Unlike stars and galaxies, quasars and AGN are multi-wavelength emitters. As a result obtaining a complete picture of an AGN is a challenging prospect requiring observations with a wide variety of telescopes. Over the past two decades, our multi-wavelength view of quasars and AGN has expanded significantly thanks to the continuing increases in sensitivity (Sanders et al. 1989, Elvis et al. 1994, Haas et al. 1998, Polletta et al. 2000, Haas et al. 2003). The variety of the resulting Spectral Energy Distributions (SEDs) grows with our parameter space (Kuraszkiewicz et al. 2003) and, while the contributing emission and absorption mechanisms are well accepted, their relative importance, particularly as a function of AGN class, remains a subject of debate.
Also hotly debated are the importance of orientation and absorption in determining the SED, the relations between the various classes and the details of the Unification picture. While surveys at many wavelengths can efficiently find AGN, these surveys provide different views of the AGN population, always selecting those brightest in a particular waveband. It can be argued that some wavebands provide a less biased view than others, (e.g. X-rays are less affected by absorption, far-IR is independent of orientation). However it is only by combining surveys in different wavelength regions that we can gain a view of the intrinsic population. Only then can we hope to answer the many open questions that remain.
With the advent of new, sensitive observing facilities at many wavelengths (X-RAY: Chandra, XMM-Newton, IR: 2MASS, SIRTF OPTICAL: 8-m telescopes, SDSS), multi-wavelength observations are now possible for a significant fraction of the AGN population. Deeper, multi-wavelength surveys are finding possible new varieties of AGN which raise the fundamental question: "What is an AGN?". Examples include the numerous, low-redshift red AGN found by 2MASS (Cutri et al. 2001) and otherwise uninteresting, X-ray loud galaxies visible with Chandra (Brandt et al. 2001). Whether these new sources are truly AGN, how they relate to "traditional" AGN and how large a fraction of the population they constitute, are major open questions which will be addressed via multi-wavelength follow-up.
Although AGN can be found in many wavebands, definitive classification is challenging without optical/IR spectroscopic data. This is particularly true given the lack of correspondence between the traditional optical class and other characteristics in increasing subsets of the population (e.g. IR/optical emission lines and X-ray flux (Genzel & Cesarsky 2000), or optical class and X-ray absorption (Wilkes et al. 2002)). The SDSS is providing an unprecedentedly large sample to relatively faint optical flux limits, the means to classify AGN with a wide range of SEDs. Similarly, SIRTF, successfully launched in August 2003, will fill the last major gap in our multi-wavelength picture of AGN reaching beyond the few bright and/or nearby sources, to the bulk of the population in the mid- and far-IR for the first time.
With this unprecedented combination of powerful, multi-wavelength observatories and the many planned and in progress surveys being carried out (GOODS: Giavalisco et al. 2003, ChaMP: Kim et al. 2003a, SWIRE: Lonsdale et al. 2003), we are poised to take great strides in our understanding of the intrinsic population of AGN, and their structure and evolution, as well as the larger question of the importance of accretion to the energy budget of the universe.
In this article, I review the observational components of AGN SEDs along with the physical structure and emission mechanisms believed to contribute in the various wavebands. The shape and variety of the SEDs, colors, flux ratios and other properties are discussed as a function of AGN properties and in comparison with models. I conclude with the prospects for answering the question: "What is an AGN?" and thus of viewing the intrinsic population and constraining AGN models.