Atmospheric Circulation Impacts of Alternative Warming Scenarios for the Eocene

Recent work in modelling the warm climates of the Early Eocene shows that it is possible
to obtain a reasonable global match between model surface temperature and proxy
reconstructions, but only by using extremely high atmospheric CO2 concentrations or more
modest CO2 levels complemented by a reduction in global cloud albedo. Understanding the
mix of radiative forcing that gave rise to Eocene warmth has important implications for
constraining Earth's climate sensitivity, but progress in this direction is hampered by
the lack of direct proxy constraints on cloud properties. Here, we explore the potential
for distinguishing among different radiative forcing scenarios via their impact on
regional climate changes. We do this by comparing climate model simulations of two
end-member scenarios: one in which the climate is warmed entirely by CO2, and another in
which it is warmed entirely by reduced cloud albedo (which we refer to as the low
CO2-thin clouds or LCTC scenario) . The two simulations have almost identical
global-mean surface temperature and equator-to-pole temperature difference, but the LCTC
scenario has ~ 11 % greater global-mean precipitation. The LCTC simulation also has
cooler midlatitude continents and warmer oceans than the high-CO2 scenario, a
tropical climate which is significantly more El Niño-like, and a substantial equatorward
shift of the mid-latitude eddy-driven jets. We show that increased precipitation in the
LCTC case is due to its lower CO2 concentration. Furthermore, the increased
precipitation in the LCTC case is concentrated in the equatorial Pacific, and drives an
extratropical response which explains the equatorward jet shift. Overall, the
differences in regional climates between the two simulations are strong enough to be
potentially detectable in the geological record.