This past summer was marked by intense wildfires, heat waves, flooding, and drought around the world. How much more extreme weather can we expect in the coming decades? It’s hard to say. Climate change is complex, and the farther ahead we look, the cloudier the projections get. Globally, this uncertainty has often been used as an excuse to put off expensive or unpopular measures like building seawalls, retreating inland, and upgrading or relocating electrical grids, transit systems, and other critical infrastructure.
To bring greater precision to climate modeling and encourage societies to prepare for the inevitable disruptions ahead, the National Science Foundation (NSF) has selected Columbia to lead a climate modeling center called Learning the Earth with Artificial Intelligence and Physics (LEAP). In collaboration with the National Center for Atmospheric Research (NCAR) and NASA’s Goddard Institute for Space Studies (GISS), the center will develop the next generation of data-driven physics-based climate models. It will also train a new wave of students fluent in both climate science and working with big datasets and modern machine-learning algorithms. The center’s larger goal is to provide actionable information for societies to adapt to climate change and protect the most vulnerable.
“Until climate models can offer more precise projections, at the regional level where planning decisions are made, it will be difficult to make the billion-dollar investments needed to adapt,” said Columbia President Lee C. Bollinger. “I can think of no better university than Columbia, with its interdisciplinary focus, to lead the way in tackling the climate prediction problem.”
The center will be led by Columbia Engineering, its Faculty of Arts of Sciences, and Lamont-Doherty Earth Observatory, in collaboration with Teachers College, Columbia Business School, the School of Social Work, and the new cross-cutting Columbia Climate School.
Global climate models agree that the planet will continue to warm in the next 40 years. But they disagree on how much, and how severe the impacts will be, from sea-level rise to an increase in floods and drought. Much of the problem comes down to trying to represent the details of complex physical and biological processes — like clouds reflecting sunlight into space or trees absorbing carbon from the air —into the models. Processes interact, and many are poorly understood.
With the help of big data and machine learning, researchers will dig deeper into these processes and update the models with new knowledge to improve climate projections. Researchers will harness existing algorithms to analyze satellite images and other large-scale observational data missing from today’s models. They will also develop new algorithms to take detailed observations and generalize them to broader contexts, discover cause and effect relationships in the data, and find better equations to describe the processes represented in the models. As new knowledge gets woven into the models, researchers will use machine-learning tools to test their predictions.
In addition to NCAR and GISS, Columbia researchers will collaborate with peers at New York University, and universities of California at Irvine, Minnesota, and Montreal to update the NSF-funded and NCAR-based Community Earth System Model.
More accurate modeling will allow researchers to gaze farther into the future, and at closer geographic range. “Let’s say I want to know the number of heat-wave days in NYC in 2050,” said center director Pierre Gentine, a professor of earth and environmental engineering. “Some models may say 10 days. Some say 20 days. How are you going to adapt your electrical grid to avoid blackouts?”
“We still have these huge cones of uncertainty,” added deputy director Galen McKinley, a professor of earth and environmental sciences who is based at Lamont-Doherty, part of the Columbia Climate School. “Our goal is to harness data from observations and simulations to better represent the underlying physics, chemistry, and biology of Earth’s climate system. More accurate models will help give us a clearer vision of the future.”
Dealing with massive data requires a modern infrastructure. In collaboration with Google Cloud and Microsoft, Ryan Abernathey, an associate professor of earth and environmental sciences based at Lamont-Doherty, will create a platform to allow researchers to share and analyze data.
To achieve its mission, the center will create a new discipline merging climate science with data science and AI. Carl Vondrick, an assistant professor of computer science, will lead the team developing new algorithms to advance the science of climate change and improve current models. Tian Zheng, a statistics professor and chair of the department, will lead a team developing undergraduate and graduate educational programs that blend climate science and data science, and classroom learning with research experience.
Climate change will affect all of humanity, yet most climate science research to date has failed to draw from the full diversity of people and ideas available. To address the imbalance, the center will make diversity, equity, and inclusion central to its research and education mission. Courtney D. Cogburn, an associate professor of social work and a member of the Data Science Institute, will oversee diversity, equity, and inclusion initiatives and the transfer of knowledge to communities most vulnerable to climate change.
“Confronting the threat of climate change will require our brightest minds and best technologies,” said Columbia Provost Mary Boyce. “The LEAP center, with its entwined research and education mission, will play a pivotal role in getting us there.”
Columbia will play a role in three of the six Science and Technology Centers that NSF chose to fund this year. In addition to LEAP, Columbia will join Woods Hole Oceanographic Institution in studying how ocean microbes impact carbon cycling and climate, and the University of Washington in developing optoelectronic materials.
This story originally appeared in Columbia News.