Introduction

The abundance of wild salmon has declined dramatically since the 1980s (Kellogg 1999, Noakes, Beamish et al. 2000, Torrissen, Jones et al. 2013, NASCO 2019). This global decline is due mainly to poor marine survival (ICES 2016) caused by the complex interaction of anthropogenic, natural biological, and physical factors, including sea-temperature change, pollution, susceptibility to predators, and post-smolt growth (Fig. 1) (Friedland, Reddin et al. 1993, Parrish, Behnke et al. 1998, Jonsson and Jonsson 2004, Friedland, MacLean et al. 2009, Chaput 2012, Friedland, Shank et al. 2014, Vollset, Krontveit et al. 2015).

Fig. 1
figure 1

Decline in salmon abundance, 1983–2016 (NASCO 2019)

Full size image

A growing body of scientific evidence indicates that the sea-lice parasite Lepeophtheirus salmonis, which thrives in salmon farms and spreads to infest surrounding salmon populations, can significantly aggravate marine mortality among wild salmon (Hilborn 2006, Costello 2009, Krkosek, Connors et al. 2011, Vollset, Krontveit et al. 2015, Johnsen, Harvey et al. 2021, Stige, Helgesen et al. 2022). However, the precise extent to which sea-lice infestations reduce adult salmon returns has been subject to scientific debate. Some studies find that farm sea lice represent a minor and irregular component of marine mortality, with insignificant population-level effects (Marty, Saksida et al. 2010, Jackson, Cotter et al. 2011, Jackson, Cotter et al. 2013a, Jackson, Kane et al. 2013b, Skilbrei, Finstad et al. 2013). However, others, using similar data, find that the presence of sea lice fuels the decline in populations (Otero, Jensen et al. 2011, Gargan, Forde et al. 2012, Krkosek, Revie et al. 2013, Shepard and Gargan 2017, Shepard and Gargan 2021).

This lack of scientific consensus around the scale of farm-lice impacts has been accompanied by political controversy around what policies are needed to protect wild salmons adequately. This is a common feature of “wicked” policy problems, which are often riddled with scientific uncertainties that fuel disagreement as to appropriate regulatory responses (Rittel and Webber 1973, Osmundsen, Almklow et al. 2017). Scientific uncertainty often spurs conflict between institutional and political players with differing values, interests, and policy preferences (Sarewitz 2004, Hoppe 2005, Lackey 2007). In the case of salmon aquaculture, disputes around the need for stricter sea-lice regulation have emerged between, on the one hand, researchers, stakeholders, and government actors that believe the available scientific knowledge on lice-induced health hazards and mortality risks for wild salmons justifies more stringent regulations, and, on the other hand, those who stress the lack of a statistically proven association between lice infestations and population declines, and thus dispute the need for reform.

This study examines the role of science in shaping aquaculture policymaking towards enhanced protection of wild salmon. It does so by conducting a comparison of sea-lice regulation in four major salmon-farming jurisdictions that have a stated responsibility to protect wild populations: Norway, Canada, Scotland, and Ireland. The article first provides an overview of sea-lice regulations in each country, focusing on farm sea-lice limits or management thresholds during periods of out-migration for juvenile salmons, the rules of threshold enforcement, and reporting and public disclosure of on-farm lice levels, and new, area-based management systems that regulate farm biomass (production volumes) based on estimated mortality risks for wild salmon. Second, it considers whether the regulations have been designed and revised to provide improved protection of wild salmons. Third, it assesses how variations in regulatory approaches relate to differences in the governments’ interpretation of the scientific basis. Notwithstanding other factors that may contribute to explaining whether or not a country has significantly revised its on-farm thresholds or adopted new regimes for area-based sea lice management—including institutional, political, cultural, or ecological factors—the article finds that the existence of consensus or controversy around the science on farm–wild interactions has shaped regulatory actions toward reform. In Norway and Scotland, the emergence of a near-consensus around the science that justifies policy change has enabled reformist political forces to adopt and implement more precautionary and strict regulation to protect wild salmon. By contrast, the persistence of controversy in Canada and Ireland appears to have hindered reformists in their calls for change and led to a lack of political commitment. The emergence of scientific agreement within the government may be an important prerequisite for significant policy reforms related to farm–wild interactions.

Methodology and data collection

The research employs a qualitative, case-study approach. The cases were selected on the basis of two criteria: that the jurisdictions have a national salmon-farming industry and have native, wild salmon populations for which they have a stated responsibility to protect. These criteria excluded Chile, a large salmon producer with no wild salmons, and the Faroe Islands, a substantial producer but with no native wild salmons—only a small population of native trout. As for Canada, the Atlantic East Coast region was excluded due problems of access: aquaculture and sea-lice regulations are worked out on an individual basis through farm management plans by the province-level government. Such farm management plans are private, protected by corporate confidentiality agreements, which hindered efforts to assess their regulatory schemes for this study.

The research draws upon a range of written sources, including peer-reviewed literature, government reports and presentations, policy and legislative documents, and written correspondence and semi-structured interviews with a total of 23 key informants. Informants were selected from relevant government bodies (the main bodies responsible for regulating aquaculture, sea lice, and wild salmon), stakeholder organizations (NGOs or corporate organizations involved in sea-lice regulation), and research (scientists and researchers studying farm sea lice on wild salmon, either from government research or key monitoring bodies, public universities, or private research institutions). The aim was to find at least one informant from each of these sectors in each jurisdiction and to select informants who were deemed particularly resourceful, knowledgeable, and experienced in their region (see Kumar, Stern et al. 1993: 1634). In the case of Ireland, the wild salmon NGO declined the request for an interview due to its strong opposition to aquaculture (and thus any aquaculture-related research). However, this NGO provided information about its positions through written correspondence.

A snowball method was used to locate additional, highly knowledgeable persons by asking all interviewees to identify other relevant informants. The Aquaculture Stewardship Council’s (ASC) Head of Standards played a facilitating role, helping to scope out, contact, and convince key informants to participate in interviews.

Interviews were conducted according to a semi-structured model, whereby the researcher prepares a set of prepared themes, issues, and questions to be covered during the interview but leaves space and time open to adjust the sequence and nature of the questions throughout (Kvale 1997, Rubin and Rubin 2005: 4). This enables the researcher to pursue relevant but unanticipated issues and information raised by informants (Kvale 1997: 72, Bauer 2000). The strategy resembles a conversational interview, in which data are gathered through a dynamic interplay and exchange of knowledge and information between informant and interviewer (Pawson 1996: 298, Kvale 1997: 29, Holstein and Gubrium 2002: 113).

All interviews were based on a common interview guide that was adjusted to the specific national or local circumstances of each case. Desk research was conducted prior to scripting the interview guides, to establish the context and an understanding of the historical and current regulations in each jurisdiction. Interviews covered the following topics: (i) politics (the positions of various actors and interests in public/political debates around salmon aquaculture, farm lice, and wild salmon protection and the focus and role of science in such debates); (ii) regulation (specific regulatory requirements and bodies, regulatory updates or lack thereof); (iii) research and stakeholders (the role of science and stakeholders in government decision-making toward reform, consensus, or controversy within government and with stakeholders); (iv) data transparency (farm reporting requirements, monitoring of wild salmon, publication of real-time data).

Interviews lasted between 45 and 90 min. All were transcribed and subjected to systematic analysis, where the content was organized according to key themes and coding words. Information provided by informants was supplemented and triangulated (see Denzin 1978, Miles and Huberman 1994) using documentary analysis, including reports, presentations, policy documents, and newspaper articles.

Due largely to the controversies surrounding views on aquaculture sea-lice regulation and wild salmon, most informants requested anonymity. Thus, the presentation of the research findings does not refer directly to individual statements made or information provided by informants during interviews or written correspondence. An anonymized list of the interview informants is provided in Table 1.

Table 1 Anonymized list of the informants
Full size table

Sea-lice regulation and the role of science in salmon-farming countries

Norway

Norway has sought to reduce sea lice-induced health hazards and mortality risks for its wild salmon populations by setting increasingly strict limits as to on-farm sea lice. During the period of out-migration for juvenile wild salmons (weeks 16–21 in southern Norway; weeks 21–26 in Northern Norway), all holders of government permits to produce salmon in specific locations are required to keep lice levels below a maximum average of 0.2 adult female L. salmonis per fish on their farm sites. When this absolute limit was adopted in 2017, it represented a tightening of the previous requirement from 2012 to keep levels below 0.5.Footnote 1 The 0.2 limit was intended to be precautionary, but the decision-making process prior to its adoption was also partly anchored in research. The Institute for Marine Research (IMR)—a neutral knowledge-provider and advisor on farm and wild fish interactions associated with the Ministry of Trade, Industry, and Fisheries—ran models simulating the effects of different thresholds. However, setting the limit at 0.2 (and not 0.1) was also a decision based on farm data, where consideration was given to what was possible, given existing sampling/counting techniques, and without necessitating excessive delousing, to minimize welfare concerns.Footnote 2 The Food Safety Authority (FSA) is responsible for enforcing compliance with the limit.Footnote 3

Since 2012, license-holders have been required to conduct weekly counts and file weekly reports to the FSA that include the average number of motile lice, mature female lice, and sedentary stages of L. Salmonis per fish in all production cages.Footnote 4 In addition, they must report the type of delousing measures used to keep levels below the set limits. The FSA publishes lice data close to real time and shares the data on a weekly basis with the industry and NGOs. Full public access is provided through two sites, lakselus.no and Barentswatch.no. Every week, the IMR, on behalf of the FSA, also compiles a sea lice and biomass report from each Norwegian production area (PA), which is shared with the industry.

Norway has adopted an area-based system for regulating aquaculture production volumes according to mortality risks for wild salmon. The Traffic Light System (TLS) from 2017 models mortality risk based on sea-lice infestation pressure within 13 delineated production areas, which are then attributed a red, yellow, or green “traffic light” based on their respective risk levels, on a biannual basis. In green areas, salmon producers are allowed to increase their biomass; in yellow areas, they are requested to maintain current production volumes; and in red areas, they are required to reduce their total salmon biomass.

Area-based, mortality risk levels are set by an expert group composed of scientists from the IMR, the Veterinary Institute (VI), and the Norwegian Institute of Nature Research (NINA).Footnote 5 These levels are based on a combination of hydrodynamic dispersion models, which predict the spread of lice larvae from production sites from reported lice levels, sea temperature, and water currents—as well as data from the national surveillance program for salmon lice on wild salmon (NALO), which are used to verify the models.Footnote 6 The conclusions of the expert group are considered by a steering group, who advise the Ministry on the final decision on traffic lights for each PA.Footnote 7 The TLS system is also linked to on-farm thresholds, as farmers within red PA may apply for an increase in site biomass if they can demonstrate compliance with the 0.1 limit.

This increasingly strict regulatory regime has been anchored in the growing scientific knowledge-base concerning the negative impact of sea lice on the health and survival of wild salmonids in Norway (see Skilbrei, Finstad et al. 2013, Svåsand, Boxaspen et al. 2015, Nilsen, Bjørn et al. 2016, Thorstad and Finstad 2018, Vollset, Dohoo et al. 2018). In 2011, the Norwegian Parliament acknowledged that sea lice from salmon farms could be a serious hazard to the country’s wild salmonids, and that it constituted a potential threat their survivalFootnote 8 (White Paper 1 S, chapter 4.1). This was reiterated by the Office of the Auditor General in its 2012 evaluation of Norwegian aquaculture, which concluded that current regulations were insufficient to combat the proliferation of sea lice, in view of the substantial growth and farm expansions. Thus, the growth of a scientific knowledge-base and related political concerns were they key drivers behind the adoption of more stringent sea-lice regulation. The emergence of a near-consensus between national research institutes and the government that farm sea lice may have hazardous, sub-lethal, and potentially lethal effects on wild salmonids also facilitated the institutionalization of researchers as regulators within the TLS.

Although industry actors have disputed the scientific justification for stricter regulation, an evaluation conducted by an independent expert committee has concluded that the regulatory system fails to communicate scientific uncertainties satisfactorily—in particular, regarding the modelled estimations of mortality thresholds (Revie, Eliasen et al. 2021)—there has been little major controversy around the evidence base for regulatory action. Research institutes, policymakers employed in the Ministry, and other bodies such as the FSA and the Fisheries Directorate, as well as the executive government, have largely agreed that sea lice represent a threat to wild populations, and thus on the need for scaling up regulatory actions.

Scotland

Scotland has set comparatively lenient thresholds for on-farm sea lice. Until recently, its sea-lice policyFootnote 9 was not designed to address interactions between farms and wild fish, but to protect the health of caged salmon. The scientific justification for on-farm thresholds did thus not relate to emerging data and evidence of farm-lice impacts on wild populations.

Scotland requires all license-holders to inform the Fish Health Inspectorate (FHI)—which is responsible for monitoring and enforcing the policy—upon exceeding an average of 2.0 adult female L. salmonis per fish. Reports of reaching the 2.0 “trigger limit” prompt increased FHI monitoring. When sea-lice levels reach or exceed an average of 6.0 adult females per fish, license-holders are required to meet the Scottish Salmon Producer Organization’s voluntary Code of Good Practices (CoGP) criteria for the period of out-migration (February 1st to June 30th).Footnote 10 They must then bring levels down to 0.5 adult female L. salmonis per fish.

These trigger levels have been updated once since their adoption in 2007. The notification threshold of 2.0 has remained unchanged, but the intervention limit was lowered from 8.0 to 6.0 in 2019. This was the result of public and political debate around the welfare of farmed fish, and not the potential hazards for wild populations.

Scotland’s sea-lice reporting regime was recently tightened with the adoption of the new Fish Farming Businesses (Reporting) order—which requires mandatory reporting for all aquaculture-production businesses. After the order entered into force in 2021, sea-lice counts must be reported on a weekly basis, irrespective of the average levels per fish; if no count is conducted, a reason must be provided. Fish farmers must report the average number of adult female (gravid and non-gravid) L. salmonis counted per fish per site in the reporting week. The government publishes sea-lice data within 2 weeks of receiving the weekly reports.Footnote 11 Furthermore, all sea-lice data are made publicly available through the Scotland Aquaculture Website.Footnote 12

Furthermore, an ongoing policy reform process has now been instigated to establish an area-based system for regulating sea-lice levels based on modelled infestation pressures in the wild. The scientific basis and justification for adopting this new regulatory scheme is the explicit acknowledgement of the need to address potentially hazardous impacts of farm sea lice on wild populations. The reform was triggered by two parliamentary inquiries by the Environment, Climate Change, and Land Reform Committee (ECCLR) and the Rural Economy and Connectivity (REC) committee in 2018, which led the government to make a political commitment to reforming the existing system. It established the Salmon Interactions Working Group (SIWG), which in April 2020 published 40 reform recommendations; in October 2021, the government issued its formal response to them. Meanwhile, the ruling Scottish National Party (SNP) pledged to thoroughly reform the regulatory framework for salmon aquaculture, with one government authority made responsible for managing farm–wild interactions.Footnote 13

Although the final outcome of this process is still undetermined, responsibility for managing risks to wild salmonids from sea lice emitted from fish farms has been assigned to the Scottish Environmental Protection Agency (SEPA). SEPA also has responsibility for issuing licenses “to pollute” under the Water Environment Controlled Activities Regulations of 2011. These “CAR licenses” set the limits for levels of pollutants that fish farms may discharge to the water environment; although they currently cover the use of sea-lice therapeutants, they do not regulate sea lice or emissions of sea-lice larvae into the wild. SEPA was instructed to build on the work of the Regulators Technical Working Group and worked closely with scientists from Marine Scotland to develop the technical details of a proposal for “a spatially based risk-assessment framework for regulating the interaction between sea lice from marine finfish farms and wild Atlantic salmon.” In 2022, SEPA consulted stakeholders on the proposal, and in 2023, it will consult on how the framework will operate in practice, before it is implemented through the CAR licensing regime.Footnote 14

There have been fewer scientific studies of interactions between wild fish and farms in Scotland than in Norway and Ireland. However, the scientific knowledge-base relevant to assessing risks for Scottish wild salmonids has grown in recent decades (Butler 2002, Butler and Watt 2003, McKibben and Hay 2004, Middlemas, Fryer et al. 2013, Susdorf, Salama et al. 2018).Footnote 15 As in Norway, a near-consensus among research institutes, politicians, and policymakers around the science of wild-farm interactions and the conclusion that sea lice represent a hazard and potential threat to wild populations have enabled political commitment to reform. Government bodies have also increasingly turned to the scientific knowledge-base to justify reform. For instance, Marine Scotland has conducted regular literature reviews, which resulted in the 2021 publication of a Summary of the Science. This has been publicly referred to by the government as the basis for its official acknowledgement of the need to reform the Scottish regime to deal effectively with hazards to wild salmonids.Footnote 16

Ireland

As in Norway, the objective of Ireland’s on-farm sea-lice threshold is to mitigate potential health hazards and mortality risks for wild salmonids. Its sea-lice policy has since 2008 required license-holders to instigate treatment or management action to reduce sea-lice levels on salmon farms when these reach or exceed an average level of 0.5 ovigerous (egg-bearing) L. salmonis per fish in the period of out-migration (March 1 to May 31). This requirement was added to an existing year-round trigger level for treatment set at an average of 2.0 lice per fish. The adoption of a 0.5 “trigger limit” was the result of processes instigated by the government, which had requested an examination and review of the existing system for sea-lice control in marine finfish farms. However, the commissioned “Sea Lice Monitoring and Control Working Group” (with representatives from the Department of Communications, Marine and Natural Resources, the Marine Institute, the Fisheries Boards, and the Irish Seafood Federation) was unable to reach consensus on recommendations for moving forward. The responsibility for aquaculture licensing was then transferred to the Department of Agriculture, Fisheries, and Food (now AFM). In 2008, they developed a pest-control strategy and worked closely with the Marine Institute (MI) on updating the treatment trigger level in the out-migration period. The 0.5 trigger level for treatment was intended to be precautionary, to ensure mitigation of potentially negative effects of farm sea lice. However, setting the level at 0.5 was also a pragmatic decision anchored in farm data, taking into consideration what was achievable. There was no scientific evidence-base or analysis conducted to justify the threshold level.Footnote 17

Independent inspectors from the Marine Institute (MI) are responsible for monitoring and enforcing the treatment threshold. Bi-weekly sampling is conducted in the sensitive period; for the rest of the year, sampling is conducted on a monthly basis. The MI compiles monthly reports of farm sea-lice levels, which include counts of the average level of ovigerous and mobile L. salmonis and Caligus elongatus per fish.Footnote 18 These reports are shared with the Department of Agriculture, Food and the Marine, and a range of other stakeholders, including the IFI and the regional fisheries board. However, live, real-time data are not publicly available. The results are reported back to farms within 5 days of inspection; if a fish farm is found to exceed the 0.5 threshold, MI will aim to report back to the farm as soon as possible.

The sea-lice threshold has not been revised since 2008, although a reform was considered in 2016, when the Department of Agriculture, Fisheries, and the Marine requested a review of the aquaculture licensing process.Footnote 19 That led to discussions between the Department and the Marine Institute (MI); however, according to the latter, there is no new scientific evidence on optimal sea-lice thresholds; whether the threshold requires an update would be a political decision.

The lack of reform thus relates to the MI’s position that the current threshold is adequately precautionary, and the ongoing scientific controversy between actors within and outside the Irish government concerning the basis for scientific evidence and the scale of farm-lice impacts. Various governmental bodies, as well as scientists studying farm–wild interactions, do not agree on how and the extent to which farm sea lice have population-level effects. MI scientists are the main knowledge providers and advisors on aquaculture to the Department; they have developed a substantial body of research,Footnote 20 concluding that farm lice represent only a small and irregular component of the marine mortality of wild populations (Jackson, Cotter et al. 2011, Jackson, Cotter et al. 2013a, Jackson, Kane et al. 2013b). On the other hand, scientists at the Inland Fisheries Ireland (IFI), responsible for protecting wild salmons and working under the Department of Environment, Climate, and Communications, conclude otherwise. They have found that sea lice-induced mortality affects Atlantic salmon returns, and that farm lice represent a serious hazard for sea-trout populations (Gargan, Forde et al. 2012, Shepard and Gargan 2017, Shepard and Gargan 2021). Moreover, the results of studies conducted by MI scientists have been disputed by independent researchers not associated with IFI (Krkosek, Revie et al. 2013).

Similarly, wild salmon stakeholders see the Irish regulation as unsuccessful in mitigating the effects of farm sea lice on wild salmonid survival. They hold that this has led to extinction of sea trout as well as significant reductions in wild salmon populations in fish-farming areas.

Moreover, there is disagreement on the need for stricter regulation: while the MI holds that no new research or analysis on optimal sea-lice thresholds has indicated the need for regulatory change, the IFI argues that the current 0.5 trigger level leads to inadequate protection of wild salmonids, and that Ireland must set stricter on-farm thresholds to ensure that sea-lice levels are closer to zero in the period of out-migration. They also argue for absolute enforcement limits for both mobile L. salmonis and Caligus elongatus and want a “total bay cap” setting a total lice load limit in aquaculture bays during spring migration. However, as the IFI has neither a formal nor informal role as advisor to the Department of Agriculture Food and the Marine, its suggestions have not been taken into consideration. Thus, the lack of regulatory reform since 2008 appears to be anchored in the circumstance whereby a scientific knowledge-base represents only side of the debate.

Canada, British Columbia

The federal government in Canada, which is responsible for both the protection of wild salmonids and for managing salmon aquaculture in British Columbia (BC), has also set on-farm sea-lice thresholds to mitigate risks for wild populations. Licensing conditions require salmon farmers to undertake delousing actions when on-farm sea-lice levels reach or exceed an average of 3.0 motile L. salmonis per fish (equals about 0.64–1.65 adult females)Footnote 21 in the period of out-migration (March 1 to June 30). Although a literature review was conducted prior to setting the 3.0 threshold, the figure arrived at was not based on research or BC-specific knowledge, but was a “best guess” of what might be precautionary at the time.

In 2020, the regulation was updated to strengthen the enforceability of the 3.0 threshold. Since then, farmers have been required to bring lice levels below 3.0 within 42 days of having exceeded the threshold. They must also notify the authorities about planned delousing measures in the pre-migration period (February 1 to 28/29), to ensure that they will be under the threshold by the first day of out-migration. The conditions apply to farmers of Atlantic and Pacific salmon (chinook and coho) and are monitored and enforced by the Department of Fisheries and Oceans (DFO).Footnote 22 These revisions were triggered by heightened political attention to sea lice-induced hazards for wild salmon in 2018, after reports of a sea-lice outbreak in Clayoquot Sound on the west coast of Vancouver Island following the emergence of resistance to SLICE (a treatment for all parasitic stages of sea lice) and the lack of alternative chemotherapeutics or mechanical delousing equipment.Footnote 23 After independent biologists and wild-salmon NGOs had alerted the media, bringing the issue to the DFO Minister’s attention, the aquaculture management division was asked to update the conditions of licensing to improve the enforceability of the threshold.

Atlantic salmon farms are required to conduct bi-weekly sampling in the pre-migration and out-migration windows and to report the average level of motile, chalimus-stage, and adult female L. salmonis, as well as the average level of adult and preadult Caligus clemensi per fish. For the rest of the year, license-holders are to report monthly. For Pacific salmon farmers, quarterly sampling is required. The DFO publishes an Industry Sea Lice Abundance Counts report (per farm), updated on a monthly basis.Footnote 24 However, there is a significant time-lag between reporting and publication. Although they must perform bi-weekly counting, license-holders are required to submit the counts to the DFO only on a monthly basis. The DFO receives the reports on the 15th of the following month, upon which they conduct a quality control, including a comparison of data with DFO-performed audits, which may take between 2 and 4 weeks. In practice, however, the reviews of reports are often bundled together quarterly, so there is a significant time-lag of up to several months in data publication.

The lack of a substantial tightening of the 3.0 threshold also appears associated with an ongoing controversy around the science of farm–wild interactions. There are opposing views among researchers working within the DFO science branch, and some DFO studies show that the physiological impact of L. salmonis on Pacific salmon species, particularly sockeye salmon, may be greater than for Atlantic salmon (Long, Garver et al. 2019). However, most DFO scientists argue that the risks to wild populations have been exaggerated (e.g. Brooks and Jones 2008). The views of in-house scientists reflect the position of the DFO aquaculture management, who argue that farm sea lice do not represent a significant threat to the abundance and population productivity of wild salmonids.Footnote 25 Thus, continuation of the 3.0 threshold is anchored in a belief shared by many DFO researchers and policymakers that the threshold remains precautionary in nature and that the population-level effects of farm sea lice are low to negligible.

However, a substantial body of research developed by independent scientists concludes otherwise, stressing the importance of sub-lethal and indirect effects of farm lice on the health of wild populations (Mages and Dill 2010, Godwin, Dill et al. 2017). These have also demonstrated the existence of significant population-level impacts, albeit through correlational studies (Krkosek, Connors et al. 2011, Connors, Braun et al. 2012). However, non-DFO scientists do not have a formal advisory role as do their in-house scientists and have historically not influenced the regulatory decision-making.

When sea-lice conditions were up for renewal in 2022, the DFO was considering arguments for stricter and more extensive regulation advocated by the independent research community and wild salmon stakeholders.Footnote 26 These have long argued that lice levels should not be allowed to remain above 3.0 for as long as 6 weeks, resulting in volatile levels in the period of out-migration.Footnote 27 Second, they have argued for setting thresholds related to the total lice load of a farm or farming area. As the size and thus biomass of many farms has grown over time, so has the abundance of sea lice—irrespective of the license-holders’ ability to keep average levels per fish below 3.0. Thirdly, they have pressed for the adoption of a management system for farms based on monitoring of salmonids and sea-lice levels in the wild. However, the DFO has remained skeptical to area-based regulation of farms on the basis of wild salmonid monitoring or modeling of risks based on sea-lice infestation levels.

Regulatory differences and how science shaped policy (in)action to enhance protection of wild salmons

This study finds substantial variations in the national regulatory regimes governing on-farm sea-lice levels in periods of wild salmonid out-migration, as well as efforts towards adopting area-based regulation of total lice loads in production areas based on sea-lice infestation levels in the wild and the estimated mortality risks for wild salmons.

Scotland has historically not designed its sea-lice regulations for the purpose of mitigating impacts of farm lice on wild populations: until recently, the aim has been to protect farmed-fish welfare. This explains the use of a higher trigger level for treatment for adult females compared to the other countries. In Norway, Ireland, and Canada (British Columbia), on the other hand, on-farm sea-lice thresholds have been adopted for the stated purpose of minimizing potential hazards to wild salmonids. Norway enforces the most stringent, absolute sea-lice limit of 0.2 mature female lice per fish. Ireland does not regulate so strictly, setting a trigger level for treatment at 0.5 ovigerous lice per fish, and Canada (BC) sets a trigger level for delousing actions at 3.0 motile lice per fish (equaling ca. 0.64–1.65 adult females).Footnote 28

As regards sea-lice reporting requirements, these countries differ considerably with respect to data-sharing arrangements and the timing of publication. In Scotland, sea-lice data are published no later than 2 weeks after recording. In Norway, the data are published close to real time; in addition, the government arranges regular industry data-sharing through meetings and report distribution. In Ireland and Canada, sea-lice reports are shared with industry and stakeholders, but the considerable time-lags in publication give rise to concerns about real-time data transparency.

Norway stands out as the only country that has implemented area-based sea-lice regulation. However, Scotland is moving in the same direction, seeking to reform its salmon farming regime thoroughly, and adopting a spatially based, risk-assessment framework. This appears to be partly influenced by Norway’s red/yellow/green “traffic light” system. While there is strong political commitment to reform in Scotland, uncertainties remain as to how on-farm lice thresholds will be updated.

However, an in-depth comparative evaluation of the different sea-lice limits and trigger-level thresholds with the aim of assessing their robustness or effectiveness in protecting the health of wild salmonids is beyond the scope of this assessment (Table 2). The various threshold levels are not directly comparable, as the jurisdictions have varying ecological and biological conditions, different wild salmonid species, lice species and sub-species, as well as highly varying aquaculture production outputs and farm-area densities. For instance, Norway has set the strictest absolute thresholds but is also the by far largest producer of farmed salmon. Ireland’s trigger levels are less stringent, but Ireland has a comparatively small-scale fish-farming industry. Canada’s BC threshold is set to protect wild populations of Pacific salmon, which are genetically distinct from Atlantic salmon.

Table 2 Comparison of regulatory regimes for aquaculture sea lice
Full size table

A comparative analysis of sea-lice regulation in these jurisdictions shows that two countries have revised or are in the process of revising their sea-lice regulations with the aim of enhancing wild salmon protection: Norway has adopted strict on-farm limits and a new, area-based management system, and Scotland is in the process of adopting a similar system. A near-consensus around the scientific knowledge-base of farm–wild interactions has also emerged. Although many aquaculture industry actors have disputed the need for stricter sea-lice management, most scientists, national research institutes, politicians, and policymaking bodies have acknowledged and largely agreed that sea lice from salmon-farm lice represent a major health hazard and potentially threat to the reproductivity of wild salmonids. The absence of pronounced conflict around scientific uncertainties appears to have enabled political efforts to adopt stricter regulation to mitigate the risks for wild salmon populations.

Ireland and B.C. Canada have not conducted substantial regulatory revisions in response to calls for improving wild salmon protection, and in these jurisdictions, there has also been pronounced political controversy around the science of farm-wild fish interactions. Indeed, the deviating positions of the various national research institutes and independent scientists on both the sub-lethal and the population-level effects of farm sea lice appear to have limited the scope for regulatory reform. In both countries, the governmental departments responsible for aquaculture management have relied largely on research, analysis, and advice from scientists who represent only one side of the scientific debate, leading them to conclude that population-level effects are small to negligible. Other research institutes or scientists have concluded differently and stressed the need to incorporate sub-lethal effects of lice on salmonids to a greater extent. However, this research has not had a significant impact on government decision-making. This is most clearly seen in Ireland, where sea-lice regulations have not been updated since 2008, despite considerable pressure from wild salmonid knowledge providers. Thus, a lack of consensus around the science involved appears to have hindered political agreement on the need for regulatory change.

Conclusions

This study has conducted a comparative evaluation of aquaculture sea-lice regulations in major salmon-farming countries with wild salmon populations they are tasked to protect. This includes an assessment of their on-farm limits/thresholds, systems for data monitoring and reporting, and area-based management. The aim has been to establish how and the extent to which the jurisdictions have implemented regulatory reforms to meet growing concerns that protection of the health and survival of their wild salmon populations has been inadequate.

Norway has the by far most stringent and extensive regime for sea-lice management, combining strict absolute on-farm limits and reporting with area-based regulation. Scotland is headed in this direction, having embarked on a thorough reform toward area-based sea-lice management. The regulatory reforms of these jurisdictions, involving significant updates towards more precautionary measures to protect wild salmonids, have been made possible by the emergence of consensus around the science of farm–wild interactions. In contrast, Ireland and Canada’s British Columbia have not seriously considered or embarked on any substantial policy reforms. The lack of significant regulatory updates in these jurisdictions appears related to the persistence of controversy around the knowledge-basis for reform. Here, scientists and policymakers still disagree on the extent of the impact of farm sea lice on wild populations, which has hindered reformist forces and led to a lack of political commitment.

This illustrates a more general phenomenon related to “wicked” policy problems (Rittel and Webber 1973): that uncertainties related to available data and knowledge become a political battleground between actors with competing policy preferences (Sarewitz 2004, Van Enst, Driessen et al. 2014). It suggests that the emergence of consensus between key actors around the knowledge-base for regulatory change may serve to facilitate reforms to deal with wicked policy problems such as aquaculture sea lice. Conversely, when scientific disparity and political conflict exist or persist, reform may be difficult. Thus, broader consensus around the science of farm–wild interactions may be an important prerequisite for the adoption of more precautionary policies for mitigating risks to wild salmon populations.