Title: Magnetic non-Gaussianity
Abstract: The standard cosmological scenario does not include magnetic fields. Since the universe before recombination was composed of a variety of
plasmas, this is a very suspect approximation; magnetic fields of some form must have existed.
With the precision of current and upcoming CMB experiments a lot of attention is being focused on the non-Gaussianity of the CMB, with the
hope of probing early universe physics (and specifically inflationary physics). To draw any conclusions about the inflationary physics it is
vital to accurately remove any contaminants. Since the energy density of a magnetic field is quadratic, the perturbations sourced by the
field are inherently non-Gaussian -- even if the field is Gaussian itself. A magnetic field is therefore an inevitable source of contamination of the primordial non-Gaussianity which must be subtracted before we all run to the papers claiming we know the inflaton in intimate detail.
Conversely, identifying a magnetic non-Gaussianity on the CMB provides an extremely powerful additional constraint on the form of an
early-universe magnetic field, over and above that provided by the standard approaches.
In this talk we consider the form of the intrinsic magnetic non-Gaussianity -- basically, the non-Gaussianity in its spatial distributions -- which is a prerequisite to knowing the CMB. We present a full set of rotationally-invariant correlations, only two of which have ever been mapped onto the CMB (and only three of which have before been shown), but all of which are needed for a full calculation. We then outline how these can be employed for parameter estimation, without absorbing the years of computer time that would otherwise be necessary.