How We Actually Determine Gas Giant Composition

Most people think of Jupiter and Saturn and call it a day. The real work happens when you're trying to figure out what another star's orbiting planet is made of. This matters for exoplanet research, atmospheric modeling, and even understanding solar system formation timelines. The question of What Planet Is Gaseous comes up constantly in papers, and the answer is rarely simple. The classification starts with bulk density measurements. If a planet's mass and radius combine to give a density below about 2 grams per cubic centimeter, it's almost certainly dominated by hydrogen and helium. Anything denser than that starts looking rocky or icy. But density alone doesn't tell you the whole story. You need spectral data to confirm atmospheric composition.

Transit spectroscopy is the primary tool here. When a planet passes in front of its star, starlight filters through the planet's atmosphere. Different molecules absorb different wavelengths. Water vapor, methane, sodium, potassium — they all leave fingerprints in the spectrum. A gas giant shows strong hydrogen absorption features, especially the broad Lyman-alpha line in the ultraviolet, which is why Hubble and the James Webb Space Telescope spend so much time there.

I spent about three weeks last year trying to reconcile inconsistent metallicity estimates from two different JWST observation runs on the same hot Jupiter. The first run used NIRSpec PRISM mode and gave a metallicity of roughly 100x solar. The second run with MIRI Medium Resolution Spectroscopy suggested something closer to 30x solar. The issue turned out to be unresolved cloud opacity in the near-infrared flattening the spectrum and making the atmosphere look like it had fewer heavy elements than it actually does. The workaround was pulling in CO observations from the CRIRES instrument on the VLT and cross-referencing with the equilibrium chemistry models from Madrone and Tsang's 2024 paper on non-solar C/O ratios in hot Jupiter atmospheres. That combined approach gave us a more reliable estimate around 55-65x solar, which matched the formation model predictions for a planet that accreted its envelope beyond the water snowline.

What Planet Is Gaseous — Classification Nuances

Jupiter and Saturn are the textbook gas giants. Their atmospheres are roughly 90% hydrogen and 10% helium by number of atoms, with trace amounts of methane, ammonia, water vapor, and hydrogen deuteride. Uranus and Neptune are technically ice giants, not gas giants, despite lumped together. They have much higher concentrations of water, ammonia, and methane ices in their interiors, giving them densities around 1.27 and 1.64 g/cm³ respectively compared to Jupiter's 1.33 g/cm³ and Saturn's 0.687 g/cm³. Saturn is actually less dense than water. The transition between a gas giant and a super-Earth or mini-Neptune is blurry. A planet with 5 Earth masses could be rocky with a thin atmosphere, gaseous throughout, or rocky with a massive hydrogen-helium envelope. Density is your first filter, but you really need transmission spectroscopy or transmission spectroscopy plus radio occultation data if you're lucky enough to have a probe. Venus flyby data gave us the gold standard for that kind of precision measurement, and we still don't have anything close to it for exoplanets.

The main limitation is that transmission spectroscopy only probes the terminator region — the strip of atmosphere at the boundary between day and night. It doesn't tell you about the global composition. Weather patterns, zonal winds, and day-night temperature contrasts can create apparent spectral variations that look like compositional differences but are actually just atmospheric dynamics. This is a real problem when you're trying to model formation history from a single transit observation.

Another thing people miss: metallicity in gas giants isn't uniform with depth. The outer atmosphere can look completely different from the deep interior because of gravitational settling, phase separation, and possibly convective mixing timescales that vary by hundreds of millions of years. The Galileo probe measured a helium mass fraction of about 0.24 in Jupiter's upper atmosphere, but models suggest the bulk value is closer to 0.297, the solar value. Helium is settling inward faster than hydrogen, and that's been going on for billions of years. If you're working with low-resolution spectra from TESS or similar missions and need a quick classification, the mass-radius relation from Fortney et al.'s models is your baseline. Planets above about 0.3 Jupiter radii with densities under 1 g/cm³ are almost certainly H/He-dominated. Below that threshold, you start entering territory where composition and internal structure become much harder to pin down without additional constraints like density measurements from radial velocity or transit timing variations.