Fig. 1: Emulator-enabled decomposition of global mean surface temperature uncertainty.
From: Rewiring climate modeling with machine learning emulators
Figure 11 from 13 showcasing the capability of emulators to decompose global mean surface temperature projection uncertainties beyond what is plausible with large simulator ensembles. Panels a, b show the EBM-KF emulator uncertainty projections for temperature change, which adds the ability to quantify uncertainty due to volcanic activity in addition to the uncertainties quantified by the equivalent simulator projections from the CMIP621 projections in Panels c, d75,76. Source: Nicklas, J.M., Fox-Kemper, B., Lawrence, C.: Efficient Estimation of Climate State and Its Uncertainty Using Kalman Filtering with Application to Policy Thresholds and Volcanism. Journal of Climate 38(5), 1235-1270 (2025). â“’American Meteorological Society. Used with permission.
