2022ApJ...935..120J

We study the relationship between the morphology and star formation history (SFH) of 361 quiescent galaxies (QGs) at redshift 〈z_obs〉 ≈ 2, with stellar mass $\mathrm{log}{M}_{* }\geqslant 10.3$, selected with the UVJ technique. Taking advantage of panchromatic photometry covering the rest-frame UV-to-NIR spectral range ( ≈40 bands), we reconstruct the nonparametric SFH of the galaxies with the fully Bayesian SED fitting code PROSPECTOR. We find that the half-light radius R_e, observed at z_obs, depends on the formation redshift of the galaxies, z_form, and that this relationship depends on M_*. At $\mathrm{log}{M}_{* }\lt 11$, the relationship is consistent with ${R}_{e}\propto {\left(1+{z}_{\mathrm{form}}\right)}^{-1}$, in line with the expectation that the galaxies' central density depends on the cosmic density at the time of their formation, i.e., the “progenitor effect.” At $\mathrm{log}{M}_{* }\gt 11$, the relationship between R_e and z_form flattens, suggesting that mergers become increasingly important for the size growth of more massive galaxies after they quenched. We also find that the relationship between z_form and galaxy compactness similarly depends on M_*. While no clear trend is observed for QGs with $\mathrm{log}{M}_{* }\gt 11$, lower-mass QGs that formed earlier, i.e., with larger z_form, have larger central stellar-mass surface densities, both within the R_e (Σ_e) and central 1 kpc (Σ_1 kpc_), and also larger M_1 kpc_/M*_, the fractional mass within the central 1 kpc. These trends between z_form and compactness, however, essentially disappear if the progenitor effect is removed by normalizing the stellar density with the cosmic density at z_form. Our findings highlight the importance of reconstructing the SFH of galaxies before attempting to infer their intrinsic structural evolution.