Astron. Astrophys. 324, 32-40 (1997)

1. Introduction

The extinction correction is crucial for any study on galaxy emissions at UV and optical wavelengths. In particular in the UV band, substantial dust extinction occurs within the disks of spiral galaxies (Buat & Xu 1996, hereafter Paper II). At the same time there is still a controversy over the question whether galaxy disks are opaque or not in the optical band (e.g. Disney et al. 1989, Valentijn 1990, Huizinga & van Albada 1992, Boselli & Gavazzi 1994, Xu & Buat 1995, hereafter Paper I). Because it is very difficult to measure the extinction directly, it has been a common practice in the literature to estimate the extinction correction using indirect tracers such as atomic and total gas column density (e.g. Donas et al. 1987, Buat et al. 1989, Lisenfeld et al 1996), although apart from our Galaxy (Savage & Mathis 1979) the existence of a local correlation between the extinction and the gas column density has only been directly verified in the Magellanic Clouds from the observations of individual stars (LMC, Koornneef 1982; SMC, Bouchet et al. 1985), plus the evidence for such a correlation in the disk of M31 from a multi-wavelength study by Xu & Helou (1996).

It is the aim of this paper to scrutinize the validity of the widely used extinction correction method based on the gas column density. The tool we exploit is the model recently developed by two of us (Paper I and Paper II) which calculates the internal extinction for individual galaxies using a radiative transfer model with the far-infrared (FIR), UV and optical fluxes as input. The reliability of the model is demonstrated by the results for the internal extinction in M31 obtained by applying this radiative transfer model (Xu & Helou 1996), which agreed very well with the direct extinction measurements in that galaxy (Bajaja & Gergely 1977; Hodge & Lee 1988). In this paper we apply the model to a large sample of the galaxies with available UV, FIR, optical, HI and CO data. We will compare the resulted extinction at B-band, , with the gas column density estimated from the HI and CO data.

The paper is organized in the following way: this introduction (Sect. 1) is followed by a description of the sample (Sect. 2); The model and results on are presented in Sect. 3. Correlation studies on and gas column densities are carried out in Sect. 4. Sect. 5 is devoted to a discussion and to the conclusion.

© European Southern Observatory (ESO) 1997

Online publication: May 26, 1998

helpdesk.link@springer.de