Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Tip
iconfalse
titleContents

Table of Contents
maxLevel2
indent20px

PS1-PSC: Probabilistic Classifications of Unresolved Point Sources in PanSTARRS1

A relatively new source of star-galaxy separation is the classification table derived by Tachibana & Miller (2018, PASP 130, 995).  The authors combined multiple sources of information from the PS1 databases (including the approaches described below) using a sophisticated machine-learning algorithm to estimate the point-source probabilities for 1.5 billion objects.  The resulting table is accessible in MAST Casjobs in the HLSP_PS1_PSC database.  It can be joined with the standard PS1 tables to get accurate classification probabilities for most objects in PS1.  Here is an example Cajobs query that selects objects within 1 arcmin of RA=10 deg, Dec=20 deg:

Code Block
languagesql
themeConfluence
-- Run this in the PanSTARRS_DR2 Casjobs context

select psc.objid, psc.ps_score, o.nDetections, o.nStackDetections
from fGetNearbyObjEq(10.0, 20.0, 1.0) f
join ObjectThin o on o.objid=f.objid
join HLSP_PS1_PSC.pointsource_scores psc on psc.objid=f.objid

See the PS1-PSC High Level Science Product page and the published paper for more details.  We have seen good results using this data to separate blended and unblended stars in crowded regions in the Galactic plane.

PSF-Kron

The simplest way is to use the difference between PSFMag and KronMag. This has the advantage that it is available for all objects in the survey and in all the main tables.

...