Use acoustic devices on fishing gear

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

Source countries

Key messages

COMMUNITY RESPONSE (0 STUDIES)

POPULATION RESPONSE (0 STUDIES)

BEHAVIOUR (16 STUDIES)

  • Behaviour change (16 studies): Twelve of 16 controlled studies (including three replicated studies) in the North Atlantic Ocean, the Fortune Channel, the South Atlantic Ocean, Moreton Bay, the Mediterranean Sea, the Celtic Sea, the Rainbow Channel, a coastal site in the UK, the Great Belt, the North Sea, Omura Bay and the Indian Ocean found that using acoustic devices on fishing nets, float lines or simulated fishing nets resulted in harbour porpoises, common bottlenose dolphins, tuxuci dolphins, finless porpoises and seals approaching nets or lines less closely, having fewer encounters or interactions with nets, or activity and sightings were reduced in the surrounding area. The other four studies found that using acoustic devices on trawl nets, float lines or simulated fishing nets did not have a significant effect on the behaviour of common bottlenose dolphins, harbour porpoises, Indo-Pacific humpback dolphins or dugongs.

OTHER (19 STUDIES)

  • Reduction in entanglements/unwanted catch (14 studies): Nine studies (including seven controlled studies and two before-and after studies) in the North Atlantic Ocean, the North Sea, the South Atlantic Ocean, the North Pacific Ocean, the Black Sea, and the South Pacific Ocean found that using acoustic devices on cod traps or fishing nets resulted in fewer collisions of humpback whales or entanglements of harbour porpoises, Franciscana dolphins, beaked whales and small cetaceans. Three studies (including two controlled studies and one before-and-after study) in the North Pacific Ocean found that using acoustic devices on fishing nets resulted in fewer entanglements of some species but not others. One controlled study in the North Atlantic Ocean found that fishing nets with a ‘complete’ set of acoustic devices had fewer entanglements of harbour porpoises, but those with an ‘incomplete’ set did not. One replicated, controlled study in the North Sea and Baltic Sea found that using acoustic devices on fishing nets reduced harbour porpoise entanglements in one fishing area but not the other.
  • Human-wildlife conflict (6 studies): Five of six studies (including six controlled studies, one of which was replicated) in the Baltic Sea, the Mediterranean Sea, the North Pacific Ocean, a coastal site in the UK and the North Sea found that using acoustic devices reduced damage to fish catches and/or fishing nets caused by common bottlenose dolphins and seals. The other study found that acoustic devices did not reduce damage to swordfish catches by California sea lions.

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 controlled, before-and-after study (year not stated) in eight pelagic areas in the North Atlantic Ocean, off the coast of Canada (Lien et al. 1992) found that cod Gadus morhua traps with acoustic devices attached had fewer humpback whale Megaptera novaeangliae collisions than traps without acoustic devices. The average number of humpback whale collisions was lower at traps after acoustic devices were installed (0.02 collisions/day) compared to before (0.4 collisions/day) or at control traps without acoustic devices (0.04 collisions/day). Average fish catches (including target species) were greater in traps with acoustic devices (686 kg/day) than those without (30–235 kg/day). In spring and summer, fishers deployed cod traps for 169 days before and 1,762 days after acoustic devices were installed. Control traps without acoustic devices were deployed for 2,223 days. Six or seven acoustic devices were attached to each trap (one at each corner, 2–3 on the leader section), 2 m below the water surface. Devices emitted sound pulses centred at 4 kHz every 3–6 seconds. Fishers recorded whale collisions and fish catches for each trap deployment.

    Study and other actions tested
  2. A randomized, controlled study in 1994 of a pelagic site in the North Atlantic Ocean, off the coast of New Hampshire, USA (Kraus et al. 1997) found that using active acoustic devices on fishing nets resulted in fewer harbour porpoise Phocoena phocoena entanglements compared with using inactive acoustic devices. The probability of at least one porpoise being entangled was lower in nets with active acoustic devices (0.0027) than in nets with inactive acoustic devices (0.025). Two harbour porpoises were entangled in nets with active devices, whereas 25 were entangled in nets with inactive devices. Catches of target cod Gadus morhuai and pollock Pollachius virens were similar in nets with active and inactive devices (data not reported). Commercial gill net fishers deployed a total of 844 net strings (each comprising 12 nets, 92 m long x 4 m deep, stretched mesh size 15 cm) with acoustic devices attached at 92 m intervals. The acoustic devices on each net string were randomly assigned as active (emitting 300 ms sounds every 4 seconds at 10 kHz; total 421 net strings) or inactive (silent; total 423 net strings). Net strings were submerged for 24 h. Onboard observers and fishers recorded porpoise entanglements and fish catches during daily hauls in October–December 1994.

    Study and other actions tested
  3. A controlled study in 1995–1997 of a pelagic area in the North Pacific Ocean, off the coast of Washington, USA (Gearin et al. 2000) found that fishing nets with acoustic devices attached had fewer entanglements of harbour porpoises Phocoena phocoena than nets without acoustic devices, but the number of harbour seal Phoca vitulina entanglements did not differ. In 1995 and 1996, harbour porpoise entanglement rates were lower in fishing nets with acoustic devices attached (0.02 porpoises/net/day) than without (0.4–0.5 porpoises/net/day). Harbour seal entanglement rates did not differ with or without acoustic devices (both 0.05 seals/net/day). In 1997, with acoustic devices on all nets, entanglement rates were 0.07 porpoises/net/day and 0.07 seals/net/day. Catches of target chinook salmon Oncorhynchus tshawytscha and sturgeon Acipenser spp. did not differ significantly with or without acoustic devices (see original paper for data). In 1995 and 1996, two pairs of gill nets (183 m long, 50–80 meshes deep) were deployed on the ocean bottom at depths of 8–12 m, spaced >300 m apart. One net in each pair had 11 acoustic devices (‘piezo buzzers’) attached at 17 m intervals 4–7 m below the surface; the other had no devices. The devices (emitting pulses every 4 seconds with peak frequencies at 3 and 20 kHz) were rotated between nets. In 1997, all four nets had acoustic devices attached. Nets were checked every 24 h on 51–61 days in July–August 1995 and 1996 and 180 days in June–August 1997.

    Study and other actions tested
  4. A controlled study in 1999 in a fjord in the Fortune Channel, Vancouver Island, Canada (Culik et al. 2001) found that using an acoustic device on a float line resulted in harbour porpoise Phocoena phocoena groups approaching less closely. The average distance of porpoise groups from the centre of the float line was greater during trials with an active acoustic device attached (530 m) compared to trials before (150 m) or after (152 m). In June–July 1999, a float line was deployed during six days before an acoustic device was attached (total 26.5 h),  five days with an acoustic device attached (emitting 300 ms pulses every 5–30 seconds at frequencies of 20–160 kHz; total 21 h), and two days after the device was removed (total 7 h). The float line (65 m long) had 10-m long weighted lines attached every 0.5 m. The acoustic device was attached to the centre of the float line, 30 cm below the water surface. Porpoises were tracked from the shore using a theodolite before (172 groups), during (44 groups) and after (22 groups) the acoustic device was attached. An acoustic detector deployed 1 m below the centre of the float line recorded porpoise echolocation clicks.

    Study and other actions tested
  5. A replicated, randomized, controlled study in 1999–2000 of multiple pelagic sites in the South Atlantic Ocean, off the coast of Buenos Aires, Argentina (Bordino et al. 2002) found that fishing nets with active acoustic devices attached had fewer entanglements of Franciscana dolphins Pontoporia blainvillei than nets with inactive acoustic devices. Entanglement rates of Franciscana dolphins were lower in fishing nets with active acoustic devices attached (0.002 dolphins/m2/hour) than in nets with inactive acoustic devices (0.01 dolphins/m2/hour). However, South American sea lions Otaria fivescens damaged fish more in nets with active than inactive devices (see original paper for data). C­atch rates of target fish did not differ between nets (active devices: 2.2 kg/m2/hour; inactive devices: 2.3 kg/m2/hour). Between October 1999 and February 2000, a total of 604 gill nets with acoustic devices attached (Dukane NetMark 1000, spaced 50 m apart) were deployed on the ocean bottom at multiple sites (number not reported). Each of 604 nets was randomly assigned as a treatment (active acoustic devices emitting pulses every 4 seconds with a peak frequency of 10 kHz; 309 nets) or control (inactive silent acoustic devices; 295 nets). Observers on board the fishing vessels recorded fish catches, entangled dolphins and sea lion damage as each of the 604 nets was retrieved.

    Study and other actions tested
  6. A replicated, randomized, controlled study in 1996–1997 of multiple pelagic sites in the North Pacific Ocean, off the coasts of California and Oregon, USA (Barlow & Cameron 2003; same fishery as Carretta et al. 2008 and Carretta & Barlow 2011) found that fishing nets with acoustic devices attached had fewer entanglements of short-beaked common dolphins Delphinus delphis and California sea lions Zalophus californianus than nets without acoustic devices, but no difference was found for eight other marine mammal species. Entanglement rates were lower for short-beaked common dolphins and California sea lions in fishing nets with acoustic devices attached (both 0.01 entanglements/net) than in nets without acoustic devices (short-beaked common dolphins: 0.07 entanglements/net; California sea lions: 0.05 entanglements/net). Numbers of entanglements did not differ for eight other dolphin, porpoise, whale and seal species (see original paper for data), although sample sizes were small. Between April 1996 and October 1997, ‘drift’ gill nets were deployed at multiple sites (number not reported) and randomly assigned as treatment nets (Dukane NetMark 1000 acoustic devices; 295 nets) or control nets (no acoustic devices; 314 nets). Acoustic devices were spaced 91 m apart and emitted 300 ms pulses with a peak frequency of 10–12 kHz. Observers on board the fishing vessels recorded entangled marine mammals as each of the 609 nets was retrieved.

    Study and other actions tested
  7. A randomized, controlled study in 2001 at a coastal site in the North Atlantic Ocean, off the coast of North Carolina, USA (Cox et al. 2003) found that fishing nets with active acoustic devices were approached within 100 m by fewer common bottlenose dolphin Tursiops truncatus groups than nets with inactive devices, although the number of dolphin groups observed within 300 m of the nets and average closest approach distances were similar. Fewer dolphin groups approached within 100 m of nets with active acoustic devices (2 of 25 groups, 8%) than inactive devices (7 of 15 groups, 47%). The average number of dolphin groups observed within 300 m of the nets and average closest approach distances did not differ significantly with active (0.4 groups/h; 47 m) or inactive (0.6 groups/h; 38 m) acoustic devices. A gill net was deployed on random days with three active acoustic devices attached (Dukane NetMark 1000, emitting regular interval pulses at 10 kHz; total 13 days) or three inactive (silent) acoustic devices (total nine days). The net (200 m long, stretched mesh size 76 mm) was deployed 300 m from a beach perpendicular to the shore in water 3–6 m deep. Two observers tracked 40 dolphin groups from the shore using a theodolite over 22 days in April–May 2001.

    Study and other actions tested
  8. A controlled study in 1996–1998 at a coastal site in the South Atlantic Ocean, near Fortaleza, Brazil (Monteiro-Neto et al. 2004) found that float lines with active acoustic devices attached had fewer tucuxi dolphin sightings Sotalia fluviatilis around them than float lines with inactive acoustic devices or no devices. On average, fewer tucuxi dolphins were sighted in two quadrats on either side of a float line with active acoustic devices attached compared to float lines with inactive acoustic devices or no devices attached (data reported as statistical model results). The average number of dolphin sightings did not differ significantly between trials within seven other quadrats that were not immediately adjacent to the float line (see original paper for data). A float line (100 m long) was deployed with active acoustic devices attached (30 trials), inactive (silent) acoustic devices attached (20 trials) and with no devices (55 trials). Each trial lasted 1–7 h. Five acoustic devices (Dukane NetMark 1000) were evenly spaced along the float line. Two observers on the shore recorded dolphin sightings within nine quadrats (0.5–0.9 km2) in a 6-km2 area surrounding the float line during each of the 105 trials between November 1996 and August 1998.

    Study and other actions tested
  9. A replicated, controlled study in 1998–2000 at 19 pelagic sites in the northern Baltic Sea, Sweden (Fjälling et al. 2006) found that using acoustic devices at salmon trap-nets resulted in an increase in intact fish catches and a decrease in damaged fish and fishing gear, likely due to reduced grey seal Halichoerus grypus predation. Trap-nets with acoustic devices had higher intact fish catches (26 kg/day), lower average quantities of fish damaged by seals and birds (4 kg/day) and fewer new holes in fishing gear (result reported from text, which does not match data in table) than those without acoustic devices (fish catches: 12 kg/day; damaged fish: 7 kg/day). Salmon trap-nets were deployed at each of 19 sites (3–9/year) with acoustic devices (total 600 fishing days, 755 trap-net lifts) and without acoustic devices (total 493 fishing days, 668 trap-net lifts). Trap-nets consisted of a leader net starting close to the shore and ending in a funnel-shaped net and fish chamber (10 x 5 x 5 m). An acoustic device (Lofitech Fishguard) was deployed on a raft next to the fish chamber with the transducer at a 5 m depth (emitting 250–500 ms pulses at 15 kHz). Eight commercial fishers recorded catch weight, numbers of fish damaged by seals and birds, and damage to fishing gear during a total of 1,423 trap-net lifts across three fishing seasons in 1998–2000.

    Study and other actions tested
  10. A controlled study in 2003 in a fjord in the Fortune Channel, Vancouver Island, Canada (Koschinski et al. 2006) found that using an acoustic device at a float line did not reduce the approach distances or time spent near the line by harbour porpoises Phocoena phocoena. Closest approach distances of harbour porpoises and time spent within 50 m of float lines did not differ significantly between lines with an acoustic device (average 25 m; 32 seconds) or without (average 28 m; 17 seconds). In August 2010, a float line was deployed on one occasion (total 2.8 h) without an acoustic device and on three occasions (total 12.5 h) with an acoustic device. The acoustic device (a CD player with an underwater transducer at a depth of 4.5 m) emitted 0.3 second pulses at 2.5 kHz. Porpoises within 50 m of the float line were tracked with a theodolite during each of the four deployments.

    Study and other actions tested
  11. A controlled study in 2002 and 2005 at a pelagic site in Moreton Bay, Queensland, Australia (Hodgson et al. 2007) found that active acoustic devices of two types deployed to simulate a fishing net had similar numbers of dugongs Dugong dugon passing between them compared to inactive acoustic devices, and dugong orientation and feeding behaviour were also similar. The number of dugongs passing between two acoustic devices was similar when the devices were active or inactive (data reported as statistical model results). The proportion of dugongs oriented towards the acoustic devices and the number of dugongs feeding within 100 m of them also did not differ significantly when the devices were active or inactive. Two acoustic devices (either 4 kHz or 10 kHz ‘BASA’ devices) were deployed 50–55 m apart at depths of 1 m below the water surface to simulate a fishing net. The devices were attached to a research vessel and an anchored floating tube close to dugong herds. Each trial comprised three 10-minute sequential treatments with both devices inactive (silent), active (emitting pulses at 4-second intervals), and inactive (silent). Ten trials were carried out in August 2002 with 10 kHz devices. Sixteen trials were carried out in July 2005 with 4 kHz devices. Dugong behaviour and feeding plumes (disturbed sediment) were recorded with a video camera attached to a balloon during each of the 26 trials.

    Study and other actions tested
  12. A replicated, randomized, controlled study in 2005 of multiple pelagic sites in the Mediterranean Sea, off the Balearic Islands, Spain (Brotons et al. 2008) found that one of three types of acoustic device attached to fishing nets reduced common bottlenose dolphin Tursiops truncatus interactions with the nets. The average interaction rate of bottlenose dolphins with nets was 70% lower when active Aquatec AQUAmark 210 devices were attached to nets than when inactive devices or no devices were attached (data reported as statistical model results). The difference in interaction rates was not significant for two other types of acoustic device: Dukane NetMark 1000 and SaveWave Dolphinsaver High-impact. Target fish yields (measured as profit) did not differ significantly between treatments (see original paper for details). A total of 1,193 gill nets were deployed on the ocean bottom at multiple sites (number not reported). One of seven treatments was randomly assigned to each net: one of three types of active acoustic device attached (Aquatec: 260 nets; Dukane: 272 nets; SaveWave: 211 nets), one of three types of inactive (silent) acoustic device attached (Aquatec: 118 nets; Dukane: 74 nets; SaveWave: 114 nets) or no device attached (144 nets). Observers on board each of 59 fishing vessels recorded dolphin interactions (sightings of dolphins around the nets or dolphin-damaged fish) in July–December 2005.

    Study and other actions tested
  13. A before-and-after study in 1990–2006 of multiple pelagic sites in the North Pacific Ocean, off the coasts of California and Oregon, USA (Carretta et al. 2008; same fishery as Barlow & Cameron 2003 and Carretta & Barlow 2011) found that using acoustic devices on fishing nets reduced the number of beaked whale (Ziphiidae) entanglements. No beaked whales were found entangled in fishing nets during the 11 years in which acoustic devices were used, whereas 33 whales of at least six species were entangled during the six years before the devices were used (see original paper for details). In 1990–1995, a total of 3,303 nets were deployed without acoustic devices. In 1996–2006, a total of 4,381 nets were deployed with acoustic devices attached at 91 m intervals (average 40 devices/net). The devices emitted 300 ms pulses at 10–12 kHz. Each of the 7,684 ‘drift’ gill nets (1,800 m long x 65 m deep) was deployed from dusk until dawn at depths of 11–90 m to catch swordfish and sharks. Observations of entangled whales were made by biologists on board fishing vessels in 1990–2006.

    Study and other actions tested
  14. A controlled study in 2001 of a pelagic area in the Mediterranean Sea, off the Balearic Islands, Spain (Gazo et al. 2008) found that using acoustic devices on fishing nets reduced damage to nets and fish caused by common bottlenose dolphins Tursiops truncatus. Fishing nets with active acoustic devices had significantly fewer holes (average 1 hole/net) than nets with inactive acoustic devices (average 8 holes/net) or no devices (average 6 holes/net). The percentage of caught fish bitten by dolphins was also lower in nets with active devices (7%) compared to inactive devices (13%) or no devices (17%), although statistical significance was not assessed. Catch rates of target red mullet Mullus surmuletus did not differ significantly between nets with active devices (0.6 kg/net), inactive devices (0.7 kg/net) and no devices (0.9 kg/net). A total of 55 trammel net deployments (each with multiple nets, 50 m long x 2 m high, tied together) were deployed across a fishing area (340 km2). One of three treatments was rotated between deployments: active acoustic devices attached (27 deployments), inactive (silent) acoustic devices attached (16 deployments) or no devices (12 deployments). Acoustic devices (Aquatec AQUAmark 100; emitting eight different signals of 5–30 second duration at 20–160 kHz) were attached at 150 m intervals. An observer on board each of three fishing vessels recorded fish catches, dolphin-damaged fish and new holes in the nets during each of the 55 hauls in September–October 2001.

    Study and other actions tested
  15. A controlled study in 1999–2007 of a pelagic area in the Northwest Atlantic Ocean, USA (Palka et al. 2008) found that fishing nets with a ‘complete’ set of acoustic devices attached had fewer harbour porpoise Phocoena phocoena entanglements than nets without acoustic devices, but nets with an ‘incomplete’ set of acoustic devices had the highest number of entanglements. Harbour porpoise entanglement rates were lower in nets with a ‘complete’ set of acoustic devices attached (0.02 porpoises/metric tons landed) than in nets with no acoustic devices (0.05 porpoises/metric tons landed). Entanglement rates were highest in nets with an ‘incomplete’ set of acoustic devices attached (0.12 porpoises/metric tons landed). In 1999–2007, acoustic devices were attached to gill nets during commercial fishing operations. Gill net strings were deployed with either a ‘complete’ set of acoustic devices attached (11 devices on each string of 10 x 92 m long nets; total 2,407 hauls), an ‘incomplete’ set of acoustic devices (<11 devices/string; total 1,065 hauls), or no devices (total 3,157 hauls). Acoustic devices emitted 300 ms pulses every 4 seconds at 10 kHz. Observers on board the fishing vessels recorded porpoise entanglements, fish catches and numbers of acoustic devices used during each of the 6,629 hauls in 1999–2007.

    Study and other actions tested
  16. A controlled study in 2006 of a pelagic area in the Mediterranean Sea, off the coast of southern Italy (Buscaino et al. 2009) found that using acoustic devices on a fishing net resulted in higher fish catches and less net damage, likely due to reduced predation by common bottlenose dolphins Tursiops truncatus. A fishing net with acoustic devices attached had higher average fish catches (5.2 kg/h) and fewer small holes (0.8 holes/50 m) than a net without acoustic devices (4.1 kg/h; 1.2 holes/50 m). Two identical gill nets (900 m long x 2.2 m deep) were deployed on the ocean bottom; one with four evenly spaced acoustic devices attached and one without. Acoustic devices (STM and SEAMed model DDD02) emitted 6-second signals at random intervals with a frequency range of 0.1–150 kHz. Researchers on board the fishing vessel recorded the presence of dolphins and fish catches in each net during 29 hauls in spring 2006. Small holes (<20 cm) were counted in both nets at the end of the experiment.

    Study and other actions tested
  17. A controlled study in 2001 at a pelagic site in the North Atlantic Ocean, off the coast of Scotland, UK (Carlström et al. 2009) found that using acoustic devices on a simulated fishing net reduced the approach distances and echolocation activity of harbour porpoises Phocoena phocoena. The average approach distance of porpoise groups from the ‘net’ was greater when acoustic devices were active (961 m) than inactive (653 m). The average number of echolocation encounters within 0–500 m of the ‘net’ was lower when devices were active (0.1–0.3 encounters/h) than inactive (0.3–0.7 encounters/h). The difference in the number of echolocation encounters was not significant at 750 m (active: 0.36 encounters/h; inactive: 0.42 encounters/h). In April–June 2001, a simulated fishing net (a 700-m lead line) was deployed on the ocean bottom with eight acoustic devices (Dukane NetMark 1000) attached at 100 m intervals. Six devices were active (emitting 300 ms pulse at 10–12 kHz every four seconds) or inactive (silent) for alternating 4-h periods. Two devices at the centre of the ‘net’ were inactive throughout. Acoustic loggers deployed at 0, 250, 500 and 750 m from the ‘net’ recorded porpoise echolocation clicks while acoustic devices were active (total 1,472 h) and inactive (total 1,352 h). Observers on the shore tracked porpoise groups with a theodolite while acoustic devices were active (11 groups during 30 h) and inactive (39 groups during 49 h).

    Study and other actions tested
  18. A controlled study in 2006 of a pelagic area in the Black Sea, Turkey (Gönener & Bilgin 2009) found that fishing nets with acoustic devices attached had fewer entanglements of harbour porpoises Phocoena phocoena than nets without acoustic devices. Harbour porpoise entanglement rates were lower in nets with acoustic devices (0.01 porpoises/day) than in those without acoustic devices (0.47 porpoises/day). Catch rates of target Black Sea turbot Schophthalmus maeoticus were higher in nets with active acoustic devices (1.1 fish/day) than in those without (0.5 fish/day). During each of 20 fishing trips, one gill net string was deployed with acoustic devices attached (Dukane NetMark 1000 emitting 300 ms signals every 4 seconds at 10–12 kHz, spaced 200 m apart) and one was deployed without acoustic devices. Each string comprised 16 nets tied together (total length 1.1 km, 160 mm mesh size). Nets were deployed at depths of 17–183 m for 168–288 h. Entangled porpoises and fish catches were recorded during each of the 20 fishing trips in March–April 2006.

    Study and other actions tested
  19. A before-and-after study in 1990–2009 of multiple pelagic sites in the North Pacific Ocean, off the coasts of California and Oregon, USA (Carretta & Barlow 2011; same fishery as Barlow & Cameron 2003 and Carretta et al. 2008) found that using acoustic devices on fishing nets reduced entanglements of two of five marine mammal species, but did not reduce damage to target broadbill swordfish Xiphias gladius catches by California sea lions Zalophus californianus. The proportion of fishing net deployments with at least one entanglement was lower for nets with acoustic devices than those without for short-beaked common dolphins Delphinus delphis (with devices: 3.2%; without: 5.7%) and northern elephant seals Mirounga angustirostris (with devices: 0.5%; without: 2.4%). The difference was not significant for northern right whale dolphins Lissodelphis borealis (with devices: 0.003%; without: 0.005%), Pacific white-sided dolphins Lagenorhynchus obliquidens (with devices: 0.005%; without: 0.003%) or California sea lions (with devices: 2.6%; without: 1.6%). In one year, the proportion of deployments with swordfish catches damaged by California sea lions did not differ significantly with (19 of 69 deployments; 28%) and without acoustic devices (38 of 124 deployments; 31%). In 1990–1998, fishing nets (1,281 in total) were deployed without acoustic devices. In 1996–2009, fishing nets (2,792 in total) were deployed with acoustic devices (≥30 devices/net at 91 m intervals, emitting 300 ms pulses every 4 seconds at 10–12 kHz). Nets (each 1.5–1.8 km long, 65 m deep, 40–60 cm mesh size) were deployed for 8–20 h between dusk and dawn by a ‘drift’ gill net fishery targeting swordfish and sharks. Onboard observers recorded mammal entanglements in 1990–2009. Sea lion damage to swordfish catches (shredding of the body) was recorded in 1997.

    Study and other actions tested
  20. A controlled study in 2008–2011 of a pelagic area in the North Atlantic Ocean, off the coast of southwest England, UK (Northridge et al. 2011) found that fishing nets with acoustic devices attached had fewer entanglements of harbour porpoises Phocoena phocoena than nets without acoustic devices. Porpoise entanglement rates were lower in fishing nets with acoustic devices attached (0.007 porpoises/haul) than in those without acoustic devices (0.02 porpoises/haul). Fishing vessels (>12 m) deployed fleets of gill nets (up to 8 km in length) with acoustic devices attached (total 999 hauls) and without acoustic devices (total 907 hauls). Dolphin Dissuasive Devices (model DDD-02, STM Products) were either attached to the middle of each section of 20 net panels (in 2008) or to the end ropes and 10 m above the anchor (in 2009–2011). Between August 2008 and April 2011, entangled porpoises were recorded during each haul by independent observers (1,709 hauls) or fishers (197 hauls).

    Study and other actions tested
  21. A replicated, controlled study in 2009–2010 of three pelagic sites in the Celtic Sea, off the coast of Cornwall, UK (Hardy et al. 2012) found that fishing nets with acoustic devices had lower echolocation activity of harbour porpoises Phocoena phocoena around them than nets without acoustic devices. Overall, 35–51% fewer harbour porpoise clicks were recorded at nets with acoustic devices (total 800–20,000 clicks/site) than at nets without acoustic devices (total 2,000–40,000 clicks/site). Only one entangled porpoise was found during the study, in a net without an acoustic device. Three commercial fishing vessels deployed pairs of ‘tangle’ nets (267 mm mesh; number not reported) between April 2009 and April 2010. Each pair had an experimental net with acoustic devices attached (Aquatec AQUAmark 100, spaced 200 m apart) and a control net with no devices. Acoustic devices emitted 400 ms pulses at 20–140 kHz. Each pair of nets was deployed for approximately five days at depths of 20–100 m to target benthic fish species. Acoustic detectors attached to the nets recorded porpoise echolocation clicks during a total of 640 days in 2009–2010.

    Study and other actions tested
  22. A controlled study in 2012–2013 of a pelagic area in the North Atlantic Ocean, off the coast of Cornwall, UK (Crosby et al. 2013) reported that fishing nets with acoustic devices had lower echolocation activity of harbour porpoises Phocoena phocoena around them than nets without acoustic devices. Results are not based on assessments of statistical significance. Overall, the number of porpoise echolocation clicks recorded was 6,321 clicks at nets with acoustic devices, compared to 34,600 clicks at nets without acoustic devices. In October 2012–March 2013, four fishing vessels (<12 m long) deployed pairs of inshore ‘tangle’ nets (22–35 cm monofilament mesh deployed flat on the seabed) with and without acoustic devices during a total of 161 days of fishing. Acoustic devices (Fishtek Banana Pingers, spaced 2 m apart) emitted 300 ms sounds at random intervals of 4–12 seconds with random frequencies between 50–120 kHz. Nets were deployed for five days at depths of 20–100 m. An acoustic logger attached to each net recorded porpoise echolocation clicks.

    Study and other actions tested
  23. A controlled study in 2006 of multiple pelagic sites in the North Sea, Denmark (Larsen et al. 2013) found that fishing nets with acoustic devices attached at two different spacings had fewer entanglements of harbour porpoises Phocoena phocoena than nets without acoustic devices. Overall, entangled porpoises were recorded in fewer hauls of fishing nets with acoustic devices attached at 455 m spacings (0 hauls) and 585 m spacings (5 hauls) than nets with no acoustic devices attached (22 hauls). Numbers of entanglements did not differ significantly between the two device spacings. Average catch rates of target hake Merluccius spp. did not differ significantly between nets with acoustic devices at 455 m spacings (29 fish/km/day) and nets without acoustic devices (30 fish/km/day; data not reported for nets with devices at 585 m spacings). Strings of 45–135 gill nets were deployed during five commercial fishing trips in July–September 2006. The nets had acoustic devices (Aquatec AQUAmark 100) attached at spacings of 455 m (24 hauls) or 585 m (43 hauls) or had no devices attached (41 hauls). Observers on board the fishing vessels recorded porpoise entanglements and hake catches within each of the 108 hauls.

    Study and other actions tested
  24. A replicated, controlled study in 2009–2011 of multiple pelagic sites in the South Pacific Ocean, northern Peru (Mangel et al. 2013) found that fishing nets with acoustic devices attached had fewer entanglements of small whale, dolphin and porpoise species than nets without acoustic devices. Average entanglement rates were lower in fishing nets with acoustic devices (0.5 cetaceans/km/h) than in nets without acoustic devices (0.8 cetaceans/km/h). Five species or species groups were entangled including dolphins, porpoises, and pilot whales Globicephala spp. (see original paper for details). Catch rates of target sharks and eagle rays Myliobatis spp. did not differ significantly with acoustic devices (26 sharks/km/h; 0.002 rays/km/h) or without (19 sharks/km/h; 0.001 rays/km/h). Six small-scale ‘drift’ net vessels carried out 43 experimental fishing trips (total 156 nets with acoustic devices) and 47 control trips (total 195 nets without acoustic devices) during 29 months in April 2009–August 2011. Acoustic devices (Dukane NetMark 1000, emitting 300 ms pulses at 10–12 kHz) were attached to the lead line of experimental nets spaced 200 m apart at a depth of 14 m. Onboard observers recorded entanglements and target fish catches during each of the 90 fishing trips.

    Study and other actions tested
  25. A randomized, controlled study in 2007–2008 at a pelagic site in the Rainbow Channel, Queensland, Australia (Soto et al. 2013) found that when a row of three active acoustic devices was deployed to simulate a fishing net, the number of Indo-Pacific humpback dolphin Sousa chinensis groups observed, the minimum surfacing distance of dolphins to a device and the number of days in which dolphins did not cross devices was similar to when three inactive devices were deployed. The number of dolphin groups observed did not differ significantly with active acoustic devices (average 4 groups/day) or inactive acoustic devices (5 groups/day). The same was true for the minimum distance between a surfacing dolphin and an acoustic device (active devices: average 41 m; inactive: 33 m) and the number of days in which dolphins did not cross the row of acoustic devices (active devices: 7 days; inactive: 3 days). A row of three acoustic devices (Fumunda acoustic alarms) was deployed across a channel to simulate a gill net. On randomly selected days, all three devices were either active (emitting 300 ms pulses every 4 seconds at 10 kHz; total 10 days) or inactive (silent; total 10 days).  Devices were attached to buoys anchored to the seafloor and submerged at a depth of 5 m in water 10–15 m deep. A total of 84 dolphin groups were observed from the shore during 20 days using a video camera attached to a theodolite in September 2007–April 2008.

    Study and other actions tested
  26. A randomized, controlled study in 2004–2005 across two coastal areas in the North Atlantic Ocean, off the coast of North Carolina, USA (Waples et al. 2013) found that when active acoustic devices were used on fishing nets, common bottlenose dolphins Tursiops truncatus interacted with the nets less and echolocated more compared to when inactive devices were used. Fewer dolphins approached within 500 m and interacted with nets with active acoustic devices than nets with inactive acoustic devices (data reported as statistical model results). Dolphins spent more time echolocating near to nets with active acoustic devices than nets with inactive or no devices. In 2004–2005, commercial fishers deployed 83 gill nets with active acoustic devices (SaveWave devices attached to the float line, 100 m apart, emitting sounds at frequencies of 5–90 kHz and 30–160 kHz) and 68 gill nets with inactive (silent) devices. Gill nets (300 m long) were deployed perpendicular to the shore. An onboard observer recorded dolphin behaviour around each of the 151 nets. In 2004, a research vessel towing a hydrophone recorded the echolocation clicks of seven dolphins within 500 m of four nets with active acoustic devices, two nets with inactive devices and one net without devices deployed by another fishery.

    Study and other actions tested
  27. A randomized, controlled study in 2009–2010 at a coastal site (water body not stated) in Scotland, UK (Harris et al. 2014) found that using an acoustic device at a bag-net reduced the number of grey seals Halichoerus grypus and harbour seals Phoca vitulina around the net and the amount of seal-damaged salmon (Salmonidae). The number of grey and harbour seal sightings/survey within 80 m of the net was lower when the acoustic device was active than when it was inactive (data reported as statistical model results). No seal-damaged salmon were found in the nets when the acoustic device was turned on, whereas 5–7% of salmon catches were damaged when the device was off. An acoustic device (Lofitech Seal Scarer) was deployed alongside a salmon double bag-net deployed 90 m offshore. In July–August 2009 and 2010, the acoustic device was randomly set as active (emitting 500 ms pulses at 15 kHz) or inactive (silent) for 12 h (2009) or 24 h periods (2010). An observer recorded seals from the shore during surveys (each lasting an average of 1.4 h) with the device turned on (34 surveys) and off (41 surveys). Fishers recorded seal-damaged catches during hauls with the device turned on (78 hauls) and off (104 hauls).

    Study and other actions tested
  28. A replicated, controlled study in 1997 at a wreck and an area of seabed in the North Sea, Denmark (Larsen & Eigaard 2014) found that using active acoustic devices on fishing nets resulted in fewer harbour porpoise Phocoena phocoena entanglements compared with using inactive acoustic devices or no devices. At both sites, harbour porpoises were found entangled in fewer nets with active acoustic devices attached (wreck: 0 nets; seabed: 1 net) than in nets with inactive or no acoustic devices attached (wreck: 8 nets; seabed: 15 nets). Catches of target cod Gadus morhua did not differ significantly with active, inactive, or no acoustic devices (data reported as statistical model results). Gill nets were deployed on the ocean bottom at depths of 20–80 m at two sites (a wreck and a flat/stony seabed). Nets had active acoustic devices attached (wreck: 1,052 nets; seabed: 5,596 nets), inactive (silent) acoustic devices attached (wreck: 1,056 nets; seabed: 5,210 nets) or no devices (wreck: 74 nets; seabed: 2,973 nets). Acoustic devices (prototype LU-1, Loughborough University, UK) were attached to nets at 70 m intervals and emitted 300 ms pulses at 40–120 kHz. Observers on board each of 14 fishing vessels recorded porpoise entanglements for a total of 592 hauls during 168 fishing days in August–October 1997.

    Study and other actions tested
  29. A controlled study in 2005 in a pelagic area of the Great Belt, Denmark (Khyn et al. 2015) found that using active acoustic devices of two types on simulated fishing nets resulted in fewer detections of harbour porpoises Phocoena phocoena compared to when devices were inactive. Harbour porpoise detection rates within each of two 0.6 km2 areas were lower when two types of acoustic device were active (‘Airmar’ devices: 40% lower; ‘SaveWave’ devices: 65% lower) compared to when devices were inactive (data reported as statistical model results). Detection rates at three control sites without acoustic devices located 2.5, 3 and 5 km away did not change significantly over the same period. Acoustic devices were deployed across two areas (each 0.6 km2). Fifty-five ‘Airmar’ devices (emitting 300 ms pulses every 4 seconds at 10 kHz) were deployed 100 m apart in one area.  Fifteen ‘SaveWave Black Save’ devices (emitting pulses of 200–900 ms every 4–16 seconds with frequency sweeps of 30–60 kHz) were deployed 200 m apart in the other. In May–June 2005, the acoustic devices were alternately activated and deactivated for six repeating cycles of 2–9 days to simulate gill net fishery deployments. Seven acoustic detectors (two in each area; three at control sites) recorded porpoise echolocation clicks during each of the two acoustic device deployments.

    Study and other actions tested
  30. A replicated, controlled study in 2013–2014 of two pelagic areas in the North Sea and Baltic Sea, Denmark and Germany (Culik et al. 2016) found that fishing nets with acoustic devices attached had fewer harbour porpoise Phocoena phocoena entanglements than nets without devices in one area but not the other. In the Baltic Sea, no porpoises were entangled in nets with acoustic devices, whereas nine porpoises were entangled in nets without acoustic devices. In the North Sea, the number of entangled porpoises did not differ significantly between nets with acoustic devices (two porpoises) or without (three porpoises). In 2013–2014, commercial fishing vessels simultaneously deployed gill nets (number not reported) with and without acoustic devices across two areas. Acoustic devices (‘PALfi’ Porpoise Alerting Devices) emitted three synthetic porpoise alert calls/minute (1.3 second sweeps consisting of 700 clicks centred at 133 kHz). Devices were attached to the headrope of gill nets, spaced 200 m apart. Fishers reported entangled porpoises. Some fishing trips were additionally monitored by onboard video equipment and scientific observers (number not reported).

    Study and other actions tested
  31. A controlled study in 2015 at a bay in the North Sea, Scotland, UK (Harris & Northridge 2016) found that using an active acoustic device alongside a bag-net reduced seal presence at the net and resulted in greater catches of undamaged fish (Salmonidae) compared to when the device was inactive. Seal presence was lower when the acoustic device was turned on than turned off (data reported as statistical model results; seal species not reported). Catch rates of fish without seal damage were greater with the acoustic device turned on than turned off (data reported as statistical model results). An acoustic device (Airmar dB Plus II) was deployed alongside a bag-net for five months in April–August 2015. The device was turned on (emitting acoustic signals; total 1,522 h) and off (silent; total 578 h) during randomly selected periods. An underwater video system recorded the presence of seals at the net with the acoustic device turned on (80 hauls) and off (39 hauls). Fishers recorded fish catches and seal damage during hauls with the acoustic device turned on (108 hauls) and off (50 hauls).

    Study and other actions tested
  32. A controlled study in 2011–2012 of a pelagic site in Omura Bay, Japan (Amano et al. 2017) found that using two acoustic devices on a fishing net reduced the number of encounters of finless porpoises (Neophocaena) with the net. Fewer finless porpoise encounters were recorded each day at the net when the acoustic devices were turned on than when they were turned off (data reported as statistical model results). Two acoustic devices (Aquatec AQUAmark 100) were attached to a fishing net (one on the upper rope of the guide net, one at the entrance of the enclosure net) at a depth of 30 cm. The net was deployed in water 10–15 m deep. Both acoustic devices were turned on (emitting 200–300 ms pulses at 20–160 kHz) or off (silent) for alternating two-week periods in April–December 2011 and 2012. A passive acoustic event recorder deployed 40 m offshore from the net at a depth of 1.5 m recorded daily encounters of finless porpoises.

    Study and other actions tested
  33. A controlled study in 2013 of a pelagic area in the Indian Ocean, northwest Australia (Santana-Garcon et al. 2018) found that trawl nets with acoustic devices attached had a similar number and duration of common bottlenose dolphin Tursiops truncatus interactions compared to trawl nets without acoustic devices. Average daily interaction rates of dolphins with trawl nets did not differ significantly between nets with acoustic devices (0.7 interactions/minute) and without (0.4 interactions/minute). The average duration of interactions also did not differ significantly with acoustic devices (1.7 minutes) or without (1.3 minutes). Three commercial vessels carried out 14 trawls with acoustic devices attached to trawl nets and 17 trawls without acoustic devices. Dolphin Dissuasive Devices (emitting random frequencies between 2 and 500 kHz) were attached on either side of an underwater video camera installed within each trawl net. All trawls were carried out during the day with a single stern trawl net towed close to the seabed in water 50–100 m deep. Video cameras recorded dolphin interactions with the nets during each of the 31 trawls in January–February 2013.

    Study and other actions tested
Please cite as:

Berthinussen, A., Smith, R.K. and Sutherland, W.J. (2021) Marine and Freshwater Mammal Conservation: Global Evidence for the Effects of Interventions. Conservation Evidence Series Synopses. 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 and Freshwater Mammal Conservation

This Action forms part of the Action Synopsis:

Marine and Freshwater Mammal Conservation
Marine and Freshwater Mammal Conservation

Marine and Freshwater Mammal Conservation - Published 2021

Marine and Freshwater Mammal Synopsis

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, terrestrial 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