WFPC2 Calibration FAQs

Q: How can I determine if I need to recalibrate my data?

A: Occasionally, improved reference files will become available after your data were calibrated with the Pipeline software. In these cases, you may want to consider recalibrating your data. There are two ways to find out what the recommended reference files are for your datasets. One way is to use the WFPC2 Calibration Files search screen on STARVIEW, which lists the reference files used in the pipeline calibration of your dataset and the recommended files for your dataset. This is a quick way to find out if there are any new files that you should consider. Note that when data are retrieved from the archive, OTFR (On The Fly Reprocessing) will automatically use the best available reference files to calibrate the data. Another way to check your calibrations is to look at the Reference Files memo which gives more detailed descriptions of all available reference files. All reference files can be retrieved with STARVIEW.

Q: Is there a way to find out about any changes to WFPC2 that might affect the calibration or analysis of my data?

A: Yes, the Chronological History of WFPC2 memo provides a brief history of decontaminations, darks and superdarks, focus changes, temperature changes, changes to the pipeline, safings, and other miscellaneous items.

Q: What do the numbers mean in the "data quality" files (.q0h, .c1h)?

A: Each non-zero number corresponds to a defect in that pixel. To find out what each number means, look at table 3.4 of the HST Data Handbook, V4.0. This table is on pg. WFPC2:3-17. It explains each of the values that you find in the WFPC2 data quality files. Note that the values are additive. For instance, if a pixel has a calibration file defect *and* a permanent camera defect, then the pixel will have a value of 6 (ie; 2+4) in the data quality file.

Q: How can I tell if my data is saturated ?

A: Look at the data quality file (c1h); bit 4 will be set if the image was saturated, for a value of 8 (+ other powers of 2 if there are other problems). Alternatively, the d0h file will have an uncalibrated value of 4095 for a saturated pixel. The value of 3790 DN is approximate, since the *calibrated* value of a saturated pixel will vary depending on the bias level, the dark current subtraction (esp. hot pixels), and the inverse flat field; so this value should be seen as a "danger threshold", and in fact, to be safe, it should be set around 3500 or so. I have seen pixels with count values below 3600 that were in fact saturated (4095 - 350 bias = 3745; multiply by flat field value of 0.95 => 3558 DN calibrated).