There are several compelling reasons for observing nearby galaxies in the ultraviolet (UV). First of all, massive, young stars emit most of their energy in this part spectrum and at least in star-forming galaxies they outshine the emission from any other stage of the evolution of a composite stellar population (e.g. Bruzual & Charlot 2003). Therefore, the flux emitted in the UV in spiral and irregular is an excellent measure of the current Star Formation Rate (SFR; Kennicutt 1998; Donas et al. 1987). In the case of quiescent elliptical galaxies the analysis of the UV upturn (the rising part of the FUV spectrum of these galaxies) promises to provide fundamental clues in our understanding of the evolution of low-mass stars on the horizontal branch. Due to its remarkable sensitivity to the physical properties of these stars, the UV upturn could be used, once fully understood, as a powerful diagnostic of old stellar populations (Burstein et al. 1988; O'Connell 1999; Yi et al. 1999; Brown 2004; Rich et al. 2005, 2006, in prep.; Boselli et al. 2005). The UV has also revealed the presence of residual star formation in a non-negligible fraction of low-redshift elliptical galaxies (Yi et al. 2005).
Second, the light emitted in the UV can be very efficiently absorbed by dust and then re-emitted at far-infrared (FIR) wavelengths. Therefore an analysis of the energy budget using a comparison of the infrared and UV emission is a powerful tool to determine the dust attenuation of light at all wavelengths (see Buat et al. 2005 and references therein). In this sense, it is worth emphasizing that dust attenuation is the most vexing problem that one has to face when analyzing the observational properties of composite stellar populations and galaxies.
Finally, the observation of nearby galaxies in the UV is fundamental if we are to understand the evolution of galaxies from the high-redshift Universe (where their properties are commonly derived from rest-frame UV observations) to the present.
There have been many attempts in the past to address some of these issues. Sullivan et al. (2000, 2001, 2004) studied the star formation histories in a relatively large and complete sample of UV-selected local galaxies, from which Treyer et al. (1998) derived the SFR density of the local Universe. The nature of the UV upturn in elliptical galaxies has been widely studied by several groups, including O'Connell (1999), Brown et al. (2000), Deharveng, Boselli, & Donas (2002). Studies on the dust attenuation in galaxies based on either photometric or spectroscopic UV studies are numerous, including Calzetti et al. (1994), Heckmann et al. (1995), Meurer et al. (1995, 1999), Buat & Xu (1996), Gordon et al. (2000, 2003), Buat et al. (2002), Roussel et al. (2005). The analysis of the UV morphology of nearby galaxies as a local benchmark for studies in the optical at high redshift have been also carried out by several authors, including Kuchinski et al. (2000, 2001), Marcum et al. (2001), Windhorst et al. (2002), Lauger, Burgarella, & Buat (2005).
However, the results of some of these studies were not conclusive mainly due to the small size of the samples used, which were not representative of the overall population of galaxies in the local Universe. This is particularly true for studies on the dust attenuation in star-forming galaxies and on the rest-frame UV morphology in nearby galaxies. In the case of the UV-upturn studies in early-type galaxies this limitation adds to the lack of spatial resolution and depth of previous UV data and, in some cases, to the availability of UV data in only one band, which leads to a loss of sensitivity to the strength of the UV upturn, best traced by the FUV-NUV color (see Gil de Paz et al. 2005 and references therein).
The availability of deep UV observations with moderately-good spatial resolution for large numbers of well-known nearby galaxies is now possible thanks to the launch of the Galaxy Evolution Explorer (GALEX) on April 28th 2003. The compilation of GALEX UV data carried out as part of this paper will allow us (and other researchers making future use of this dataset) to provide fundamental clues for solving some of the still many open questions regarding the UV properties of galaxies in the local Universe. In particular, we will show how the strength of the UV upturn is function of the stellar mass of the galaxy, with more massive elliptical galaxies showing stronger UV upturns. We will also demonstrate that in a sample like ours, which adequately represents the bulk of the galaxy population in the local Universe, the slope of the UV continuum is well-correlated (although with a significant dispersion) with the infrared-to-UV ratio and, therefore, with the UV extinction, and that the (FUV-K) color provides and excellent segregation between early-type (ellipticals and lenticulars) and late-type (spirals and irregulars) galaxies.
In this "The GALEX Ultraviolet Atlas of Nearby Galaxies" we present surface photometry in the two GALEX ultraviolet (FUV & NUV) bands, providing integrated photometry and structural parameters for a total of 1034 nearby galaxies, including extensively-studied objects like M 31, M 32, M 33, M 51, M 81, M 82, M 83, M 87, M 101, etc. We compare the UV properties of this sample with corollary data in the optical, NIR, and FIR, available for the majority of the galaxies in the Atlas. This comparison allows us to obtain insight into fundamental correlations such as the `red sequence' found in the color-magnitude diagram of ellipticals and lenticulars, and a better definition of the IRX- relation in normal star-forming galaxies.
In Section 2 we extensively describe the sample of galaxies. Section 3 provides a summary of the GALEX observations. The analysis and results are given in Sections 4 & 5, respectively. The conclusions are summarized in Section 6.