When technology meets nature: Smarter eel passage at hydroelectric power plants

Photographer: Jörgen Wiklund

Many hydroelectric power plants hinder the natural migration of eels, threatening populations and creating both ecological and economic challenges. How can we make passage measures more effective and sustainable? This webinar presents new research that uses machine learning to understand eel behavior.

The results from the project Using machine learning for improved eel downstream passage design could be used to develop better measures for eel passage at hydroelectric power plants throughout Sweden. It acts as a first step toward predictive models that can test and improve measures before they are built.

In this lunch webinar, Joschka Wiegleb from Karlstad University, will talk about the movement patterns of silver eels and how they are affected by environmental factors. David Aldvén, Biologist at Vattenfall Research and development in Älvkarleby, will participate as an industry representative, discussing with Joschka Wiegleb how the projects’ findings can be implemented more quickly and efficiently.

The project is based on a unique dataset from the river Ätran in 2017, where 98 silver eels were tagged with acoustic transmitters with depth sensors and tracked as they passed through the Herting power plant, both via a deflector with an escape opening and a natural fishway. Now, the researchers are building a tool based on an Artificial Neural Networks (ANN), that learns to predict the eel downstream movement in response to the present hydraulic and weather conditions. Through high-resolution data and advanced analysis, this project will improve the understanding of eel behavior and support both ecological and economic decisions.

The webinar is aimed at researchers, authorities, consultants, and anyone working with hydropower, fish connectivity, and environmental measures. You will gain insights into how modern technology and advanced data analysis can contribute to more sustainable solutions for hydropower and nature.