© 2006 International Council for the Exploration of the Sea
Acoustic harassment devices reduce seal interaction in the Baltic salmon-trap, net fishery
a National Board of Fisheries SE-178 93 Drottningholm, Sweden
b University of Aarhus, Department of Zoophysiology C. F. Moellers Allé Building 131, DK-8000 Aarhus C, Denmark
c National Board of Fisheries Box 423, SE-401 26 Gothenburg, Sweden
*Correspondence to A. Fjälling: tel: +46 8 6990622; fax: +46 8 6990650. e-mail: arne.fjalling{at}fiskeriverket.se.
Acoustic harassment devices (AHDs) were deployed at salmon-trap nets in the Baltic Sea to reduce gear and catch damage by grey seals (Halichoerus grypus). The AHDs emitted pulses of 250–500-ms duration, worked at a frequency of 15 kHz, and a source level of 179 dB re 1 µPa rms at 1 m. AHDs were deployed during three consecutive fishing seasons. Catches were significantly higher in traps with AHDs (25.5 kg d–1) than in controls (12.0 kg d–1), and catch damage was less (3.5 vs. 6.7 kg d–1). These results persisted over and between fishing seasons, but late in the season damage to the catches was common also in traps with AHDs. This study shows that the AHD may be a complementary mitigation tool in the seal–fishery conflict in certain types of fisheries, even though it is technically demanding, and for environmental reasons should be used with great care.
Keywords: conflict, damage, fishery, grey seals, mitigation, salmon
Received 18 December 2005; accepted 28 June 2006.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionConclusionsReferences |
|---|
In the Baltic Sea, a rapidly increasing population of grey seals (Halichoerus grypus) causes catch losses and gear damage in coastal static gear fisheries for salmonids (Fjälling, 2005; Westerberg et al., 2006). The damage is of such a magnitude that it threatens the survival and sustainability of many small-scale fishery operations of great local economic importance (Westerberg et al., 2000). Many tests of mitigation methods, with varying success rates, have been made to try to find solutions to these problems (Lunneryd et al., 2003; Westerberg et al., 2006). Here we focus on the results of displacing seals away from fishing traps using sounds of sufficiently high intensity to cause discomfort or pain and hence leading to avoidance. The equipment used is known as an acoustic harassment device (AHD).
Commercially available AHDs operate in a frequency range of 11–17 kHz and with source levels of 187–195 dB re 1 µPa rms at 1 m (Richardson et al., 1995). Previous studies of the effects of AHDs have focused on the acoustic output and possible impact on the hearing and behaviour of marine mammals (Mate and Harvey, 1986; Richardson et al., 1995; Johnston and Woodley, 1998; Oleisuk et al., 2002). The only long-lasting effects from AHDs that have been scientifically documented are observations of a diminishing effect over time, often attributed to the seals becoming used to the sound (Jefferson and Curry, 1995; Richardson et al., 1995). Still, many aquaculture farmers in Europe and Canada do use AHDs to protect their operations against seals (Taylor et al., 1997; Johnston and Woodley, 1998), indicating that there may be a long-lasting protective effect of such devices at least in some cases. Also, Yurk and Trites (2000) successfully scared harbour seals (Phoca vitulina) away from a river using AHDs, and the effect seemed not to have decreased after a two-week observation period. However, the effects of AHDs upon catch loss and gear damage in fisheries have previously been studied only on a relatively short-term basis (Mate and Harvey, 1986). Instant and time-dependent effects may differ considerably between the type of AHD, fishing gear, seal species, and many other factors.
This study aimed to measure the long-term effectiveness of AHDs in the Baltic salmon-trap, net fishery. A series of field trials was carried out in the northern part of the Baltic Sea. Catch and gear damage data collected by fishers were used to evaluate the effects of the AHD. We also present modifications of an existing AHD model, aimed to solve the high electric power demand problems that face AHD installations at remote fishing gear. Finally, we attempt to evaluate the economic benefits of using AHDs in this fishery.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionConclusionsReferences |
|---|
Technical specifications and modifications of the AHD
The AHD used in the tests was manufactured by LofiTech AS (Tromsø, Norway) and distributed by SIMRAD under the trade name "Fishguard". The electronic circuitry generates 250–500-ms-long, sinusoid signals of around 15 kHz, with a source level of 179 dB re 1 µPa rms at 1 m (Figure 1).
|
Originally, the unit produced 500-ms pulses that were semi-randomized within groups of a few seconds duration, with a fixed interval between groups of around 90 s. Hardware modifications were made to all units to improve the randomization of pulses and intervals, to shorten the pulse length, and to distribute pulses more evenly over time. The first modification was made to make it difficult for seals to anticipate the arrival of a pulse train and the last to decrease the energy consumption of the AHDs without affecting how the seals would perceive the loudness of the signal. The energy integration time of a seal's ear is most likely to be <100 ms (see the discussion in Richardson et al., 1995), so reducing the signal pulse length from 500 to 200 ms should not affect the perceived intensity of the signal by the seal.
An unmodified AHD of this model uses on average 2.3 A at 12 V DC input when emitting 17 pulses per 100-s cycle. This equals a daily energy consumption of 55 Ah, so a 70-Ah car battery, allowing a maximum of 30% discharge per cycle before life expectancy is reduced, may last in theory approximately 9 h. The high energy consumption while producing sounds (28 A, equalling 300 W) causes losses through the internal resistance in the batteries, leading to a 
5% increase in energy consumption. When a moderately discharged battery is recharged, the efficiency stays high at about 90%. The efficiency is lower, about 60%, when a well discharged battery is recharged. We developed a system for recharging the batteries as often as possible using different kinds of charging systems, e.g. solar panels and wind generators. We also supplied replacement battery packages for fishers, and in a few instances a cable from shore was used to power the AHD. The standard car batteries were later replaced by 105-Ah heavy-duty batteries. With the shortened sound pulses of the modified AHD, these batteries theoretically allowed for 24–36 h of operation before recharging. The AHDs were running 24 h d–1 except when batteries were changed or there were technical problems.
The AHDs, equipped with solar panels or wind generators, were mounted on a triangular raft (Figure 2a). Those with replacement batteries and the units with mains connections from shore were each mounted in a plastic barrel, which in turn was fitted into a floating ring consisting of a truck tyre filled with two-component polyurethane foam (Figure 2b). The transducers were lowered to about 5 m deep and positioned at the side of the trap near the fish chamber (Figure 3a).
|
|
The type of fish trap used in the study (a "combi trap") is basically a Scottish salmon trap with reduced mesh size (Figure 3a). This is the standard fishing trap for salmonids and coregonids in the region where these tests were conducted (Anon., 2001; Lehtonen and Suuronen, 2004). A leader net several hundred metres long starts close to shore and runs perpendicular to the coast, ending with large funnel-shaped nets leading into the fish chamber. The dimensions of this part of the trap vary, but are usually about 10 x 5 x 5 m. As the fish chamber is situated several hundreds of metres from the shore, the AHD cannot easily be powered by a cable from land. We therefore decided to use a self-contained raft with batteries and other power sources onboard.
Data collection and processing
To test the AHDs in the Baltic salmon-trap, net fishery, eight commercial fishers were engaged (Table 1). Their commitment included keeping detailed records of catches and seal damages to fish and gear, as well as daily equipment maintenance. At each emptying of a trap, the catch was recorded both as the number of fish and as the total weight of each species. The number of damaged fish was determined and judged as being the result of seal or bird activity. Large fish (salmon, Salmo salar, or large sea trout, Salmo trutta) bitten by seals typically lose large parts of the body, whereas smaller fish species are usually consumed whole. The skin of larger fish is often peeled off, and in many cases only the head and the spine remain. Fish damaged by seabirds (mostly whitefish, Coregonus spp., and small sea trout) are usually pecked in the neck. The number, size (<100 mm, >100 mm, large tears), and position of mechanical damage to the gear were also recorded. Controls on data quality were made in the field (parallel protocols) by research personnel, and during analysis by statistical means (frequency distribution of first and last digits tested for deviations from Benford's law and randomness; Hill, 1998). All data were entered into a logbook (described in Lunneryd et al., 2005).
|
Up to nine AHDs were deployed per year during the test period from 1998 to 2001 in the northern Baltic Sea (Figure 3b). Not all sites were the same in all of the years; there were 19 usable data sets in total (Table 1). Several that did not meet the criteria below were discarded. Data series (one trap during one season) were deemed usable and were used for analysis if (i) they contained both days with AHDs in operation as well as days without, and (ii) neither group represented more than about 80% of the days in that data series. This was done to cancel out systematic errors from the choice of site for AHDs, and to generate data from the same traps both with and without AHDs. The AHDs were assumed to be in operation as long as there was no direct evidence of a technical failure (i.e. when no signal could be detected, or when the signal was very weak).
A trap lift was defined as being disturbed by seals if damaged fish or the remains of fish were observed in the fish chamber or elsewhere in the trap, or if there were new holes or tears in the trap. Each lifting of a trap was classified as belonging to one of four categories, and assigned a corresponding catch code:
- Code A – lift only containing intact fish and having no damage to the gear;
- Code B – lift with some intact fish but also with seal damage to fish or gear;
- Code C – lift with no intact fish and with seal damage to fish or gear; and
- Code D – empty traps with no gear damage.
- Code B – lift with some intact fish but also with seal damage to fish or gear;
The estimated losses (weights) were calculated per trap lift from the observed number of damaged fish multiplied by the average weights. For salmon, the mean weight per week was used because there was a negative trend in weight over the season (weight = –0.49 x week number + 20; r2 = 0.29). For sea trout and whitefish, there was no trend in the weight of the fish caught during the season. As there was a regional variation, at least for whitefish, the losses were calculated from the average weights for each data set, or for the whole season if required.
The individual weights for salmon and sea trout were estimated from the total weights for that species for a specific day, divided by the number of fish. Individual weights for whitefish were not calculated or used because the numbers caught were uncertain since fishers often estimated the number of whitefish from the weight of the catch.
Fishing effort was defined as the number of days from one lift to the next. When the number of lifts per day exceeded one, fishing effort was adjusted accordingly, assuming that the intervals between lifts were evenly distributed during that day.
A Mann–Whitney U-test was used to test catches and damages statistically. Differences in frequencies (i.e. for catch codes) were tested using a 
2 test. Catch figures were not normally distributed, so a bootstrap resampling procedure (1000 repetitions) using bias-corrected confidence intervals (Haddon, 2001) was used.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionConclusionsReferences |
|---|
The fishing effort (trap-days) for traps with an AHD in operation was 600 (755 lifts). The number without an AHD was 406. During another 87 trap-days the AHD did not function, and because the mean catch and damage for these days did not differ significantly from days without AHDs, they were included in the latter group (altogether 493 trap-days, 668 lifts). The average duration of a fishing event was 0.79 days for traps with an AHD, and 0.72 days for traps without an AHD. The 19 data series were distributed in time according to Figure 4.
|
Effect of AHDs on the size of landed and damaged catches
The mean daily catch landed intact was significantly higher for traps with an AHD (25.5 kg d–1, n = 755) than for traps without one (12.0 kg d–1, n = 668, p < 0.001; Table 2). The traps with AHDs had fewer lifts with zero catches and more with very large catches. The mean quantity of damaged fish was less for traps with AHDs (3.5 kg d–1) than for traps without one (6.7 kg d–1, p < 0.01). The traps with AHDs had more lifts with zero or only slight damage, and fewer with a lot of damage (Table 3).
|
|
The distribution of catch codes differed between traps with and without AHDs. The proportion of undisturbed lifts with unharmed catch (code A) was higher in traps with an AHD (44% vs. 24%). The proportion of disturbed lifts that yielded at least some unharmed catch as well as damaged fish (code B) was also higher (30% vs. 23%). The proportion of disturbed lifts that did not have any unharmed catch (code C) was lower (9% vs. 18%) in traps with an AHD. The proportion of lifts that had no catch, but also no signs of seal visits (code D), was also lower (17% vs. 34%). All these differences were statistically significant (
2 test, p < 0.001). The frequency of holes in net panels was significantly lower in traps with AHDs than in those without (p < 0.001); in terms of other damage, there was no difference between traps with and without AHDs (Table 4).
|
Changes in the effect of an AHD throughout the fishing season
The mean landed catch in traps with an AHD was higher throughout the season, and increased over time (Figure 5a). Fishing was continued for 2–3 weeks longer in autumn in traps with AHDs. At the beginning of the season, the quantity of damaged fish was much higher in traps without an AHD. In the middle of the fishing season, the quantity of damaged fish was roughly equal between traps, and in the late season it remained high in traps with an AHD (Figure 5b). Most traps without an AHD had by then been brought ashore because of the intensity of the seal attacks and small size of the landed catches.
|
AHD efficiency between consecutive years
Traps with an AHD caught more fish in all three years (U-test, p < 0.05, p < 0.001, p < 0.001), and damage was less (U-test, p < 0.001, p = 0.15, p < 0.01; Figure 6a, b).
|
Losses to seabirds
The numbers of bird-damaged salmon and sea trout were lower in traps with AHDs, but not significantly so (0.002 per unit of effort vs. 0.05, p = 0.098, p = 0.3), but the number of bird-damaged whitefish was significantly higher in traps with AHDs (1.04 vs. 0.82, p = 0.002).
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionConclusionsReferences |
|---|
AHDs had a positive effect in the Baltic salmon-trap, net fishery by way of larger landed catches (less damaged) and less gear damage. It is also important to consider whether the positive effects exceed the economic costs. The initial investment in an AHD is around
3500. The value of the catch per day was calculated from a minimum first-hand sale value of 4.30 per kg for salmon and sea trout, and 3 per kg for whitefish (Table 5). With an estimated fishing season of 55 days, the mean duration of the fishing season for the traps in the study, the gross return for a trap without an AHD would be about 2600, and for a trap with an AHD, 5200. The payback time for an AHD should then be less than two years. This crude balance does not take into consideration either the costs of AHD maintenance or the costs of the extra work operating the AHD, or the reduced work costs in repairing gear. Also, there may be an extra gain not considered here if the AHD allows a longer fishing season and less wear to the fishing gear.
|
Quality controls indicated that the data recorded by fishers were reliable. In addition, the fishers had a great interest in the results of the investigation and were highly motivated in their efforts. On the other hand, the experimental design may have had other consequences on the results. One is that some AHDs were moved back and forth between traps in the same area. The seals therefore usually had an alternative to the trap with the AHD. The AHD-induced differences in catch, quantity of damage to fish, and damage to gear might therefore have been larger than if the traps had been in separate areas. Some fishers not using AHDs in the neighbourhood of ongoing trials claimed that seal damage increased during the trials. The "protective effect" of AHDs (53% reduction in catch when traps without AHDs are used; Table 2) is similar to the estimated total loss when seals visit salmon traps (57%; Fjälling, 2005). In addition to the directly measurable losses, there may be hidden losses, obvious neither to the fisher nor to the observer. Such hidden losses could be caused by seals taking fish from the trap whole, without leaving any sign of their visit (Fjälling, 2005).
The study also showed variations in the behaviour of seals. Seals rapidly become accustomed to sound sources, even those at very high intensity (Jefferson and Curry, 1995; Richardson et al., 1995). During the present trials, fishers reported that some seals seemed to learn to utilize the 1.5-min silent interval at those AHDs having the original pulse pattern, apparently keeping their heads above the water while the AHD was pinging, then diving into the traps during the silent interval. Knowledge of this behaviour was put to use by some fishers, who shot seals when they exposed their heads during the AHD ping.
Sometimes, individual seals, especially very large, old males, did not respond to the AHDs. This could be because they were used to the sound (Jefferson and Curry, 1995), or perhaps to age-related hearing impairment (Shaughnessy et al., 1981; Schusterman et al., 2002). There is also a risk of causing a permanent hearing-threshold shift (PTS) by prolonged exposure to the sound of an AHD. The AHD sounds used in this project are approximately 110 dB above the auditory threshold of earless seals in the frequency range of the signal (Richardson et al., 1995). There are no data on the noise level needed to induce PTS in marine mammals; it may be a function of the intensity of the received sound, and also of its duration as well as how often it is repeated (Madsen, 2005). There is a possibility that positive interference in sound fields creates local areas of very high sound pressure. Field studies are needed to evaluate these risks, through detailed measurements of the sound fields near AHDs and of how seals are exposed to them. Such studies should take the seals' motivation into consideration. The sound level might be high enough to induce discomfort and avoidance at distances on the order of 10–100 m from the device, but this may be counteracted by the seals' own great motivation to eat the fish in the trap. Therefore, an absolute avoidance distance probably does not exist; see the discussion in Jacobs and Terhune (2002), and Terhune et al. (2002).
Other studies (e.g. Johnston and Woodley, 1998) have raised concerns on how AHDs around aquaculture operations may affect the behaviour of co-existing cetaceans, such as harbour porpoises. As cetaceans only rarely appear along the northern Baltic coast, and given that the high frequency sounds emitted have limited range, unwanted effects on other animals should be limited. Nevertheless, the increased underwater noise levels resulting if AHDs were to be used extensively, could negatively influence wildlife in areas where animals that can hear the signals are present. To date, there are to our knowledge only two studies on how AHD signals propagate in the coastal environment: those of Jacobs and Terhune (2002), and Terhune et al. (2002). More data are needed to evaluate the possible adverse effects of extensive AHD usage.
Fishers were instructed to deploy the AHDs a few weeks before setting the traps, but this did not happen. The AHDs were actually deployed some weeks after the traps were set, when seal visits had already become a problem. The initial AHD-free period perhaps allowed the seals' motivation to forage at traps to become unwittingly high, perhaps accelerating their familiarity with the sound.
The fact that seal activity around fishing gear often goes undetected was confirmed by underwater video in a previous study (Lunneryd et al., 2003), and by other analyses of the seal–fisheries interaction (Fjälling, 2005). Days with no or little catch and without signs of seals may be interpreted as poor fishing periods by fishers. The presence of seals taking fish from the traps or disturbing the catching process without their being detected by fishers is supported by the data collected here. It is also supported by a nearly significant tendency that traps with AHDs had better catches during days with no signs of seal visits than did non-disturbed traps without AHDs (p = 0.06, Table 2).
The differences in catch damage caused by seabirds in traps with and without AHDs are more difficult to explain. Owing to the way their middle ear is constructed, all the seabirds tested to date cannot hear above 12 kHz in air (Dooling et al., 2000). Very little is known about the ability of seabirds to hear underwater. Even though the energy spectrum of the AHD signals used here is centred at around 15 kHz, it contains frequency components down to 7 kHz (Figure 1), which may be audible to seabirds underwater. Hence, assuming that diving seabirds may hear and be deterred by the sounds from the AHD, the difference in losses to seabirds can possibly be explained by how seabirds get to the fish. Whitefish often try to pass over the cork leader of the net at the surface of the water, where they may be taken by seabirds without diving, but salmon and sea trout only rarely come so close to the surface. For those fish, therefore, seabirds must dive deeper and perhaps then expose themselves fully to the sound.
The use of AHDs has not become widespread in Swedish set-trap fisheries for salmonids despite the good results demonstrated here. This is probably due to the successful development of a seal-safe salmon trap (Lunneryd et al., 2003) and to technical imperfections in available AHDs. There are, however, two recent developments where the AHDs may offer a solution. The first is the increasing number of reports of seals taking up residence in the entrance of the seal-safe trap, and disturbing the catch process by acting as "goalkeepers". An AHD positioned in an appropriate place should limit this behaviour. The second development is that there are positive reports of AHDs being used in combination with removing those seals that have become used to them. One fisher who culled 11 grey seals, most of them large males, at a severely disturbed salmon trap in a river mouth, experienced very little seal interference (Sjöström, pers. comm.) for the next 3–4 years. At Ballard Locks, Washington, an AHD was not able to deter pinnipeds from feeding on migrating salmon near a fish ladder, but hindered the recruitment of new pinnipeds (NOAA, 1997). The subsequent removal of habituated pinnipeds resulted in little or no predation taking place (London et al., 2002).
|
Conclusions |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionConclusionsReferences |
|---|
- The AHDs had a positive effect in the Baltic salmon-trap, net fishery in terms of bigger catches and lessened damage to catch and gear.
- The effect was stable over time, over both season and three consecutive years.
- The positive effect was heavily dependent on careful deployment and maintenance of the equipment. Further technical development is needed if AHDs are to be a realistic and sustainable option for these fisheries.
- As there are no cetaceans in the northern Baltic, the non-target environmental impact of AHDs is probably limited.
- Measurements of the sound propagation of AHD signals are needed to evaluate the effects on marine mammal hearing.
|
Acknowledgements |
|---|
This paper is based on the combined efforts of several fishers who attended the equipment doing what they possibly dislike most: completing forms. Without their patient work this study would not have been possible. Sven-Gunnar Lunneryd, Mats Amundin, and Jack Terhune made valuable comments on the manuscript. The project was financed by the Swedish Environmental Protection Agency and the Swedish Board of Fisheries. MW was funded by the EU 5th framework project FRAP.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionConclusionsReferences |
|---|
-
Anon. (2001) Småskaligt kustfiske och insjöfiske – en analys. 101-800-00, Fiskeriverket, Göteborg.
Dooling R.J., Lohr B., Dent M.L. (2000) Hearing in birds and reptiles. In Dooling R.J., Fay R.R., Popper A.N. (Eds.). Comparative Hearing: Birds and Reptiles(Springer, New York) pp. 308–359.
Fjälling A. (2005) The estimation of hidden seal-inflicted losses in the Baltic Sea set-trap salmon fisheries. ICES Journal of Marine Science 62:1630–1635.
Haddon M. (2001) Modelling and Quantitative Methods in Fisheries(Chapman and Hall/CRC Press, Boca Raton, FL).
Hill T.P. (1998) The first-digit phenomenon. American Scientist 86:358–363.[CrossRef][Web of Science]
Jacobs S.R. and Terhune J.R. (2002) The effectiveness of acoustic harassment devices in the Bay of Fundy, Canada: seal reactions and a noise exposure model. Aquatic Mammals 28:147–158.
Jefferson T.A. and Curry B.E. (1995) Acoustic methods of reducing or eliminating marine mammal–fishery interactions: do they work? Ocean and Coastal Management 1:41–70.
Johnston D.W. and Woodley T.H. (1998) A survey of acoustic harassment device (AHD) use in the Bay of Fundy, NB, Canada. Aquatic Mammals 24:51–61.
Lehtonen E. and Suuronen P. (2004) Mitigation of seal-induced damage in salmon and whitefish trapnet by modification of the fish bag. ICES Journal of Marine Science 61:1195–1200.
London J. M., Lance M. M., Jeffries S. J. (2002) Observations of harbor seal predation on hood canal salmonids from 1998 to 2000. Washington Department of Fish and Wildlife, Washington.
Lunneryd S.G., Fjälling A., Westerberg H. (2003) A large-mesh salmon trap; a way of mitigating seal impact on a coastal fishery. ICES Journal of Marine Science 60:1194–1199.
Lunneryd S-G., Hemmingsson M., Tärnlund S., Fjälling A. (2005) A voluntary logbook scheme as a method of monitoring the by-catch of seals in Swedish coastal fisheries. ICES Document CM 2005/X: 04. 10 pp.
Madsen P.T. (2005) Marine mammals and noise: problems with root mean square sound pressure levels for transients. Journal of the Acoustical Society of America 117:3952–3957.[CrossRef][Web of Science][Medline]
Acoustical Deterrents in Marine Mammal Conflicts with Fisheries Mate B. R. and Harvey J. T. (1986) Acoustical deterrents in marine mammal conflicts with fisheries. Oregon Sea Grant, Newport, Oregon. 116 pp.
NOAA. (1997) Impacts of California sea lions and Pacific harbor seals on salmonids and on the coastal ecosystems of Washington, Oregon, and California. NMFS-NWFSC-28, National Marine Fisheries Service, Seattle.
Oleisuk P.F., Nichol L., Sowden M.J., Ford J.K.B. (2002) Effect of the sounds generated by an acoustic deterrent device on the relative abundance and distribution of harbour porpoise (Phocoena phocoena) in Retreat Passage, British Columbia. Marine Mammal Science 18:843–862.[CrossRef][Web of Science]
Richardson W.J., Greene J.C.R., Malme C.I., Thomson D.H. (1995) Marine Mammals and Noise(Academic Press, San Diego).
Schusterman R.J., Southall B., Kastak D., Kastak C.R. (2002) Age-related hearing loss in sea lions and their scientists. Journal of the Acoustical Society of America 111:2342–2343.
Shaughnessy P.D., Semmelink A., Cooper J., Frost P.G.H. (1981) Attempts to develop acoustic methods of keeping Cape fur seals Arctocephalus pusillus from fishing nets. Biological Conservation 21:141–158.[CrossRef][Web of Science]
Taylor V.J., Johnston D.W., Verboom W.C. (1997) Acoustic harassment device (AHD) use in the aquaculture industry and implications for marine mammals. Proceedings of the Institute of Acoustics 19:267–275.
Terhune J.M., Hoover C.L., Jacobs S.R. (2002) Potential detection and deterrence ranges by harbour seals of underwater acoustic harassment devices (AHD) in the Bay of Fundy, Canada. Journal of the World Aquaculture Society 33:176–183.[CrossRef][Web of Science]
Westerberg H., Fjälling A., Martinsson A. (2000) Sälskador i det svenska fisket. Beskrivning och kostnadsberäkning baserad på loggbokstatistik och journalföring, 1996–1997. Fiskeriverket Rapport 3:4–38 (in Swedish with English summary).
Westerberg H., Lunneryd S.-G., Wahlberg M., Fjälling A. (2006) Reconciling fisheries activities with the conservation of seals through the development of new fishing gear: a case study from the Baltic fishery – grey seal conflict. Transactions of the American Fisheries Society (World Fisheries Congress Proceedings) pp. 587–597.
Yurk H. and Trites A.W. (2000) Experimental attempts to reduce predation by harbor seals on out-migrating juvenile salmonids. Transactions of the American Fisheries Society 129:1360–1366.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Vilata, D. Oliva, and M. Sepulveda The predation of farmed salmon by South American sea lions (Otaria flavescens) in southern Chile ICES J. Mar. Sci., November 6, 2009; (2009) fsp250v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. M. Graham, R. N. Harris, B. Denny, D. Fowden, and D. Pullan Testing the effectiveness of an acoustic deterrent device for excluding seals from Atlantic salmon rivers in Scotland ICES J. Mar. Sci., June 1, 2009; 66(5): 860 - 864. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||