REVIEW AND SYNTHESIS OF THE ENVIRONMENTAL IMPACTS OF AQUACULTURE
CHAPTER THREE EFFECTS OF DISCHARGES OF MEDICINES AND CHEMICALS FROM AQUACULTURE
SEA LICE MEDICINES
3.1 A recent review of the availability and use of chemotherapeutic sea lice control products identified eleven compounds representing five pesticide types being used internationally on commercial salmon farms in the period 1997 to 1998. These included two organophosphates (dichlorvos, azamethiphos); three pyrethrin/pyrethroid compounds (pyrethrum, cypermethrin and deltamethrin); one oxidizing agent (hydrogen peroxide); three avermectins (ivermectin, emamectin and doramectin) and two benzoylphenyl ureas (teflubenzuron, diflubenzuron). Of these, six compounds were available for use in the UK (dichlorvos, azamethiphos, cypermethrin, hydrogen peroxide, ivermectin and emamectin). Dichlorvos and ivermectin are not known to be used in Scotland, and hydrogen peroxide, which degrades rapidly to water and oxygen, is not considered to be a hazard to marine life.
3.2 This report concentrates on four compounds currently licensed for use as sea lice medicines in Scotland: the bath treatments, azamethiphos and cypermethrin; and the in-feed treatments, emamectin benzoate and teflubenzuron. Of these, cypermethrin and emamectin benzoate are most widely used and are, therefore, considered to present the greatest environmental risk. Bath treatments involve the discharge of dissolved medicine into the water column after the treatment period. In-feed treatments are ingested by the fish and then excreted over a period of time with most of the losses occurring to the sediments rather than the water column.
3.3 Currently azamethiphos use for sea lice control on salmon farms is limited and will probably continue to decline as the use of in-feed treatments increases. At present, azamethiphos is most often used in conjunction with cypermethrin treatments when lice numbers necessitate control measures but farms have reached their discharge consent limits for cypermethrin. Field studies in Scotland using deployed mussels and lobster larvae indicate that effects on marine organisms in the vicinity of treated cages are unlikely. A dispersion and toxicity study undertaken in the Lower Bay of Fundy, New Brunswick, at sites displaying a range of dispersive energy conditions concluded that azamethiphos presented a low to moderate environmental risk. The risks of short or long term adverse environmental effects resulting from the use of azamethiphos for sea lice control are considered to be low as toxicity values are well above both concentrations predicted following sea lice treatments and Environmental Quality Standards (EQS).
3.4 Cypermethrin is widely used for sea lice control in Scotland and a considerable amount of information is available on its dispersion, fate and ecotoxicity. Dispersion modelling and field based studies focussing on single treatments indicate that cypermethrin released following a bath treatment will be rapidly diluted in the receiving environment, with the majority adsorbed onto particulate material, which settles to the sea bed. This absorption process takes several hours by which time the discharge plume is spread over a wide area. Sediment concentrations are, therefore, generally so low as to be undetectable. Both water column and sediment cypermethrin concentrations predicted following single releases are lower than Environmental Quality Standards (EQS), and are therefore unlikely to result in toxic effects. However, a recent study concluded that even a single cage application of cypermethrin has the potential to create a plume of up to 1 km 2 that may retain its toxicity for several hours. In that study, water samples collected up to 5 hours post-treatment were toxic to the benthic amphipod, Eohaustorius estuarius, causing immobilisation during 48 hour exposures. This has potential ecological implications because, in reality, cypermethrin treatments involve multiple releases daily, usually over several consecutive days. The potential for cypermethrin concentrations to exceed water and sediment EQS is, therefore, increased during multiple treatment events. Consequently the environmental risk associated with cypermethrin use is greater. SEPA account for this by setting 3 hour and 24 hour EQSs. The dispersion, fate and cumulative effects of multiple treatment releases on the marine environment remain unknown and require further investigation.
3.5 Sediment associated organisms are most likely to be affected by cypermethrin as it binds strongly to organic particles and solids, and is rapidly adsorbed by sediments. Such particle binding ameliorates toxicity by reducing bioavailability. For example, the tissue concentrations of cypermethrin in Daphnia have been examined as a proportion of sediment concentration and were found to decrease with increasing organic carbon content, indicating decreases in bioavailability. This was a freshwater study, but has implications for the organically enriched sediments below fish farm cages in terms of cypermethrin bioavailability and toxicity to benthic invertebrates.
Emamectin benzoate (Slice)
3.6 Emamectin benzoate use for sea lice control is increasing in Scotland and, in many loch systems, strategic treatments are being undertaken simultaneously at several farm sites. There is very little information available on the environmental fate and ecological effects of emamectin benzoate in the marine environment.
3.7 The organisms most likely to be affected by emamectin benzoate are those closely associated with the sediment as emamectin has low water solubility and a high potential to be adsorbed and bound to suspended particulate material. Much of the emamectin reaching the sediments will be associated with particulate material in the form of fish faeces and uneaten fish food. Emamectin remains in the sediments for a considerable period of time having a half life ( i.e. the time taken for the concentration to diminish by 50%) of around 175 days.
3.8 Benthic communities in the organically enriched sediments below fish farm cages are generally dominated by small worms, which play a vital role in remineralising waste products. A recent study on the effects of emamectin benzoate on infaunal polychaetes indicated that predicted sediment concentrations are unlikely to adversely affect polychaete communities below fish farm cages. Sediment emamectin concentrations causing significant mortality to the capitellid worms that typically dominate sediments beneath fish farms were also considerably higher than the EQS.
3.9 Emamectin benzoate water column concentrations are expected to be considerably lower than sediment concentrations and are unlikely to pose a risk to planktonic organisms. Results from laboratory toxicity tests support this conclusion, with acute toxicity values orders of magnitude higher than the maximum allowable water concentration of 0.22 ng L -1.
3.10 The environmental risk of emamectin benzoate to the marine environment is considered to be low to moderate. However, there is relatively little information available on the toxicity of this chemical to marine benthic invertebrates in particular, and little is known about the potential long-term impacts of this chemical on the marine environment.
3.11 Discharge consents are being granted for the use of teflubenzuron as a sea lice medicine in Scotland, but it is not being widely used, primarily because it is not effective against adult sea lice. There is very little information available on the environmental fate and ecological effects of teflubenzuron in aquatic environments. The specific mode of action of teflubenzuron means it is highly toxic to aquatic crustacean invertebrates, but low in toxicity to fish, mammals and birds. As with emamectin benzoate, it is likely that the sediments will act as a sink for teflubenzuron and so sediment associated organisms are more likely to be affected by this chemical.
3.12 To our knowledge, there are no data on the toxicity of teflubenzuron to marine invertebrates in the published literature and the suitability of sediment quality standards in particular, are unknown. A recent study, investigating the toxicity of sea lice chemotherapeutants to non-target planktonic copepods, determined acute toxicity values for planktonic marine copepods exposed to teflubenzuron that are orders of magnitude higher than water column EQS.
3.13 Teflubenzuron is predicted to be only directly toxic to crustacean invertebrates in marine ecosystems. However, the potential exists for indirect effects such as increases in primary productivity and changes further up the food chain. Direct and indirect ecosystem-level effects of the structurally similar benzoylurea insecticide, diflubenzuron, have been observed in freshwater mesocosms in the USA. Monthly and bimonthly applications of 10 µg L -1 diflubenzuron reduced zooplankton abundance and species richness, causing algal biomass to increase because of decreases in invertebrate grazing. Significant declines were also observed in juvenile bluegill biomass and individual weight, probably because of decreases in invertebrate food resources.
3.14 It is difficult to predict the ecological risk of teflubenzuron to the marine environment because of the current lack of information. Results from field studies referred to in SEPA's environmental risk assessment suggest that the use of teflubenzuron for sea lice control may present a moderate to high environmental risk. It seems unlikely that teflubenzuron will be widely used for sea lice control in Scotland, but if use does increase, investigation into the potential long-term impacts of this chemical on the marine environment is recommended.
3.15 Antimicrobial compounds such as oxytetracycline, oxolinic acid, trimethoprim, sulphadiazine and amoxycillin are administered to farmed salmon as feed additives to treat bacterial infections. In general, salmon farming is one of the least medicated forms of agriculture; compared with factory beef, poultry and pork production, antibiotic usage in fish farms is small and continues to decline. Antibiotics are not used on a continual, long-term basis as they often are in other types of animal husbandry. Rather, they are used intermittently for short periods (5 to 14 days) to control outbreaks of disease.
3.16 Most antimicrobial compounds readily associate with particulate material and residues are often found in the organically enriched sediments below farms that have treated fish with antibiotics, although the area of sediments containing measurable residues is generally very localised.
3.17 Concerns relating specifically to antibiotic usage by the aquaculture industry are:
- Development of drug resistance in fish pathogens
- Spread of drug resistant plasmids to human pathogens
- Transfer of resistant pathogens from fish farming to humans
- Presence of antibiotics in wild fish
- Impact of antibiotics in sediments on: rates of microbial processes; composition of bacterial populations; relative size of resistant sub-populations.
3.18 The environmental risk of antimicrobial compounds used by the aquaculture industry is considered to be very low. Antibiotic usage in aquaculture is insignificant compared with agricultural use and, because of the development of vaccines, continues to decline.
3.19 Of the metals present in fish farm sediments, elevated concentrations of copper and zinc have been reported in Scotland and Canada. The principal sources of these metals are antifoulant paints and fish feed.
Metals in antifoulants
3.20 Antifoulant products are painted or washed onto fish farm nets and structures to slow the build up of fouling organisms. Currently, 19 of the 24 antifoulant products registered for use in Scottish aquaculture are copper based, either as copper, copper oxide or copper sulphate. These copper-based products exhibit effective antifouling activity against barnacles, tube worms and most algal fouling species. Two types of antifoulant paint (water based or spirit based) may be applied to fish farm nets at washing sites remote from the farm. When the nets are placed back in the water at the farm, copper can be released from the paints, producing metallic slicks. It is likely that copper is also released in soluble and particulate form from paint on the metal cage structures.
3.21 The use of copper-based antifoulants is likely to increase and there may be reason for concern because of the accumulation of copper in sediments below fish farms, and its potential toxicity to benthic organisms.
Metals in fish feed
3.22 Metals present in fish feed are either constituents of the meal from which the diet is manufactured or are supplemented as a mineral pre-mix for perceived nutritional requirements. The meal constituents, together with the mineral pre-mixes, are composed of various trace and heavy metals, providing copper, zinc, iron, manganese, as well as cobalt, arsenic, cadmium, fluorine, lead, magnesium, selenium and mercury. Concentrations of copper and zinc in feeds produced for Atlantic salmon range from 3.5 to 25 mg Cu kg -1 and 68 to 240 mg Zn kg -1. However, the estimated dietary requirements of Atlantic salmon for these elements are 5 to 10 mg Cu kg -1, and 37 to 67 mg Zn kg -1. Therefore, it would appear that the metal concentrations in some feeds are unnecessarily high as they exceed salmon dietary requirements.
3.23 Sediment copper and zinc concentrations measured at fish farms surveyed by the SEPA West Region in 1996 and 1997 were compared with proposed sediment quality criteria to assess the potential for adverse effects caused by elevated metal concentrations. Sediments directly beneath the cages and within 30 m of the farms were severely contaminated by copper and zinc at 7 of the 10 farms surveyed, with "probable" adverse effects predicted on the benthic invertebrate community at these sites.
3.24 The long-term ecological implications of high metal concentrations in fish farm sediment are unknown. Sediment biogeochemistry and physical characteristics influence the accumulation, availability and toxicity of sediment contaminants such as trace metals to benthic invertebrates. Even when metal concentrations in sediments substantially exceed background levels, metal bioavailability may be minimal and adverse impacts may not occur. Organically enriched fish farm sediments characteristically have a high biological oxygen demand and negative redox potential; conditions leading to sulphate reduction. Under these conditions, metals such as copper and zinc are less likely to be biologically available. However, disturbance of the sediments by strong currents or by trawling could cause the sediments to be redistributed into the water column, leading to re-mobilisation of the metals so that they become available for uptake. It is possible that elevated copper and zinc concentrations, in combination with high levels of other potentially toxic substances such as sulphides and ammonia, could represent a significant barrier to the recolonisation of benthic sediments when fish farm sites are fallowed. Sediment chemical remediation when a fish farm site is fallowed, in particular, degradation of organic material and reductions in sulphide concentrations, may increase metal bioavailability in the sediments, and might also result in the release and further dispersal of metals away from fish farm sites.
3.25 There is currently insufficient information available to determine the long-term effects of medicine and antifoulants use. Further research is required into the effects of these products over the long term, particularly where multiple sources enter the same marine area. In the short term, the environmental risk is considered to be low if sea lice medicines and antifoulants are used according to regulatory guidelines but, in the case of antifoulants, more information is required relating to their use by the aquaculture industry.
3.26 The following concerns and areas for future research relating to these chemicals and their potential environmental impacts have been identified:
- More information is required on the toxicity of emamectin benzoate, teflubenzuron, copper and zinc to benthic organisms commonly found Scottish sea lochs.
- More information is required on the long-term effects of cypermethrin, emamectin benzoate, copper and zinc on sediment associated organisms. In particular:
- What proportion of the chemicals, particularly the metals, present in fish farm sediments is bioavailable?
- Is there potential for these chemicals, particularly the metals, to accumulate up the food chain.
- What happens when a site is fallowed and the sediment biogeochemistry changes?
- Do the chemicals that have accumulated, and are possibly not biologically available in the organically enriched sediment, become bioavailable as chemical remediation occurs? Are they released, and do they disperse over a wider area? Do they prevent recolonisation of impacted sites?
- More information is required on the dispersion, fate, and potential long-term effects of multiple cypermethrin treatments (at single and multiple farm sites) within a loch system.
- More information is required on the potential effects of concurrent emamectin treatments at several farm sites within a loch system.
- Antifoulant usage by the aquaculture industry should be quantified.
- Copper and zinc concentrations, speciation, and toxicity in fish farm sediments needs to be investigated.
- Better understanding of salmon metal dietary requirements is needed to reduce metal concentrations in feed and consequent metal input into the marine environment.