Computational Model for Best Dam Placement

As the world is moving away from non-renewable energy sources, hundreds of hydropower dams have been proposed as an alternative in the ecologically sensitive and largely untapped resource of the Amazon basin. However, hydropower is not completely green, as about 10% of dams emit just as much greenhouse gas (GHG) per unit energy as traditional fossil fuel plants, and some existing dams in the Amazon basin emit 10 times more than traditional energy plants (Almeida et al., 2019).  To optimize hydropower dam placements and ensure they will be a clean source of energy, in 2019 a Cornell-lead team of ecologists and computer scientists worked with researchers in South America to develop an artificial intelligence computational model to find the best possible geographical locations amongst 350 possible dam sites. The AI model considers the ecology of the entire Amazon basin to protect biodiversity while generating the combination of dam sites producing the least amount of greenhouse gas per energy unit. This is crucial to maintain biodiversity and protect the region fondly nicknamed “Earth’s Lungs”. Interesting results have revolved around dam construction at higher elevations because less land needs to be flooded to generate the same unit of energy. Whether dams are environmentally friendly or not is still a hot debate, but this new AI computation model could be key to reducing carbon emissions from dam networks in the tropics. 

Almeida, R. M., Shi, Q., Gomes-Selman, J. M., Wu, X., Xue, Y., Angarita, H., … Flecker, A. S. (2019). “Reducing greenhouse gas emissions of Amazon hydropower with strategic dam planning“.

Lefkowitz, M. (2019, September 19). “AI helps shrink Amazon dams’ greenhouse gas emissions“.


Artificial Intelligence, Climate Change, Internet of Things, Pollution