Exoplanet atmospheres contain less methane than they should. Chemical equilibrium models predict specific CH₄ abundances based on temperature, pressure, and composition. Observations consistently find less. The standard fix was to invoke photochemistry — ultraviolet light from the host star breaking methane apart in the upper atmosphere. It works for hot Jupiters close to their stars but fails for cooler, more distant planets where UV flux is low.

The actual fix, shown by Fortney et al. (arXiv 2604.01672), is that the interiors are hotter than models assumed. Higher internal temperatures drive vertical mixing that dredges up CO and CO₂ from deep layers, shifting the carbon budget away from methane. The atmosphere is not out of equilibrium because something external is destroying CH₄. It is out of equilibrium because something internal — excess heat from formation, tidal dissipation, or radioactive decay — is pushing the chemistry from below. The atmospheric composition is a thermometer for the planet’s interior.

Fusion tokamaks present a parallel diagnostic puzzle. Particle distributions measured at the plasma edge have long shown asymmetries that standard transport models could not explain. The discrepancy — where particles appeared at the boundary versus where models predicted — persisted for years. The resolution came from recognizing that the edge measurement was encoding information about interior transport mechanisms: the way particles move through the core determines where they arrive at the surface. The surface anomaly was a projection of the interior dynamics onto the boundary.

In both systems, the accessible measurement — atmospheric spectrum, edge particle distribution — initially looked anomalous because it was compared to models that assumed the interior was in its simplest possible state. The anomaly vanished when the interior was allowed to be interesting.

The surface is a window into the interior, but only if you know which discrepancy to trust.