There are a couple things floating around the blogosphere that seem to be getting effect and cause mixed up. Pressure and density in the atmosphere and the greenhouse effect.
Pressure, density and composition of the atmospheres are important parts of the atmospheric effect of course. With no atmosphere there would be no effect and the more atmosphere the greater the effect. Greenhouse effect and atmospheric effect are not identical.
Basic heat transfer is conductive, convective, radiant and in Earth's case, latent with convective.
Conductive transfer in gases is based on the collisions of the gas molecules. More collisions, more transfer of heat. A denser atmosphere means more molecules are available for collision so more heat can be transferred. Pressure, mainly the pressure gradient once you have density, controls the convective or rate of rising warmer air versus cooler air. More density implies higher viscosity which means a slow rate of convection.
Density and composition impact radiant transfer. If the atmospheric composition includes radiant absorbing gases in the Planck envelop of the black body temperature of the surface, great density of those gases maximize absorption of those wavelengths.
Latent transfer leap frogs regions of the atmosphere as convection moves the latent heat accumulation from one temperature/pressure to another where the latent energy can be released or gained.
That is ridiculously simple because that part is ridiculously simple.
Radiant emission and absorption are limited by the absorption spectrum of the combined atmospheric gases. If CO2 absorbs 15 percent of the radiation from the surface it can only impact 15 percent of the of the outgoing radiation, all other things left constant :) This is were it is not longer so simple.
In a dense portion of the atmosphere, molecules excited by absorption have little chance of emitting in the spectrum. They collide with other molecules so there is conductive transfer of that energy which may broaden the spectrum of that packet of energy. In other words, spectral broadening is dependent on conductive transfer potential.
Venus has a dense atmosphere of CO2 at high pressure. At low altitude, the path length of radiant transfer is basically nil. The lowest millimeter of the atmosphere can absorb outgoing radiation, but conductive transfer rules in the highly viscous surface atmosphere. With conductive transfer ruling, the lowest portion of the atmosphere is effectively iso-conductive. Radiant transfer is small versus conductive transfer. The number of collisions serves to broaden the absorption spectrum of CO2 to the point that all available lines of absorption are filled. The CO2 at the surface is nearly a perfect mirror for all wave lengths. On the whole, Venus reflect 60 percent of the visible light and absorbs nearly 100 percent of the longwave radiation. In a way, this is runaway atmospheric effect, not necessarily runaway greenhouse effect.
The greenhouse effect is amplification of solar in coming radiation after absorption by the black or gray body. Atmospheric effect is the retention of any energy that increases surface temperature above the evident outgoing longwave implied temperature. If Venus were only at maximum greenhouse effect, its surface temperature would be approximately 440K, but its surface is nearly 770 K degrees. The additional 300 or so degrees is due to the internal core temperature being felt at the surface by nearly perfect atmospheric insulation.
This can be incorrectly thought of a greenhouse effect. It is incorrect because not enough solar radiation reaches the lower atmosphere to be absorb and then amplified by the greenhouse gases, CO2. In the higher atmosphere where the temperature approaches 440K, the greenhouse effect begins. Call it a matter of semantics, but Earth illustrates the reason for the terminology.
On Earth there is the addition of latent heat because of water. Considering that Earth has a surface black body temperature of 288K degrees it would radiate at 390Wm-2. Approximately 80 Wm-2 of that value leaves the surface as latent heat, so the effective surface radiant flux is 310Wm-2. Convection accounts for approximately 25 Wm-2 of surface cooling. Of the remaining 285Wm-2, approximately 40 Wm-2 does not interact with greenhouse gases above the troposphere, this is the atmospheric window. After removing that 40 Wm-2, 245 Wm-2 is emitted from the surface that may interact with greenhouse gases at varying levels in the atmosphere, mainly near the surface where the density of the atmospheric molecules is greatest. At the surface, conductive, convective and latent all interact to produce surface cooling at a relatively steady rate.
The physics does not end with the heat transfer. The concentration of the gases in the atmosphere varies with temperature of the atmosphere and oceans. Each gas has a vapor pressure so they are attempting to maintain an equilibrium with the oceans concentration at that temperature. Gas molecules with their energy flow in and out of the oceans constantly with changing temperatures and pressures.
When air warms it can hold more water vapor. While the greenhouse gases absorb radiation they transfer that energy with collision, conductive transfer, spreading the warming. The air can also contact the surface absorbing energy via collision. This reduces the vapor pressure for water vapor which gains energy by mainly conductive means to evaporate. While radiant energy from the sun provides the energy for evaporation, down welling or return radiant energy from the greenhouse gases cannot provide the majority of the energy for evaporation. Why?
The same greenhouse gases that absorb surface radiation in their spectra block the return of that radiation in sufficient quantity from more than a meter above the surface. Because collisions broaden the greenhouse spectra and collisions decrease with altitude as pressure decreases, the emission spectrum of the atmosphere changes with density and pressure.
If you compare Earth and Venus, the atmosphere of Earth is less viscous allowing more rapid cooling of the surface by convection while the limited rate of convection on Venus allows more rapid conductive transfer or local cooling. Conduction radiates in all directions just as radiant transfer does only on Venus, conductive transfer is more rapid.
So atmospheric pressure is important, all factors need to be considered or it is no better than the radiant only greenhouse theory.
Dr. Roy Spencer has a post on the same subject, Why Atmospheric Pressure Cannot Explain the Elevated Surface Temperature of the Earth. In his post he notes that weather, tends to cool the surface and that without weather, convection and surface evaporation, the surface may be 70C instead of 33 C warmer. That agrees pretty well with my estimate of what the no greenhouse Earth surface would be.
With no greenhouse gases and an albedo of 0.30 at least of half of that 0.3 would be due to the atmosphere, clouds and solar scattering by oxygen and nitrogen. If the distribution of atmospheric to surface reflection and absorption we the same for the no greenhouse gas Earth, the surface absorption would be approximately 175 Wm-2 which would have a corresponding temperature of about 235K degrees which is 52 degrees C lower than the present approximate temperature. Greenhouse gases account for approximately 52 degrees of warming, countered by 20 degrees of latent cooling and roughly 12 degrees of conductive/convective cooling. All three are in balance so while CO2 will increase the radiant warming some degree, it will also increase the conductive/convective cooling over a longer time scale. Water evaporation will initial increase with the additional CO2 warming attempting to maintain the balance. Reduced solar surface absorption reduces the impact of the greenhouse gases by removing a portion of its source of energy which in turn would reduce water vapor evaporation which is the caveat.
Water vapor cools the surface as it evaporates but is the primary greenhouse gas and also the major source of surface and atmospheric albedo. Changing weather patterns with reduced solar will determine how much cooler the average surface temperature may become with a prolonged solar minimum. If the cooling portion of water vapor feed back is dominate, the conductive cooling will not decrease as much as the radiant and there will be more cooling, possibly another little ice age if the solar minimum is long enough.
So there does not yet appear to be a unified theory of climate. Which is not surprising with such a complex system.