Astron. Astrophys. 335, 421-430 (1998)

3. The data

Fig. 1 shows a few examples of LSB galaxy HII region spectra. The resulting data (uncorrected for foreground Galactic extinction which is in most cases less than 0.05 mag in B) are presented in Tables 1 and 2.

Fig. 1. Some examples of typical spectra of LSB galaxy HII regions.

Table 1 contains data on the hydrogen Balmer emission lines. Column 1 gives the name of the galaxy. Column 2 contains the HII region identification. The first number refers to the night the spectrum was taken; the second number is an arbitrary ranking number based on the position of the HII region along the slit.

Table 1. Balmer line data

Column 3 contains the H flux in units of erg cm^-2 s^-1. The RMS error in the flux in the same units is given in Column 4. Columns 5, 6 and 7 contain the equivalent widths in Angstrom of the three brightest Balmer lines. The low continuum levels do however make the determination of equivalent widths uncertain. In general these could be determined only to an accuracy of 5 Angstrom at best. Any correction for absorption by the underlying stellar population ( Angstrom for normal galaxies [McCall et al. 1985]) will thus be negligible compared to the uncertainties in the widths themselves.

Column 8 contains the reddening coefficient c, as defined in e.g. Osterbrock (1989). c is related to by . Column 9 contains the error in c. c was determined from the ratio between the H and H fluxes and by assuming that the HII regions were Case B HII regions with a temperature of 10,000 K (see Osterbrock 1989). We will show in Sect. 4.1 that this is a justified assumption. The interstellar extinction curve by Savage & Mathis (1979) was used. The calibration uncertainty of 10% mentioned above results in an uncertainty of 0.2 in the reddening coefficient. In a few cases negative reddenings were found. Modestly negative reddenings can usually be explained by assuming a higher temperature for the nebula combined with zero reddening. A few very negative reddenings could not be explained satisfactorily. These are noted with colons in Table 1. The spectra with negative reddenings generally have low H and H fluxes thus introducing additional uncertainties. Negative reddenings have been set to zero in any further analysis.

The errors in the strong lines are shot noise limited, while the errors in the faint lines are dominated by the read-out noise of the CCD. An estimate of the uncertainty in the flux of the H lines was made by adding in quadrature the shot noise in the line, the shot noise in the continuum and the read-out noise of the CCD, which is erg s^-1 cm^-2 for the red images, and erg s^-1 cm^-2 for the blue ones. To ascertain that these were realistic estimates of the uncertainties, we also determined the uncertainties in the H flux by making two additional measurements of the line-flux, with the continuum levels systematically offset by and of the continuum. Both uncertainties agreed within a factor of two with each other, where for all but the strongest lines the shot noise value was slightly smaller than the offset-continuum value. The shot noise values are given in Column 4 of Table 1.

Table 2 contains the ratios of the fluxes of the other bright emission lines with respect to the H flux. If additional faint lines were measured in the spectrum the HII region number is marked with a star. These extra line fluxes are given in the lower panel of Table 2.

Table 2. Line ratios

In a few spectra the H line was not detected. Fluxes of the lines that were detected in these spectra are given in Table 3. Also given in Table 3 is a (of the continuum) upper limit of the H flux. All flux values are expressed in units of erg s^-1 cm^-2.

Table 3. Miscellaneous fluxes

© European Southern Observatory (ESO) 1998

Online publication: June 18, 1998
helpdesk.link@springer.de