New technology could soon help protect fish farmers around Africa’s largest lake by providing early warnings before dangerous water conditions lead to mass fish deaths.
Researchers from King’s College London recently recorded the warning signs of a major low-oxygen event in Lake Victoria only hours before local communities reported large numbers of dead fish.
The finding highlights how real-time monitoring could one day give fish farmers enough time to act before disasters strike.
The discovery happened shortly after researchers installed pilot water-quality sensors at aquaculture sites near Dunga Beach in Kisumu, Kenya.
The sensors were designed to continuously measure water conditions, including the amount of dissolved oxygen, which fish need to survive.
One day, the sensors detected something unusual. Dissolved oxygen levels suddenly dropped to almost zero. At first, the research team thought the equipment might be faulty because the readings were so extreme.
However, the following morning, local fish farmers reported that many fish had died in the same area where the sensors had detected the oxygen crash.
The project is part of the FRESH-WQ program, a partnership between King’s College London, the Kenya Marine and Fisheries Research Institute (KMFRI), and the African Center for Aquatic Research and Education (ACARE).
The goal is to create an affordable early warning system that combines water sensors, satellite images, and artificial intelligence (AI) to predict dangerous water conditions before they cause major losses.
Fish need oxygen dissolved in the water, just as humans need oxygen in the air. When oxygen levels become too low, a condition called hypoxia develops. Fish can quickly become stressed, suffocate, and die if oxygen levels remain dangerously low. These events can be devastating for fish farmers who rely on healthy fish for their income and food supply.
Scientists believe that pollution, climate change, and other environmental pressures may increase the risk of low-oxygen events in Lake Victoria and other freshwater lakes. As a result, finding ways to predict these events has become increasingly important.
Unlike weather forecasting, which can warn people about storms, floods, or droughts, there are very few systems that can predict changes in water quality. This is especially true in many parts of Africa, where monitoring equipment is limited.
During the project, researchers also met with fish farmers, community members, and local experts to learn how they currently monitor lake conditions. Many said they rely on traditional knowledge, such as watching changes in the water’s color, observing fish behavior, and noticing other natural signs. While these methods are valuable, participants said they would greatly benefit from receiving warnings days or even weeks before dangerous conditions develop.
The research team hopes to build a forecasting system that combines information from underwater sensors, satellites, and AI. By analyzing these different sources together, the system could recognize patterns that signal worsening water quality and send early alerts to local communities.
Researchers believe this technology could help protect fisheries, support aquaculture businesses, and improve water management across the African Great Lakes region.
For the scientists involved, the project is about much more than developing new technology. They hope it will strengthen partnerships with local communities, encourage citizen science, and train the next generation of researchers and engineers.
Most importantly, they believe that giving people advance warning of environmental hazards could help protect both livelihoods and the freshwater ecosystems that millions of people depend on.
