upper waypoint

Stanford AI Predicts a Scorching Future — Even If Greenhouse Gas Goals Are Met

Save ArticleSave Article
Failed to save article

Please try again

Hoover Tower on the Stanford University campus on April 9, 2019. Two new Stanford University studies using artificial intelligence suggest Earth’s hottest years ahead will likely shatter records, leading to more extreme climate conditions for California. (Rachael Myrow/KQED)

Even after some of Earth’s warmest years in history, two new climate studies from Stanford University suggest that the hottest years ahead will likely shatter existing records — even if greenhouse gas emissions are slashed.

The research comes as this year is likely on track to beat out 2023 as Earth’s hottest year on record. This summer was the warmest in the world for California and at least three Bay Area cities. And as average temperatures continue to climb, more extreme climate conditions will be likely.

For the first study, published in Geophysical Research Letters, the researchers trained artificial intelligence using historical temperature observations alongside a range of temperature and greenhouse gas data from global climate models.

Sponsored

They found that “the world’s virtually certain to cross” the Paris Climate Agreement’s threshold of limiting global temperature increases to 1.5 degrees Celsius to curb the worst impacts of climate change. They also found about a 50% chance of crossing the 2 degrees Celsius threshold. That’s even if humanity meets current goals of rapidly reducing greenhouse gas emissions to net zero by 2050.

“We’ll likely have an individual year that is at least as hot as what we faced in 2023,” said Noah Diffenbaugh, a Stanford climate scientist and coauthor of both studies. The record-breaking heat of 2023 influenced climate patterns, leading to marine heat waves, droughts, wildfires and significant flooding worldwide.

“We’re arguably not keeping up with the climate change that’s already occurred,” he added.

For the second study, published in Environmental Research Letters, the researchers trained artificial intelligence to predict how long until different regions reach different warming thresholds. They found that western North America has a high likelihood of meeting the 2-degree Celsius threshold by around 2030 and is “virtually certain” to reach that level by 2040.

“This is earlier than what has been projected based on just the climate model projections,” Diffenbaugh said. “We find a narrower uncertainty and, in many cases, an earlier crossing of these warming thresholds.”

Diffenbaugh said the findings are a big deal for California and the Bay Area because he expects flood-inducing storms, sea level rise, heat waves and wildfires to worsen as climate change heats the planet.

That means state and local governments, as well as communities, will need to prepare for a warmer world and more extreme weather scenarios in the coming years.

“These results suggest that even in the best-case scenario for reducing emissions, people in ecosystems are still very likely to face conditions much more severe than what we’ve been exposed to so far,” he said.

lower waypoint
next waypoint