2018ApJ...856..116P

We improve the accuracy of photometric redshifts by including low-resolution spectral data from the G102 grism on the Hubble Space Telescope (HST), which assists in redshift determination by further constraining the shape of the broadband spectral energy distribution (SED) and identifying spectral features. The photometry used in the redshift fits includes near-infrared photometry from FIGS+CANDELS, as well as optical data from ground-based surveys and HST ACS, and mid-IR data from Spitzer. We calculated the redshifts through the comparison of measured photometry with template galaxy models, using the EAZY photometric redshift code. For objects with F105W < 26.5 AB mag with a redshift range of 0 < z < 6, we find a typical error of Δz = 0.03 * (1 + z) for the purely photometric redshifts; with the addition of FIGS spectra, these become Δz = 0.02 * (1 + z), an improvement of 50%. Addition of grism data also reduces the outlier rate from 8% to 7% across all fields. With the more accurate spectrophotometric redshifts (SPZs), we searched the FIGS fields for galaxy overdensities. We identified 24 overdensities across the four fields. The strongest overdensity, matching a spectroscopically identified cluster at z = 0.85, has 28 potential member galaxies, of which eight have previous spectroscopic confirmation, and features a corresponding X-ray signal. Another corresponding to a cluster at z = 1.84 has 22 members, 18 of which are spectroscopically confirmed. Additionally, we find four overdensities that are detected at an equal or higher significance in at least one metric to the two confirmed clusters.