Fish welfare

Introduction

Aquaculture is one of the few types of animal farming that has grown continuously over recent decades, by about 10% annually on an international level (Moffitt & Cajas-Cano 2014). However, as production increases and new methods appear, such as aquaponics, we have been witness to more problems related with fish health and welfare. Although it may seem surprising, more than 1300 scientific articles have been published on fish welfare since 1990 (see Table 2). Not all those studies deal with commercially produced species, but in general the number for all fish is comparable to or higher than some other species like sheep, horses or poultry.

Table 2: Summary of publications on animal welfare for different species of farm animals (based on a search in the Web of Science for the years 1990-2017)

One of the first scientific reviews of fish welfare was by Conte (2004) from the University of California at Davis, followed a few years later by two groups from the United Kingdom (Huntingford et al. 2006 and Ashley 2007). In his review, Conte (2004) underlines that fish farmers already know that welfare is important and that stress must be minimized since fish have specific requirements in terms of handling and environment outside of which they will not thrive or survive. That is to say, compared to terrestrial animals, fish are more demanding in terms of growing conditions and can be stressed easily, so much so that they can also die easily. Huntingford et al. (2006) summarize the main arguments for believing that fish can feel pain. Fish are complex beings that develop sophisticated behaviour, so the authors believe they can probably suffer, although it may be different in degree and type than for humans. That review ends up identifying four main critical areas when considering fish welfare: assuring that fish are not kept without water or food; assuring that producers provide good water quality and equipment; that their movements or behaviour are not restricted; and that mental and physical suffering be avoided. In his review, Ashley (2007) starts with a description of the industry and the critical points that may compromise fish welfare, including fish density in cages and problems with aggression. For example, some species, like tilapia, are more aggressive when kept at low densities than at higher densities. Importantly, Ashley (2007) provides a table of the main welfare problems in fish which is 7 pages long. In conclusion, there is a lot of scientific literature about fish welfare and several critical areas have been identified. However, regarding aquaponics, there are very few studies about the welfare of fish bred together with plants, but we can learn from other studies about the welfare of fish kept in small-scale recirculation systems.

Legislation in the EU

In Europe, any animal kept for the purpose of farming must comply with Directive 98/58/EC, which is a law that sets down several minimum conditions for adequate animal welfare for vertebrates. Although fish are technically included in that Directive, they are practically exempt due to our lack of knowledge about fish welfare, so there are no specific requirements for minimum conditions for fish used in aquaculture. Since 2006, several reports have been published in Europe, for example by the European Council of the European Food Safety Authority (EFSA), which give scientific recommendations for the most common species used in aquaculture. Overall, at least in Europe, there seems to be general agreement that fish undergo stress when oxygen levels are low and when they are taken out of water, and that chronic stress in fish compromises the immune system and can make them more vulnerable to disease.

Specific measures to evaluate welfare

Studies on fish welfare began later than for other farm animal species, in part because aquaculture is a younger animal production science and also since it was unclear to many whether fish can feel pain. Until recently, fish were not considered to be sensitive animals, but that situation has been changing. Sneddon (2003) was one of the first to prove that trout have pain receptors (nociceptors) on their face and jaw. She proved that those receptors respond to stimuli which are potentially damaging and send nervous signals to the spinal cord and brain. In addition, it appears that trout are aware of pain since they change complex behaviour when given a noxious substance, but revert to normal behaviour when given morphine (which essentially eliminates the pain). Those findings have also been confirmed in other species such as goldfish, where anxiety and fear decrease when they are given doses of morphine (Nordgreen et al. 2009). On the other hand, other scientists like Rose (2002) argue that fish cannot feel pain like humans since they lack a neocortex. Thus, they are probably not conscious about their pain in the same way as we are, although they react to pain in a similar manner. Whatever the case may be, both sides agree that fish can be stressed and that they have evolved a complex physiological response to stressors. Dawkins also makes the important point that everyone should worry about animal welfare whether or not they are conscious, simply because poor animal welfare leads to diseased and unhealthy fish, which has negative effects on farmers and consumers (Dawkins 2017).

The HPI axis and the stress response

The cascade of neuroendocrine activities that are released in fish after they become aware of a stressor is very similar to the responses seen in other vertebrates. As in mammals, the immediate neuroendocrine response is called the primary response and consists of nerve signals which release adrenaline and noradrenaline from chromaffin cells (at the head kidney), whose equivalent in mammals is the adrenal medulla (Figure 5). After the primary response, there is a slower secondary response which takes 2-15 minutes to activate the hypothalamo-pituitary-interrenal axis, or HPI axis (Sumpter et al. 1991), which in mammals is called the hypothalamo-pituitary-adrenal axis or HPA.

The hypothalamus produces corticotropin releasing hormone (CRH) which stimulates the production of adrenocorticotropic hormone (ACTH) by the anterior pituitary, also called the adenohypophysis. ACTH is released into the blood stream and simulates the production of cortisol by the interrenal tissue (also associated with the kidneys in fish), which corresponds with the adrenal cortex in mammals (Okawara et al. 1992). The secondary response includes an increase in heart frequency, greater oxygen uptake by the gills, and an increase in glucose concentration in plasma via glucogenolysis (Pickering & Pottinger 1995). Both the primary and secondary response systems help to maintain homeostasis after a stress by providing energy and increased levels of oxygen to the brain so that the body can adjust and return to normal or basal metabolic function.

Although there is no simple relationship between stress and welfare, we know that they are related and that the response to a stressor can be used to give an idea about the degree of the challenge.

With that in mind, it is always preferable to consider several indicators at the same time, including growth indices, immune system response, and other physiological indicators.

Figure 5: The HPI axis in fish and the cascade of responses to a stressor (source M. Villarroel) (CRH = corticotropin releasing hormone, ACTH = adrenocorticotropic hormone)

Operational welfare indicators

On an industrial level, a new approach is being developed to analyse fish that involves interactions between scientists studying animal welfare and companies that strive to be more efficient. Together they are developing operational welfare indicators (OWI). A good example for salmon is the manual presented by Noble et al. (2018) that tells farmers how to evaluate on a commercial level the immediate environment, different groups of fish, and individual fish. As mentioned above, many scientific articles have been published on fish welfare, most of which are based on observations made in the lab. OWI are practical indicators that are used on the farm and can be easily explained and repeated. OWI can be separated into two large groups: those more related to the environment; and those related to the fish. The latter can be applied to groups of fish, or individually. Finally, individual indicators can include laboratory analyses which are less operational per se but can provide useful information in the short term (see Figure 6). OWI can provide an idea of the current status of production in terms of the needs of the fish and their welfare. In parallel, they can be used to help develop good practice and to identify critical points that can have a negative effect.

Figure 6: Summary of the operational indicators used on fish farms, including indicators that vary with the environment and the animal. The animal-based indicators can be based on groups of fish or on individuals, and individual indicators can include laboratory analyses

In general, aquaculturists use feeding as an indirect indicator of welfare. That is, one approaches the tank and provides food, and the fish respond by going to the surface and eating, which is a good sign. If the fish do not come to eat, they have lost their appetite for some reason and more information is needed. Although there is plenty of equipment that can be purchased to feed fish automatically, it is recommended to feed fish at least once a day by hand in order to get an idea about how they are doing. If the fish do not eat, that will affect their weight gain, which is also relatively easy to measure. Another operational indicator that is common in fish farms is the coefficient of condition in live weight (the live weight in grams divided by the cube of the fork length in centimetres cm³). It indicates nutritional status (Bavčević et al. 2010), and gives an idea about the amount of intraperitoneal fat. The hepato-somatic index (HSI) is defined as the ratio between the weight of the liver and the live weight. During periods of fasting, the needs for energy are met mostly by mobilising glycogen reserves from the liver, while the fat reserves are left more or less untouched during the first few days (Peres et al. 2014). Thus, HSI can be used to indicate energy reserves since the liver is an important regulator of nutrient use in fish (Christiansen & Klungsøyr 1987).

Copyright © Partners of the Aqu@teach Project. Aqu@teach is an Erasmus+ Strategic Partnership in Higher Education (2017-2020) led by the University of Greenwich, in collaboration with the Zurich University of Applied Sciences (Switzerland), the Technical University of Madrid (Spain), the University of Ljubljana and the Biotechnical Centre Naklo (Slovenia).