Photometric Redshift Calibrations for LSST

redshift of each photo-z bin must be known to better than ~0.002(1+z) if cosmological measurement errors are not to be degraded. We are obtaining deep imaging in LSST filters using Subaru/SuprimeCam to forecast LSST precision. We investigate the use of deep multiwavelength photometry of a small portion of the LSST area as an intermediate calibrator between faint galaxies and a bright subsample with spectroscopy. We also explore using the cross-correlation between galaxies in a photometric redshift bin and a bright spectroscopic sample in order to reconstruct the photo-z redshift distribution in detail, even if the spectroscopic data is highly incomplete. With currently planned samples these methods can reach LSST calibration goals. Photometric Redshift Calibrations for LSST