What Will the Climate Look Like in 10 Years?

Elizabeth Barnes Explores how Data-driven Models can Improve Forecasts from Days to Decades

By Molly Glass, The Brink

Predicting climate is incredibly complex because it depends on huge, constantly changing datasets like temperature, wind, and ocean conditions. Since both AI and Earth science deal with massive data and pattern-finding, combining them can help scientists better understand how Earth’s systems interact and improve forecasts of weather and climate over longer timescales.

“As I tell my students, if you’re trained as an Earth scientist, you’re already trained to think about big data—to think about finding relationships, to understand that correlation is not causation,” says Elizabeth “Libby” Barnes, the inaugural Dalton Family Chair in Environmental Data Science & Sustainability at Boston University, a Professor of Computing and Data Sciences, and Earth & Environment, and a Faculty Affiliate of the Hariri Institute for Computing. “So, when you learn how to use AI tools, you immediately start to see the connections to your own work, because that’s what we do every day.”

Boston University researcher Elizabeth “Libby” Barnes and her team are using AI to investigate how Earth’s interconnected systems interact and evolve with the aim of advancing climate and extreme weather predictions across timescales from days to decades.

The researchers have found large-scale data models can be useful in predicting hurricanes across years and seasons. They’ve also developed a way to combine machine learning and analog forecasting to improve multiyear climate predictions. And they’ve collaborated on a project that fused climate model simulations and trade data to assess how extreme weather in one country might ripple out and affect many others.

Learn more in The Brink story by Molly Glass here.