About This Article
1. Notes for Phase 2 Submissions
J. Green, A. O’Connor
The HST Cycle 32 proposal results should be coming out in the next few weeks, and observers will have about a month to prepare their Phase 2 submissions. Here are a few notes and tips on Phase 2 submissions for Cycle 32:
- New 4-point dither patterns for improved dithering in WFC3-ACS prime-parallel programs are available as POS-TARGs that users can input into APT; see WFC3 ISR 2023-05.
- Remember to choose the appropriate postflash level to mitigate charge transfer inefficiency (dark + sky + postflash should total 20e-/pix; see WFC3 ISR 2021-09).
- Two new XML overlay files (created by WFC3 team member Peter McCullough) have been added for users to explore the potential for the Dragon’s Breath and Figure 8 ghost artifacts on UVIS images. They can be loaded through the field of view (FoV) tab in the Aladin file viewer, on top of the planned visit. Users can then tweak the overlay or their observation ORIENT to explore whether particular ORIENT restrictions might compromise their science results. If the user wishes to restrict their observation ORIENT but did not include this in the accepted Phase 1 proposal, they must appeal to the Telescope Time Review Board (TTRB) to apply this additional scheduling condition.
- Remember that anything not explicitly included in the accepted Phase 1 proposal must be approved. If a major change (ie. one that affects schedulability), a TTRB request is required.
2. New Release of WFC3 Data Handbook
A. Pagul, I. Rivera
The WFC3 team has updated the WFC3 Data Handbook (DHB) in an effort to better support the community and provide the latest findings, characterizations, and calibrations for WFC3 data. Updates to the previous edition of the Data Handbook are described in What's New in this Revision. Some highlights include:
- Updated UVIS superbias and saturation map reference file updates (Section 5.2.1).
- Updated discussion of UVIS CTE, including a short description of x-CTE (Chapter 6).
- Description of 4 new notebooks that are designed to aid users in working with IR file structures and visualization, as well as identifying and correcting issues with data (Section 7.10).
- Updated Photometry section, including a checklist of steps for measuring photometry of sources on images collected with the WFC3 UVIS and IR detectors, and a description of each (Section 9.1).
- Updated discussion of HSTaXe and slitlessutils for grism extractions (Section 9.5.6).
3. Revisiting x-CTE in WFC3/UVIS
J. Anderson
The harsh radiation environment of HST’s low-earth orbit exposes its CCD detectors (such as WFC3/UVIS) to energetic cosmic rays that can cause damage to the silicon lattice of the detector. This damage can prevent the free-flow of electrons through the lattice during readout. Electrons can get temporarily trapped in some pixels then released later into “upstream” pixels. This phenomenon is referred to as charge-transfer-efficiency (CTE) losses, and losses get progressively worse as the camera spends more and more time in orbit.
Most discussion of CTE losses involves the parallel transfer of charge in the y-direction down the chip columns. Parallel CTE losses for faint sources on low backgrounds have become an increasing concern for WFC3/UVIS. Thanks to the way the CCD’s electrodes confine the pixel-packets of electrons within pixels and move them through the silicon, the first few electrons in a pixel are much more susceptible to traps than the hundredth or thousandth electrons. When the background is extremely low (below 20 e-), faint sources can find the majority of their electrons trailed beyond recognition. For this reason, WFC3/UVIS images should always be taken with a minimum level of background (see WFC3 ISR 2021-09) to prevent pathological losses. Although losses are extreme for faint sources on low backgrounds, all sources suffer some CTE losses and users must fold in these losses into their analysis, either by using the pixel-corrected products or by implementing some post-measurement corrections (WFC3 ISR 2021-13 or WFC3 ISR 2024-04).
During readout, charge is parallel-shifted down into the serial register then shifted horizontally to the readout amplifier. Although the serial shifting of charge is more than 2000x faster than the parallel shifting, charge traps in the serial register can delay electron transfer and cause CTE losses, which manifest themselves as trailing in the serial direction. This effect was first explored in WFC3 ISR 2014-02, where it was shown to be at worst a 0.003-pixel effect, but since the detector has now spent more than three times longer on orbit, we decided to revisit this effect.
X-CTE losses are more subtle than Y-CTE losses, so it is harder to see visually in individual images. Figure 1 shows an extreme case of an intensely saturated cosmic ray hit in super-low-background bias image. It is clear that there is a moderate amount of flux released in the first 20 or so pixels and a low-level trail that goes out to almost 2000 pixels (wrapping around from one row to the next). In WFC3 ISR-2024-07, we use hot pixels, stars, and cosmic rays to explore all aspects of X-CTE. As of 2024, we find that, whereas Y-CTE affects bright sources at the 5% level and can affect faint sources by more than 50%, X-CTE affects bright sources at the ~1% level and faint sources at the 3% level. X-CTE trails have a sharp component that releases most of the trapped charge within one or two transfers, but there is also a broad component that releases charge over ~2000 transfers, often bleeding into the row above. Thanks to the sharp-trail component, X-CTE has a more significant effect on astrometry than on photometry. We develop a stand-alone pixel-based correction for X-CTE and make it available on the CTE-tools website: it can be operated on raw images, flt/flc images, or internally within hst1pass. They have not yet been folded into the calwf3 pipeline. Figure 2 shows the pixel-based correction for the extreme event of Figure 1.
4. WFC3 and WFPC2 PSF Update
F. Dauphin, J. Anderson, V. Bajaj
We refreshed the WFC3 PSF Database by remeasuring sources in every external WFC3 observation taken between 2009 and 2022 using hst1pass, also including saturated sources. As of February 2024, the database contains over 57 million source measurements, including pixel position, flux, and quality of fit. Specifically, it contains 32 million WFC3/UVIS sources (29 million unsaturated and 3 million saturated), and 25.2 million WFC3/IR sources (25 million unsaturated and 0.2 million saturated). These star-image cutouts can be used to generate custom PSF models appropriate for specific observations (e.g. filter, detector, focus-value, etc.). These can also be used to build delta-PSFs, which are the difference between custom models and the static library models. The custom PSF models are expected to be more representative of specific observations compared to the static library models, and hence may help in accurate photometry/astrometry.
A similar database is available for WFPC2, containing 25 million sources (15 million unsaturated and 10 million saturated) from external observations collected during the instrument’s lifetime.
The database of WFC3 observed PSFs are available on the MAST Portal. To access the WFC3 PSF database:
- Select “WFC3 PSF” under “Select a collection” (a similar option is available for the WFPC2 database).
- For WFC3, choose either “UVIS” or “IR” under “Waveband”.
- Click “Advanced Search” to select specific PSF parameters to search the database.
- Note that there is a pre-selected “Good Quality PSF Subset” (298 PSFs for UVIS, 163 PSFs for IR, 60 PSFs for WFPC2), which may serve as a guide to select the various parameters.
Once satisfied with the advanced searching parameters, click "Search". The PSFs can be visually inspected by clicking the “Show Cutout”, or by clicking "Album View" and "Cutout". The download option allows users to download the raw, and the calibrated (flt and flc) PSF image files.
A Jupyter notebook demonstrating a Python API to query the databases will be provided in the future.
5. A New Notebook for Point Spread Function Modeling
M. Revalski, F. Dauphin, V. Bajaj, J. Anderson
The WFC3 team has developed a Jupyter notebook that allows users to generate Point Spread Function (PSF) models for WFC3, ACS, and WFPC2 observations. A key strength of this new tool is the functionality to work with both individual exposures (FLTs/FLCs) as well as drizzled data products (DRCs/DRZs). Users are provided with several modeling options depending on their science goals and available data, including 1) downloading library PSF models for high-precision stellar photometry and astrometry, 2) extracting and stacking stars for characterizing extended wing emission and diffraction spikes, and 3) querying the MAST PSF image library to construct models for sparse fields by stacking stars from other observations.
The figures below highlight key aspects of the modeling options available in the notebook. The first figure displays a small region containing four bright stars from observations near the outskirts of Omega Centauri with HST WFC3/UVIS in the F606W filter. The three panels show the data, model, and residuals after PSF subtraction using a pre-existing PSF model provided by the WFC3 team. The second figure displays a PSF model for the F606W filter that was constructed by retrieving several dozen stars from the MAST PSF library cutout service using various observations, which were then aligned to sub-pixel precision and median stacked, which allows for PSF models to be created for any filter across WFC3, ACS, and WFPC2.
This tool is now available in the HST Notebooks Github repository under the WFC3 section, and is listed as hst_point_spread_function.ipynb.
6. WFC3/UVIS G280 Flux Calibration Updates
D. Som, B. Kuhn
The absolute sensitivity calibration of the WFC3/UVIS grism G280 has been updated using the most recent CALSPEC observed spectral energy distribution (SED) of GD-71. The previous flux calibration for the G280 grism (WFC3 ISR 2020-09) utilized v.009 of the CALSPEC model SED ("gd71_mod_009") for GD-71 to derive sensitivity curves for the various G280 spectral orders and for each UVIS detector. The present update to these sensitivity curves makes use of the latest observed CALSPEC composite SED ("gd71_stiswfcnic_004") which was normalized to the latest models for the primary CALSPEC spectrophotometric standard white dwarfs and the updated Vega gray flux (3.47 x 10-9 erg s-1 cm-2 A-1 at 5556 A in air, Bohlin et al. 2020). Figure 5 compares the updated sensitivity curves with their previous estimates for the +1 spectral order on both UVIS CCDs. The updated curves are universally lower compared to the older ones, with ~2% difference between them at ~1900 A and the difference gradually falls to ~1% at 7000 A. The latest sensitivity curves, therefore, will result in higher extracted flux.
Figure 5: Comparison of the previous version ('2021', in blue) of WFC3/UVIS G280 sensitivity with the current ('New', in red), updated sensitivity derived using the latest CALSPEC composite spectrum of GD-71, shown for the +1 spectral order on CCD1 (left) and the +1 order on CCD2 (right). The bottom panel in each plot show the percent difference between the old (2021) and the updated sensitivity estimates.
The updated sensitivity files and the latest configuration files (version 2.5) modified to point to these sensitivity files are now available on Box and the Grism Data Analysis page. Spectral extraction cookbooks demonstrating the use of these configuration and sensitivity files with HSTaXe can be found here.
7. New Documentation
ISR 2024-01: Sensitivity Evolution of WFC3/IR Using Spatial Scanning Photometry and Grism Spectrophotometry - D. Som, R. Bohlin, J. Mack, V. Bajaj, A. Calamida
ISR 2024-02: Improvements and Updates to the WFC3/IR Bad Pixel Tables Cycle 28-30 - K. Huynh & H. Khandrika
ISR 2024-03: WFC3/UVIS Guide Star Failure Classification with Machine Learning - M. Jones & F. Dauphin
ISR 2024-04: WFC3/UVIS External CTE Monitoring 2009-2024 - B. Kuhn
ISR 2024-05: The HST Focus Monitor Program 2019-2023 - L. Dressel & I. Rivera
ISR 2024-06: Time-Dependent Sensitivity of the WFC3/IR Detector (2009 - 2023) - M. Marinelli, V. Bajaj, A. Calamida, J. Mack
ISR 2024-07: Revisiting x-CTE in WFC3/UVIS - J. Anderson
The complete WFC3 ISR archive is available here. Additional information about WFC3 calibration, performance, data analysis, software tools, and more can be found online.
Need help? stsci.service-now.com/hst
To subscribe to the WFC3_NEWS list, please send a blank email to WFC3_NEWS-subscribe-request@MAILLIST.STSCI.EDU
To unsubscribe from the WFC3_NEWS list, please send a blank email to WFC3_NEWS-signoff-request@MAILLIST.STSCI.EDU