In “LSST/ From Science Drivers to Reference Design and Anticipated Data Products”, there is a Table 3 which lists the astrometric uncertainties for magnitudes 20 and fainter. The text generally describes the process that was used to develop these values. Is there another document that takes this table down to magnitude 17? As a comparison, GAIA shows a 20 fold improvement in astrometric uncertainty from mag 21 to mag 17.