During the Early Eocene, about 55 millions years ago, the climate of the Earth was characterized by radically different conditions than today: atmospheric CO2 exceeding 1000ppm, 10-15°C higher mean global surface temperature (GST) and strongly reduced pole-to-equator temperature gradient.
On top of this extraordinary mean state, the Earth was struck by a series of sudden global warming events, known as hyperthermals, which lasted a few millennia and saw further GST rise by as much as 5°C. They were entirely natural climatic events and were driven by an estimated 2000-5000 Gton carbon release into the atmosphere. Hyperthermals represent the fastest carbon release in the paleoclimatic records and our comprehension of their onset and decay is still partial, also because of the limited spatial and temporal frequency of proxy data.
A powerful tool to investigate hyperthermals - and Eocene climate in general - is represented by numerical Earth System Models (ESMs). Due to their capacity of filling gaps in proxy data and to the opportunity they provide for studying climate interactions under different mean states, ESMs are rapidly gaining ground for paleoclimatic applications.