LSST Science Data Quality Analysis Subsystem Design

Abstract

The Large Synoptic Survey Telescope (LSST) will have a Science Data Quality Analysis (SDQA) subsystem for vetting its unprecedented volume of astronomical image data. The SDQA subsystem inhabits three basic realms: image processing, graphical-user-interface (GUI) tools, and alarms/reporting. During pipeline image processing, SDQA data are computed for the images and astronomical sources extracted from the images, and utilized to grade the images and sources. Alarms are automatically sent, if necessary, to initiate swift response to problems found. Both SDQA data and machine-determined grades are stored in a database. At the end of a data-processing interval, e.g., nightly processing or data-release reprocessing, automatic SDQA reports are generated from SDQA data and grades queried from the database. The SDQA reports summarize the science data quality and provide feedback to telescope, camera, facility, observation-scheduling and data-processing personnel. During operations, GUI tools facilitate visualization of image and SDQA data in a variety of ways that allow a small SDQA-operations team of humans to quickly and easily perform manual SDQA on a substantial fraction of LSST data products, and possibly reassign SDQA grades as a result of the visual inspection.