Last week, the Bezos Earth Fund announced the winners of its $30 million AI Grand Challenge Awards, where 15 trailblazing projects were selected to receive a sliver of the funding, including two Yale University teams developing novel climate solutions.
The Yale-affiliated teams are working on a forecasting system for marine carbon dioxide removal (mCDR) solutions, as well as a methane emissions reduction approach for livestock operations.
Each of the teams is set to receive $2 million from the Bezos Earth Fund within the next two years, in addition to an initial $50,000 award secured by each of the teams earlier this year.
One of the awarded teams is a research group in the Yale Center for Natural Carbon Capture (YCNCC) and the Department of Earth & Planetary Sciences in Yale’s Faculty of Arts and Sciences (FAS) that has come up with an AI-enabled model to predict the carbon removal potential of ocean-based CO2 solutions.
Led by Elizabeth Yankovsky, assistant professor of Earth and planetary sciences, and YCNCC associate research scientist Luke Gloege, the team is developing an mCDR forecasting “Stack” that adds a machine learning component to MRV simulations.
This tool unifies and builds onto existing AI and physics models from various disciplines, combining a new, graphics processing unit (GPU)-optimized ocean model and a biogeochemistry model that are tailored to CDR simulations, utilizing AI-derived atmospheric forecasts developed through NVIDIA systems.
As a result, mCDR Stack users will be able to see a day-by-day forecast of net CO2 removed, along with the durability of that removal over a user-indicated amount of time.
The Stack also sets an industry precedent by providing decadal-scale forecasts in hours and millennia-scale forecasts in days.
Relevant: The Bezos Earth Fund And RMI Publish A Detailed Roadmap For Scaling CDR By 2050
The second Yale project to win the Bezos Earth Fund award is a partnership between the Alliance of Biodiversity International and the International Center for Tropical Agriculture (CIAT) and Purushottam Dixit of the Yale School of Engineering & Applied Science, which is working on a “Rumen Digital Twin” that leverages AI to reduce livestock methane emissions.
Using a Yale-developed AI component and publicly available data, this tool will allow users to create virtual, detailed livestock groups, enabling the identification of characteristics and conditions under which specific feed additives or components will most effectively reduce methane emissions.
Read more: Ocean Visions Is Expanding The Global Conversation On Ocean Climate Solutions
