next up previous
Next: Ground-based near-UV through near-IR Up: HST-UV spectroscopy: results and Previous: Wind absorption lines

Extinction Estimates and Intrinsic Luminosities

We have demonstrated in the previous sections that the UV light in these Seyfert 2 galaxies is dominated by the starburst component. Thus, the continuum flux distribution can be used to estimate the reddening. The fluxes at 1500 Å measured in the GHRS aperture are 3.311#1110-15 erg s-1 cm-2 Å-1, 4.711#1110-15 erg s-1 cm-2 Å-1 and 1.6511#1110-15 erg s-1 cm-2 Å-1 for NGC 7130, NGC 5135 and IC 3639, respectively. These fluxes represent the 38%, 49% and 15% of the IUE flux in the 1432-1532 Å band measured by Kinney et al (1993). Leitherer & Heckman (1995) have shown that the UV energy distribution arising from a starburst has a spectral index (9#9) in the range -2.6 to -2.2 (F 59#59 if the starburst is younger than 10 Myr (see also Meurer et al 1995,1997)). This spectral index is independent of the metallicity and IMF. Therefore, any deviation from the predicted spectral index can be attributed to reddening.

First, the spectra are corrected for Galactic extinction using the MW extinction law and the Galactic E(B-V). Then, the corrected spectra are further dereddened to correct for the internal extinction in the starburst. This is done using the empirical Calzetti et al (1994) extinction law, such that the slope of the corrected spectrum (log F60#60 vs log 1#1) matches the slope of the synthetic spectra that fit the wind absorption lines. Figures 14-16 show the spectra corrected by reddening and one of the best models that fits the SiIV line. These models also fit the absorption part of the profiles of CIV and NV well. Note also that there is good agrement between the synthetic photospheric absorption lines (SV, CIII and SiIII) and the observations. The resulting values of the reddening are listed in Table 4 and the extinction-corrected UV luminosities are given in Table 5. The typical total UV extinctions are 2 to 3 magnitudes, and the implied extinction-corrected bolometric luminosities are of-order 1010 L2#2. Note that these values pertain only to the light within the relatively small GHRS aperture. We have followed an analogous procedure to determine the extinction and extinction-corrected UV luminosities for the IUE spectra (taken through an aperture about 70 times larger in projected area). These values are also listed in Tables 4 and 5. The bolometric luminosities implied by the extinction-corrected IUE spectra are typically a factor of 3 to 10 larger than for the GHRS spectra, and are quite similar to the total IR luminosities measured by IRAS (see Table 1 and 5).

In the previous section, we have seen that the line profile synthesis technique has not provided a unique solution. Instead, we have obtained a few models which are compatible with the profile of the wind lines. To better constrain the solution, we can use the UV continuum luminosity at 1500 Å. With this luminosity, we can predict the number of ionizing photons, Q, for each model, and the mass of the burst. The predicted ionizing photons can be compared with the values derived from the Balmer recombination lines. Table 5 gives the mass of the nuclear starburst and its predicted ionizing luminosity, based on the extinction-corrected UV luminosity (as measured through the GHRS aperture).


next up previous
Next: Ground-based near-UV through near-IR Up: HST-UV spectroscopy: results and Previous: Wind absorption lines
Rosa Gonzalez
1998-06-20