MAPCOR and DB_MAPCOR Purge Algorithms

T. Evans - IPAC

Links in brief:

All bright stars found in the 2MASS fields cause artifacts on the detector arrays.  These artifacts can be detected by the 2MASS data processing as point and extended sources, or can affect the brightness of real point and extended sources.  MAPCOR and DB_MAPCOR are the subsystems that find and flag these artifacts or sources affected by the artifacts in the point source lists.  The main difference between these subsystems is that MAPCOR works inside the processing pipeline, on one band of one scan at a time, and DB_MAPCOR works on the database of point sources, which contains many observing nights and scans.  (For more information on the types of artifacts possible, please see this page.)

MAPCOR Processing

MAPCOR, as stated above, only works on one band of one scan at a time; thus it only has data from the point sources found in that band of that scan.  The possible "parent" sources, sources which cause artifacts, are the stars in the input point source list brighter than or equal to certain magnitudes:

Artifacts from fainter sources will be too faint to be extracted from the images.  MAPCOR, using algorithms dependent on the position and brightness of the "parents", looks for sources caused or affected by diffraction spikes, persistence, glints, stripes, and confusion found inside the scan and flags them appropriately (for artifact definitions see this page).

DB_MAPCOR Processing

MAPCOR cannot always find all artifacts in the scans with their "parent" sources.

  1. Any reasonably bright star located right on the edge of a scan may be extracted in one scan but not its overlapping scan, but it can still cause artifacts in both scans.
  2. Not all bright stars are actually detected and extracted within the point source processing pipeline -- especially sources saturated in the 51-ms Read1 exposure.  A prime example of this is beta Pegasus.  Of course, if a star isn't found in the scan source lists, MAPCOR can't find the artifacts it creates.
  3. Even when a very bright star is found in the source lists, it often has a poor-quality mag, which means MAPCOR with its magnitude-dependent algorithms may not have correctly found all of its artifacts.
  4. Since MAPCOR works only within a scan, it will not find any artifacts in one scan that are caused by a very bright star in a nearby scan (for example, beta Peg's diffraction spikes).
DB_MAPCOR works on the database of point sources to do final cleanup of diffraction spike and confusion artifacts only.  Therefore, it has access to "parent" sources which are not found in the same scan as their artifacts, as well as data on very bright Read1-saturated stars compiled from published lists of IR photometry.  It is run with these two different input "parent" source lists and parameter values:
  1. "Edge parent" processing, using the first type of "parents" above and the same algorithms and input parameter values as MAPCOR.  These "parents" are stars with J<=10.5, H<=10.0, or Ks<=9.5 which are located on the edge of a scan and therefore not extracted in that scan, but are found on an overlapping scan(s) in the database.  DB_MAPCOR uses the latter scan's data for the "parent" to clean up its artifacts in the former scan.
  2. "Very bright parent" processing, using the compiled "parent" list of data for stars with J, H, or Ks mag <= 5.0 and slightly different algorithms and input parameter values.  This processing conservatively removes artifacts caused by very bright stars as discussed in numbers 2-4 above.

e-mail: T. Evans
Last modified: 13 Jul 2000