Sea around Scandinavia (Madsen et al. 2006) found that modifying the design of a flatfish trawl reduced the unwanted catch of Atlantic cod Gadus morhua, compared to a standard design. Across all trails, total catch numbers of all cod and undersized (<40 cm) cod were reduced in the modified trawl design compared to conventional trawls, by 15–75% in four of four cases, and by 50–80% in three of four cases (the fourth case showed an increase of 24%), respectively. In addition, there were no differences in catch numbers of undersized individuals of the target flatfish species plaice Pleuronectes platessa and flounder Platichthys flesus. However, catches above the minimum landing sizes were higher in modified trawls in one of one plaice (>27 cm: 15%) and one of three flounder (>25 cm: 46%) comparisons. Experimental trials were carried out in June 1996 in the Skagerrak and the North Sea and in December 1996 and January 1997 in the Baltic Sea. Trials compared two modified trawl designs to conventional trawls targeting plaice (Skagerrak/North Sea) and flounder (Baltic Sea). Modifications included a triangle of large mesh (400 mm) at the trawl opening, reduced flotation to keep a low vertical opening, a long headline to increase the seabed area swept by the trawl, and different configurations of square mesh panels/windows (see original paper for full gear specifications).

    Study and other actions tested
  8. A replicated, paired, controlled study in 2005–2006 in an area of seabed in the northwest Atlantic Ocean, USA (Beutel et al. 2008) found that using a modified design of bottom fish trawl (large mesh in the front sections) reduced the catches of most non-target fish species, compared to a conventional trawl. Total catch weights were lower in the modified trawl compared to the conventional trawl for 15 of 19 non-target fish species, including cod Gadus morhua (the main unwanted species) and several other commercial bottom species with stock levels of concern (see original paper for species individual data). Total catch weights of haddock Melanogrammus aeglefinus (the target species) were similar between trawl types (modified: 12,580, standard: 14,327 kg), and the ratio of haddock by weight to that of four key non-target fish species was increased in modified trawls from less <1 haddock/non-target species to 151 haddock/non-target species. Data were collected from 100 parallel trawl deployments on two vessels during four fishing trials from June 2005 to April 2006, in and around a closed area on the Georges Bank (37–154 m depth). Vessels towed side-by-side, one using a modified trawl design constructed with large mesh (240 cm) front sections (‘Eliminator TrawlTM’), and the other a conventional trawl. Full details of the trawl designs are provided in the original study.

    Study and other actions tested
  9. A replicated, controlled study in 2007 of an area of seabed in an estuary flowing into the Tasman Sea, Australia (Broadhurst et al. 2009) found that modifying a prawn trawl by separating the codend into two compartments (double codend) did not reduce the catches of discarded finfish, compared to a conventional single codend. Average catch weights of discarded finfish species were similar for double and single codend trawls (double: 0.9 kg, single: 1.0 kg). In addition, average retained and discarded weights of the target school prawn Metapenaeus macleayi were both similar between codend types (double: 0.7–9.3 kg, single: 0.8–9.5 kg). In October 2007, a total of 24 trawl deployments (1 h) were done on a prawn fishing ground in Lake Wooloweyah, Clarence River estuary, New South Wales. Two codend types were tested: one modified with two compartments (12 hauls), and one with a single codend (12 hauls). Both codend types were square mesh (27 mm) and had a size-sorting escape grid (see original paper for gear specifications).

    Study and other actions tested
  10. A replicated, paired, controlled study in 2010 in an area of seabed in the North Pacific Ocean, off Oregon, USA (Hannah et al. 2011) found that modifications to the footrope of prawn trawl gear reduced the capture of unwanted eulachon Thaleichthys pacificus, and other unwanted finfish species, compared to a conventional trawl design. The average weight of unwanted catch was reduced in the modified trawl compared to the conventional trawl, by 34% for eulachon (modified: 1,891, standard: 2,858 g/haul), by 96% for slender sole Lyopsetta exilis (modified: <1, standard: 4 kg/haul), by 97% for other small flatfish (modified: <1, standard: 2 kg/haul) and by 80% for darkblotched rockfish Sebastes crameri (modified: 13, standard: 61 g/haul). However, there were no differences between gear types for whitebait smelt Allosmerus elongatus (modified: 18, standard: 29 g/haul) and Pacific herring Clupea pallasii (modified: 1, standard: 1 kg/haul). In addition, target ocean shrimp Pandalus jordani catches were not statistically different between gear types (see paper for data). Experimental trials were conducted from 26 paired deployments on a double-rigged shrimp trawler in June 2010 (99–148 m depth, 1.6–1.8 knots, 45–60 min). Two trawl types were tested simultaneously: a modified trawl with the central section of the groundline removed and drop chains attached to the central section to help stabilise it, and a standard trawl with a complete groundline (see original paper for gear specifications). Both trawl types had a rigid size-sorting escape grid (19 mm bar spacing).

    Study and other actions tested
  11. A replicated, paired, controlled study in 2009 of two areas of seabed in the North Sea off the Shetland Islands, UK (Kynoch et al. 2011) found that modifying the forward sections of a bottom trawl net (increases in mesh sizes) resulted in reduced catch rates of all sizes of four of seven commercial fish species, increased catch rates of one species in one of two cases, and for the other two species the effect varied with fish size, compared to standard forward sections. Overall, increased mesh sizes of 300 mm and 600 mm caught fewer Atlantic cod Gadus morhua (by 49% and 75%), megrim Lepidorhombus whiffiagonis (by 79% and 93%), ling Molva molva (by 36% and 68%) and hake Merluccius merluccius (by 28% and 53%) at all lengths, relative to standard mesh sizes of the forward sections (120/160 mm). Relative catch of haddock Melanogrammus aeglefinus was similar with the 600 mm gear (5% difference) but higher by 42% with the 300 mm gear compared to the standard. The 300 mm and 600 mm gears caught fewer monkfish Lophius piscatorius below 76 cm and 83 cm respectively (and similar catch of fish above these sizes) and saithe Pollachius virens above 53 cm (data reported as statistical results). Trials were conducted on two Shetland fishing grounds 30 nautical miles apart (104–150 m and 163–185 m depths), by a twin-rig trawler in June and July 2009. The 600 mm mesh was deployed only in one area (14 hauls) and the 300 mesh was trialled in both areas (30 hauls in total). Both modified trawl types were fished alongside a standard bottom trawl (see original paper for gear specifications).

    Study and other actions tested
  12. A replicated, paired study in 2011–2012 of sandy mud bottom in an river/estuary flowing into the Tasman Sea, Australia (Broadhurst et al. 2012) found that modifying the length and number of panels in the body of a prawn trawl net reduced the unwanted catch of one of four main fish species caught in shorter compared to longer trawls. Average catch rates of southern herring Herklotsichthys castelnaui were lower in shorter trawl designs than longer designs (short: 4 fish/ha, long: 1–2 fish/ha), irrespective of panel number. But there were no differences between trawl types for Ramsey's perchlet Ambassis marianus, narrow banded sole Synclidopus macleayanus or yellowfin bream Acanthopagrus australis (short: 1–7 fish/ha, long: 1–6 fish/ha). Numbers of target school prawns Metapenaeus macleayi were lower in shorter trawls (short: 1,107 ind/ha, long: 2,247 ind/ha). Sampling was conducted in December 2011 and January 2012 in the Clarence River, New South Wales, using a local prawn trawler. Four designs of trawl were tested (see original paper for gear specifications). All were identical except for their body length/side taper (short 35°, or long 25°) and the number of panels (two or four). Twenty paired deployments with each trawl design were completed. Codend catches were sorted and weighed by species. Data on the main target and non-target species were analysed.

    Study and other actions tested
  13. A replicated study in 2012 of sandy mud bottom in an estuary flowing into the Tasman Sea, Australia (Broadhurst et al. 2013) found that changes to the configuration (number of nets, one to four) of prawn trawl gear resulted in lower overall catches of non-target fish species in triple trawls, and fewer of one of six of the main non-target individual fish species in multi-rigged trawls compared to a single trawl net. Average weight, but not number, of all non-target catch (more than 95% of which was just six fish species) differed between trawl configurations and was lower only in triple-rigged trawls (0.5 kg/ha) compared to single-rigged trawls (1.0 kg/ha), but similar for all other comparisons between trawl types (double-rigged: 0.7 kg/ha, quadruple-rigged: 0.7 kg/ha). Individually for the main six fish species caught, the catch rate of only yellowfin bream Acanthopagurus australis differed between trawl designs and was lower in double-rigged, triple-rigged and quadruple-rigged (2 ind/ha) trawls than in single trawls (4 ind/ha). But there were no differences between trawl designs in the catch rates of five other unwanted fish species, fork-tail catfish Arius graeffei, narrow-banded sole Synclidopus macleayanus, bullrout Notesthes robusta, silver biddy Gerres subfasciatus and mulloway Argyrosomus japonicus (see original paper for species individual data by gear type). Catch rates of the target school prawn Metapenaeus macleayi were similar across trawl configurations, (single: 600, double: 750, triple: 900, quadruple: 1,450 ind/ha), although single-rigged trawls retained larger individuals. In March and May 2012, a total of 36 experimental deployments/trawl configuration were made in the Clarence River estuary (3–18 m depth) of four trawl configurations: single, double, triple or quadruple-rigged. All used 45 mm mesh, but each configuration had different technical specifications (see original paper for details).

    Study and other actions tested
  14. A replicated, controlled study in 2013 of an area of sand and mud bottom in an estuary off the Tasman Sea, Australia (McHugh et al. 2014) found that modifying the configuration (spreading mechanism) of prawn trawls resulted in reduced overall catches of unwanted fish in one of four configurations. Average catch rate of unwanted fish by number was lower in otter trawls without sweep wires (62 fish/40 min) than with (96 fish/40 min), and was similar to beam trawls both with and without a horizontal wire across the trawl mouth (with: 59 fish/40 min, without: 73 fish/40 min). Catches of target school prawn Metapenaeus macleayi were similar in otter trawls with and without sweep wires (with: 2,600, without: 2,100 ind/40 min), but lower in beam trawls with a horizontal wire (1,200 ind/40 min) than without (1,600 ind/40 min). In summer (austral) 2013, a total of 36 trawl deployments in Lake Wooloweyah estuary in New South Wales (at 1–2 m depth) were done with each of four trawl designs: an otter trawl, with and without sweep bridles, and a beam trawl with and without a horizontal wire. Sweep wires were removed to test reduction in the herding of fish into the net and the horizontal wire was designed to produce a fish escape response. Full details of the trawl specifications are provided in the original study.

    Study and other actions tested
  15. A replicated, paired study in 2014 of an area of shallow sandy mud bottom in an estuary lagoon site in off the Tasman Sea, Australia (Broadhurst et al. 2015) found that modifying the design of prawn trawls (body length/side taper) reduced the unwanted catch of only one of seven main fish species caught, and for most of the rest, catches varied with light conditions (day/night). Catch rate of only one of seven unwanted fish species, Australian anchovy Engraulis australis, was lower in a short trawl (0–2 fish/ha) than a long trawl (1–3 fish/ha), irrespective of day or night sampling times, and southern herring Herklotsichthys castelnaui catch rate was lower in the short trawl during daylight only (short: 48 fish/ha, long: 68 fish/ha). However, regardless of trawl type, catch rates of forktail catfish Arius graeffei, pinkbreast siphonfish Siphamia roseigaster, and yellowfin bream Acanthopagrus australis, were lower during the day than night, whereas whitebait Hyperlophus vittatus catches were lower during the night. Neither trawl type nor sampling time affected catch rates of Ramsey's perchlet Ambassis marianus (see original paper for species individual data by trawl type/sampling time). In addition, the catch rates (weight and numbers) of the target species school prawns Metapenaeus macleayi were lower in the short trawl than the long trawl (short: 7 kg/ha, long: 9 kg/ha). In March and April 2014, a total of 44 paired deployments (45 min) were made in Lake Wooloweyah, Clarence River estuary, New South Wales, with one short and one long prawn trawl, during six days and four nights. Trawls nets were identical apart from different configurations of wing and body tapers (see original paper for gear specifications).

    Study and other actions tested
  16. A replicated, paired, controlled study in 2013–2014 of an area of sand and mud bottom in the Clarence River estuary, Tasman Sea, Australia (Broadhurst et al. 2015) found that modifying a prawn trawl design (wing mesh orientation and hanging ratio) typically reduced the overall catches of unwanted fish by number, but not by weight, compared to a conventional design. Overall catch numbers of unwanted fish were reduced in three of the four modified trawls compared to the conventional trawl (diamond mesh wings, both loose and tight: 35 fish/ha, loose square mesh: 37 fish/ha, conventional: 55 fish/ha), but were similar in the tight square mesh trawl (40 fish/ha). By weight, catch rate of all non-target fish species was reduced only with loose diamond wings (0.7 kg/ha) compared to the conventional trawl (1.1 kg/ha), and was similar between the other three designs and the conventional trawl (0.8 kg/ha). In (austral) summer 2013/2014, four novel trawl configurations (35 mm diamond or square mesh wing/side panels, and loose or tight hanging ratios) were tested against a conventional trawl with 41 mm diamond mesh wings. Twenty-four deployments of each trawl design were conducted on a twin-rigged trawler, paired with another design. All trawls also had a size-sorting escape grid. Full trawl details are provided in the original paper.

    Study and other actions tested
  17. A replicated, controlled study in 2011 of a fished area of seabed in the Mediterranean Sea, Turkey (Özbilgin et al. 2015) found that three different designs of bottom trawl codend woven by machine instead of by hand, improved size-selectivity and reduced the discarded catches of five commercial fish species compared to a commonly used commercial hand-woven codend. The length at which fish had a 50% chance of escaping was higher in machine-woven codends than hand-woven codends for: red mullet Mullus barbatus (machine: 8–14 cm, hand: 7 cm); brushtooth lizardfish Saurida undosquamis (machine: 23–28 cm, hand: 8 cm); common pandora Pagellus erythrinus (machine: 12–15 cm, hand: 8 cm); goldband goatfish Upeneus moluccensis (machine: 12–21 cm, hand: 5 cm); and Randall’s threadfin bream Nemipterus randalli (machine: 10–14 cm, hand: 6 cm). In addition, the proportions of undersized fish retained for the five species were lower in machine codends compared to the hand codend design, however, there were also increased losses of commercial sizes for some species (see original paper for full data by species). Trials were done with three alternative machine-woven trawl codends (40 mm square mesh; 44 mm and 50 mm diamond mesh) and a commercial hand-woven codend (44 mm diamond mesh) on fishing grounds in Mersin Bay between January-December 2011. Data were collected from a total of 87 individual deployments (20–23 of each of the four codends) onboard a commercial trawler. Tow duration was 80–220 minutes at a speed of 2.3–2.8 knots. A small-mesh cover was fitted over each codend to collect the escaped fish. Catches in the codends and covers were sampled.

    Study and other actions tested
  18. A replicated, paired, controlled study in 2014 of a fished area of seabed in Moreton Bay in the Coral Sea, Australia (Balash et al. 2016) found that using a different design of prawn trawl (a ‘W-trawl’) reduced the quantities of unwanted fish in one of four cases, compared to a conventional trawl design. The catch rates of unwanted fish were lower in the first of four designs of W-trawl configuration compared to the conventional trawl (original: 5 kg/ha: conventional: 2 kg/ha). However, there were no differences in catch rates of unwanted fish between three subsequent modifications to the original W-trawl design and the conventional design (modified: 1–2 kg/ha, conventional: 2–3 kg/ha). In addition, target shrimp Penaeidae catches were reduced in all W-trawl designs compared to the conventional, by 27–80%. Trials were carried out by a double-rigged commercial vessel on trawl fishing grounds in Moreton Bay in February 2014. One of four designs of W-trawl were deployed in paired tows with a conventional trawl design (Florida Flyer). A total of 45 paired deployments were made (10 to 13 of each W-trawl design) and all trawls had a 42 mm diamond mesh codend (see original paper for gear specifications).

    Study and other actions tested
  19. A replicated, paired, controlled study (year not provided) of sand and mud bottom in an estuary off the Tasman Sea, New South Wales, Australia (Broadhurst et al. 2016) found that modifying a prawn trawl net (knot orientation) reduced the catches of unwanted fish species compared to a conventional trawl. Overall catch rate of unwanted fish species was lower in a modified trawl with a negative knot angle of attack than the conventional positive attack angle (modified: 90 fish/ha, conventional: 110 fish/ha). For the five most abundant non-target fish species caught, individual catch rates were lower with negative angles of attack for three: yellowfin bream Acanthopagrus australis (modified: 4, conventional: 8 fish/ha), southern herring Herklotsichthys castelnaui (modified: 2, conventional: 8 fish/ha) and silver biddy Gerres subfasciatus (modified: 0.2, conventional: 1.3 fish/ha); but similar for forktail catfish Arius graeffei (modified: 80, conventional: 88 fish/ha) and Ramsey’s perchlet Ambassis marianus (modified: 0.4, conventional: 0.7 fish/ha). Target school prawn Metepenaeus macleayi catches were also lower with the modified knot orientation (modified: 675, conventional: 775 ind/ha). Fishing trials were done in the austral summer in the Clarence River estuary (2–20 m depth) using a local trawler. Knot-force direction in the top and bottom net panels was compared by turning a conventional trawl with a positive angle of attack inside out to create a negative angle of attack (see original paper for full gear specifications). A total of 24 paired deployments (45 min) of both trawl types were completed. The authors reported that altering the knot orientation may have affected the overall geometry of the trawl during fishing (e.g. a lower position in the water column and lower headline height).

    Study and other actions tested
Please cite as:

Taylor, N., Clarke, L.J., Alliji, K., Barrett, C., McIntyre, R., Smith, R.K., and Sutherland, W.J. (2021) Marine Fish Conservation: Global Evidence for the Effects of Selected Interventions. Synopses of Conservation Evidence Series. University of Cambridge, Cambridge, UK.

Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Marine Fish Conservation

This Action forms part of the Action Synopsis:

Marine Fish Conservation
What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 18

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.


Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape Programme Red List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Bern wood Supporting Conservation Leaders National Biodiversity Network Sustainability Dashboard Frog Life The international journey of Conservation - Oryx British trust for ornithology Cool Farm Alliance UNEP AWFA Butterfly Conservation People trust for endangered species Vincet Wildlife Trust