The behavior and welfare of neglected species: some examples from fish and mammals

Buffalo welfare issues and limitations for production

Before being domesticated, buffaloes were able to roam freely in the wild (Deb et al., 2016; Zhang, Colli & Barker, 2020), and used to spend most of their time resting, feeding, and ruminating. Bathing and wallowing are two main buffalo behaviors used to regulate body temperature and protect against ectoparasites in extensive farming (Mora-Medina et al., 2018). Buffalo welfare is generally expected to be better in extensive farming because the animals are free to express natural behaviors e.g., wallowing (Fig. 3) and browsing (Figs. 4A, 4B), (Mora-Medina et al., 2018). Being able to engage in natural behaviors allows animals to fulfill their own needs. A wide range of natural behaviors are linked with positive emotions, and their performance directly enhances animal welfare (Špinka, 2006; Browning, 2020). Furthermore, engaging in natural behavior has long-term benefits for animals, including enhanced ability to deal with social and physical problems (Špinka, 2006; Browning, 2020). However, conventional farming methods are being replaced by more intensive systems designed for dairy cattle, without access to water or mud for wallowing (Napolitano et al., 2013).

Intensive farming has exposed buffaloes to changing environments that introduce psychological and physical stressors previously unknown to them, such as confinement, absence of grazing, poor cleanliness etc. (de Rosa et al., 2009b). Buffaloes are often now kept in dairy farm sheds (5–10 m² per animal) with space availability in an adjacent outdoor farmyard (8–14 m² per animal), (Napolitano et al., 2013). In modern farming, confinement discourages natural behaviors such as wallowing (Fig. 3), and grazing (Fig. 5). Furthermore, space restrictions in more intensive systems results in the expression of undesirable behaviors such as teat suckling of other animals and excessive aggression, resulting in risk factors for injuries and poor welfare (Napolitano et al., 2013). Machine milking may involve appropriate equipment performance and calf detachment (removal of calf from mother), (Napolitano et al., 2013). The negative behavior of stockpersons, such as hitting, slapping, and yelling, can cause physical and emotional stress, leading to milk letdown problems, elevated heart rates, higher cortisol levels, and retained milk (Rushen, De Passille & Munksgaard, 1999; Napolitano et al., 2013). Other welfare issues associated with intensive buffalo farming include higher mortality levels, probably due to immune suppression caused by stress and poor housing conditions. Intensive farming also leads to the progression of abnormal behaviors, such as allosuckling and tongue rolling (de Rosa et al., 2009b).

Cattle were the first animals to be incorporated into intensive farming systems, with advancements like machine milking, and housing specifically developed for them (de la Cruz-Cruz et al., 2019). Currently, the same techniques are also being used for buffalo milk production. By gaining a deeper comprehension of buffalo behavior, physiology, and health, and making necessary adaptations (for example, housing, nutrition, and design of milking parlors) to their production systems, there is potential to enhance their welfare (Deb et al., 2016). It is important to recognize the significance of welfare throughout the entire production chain to enhance their welfare and production efficiency, as buffalo welfare extends beyond their ability to produce meat, milk and work as draught animals (Mora-Medina et al., 2018). Moreover, as buffaloes can adapt better to hotter than colder environments (de Rosa et al., 2009a), this feature makes them an important future resource as global temperatures increase.

Monitoring buffalo welfare

More attention needs to be focused on housing since it not only limits the natural buffalo behaviors, but it can also induce physiological changes related to stress due to limited space (Mora-Medina et al., 2018). Water pools in traditional farming systems have been phased out in favor of spray systems in order to minimize the time spent cleaning the udder, enhance milk hygiene, and decrease disease transmission during milking, such as mastitis (Rosa et al., 2007). However, buffaloes with access to a pool had a higher level of fertility and shorter calving intervals than those with shower stalls to help with body temperature regulation (de Rosa et al., 2009b; Napolitano et al., 2013). Providing buffaloes with approximately 36–40 m² space per animal along with access to vegetation and pools is important as those conditions allow them to express natural behaviors (for example, grazing and wallowing). Appropriate space per animal is also linked to increased milk production (Rosa et al., 2007). Therefore, to enhance buffalo welfare, housing should include water pools, adequate space, and access to natural grass (de Rosa et al., 2009b; Napolitano et al., 2013).

Animal cleanliness reveals information about their comfort, pleasure, and standard of care by stockpersons (Winckler et al., 2003; Napolitano et al., 2013). Dirty skin can minimize antigerm and thermoregulatory features of the skin, as well as causing epidermal inflammatory disorders (Winckler et al., 2003; Napolitano et al., 2013). A buffalo cleanliness score has been developed but its repeatability was relatively low compared to cattle (de Rosa et al., 2009b). This could be due to their unique wallowing behavior, performed in order to get protection from biting insects and solar radiation. Thus, the appearance of mud on buffaloes can be potentially regarded as a positive trait, while a thick and compact layer of dung on their bodies may be considered a potential health risk (Napolitano et al., 2005, 2013).

Buffalo body condition score (BCS) is used to estimate body energy stores by scoring the subcutaneous fat components of animals rather than live weight fluctuations (de Rosa et al., 2009b; Napolitano et al., 2013). This method entails assigning three distinct scores to each buffalo based on the categories of ‘very lean,’ ‘regular,’ and ‘very fat.’ BCS can be used to assess malnutrition and welfare, with an increased proportion of skinny animals associated with lower welfare (de Rosa et al., 2009b). Morphologically, buffaloes are more comparable to dual-purpose cattle (raised both for milk and meat purpose) than dairy cattle (Napolitano et al., 2013).

Damaged skin shows the effect of the surrounding environment on the body of an animal (de Rosa et al., 2009b; Napolitano et al., 2013). Contact with hard surfaces, drags against walls, feed shelves and cubicle dividers can cause skin damage (Napolitano et al., 2013). Thus, while assessing buffalo welfare, it is important to check for any swelling, callosity, and lesions on various body areas. These measurements reflect inappropriate management, feeding racks and housing design (Napolitano et al., 2013).

Higher buffalo mortality can result from poor management and bad welfare, leading to economic losses for farms (Napolitano et al., 2013; Deb et al., 2016). Important buffalo diseases include enteritis, respiratory issues, and mastitis, which lead to high mortality and culling. These are generally linked to nutritionally unbalanced feed fed to buffaloes (de Rosa et al., 2009b; Napolitano et al., 2013). Therefore, guidelines for assessing the buffalo welfare on farms must consider parameters that reveal their health and diseases. Animal health and disease information should be recorded on farms for animal welfare assessment, as mortality and culling due to illness and accidents are important indicators to quantify buffalo welfare (de Rosa et al., 2009b; Napolitano et al., 2013).

Buffalo calves that are kept in confined spaces (1.2 m² per animal), rest and lie for less time, and have more bent legs while lying as compared to those living in larger areas (2.1 m² per head) (de Rosa et al., 2009b). Animals also show decreased levels of lying idle, which is likely to represent a form of resting (de Rosa et al., 2009b). Cross-suckling occurs in both calves and adults, resulting in severe injuries to udders, prepuce and navel, as well as milk loss in dairy animals (Napolitano et al., 2013).

Amid decreased space allocation (up to 5–10 m² per animal) due to farm intensification, agonistic interactions are becoming an important problem for buffalo farming (de Rosa et al., 2009b; Napolitano et al., 2013). Buffaloes are also usually not dehorned and are free to engage in agonistic behaviors. However, social contacts are more likely inside a confined area, and flight possibilities of weak animals are greatly decreased (de Rosa et al., 2009b; Napolitano et al., 2013). Therefore, when the available space is too limited, the risk of abnormal behavior, such as cross-suckling, greatly increases and can compromise welfare. (de Rosa et al., 2009b; Napolitano et al., 2013). Primiparous buffaloes are frequently mixed in with multiparous animals. Primiparous animals are usually have the lowest social status and therefore, they have a higher likelihood of udder and skin lesions due fighting for food, and water and space limitations (de Rosa et al., 2009b; Napolitano et al., 2013).

Overall, to establish a buffalo welfare monitoring system, it is essential to implement standardized methods for translating welfare related indicators into consumer-friendly information. Housing, body cleanliness, body condition score, lameness, normal social relationships, mortality rate, abnormal behaviors, and positive welfare indicators are essential measures to monitor buffalo welfare (Hafez, Badreldin & Shafei, 1955; de Rosa et al., 2009b). The most urgent research priorities for buffalo should be machine milking, housing, behavior, physiology, and standardization of farming and welfare monitoring systems (de Rosa et al., 2009b; Napolitano et al., 2013).

Camel production systems

There are three major camel production systems, i.e., nomadic, sedentary, and modern farming (Ahmad et al., 2010; Camelicious Farm, 2022). These systems are primarily determined by number and purpose of animals, climatic conditions, the topography of the land, plant growth phenology, and water sources (Ahmad et al., 2010).

Camel welfare issues and limitations for production

Nomadic camel welfare issues and limitations for production have been reported from various regions of Africa, the Middle East and South Asia (Ahmad et al., 2010; Dioli, 2022). Camels are often subjected to periods of hunger and starvation caused by erratic rainfall and decrease in forage availability. Moreover, drinking water points are usually several walking days away from the foraging areas. Camel night enclosures are often not appropriately built and routinely subjected to thermal as well as predator stresses. Furthermore, clinical veterinary services may be unavailable in nomadic pastoral areas. Diseases, parasites, and the pain and distress that can be associated with disease are thus a regular occurrence, adversely affecting camel health and welfare (Dioli, 2022).

Traditional husbandry methods (nomadic and sedentary systems), inadequate marketing of camel products, and calf mortality rates remain key barriers to overall camel production and welfare (Faraz et al., 2021). Moreover, inadequate marketing of camel products results in lesser income for camel owners, which forces them to compromise on their housing quality and other farm operations. This ultimately impacts the living standards and welfare of camels (Faraz et al., 2021). Ethno-veterinary medicines are commonly used as primary health care, influencing camels’ overall health and production potential. It is crucial for concerned stakeholders and authorized institutions to evaluate the immediate needs, education, knowledge, and husbandry expertise of indigenous communities. This assessment will facilitate support for farmers and the enhancement of animal welfare in Low and Middle Income Countries (LMICs), (Faraz et al., 2021).

Camel welfare assessment

Some camel welfare assessment guidelines have very recently been designed for intensive or farm-based animal production systems (Kaurivi et al., 2020; Dioli, 2022). However, they cannot be used directly to assess welfare issues for extensively managed or nomadic livestock (Kaurivi et al., 2020). Restraining methods, milking practices, weaning and fostering, calf acceptance, packing and riding practices, slaughtering methods and traditional ethno-veterinary treatments are some of the most common areas of camel production where their welfare may be compromised (Dioli, 2022).

Some of the most important factors to consider for herd level camel welfare assessment are proper feeding, housing, appropriate health and the ability to engage in normal behaviors (Padalino & Menchetti, 2021). There should be a proper feeding management framework for the herd for the provision and availability of nutritionally balanced feed. There must be a sufficient number of clean feeding points and feeding behavior of animals should be assessed (Padalino & Menchetti, 2021; Dioli, 2022). Comfortable housing should be provided for animals with appropriate bedding and space allowance. The behavior of animals needs to be considered while resting for welfare assessment (Kaurivi et al., 2020; Padalino & Menchetti, 2021). Animals must be free from injuries and diseases and there is a need to check for sick animals, type of injury and disease. Finally, there should be check regarding behavior of animal for welfare assessment, e.g., social behavior, aggressive behavior, stereotypies etc. (Padalino & Menchetti, 2021).

Individual animal level welfare assessment usually comprises a visual inspection to evaluate camel behavior and health status. Some of the most important points to check while assessing individual camel welfare are body condition score, presence of disease, resting behavior, and injuries due to tethering and hobbles (Padalino & Menchetti, 2021; Zappaterra et al., 2021). Moreover, camel feeding and drinking behavior should be examined. The good feeding and drinking principles are adequate nutrition and lack of prolonged thirst. Hunger and thirst can appear not only when feed and water are unavailable, but also when they are inaccessible or do not meet the behavioral and physiological needs of the animals (Padalino & Menchetti, 2021).

Welfare issues of donkeys and mules

The welfare standards of working donkeys and mules remain very poor (Valette, 2015; Davis, 2019). Some of the most common reasons for poor welfare are improper shelter, food, and water; high workload, poor handling (e.g., driving and whipping) and harmful practices such as nostril slitting (Davis, 2019; Bukhari et al., 2023). In addition to this, a lack of supporting infrastructure (healthcare, good farriers, and saddlers), harsh environmental conditions, a lack of recognition in legal systems, marginalization, and absence of program enforcement also contribute to compromised working donkey and mule welfare (Valette, 2015; Davis, 2019).

Overloading of donkeys and mules is one of the many issues that may lead to reduced welfare, which is a global concern (Bukhari et al., 2022). Their most common welfare issues include overloading, poor physical condition, lameness, dehydration, heat stress, exhaustion, malnutrition, and other catastrophic injuries (Valette, 2015; Bukhari, McElligott & Parkes, 2021; Khan et al., 2022). The severity of welfare problems depends on the type of work the animals perform. For example, brick kiln work (Fig. 8) appears to be linked with increased welfare problems in working equids compared to other work areas, for example, agriculture (Mitra & Valette, 2017). The welfare of working equids can be improved through the collective actions of both equid-owning communities and supporting organizations (Bukhari et al., 2023).

Changes in consumer habits also have an impact on livestock welfare and behavior. Chinese medicine practitioners use gelatinous material extracted from the donkey hide as an elixir (Bukhari et al., 2023). Traditionally, this material, known as ‘ejiao’, is believed to improve strength, treat anemia, and counter some psychological issues, including insomnia. Initially, the demand for donkey products was met by slaughtering large numbers of donkeys within China, resulting in a reduction in the Chinese donkey population. Increasing demand for donkeys has led to increased farming in China itself, as well as other countries such as Australia, Brazil, Kenya, Pakistan, and South Africa (Davis, 2019; Tatemoto et al., 2021). However, farming systems need to be improved to consider the welfare and behavioral needs of donkeys, as this mode of production is relatively new. Important behavioral needs include the freedom to graze and browse, as well as the opportunity to engage in normal behaviors such as forming social bonds with other animals (Davis, 2019; Tatemoto et al., 2021). Donkeys, being hindgut fermenters with a simple stomach, are adapted to “trickle feeding,” which involves consuming small amounts continuously to meet their nutritional requirements. If donkeys are only fed twice a day (a common practice on intensive farms), their body condition and gastrointestinal health may suffer, and their nutritional needs may not be adequately met based on their stage of production (lactation, pregnancy, working, etc.), (The Donkey Sanctuary, 2023). Consequently, it remains uncertain whether intensive farming is suitable for this species, and evidence-based research is necessary to address these inquiries.

Every year, millions of equids are slaughtered (Tatemoto et al., 2021). Equine meat consumption is deeply rooted in the culture of certain countries, such as France and Japan, as well as in LMICs like Chile, where it is popular due to its affordability compared to beef (Rubio Lozano et al., 2020; Fletcher et al., 2022). However, there is a lack of consistency in equine slaughter guidelines worldwide, and published data on the topic are generally insufficient. This hampers their practical implementation and may compromise animal welfare, particularly in the moments leading up to slaughter (Fletcher et al., 2022). In LMICs, there is a scarcity of published studies that assess the suffering of animals during the slaughter process, especially for donkeys and mules (Davis, 2019; Tatemoto et al., 2021; Fletcher et al., 2022). This highlights the urgent need for research to establish an evidence base that can provide better guidance in this specific field. Furthermore, slaughterhouses should adapt their practices to accommodate the distinct needs of donkeys and mules, including access to food and water, efficient restraining, stunning, and slaughtering methods. However, there have been no studies conducted to examine the effects of these practices.

Donkey and mule welfare assessment

Two specific welfare assessment tools exist for donkeys and mules. Firstly, the Equid Assessment, Research and Scoping (EARS) tool developed by the Donkey Sanctuary (Raw et al., 2020) and secondly, Standardized Equine Based Welfare Assessment Tool (SEBWAT) developed by the Brooke (Browne, 2012). The purpose of these tools is to provide an overview of the general welfare condition of animals, both individually and at a group level (Browne, 2012). The data generated by using these tools helps to determine the nature and prevalence of welfare issues occurring within an animal population. Secondly, they help to compare the welfare status between various locations and communities of interest. Third, they determine which groups of animals have the poorest welfare and the seasonal variation in welfare issues.

However, these protocols should be adapted to the context in which they are used since welfare issues differ in different parts of the world and even different parts of the same country (de Santis et al., 2021; Bukhari et al., 2022). For example, for working donkeys in LMICs resource-based protocols are not practical, while animal-based behavioral measures are particularly appropriate. Protocols should be easy to learn, low cost and rapid in their execution (de Santis et al., 2021). Behavioral measures for welfare assessment consist of the general attitude of donkeys and response to human approach (Table 2).

Lumpfish welfare issues

When used in Atlantic salmon production, lumpfish are placed into sea cages and are not only held with salmon but may also be held with wrasse also used as cleaner fish. Their diet in the sea cage consists of parasitic sea lice and they are confined to the shallow coastal conditions they are housed in. High mortalities have been reported for lumpfish and other cleaner fish species. One study documented that 27% of lumpfish died over a 12 week period conducted in Norway (Geitung et al., 2020). Cleaner fish mortalities in sea cages range from 18% to 48%, with some farms sustaining 100% mortality (or loss for unknown reasons) (Nielsen et al., 2014). Over 65% mortality of ~193,000 cleaner fish was recorded in 12 commercial salmon sea cages (Bui et al., 2018) and a recent industry report registered mortality at 42% (Stien et al., 2020). Given this high loss of life, it is crucial that we understand how to improve their health and welfare in these production systems.

A study exploring welfare, growth and survival of lumpfish demonstrated that these animals occupy shallower, cooler and less saline areas of sea cages (Geitung et al., 2020). One of the impacts of climate change is that oceanic temperatures are rising and thus lumpfish may be unable to seek cooler temperatures in the confines of a sea cage especially during heat waves. These animals are cold blooded and thus the environmental temperature has a substantial impact on their physiology. Heart rate measurements in response to increased temperature and exercise demonstrated that lumpfish have a relatively poor response with a low maximal heart rate and thus these fish are unlikely to perform well under conditions when temperatures rise (Zrini, Sandrelli & Gamperl, 2021). Preference studies show adult lumpfish prefer 6–7 °C but on land production of these animals is typically 10 °C (Mortensen et al., 2020). During a heatwave in Newfoundland, surface water temperatures rose from 10–12 °C to 18–19.5 °C. During this period the salmon within the cage reduced their depth to cooler areas (Gamperl, Zrini & Sandrelli, 2021). If lumpfish were present in this scenario, these temperatures would be much higher than their preferred range but they may also have to compete for space at cooler, lower depths with the relatively larger and more active salmon. Lumpfish used as cleaner fish are unable to choose their own habitat and may be confined in temperatures much higher than their temperature preference (Hvas, Nilsen & Oppedal, 2018; Mortensen et al., 2020). It is currently not known whether the thermal environment has an impact on the welfare of lumpfish in sea cages, but this is a crucial question that requires further study. Climate not only affects fish but can also lead to storms creating rough seas that might impair the welfare of lumpfish (Zrini, Sandrelli & Gamperl, 2021), especially if there is little for them to cling to within a sea cage. Other challenges in the sea cage environment linked to climate change concern interactions and dangers from other species such as the increasing prevalence of stinging jellyfish blooms. A recent report in the media stated that the incidence of micro jellyfish blooms substantially impaired the welfare of Atlantic salmon within sea cages and could lead to high mortality (The Herald, 2023). However, it was not reported whether lumpfish were present or affected by these jellyfish bloom events; thus more information on how climate related changes in not only abiotic factors but also biotic or biological interactions with other species is needed.

Wild adult lumpfish are often caught during the spawning season to act as brood stock in captivity (Powell et al., 2018), where eggs are hatched by being held in incubators or hatching boxes. After hatching, lumpfish are typically fed commercial diets or live brine shrimp (Artemia salina) before being transferred to a growing farm, when they reach a size of 50mm (Powell et al., 2018). Lumpfish are typically reared on land in either tanks supplied by seawater from the marine environment in an open or flow through system or in a recirculating aquaculture system (RAS) (Dahle et al., 2020). The water used in RAS is either taken from the marine environment or is artificial seawater that is recirculated through a filtration system to keep water quality optimal. These are closed or semi-closed since to maintain salinity either new salt water or freshwater must be added and to maintain water quality some water has to be replenished at regular intervals to remove the accumulation of nitrate (end product of the denitrification of ammonia and nitrite). RAS is considered to be more environmentally friendly and sustainable, but it requires electricity or other power supplies to fuel electrical equipment e.g. water pumps and other filtration mechanisms. In 2018, 31 million lumpfish were farmed for use in sea cages in Norway. The number of cleaner fish farms in Norway mostly rear lumpfish and the numbers have increased from five to 31 from 2014 to 2019 (Fish Monitoring Centre, 2019). Few studies have explored the welfare of lumpfish from hatcheries to growing farms. One study reported housing conditions where the animals were transported from Iceland to Scotland and held in two flow-through larval rearing tanks. These tanks were 1.3 m³ round PVC black tanks (150 cm diameter × 80 cm depth) at stocking densities of 10–15% (~12 kg/m³). The temperature ranged from 8.2 °C (January) to 14.2 °C (October) suggesting a flow through system since temperature was not constant (Rey et al., 2021). No enrichment is described but it is possible these lumpfish could adhere to the bottom or sides of the tanks to meet their behavioral needs since wild individuals at this development stage are found adhering to rocks or algae. One study exploring the provision of shelters as enrichment within lumpfish tanks found that shelter provision reduced growth and body condition if food was delivered continuously rather than in pulses (Johannesen, Joensen & Magnussen, 2018). Reports have suggested that lumpfish prefer a black background rather than the typical blue or light grey aquaculture tank color (Garcia de Leaniz et al., 2015). There is still much to learn about the use of enrichment in lumpfish production.

Once lumpfish reach 20g they are deployed into Atlantic salmon sea cages (Imsland et al., 2016). They can be stocked at densities of 10–15% within the sea cages with studies reporting the use of artificial kelp (seaweed) to allow the lumpfish a surface to adhere to but also provide enrichment for the salmon within the cage (Imsland et al., 2020). Lumpfish spend a large proportion of time foraging during the day. They also have been observed affixed to floating seaweed or hovering under the seaweed (Imsland et al., 2014a). During the night, observations suggest they group together on smooth plastic or concrete substrates (Imsland et al., 2015a). Thus, providing suitable resting substrates seems key in meeting their behavioral needs. Lumpfish are more active in sea cages with salmon present. This is not a result of aggressive interactions between the two species and so they appear to coexist without having negative impacts upon each other (Imsland et al., 2014b). Indeed, lumpfish do eat sea lice from the salmon body, and one could then infer a mutualistically beneficial relationship between the two species in this artificial context. However, it has been estimated that one third of lumpfish suffer mortality from starvation (Powell et al., 2018). The ingestion of sea lice does appear to increase over time since one study showed that the percentage of lumpfish feeding on sea-lice increased from ~15% at day 11 to ~35% at day 77 (Imsland et al., 2015b). Thus, identifying these parasites as food may be a learned behavior that could be “trained” to juvenile lumpfish before deployment into sea cages. Researchers found that younger lumpfish fed more readily on sea lice and showed a higher weight gain than larger size classes (Imsland et al., 2016). Indeed reports suggest that when lumpfish reach a size greater than 200 g their impact on sea lice is negligible (Imsland & Reynolds, 2022). Studies suggest variable success with use of lumpfish as a parasite control tool with 9% to 97% reduction of sea lice reported (Imsland et al., 2018; Overton et al., 2020). However, given the very high mortality rates of lumpfish one could pose the ethical question as to whether the use of one sentient being that has a ~33% of mortality in the first few weeks of being placed in a sea cage is worth it if the sea lice levels on another sentient being are not significantly reduced? To combat the loss of lumpfish through starvation supplementary feeding occurs via pelleted food for the lumpfish. Pelleted food is sporadically added to the cage and can be lost if not immediately consumed so one study explored the use of hanging food blocks where lumpfish has access to food over a longer time period (Imsland et al., 2020). These blocks did result in improved health and survival.

Another problem that arises within the sea cage that is of significant welfare concern is the lumpfish are held in close proximity with parasitized salmon. Sea lice do not just parasitize salmon but also attack any species of cleaner fish within the sea cage including lumpfish (Garcia de Leaniz et al., 2022). The welfare issues are three-fold: firstly, the damage that the sea lice do to the animal and secondly, the salmon are subjected to chemical and mechanical treatments used to remove sea lice that can lead to mortality in some cases thus there is a reduction in the lumpfish “diet” after treatments and thirdly; the lumpfish are subject to the treatments that affect their welfare. It has been reported that all animals in the sea cage undergo these treatments, and there is a notable rise in lumpfish mortality following mechanical delousing procedures (Imsland et al., 2020). The researchers attribute this increase to the mechanical delousing methods employed, suggesting that lumpfish should not be included in mechanical sea lice treatment.

Lumpfish welfare assessment

Growth rates, body condition and survival (or mortality) rates have been used by several studies at different developmental stages of lumpfish including larval, on growing and adult lumpfish (Powell et al., 2018; Dahle et al., 2020; Geitung et al., 2020; Rey et al., 2021). Of course, the premise used here is that if the animals are growing, are in better condition and are surviving between different experimental groups one can identify the optimal welfare conditions to house or farm lumpfish in. Whilst this does indeed have value, this does not address mental wellbeing. Researchers explored four methods of measuring body condition and found it did not relate to fin damage and suggested that in this case the tested methods were not useful to measure damage from aggressive interactions (Rey et al., 2021).

Wounds or sores on the lumpfish body can reflect general health, aggressive behavior and/or parasite or disease prevalence. Other measures can be used to assess health such as the incidence of cataracts, missing eyes or eye ulceration or swelling, spinal deformities or deformities to the suction disc (Imsland et al., 2020). Fin damage also indicates potential welfare issues and can be caused by poor water quality, injuries or aggression. By quantifying the extent of body and/or fin damage a health score can be attributed to lumpfish held in different contexts. Scientists derived a health score from the assessment of body wounds and fin damage which they propose can be used as a means of understanding the state of the animals as well as deciding when to euthanize (Imsland et al., 2020).

Lumpfish fin damage is seen in larvae and juveniles due to agonistic interactions in captive conditions or induced by stress (Gutierrez Rabadan et al., 2021). Fin damage prevalence appears to be reduced in the sea cage environment (Brooker et al., 2018; Gutierrez Rabadan et al., 2021). The main cause of fin damage in sea cages has been attributed to aggression but since space is greater and there is possibly less competition due to the abundance of sea lice, fin damage is seen at lower levels than in the hatchery or on growing farms (Eliasen et al., 2020; Gutierrez Rabadan et al., 2021). Fin damage scales can provide a means of quantitively assessing welfare. For example, scores can be made of the dorsal, tail (caudal) or anal fins. Recently researchers used a four point scale where a value of 0 = no damage, 1 = marginal biting or fin splitting, 2 = major distal fin ray loss, 3 = complete absence of fin (Rey et al., 2021). This study found that fin damage is most prevalent at young developmental stages due to high aggression. Fin damage poses a serious risk to welfare as it is likely to cause pain and can lead to compromised movement and infections in the damaged fin tissue (Rey et al., 2021). Therefore, it is imperative to explore strategies aimed at reducing aggression and minimizing fin damage.

Very little is known about the behavioral repertoire of lumpfish and so there is a dearth of knowledge on this subject (Garcia de Leaniz et al., 2022). First, we need to understand what is normal behavior before we can understand what is abnormal. Aggression is high in hatchery and on growing farms and this declines in the sea cage environment. From the limited study of lumpfish behavior in sea cages it appears as if most of the daylight period is spent foraging with short bouts of resting whereas lumpfish aggregate to rest on smooth surfaces during the night period (Imsland et al., 2014a). Five behavioral indicators have been proposed as useful for on farm assessment and they include loss of appetite, erratic swimming, lethargy, increased aggression and gasping (usually seen at the surface and may be an attempt to increase oxygen uptake) (Gutierrez Rabadan et al., 2021; Garcia de Leaniz et al., 2022).