Reassessing Last Glacial Maximum Cooling With Proxy‐Constrained Emergent Relationships

A key uncertainty in climate science lies in reconstructing global temperatures from sparse proxy data. Here, we apply an emergent constraint framework that combines statistical relationships in paleoclimate model simulations with proxy data to estimate difficult-to-reconstruct climate variables. Using a new multiproxy dataset of tropical land temperature reconstructions for the Last Glacial Maximum (LGM), we estimate a tropical terrestrial temperature change of -4.8°C (-5.6 to -4.0°C, 95% Confidence Interval, CI). Leveraging model-derived statistical relationships, we use the new terrestrial temperature reconstruction to constrain estimates of tropical sea surface temperature (-2.9°C; -3.6 to -2.3°C, 95% CI) and global mean surface temperature (-5.9°C; -7.3 to -4.7°C, 95% CI). These results suggest a colder LGM climate than estimates based heavily on the MARGO SST dataset. This method can be adapted for different time periods and climate variables, providing a powerful and flexible framework for integrating proxy and model information.