Data Metrics Database

Plan for what fields there should be in data metrics database.

Image and header keyword associated fields

• Fits file name.
• CCDSUM -- Effectively tells you the binning of image. Can translate values such as '1 1' to 1x1 binning, etc.
• NCCDS -- Number of CCDs. May only be necessary for 90Prime?
• NAMPS -- Number of amplifiers.
• DATE-OBS -- UTC Calendar data when shutter opened. Should be combined with TIME-OBS.
• TIME-OBS -- UTC time at start of exposure. Should be combined with DATE-OBS.
• JD or JULIAN -- Julian date of observation.
• LST -- Local Sidereal time of observation.
• EXPTIME -- Exposure time in seconds.
• IMAGETYP -- 'object' or 'bias' or 'flat'. Is there something else we want? Could imagine: 'DOMEFLAT', 'SKYFLAT' 'FOCUS'
• OBJECT -- Nominally the object name, but also includes 'Focus', 'Zero' and similar things. The OBJECT and IMAGETYP keywords may play together in combination for aspects of the pipeline. For instance, if IMAGETYP='object' and OBJECT='focus' then we might not want to measure the image quality, sky background, etc.
• INSTRUME -- Instrument name
• DEWAR or DETNAME -- This will help us keep track of old vs. new 90Prime?
• OBSERVER -- Who took the data at the telescope?
• PIXSCAL1, PIXSCAL2 -- pixels scale of detector. Is this too redundant, can we glean this info from CCDSUM and INSTRUME, DETNAME?
• GAIN keywords -- for 90Prime, these all seem to be in the primary header. Does GAIN4 keyword correspond to IM4 extension?
• RDNOIS keywords -- same as above.
• PROPID -- proposal ID.
• READ_SPD -- Readout speed. Is it in kilopixels/s?
• AIRMASS -- Airmass of observations. Is this at beginning of observation, or end?
• AZIMUTH -- Telescope azimuth (degrees)
• ELEVAT -- Telescope elevation (degrees)
• DOME -- Dome position (degrees)
• FILTER -- Filter name
• RA, DEC -- strings for RA and DEC commanded position. Could/should be checked agains CRVAL1, CRVAL2
• FOCUSVAL -- Focus value of three LVDT's. Do they correspond to LVDT A, B, C? Is this from beginning of exposure?
• HA -- Hour Angle
• INDP -- Inside dewpoint
• INH -- Inside humidity
• INT -- Inside temperature
• MCELLDP -- Mirror Cell Dewpoint
• MCELLH -- Mirror Cell humidity
• MCELLT -- Mirror cell temperature
• CRVAL1, CRVAL2 -- this specifies the RA DEC in degrees. Seems to be the same in all image extensions, but is not in primary extension.

Many of the keywords above, especially those recording temperature, humidity, etc are not in all images. It is not clear why.

Bias or zero frames

In addition to keywords above, we will need:

• Mean value -- Mean value of bias frame after overscan subtraction. Should be measured for each amp on each chip.
• Standard deviation -- standard deviation of bias frame. Should be measured for each amp on each chip.
• Number of bad pixels -- Maybe choose this as number of pixels more than 10 standard deviations away from mean.

Flatfield frames

In addition to keywords above:

• Modal value of flat field in all amps. After overscan subtraction.
• Mean value of flat field in all amps.
• Standard deviation in all amps.

Focus frames

Other than the header keywords, I think we can ignore focus frames.