Cosmic Microwave Background
I started working on the Cosmic Microwave Background (CMB) with the WMAP satellite, launched in 2001. The CMB is the earliest image of the Universe that we can observe, the relevant physics is relatively simple, linear and well understood. The CMB offers an uniquely clean window into the linear universe (as it was 380000 years after the big bang) and a robust standard ruler to determine the overall Universe Geometry and its peresent age.
CMB observation were key in confirming the LCDM as the standard cosmological model.
The recently launched Planck satellite will be the "ultimate" experiemnt for the primary CMB signal. By primary here I mean as the Universe was at the last scattering surface. By ultimate here I mean cosmic variance dominated. The blessing and curse os cosmology is that we only have one visible Universe but we try to infer the properties of a much larger --"ensamble"- Universe of which our observable one is just a small patch. There is therefore a fundamental limitation to any cosmological error-bars that is how representative of the whole our observable patch is. This is called cosmic variance and it is somewhat a curse. This is however also a "blessing" and it means that in cosmology we can make "ultimate" experiments: once an experiment is cosmic variance dominated over the full (extra-galactic) sky , no repeated experiment can improve over it (unless one is looking at transients and thus opens the time-domain window).
Planck will also improve significantly CMB polarization measurements (see more below).
After the spectacular CMB results of the past 7 years or so, and in anticipation of Planck's results it is interesting to ask what's next?
One approach is to use the primary CMB as a backlight illumination of the foreground Universe. The imprint of the late-time universe on the CMB (the so-called secondary CMB effects) are a potentially powerful probe of the Universe's history and evolution. In principle it is possible to learn about the physics of reionization, the growth of cosmological structures, the evolution of dark energy, even neutrino properties or the nature of the initial conditions. In some of these cases complicated astrophysics come into play and it is still being investigated if that can be modeled well enough to extract the signal one is after.
The other approach is to measure, down to the cosmic variance limit if possible, the CMB polarization. The fact that CMB light should be polarized was predicted shortly after the observations of the CMB, but it was not until 2002 that it was measured (statistically) by the DASI experiment and until 2005 that WAMP produced the first full sky maps of the CMB polarization.
Polarization carries signal not only about the process of reionization but also about the physics of the early Universe and inflation.
For example, the characteristic large-scale anti-correlation between temperature and polarization as measured by WMAP in 2003 indicated the presence of super-horizon, adiabatic perturbations, as predicted by inflation.
More strikingly, CMB polarization can in principle enable us to see beyond the last scattering surface (at recombination, 380000 years after the big bang, the Universe became transparent to electromagnetic radiation, that's why we can see the CMB but not beyond it). In fact the universe is transparent to gravity waves: inflation is predicted to have generated a stochastic background of gravity waves. These are tensor-modes primordial perturbations that leave a characteristic imprint in the CMB polarization pattern. The next generation of --proposed-- CMB experiments (BPol, CMBPol, COrE) has the goal of measuring just that. Such signal, combined with the information on the CMB temperature anisotropies, can probe directly the energy scale of inflation and therefore the physics of the early Universe.
There are, however, challenges.
First, even in an ideal case, the signal is small and weak gravitational lensing of large-scale structures between us and the last scattering surface tend to swamp somewhat the primordial polarized signal.
More importantly, we live in a galaxy, which electromagnetic emission is polarized: this signal is well above the primordial CMB one.
Thus a thorough "cleaning" of all these spurious signals must be implemented before being able to go after the primordial one. This will be extremely challenging.
More about this can be found for example in the following two references
White paper in support of the BPol satellite proposal;
and the flollowing presentation.