WFPC2 Photometry FAQs

Q: Are my data flux calibrated?

A: The pipeline calibrated data are not flux calibrated. The data are in units of Data Numbers (DN). The formula to convert from DN to flux density is:

    Flux Density = DN x PHOTFLAM / EXPTIME

where PHOTFLAM and EXPTIME are header keywords.

Q: Where can I find the throughput curves for the WFPC2 filters?

A: The throughput curves are available electronically at the WFPC2 Filter Throughput Directory. These can also be created with the 'calcband' task in the 'stsdas.synphot' package. For example, to get the throughput curve for the F555W filter type:

    calcband 'wfpc2,f555w,a2d7' out.tab

The 'a2d7' parameter is used to specify a gain of 7 in the WFPC2 observing mode. Similarly, 'a2d15' can be used to specify a gain of 15. Please note that the throughput tables are coarsely sampled. To create a well sampled table, you will first need to create a wavelength table using the 'genwave' task in the 'stsdas.hst_calib.synphot' package. Set the 'minwave', and 'maxwave' parameters to the starting and ending wavelengths of the filter (this information can be found in the WFPC2 Instrument Handbook), then set the 'dwave' parameter to a sampling interval that will produce 100-200 points over the wavelength range. Use the output table from 'genwave' in the 'wavetab' parameter of the 'calcband' task.

The result of 'calcband', out.tab in this case, is an STSDAS table which has two columns, WAVELENGTH and THROUGHPUT. STSDAS tables can be read and manipulated with the 'stsdas.toolbox.ttools' package. For more information on SYNPHOT, consult the Guide to WFPC2 SYNPHOT Tables.

Q: For the narrow band filters, 'calcphot' gives different FWHM and Pivot Wavelength values than those given in the Instrument Handbook. What causes the difference?

A: The main reason for this is that the default wavelength range that 'calcphot' uses undersamples the narrow band filters. 'calcphot' does not take into account the fact that the narrow band filters have red and blue leaks. The leaks do not contribute much to the overall throughput, but they greatly widen the wavelength range of the throughput. Therefore, the default wavelength range that 'calphot' uses includes only a few sample points in the range of interest. The temporary solution to this problem is for the user to create a wavelength set to ensure good coverage of the narrow filters. As described in the previous question, the 'genwave' task can be used to create a sampling interval that will produce ~100-200 points over the wavelength range. The output table from 'genwave' should then be used in the 'wavetab' parameter of 'calcphot'.

Q: How do I calculate a magnitude of an object in my WFPC2 field?

A: Basically, you need to determine a zeropoint (ZEROPT) for the passband you are interested in, using one of the three methods outlined below, and then use the normal formula:

     m = -2.5 x log10(COUNTS) + ZEROPT

where COUNTS is the number of counts from your image measured in the same way as your ZEROPT (e.g., the amount of light within a 0.5" radius; either per second or for the total exposure time, etc.). You can easily convert to per total exposure time by adding 2.5 x log₁₀(EXPTIME) to your zeropoint.

In practice, a number of "corrections" may need to be considered depending on the level of accuracy you require (e.g., CTE effect,contamination and red leaks for the UV filters, variable gains on different chips, color terms, geometric distortions). These are discussed in detail in the Holtzman et al. article under Method #2 below.

METHOD #1 - Use ground based aperture photometry of objects in your field to establish the magnitude zeropoint. Several software systems are available for doing aperture photometry (e.g., the APPHOT package in IRAF). This approach has the advantage that the measurement is made under exactly the same conditions as your program objects (e.g., same temperature, amount of contamination, gain, etc.). Disadvantages are that accurate photometry might not be available for objects in your field, background subtraction is required, and effects such as CTE and geometric distortion still need to be considered.

METHOD #2 - Determine the zeropoint using the PASP photometry paper by Holtzman et al. Here are some highlights:

• Conversions to the Landolt UBVRI (i.e., Johnson UBV plus Cousins RI) photometric system are provided.
• When gain = 7 (the most commonly used mode), the zeropoints are different for each of the four chips.
• The time dependence of the UV throughput is determined.
• A nice photometric calibration cookbook section is provided.

Another useful reference is the WFPC2 IDT's data reduction summary.This is similar to how the STScI pipeline works, but not identical. Of particular interest for people doing photometry are instructions for making a pixel area correction when performing integrated photometry (as opposed to surface photometry).

See also Photometry with the WFPC2 (PostScript, 500 Kbytes), by Whitmore, detailing the various methods for doing photometry with the WFPC2.

METHOD # 3 - Use the photometry information supplied in the header of the calibrated science data file (.c0h). The equation for calculating a WFPC2 zeropoint is:

     ZEROPT = -2.5 x log10(PHOTFLAM) + PHOTZPT

where PHOTZPT, and PHOTFLAM are header keywords.

A potential problems with this approach is that the resulting magnitude is in the STMAG system (see p. 15 of the SYNPHOT User's Guide), which is different than conventional photometric systems. A table for making the conversion from STMAG to the Johnson UBVRI and Cousins RI photometric systems is provided in the WFPC2 Photometry Cookbook.

SYNPHOT commands needed to make the conversion for other bandpasses are also included in the cookbook.

Preliminary checks indicate that the zeropoints obtained from Method #2 and Method #3 are in fair agreement, with typical differences of about 0.05 mag. In the future, SYNPHOT will be updated with new filter and response tables which should reduce these differences.