WFPC2 Instrument Handbook for Cycle 11
The process of correcting for the effect of the variation in the sensitivity of the WFPC2 with field position is known as flat-fielding, or flattening. For ground-based observations, usually a "flat field" (an exposure of a spatially uniform source) is observed through the telescope with the desired filter. Unfortunately, there is no uniformly illuminated target available on-orbit. Instead, several assets are available to estimate the flat field and monitor any changes -- these include pre-launch SLTV optical stimulus flats, Earth flats, calibration channel flats (VISFLATS), and internal flats (INTFLATS).
During SLTV (System Level Thermal Vacuum) ground tests of WFPC2, flat fields were obtained using both the calibration channel and the WFPC2 optical stimulus (HST simulator). The later provided a close approximation to a uniform target as viewed through HST, and are a prime ingredient for the final calibration flats.
The Earth is an imperfect flat field target because it is too bright for the WFPC2 in the broad-band green and red filters. In addition, the rapid motion of the HST creates streaks across the flat field images, though the streaks can be removed by combining multiple Earth observations with the streaks at different angles on the CCDs. An extensive discussion of the generation of Earth flat fields is available in Chapter 6 of the WF/PC-1 IDT OV/SV Report, as well as in the History records of the flat field reference files themselves.
While imperfect, Earth flats are an important part of the flat field calibration; they provide corrections to the SLTV flats for any differences between the SLTV optical stimulus illumination, and the OTA illumination pattern. Flat fields in narrow bandpass filters are obtained using the sunlit Earth (Target = EARTH-CALIB) as part of routine calibration. These are used primarily to remove the low spatial frequency effects in the calibration flats.
Flat field calibration files have been generated for all filters by combining information from the SLTV test flats (which are good for all but the lowest spatial frequencies), and on-orbit Earth flats obtained for a small subset of narrow band filters (F375N, F502N and F953N). These Earth flats are used to correct low spatial frequency errors in the ground-based SLTV flats, which result from imperfect simulation of the HST OTA illumination pattern. These Earth flats taken regularly during available occultation time periods (i.e., no impact to science observations).
There are also two types of on-board flats available in WFPC2, which can be used to monitor changes in the flat field. The calibration channel (VISFLAT system) produces a reasonably flat illumination pattern down to about 1800Å. It works by imaging an illuminated diffuser plate onto the WFPC2 exit pupil (relay secondary) by means of an MgF2 lens. Two lamps provide optical and FUV illumination, yielding a flat field which resembles the input beam from the OTA between 1600Å and 10000Å. The system is mounted outside, but adjacent to, WFPC2, and light is directed into WFPC2 via a mechanically actuated flip mirror. A second system is much cruder, but provides a measure of redundancy: the internal flat system (INTFLAT system) consists of small lamps which, when commanded on, illuminate the back side of the shutter blade. The INTFLAT illumination pattern is not very uniform, but provides a robust backup capability.
The calibration channel data (VISFLATS) are used to monitor time dependent changes in the flat fields; only small changes have been seen to date in the visible filters. INTFLATS are also taken on a routine basis, and provide a redundant monitor capability. As of this writing (June 2001), both types on internal flats have been used only as monitors, with no corrections being made to the actual calibration files.
During early 1996 flat fields for most filters redward of F300W were updated using an improved illumination pattern derived from large numbers of streak flats. Corrections were typically 1%, though the outermost corner of WF3 showed a 7% correction. Regions near the CCD centers have very small corrections, ~0.3% RMS over 600x600 central pixels. The improved flats compared very favorably with sky flats. Internal consistency checks against the JHU Medium Deep Sky Survey sky flats indicates that the revised flats are better than 0.3% on spatial scales of 0.1" to 80".
Note that the flat fields presently used in the pipeline are based on gain 14 data. The gain ratios are not constant from chip to chip, and therefore a small correction to photometric results derived from gain 7 data should be applied (see Table 4.4). (See Biretta 1995 for further discussion of WFPC2 flat fields; also see the HST Data Handbook.)
Space Telescope Science Institute
Voice: (410) 338-1082