HST Data Handbook for WFPC2

5.6 Accuracy of WFPC2 Results

---

Table 5.9 summarizes the accuracy to be expected from WFPC2 observations in several areas. The numbers in the table should be used with care, and only after reading the relevant sections of this handbook and the documents referenced therein; they are presented in tabular form here for easy reference.

Table 5.9: Accuracy Expected in WFPC2 Observations

Procedure	Estimated Accuracy	Notes
Calibration (flatfielding, bias subtraction, dark correction)
Bias subtraction	0.1 DN rms	Unless bias jump is present
Dark subtraction	0.1 DN/hr rms	Error larger for warm pixels; absolute error uncertain because of dark glow
Flatfielding	<1% rms large scale	Visible, near UV
	0.3% rms small scale
	~10%	F160BW; however, significant noise reduction achieved with use of correction flats
Relative photometry
Residuals in CTE correction	< 3% for the majority (~90%) of cases
	up to 1-% for extreme cases (e.g., very low backgrounds)
Long vs. short anomaly (uncorrected)	< 5%	Magnitude errors <1% for well-exposed stars but may be larger for fainter stars. Some studies have failed to confirm the effect (see Long vs. Short Anomaly (non-linearity))
Aperture correction	4% rms focus dependence (1 pixel aperture)	Can (should) be determined from data
	<1% focus dependence (> 5 pixel)
	1-2% field dependence (1 pixel aperture)
Contamination correction	3% rms max (28 days after decon) (F160BW)
	1% rms max (28 days after decon) (filters bluer than F555W)
Background determination	0.1 DN/pixel (background > 10 DN/pixel)	May be difficult to exceed, regardless of image S/N
Pixel centering	< 1%
Absolute photometry
Sensitivity	< 2% rms for standard photometric filters	Red leaks are uncertain by ~10%
	2% rms for broad and intermediate filters in visible
	< 5% rms for narrow-band filters in visible
	2-8% rms for UV filters
Astrometry
Relative	0.005" rms (after geometric and 34th-row corrections)	Same chip
	0.1" (estimated)	Across chips
Absolute	1" rms (estimated)