General Remarks and Limitations:
This document is meant to provide a quick overview of the primary differences between ACS and WFPC2. Much more detailed and accurate information can be found in the Instrument Handbook.
Observations designed for ACS can often be migrated to WFPC2 or NICMOS, provided one can accept some loss of sensitivity, resolution, or field-of-view. Exposure times and numbers of images can be increased to recover these losses, though whether the increase is too high, or whether the science is too gravely impacted are issues for consideration. Some observations will be very difficult or impossible to perform on WFPC2. Among these are imaging very faint targets (V>28), very high precision photometry (better than 2%), grisms, or coronography.
The WFPC2 field-of-view is approximated by a 150" x 150" square with one quadrant absent (L-shape defined by WF2, WF3, and WF4). The ACS WFC field is approximately 202" x 202" and the HRC is 27" x 27". The fields are illustrated in the figure below. The field of view of the WFPC2 PC1 CCD is 32" x 32".
Those wishing to image a large target or field may need to mosaic or "tile" multiple WFPC2 pointings. The most efficient method is to place WF3 into the missing quadrant of the prior exposure, though this produces an irregular shaped mosaic. A less efficient alternative is to use WF2 & WF3 in multiple pointings to build-up a rectangular mosaic. NB: for phase 1 proposals it is NOT necessary to give the RA and Dec for each mosaic pointing; it is sufficient to specify the region center and the total number of exposures in the mosaic.
Pixel Sizes and Resolution:
The WFPC2 WFC pixels are twice as large as those on the ACS WFC, hence the Point Spread Function (PSF) sampling is poorer, and the effective resolution is lower. Some information on finer scales can be recovered by sub-pixel dithering, but the full ACS sampling cannot be recovered. The ACS HRC pixels were significantly smaller than any on WFPC2, though again some fine-scale information can be recovered by sub-pixel dithering. See the WFPC2 Dither/Drizzle page for more details.
ACS: WFC = 0.05" HRC = 0.027" WFPC2: WF2,WF3,WF4 = 0.10" PC1 = 0.046"
Nearly all broadband filters on ACS have close counterparts on WFPC2. There are also numerous medium and narrow band filters on WFPC2 covering important bands and spectral lines. Ramp filters and polarizers are also available.
Both ACS and WFPC2 have ramp filters which can be used when no standard filter is available at a desired wavelength. In some cases an observation planned for an ACS ramp filter might be performed on a WFPC2 standard narrow band filter; there are 21 narrow band filters available on WFPC2. The ACS ramp filter fields of view vary in size, but 70" x 20" is typical. Most WFPC2 narrow band filters cover the entire WFPC2 field of view (some so-called "quad" filters are smaller); this large field of view can be a significant advantage when imaging large targets or regions vs. using the smaller ACS ramp filters. In other cases, it may be necessary to move an ACS observation to a WFPC2 ramp filter; proposers should be aware that the WFPC2 ramp filters have an unvignetted field of view of only ~10" x ~10".
Exposure Time Calculations:
ACS utilized a red-optimized CCDs in the WFC, and a blue optimized CCD in the HRC. In comparison WFPC2 has a single type of CCD with generally lower QE than those in ACS. WFPC2 has no health & safety issues for observing bright targets.
The table below gives a few comparative exposure times required to reach the SNR=10 for a G0V star. At middle wavelengths WFPC2 exposures tend to be a factor ~4 longer, but at short and long wavelengths, and for sky-limited cases the ratio becomes much larger. (Aperture diameter 0.2" for ACS, but 0.3" for WFPC2 to allow for re-sampling the larger WF pixels. WF CCDs assumed for WFPC2. Sky V=22.7. Nominal CR-SPLITS assumed. For the two emission line calculations, RAMP/F375N and F658N/F656N, we assume 10-15 erg cm-2 s-1.) CTE effects are ignored in this table, and can be very important in situations with faint targets and/or low background counts. See CTE discussion below.
Filter V Mag Exposure time to reach SNR=10 ACS/WFPC2 ----- ----------------------------- ----------- ACS WFPC2 RATIO ------- ------- ----- F220W/F218W 18 665s 22000s 33 F250W/F255W 19 221s 8000s 36 F330W/F300W 20 76s 400s 5.3 RAMP/F375N -- 618s 15000s 24 F435W/F450W 24 169s 600s 3.6 26 1480s 6400s 4.3 28 29000s 180000s 6 F606W 24 47s 180s 3.8 26 370s 1800s 4.9 28 6500s 52000s 8 F658N/F656N -- 73s 260s 3.6 F814W 24 69s 360s 5.2 26 560s 5100s 9 28 11300s 165000s 15 F850LP 24 235s 1800s 8 26 2066s 32000s 16 28 60000s --- --
We give one example for an extended target with V=23 per square arc second using F606W and SNR=10 per 0.1" square -- the ACS exposure time is 1400s and that for WFPC2 is 4600s.
Exposure time estimates depend on many variables. We recommend running the WFPC2 Exposure Time Calculator to obtain the best estimates for your observation.
Note: as of 2/1/2007 14:00 EST the WFPC2 ETC will default to a circular aperture, to make it more similar to the ACS ETC. There are still some differences however: the ACS ETC defaults to a 0.2" radius aperture, whereas WFPC2 defaults to 2 pixel radius. The previous WFPC2 default of Optimal PSF Weighting can still be manually selected in the ETC, if desired.
WFPC2 suffers significant Charge Transfer Efficiency problems from on-orbit radiation damage. These effects will be smallest for long exposures in broadband filters such as F606W, F625W, F785LP, F814W, where the sky background serves to "preflash" the CCD. For example, stellar images near the center of the CCD in a 600 second F606W exposure will suffer roughly the CTE losses shown in the table below. Correction equations available, and these can be applied provided the target is well-detected. After correction the photometry will be accurate to a few percent, though of course, the signal-to-noise ratio is degraded by loss of counts.
However, the problem becomes more severe for narrow and UV filters where there is little or no sky background. For example, in a 600 second F656N exposure, the CTE losses at the CCD center for stellar targets are roughly:
Star total CTE loss CTE loss counts F606W F656N 1000 DN 8% 13% 100 DN 20% 32% 10 DN 30% 51%
Observers who require very accurate photometry or detection of faint objects are urged to estimate the CTE effects for their observation. Several tools for estimating CTE effects on stellar targets are available. Of the three tools, CTE tool #1 contains the most recent results. It will require the following inputs: MJD -- use 54500 for mid-Cycle 16; the source counts -- use the value of "object_counts" near the bottom of the ETC output page; the CCD X and Y positions -- use (400, 400) for the CCD center, or (800, 800) if you need the worst case for an object anywhere in the field-of-view; the background counts -- use the value of "sky_counts" again from the bottom of the ETC output page; and finally the stellar magnitude of the target -- this only need be very approximate. It will then output XCTE and YCTE, which are the magnitude changes due to two types of CTE loss, or equivalently the fraction of counts lost to CTE. For example, XCTE=0.01 and YCTE=0.17 imply 0.18 magnitudes loss or ~18% loss in counts.
One can use this estimated CTE value in several ways. In this example an observer might want to expose 18% longer to ensure the desired signal-to-noise ratio. Those requiring accurate photometry should be aware that this 18% effect will need correction during data analyses, and that the corrections are likely imperfect. The accuracy of the CTE corrections is not well established, but an uncertainty of 10% to 20% is likely. Hence the photometric uncertainty due to CTE (even after correction) would be in the range 2% to 4% for this example.
For extended targets the effects are expected to be much less, as the leading edge of the image effectively preflashes the CCD during readout.
For small targets (few arcseconds in size), CTE losses can be greatly reduced by placing the target in the CCD corner nearest the readout amplifier. In this way CTE losses can be reduce by a factor of ~4 from the values given above.
Pre-flashing is another way of reducing CTE losses in WFPC2, though it will add noise to the images, and increase overhead times by about 3 minutes per exposure.
The WFPC2 CCDs saturate around 53000 electrons, whereas the ACS/HRC CCD saturates near 140000 electrons. This will make simultaneous observation of very bright and very faint objects more difficult with WFPC2.
WFPC2 WF4 CCD Anomaly:
The WF4 CCD, one of the 3 wide field camera CCDs, suffers an anomaly in the readout electronics whereby the CCD amplifier gain and bias levels are unstable. This results in altered photometry and faint horizontal stripes (<1 DN) in the background. The error is fairly well characterized and the photometry can be corrected to a few percent accuracy during post-processing. The stripes are easily removed by spatial filtering in fields with few objects; however, in crowded fields or in the presence of extended objects filtering will be difficult. Details are available in the WFPC2 Instrument Handbook and in WFPC2 ISR_2005-02.
WFPC2 observations tend to have larger overhead times than ACS. A WFPC2 exposure requires a minimum of 2 minutes regardless of the exposure time, and an additional 1 minute if a filter change is needed before the exposure. This limits the maximum number of WFPC2 exposures per 53m orbit to ~20.
There are some special circumstances where a proposer may wish to consider a change to NICMOS instead of WFPC2. These could include coronagraphy, polarimetry and photometry in the far-red optical/NIR region. The NICMOS Instrument Handbook provides details of these observing modes.
Version 1.3 2/5/2007 JB