Presets
Playback
Parameters
Camera
Layers
Quality
Model check
The science
What the model solves. The planet absorbs starlight and must radiate the same power: global-mean absorbed flux S(1 − A)/4 sets the equilibrium temperature Teq = 278.5 K × (1 − A)1/4 L1/4 / √a (A Bond albedo, L stellar luminosity in solar units, a orbit in AU). An atmosphere that is transparent to starlight but opaque to infrared warms the surface above Teq. With a grey infrared optical depth τ(p) = τs(p/ps)n, radiative equilibrium in the Eddington two-stream approximation gives T4(τ) = Teq4(1/2 + 3τ/4): the top of the atmosphere (skin) sits at Teq/21/4, and a purely radiative surface at Teq(1 + 3τs/4)1/4.
Convection. Deep in a thick atmosphere the radiative profile falls off with pressure faster than a rising air parcel cools, so the column overturns. The code marches down from the top; where dlnT/dlnp would exceed κ = R/cp it stops following the radiative profile and continues along the dry adiabat T = Tt(p/pt)κ to the surface. That gives the tropopause (pt, Tt) and the surface temperature. The dry lapse rate is g/cp = gκμ/R.
Altitude, scale height, transit signal. Hydrostatic balance gives dz = (R T / (μ g)) dln p, integrated numerically from the surface; the scale height is H = kT/(μ mu g). In transmission a clear atmosphere adds about 2 N H Rp / Rs2 to the transit depth, with N of roughly 5 scale heights; the readout uses the temperature near the 1 millibar level and typical stellar radii (1, 0.73, 0.29, 0.12 solar). This is why a hot, light, hydrogen-rich envelope around a small star is the easiest to observe.
Honest simplifications. The absorber is grey (real greenhouse gases absorb in narrow bands, and a doubling of CO2 gives roughly a logarithmic, not linear, forcing); the optical depth is a free number, not computed from gas amounts. For Earth, Venus, Mars and Titan it is tuned so that the model returns the measured surface temperature (288, 737, 214 and 94 K), so those four are calibrations, not predictions; the independent tests are the skin temperature, the lapse rates, the hydrostatic altitudes, the scale heights and Teq (checks below). Water phase change, clouds, ice-albedo feedback, latent heat (the real Earth lapse rate is the smaller moist value, about 6.5 K/km), a stratospheric temperature inversion from ozone and a day-night heat-transport limit are all left out; atmosphere height on the 3D globe is exaggerated for visibility and the shell colours are schematic.
Checks. (1) skin temperature, bare-rock limit and monotonic greenhouse warming; (2) dry adiabatic lapse rates of four worlds against measured values; (3) numerical hydrostatic altitudes against closed forms; (4) scale heights and equilibrium temperatures against published values.