Veterinary and Animal Science 11 (2021) 1001672Aquaculture industry sectors Scottish aquaculture is dominated by Atlantic salmon (Salmo salar) production, however rainbow trout (Oncorhynchus mykiss), mussels (Mytilus spp.) and Pacific oysters (Crassostrea gigas) also account for significant sectors (Table 1). The scale and structures of production differ between sectors, in particular salmon production is handled by a small number of large multinational companies (Asche, Cojocaru, & Roth, 2018), while other species are farmed by many small companies. This has important implications for the access of different sectors to resources and markets. Small companies with fewer resources are more vulnerable to running out of funds, and may also have more difficulty getting product to markets. Almost all the aquaculture sectors move ova or live animals between different sites; for salmon movement from freshwater to seawater is inherent in their life cycle. Movements are often over significant dis-tances; mean distances salmon are moved in the freshwater phase and from freshwater to seawater are 81 and 201 km respectively (Wallace, Munro, Murray, Christie, & Salama, 2016), marine phase mean move-ments are only 10 km. A large proportion of aquaculture production is dependant on imported ova (Table 1) the movement of which along with animals requires resources and the movement of people, so could be vulnerable to disruption due to restrictions aimed at reducing the spread of COVID-19. Production of finfish also requires that feed pellets are supplied to the farms and this uses fish meal, oil and soya that is sourced globally (Asche et al., 2018). Disruption of this chain could quickly lead to welfare issues, as could disruption in the supply of vaccines and medi-cines and the ability of fish health professionals such as vets and fish health inspectors to conduct visits. In contrast, shellfish, filter their own food consuming plankton, thus require much less labour and are not vulnerable to supply chains. Authorities have applied temporary relax-ation of regulations to ensure medicines are available (SEPA, 2020a). Pathways to impact of COVID-19 on aquatic animal disease management Animal health and welfare are key issues causing problems for development of sustainable aquaculture (Jones et al., 2015; Murray & Peeler, 2005; Stien et al., 2013). We describe the routes of impact of COVID-19 on fish health and welfare assessing means by which impact occurs, the end-point indicators of that impact and sources of data to assess that impact (Fig. 1). COVID-19 and associated management measures to protect human health lead to reduced availability of labour and economic resources for fish health and welfare which is managed under regulation and a Code of Good Practice (COGP, 2015). These potential impacts are identified as occurring through two pathways: (i) harvesting and stocking which affects disease management (Pettersen et al., 2015) and population densities which are important drivers of disease and parasites (Anderson & May, 1979; Moriarty et al., 2020; Murray & Peeler, 2005); and (ii) through surveillance and veterinary interventions for control of diseases and parasites (Oidtmann, Peeler, & Lyngstad, 2013). Regulation has been partially relaxed on both biomass limitation and sea lice to allow for COVID impacts (SEPA, 2020a, 2020b), which may lead to additional negative environmental impacts (Fig. 1). In the medium to long term under-resourced management and poor environmental conditions can result in increases in parasites and dis-eases that reduce fish health and welfare, as human health and envi-ronmental effects can work synergistically to impact animal health and welfare (Rüegg et al., 2018). Mortality and parasitic sea lice prevalence are considered key indicators for fish welfare (Stien et al., 2013) that provide data on potential impacts. We consider three specific endpoints, parasitic sea lice, endemic diseases and notifiable diseases as detailed in Fig. 1. The direct and acute impacts of COVID-19 are to reduce labour (either through restrictions or illness in the workforce), disrupt supply lines and logistics (again through restrictions or illness) and in the me-dium to longer term to reduce economic resources available to the production company. These are further reduced by increased costs associated with protecting employee health (e.g. PPE, testing). This may mean increased timeframes are required or a temporary reduction in resources available for protection of fish health. Regular surveillance is vital to the timely targeting of treatments and controls for pathogens. This takes a range of forms, including daily on- site observations by staff and more occasional visits by vets or fish health professionals. Absence of the individuals, either because staff are unavailable or due to restrictions on site access has an impact on active surveillance. Reporting of data through passive surveillance is important for assessing COVID-19 impacts on animal health which can aid in strengthening policy against future impacts, and inform processes for contingency planning. Intervention to manage pathogens can be directed with information from surveillance but requires considerable input of labour and invest-ment, notably for sea lice management (Overton et al., 2019) which typically costs 9% of farm revenue (Abolofia, Asche, & Wilen, 2017). Options for management of infectious diseases caused by micro patho-gens are more limited but can also be expensive, for example £25 M from an outbreak of ISA in the 1990s (Hastings et al., 1999). A key activity for aquaculture production that may be impacted is harvesting. Harvesting requires both local labour and also access to lo-gistics and markets that may be international. Delayed harvesting weakens the finances of companies, which could reduce resources and also increases the biomass of fish on the farms, increasing the risk of mortality (Moriarty et al., 2020) and higher rates of sea lice treatment (Murray & Hall, 2014). Early harvesting is also a disease management practice which can maximise potential production (Pettersen, Rich, Bang Jensen, & Aunsmo, 2015) this may not be practicable under COVID-19 restrictions. Potential outcomes of inability to manage pathogen loads are increased sea lice infestation, increases in endemic disease losses and outbreaks of notifiable diseases. These impacts are different and measured by different data sets. Sea lice are rated as amongst the most important pathogens for sustainable aquaculture (Jones et al., 2015) and management to reduce lice numbers is required with government intervention should loads indicate improvements in mitigation are required (Marine Scotland, 2019). Increases in lice numbers feed back to increased management Table 1 Structure of the Scottish aquaculture industry in 2018. Data from Munro 2020a, Munro, 2020b. FW =Freshwater, SW =seawater. Atlantic salmon Rainbow trout Mussels Pacific oysters Production (tonnes/y) 203,881 7,405 6,699 369 Number of production sites 226 52 113* 41* Number of production companies 11 22 31 31 Ova imported 85% 99% Wild spat †First sale value £1074M £36.2‡£6.2M £1.6M Feed input Yes Yes No No Environment FW then SW FW and SW SW SW *There were 165 producing shellfish sites in 2019 but these were not sepa-rated by species in Munro, 2020b, but 69% in Scotland’s Aquaculture Website (http://aquaculture.scotland.gov.uk/) licenced for mussels and 25% licenced for Pacific oysters. ‡Figure not published regularly, value estimated from Marine Scotland (2016). †Data not in Munro, 2020b, but see Murray, Munro & Matejusova, 2020, Pacific oyster spat is mostly sourced from England or the Channel Islands, there is no hatchery in Scotland. A.G. Murray et al.