A dichotomy of the mass–metallicity relation of exoplanetary atmospheres demarcated by their birthplace

Abstract Atmospheric observations by JWST raise growing evidence that atmospheric metallicity exhibits an anti-correlation with masses of giant exoplanets. While such a trend was anticipated by planetesimal-based planet formation models, it remains unclear what kind of atmospheric metallicity trends emerge from pebble-based planet formation. Moreover, while recent studies of solar system Jupiter suggest that the uppermost observable atmosphere may not represent the bulk envelope composition, it remains uncertain how the envelope inhomogeneity influences the atmospheric metallicity trend. In this study, we develop disk evolution and planet formation models to investigate the possible atmospheric metallicity trends of giant exoplanets formed via pebble accretion and how they depend on the metallicity inhomogeneity within the envelope. We find that pebble-based planet formation produces two distinct mass–metallicity relations depending on planetary birthplace. Planets formed beyond the H$_2$O snowline exhibit a mass–metallicity anti-correlation similar to that predicted by planetesimal-based models if their atmospheres are fully convective. This anti-correlation disappears if the convective mixing is inefficient. In contrast, planets formed inside the H$_2$O snowline show a shallower mass–metallicity anti-correlation, regardless of the efficiency of atmospheric mixing. We test different initial disk properties and fragmentation threshold velocities of dust particles, demonstrating that the dichotomy of the mass–metallicity relation is robust against these uncertainties. Many gas giants observed by JWST observations lie around the mass–metallicity relation predicted for formation at close-in orbits, although some planets with sub-stellar atmospheric metallicity appear to require unmixed envelopes and formation beyond the H$_2$O snowline. We also examine the relationship between bulk and atmospheric metallicity and find a clear correlation that closely follows atmospheric metallicity that is comparable to bulk metallicity. Our findings will help future surveys of exoplanetary atmospheres by JWST and Ariel to shed light on where close-in giants come from on the basis of the mass–metallicity relation.