A new study using AI technology reveals that global warming is likely to surpass the critical threshold of 2 degrees Celsius, even with accelerated efforts to reduce greenhouse gas emissions. The findings highlight the urgency for both decarbonization and climate adaptation to mitigate severe impacts.
A recent study using artificial intelligence (AI) has revealed that the global target of limiting warming to 1.5 degrees Celsius above pre-industrial levels is nearly unattainable. Results from the study, published in Geophysical Research Letters, indicate that the hottest years ahead are poised to break existing heat records. The researchers report a 50% chance that global warming will breach 2 degrees Celsius, even if global aims to achieve net-zero emissions by the 2050s are met.
“This study suggests that, even in the best case scenario, we are very likely to experience conditions that are more severe than what we’ve been dealing with recently,” co-author Noah Diffenbaugh, a professor of Earth system science at Stanford Doerr School of Sustainability and a senior fellow in the Stanford Woods Institute for the Environment, said in a news release.
His collaborator, Elizabeth Barnes, a professor in the Department of Atmospheric Science at Colorado State University, emphasized the capability of AI in refining climate projections.
“AI is emerging as an incredibly powerful tool for reducing uncertainty in future projections. It learns from the many climate model simulations that already exist, but its predictions are then further refined by real-world observations,” Barnes said in the news release.
Earth’s Hottest Years Yet
Experts have warned for years about exceeding the 1.5 degrees Celsius threshold. The new study predicts that 2024 will be hotter than 2023, potentially making it the hottest year on record. Temperatures could exceed 1.5 degrees Celsius above pre-industrial levels, nearly 2.7 degrees Fahrenheit.
By harnessing AI, Diffenbaugh and Barnes analyzed temperature and greenhouse gas data from extensive climate model simulations. They calibrated their AI model with actual historical temperatures and various greenhouse gas emission scenarios.
The findings highlight that there is a nine-in-ten chance this century’s hottest year will be at least half a degree Celsius hotter, even with swift decarbonization efforts.
Potential Global Impact
The urgency of these findings comes as world leaders have already pledged, under the 2015 Paris Climate Agreement, to keep long-term warming well below 2 degrees while striving to avoid 1.5 degrees. The latest findings, however, suggest that achieving these ambitious goals might be too optimistic.
A companion paper, also published in Environmental Research Letter and co-authored by Sonia Seneviratne, a professor of environmental science at ETH-Zurich, indicates that regions such as South Asia, the Mediterranean, Central Europe and parts of sub-Saharan Africa might see temperatures exceeding 3 degrees Celsius by 2060 if emissions increase unabated.
The Necessity of Adaptation
As global temperatures rise, extreme weather events have become more frequent and severe. Diffenbaugh reflects on the importance of preparing for these severe impacts.
“Our results suggest that even if all the effort and investment in decarbonization is as successful as possible, there is a real risk that, without commensurate investments in adaptation, people and ecosystems will be exposed to climate conditions that are much more extreme than what they are currently prepared for,” he added.
The findings underscore the need to invest in both reducing emissions and making human and natural systems more resilient to climate impacts such as extreme heat, intensified droughts and heavy precipitation.
These insights add a critical dimension to our understanding of the future climate scenario, illustrating the indispensable role of AI in refining predictions and guiding strategies to combat climate change.