{"679801":{"#nid":"679801","#data":{"type":"news","title":"At the Intersection of Climate and AI, Machine Learning is Revolutionizing Climate Science","body":[{"value":"\u003Cp dir=\u0022ltr\u0022\u003EExponential growth in big data and computing power is transforming climate science, where machine learning is playing a critical role in mapping the physics of our changing climate.\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003E\u0026nbsp;\u201cWhat is happening within the field is revolutionary,\u201d\u0026nbsp;says\u0026nbsp;\u003Ca href=\u0022https:\/\/eas.gatech.edu\u0022\u003ESchool of Earth and Atmospheric Sciences\u003C\/a\u003E\u003Cstrong\u003E\u0026nbsp;\u003C\/strong\u003EAssociate Chair and Professor\u0026nbsp;\u003Ca href=\u0022https:\/\/sites.gatech.edu\/annalisabracco\/\u0022\u003E\u003Cstrong\u003EAnnalisa Bracco\u003C\/strong\u003E\u003C\/a\u003E, adding that because many climate-related processes\u0026nbsp;\u2014 from ocean currents to melting glaciers and weather patterns\u0026nbsp;\u2014 can be described with physical equations, these advancements have the potential to help us understand and predict climate in critically important ways.\u0026nbsp;\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003EBracco is the lead author of a new review paper providing a comprehensive look at the intersection of AI and climate physics.\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003EThe result of an international collaboration between Georgia Tech\u2019s Bracco,\u0026nbsp;\u003Cstrong\u003EJulien Brajard\u003C\/strong\u003E (Nansen Environmental and Remote Sensing Center),\u0026nbsp;\u003Cstrong\u003EHenk A. Dijkstra\u003C\/strong\u003E (Utrecht University),\u0026nbsp;\u003Cstrong\u003EPedram Hassanzadeh\u003C\/strong\u003E (University of Chicago),\u0026nbsp;\u003Cstrong\u003EChristian Lessig\u003C\/strong\u003E (European Centre for Medium-Range Weather Forecasts), and\u0026nbsp;\u003Cstrong\u003EClaire Monteleoni\u003C\/strong\u003E (University of Colorado Boulder), the paper, \u2018\u003Ca href=\u0022https:\/\/www.nature.com\/articles\/s42254-024-00776-3\u0022\u003EMachine learning for the physics of climate\u003C\/a\u003E,\u2019\u0026nbsp;was\u0026nbsp;recently published in\u0026nbsp;\u003Cem\u003ENature Reviews Physics\u003C\/em\u003E.\u0026nbsp;\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003E\u201cOne of our team\u2019s goals was to help people think deeply on how climate science and AI intersect,\u201d Bracco shares. \u201cMachine learning is allowing us to study the physics of climate in a way that was previously impossible. Coupled with increasing amounts of data and observations, we can now investigate climate at scales and resolutions we\u2019ve never been able to before.\u201d\u003C\/p\u003E\u003Ch3\u003E\u003Cstrong\u003EConnecting hidden dots\u003C\/strong\u003E\u003C\/h3\u003E\u003Cp dir=\u0022ltr\u0022\u003EThe team showed that ML is driving change in three key areas: accounting for missing observational data, creating more robust climate models, and enhancing predictions, especially in weather forecasting. However, the research also underscores the limits of AI \u2014 and how researchers can work to fill those gaps.\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003E\u201cMachine learning has been fantastic in allowing us to expand the time and the spatial scales for which we have measurements,\u201d says Bracco, explaining that ML could help fill in missing data points \u2014 creating a more robust record for researchers to reference. However, like patching a hole in a shirt, this works best when the rest of the material is intact.\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003E\u201cMachine learning can extrapolate from past conditions when observations are abundant, but it can\u2019t yet predict future trends or collect the data we need,\u201d Bracco adds. \u201cTo keep advancing, we need scientists who can determine what data we need, collect that data, and solve problems.\u201d\u003C\/p\u003E\u003Ch3\u003E\u003Cstrong\u003EModeling climate, predicting weather\u003C\/strong\u003E\u003C\/h3\u003E\u003Cp dir=\u0022ltr\u0022\u003EMachine learning is often used when improving climate models that can simulate changing systems like our atmosphere, oceans, land, biochemistry, and ice. \u201cThese models are limited because of our computing power, and are run on a three-dimensional grid,\u201d Bracco explains: below the grid resolution, researchers need to approximate complex physics with simpler equations that computers can solve quickly, a process called \u2018parameterization\u2019.\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003EMachine learning is changing that, offering new ways to improve parameterizations, she says. \u201cWe can run a model at extremely high resolutions for a short time, so that we don\u2019t need to parameterize as many physical processes \u2014 using machine learning to derive the equations that best approximate what is happening at small scales,\u201d she explains. \u201cThen we can use those equations in a coarser model that we can run for hundreds of years.\u201d\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003EWhile a full climate model based solely on machine learning may remain out of reach, the team found that ML is advancing our ability to accurately predict weather systems and some climate phenomena like El Ni\u00f1o.\u0026nbsp;\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003EPreviously, weather prediction was based on knowing the starting conditions \u2014 like temperature, humidity, and barometric pressure \u2014 and running a model based on physics equations to predict what might happen next. Now, machine learning is giving researchers the opportunity to learn from the past. \u201cWe can use information on what has happened when there were similar starting conditions in previous situations to predict the future without solving the underlying governing equations,\u201d Bracco says. \u201cAnd all while using orders-of-magnitude less computing resources.\u201d\u003C\/p\u003E\u003Ch3\u003E\u003Cstrong\u003EThe human connection\u003C\/strong\u003E\u003C\/h3\u003E\u003Cp dir=\u0022ltr\u0022\u003EBracco emphasizes that while AI and ML play a critical role in accelerating research, humans are at the core of progress. \u201cI think the in-person collaboration that led to this paper is, in itself, a testament to the importance of human interaction,\u201d she says, recalling that the research was the result of a workshop organized at the\u0026nbsp;\u003Ca href=\u0022https:\/\/www.kitp.ucsb.edu\/\u0022\u003EKavli Institute for Theoretical Physics\u003C\/a\u003E \u2014 one of the team\u2019s first in-person discussions after the Covid-19 pandemic.\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003E\u201cMachine learning is a fantastic tool \u2014 but it\u0027s not the solution to everything,\u201d she adds. \u201cThere is also a real need for human researchers collecting high-quality data, and for interdisciplinary collaboration across fields.\u003Cstrong\u003E\u0026nbsp;\u003C\/strong\u003EI see this as a big challenge, but a great opportunity for computer scientists and physicists, mathematicians, biologists, and chemists to work together.\u201d\u003C\/p\u003E\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003E\u003Cem\u003E\u003Cstrong\u003EFunding\u003C\/strong\u003E: National Science Foundation, European Research Council, Office of Naval Research, US Department of Energy, European Space Agency, Choose France Chair in AI.\u003C\/em\u003E\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003E\u003Cem\u003E\u003Cstrong\u003EDOI\u003C\/strong\u003E:\u0026nbsp;\u003C\/em\u003E\u003Ca href=\u0022https:\/\/doi.org\/10.1038\/s42254-024-00776-3\u0022\u003E\u003Cem\u003Ehttps:\/\/doi.org\/10.1038\/s42254-024-00776-3\u003C\/em\u003E\u003C\/a\u003E\u003C\/p\u003E\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp dir=\u0022ltr\u0022\u003EA Georgia Tech-led review paper recently published in\u0026nbsp;\u003Cem\u003ENature Reviews Physics\u003C\/em\u003E is exploring the ways machine learning is revolutionizing the field of climate physics \u2014 and the role human scientists might play.\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"A Georgia Tech-led review paper recently published in\u00a0Nature Reviews Physics is exploring the ways machine learning is revolutionizing the field of climate physics \u2014 and the role human scientists might play."}],"uid":"35599","created_gmt":"2025-01-22 17:43:30","changed_gmt":"2026-01-01 18:31:44","author":"sperrin6","boilerplate_text":"","field_publication":"","field_article_url":"","location":"Atlanta, GA","dateline":{"date":"2025-01-22T00:00:00-05:00","iso_date":"2025-01-22T00:00:00-05:00","tz":"America\/New_York"},"extras":[],"hg_media":{"676086":{"id":"676086","type":"image","title":"Researchers launch a a lightweight, balloon-borne instrument to collect data. \u0022To keep advancing, we need scientists who can determine what data we need, collect that data, and solve problems,\u0022 Bracco says. (NOAA)","body":"\u003Cp\u003EResearchers launch a a lightweight, balloon-borne instrument to collect data. \u0022To keep advancing, we need scientists who can determine what data we need, collect that data, and solve problems,\u0022 Bracco says. (NOAA)\u003C\/p\u003E","created":"1737567826","gmt_created":"2025-01-22 17:43:46","changed":"1737567826","gmt_changed":"2025-01-22 17:43:46","alt":"Researchers launch a a lightweight, balloon-borne instrument to collect data. \u0022To keep advancing, we need scientists who can determine what data we need, collect that data, and solve problems,\u0022 Bracco says. (NOAA)","file":{"fid":"259801","name":"noaa-5hZJVGPG6vo-unsplash.jpg","image_path":"\/sites\/default\/files\/2025\/01\/22\/noaa-5hZJVGPG6vo-unsplash.jpg","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2025\/01\/22\/noaa-5hZJVGPG6vo-unsplash.jpg","mime":"image\/jpeg","size":2094496,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2025\/01\/22\/noaa-5hZJVGPG6vo-unsplash.jpg?itok=KR8SZhoH"}}},"media_ids":["676086"],"groups":[{"id":"1188","name":"Research Horizons"},{"id":"367481","name":"SEI Energy"},{"id":"1280","name":"Strategic Energy Institute"}],"categories":[{"id":"153","name":"Computer Science\/Information Technology and Security"},{"id":"144","name":"Energy"},{"id":"154","name":"Environment"},{"id":"150","name":"Physics and Physical Sciences"},{"id":"135","name":"Research"}],"keywords":[{"id":"192258","name":"cos-data"},{"id":"192254","name":"cos-climate"},{"id":"192252","name":"cos-planetary"},{"id":"187915","name":"go-researchnews"},{"id":"186858","name":"go-sei"}],"core_research_areas":[{"id":"193655","name":"Artificial Intelligence at Georgia Tech"},{"id":"39531","name":"Energy and Sustainable Infrastructure"},{"id":"193653","name":"Georgia Tech Research Institute"}],"news_room_topics":[{"id":"71911","name":"Earth and Environment"}],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003EWritten by \u003Ca href=\u0022mailto: sperrin6@gatech.edu\u0022\u003ESelena Langner\u003C\/a\u003E\u003C\/p\u003E","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}