CALMON Version 3.0
The following are proposed changes to CALMON for Version 3.0 processing:
1. The following adjustments to the CALMON thresholds are proposed to facilitate the QA process and allow more nights to pass the first time. This decision was made by the QA analysts based on their extensive experience with the current processing. The parameters to alter are:
|
|
Threshold |
Version 3 Threshold |
|---|---|---|
|
calibration block (NOT for linear fit solutions ) Calibration strategy 1 pre-971011 |
|
|
calibration block (NOT for linear fit solutions ) Calibration strategy 2 post-971011 |
|
|
|
|
|
2. A table is proposed to assist the QA process.
CAL06: This file will contain specific information for the QA analyst regarding each scan within the night. Each row will provide all the necessary information to judge the quality of a single scan. It will list both cal and survey scans. Much of this data are already in cDATEH.cal file but are reformatted in CAL06 to assist the analyst in assigning scan quality.
3. To eliminate the problem of rejecting an entire set of calibraton scans due to a single aberrant standard, the aberrant standard must be identified and removed within a calibration scan Go here to see document on proposal This requires a change to STDMATCH routine that read in the input record and pulls off specific column information for CALMON. It could also require a significant change to CALMON.
4. To assure the quality of the standards accepted it has been proposed to check additional quality indicators before accepting or rejecting the standard aperition. Currently, [j,h,k]_prg_flg is checked for a "fill" indication . The new criteria would require that this flag be "000" in ALL bands to be accepted. This will eliminate any standard that happens to be affected by a spurious anomaly.
Other flags have been investigated. There was a proposal to use the [j,h,k]_psfchi to assist in determining quality of the magnitude measure. However in examining the delta (True-Obs) magnitudes verses psfchi values for a single standard, FS4 at ra=28.65715 and dec=0.71721, this does not appear to be a sufficient indicator of magnitude quality. See plots for J , H and K . The second moment ratio was also examined to see if there was a relationship with large delta magnitudes for the same standard, FS4. There appears to be no correlation, see plot for K . In conclude, only the prg_flg will be examined to determine quality of standards.
No additional quality indicators need be extracted from the input file, CALMON would only have to implement the new quality criteria.
NOTE: This is not a major change to CALMON. IF any band fails the test prg_flg == "000" test, the entire aperition is tossed.
5. To minimize noise in the zero point values assigned to calibration scans, compute and use the AVERAGE of all zero points from the 6 calibration scans. Currently, each individual scan uses its own derived zero point. This is not a major change to CALMON.
6. To assist QA in forcing breaks in the nightly calibration I propose replacing the CAL in the LGO file with a ECL designation. This will force CALMON to end the calibration block at this scan and treat every scan after as a new calibration block. This is a moderate logic change to CALMON and a change to DATEH.lgo file
7. To track the extinction correction use time-variable extinction coefficients. Currently, the extinction coefficients are constants set in the namelist. For a time-variable extinction value to be used, there must be either a table look up or a function to derive the coefficients. In either case, this will require a change in how these values are set in CALMON.
8. Improve secondary standard list by removing all eclipsing binaries. The set will be generated by Robert Hurt and Sergei.
9. TBD - Implement color-corrections, if necessary.
Comments and feedback: Sherry Wheelock
email: slw@ipac.caltech.edu Last update: 15 September 2000 slw@ipac.caltech.edu