Conversation with Phil Brooke of Compassion in World Farming

Phil Brooke is the Research and Education Manager at Compassion in World Farming (CIWF). He spoke with ACE Researcher Kieran Greig on March 12, 2018. This is a summary of their conversation.

How is CIWF involved in fish welfare?

Compassion in World Farming (CIWF) has been interested in fish welfare for decades. The first report was in 1992, before I began working with CIWF. We now have a professional team of three full-time equivalents working on fish welfare, and I mentor that team.

Generally, what is the state of fish welfare today?

There was recently an EU report on the state of welfare in fish transport to slaughter, which was measured against the World Organisation for Animal Health (OIE) standard guidelines. Broadly speaking, EU legislation requires that cruelty to fish is avoided in transport and slaughter, but doesn’t provide details of how this should be achieved. OIE has published some guidelines for this. The EU has been checking the extent to which the key producers have been complying with OIE guidelines for transport and slaughter. The conclusion they reached was that the transport conditions are quite good. However, compliance with slaughter was poor or lacking, and only occasionally compliant. The British salmon industry was compliant, but that was the only one in the EU. Therefore, improving welfare of fish during slaughter is clearly an example of low-hanging fruit—it’s one place where we can make a big difference in reducing the amount of animal suffering.

Are there are significant issues or problems in the OIE guidelines?

The OIE guidelines are very common sense, but what they lack is detail. They give general rules, but don’t provide parameters for achieving those goals. If you follow their rules you’ll achieve good welfare, but you need to know a lot about each individual species to meet to those rules.

The RSPCA offers one of the best assurance schemes. They cover salmon and trout and offer good models. However, they don’t go far enough in all respects, so we have proposed five additional points that go beyond the RSPCA standards. We also believe there shouldn’t be killing of predators (and sometimes seals and cormorants) in the course of protecting farmed fish. Farmers should be choosing species which naturally live in ponds, and we believe in rearing vegetarian and omnivorous species rather than carnivorous ones.

On slaughter, we would like more guidelines, especially for individual species. There are already some general rules that say you must avoid suffering (more or less), but there needs to be more enforcement. OIE rules, which almost every country in the world has signed up for, are very helpful but they are only guidelines, not rules.

What are some of the best methods for killing farmed fish as humanely as possible?

Chapter 7.3 of OIE’s aquatic code outlines the general principles of slaughtering any species. For example, the final paragraph in Part 1 states that “farmed fish should be stunned before killing” and “should ensure immediate and irreversible loss of consciousness. If the stunning is not irreversible, fish should be killed before consciousness is recovered.” The EU interprets “immediate” as less than one second. So there are methods of stunning that are promising.

Stunning methods must be followed by an effective killing method. For example, salmon can receive an automatic percussive stun followed by a gill cut which leads to death. Where electrical stunning is followed by a gill cut, there is a risk that the animals will regain consciousness before they die. The recent European Commission report therefore recommended that electrical stunning should be followed by decapitation. Another alternative is to combine systems where an electrical stun is followed by a percussive stun, the two methods together reducing the chance of a mis-stun or of the animals recovering consciousness.

The OIE guidelines do not really go into species-specific details but each species requires its own parameters. There are some species-specific suggestions for methods though. A free-bullet or spiking (also called Ikejime) is used for Tuna. For large salmon, percussive stunning is used, whereas for small trout one would use electrical stunning. Catfish and eel are some of the the hardest fish to kill humanely. Some species are more commonly slaughtered humanely than others, for example salmon, but there can be mistakes. Electrical stunning is humane if done properly.

There is one other method that is not listed there that could potentially be humane but is not permitted in the EU at present—and that is using a food grade anesthetic. Australia and New Zealand use something similar to clove oil, which broadly speaking anesthetizes the fish with minimal stress involved. There are people that think this could be the best method of stunning, but they are not used because none of these products are licensed for use on farmed animals in the EU.

The OIE guidelines outline all of the other methods which they advise against, too. For example, chilling fish in a mixture of ice and water (which is widely used in the EU) or letting the fish asphyxiate via carbon dioxide is especially horrible for fish. CIWF has seen pictures in which fish are writhing 20 minutes after carbon dioxide has been applied. The OIE guidelines states “these methods should not be used if it is feasible to use the methods described above,” which is strong language for the OIE. The best stunning methods are the ones which have first been tested in the lab to determine the parameters that should be used.

How do farmers decide if a fish has been stunned properly?

The OIE guidelines also provide general principles to check if the animal is unconscious after stunning. There are not necessarily proper arrangements for checking if the stunning machines work properly. Big manufactures like Optimar work with institutes to check that their machines work properly. However, a small producer in Eastern Europe might not have that kind of check. One of the things that we need very much in the EU is a system for checking that equipment works properly, and protocols to follow for those using the equipment.

Other manufacturers check that animals have no behavioural evidence of consciousness in the field, and they may check later in the lab via EEG (electroencephalography). An EEG measures electrical activity in the brain. A current is applied and one can measure the brain activity before and after. This allows you to work out if the animal is unconscious or has recovered. It is best to be able to test using more than one method—the EEG is the gold standard method for checking that the behavioural test is accurate, but it can only be conducted in the lab.

Behavioural methods can be used in the field to check for rhythmic breathing or reflexes. They can also be used to check for vision by rolling the animal and seeing if they are trying to look at you. However, there are cases where no behavioural response is detected in the field but brain activity is still detected in an EEG two hours later, hence the need to apply both methods.

Establishing parameters to detect consciousness is important since these will vary by species. Broadly speaking the best methods of developing stunning and killing methods should be done in a lab experiment to determine parameters for the voltage and duration of application, among other things. Initial questions to ask are (i) whether producers have established parameters and (ii) if they are using these parameters and checking for behavioural evidence of consciousness after stunning in the field.

There is a need to develop good stunning and slaughter protocols for channel catfish, the main fish farmed in the US. Catfish are generally difficult to kill humanely, though protocols have been developed in the Netherlands for the North African catfish.

Apart from slaughter reforms, what are some other promising interventions or other ways of changing farming practices that would significantly improve fish welfare?

The first thing would be to enrich the environment. One needs to understand the ethology of species to know how they should be kept. There has been some recent research done at the University of Sterling with salmon. Curtains were placed into the water dividing the cage into lots of small corridors rather than one big cage, and the initial result was that the fish were dramatically less fearful.

There are also levels of dissolved oxygen which are preferable and which vary between species. One should ensure that the water temperature doesn’t change too much. In the sea, there isn’t suddenly a 10 degree change in temperature, which you might get in the air. Fish in general are not well adapted to sudden changes in temperature.

Another key principle is to ensure that the behavioural needs of the fish are provided. Many species may not be adapted to life in a confined area. For example, Atlantic salmon migrate thousands of miles in their lives—should we keep them in cages which prevent them from fulfilling this instinct? Solitary species are likewise unlikely to be adapted to being kept in crowded groups. Some species need access to the sea bottom. For example, halibut which live on the bottom of the sea need to have sediment at the bottom of their environment, not a cage. Otherwise the halibut will end up underneath each other because they’re trying to burrow into the sediment.

The next important thing is stocking density. There is evidence that above a certain level, high stocking density is bad for fish welfare. A paper by Turnbull et al. looking at salmon suggests a stocking density of 22–23 kg per cubic meter. The RSPCA thinks that the stocking density should be as low as 15 kg per cubic meter. For fish that naturally shoal, they might be happy to be in a high density system.

Effects of stocking density can be complicated in species that defend territories where they can, but otherwise shoal for mutual defence. For example, salmon and trout will try to protect a small area of a stream or river and they will be aggressive towards intruders. That means low stocking density would lead to more aggressive behaviour because with fewer other fish in the surrounding area, fish would be able to form and defend territories. In practice there are often more fish than available territory. Those who do not secure territory will shoal together for protection from aggressive territory-holding salmon as they swim through, because there’s security in numbers. This doesn’t mean a high stocking density is the best—there’s a threshold for any species. The higher above the shoal threshold, the more we see aggression and fin damage. As a rule of thumb, it is best to give fish more space so long as it doesn’t result in more aggressive territorial behaviour.

One top welfare issue that is often ignored—or only discussed under environment or food security—is the question of feed. No one would keep a lion as a source of food. A large amount of grain would be needed to feed a large number of pigs to feed a small number of lions. However, nearly all marine and freshwater fish that people want to eat are carnivorous. That means their diet has to contain fish meat and fish oils. This results in the killing of wild fish for feed such as menhaden, blue whiting, anchovies, and sand eels who die horribly and in very large numbers. The website fishcount.org.uk has estimated that 400 billion wildfish are reduced to fishmeal per year, without humane slaughter. This is a major welfare issue.

One could argue that fish such as anchovies and blue whiting are already fished to the maximum anyway so not using them in farming wouldn’t reduce the number who are actually used. As a long-term aim though, we should be trying to free farming of its need to kill fish superfluously. Fishmeal and oil could be produced from trimmings. When you kill a trout or a salmon, there are heads, guts, bones, and tails that people don’t eat and that generate fishmeal and oil. If the demand for this can be reduced, in the short term it will be sold to someone else, but in the long term this will be a welfare improvement. Something to watch for is that the feed doesn’t result in health or welfare problems. Farmed fish should be fed regularly, and they need to be fed a diet that doesn’t cause digestive problems. CIWF wants to reduce the use of fishmeal in feed, but if you give them too much soya instead, it can cause digestive problems. These problems can be reduced with better processing.

Another significant issue is the high rate of mortality. Take salmon for instance; over the last decade there has been a large increase in their death rate. In 2002, 4,500 tonnes of salmon died, but last year this was 25,000 tonnes. This is not due to an increase in farming—rather, it is caused by a range of fish diseases such as Amoebic Gill Disease. There has also been an increase in parasites such as fish lice. In the past these were managed using antiparasitic medicines, but these are toxic to marine invertebrate life and the lice are becoming immune to them.

More recently, farmers have experimented with mechanical treatments such as the Thermolicer. The fish are pumped out of their cage into a bath of warm water which is 20 degrees or so warmer than their environment. This is highly stressful and a proportion of the fish die in the process. Other stressful treatments such as scrubbing the fish or applying jets of pressurised water have also been tried to remove the lice. Some have suggested the answer is to take fish out of the sea and keep them in recirculation systems. Unfortunately, this is likely to require much higher stocking densities to make these systems economic. It is also what we did to pigs and chickens when we took them out of fields and into sheds—we call it factory farming. We don’t want to do the same thing to our fish. A better way would be to keep fish in their natural environments but not to keep so many, especially in environments where lice are a problem.

Any further thoughts on guidelines for higher welfare systems?

If there is such a thing as a good system for keeping fish, it would match the conditions in which they are naturally reared. Any good scheme requires good input rules: handling should be minimised, especially any time out of water; fish should have space; the environment should meet any needs for shelter and bottom substrate; the water quality should be good with minimum requirements for oxygen levels and limits to the levels of carbon dioxide and ammonia. There should also be breeding rules to ensure that the process of selectively breeding fish to grow faster doesn’t result in bad welfare.

There should also be welfare measurements in place for welfare outcomes. For example, these could include mortality rates, frequency of damaged fins, parasite loads, use of antibiotics and other medicines, how crowded the fish are when handled, and the percentage of mis-stuns during slaughter.

Are there are any other examples of promising ways to enrich the environment?

For any given species we would need to know what their natural stocking density is, whether they naturally shoal or are a solitary species, whether varying the food they eat might help them, and whether they naturally live in the middle or bottom of the water in the sea or a lake. Shelter should be provided for the fish to feel safe.

How can technology be used to decrease farmed fish suffering?

There are some people who’ve developed a laser for killing parasites. As a general view of farming practices, for instance for carp, we would prefer to see extensive systems where the fish retrieve food from their environment, rather than highly intensive stocking densities where farmers have to provide the food. People should farm vegetarian species or at least omnivorous species that can find their own food in the environment, like carp or tilapia.

What additional research would you most like to see?

We need proper humane slaughter parameters for each sort of fish. After that, further work on the ethology of fish and environmental enrichment would be useful so that we can better provide for the mental needs of fish.

When advocating for these changes, will you advocate for one reform at a time, or will this be a package?

We need comprehensive species-specific legislation to protect farmed fish during rearing, transport, and slaughter. This may be one piece of legislation or many. The legislation needs to be flexible to allow for improvements to be made as our understanding of the needs of fish increases.