Refining Phanerozoic Extreme Climate Simulations with Equilibrium Climate Sensitivity (ECS) in cGENIE

Equilibrium climate sensitivity (ECS), defined as the response of the global mean surface temperature response to a sustained doubling of atmospheric CO₂ at equilibrium, is a key metric for quantifying the Earth’s climate sensitivity to greenhouse gas emissions. Accurate ECS estimates are therefore fundamental for reliable simulations of the long-term carbon cycle. cGENIE, as an Earth system model of intermediate complexity that integrates ocean circulation, atmospheric energy balance, and global biogeochemical cycling, is widely used to investigate cross-sphere carbon cycle evolution and long-term climate feedback mechanisms. However, previous cGENIE studies have assumed a fixed climate sensitivity (with a default radiative forcing of 4 W m^-2 per CO₂ doubling), which often led to inaccurate surface temperature estimates compared with proxy reconstructions, limiting the model’s ability to capture state-dependent climate feedbacks. Here we use fully coupled models (e.g., HadCM3 and CESM) to derive the relationship between atmospheric CO₂ concentrations and ECS throughout the Phanerozoic. These simulations are considered to closely match proxy reconstructions of temperatures. We then incorporate state-dependent climate sensitivity into cGENIE to enhance its representation of climate feedbacks across varying CO₂ levels. Our results show that temperature simulations using the unmodified cGENIE model exhibit substantial discrepancies for periods of rapid cooling and warming, such as the Late Ordovician and the PETM. However, incorporating state-dependent climate sensitivity substantially reduces the discrepancy between simulated and proxy-reconstructed surface temperatures. These findings highlight the importance of accounting for state-dependent climate sensitivity in Earth system models, both for accurately reconstructing past climate extremes and for improving projections of future climate change.