WISE Software Interface Specification; icl01; (WSDC D-I101) Written by F. Masci, version 2.0, 03/05/2010 Interface Name: Bad-pixel / Frame Processing Status Mask Type of Interface: FITS-Formatted 32-bit pixel status image recording prior information from ground characterization and dynamic masking from single frame processing. Read By: MDET, WPHOT _____________________________ K. Marsh, T. Jarrett (Source Detection and Photometry modules) AWAIC, AWOD _____________________________ F. Masci (A WISE Astronomical Image Co-adder) (A WISE Outlier Detector) ScanSync PSF ____________________________ J. Fowler (Software for generating PSFs for PSFMoments program to assess scan-mirror synchronization) Written/Updated By: ICAL ____________________________________ F. Masci (Instrumental Calibration Pipeline modules) SFPRex and MFPRex _______________________ T. Conrow (Single-Frame Pointing Refinement Estimation and Multi-Frame Pointing Refinement Estimation wrappers, primarily updates to FITS header with WCS information) DESCRIPTION: This is an image mask used to store pixel status information applicable to a frame acquired from a single exposure of the WISE FOV. It will store "static" bad-pixel information set apriori on the ground as well as details of "dynamic" processing. The mask will be initialized in the ICAL pipeline and progressively updated therein. It will be used downstream for omitting specific pixels which have been tagged as "bad" or unsuitable for scientific use, and will serve as an informational diagnostic for track-back analyses and quality assurance. The mask is a single plane of 1016 x 1016 pixels for bands 1, 2 and 3; and 508 x 508 pixels for band 4. This corresponds to the active region of a WISE frame, i.e., with the reference border discarded. The pixels are represented as 32-bit signed integers (BITPIX = 32). The status of certain conditions will be assigned to individual bits. Useable bits are numbered from zero (the LSB) to 30. The sign bit, bit 31, is not used. Each status bit describes a specific condition of the pixel; if the bit has the value 1, the condition is in effect; zero implies that the condition is not in effect. FITS HEADER: The output FITS header contains the following lines, where angle brackets indicate information whose literal content depends on actual execution circumstances. SIMPLE = T BITPIX = 32 NAXIS = 2 NAXIS1 = 1016 NAXIS2 = 1016 FRMFILE = '.fits' / Name of corresponding output frame file HISTORY = v. : ~ ~ ~ ~ END NOTES: * The HISTORY line in the above header shows the name, version, and local execution date/time of the module or subsystem that read/updated the mask. There shall be one history line for every module/subsystem that read and/or updated the mask. * The above header pertains to either band 1, 2 or 3 where NAXIS1 = NAXIS2 = 1016. A band 4 frame will have NAXIS1 = NAXIS2 = 508. * The header also contains a copy of all (non-mask/FITS specific) cards from the final processed frame. This may include information from steps after the instrumental calibration pipeline has run, e.g., a refined WCS from the SFPRex and MFPRex modules. This will assist in track-back analyses. FILE NAME: There will be one pixel status mask for each processed frame. Assuming that a frame processed through the instrumental calibration pipeline has the generic name: _w-int-1a.fits, the corresponding mask shall be named _w-msk-1a.fits. These products will be generated by the ICAL pipeline. STATUS BIT DEFINITIONS: Bit # Condition ----- ----------------------------------------------------------------- 0 from static mask: excessively noisy due to high dark current 1 from static mask: excessive read noise not due to high dark current 2 from static mask: dead or very low responsivity 3 from static mask: low responsivity 4 from static mask: high responsivity 5 from static mask: saturated anywhere in ramp 6 from static mask: high, uncertain, or unreliable non-linearity 7 from static mask: broken pixel or -'ve SUR (raw frame value = 32767) 8 reserved 9 broken pixel or intrinsically -'ve SUR (downlink value = 32767) 10 saturated in sample read 1 (downlink value = 32753) 11 saturated in sample read 2 (downlink value = 32754) 12 saturated in sample read 3 (downlink value = 32755) 13 saturated in sample read 4 (downlink value = 32756) 14 saturated in sample read 5 (downlink value = 32757) 15 saturated in sample read 6 (downlink value = 32758) 16 saturated in sample read 7 (downlink value = 32759) 17 saturated in sample read 8 (downlink value = 32760) 18 saturated in sample read 9 (downlink value = 32761) 19 reserved 20 reserved 21 new/transient bad pixel from dynamic masking (tempcal) 22 flat-fielding (responsivity correction) unreliable (ical) 23 sky-offset correction unreliable (tempcal) 24 reserved 25 contains probable latent flux (tempcal) 26 non-linearity correction unreliable (ical) 27 contains cosmic-ray or outlier that cannot be classified (awod) 28 contains positive or negative spike-outlier (ical) 29 reserved 30 reserved 31 not used: sign bit NOTES: * bits 0 - 7 (the first eight) are propagated from a static (8-bit) mask. This is initially created from characterization of lab data and will be updated periodically during mission operations. The remaining bits (8 - 30) are set dynamically on a per-frame basis in the WSDS. * bit 3: low responsivity refers to a pixel which is not quite dead. Truly dead or unresponsive pixels are covered by bit 2. * bit 21: new/transient bad pixels are detected and flagged using dynamic masking in the sky-offset calibration pipeline (tempcal module). * bits 22 and 23: these rely on calibration products dynamically generated from independent on-orbit data, presumably close in time to the science frame. Therefore, additional transients and outliers will affect the quality of these calibrations. * bit 27: this is a generic condition pertaining to any outlier detected using the temporal frame stacking method (awod module). It will contain transients inadvertently missed by other methods, as well as sources that have moved or varied appreciably between frames. * bit 28: single spike outliers are detected at the frame level (1a) in instrumental calibration. There may be redundancy with outliers detected using the temporal method (recorded in bit 27).