2.1.1. Observational Considerations

It is easy to obtain a high-quality spectrum of an H II region in a nearby galaxy. It is not so easy to obtain a high-quality analysis afterward. Photon statistics is not the entire story in CCD spectroscopy. Additional random errors creep in during the flat-fielding and photometric calibration stages. It is difficult to flat-field a CCD frame to better than 1% even in imaging observations, where the most precise flat-fields involve matching the color of the target to that of the flat-field source (the night-sky for deep imaging - see Tyson 1986). Spectroscopists rarely observe such practices. In typical H II region spectroscopy, the flat-field is often obtained by combining an internal lamp to map the pixel-to-pixel sensitivity variations with a twilight sky observation to fit the slit vignetting. Both fill the slit in a different way than the object, which is an important consideration for the correction of interference fringing in the red. Flat-fields repeatable to 1% precision can be obtained over limited areas of a CCD spectrum, but the precision can be worse over regions where the lamp source is weak (in the blue part of the spectrum for example) or vignetting is strong.

The photometric calibration also contributes to the uncertainty of the measured spectrum. Flux standard stars are typically measured at widely spaced wavelengths (50 Å is common), and the sensitivity function of the instrument is determined by fitting a low-order polynomial or spline to the flux points. Such fits inevitably introduce low-order "wiggles" in the sensitivity function, which will vary from star to star. Based on experience, the best spectrophotometric calibration yield uncertainties in the relative fluxes of order 2-3% for widely-spaced emission lines; the errors may be better for ratios of lines closer than 20 Å apart. Absolute fluxes have much higher uncertainties, of course, especially for narrow-aperture observations of extended objects.

Another source of concern is the extended nature of H II regions and patchiness in interstellar reddening, which affects the measured line ratios. H II region spectra are often presented as integrated over the source. Reddening by dust is patchy everywhere we look, so the effects on the H II region spectrum must vary from point-to-point if we look at the spatial distribution. Although the spectrum of an H II region may be dominated by the areas with the highest surface brightness, it may be possible for a bright but obscured area to be given low weight, or for a region with a lower-quality spectrum to have an inflated surface brightness because of poorly-measured extinction. Thus the patchy nature of dust reddening must introduce additional uncertainty into measured line ratios. Spatially-resolved measurements are encouraged whenever possible.

The highly opinionated point here is that anyone who presents measured emission line strengths with uncertainties of 1% or less is probably not adding in all the error sources. Five percent uncertainties are probably more realistic for the brightest emission lines observed. Note that this level of precision is more than adequate for abundance measurements for most astrophysical problems.