South SFV Today

Researchers at Stanford and Colorado State University have developed a rapid, low-cost approach for studying how individual extreme weather events have been affected by global warming. Their method, detailed in an August 21 study in Science Advances, uses machine learning to determine how much global warming has contributed to heat waves in the U.S. and elsewhere in recent years. The approach proved highly accurate and could change how scientists study and predict the impact of climate change on a range of extreme weather events. The results can also help guide climate adaptation strategies and are relevant for lawsuits seeking compensation for damages caused by climate change.

“We’ve seen the impacts that extreme weather events can have on human health, infrastructure, and ecosystems,” said study lead author Jared Trok, a PhD student in Earth system science at the Stanford Doerr School of Sustainability. “To design effective solutions, we need to better understand the extent to which global warming drives changes in these extreme events.”

Trok and his co-authors trained AI models to predict daily maximum temperatures based on regional weather conditions and the global mean temperature. For training the AI models, they used data from a large database of climate model simulations extending from 1850 to 2100. Once trained and verified, the researchers used actual weather conditions from specific real-world heat waves to predict how hot these heat waves would have been if they occurred under different levels of global warming. They then compared these predictions at different global warming levels to estimate how climate change influenced the frequency and severity of historical weather events.

The researchers first applied their AI method to analyze the 2023 Texas heat wave, which contributed to a record number of heat-related deaths in the state that year. The team found that global warming made this historic heat wave 1.18 to 1.42 degrees Celsius (2.12 to 2.56 F) hotter than it would have been without climate change. They also found that their new technique accurately predicted the magnitude of record-setting heat waves in other parts of the world, with results consistent with previously published studies.

Based on this analysis, researchers used AI to predict how severe future heat waves could become if similar weather patterns occurred under higher levels of global warming. They found that some of the worst heat waves experienced over the past 45 years in Europe, Russia, and India could happen multiple times per decade if global temperatures reach 2°C above pre-industrial levels—global warming is currently approaching 1.3°C above pre-industrial levels.

“Machine learning creates a powerful new bridge between actual meteorological conditions causing specific extreme weather events and climate models enabling us to run more generalized virtual experiments on Earth's system,” said study senior author Noah Diffenbaugh, Kara J Foundation Professor at Stanford Doerr School of Sustainability.

The new AI method addresses limitations of existing approaches by using actual historical weather data when predicting effects of global warming on extreme events without requiring expensive new climate model simulations since AI can be trained using existing simulations.

The team plans further application across various extreme weather events while refining AI networks for improved predictions including quantifying uncertainty ranges fully.

“We’ve shown that machine learning is a powerful tool for studying impacts of global warming on historical weather events,” said Trok.“We hope this study promotes future research into using AI for understanding human emissions' influence on extreme weather aiding better preparation.”

Diffenbaugh is also Kimmelman Family Senior Fellow at Stanford Woods Institute for Environment; co-authors include Elizabeth Barnes & Frances Davenport from Colorado State University funded by Stanford University & U.S Department Energy.