A Systematic Assessment of Total Stellar Mass Measurements with Spatially Resolved Photometry and Medium Bands at 1 < z < 9: The Impact of Mass-to-Light Variations and Outshining

JWST NIRCam spatially resolved photometry enables studies of galaxies’ stellar populations and stellar masses on sub-kpc scales. We use JWST (CANUCS, Technicolor, JUMPS) and HST data, covering wide- and medium-band UV-IR filters to study the stellar mass bias between spatially resolved and integrated photometry of $\sim$3000 galaxies at 1 $<z_{\mathrm{p}\mathrm{h}\mathrm{o}\mathrm{t}}<$ 9. The 2D parameter maps reveal spatial variations in stellar mass and populations, showing that integrated measurements systematically underestimate stellar masses. We find that the stellar mass bias remains small for galaxies with specific star formation rate $\mathrm{s}\mathrm{S}\mathrm{F}\mathrm{R}<{10}^{-9}$ yr${}^{-1}$, but increases for galaxies with $\mathrm{s}\mathrm{S}\mathrm{F}\mathrm{R}>{10}^{-8.5}$ yr${}^{-1}$. For these high-sSFR galaxies, the median offset reaches $\mathrm{\Delta }{\log }_{10}(M_{\ast })$ $\sim$ -0.36 dex with the largest bias occurring in low-mass ($M_{\ast }\le {10}^{9.5}{M}_{\odot }$) galaxies. These results are consistent with the outshining of older populations by recent star formation. High-sSFR galaxies also show spatial variations in their mass-to-light ratios ($M_{\ast }\mathrm{/}{L}_V$): the stellar mass bias decreases with $\sigma [{\log }_{10}(M_{\ast }\mathrm{/}{L}_V)]$ for galaxies with ${10}^{-8}\le \mathrm{s}\mathrm{S}\mathrm{F}\mathrm{R}[{\mathrm{y}\mathrm{r}}^{-1}]<{10}^{-9}$, with the strongest trend for ${10}^{-7}\le \mathrm{s}\mathrm{S}\mathrm{F}\mathrm{R}[{\mathrm{y}\mathrm{r}}^{-1}]<{10}^{-8}$, quantified by the Spearman correlation coefficient ${\rho }_{\mathrm{S}}=-0.37$. While medium bands are important for constraining stellar masses from integrated photometry, spatially resolved photometry reduces the impact of their absence, with dispersion offsets of $\mathrm{\Delta }{\log }_{10}(M_{\ast })$ = 0.33 dex for integrated photometry but 0.15 dex for resolved photometry when only wide-bands are included.