Data model: calibPhotomGlobal

General description: This file contains the ubercalibration results for field in a given RUN/CAMCOL. Final calibration is given as the nanomaggies per count, the factor by which to multiply the raw counts of each object:

nmgypercount = aterm × 10^{0.4 × [kterm+kdot× (t-t₀)]× airmass} × flat

Note that application of the full calibration requires the flat-field vectors in the second HDU as well.

Naming convention: calibPhotomGlobal-RUN6-CAMCOL.fits

Approximate size: 200 kbytes.

File type: FITS

Read by products: photoop

Written by products: photoop

HDU1

Required column names:

• run (int32): drift scan run number
• rerun (string): reprocessing rerun number
• camcol (int32): drift scan camcol number
• field (int32): drift scan field number
• image_status[5] (int32): bitmask describing image status in each band
• calib_status[5] (int32): bitmask describing calibration status in each band
• tai[5] (float64): TAI-format time each band was observed in (ugriz)
• ref_tai[5] (float64): reference time in TAI-format for time variation (kdot)
• airmass[5] (float32): airmass at time of observation in each batch
• aterm[5] (float32): "a-term" in calibration equations in each band
• kterm[5] (float32): "k-term" in calibration equations in each band (at reference time)
• kdot[5] (float32): time variation of "k-term" in calibration equations in each band
• nstars_offset[5] (int32): number of stars used in calculating field offset
• field_offset[5] (float32): magnitude offset of ubercalibration relative to input (???)
• nmgypercount[5] (float32): nanomaggies per count in each band
• nmgypercount_ivar[5] (float32): inverse variance of nanomaggies per count in each band

Example file: calibPhotom-004948-1.fits

HDU2

The second HDU contains a 2048x5 image of the flat-fields in ugriz, in that order. Because of the drift-scan nature of the runs, the flat-field is a one-dimensional function of column (colc). One needs to use the photometric terms defined by aterm, kterm, and kdot, and then multiply the flat field in as well, as in the equation above.

Note that in the ubercalibration, the flat-fields are not constrained to average to unity, though they are always within a few percent. Thus, if you want an approximate calibration for a field, you still need to account for the value of the flat.