Fish Welfare: Who Do We Help First?

Bahasa Indonesia | Tagalog | ไทย

Studying something as broad as “fish welfare” is like studying “mammal welfare”—mice and elephants have very different needs for food, water, and shelter. In the same way, different fish species have different needs for the temperature, pH, and oxygen levels in their water. Despite farming tens of billions of fishes per year, we hardly know the most basic factors contributing to their welfare. In this paper, researchers at the Fish Welfare Initiative studied these factors, as well as the most effective welfare improvements for several different species.

The researchers began by examining 26 different countries where fish farms operate. They evaluated these countries based on several criteria, like the ease of starting a new welfare organization, the ability to influence local governments, and local attitudes toward fish welfare. They found six countries that looked promising for welfare improvements: India, Bangladesh, Vietnam, Indonesia, Taiwan, and the Philippines. 

The researchers narrowed their focus further on the fishes most commonly farmed in these countries: catla, striped catfish, rohu, nile tilapia, milkfish, and torpedo-shaped catfish. These fishes are farmed in such large numbers because they can survive in harsh conditions. However, this just means that they won’t die as easily from their conditions, not that they won’t suffer. A true improvement to fish welfare wouldn’t just mean keeping them alive—it would mean preventing them from suffering.

The researchers then looked at five different factors to determine which species would benefit the most from a welfare improvement: sensitivity, poor living conditions, total farmed fish population, neglectedness, and tractability. They named tractability as the most important—an intervention was considered tractable if there was solid evidence that it could make progress toward better welfare. The researchers also intentionally left out any factors related to sentience. One of their goals was to help shape the future of fish welfare research, and they didn’t want to suggest that we should check each species for sentience before studying welfare improvements. Fish sentience is still debated in some parts of the scientific community, but proving it shouldn’t be a prerequisite.

The researchers then applied a weighted factor model, in which each factor was given a weight that contributes to a final score. For example, “tractability” was given a weight of 25% and “poor living conditions” was given a weight of 20%, showing that tractability is slightly more important for improving welfare. More specifically, time spent researching tractability is slightly more valuable than time spent researching living conditions. Based on further research, they assigned scores from one to three to each factor for each species. For example, the milkfish was given a score of three for the factor “poor living conditions,” suggesting that milkfish living conditions are exceptionally poor. Applying the weights, they calculated an overall score for each species that determined how effective a welfare intervention would be. You can find a link to all of these calculations here. Their results showed that catla would likely benefit the most from a welfare intervention due to the large number of farmed catla (estimated to be anywhere from 1.5 to 9.9 billion fishes in 2018) and their poor living conditions on fish farms.

Overall, the researchers found that there wasn’t a lot of information available about welfare conditions on fish farms. Fish farmers don’t often document and publish the conditions on their farms, and there’s been very little research into conditions as simple as the optimal water temperature for given species. Rather than creating a definitive guide to the best fish welfare improvements, the researchers hope that their work will help bring more attention to the understudied cause of fish welfare. As more organizations begin researching fish welfare, we’ll have a much better picture of what healthy fish living conditions are, and be able to craft appropriate advocacy interventions.