About This Article
1. Ultra-Violet Color Term Transformations
A. Calamida.
The UVIS1 and UVIS2 detectors have different quantum efficiencies in the ultra-violet (UV) regime (lλ < 4,000 Å): count rate ratios change as a function of the spectral type of the source. When calibrating photometry of stars cooler than Teff ~ 30,000 K in the UV filters (e.g. when observing open and globular clusters, resolved local group galaxies, Galactic stellar populations), color term transformations need to be applied to UVIS2 magnitudes. Sources of any spectral type observed on one detector only will not require any magnitude offset.
The left panel of the figure below shows the synthetic ST magnitude difference, UVIS1 – UVIS2, for a sample of CALSPEC stars of varying spectral type (the white dwarfs (WDs) include G191B2B, GD71 and GD153), as computed for three UV filters, F218W, F225W, F275W. Cool red sources such as the G-type star P330E measured on UVIS2 have a magnitude offset relative to UVIS1 up to ~0.08 mag when observed with the F225W filter. F-type stars such as HD160617 have a magnitude offset of ~0.04 mag.
The right panel of the figure shows the ST magnitude difference for the F225W and F275W filters for ω Cen Extreme Horizontal Branch (EHB), Horizontal Branch (HB), Main Sequence (MS), and Red-Giant Branch (RGB) stars measured on different detectors and amplifiers, A (UVIS1) and C (UVIS2). WFC3 observations of ω Cen validate the results of synthetic photometry: red stars (RGBs) observed on UVIS2 have a magnitude offset up to 0.08 mag relative to UVIS1.
Calamida et al. 2018 ( WFC3 ISR-2018-08) provides lookup tables with color term transformations to be applied to UVIS2 magnitudes when observing with three UV filters, F218W, F225W, and F275W. UVIS2 magnitudes are scaled to UVIS1 by the WFC3 calibration pipeline (cal_wf3) using UVIS2 to UVIS1 modified inverse sensitivity ratio, PHTRATIO. PHTRATIO is derived using photometry of the CALSPEC WDs and is valid for hot stars, Teff > 30,000 K. For cooler stars, when observing with UV filters, PHTRATIO is not equal to the ratio of the two detectors' count rates but changes with the stellar spectral type. Photometry for cooler stars measured on UVIS2 needs to be corrected by applying a magnitude offset according to their color, if available, or temperature or spectral type.
Before applying the offset to magnitudes measured on UVIS2, photometry must be calibrated using UVIS1 inverse sensitivities provided here.
STMag (UVIS1) = -21.1 -2.5 x log(PHOTFLAM) STMag (UVIS2) = -21.1 -2.5 x log(PHOTFLAM) + Delta (Mag)
where Delta (Mag) = Mag(UVIS1 – UVIS2) is listed in lookup tables in WFC3 ISR-2018-08. For more details and to consult the color transformation tables please see the referred ISR.
2. Update on Jitter
J. Anderson, S. Baggett.
We reported in the last STAN on the increased jitter in the PSF. Since then, in Oct 2018, the problematic gyro-2 has failed and gyro-3 has been brought into the loop (see Newsletter for more details). Monitoring of the WFC3 jitter (jif) files shows that as a result of these gyro changes, the jitter RMS values have dropped to ~7 mas and ~2.5 mas in V2 and V3, respectively. The plot below summarizes all WFC3 results, from 2009-2018. At right, near the end of 2018, are results from data taken under the current gyro trio 3,4,6. The 7 mas does represent a detectable broadening in the PSF but it should not have a significant impact on photometry or astrometry; only high-precision PSF-fitting will be impacted.
3. New DrizzlePac Tutorials Available
J. Mack, S. Hoffmann, C. Martlin, R. Avila, V. Bajaj, M. Cara, T. Desjardins, C. Shanahan
Improved drizzling tutorials are now available as Jupyter notebooks and are compatible with the latest STScI software distributed via AstroConda. Prior drizzling examples were included in the 2012 DrizzlePac Handbook , just after MultiDrizzle was replaced, and supplemental examples were posted to the DrizzlePac Webpage in 2015 to support enhanced features in DrizzlePac 2.0. The new interactive notebooks consolidate information from these prior examples to form a more cohesive set, and any references to outdated software, such as PyRAF, have been removed and replaced with python functionality.
The notebooks contain live code and visualizations, making them an ideal training exercise for new users. Each tutorial includes blocks of code demonstrating how to download the calibrated files from MAST, how to align frames and update the image world coordinate system, and how to enhance the scientific value of the drizzled data products using advanced reprocessing techniques.
Nine new notebooks are available within the 'DrizzlePac' directory in the 'Notebooks' folder found at the STScI Notebooks GitHubrepository:
- Initializing DrizzlePac
- Aligning observations obtained in different HST visits
- Aligning HST images to an absolute reference catalog (e.g. GAIA, SDSS)
- Drizzling WFPC2 data to use a single zeropoint
- Improving alignment with DS9 exclusion regions
- Masking satellite trails in DQ arrays prior to drizzling
- Optimizing the image sampling for dithered datasets
- Sky matching features for HST mosaics
- Aligning HST mosaics observed with multiple detectors
Additional tutorials will be added as new software functionality becomes available, especially for advanced use cases. For additional assistance with DrizzlePac tools, users may submit a ticket to the STScI Help Desk or send email to help@stsci.edu.
4. Improved UVIS Drizzled Products from MAST
J. Mack.
Starting in January 2019, improved drizzling parameters will be utilized by MAST to generate visit-level drizzled data products for the UVIS detector. With recent changes to the UVIS bad pixel tables described below (see the STAN article below on 'Reclaiming WFC3/UVIS hot pixels'), the data quality (DQ) flags now discern between unstable and stable hot pixels, the latter of which are corrected when subtracting the dark reference file. Thus, pixels identified as hot and stable (DQ flag=16) may now be treated as 'good' data when drizzling, and those identified as unstable (DQ flag=32) should be treated as 'bad'.
The new MDRIZTAB reference table (2ck18260i_mdz.fits) contains a set of default parameters which are called by 'AstroDrizzle' to correct for geometric distortion, flag cosmic-rays, and combine sets of exposures, and these parameter values depend on the UVIS filter and the number of input exposures. With changes to the DQ flag definitions, the parameters 'driz_sep_bits' and 'final_bits', which define DQ flags for drizzle to ignore (e.g. to treat as good), are now set to a value of 336 (the sum of 16+64+256) so that stable hot pixels, warm pixels, and full-well saturated pixels will not be rejected when combining exposures. These new flags are valid for UVIS observations obtained after Nov 08 2012, when the dark calibration program began using post-flash to mitigate hot pixel trailing due to poor charge transfer efficiency at low background levels. Prior to this date, all hot pixels (stable or unstable) are flagged with a DQ value of 16 should be treated as bad.
Other improvements to the default parameters are similar to those implemented by ACS/WFC in early 2017, when tests were performed to optimize point source photometry for drizzled data (ACS ISR 2017-02). These include a new row for N=4 images, e.g. observations acquired using a 4-point dither, which previously defaulted to the N=2 parameters. For single exposures (N=1), drizzled photometry is noticeably improved when all DQ flags are ignored (ACS ISR 2007-02, Figure 4), and the parameter 'final_bits' has been set to a value of 65535 (the sum of all bits). This prevents 'AstroDrizzle' from interpolating over flagged pixels in single exposures, which has been shown to impact the photometric accuracy when flagged pixels lie within a few pixels of the center of a star. Additionally, the parameter 'driz_cr_scale', used to fine-tune the cosmic-ray rejection, has been relaxed for N=2 and N=4 exposures in order to avoid clipping the cores of stars in dithered exposures when they are positioned at the center of a pixel in one exposure and at the corner of a pixel in the second exposure and thus have slightly different radial profiles. While it is still recommended that each user reprocess their data to find the best 'AstroDrizzle' configuration for their specific program, this improved MDRIZTAB will provide users with higher quality drizzled products from the archive.
5. Updates on 2018 UVIS Darks
C. Martlin, J. Medina.
There has been a re-delivery of all UVIS pipeline darks (*_drk.fits, *_dkc.fits) used by WFC3/UVIS observations taken from December 2017 to June 2018. Due to a software mistake, the dark reference files during this time period were made using the November 2017 anneal period masterdark file. As described in detail in WFC3 ISR-2018-15, dark reference files during any given anneal period should use the calculated average value of the previous anneals’ masterdark to populate the normal i.e. non-hot pixels. Since the overall dark current increases very slowly over time (~1 electron/hr/year), the difference from one anneal period to the next in the masterdark averages is small, but over time it became clear that the software was using the same value for several months in a row. The code has been corrected and the ~7 months of affected pipeline darks regenerated and delivered.
Figure 1, the top figure below, shows the difference in the median dark current in electrons per hour that should have been used for each anneals’ reference files compared to the median dark current that was incorrectly used in the first delivery (Note: The anneal cycles are denoted by the alternating shaded/un-shaded columns). As the figure shows, the differences range from about -0.2 electrons/hour for chip 2 February 2018 darks to, at most, 0.6 electrons/hour for chip 2 in July 2018. In Figure 2, the bottom one below, we show the actual value of median dark current in electrons per hour that is applied in the re-delivered reference darks for each anneal in blue versus the incorrect median dark current in electrons per hour that was applied to the dark reference files in the first delivery in red. Observers with data from December 2017 - June 2018 who wish to have their data automatically reprocessed with the new darks can re-download their data from the archive. Alternatively the updated dark reference files are also available through CRDS.
6. Reclaiming WFC3/UVIS Hot Pixels
M. Bourque, D. Borncamp, S. Baggett, T. Desjardins, & N. Grogin.
Based on a detailed pixel-by-pixel analysis of on-orbit UVIS dark frames, we find that the majority of hot pixels (~8 % of each chip in 2018) are in fact quite stable over time and as such, are able to be calibrated. A small portion of pixels (<1%), some cold and some hot, are genuinely unstable and should not be used (see figure below).
To encode this distinction in the calibration pipeline, we employ the previously-unused data quality (DQ) flag of 32 to mark all unstable pixels in monthly bad pixel tables. Hot pixels will continue to be assigned the customary DQ flag value (16) in the daily pipeline dark reference files. With the additional DQ flag, WFC3/UVIS users will have enhanced control over which pixels to include, and which to exclude, in their analyses.
The new bad pixel tables for science data taken since Nov 2012 have been installed in the automated calibration pipeline; observers wishing to have their data automatically re-calibrated with the new files can re-download their data from MAST. Observers who work directly with the flt/flc files can continue to choose which DQ-flagged pixels to discard and whether to make use of the new flag definitions. For observers who work with AstroDrizzle products, there is a new MDRIZTAB parameter file which will retain the stable hot pixels in the final image stack for observations after Nov 2012 (see the prior STAN article 'Improved UVIS Drizzled Products from MAST').
For more details on the pixel analysis, please see ISR 2018-15 (Bourque et al.) For a description of DQ flag values in WFC3/UVIS data prior to this change, please see the Data Handbook (Gennaro et al.).
7. New Documentation
- ISR 2018-09 WFC3/IR: Time Dependency of Linear Geometric Distortion – M. McKay, V. Kozhurina-Platais.
- ISR 2018-10 Updates to the WFC3/UVIS Filter-Dependent and Geometric Distortions – C. Martlin, V. Kozhurina-Platais, M. McKay, E. Sabbi.
- ISR 2018-11 UVIS Flat Fields Affected by Shutter-Induced Vibration – H. Kurtz, P. R. McCullough, S. Baggett.
- ISR 2018-12 New Calibration in Cycles 23-26 & Detector Monitoring Results over the WFC3 Lifetime – J. Mack & the WFC3 Team.
- ISR 2018-13 Linear Reconstruction of Grism Spectroscopy I. Simulation and Extraction Examples – R. E. Ryan, S. Casertano, N. Pirzkal.
- ISR 2018-14 Focus-Diverse PSFs for Five Commonly Used WFC3/UVIS Filters – J. Anderson.
- ISR 2018-15 Using Dark Images to Characterize Pixel Stability in the WFC3/UVIS Detector – M. Bourque, D. Borncamp, S. Baggett, T. Desjardins, N. Grogin.
- ISR 2018-16 WFC3/UVIS - Temporal and Spatial Variations in Photometry – H. Khandrika, S. Deustua, J. Mack.
- ISR 2018-17 WFC3/UVIS Gain Stability Results for Cycles 24 and 25 – J. Fowler.