How far will global warming go, according to artificial intelligence

A disturbing new study, based on artificial intelligence forecasts, indicates that even a rapid cessation of anthropogenic carbon dioxide emissions into the atmosphere will not prevent global warming from exceeding 1.5 degrees Celsius. global temperature shortly before the Industrial Revolution began. Everything indicates, therefore, that in the coming years more and more new heat records will be broken.

This study was carried out by Noah S. Diffenbaugh, from Stanford University in California, and Elizabeth A. Barnes, from Colorado State University, both institutions in the United States. It is titled “Data-driven Predictions of Peak Warming Under Rapid Decarbonization” and has been published in the academic journal Geophysical Research Letters.

Another conclusion of the study is that there is a 50% chance that global warming will exceed 2 degrees Celsius, even if humanity meets current goals to rapidly reduce net greenhouse gas emissions to zero by the 2000s. 2050.

Several previous studies, including assessments by the Intergovernmental Panel on Climate Change, have concluded that decarbonization, at the target pace, would likely keep global warming below 2 degrees.

“In recent years we have observed an increase in the incidence of heat waves, torrential rains and other extreme phenomena around the world. “This study suggests that even in the best-case scenario, we are very likely to experience more severe conditions than we have experienced recently,” says Diffenbaugh.

An area of ​​Earth (Indian Ocean) seen from space. (Photo: NASA)

For the new study, Diffenbaugh and Barnes trained an artificial intelligence system to predict how far global temperatures could rise, depending on the pace of decarbonization.

To train the artificial intelligence, the researchers used temperature and greenhouse gas data from vast archives of climate model simulations. In order to predict future warming, however, they gave actual historical temperatures as input, along with several widely used scenarios for future greenhouse gas emissions.

“Artificial intelligence is emerging as an incredibly powerful tool to reduce uncertainty in predictions. “It learns from the many climate model simulations that already exist, but its predictions are then refined with real-world observations,” says Barnes.

The study adds to a growing body of research that indicates the world has almost certainly missed its chance to reach the most ambitious goal of the 2015 Paris Climate Agreement, in which nearly 200 nations pledged to do the same. needed to help keep long-term warming “well below” 2 degrees, while striving to avoid 1.5 degrees.

Another recent study predicts that warming in some areas of the Earth will likely exceed 3 degrees Celsius by the 2060s. This study is the work of Diffenbaugh, Barnes and Sonia Seneviratne, the latter of the Swiss Federal Institute of Technology in Zurich.

Affected areas will include, among others, South Asia, the Mediterranean region, Central Europe and parts of sub-Saharan Africa.

These predictions are based on a scenario in which greenhouse gas emissions continue to increase following the current pattern. (Fountain: NCYT by Amazings)