We use an advanced RCE model that incorporates PRRTM. This is a planetary version of RRTM that has been adapted by AER to handle higher pressures of CO2 (up to 11 bar). Our motivation was to explain the breakout of the Earth’s climate from its Snowball state, once volcanic emission of CO2 had increased surface pressure to ≈1.4 bars. We find that while pressure broadening and collision-induced absorption are important, they are insufficient to explain the de-glaciation. Instead, the non-radiative factor of convective deepening, which dominates as surface pressure rises, provides the required additional warming.
As atmospheric mass increases further, the surface temperature is dominated by convective deepening, provided the surface radiation budget remains positive. With this assumption, our RCE model can reproduce the atmospheric temperature profile of Venus, despite its 90 bar surface pressure. However, we find an apparent paradox if we use the existing data on CO2 absorption. There appear to be no experimental data corresponding to surface conditions on Venus, with the closest we found being for 36 bar and 470 K. However, even a generous extrapolation from this down to the surface fails to provide the necessary absorption coefficients in the ‘window’ regions. This implies the existence of CIA by other species, the next most abundant of which is N2.