HST Grism Observations of a Gravitationally Lensed Redshift 10 Galaxy

Abstract

We present deep spectroscopic observations of a Lyman-break galaxy candidate (hereafter MACS1149-JD) at $z\sim 9.5$ with the $\text{Hubble}$ Space Telescope ($\text{HST}$) WFC3/IR grisms. The grism observations were taken at 4 distinct position angles, totaling 34 orbits with the G141 grism, although only 19 of the orbits are relatively uncontaminated along the trace of MACS1149-JD. We fit a 3-parameter ($z$, F160W mag, and Ly$\alpha$ equivalent width) Lyman-break galaxy template to the three least contaminated grism position angles using an MCMC approach. The grism data alone are best fit with a redshift of $z_{\mathrm{grism}}=9.53_{-0.60}^{+0.39}$ ($68\%$ confidence), in good agreement with our photometric estimate of $z_{\mathrm{phot}}=9.51_{-0.12}^{+0.06}$ ($68\%$ confidence). Our analysis rules out Lyman-alpha emission from MACS1149-JD above a $3\sigma$ equivalent width of 21 Å, consistent with a highly neutral IGM. We explore a scenario where the red $\text{Spitzer}$/IRAC $[3.6]-[4.5]$ color of the galaxy previously pointed out in the literature is due to strong rest-frame optical emission lines from a very young stellar population rather than a 4000 Å break. We find that while this can provide an explanation for the observed IRAC color, it requires a lower redshift ($z\lesssim 9.1$), which is less preferred by the $\text{HST}$ imaging data. The grism data are consistent with both scenarios, indicating that the red IRAC color can still be explained by a 4000 Å break, characteristic of a relatively evolved stellar population. In this interpretation, the photometry indicate that a $340_{-35}^{+29}$ Myr stellar population is already present in this galaxy only $\sim 500\mathrm{Myr}$ after the Big Bang.

Accepted to ApJ. This is the accepted version