Why air cools as it rises and why mountaintops are cold. A parcel of air lifted into thinner surroundings expands, doing work against the lower pressure — and with no heat let in, that work is paid for in temperature. For dry air the price is exact: Γd = g/cp ≈ 9.8 K per km. Down the center, data flows: constants go in, the engine lifts the parcel, the profile comes out. The blue team builds and defends it; the red team tries to break it.
source Wm. Thomson (Lord Kelvin), On the Convective Equilibrium of Temperature in the Atmosphere, Memoirs of the Manchester Literary & Philosophical Society, 1862 — no stable primary link (AMBER). Rendered, not quoted.
A rising parcel exchanges no heat with its surroundings (adiabatic). The first law then reads cp dT = (1/ρ) dP, and hydrostatic balance dP = −ρg dz gives it away:
Γd = −dT/dz = g/cp. Nothing is looked up — with g = 9.81 m/s² and cp ≈ 1005 J/(kg·K) the rate falls straight out. Equivalently, potential temperature θ = T(P₀/P)R/cp is conserved: T·P(1−γ)/γ holds constant along the lift.
Live values from the engine's constants:
| quantity | value |
|---|
The lapse rate is the vertical temperature structure that rides on the vertical pressure structure. Air thins with height because the barometric formula drops P exponentially — P = P₀e−z/H — and it is into that thinning that the parcel expands.
Add water and the story bends: condensing vapour dumps latent heat back into the parcel, so the moist rate is smaller (~5–6 K/km). Each sphere is the next one's premise: pressure drives the lift, latent heat softens the fall.
The blue team's live check: recompute Γd from g/cp, confirm the moist rate is smaller, and re-derive the stability sign. If red tampers with the formula, this badge is where it shows.
The lift needs only three numbers and a rule. Gravity pulls, heat capacity resists cooling, and "no heat in" closes the books:
| symbol | meaning | value |
|---|---|---|
| g | gravity | 9.81 m/s² |
| cp | heat capacity (dry air) | 1005 J/(kg·K) |
| γ | cp/cv (diatomic) | 1.40 |
| Q | heat exchanged | 0 — adiabatic |
"Adiabatic" = the parcel is lifted faster than it can trade heat with its neighbours. That single assumption is the whole game — and it is what you feed the panel below.
Dry parcel: Γd = g/cp. No condensation — the steepest cooling.
Move any control — the parcel curve, the environment curve and the stability verdict are computed from g/cp on the spot, never looked up.
What the machine produces, proven: the dry rate 9.76 K/km straight from g/cp; a moist rate near 5 K/km once latent heat is paid back; and a stability verdict from the sign of Γd − Γenv. The current air's numbers are above; these totals are the output.
The blue team's witness (left) confirms these numbers live; the red team (right) tries to make them wrong.
And the environmental lapse rate is weather, not a law — inversions, fronts and the boundary layer bend it hour to hour. The panel lets you set it precisely because nature will not. Long-range projections of how these profiles shift are modelled, not derived here — AMBER.
"Temperature always drops with height." Cut. Inversions reverse it, and above the tropopause the stratosphere warms with height. The lapse rate is a tendency of the troposphere, not a rule of the whole column.
"Air cools with altitude because it's farther from the warm ground." Cut. A lifted parcel cools because it expands and does work — adiabatic, not conductive. Distance from the ground is not the mechanism.
"The moist rate is fixed at 6 K/km." Kept, corrected. It is smaller than dry and variable (~4–7 K/km); the engine holds a representative 5 K/km and flags it as such.
The red team's move: flip the formula to Γd = cp/g — the ratio inverted — and try to pass it off as the lapse rate. The blue team's witness (window 7) is watching.
Invert the ratio and Γd jumps from ~9.76 K/km to ~102,000 K/km — physically absurd, orders of magnitude wrong. The witness recomputes, the g/cp check fails, and the badge turns red. Nothing is faked; the attack is real and it is caught.