|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
 |
 |
|
|
|
Release of CALACS 2012.2 and new CTE-corrected data productsA new version of the data calibration pipeline CALACS has been released. It includes corrections for charge transfer efficiency (CTE) degradation and the electronic artifacts introduced by the repair of the Wide Field Channel (WFC) during Servicing Mission 4 (SM4). Users will see new data products when they retrieve ACS data from the Mikulski Archive for Space Telescopes (MAST).Please note that to retrieve the new CTE-corrected data products, users should request “calibrated” data. The new extensions (see below) will not be present in the "Files Extensions Requested" menu until after the MAST database has been reprocessed, which will occur in July-August. We anticipate a high demand from users wishing to re-process their ACS data in the next few weeks. The new CALACS takes ~ 40% longer to process an image and retrieval times may therefore be longer than normal. We ask that users who wish to submit large requests (more than 200 datasets) to please notify the Help Desk (help@stsci.edu) and we will ensure that these requests are handled properly by our processing and archive teams. The new CALACS is also available under IRAFX. Users wishing to reprocess data themselves should request the raw and reference files from MAST, and then run CALACS under HSTCAL (not STSDAS) in IRAFX (see instructions here). Pixel-based CTE correction The pixel-based CTE correction scheme is based on the work of Anderson & Bedin (2010, PASP, 122, 1035) but has been modified to include the time and temperature dependence of CTE losses (Ubeda & Anderson, ACS ISR 2012-03). An improved correction at low signal and background levels has also been incorporated as well as a correction for column-to-column variations. Bias shift, bias stripe, and crosstalk corrections The new CALACS contains corrections for three artifacts introduced by the new electronics installed during SM4. The "bias shift" (Golimowski et al. 2012, ACS ISR 2012-02) is a signal-dependent phenomenon associated with the CCDs' external pre-amplifiers and the dual-slope integrators within the CCD Electronics Box Replacement. The pixel-to-pixel bias level of each pixel is offset by 0.02–0.3% of the pixel signal, and this offset decays slowly in the serial (horizontal) clocking direction. Such small bias shifts, while unimportant for most ACS imaging, can be a significant hindrance for studies of faint, extended sources in the proximity of very bright or highly exposed sources. All full-frame post-SM4 ACS data are now corrected for this effect. "Bias striping" is the second artifact introduced by the SM4 repair. It is a low-amplitude, horizontal striping caused by electronic "1/f" noise in a reference voltage inside the replacement electronics. Grogin et al. 2011 (ACS ISR 2011-05) recently described this effect and how to correct for it using the image area. The standard deviation of the bias striping is 0.9 electrons, compared with the WFC read noise of ~4 electrons. The new CALACS corrects for this effect using the physical prescan region of all four quadrants rather than the image area discussed in Grogin et al. (2011). This new method is feasible because the bias-striping pattern is uniform across the quadrants of a given WFC CCD and is mirror-reflected across the CCD gap. Thus, all four prescan regions can be co-added and fitted together to provide an accurate correction for the bias striping. This task is done after the bias-shift correction, because this latter effect can affect the counts of the prescan regions. The third artifact that is corrected in the new CALACS is crosstalk between the amplifiers as the image is read out. Crosstalk has been present at a low level in WFC since the camera was first installed (Giavalisco 2004, ACS ISR 2004-13). This effect depends on the gain setting and a correction is now provided for post-SM4 observations taken with the default gain setting of 2 (Suchkov et al. 2010, ACS ISR 2010-02). The new CALACS The new CALACS includes additional steps for full-frame images only. First, all images are corrected for the signal-dependent bias shift, cross talk, and bias striping. The new CALACS then has two branches.
The first branch contains traditional CALACS processing using standard darks (_DRK reference files) to produce standard data products: _CRJ, _FLT and _DRZ.fits files. The new second branch will correct for CTE degradation, perform dark correction using new CTE- corrected darks (_DKC files), and then standard processing to produce new data products, called _CRC, _FLC and _DRC.fits files. The user will be able to choose whether to use the CTE-corrected or standard data products. In summary, there will be three pairs of data products from the new CALACS:
_CRJ = Cosmic-ray corrected FITS image
_FLT = Flat-field corrected FITS image
_DRZ = Drizzled FITS image
Note: any data-file type ending in "C" is the CTE-corrected equivalent of a standard calacs file. We welcome comments from users via help@stsci.edu. Note: we have not yet tested the efficacy of the CTE-correction algorithm for data taken with the grisms. We therefore ask users to compare old and new data products carefully for any grism data. Documentation on pixel-based CTE correction algorithm implemented in CALACS How accurate is the pixel-based CTE correction? ACS/WFC File Suffixes, Primary Header Keywords, and Location of New Reference Files Last modified on May 16, 2012.
|
|
| |||||||||||||||||
