3 Constructing observed galaxy SEDs

A major development in the last decade has been the advent of new observing facilities and large surveys at all wavelengths of the spectrum, enabling astronomers for the first time to observe the full SEDs of galaxies from the UV to the FIR, from the local universe out to redshifts beyond 6. While databases such as those referenced in Table 1 make it tempting to simply go ahead and fit full galaxy SEDs, it is important to pause for a moment and review which are the difficulties associated with the construction of a single SED. Indeed, the “true SED” of a galaxy as defined in the models considers the sum of all photons emitted from inside the volume defining the galaxy. To make the observed SED of a galaxy, however, this ”true SED” is then filtered through the spectral response curve of the instruments and is redistributed spatially over the point spread function (PSF). Additionally, the measurement process not only adds noise, but also makes it necessary to join data from different instruments.

The construction of multi-wavelength SEDS is a complex and rich subject and a dedicated review would be a welcome addition to the literature. In keeping with the scope of the present text we can here only give a very cursory treatment of the issue. As a starting point for further reading we suggest consulting some of the major multi-wavelength surveys and their overview articles provided in Table 1 below.

 3.1 Spectral response curve and resolution
 3.2 Spatial resolution, aperture bias and matching
 3.3 Examples of multi-wavelength datasets
  3.3.1 The Spitzer Local Volume Legacy - spatially resolved SEDs
  3.3.2 The Herschel ATLAS - unresolved SEDs
  3.3.3 The SWIRE templates
  3.3.4 Further examples