According to AI-assisted analyses, the climate crisis could hit us even faster than previously assumed. Temperatures in Europe could rise by at least three degrees compared to pre-industrial levels by 2060 if greenhouse gas emissions continue to increase. This rapid rise is also anticipated in other regions of the world. The AI analysis, which draws on ten global climate models and is refined using observational data, suggests that AI is a powerful tool for reducing uncertainty in future projections. The analysis is based on the socio-economic pathway SSP3-7.0 from the assessment report of the Intergovernmental Panel on Climate Change (IPCC) and observed temperature anomalies from 2023.

"Reducing uncertainty"

Observational data also refines projections from recent years, as the team led by Elizabeth Barnes from Colorado State University in Fort Collins reports in the journal "Environmental Research Letters". "AI is emerging as an incredibly powerful tool for reducing uncertainty in future projections," said Barnes.

The global goal of limiting warming to 1.5 degrees above pre-industrial levels appears virtually out of reach in this projection, as does the two-degree threshold. Even in the most optimistic scenario, which envisages a rapid reduction of greenhouse gas emissions to net zero by the 2050s, warming could exceed these thresholds. The researchers stress that efforts to adapt to changing climate conditions are urgently needed, as significant impacts are to be expected even if greenhouse gas emissions are successfully reduced.

The fact that 2024 is likely to be the warmest year on record makes the AI-based projections worth considering as at least one plausible future scenario.

Read everything about the study in Environmental Research Letters: "Combining climate models and observations to predict the time remaining until regional warming thresholds are reached"