Data model: photoMatchPlate

General description: These files contain the match of each plate to SDSS photometry, sorted in the same way as the spZbest files (that is, by fiber number). There are two matches: a pure positional match, and a flux-based match. The flux-based match is the default to use. The positional match finds the nearest primary object within 2 arcsec, or if there is none, identifies the primary field covering the given position. The actual data for the flux-based match can be found in the photoPlate file.

The flux-based match identifies objects that failed the non-positional match, and things for which the aperture flux of the child is low relative to the fiberflux (which is based on the parent). For such objects, it asks what parent (if any) covers the desired match position. If no parent exists, there is no valid match at all. If a childless parent exists, that is used as the match. If a parent with multiple children exists, images are made of all the children based on the atlases, and child that contributes the most flux within 1.5 arcsec radius of the match location is selected. For real objects this procedure works remarkably well.

The status of the match is recorded in the FLUXMATCH_STATUS bitmask, whose bits have the following meanings:

0 - ORIGINAL_FLUXMATCH - used the original match (which exists)
1 - FIBER_FLUXMATCH - flagged due to fiberflux/aperflux issue
2 - NONMATCH_FLUXMATCH - flagged due to non-match
3 - NOPARENT_FLUXMATCH - no overlapping parent in primary field
4 - PARENT_FLUXMATCH - overlapping parent has no children, so used it
5 - BRIGHTEST_FLUXMATCH - picked the brightest child

The original positional match information is stored in ORIG_ID, ORIG_RA, ORIG_DEC. (The flux-based match is ALWAYS taken from the same primary field).

In this directory, we also put QA plots for objects that have significant corrections: any spectrum with an OBJTYPE that isn't "SKY" but don't have FLUXMATCH_STATUS equal to 1. For example, this QA plot shows an example where we targeted a fleck from large galaxy on the sky, and the flux-based matching correctly matches it to the full object (which dominates the flux in the fiber).

Naming convention: photoMatchPlate-PLATE4-MJD.fits

Approximate size: 100 kbytes.

File type: FITS

Read by products: sas

Written by products: sas

Required header keywords:

• MATCH_V (string): version of photoop used for match
• S_REDUX (string): value of SPECTRO_REDUX during call to match
• P_CALIB (string): value of PHOTO_CALIB during call to match
• P_RESOLVE (string): value of PHOTO_RESOLVE during call to match

Required column names:

• RUN (int32): SDSS drift scan run of match (or overlapping primary field if no match)
• CAMCOL (int32): SDSS drift scan camera column of match (or overlapping primary field if no match)
• FIELD (int32): field within SDSS drift scan of match (or overlapping primary field if no match)
• ID (int32): object id within SDSS field of flux-based match (or 0 if no match)
• THING_ID (int32): unique integer associated with primary matching object (see the thingIndex file)
• RERUN (string): reprocessing number for reductions associated with the SDSS run of this match (or of primary field if no match)
• NOBSERVE (string): number of observations of the location of the flux-based match, or if no match of the desired match location (MATCH_RA, MATCH_DEC).
• NDETECT (string): number of detections associated with the flux-based match
• RA (float64): right ascension, J2000 deg, for flux-based match (or 0 if none)
• DEC (float64): declination, J2000 deg, for flux-based match (or 0 if none)
• MATCH_RA (float64): right ascension, J2000 deg, of desired match location
• MATCH_DEC (float64): declination, J2000 deg, of desired match location
• ORIG_ID (int32): object id within SDSS field of positional match (or 0 if no such match)
• ORIG_RA (float64): right ascension, J2000 deg, for flux-based match (or 0 if none)
• ORIG_DEC (float64): declination, J2000 deg, for flux-based match (or 0 if none)
• APERFLUX3_R (float32): aperture flux within 3 arcsec diameter aperture in the r-band (in fact, an interpolation between APERFLUX[2,2] and APERFLUX[3,2] from the photoObj files); this value corresponds to a measurement of child associated with the positional match (that is the ORIG_ID object); if no positional match, set to 0
• FIBERFLUX_R (float32): fiber flux within 3 arcsec diameter aperture in the r-band, again for the ORIG_ID object; however, photo measures this value from the parent, but at the location of the child center; thus, the flux-based code looks for cases where APERFLUX3_R is much less than FIBERFLUX_R to flag for further checking
• APERFLUX3_R_TOTAL (float32): flux measured from the parent object atlas image within a 3 arcsec aperture of (MATCH_RA, MATCH_DEC), for reference (not used in the flux-based matching algorithm); if this doesn't need to be checked, it is set to zero
• APERFLUX3_R_MATCH (float32): flux measured from the flux-based match object atlas image within a 3 arcsec aperture of (MATCH_RA, MATCH_DEC); the flux-based matching selects the child with the largest value of this flux; if this doesn't need to be checked, it is set to zero
• FLUXMATCH_PARENT (int32): parent that the flux-based matching used in its checks (0 if none --- that is, if the positional match worked and had a reasonable ratio of APERFLUX3_R/FIBERFLUX_R, or if the positional match failed but there was no parent overlapping (MATCH_RA, MATCH_DEC) in the primary field
• FLUXMATCH_STATUS (int32): bitmask reporting status of the match (see description above)

Example file: photoMatchPlate-2292-53737.fits