Science Pipelines Walkthrough: DRP Jim Bosch, DRP Scientist
Future Pipeline Flow Overview LSSTJTM - 2017-03-06 - Glendale, CA 2
ImChar Image Characterization ISR, PSF moding, initial background estimation, initial astrometric and photometric calibration, and Sources visits are independent, CCDs are mostly independent LSSTJTM - 2017-03-06 - Glendale, CA 3
JointCal Joint Calibration refine astrometric and photometric calibration refine PSF star positions all visits in an area of sky LSSTJTM - 2017-03-06 - Glendale, CA 4
Coaddition and Difference Imaging Make coadds (including templates) Find junk Refine background estimation Measure DIASources LSSTJTM - 2017-03-06 - Glendale, CA 5
Coadd Processing Define Objects by Deblend pixels using a associating DIASources set of coadds. and detection from a variety of coadds. Measure Objects on the coadds. LSSTJTM - 2017-03-06 - Glendale, CA 6
MultiFit Fit individual Objects or groups of Objects using all visit-level images that overlap their position. Includes galaxy model and moving point-source model for every Object. Includes Forced Photometry. LSSTJTM - 2017-03-06 - Glendale, CA 7
Postprocessing ● Estimate photometric redshifts. ● Find Solar System Objects, estimate orbits. ● Generate geometric depth/completeness masks. ● Run star/galaxy classification and variability characterization metrics. ● Apply calibrations to raw measurements. LSSTJTM - 2017-03-06 - Glendale, CA 8
Lower Levels ● All of these pipelines are (conceptually) CmdLineTask s (or SuperTask s, in the future). ● Below that level, LDM-151 also describes reusable "Algorithmic Components" that can generally be mapped to regular Task s: e.g. source detection, image subtraction, astrometric matching. ● And below that level, we have "Software Primitives": low-level data structures like Image s and Psf s.
Recommend
More recommend
