Artificial intelligence (AI) can be harnessed for sustainable hydropower development across the entire Amazon basin, according to new research in Science from an international team of researchers.
The work shows how AI can identify proposed hydropower plants that are likely to be particularly detrimental to the environment. It also reveals the lost environmental benefits from the 158 existing dams in the basin, which were originally developed without consideration for the cumulative effects they would have on basin-wide ecosystems — and the people who depend on them. The lost environmental benefits include access to food sources, clean water, cultural practices and livelihood support for the 30 million rural and urban people who are sustained by the Amazonian river systems.
“The Amazon basin has suffered the effects of uncoordinated energy development,” said Héctor Angarita, postdoctoral scholar at the Natural Capital Project and co-author on the paper. “To maintain the wealth of ecosystem and societal values that it provides both regionally and globally, there needs to be a coordinated effort across the Amazon countries in the development of their own energy portfolios. This research provides a tool to achieve that goal.”
The team set out to tackle a key challenge of working in the Amazon: the sheer size and diversity of the river basin. They wanted to understand what would happen in an unlimited number of different energy development scenarios — factoring in climate, energy and ecosystem health. They discovered that by taking advantage of the interconnectedness of river systems, they could design new AI algorithms to break the problem into parts without losing sight of the entire system. Their algorithms allow scientists, government planners and leaders to explore unprecedented levels of detail and accuracy to inform renewable energy development decisions.
Led by Cornell University biologist Alexander Flecker and Cornell computer scientist Carla Gomes, the project considers six socio-environmental criteria for optimization of the existing 158 dams and the more than 350 proposed hydroelectric dams in the Amazon basin: river flow; river connectivity; sediment transport; fish diversity; greenhouse gas emissions; and energy production.
“AI is being used by Wall Street, by social media, for all kinds of purposes,” said Gomes. “Why not use AI to tackle serious problems like sustainability?”
This research emphasizes the importance of strategic planning at the scale of the entire Amazon basin. It provides insights about the largest and most biodiverse transboundary river basin in the world, which spans eight countries, and shows why international cooperation is important for basin-wide planning in the Amazon.
“One of the frontiers we are pushing in this paper is around scalability. The AI can contribute to a shared dialogue across scales — meaning that a country can assess their proposed energy portfolio to the entire Amazon river basin, and vice versa” said Angarita. “It enables a type of coordination that hasn’t been accessible before now.”
The new work demonstrates how AI can be applied to other renewable energy projects as societies worldwide seek to transition away from fossil fuels.
The paper features a total of 40 co-authors from more than two dozen academic institutions in the U.S., Europe and South America, along with NGOs including The Nature Conservancy and the Wildlife Conservation Society.
“What makes this work special is that we’re bringing together so many different areas of expertise, and real Amazon experts, in areas such as ecology, fisheries biology, hydrology, social science, and computer science and AI,” said Flecker.
For additional information, see this Cornell Chronicle story.