Challenges in global climate models to represent cloud response to aerosols: insights from volcanic eruptions

Abstract

Abstract Aerosol-cloud interactions (ACI) remain a major source of climate uncertainty due to missing large-scale observational constraints. Such a constraint, with global cloud representativeness, has recently been developed based on the Holuhraun-2014 volcanic eruption from machine learning with satellite observations. Here, we confront this large-scale observational constraint against six diverse global climate models to advance our understanding of ACI simulation uncertainty. We show that marine liquid cloud optical depth responses to aerosols are reasonably well simulated, although through compensating errors. However, all models largely underestimate cloud cover responses to aerosols, with five of them outside the 90% confidence level. This persistent bias remains despite tuning five distinct cloud schemes and testing various key cloud processes. Such bias in cloud cover response is a major driver of simulation uncertainty in ACI cooling and needs to be addressed urgently to improve climate projections and estimations of climate sensitivity.