An effort to build a climate model using a modern code base that can leverage modern data sets has snapped up funding from the Massachusetts Institute of Technology. The Climate Modeling Alliance (CliMA) will receive an unspecified “multiyear” funding from MIT that promises to deliver more accurate results to public and private entities on a higher scale as climate change fuels increased weather catastrophes.
“This project leverages advances in artificial intelligence, machine learning, and data sciences to improve the accuracy of climate models and make them more useful to a variety of stakeholders — from communities to industry,” said a statement from MIT issued last week.
According to the proposal, CliMA aims to build a “digital twin” model of Earth’s climate system — including atmosphere, ocean, and land physics — that will be continuously updated with real-time, global observational data that is filtered through machine learning.
The project will then develop surrogate models for regional hazards like Florida hurricane that will replace “simplified reduced-form models [that] exist and are currently used by decision makers.”
The final step in model’s development is rolling out to public and private market users in “test cases” that will inform the climate portion of their risk model software.
“Our software tools tailored to stakeholder needs will provide the basis for data-driven decisions about climate-related infrastructure investments and disaster preparedness—an outcome that is estimated to be worth trillions of dollars in socioeconomic value.”
The CliMA model was one of five projects that received awards from MIT’s first Climate Grand Challenges competition.
Learn more, Computing our climate future (MIT News)
Climate model code is so outdated, MIT starts from scratch (The Register)
