6 The spectral energy distribution

Figure 4 shows the continuum spectral energy distribution (SED) of Mrk 279 from the far-IR to the X-rays. The IRAS data points are the average of 6 pointed observations reported by Edelson & Malkan (1987). The near-IR data are from Spinoglio et al. (1985) after subtraction of the stellar light (Sect. 3.2). The R and B band fluxes (Granato et al. 1993) and the mean 5100Å-flux (this paper) have also been corrected for the underlying galaxy contribution and de-reddened. The 1500 Å data point represents the average of 26 observations made with IUE (Rodriguez et al. 1998) between 1978 and 1991. It has been corrected for foreground reddening using $N_{\rm H} = 1.6~10^{20}$cm^-2 (Elvis et al. 1989). The large error bar reflects the strong variability of Mrk 279 at UV wavelengths. The EXOSAT X-ray data are from 1983 and 1984 and are best described in terms of a broken power-law (Ghosh & Soundararajaperumal 1992), while the 1994 data from ASCA are modeled with a unique power law in the Tartarus Database (http://tartarus.gsfc.nasa.gov/).

Since this SED is constructed from data collected over $\sim$19 years, we caution that it may be distorted by variability. Variability is important in shaping the X-ray, ultraviolet, and optical spectrum, but is much less significant at longer wavelengths (see Sect. 7). Bearing these limitations in mind, it is still possible to draw some general conclusions which are not affected by flux variations at short wavelengths.

The Mrk 279 SED displays three broad maxima or ``bumps''. The first maximum occurs in the far-IR at wavelengths $\geq$25$\mu$m. Given the large IRAS ($\sim$1') apertures and the cold color temperature of the far-IR bump, the 100$\mu$m and 60$\mu$m fluxes are probably dominated by cold dust from the host galaxy's ISM. Hence, the far-IR ``bump'' is most likely not related to the AGN itself. The second maximum is the usual ``big blue bump'' which dominates the SED of type 1 AGNs from the optical to the soft X-rays. It is usually identified as thermal emission from an accretion disk. In between these two maxima lies a third and smaller bump which extends from $\sim$1$\mu$m to $\sim$15-20$\mu$m. We tentatively identify this mid-IR bump as thermal emission from dust in the putative molecular torus and/or from dust in the NLR, as discussed below.