B.1 Overview

WFC3 images exhibit significant geometric distortion, similar to that seen in ACS images. The required folding, with powered optics, of the light paths in both channels to fit within the instrument’s optical-bench envelope results in substantial tilts of the focal surfaces with respect to the chief rays. The WFC3 UVIS detector is tilted at ~21° about one of its diagonals, producing a rhomboidal elongation of ~7%. The IR detector has a ~24° tilt about its x-axis, creating a rectangular elongation of ~10%.

If these were the only distortions they would not present much difficulty: their impacts on photometry, mosaicking, or dithering could be computed simply. More problematic, however, is the variation of plate scale across each detector. For the WFC3 UVIS and IR channels, this variation in plate scale amounts to a change of 3.5% in x and y, and 2% in x and 6% in y, respectively, over the full field. Hence the area on the sky covered by a pixel varies, by about 7% for the UVIS channel and about 8% for the IR channel. Allowance for this change in plate scale must be made in photometric reductions of WFC3 data that have not been corrected for distortion. Further details are available in WFC3 ISR 2010-08 and at the pixel area map section of the WFC3 website: http://www.stsci.edu/hst/instrumentation/wfc3/data-analysis/pixel-area-maps

Dithering and mosaicking are further complicated by the fact that an integer pixel shift near the center of the detector translates into a non-integer displacement for pixels in other locations. Even this is not a fundamental difficulty, but implies some computational complexity in registering and correcting images. All of these considerations make it necessary to obtain accurate measurements of the distortions. The orientations of the WFC3 detector edges for both detectors are at approximately 45° with respect to the V2 and V3 coordinate axes of the telescope. Figure 2.2 shows the WFC3 apertures in the telescope’s V2,V3 reference frame. For a telescope roll angle of zero this would correspond to an on-sky view with the V3 axis aligned with north and the V2 axis with east. See Section 6.2.2 of the Phase II Proposal Instructions, which gives detailed information on the relationship between detector coordinates, spacecraft coordinates, and ORIENT.

The first on-orbit measurements of the geometric distortion for the WFC3 detectors were made in 2009 during SMOV (Servicing Mission Observatory Verification). Astrometric fields in 47 Tuc (NGC 104) and the LMC were observed with multiple offsets in Program 11444 (UVIS, filter F606W) and Program 11445 (IR, filter F160W). Geometric distortion solutions were derived from this data (WFC3 ISR 2009-33; WFC3 ISR 2009-34) and entered into Instrument Distortion Correction (IDCTAB) files to support the use of MultiDrizzle¹ to produce distortion-corrected images. In these initial IDCTABs, the solutions for filters F606W and F160W were applied to all UVIS and IR filters, respectively. There are additional small filter-dependent offsets in distortion, so exposures made with other filters during SMOV and in subsequent calibration programs observing Omega Centauri were used to derive improved solutions for more commonly used filters (WFC3 ISR 2012-07; WFC3 ISR 2018-10). Table B.1 lists all IDCTAB deliveries for IR and UVIS. 

Variations in the thickness of a given filter modify the distribution of the light beam, which can cause an apparent displacement of the observed sources, potentially contributing to geometric distortion. In most cases, the relative displacement of stars in exposures made with different filters due to this non-coplanarity of the filters is ~0.01 arcsec for IR (WFC3 ISR 2010-12) and ~ 0.02 arcsec for UVIS (WFC3 ISR 2012-01).

Distortion solutions can be complicated by detector defects. As in the case of the ACS/WFC detector, a lithographic-mask pattern was imprinted onto the WFC3/UVIS CCDs during the manufacturing process, causing filter-independent pixel-grid irregularities that are too complicated to be easily modeled by high order polynomials (WFC3 ISR 2009-33; WFC3 ISR 2013-14). To correct these fine-scale pixel variations and therefore improve astrometric accuracy, DrizzlePac calls a 2-dimensional lookup table (the D2IMFILE reference file), before applying filter-dependent distortion corrections (WFC3 ISR 2014-12; Section 1.3, DrizzlePac Handbook). 

After correcting for the lithographic-mask pattern, imperfections of the filters themselves manifest as a complex fine-scale structure of systematic errors, varying in amplitude across a CCD chip (WFC3 ISR 2014-12; WFC3 ISR 2018-10). Corrections for these residual non-polynomial distortions are stored in a filter-dependent lookup table (NPOLFILE reference file). As illustrated in WFC3 ISR 2014-12, correcting for both the lithographic-mask pattern and the filter-dependent residual systematics reduced astrometric errors to the level of ~1 mas for many UVIS filters.

The WFC3 instrument team continually monitors and evaluates the temporal stability of UVIS and IR distortion solutions. Previously, observations of Omega Centauri indicated that the WFC3/UVIS distortion solution for F606W was stable (i.e. did not meaningfully change over time), as discussed in WFC3 ISR 2015-02, WFC3 ISR 2018-10, WFC3 ISR 2019-09, and WFC3 ISR 2021-07. Similarly, in WFC3 ISR 2018-09, exposures of Omega Centauri in the WFC3/IR F160W filter indicated no significant time-dependence in the geometric distortion solution. 

Recent analysis for both UVIS (WFC3 ISR 2024-15) and IR (WFC3 ISR 2025-04) took a more expansive approach, aligning thousands of images in five commonly-used filters² to the Gaia DR3 catalog instead, enabling significantly larger and more complete datasets. For UVIS, the geometric distortion solution was found to have a small overall time-dependent change with respect to Gaia DR3, and one component of the geometric distortion (the linear plate scale term) exhibited a filter-dependent offset of up to 0.3 pixels (WFC3 ISR 2024-15). The IR detector only exhibited temporal evolution for the linear shift term, which is dominatated by uncertainties in the telescope pointing (WFC3 ISR 2025-04). In the future, IDCTABs could potentially be updated to address small offsets and incorporate time-dependent distortion solutions. However, precise astrometry would still require the user to manually align images with DrizzlePac, which mitigates the slight time-dependent changes in the geometric distortion. 

Delivered IDCTABs with distortion solutions for WFC3/IR filters and WFC3/UVIS filters are listed in Table B.1. Any IR filter or UVIS filter not listed in the delivered IDCTABs is still using the solution derived for F160W and F606W, respectively. All UVIS filters listed in the table have accompanying NPOLFILE reference files available. The NPOLFILE reference files are the 2-D look-up-tables of the filter-dependent fine scale distortion (WFC3 ISR 2014-12; WFC3 ISR 2018-10).

Table B.1: IDCTAB (Instrument Distortion Correction Table) deliveries for the IR and UVIS filters

¹MultiDrizzle has since been replaced by DrizzlePac

²In UVIS, 7,491 total images across F275W, F336W, F438W, F606W, and F814W. In IR, 9,150 total images across F105W, F110W, F125W, F140W, and F160W.