Exposing Line Emission: A First Look At The Systematic Differences of Measuring Stellar Masses With JWST NIRCam Medium Versus Wide Band Photometry

Abstract

Photometrically derived stellar masses are known to suffer from systematic uncertainties, particularly due to nebular emission contributions to the spectral energy distribution. Using \emph{JWST} NIRCam imaging from the CAnadian NIRISS Unbiased Cluster Survey (CANUCS), we introduce a comparison study of photometrically-derived redshifts and stellar masses based on two photometric catalogs of the same field spanning $\sim$0.4-4.5$\mu$m: one consisting solely of wide band photometry, and another employing a combination of wide and medium band photometry. We find that \tilda70\% of galaxies have consistent photometric redshifts between both catalogs, with median stellar mass difference between the two catalogs of \lessthan\ 0.2 dex across all redshift bins. There are however a subset of galaxies (5\% at z\tilda2 up to 15\% at z\tilda6) where wide bands underestimate star formation rates and infer older stellar populations, leading to median stellar mass differences of \tilda0.7 dex. Examination of the SEDs for galaxies with inconsistent photometric redshifts shows this is caused by the inability of the wide bands to distinguish continuum emission from emission lines. Computing a stellar mass density with our sample we find that it is potentially underestimated using wide-band photometry by \tilda10-20\% at z \lessthan\ \ 4, and potentially overestimated by as much as a factor of 2-3 at z \greaterthan\ 5. These systematic differences caused by the poor spectral resolution of wide bands have implications for both ongoing and future planned observing programs which determine stellar mass and other physical properties of high redshift galaxies solely via wide band photometry.