STScI Logo

Hubble Space Telescope
WFC3 STAN Issue 25, June 2017

WFC3 Space Telescope Analysis Newsletter - Issue 25, June 2017

Late-breaking news about WFC3 is posted here.

Previous issues of the WFC3 STAN are archived here.


  1. Cycle 25 Phase II Deadline Reminder and Updates to APT
  2. Improved UVIS Photometry
  3. WFC3 Grism Calibration Updates
  4. New Bias and Post-Flash Calibration Reference Files
  5. Possible Overlaps Between Blobs, Grism Apertures, and Dithers
  6. WFC3/UVIS External CTE Monitor: 2016 Updates on Coefficients and Analysis Pipeline
  7. Long-term Stability of the Post-Flash LED Lamp
  8. New Documentation

1. Cycle 25 Phase II Deadline Reminder and Updates to APT

L. Dressel

Observers who are awarded time on HST for Cycle 25 should note that the submission deadline for Phase II proposals is July 20, 2017. Information regarding Phase II development and submission is posted here. Cycle 25 phase II proposals must be submitted through the latest version of the Astronomers Proposal Tool (APT 25.2). Updates included in this version of APT are described here.

2. Improved UVIS Photometry

J. Mack & S.E. Deustua

On Jun 19, 2017, an improved photometric calibration was implemented in MAST for all 42 WFC3/UVIS full frame filters. Chip-dependent inverse sensitivity values reported in the image header (PHOTFLAM, PHTFLAM1, PHTFLAM2) revert back to the ‘infinite’ aperture, which is defined at a radius of 6 arcseconds (151 pixels), and these supercede the values implemented in February 2016 which were reported for an 0.3962 arcsecond (10 pixel) aperture. The improved calibration is described in WFC3 ISR 2017-14 and includes better polynomial fits to the wavelength-dependent components of the throughput response and improved models for the three white dwarf standards.

The revised calibration is implemented via a new image photometry reference table (IMPHTTAB = 1681905hi_imp.fits) and is concordant with the current synthetic photometry tables available in the calibration reference data system (CRDS), which are described in WFC3 ISR 2016-07. Due to the change in aperture convention, the new keyword values are ~10% smaller than the 2016 values. Comparing values at the same aperture, the new values differ by ~0.5% on average. The 2017 VEGAmag zeropoints change by up to 0.1 mag in the UV compared to 2016 and are calculated using the CALSPEC STIS spectrum for Vega rather than the Vega model, which differs significantly at wavelengths shorter than 3000 .

This latest calibration is highlighted on the WFC3 main photometry webpage, which provides a link to the new UVIS tables for both the infinite and 10 pixel (0.3962 arcsecond) aperture. For observations spanning multiple epochs, users are advised to verify that the same IMPHTTAB reference file was used to process all data.

For example, if the data are retrieved from MAST at different times (e.g. after the execution of each visit), it may be possible to observe systematic differences due to changes in the latest reference files, or more generally, to the entire data processing flow. Users have two options: 1) retrieve all pertinent datasets for the given science program from MAST or 2) reprocess the RAW files offline with a self consistent version of calwf3 and reference files. See the WFC3 Data Handbook, section 3.7 for manual reprocessing instructions.

3. WFC3 Grism Calibration Updates

N. Pirzkal, R. Ryan, G. Brammer

The trace and wavelength calibrations of the WFC3 grisms have been updated. New field dependent solutions for the G102 and G141 grisms, using all of the available calibration as well as monitoring observations, have been determined. The calibration of these grisms was improved to better than 0.1 pixel for the trace dispersion and better than 0.5 pixel in wavelength (~10A and ~20A for the G102 and G141 grisms, respectively). The extent of the field of the view edges was also determined and are -22 arcseconds and +10 arcseconds on the left and right edge of the IR detector, respectively. This is described in ISR 2016-15.

New field dependent trace and wavelength calibrations for the UVIS G280 grism were also developed and released. The +1 and -1 orders on the CHIP1 as well as the CHIP2 detectors were calibrated. The accuracy of the wavelength calibration of the +1 and -1 orders is estimated to be ~7A, which is approximately half the size of a UVIS resolution element. This is described in ISR 2017-20.

An improved method to handle the variable background levels during long WFC3 grism exposures was described in ISR 2017-05. The method allows for the variable HeI component to be removed from individual read-outs and therefore allows for the use of the on-the-ramp fitting with CALWF3. The results are an improved signal-to-noise in the final calibrated images as well as the ability to use on-the-ramp fitting to detect and correct for cosmic rays in observations affected by a large amount of time-variable He I light from the Earth sky.

4. New Bias and Post-Flash Calibration Reference Files

H. Kurtz, M. McKay, S. Baggett

New Reference files for the post-flash and bias were delivered to CRDS on June 16, 2017. These files have no significant change from the previous reference files. The specifics of these reference files can be read about here.

2009 - 2016 Bias Reference Files

New UVIS bias reference files from 2009 through 2016 have been incorporated to the Wide Field Camera 3 calibration pipeline. The current files will be used for the bias correction (BIASCORR) for all UVIS full frame, subarrays and quad arrays. The current reference files were constructed using significantly more files than the previous reference files for a higher signal to noise ratio. The average differences from previous to the current reference files is ~0.01 electrons/pix. For more information see the history and comment section located in the header of the current reference files.

UVIS Bias Reference Filenames (BIASFILE) and Pedigrees
Pedigree Full-frame Subarray & quad arrays
July 26, 2009 – Dec 31, 2009 1681902ti_bia.fits 16819033i_bia.fits
Jan 1, 2010 – May 27, 2010 16819037i_bia.fits 1681903bi_bia.fits
May 27, 2010 – Dec 31, 2010 1681903ei_bia.fits 1681903ii_bia.fits
Jan 1, 2011 – Dec 31, 2011 1681903mi_bia.fits 1681903qi_bia.fits
Jan 1, 2012 – Dec 31, 2012 16819040i_bia.fits 16819043i_bia.fits
Jan 1, 2013 – Dec 31, 20139 16819047i_bia.fits 1681904bi_bia.fits
Jan 1, 2014 – Dec 31, 2014 1681904fi_bia.fits 1681904ji_bia.fits
Jan 1, 2015 – Dec 31, 2015 1681904mi_bia.fits 1681904qi_bia.fits
Jan 1, 2016 – Dec 31, 2016 16819050i_bia.fits 116819053i_bia.fits

Post-Flash Reference Files

We update the current low and medium current post-flash reference files as reported in ISR-2017-13. The post-flash reference files have less noise than the previous ones as they combine 120 images for each shutter. Based on the 100 second flash high signal-to-noise images used to generate the reference files, the post-flash LED output appears to be decreasing slightly, about 0.2% per year. However this decrease is not seen at the significantly lower flash levels used for typical GO science data (see article 7 below). Currently no time-dependence has been implemented into the post-flash reference files pending corroboration of this drift in future monitoring data.

5. Possible Overlaps Between Blobs, Grism Apertures, and Dithers

R. Ryan & P.R. McCullough

In a recent ISR, Ryan & McCullough (2017) investigated the possibility that an IR blob may impact spectrophotometry of sources on, or near, the grism aperture reference points. They examine the standard WFC3/IR dither patterns with the list of known blobs, and conclude that only there are only two blobs that may affect the GRISM1024 aperture with the WFC3-IR-DITHER-BOX-UVIS dither pattern. These effects are expected to be negligible, but we provide a brief Python script so users may test customized dither patterns, reference positions, or high-value positions on the detector. The script can be downloaded here.

6. WFC3/UVIS External CTE Monitor: 2016 Updates on Coefficients and Analysis Pipeline

C.M. Gosmeyer & S. Baggett

The evolution of the charge transfer efficiency (CTE) of Hubble’s WFC3/UVIS detector has been monitored since 2009, using observations of star clusters NGC 104 and NGC 6791. In 2016 we updated the analysis pipeline, improving the cross-identification of sources and adding a capability to study CTE loss mitigation by use of flash levels 6 - 116 e-/pix.

All results discussed in ISR 2017-09 use this new analysis pipeline, with data spanning from October 2009 to August 2016. Topics include:

  1. Flux loss measurements as function of source brightness, observation date, background level, and vertical distance from the readout amplifier (see Figures 8-11 in the report). Current flux losses for images with the recommended minimum 12 e-/pix background are 5 - 15%, depending on source brightness. These losses are further reduced by using the CTE-corrected images available from the Mikulski Archive for Space Telescopes (MAST).
  2. Updated coefficients for estimating flux corrections on point-source photometry (see Table 1 in the report or download them as text files).
  3. CTE trends with time and flash level (Figures 12-13 and 14-15 , respectively).
  4. A flattening in the CTE degradation rate that may be correlated with increased solar activity.
  5. Percentages of sources recovered using CTE-corrected images (Figure 16). Applying the CTE correction to a non-post-flashed image can recover nearly 50% of sources at the low flux end. In post-flashed data, where the CTE loss has already been partially mitigated, the CTE correction can recover nearly 10% of sources at the low flux end.
  6. The use of IRAF- vs Python-based photometry (pages 11-13).

7. Long-term Stability of the Post-Flash LED Lamp

C. Martlin & S. Baggett

We report in ISR 2017-03 the results of a study of the long-term stability of the LED lamp used in post-flashing WFC3/UVIS images. Having analyzed 644 sub-array exposures taken over the course of just over 3 years we find no significant long-term trends in the LED lamp brightness. The average percentage change per year over all FLASH level and shutter combinations is found to be 0.15% per year with an uncertainty of +/- 0.24% per year. The maximum measured percentage change per year was 0.39% for the highest FLASH level of 20 e- on shutter A which would indicate a change of less than 1 e-/pixel over the roughly 3 years since post-flashing began.

There are, however, occasional short-term deviations in the LED output. The average minimum and maximum outliers over all FLASH level and shutter combination subsets are 4.1% below and 3.5% above, respectively, the mean of that subset. Investigation of these outliers has revealed no systematic cause for the excursions and we recommend that users continue using the post-flash as they would have previously.

8. New Documentation

The archive of all WFC3 Instrument Science Reports (ISRs) is here.

The latest version (4.0) of the WFC3 Data Handbook is here.

Need help? Start here.

To subscribe or unsubscribe to the STAN, send a message to with a blank subject line and the following in the body: [un]subscribe wfc3_news

The Space Telescope Science Institute is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555.