WISE Software Interface Specification; sfp01; (WSDC D-I102) Written by F. Masci, version 1.5, 7/30/2008 Interface Name: SFPRex and MFPRex / AWAIC FITS WCS-keywords Interface Type of Interface: FITS frame header input specification for WISE co-adder: level-1B products with refined WCS. Read By: AWOD & AWAIC _____________________________ F. Masci (A WISE Outlier Detector) (A WISE Astronomical Image Co-adder) Written/Updated By: SFPRex, MFPRex and PCAL Wrappers _________ T. Conrow (Single-Frame Pointing Refinement Estimation, Multi-Frame Pointing Refinement Estimation, Photometric Calibration) DESCRIPTION: The main outputs from the SFPRex and MFPRex modules are updates to metadata tables with refined astrometric and WCS pointing information. This information will be written to level-1B FITS frame headers by the SFPRex and MFPRex wrappers. The level-1B FITS headers also contain metadata propagated from the instrumentally calibrated level-1A FITS headers. Information in the level-1B FITS headers will be used downstream for single frame (single epoch) point source detection and estimation (SDex), band-merging, background matching, and co-addition (AWAIC). FITS HEADER: An input image frame for AWAIC will be a file in standard FITS format and the header will consist of WCS keywords consistent with the FITS-WCS convention. These keywords are understood by standard WCSLIB routines, numerous analysis tools, and image viewers. An example (minimal) FITS-WCS header specification is given below. Only those WCS keywords required by AWAIC are shown. Additional ancillary and processing information, as well as comment and history lines may be included. The keyword values are not necessarily realistic, however, it is recommended that the quoted precision where shown, be adhered to. EXAMPLE: SIMPLE = T / file does conform to FITS standard BITPIX = -32 / number of bits per data pixel NAXIS = 2 / number of data axes NAXIS1 = 1016 / length of data axis 1 NAXIS2 = 1016 / length of data axis 2 CRVAL1 = 114.4170312243 / [deg] RA at CRPIX1,CRPIX2 for EQUINOX CRVAL2 = 65.5901640421 / [deg] Dec at CRPIX1,CRPIX2 for EQUINOX PA = -73.2765940245 / [deg] Pos. angle of axis 2 (EofN) for EQUINOX EQUINOX = 2000.0 / [year] Equinox of WCS CD1_1 = -9.7331869193567E-05 / [deg/pixel] CD matrix element 1_1 CD1_2 = -3.2468690061352E-04 / [deg/pixel] CD matrix element 1_2 CD2_1 = -3.2480736664632E-04 / [deg/pixel] CD matrix element 2_1 CD2_2 = 9.7537985645554E-05 / [deg/pixel] CD matrix element 2_2 CTYPE1 = 'RA---SIN-SIP' / Projection type [with distortion] for axis 1 CTYPE2 = 'DEC--SIN-SIP' / Projection type [with distortion] for axis 2 CRPIX1 = 508.5 / Reference pixel at CRVAL1,CRVAL2 along axis 1 CRPIX2 = 508.5 / Reference pixel at CRVAL1,CRVAL2 along axis 2 CRDER1 = 0.0003025099 / [deg] Uncertainty in CRVAL1 CRDER2 = 0.0003024249 / [deg] Uncertainty in CRVAL2 UNCRTPA = 0.0017481252 / [deg] Uncertainty in position angle UNCRTS1 = 0.0000001234 / [deg] Uncertainty in pixel scale for axis 1 UNCRTS2 = 0.0000001234 / [deg] Uncertainty in pixel scale for axis 2 CSDRADEC= 0.0000078193 / [deg] Costandard deviation in RA and Dec MAGZP = 20.8403 / [mag] photometric zero point MAGZPUNC= 0.0500 / [mag] photometric zero point uncertainty A_ORDER = 3 / polynomial order, axis 1, detector to sky A_0_2 = -6.341484E-05 / axis 1 distortion coefficient of v^2 A_0_3 = -1.458834E-08 / axis 1 distortion coefficient of v^3 A_1_1 = 2.204371E-04 / axis 1 distortion coefficient of u.v A_1_2 = 1.025434E-07 / axis 1 distortion coefficient of u.v^2 A_2_0 = -1.001876E-04 / axis 1 distortion coefficient of u^2 A_2_1 = -2.062207E-09 / axis 1 distortion coefficient of u^2.v A_3_0 = 8.086817E-08 / axis 1 distortion coefficient of u^3 A_DMAX = 1.625876 / [pixel] maximum correction B_ORDER = 3 / polynomial order, axis 2, detector to sky B_0_2 = 2.511898E-04 / axis 2 distortion coefficient of v^2 B_0_3 = 6.642542E-08 / axis 2 distortion coefficient of v^3 B_1_1 = -2.703651E-05 / axis 2 distortion coefficient of u.v B_1_2 = 5.469226E-08 / axis 2 distortion coefficient of u.v^2 B_2_0 = 3.103393E-05 / axis 2 distortion coefficient of u^2 B_2_1 = 4.404894E-08 / axis 2 distortion coefficient of u^2.v B_3_0 = 3.954436E-08 / axis 2 distortion coefficient of u^3 B_DMAX = 1.271021 / [pixel] maximum correction AP_ORDER= 3 / polynomial order, axis 1, sky to detector AP_0_1 = 7.738011E-07 / axis 1 distortion coefficient of V AP_0_2 = 6.345607E-05 / axis 1 distortion coefficient of V^2 AP_0_3 = -3.144725E-08 / axis 1 distortion coefficient of V^3 AP_1_0 = -2.824433E-06 / axis 1 distortion coefficient of U AP_1_1 = -2.203247E-04 / axis 1 distortion coefficient of U.V AP_1_2 = 1.779848E-08 / axis 1 distortion coefficient of U.V^2 AP_2_0 = 1.000876E-04 / axis 1 distortion coefficient of U^2 AP_2_1 = -7.421441E-08 / axis 1 distortion coefficient of U^2.V AP_3_0 = -5.308015E-08 / axis 1 distortion coefficient of U^3 BP_ORDER= 3 / polynomial order, axis 2, sky to detector BP_0_1 = 9.438048E-07 / axis 2 distortion coefficient of V BP_0_2 = -2.510898E-04 / axis 2 distortion coefficient of V^2 BP_0_3 = 6.101012E-08 / axis 2 distortion coefficient of V^3 BP_1_0 = 2.048981E-06 / axis 2 distortion coefficient of U BP_1_1 = 2.735527E-05 / axis 2 distortion coefficient of U.V BP_1_2 = -8.480797E-08 / axis 2 distortion coefficient of U.V^2 BP_2_0 = -3.104517E-05 / axis 2 distortion coefficient of U^2 BP_2_1 = -1.130328E-08 / axis 2 distortion coefficient of U^2.V BP_3_0 = -4.741393E-08 / axis 2 distortion coefficient of U^3 END NOTES: 1. It is recommended that the projection type (CTYPE1, CTYPE2) be represented as an orthographic projection (SIN). Given the relatively small WISE FOV, the projection type makes little difference, however, output Atlas images will adopt the SIN projection since it admits "more" sky to be mapped onto a flat plane compared to the TAN projection. Furthermore, the suffix "-SIP" at the end of the CTYPE keyword values stands for "Simple Imaging Polynomial". This indicates that the coefficients of a distortion polynomial are present in the header and should be used when transforming from pixel coordinates to sky and vice versa. 2. The above example shows the coefficients of a third-order distortion polynomial. There may be fewer or more coefficients depending on the nature of the distortion and the degree to which it can be characterized. The A_i_j and B_i_j coefficients are for transforming pixel coordinates to the sky, and the AP_i_j, BP_i_j are for the reverse transformation, i.e., to enable fast inversion by WCSLIB software. u and v are relative pixel coordinates: u = x - CRPIX1 and v = y - CRPIX2. In general, a distortion polynomial along any axis can be represented: i<=ORDER,j<=ORDER P(u,v) = SUM [C_i_j.u^i.v^j], with i + j <= ORDER, i=0,j=0 where ORDER can be any of the values of the following keywords: A_ORDER, B_ORDER, AP_ORDER or BP_ORDER. The C_i_j can be any of the coefficients: A_, B_, AP_, or BP_. Note that the first order pixel-to-sky coefficients: A_0_1, A_1_0, B_0_1, B_1_0 are implicitly encoded in the CD_i_j keywords (see 4 below). For the reverse transformation involving AP_ and BP_, these orders are allowed. For more details, see: http://wise2.ipac.caltech.edu/staff/fmasci/codeVdist.html http://ssc.spitzer.caltech.edu/postbcd/doc/shupeADASS.pdf 3. The distortion coefficient keyword values will be generated by an offline calibration subsystem: "gnDSTR" and inserted into FITS headers upstream, i.e., prior to SFPRex processing. 4. The CDi_j keywords encode rotation, scaling and a skew term. The skew is a shear of the Y axis to produce a non-orthogonal coordinate system. It can be treated as a constant and part of the distortion calibration. 5. The following keywords must never be used in place of (or used together with) the CDi_j keywords when distortion is present: CDELT1, CDELT2 and CROTA2. These keywords are reserved for image frames with no distortion (e.g., Atlas image co-adds and frames that have been corrected for distortion). 6. The MAGZP keyword represents the calibrated photometric zero point in magnitudes. This will be computed and inserted (or updated) into each level-1B FITS frame header by the PCAL subsystem. This keyword will be needed in the multi-frame (co-adder) throughput matching step.