Figure 1 (IMAGE)

Pohang University of Science & Technology (POSTECH)

Figure 1

Caption

a. Time series of the observed annual global mean air temperature (AGMT) anomalies during 1980- 2020 (black line) and the estimated AGMT obtained by using daily precipitation fields from the satellite-based observations and ERA5 reanalysis (colored lines) as inputs of deep learning model. Deep-learning produced global warming intensities are well consistent with the observations and show exceedance of a 95% confidence range of nature climate variability (dashed black horizontal lines) since 2015. b. The difference in the daily precipitation standard deviations between 2016–2020 and 2001–2005. Increased precipitation variability indicates more extreme behavior of daily precipitation. Black boxes represent hot spot regions for global warming influences identified by an explainable deep learning technique.

Credit

POSTECH

Usage Restrictions

None

License

Licensed content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.