Gravitational lensing of the CMB

CMB photons are gravitationally deflected on their way to us, smearing the primordial peak structure by mixing scales. The deflection happens coherently across tens of arcminutes to degrees (the peak of the line-of-sight-integrated matter distribution), also imprinting characteristic non-Gaussian structure. CMB lensing is most effective halfway from the last scattering surface, and hence traces cosmological evolution to earlier times than e.g., galaxy weak lensing can. Together with the large sky areas covered by ongoing and upcoming experiments, this offers a unique window into the intermediate redshift Universe, promising unprecedented insights into the geometry and growth of structure in the universe. I am currently working on the lensing analysis of the 2500 square degrees of SPT data, which should deliver a ~25-30 sigma lensing measurement and the first interesting constraint (maybe a detection) of the neutrino mass scale. Novel telescopes built to detect the unique features imprinted by lensing into the CMB polarization fluctuations, such as PolarBear (see below) simultaneously will provide novel insights into inflationary physics. 

Epoch of Cosmic Reionization

The Epoch of Reionization (EoR) is a pivotal stage of cosmological structure formation, marking the birth of the first objects massive enough to yield significant numbers of ionizing photons. We are currently beginning to constrain the evolution of the ionized hydrogen fraction as a function of time, entering an exciting new era of astrophysics. 

Modeling reionization is challenging, as it requires simultaneous resolution of the small galaxies likely responsible for the bulk of ionizing photons, as well a covering hundreds of millions of lightyears in order to capture the typical extends of ionized regions percolating through the cosmic web.

I have developed highly efficient algorithms connecting the ionization morphology to the properties of sources and sinks, that have found wide usage in modeling various reionization observables.

One probe of reionization that I’m currently excited about is small scale CMB data generated by the South Pole Telescope. We recently published the power spectrum analysis of the combined 2008&2009 data sets, placing the strongest constraints to date on the microwave and infrared backgrounds on arcminute scales at mm wavelengths. We were able to use this data set to constrain, for the first time, the epoch of reionization via the small scale kinetic SZ (Doppler) effect from ionized bubbles embedded into large scale velocity flows. Our data indicate that reionization was less extended than what predicted by most models. Under most conservative assumptions about systematic contaminants, the constraint on the ionization fraction shown in the green contours (68/95% CL) is obtained.  Most excitingly, this is just the beginning of this powerful reionization probe. We are now beginning to interpret the combination of the full SPT survey (>3 times more data than analyzed so far), Herschel SPIRE followup of our deepest 100 square degrees. A conservative forecast for the sensitivity of this data set combination (also folding in a Planck forecast for the constraint on the integrated optical depth), is shown in the blue contours.

Inflation/High Energy Physics

One of our main goals with PolarBear will be to gain novel insights into inflationary physics. A stochastic background of tensor modes is one of the most robust predictions of inflation.  A tensor signal should be observable in CMB polarization by PolarBear if the tensor-to-scalar ratio is greater than 0.01 (compare the model curve in right panel in the Figure below). Interestingly, this level of gravity waves seems to be tied to Planck-scale physics, hence finding r to lie either higher or lower will have important consequences on studies of the particle/field nature of inflation, grand unified theory scale physics, string theory, and quantum gravity. In slow-roll theories of inflation, r > 0.01 corresponds to super-Planckian evolution of the inflaton field. Measuring r > 0.01 would be evidence for UV-complete treatments such as string theory realizations of large-field inflation (e.g., axion monodromy). On the other hand, a detection of r< 0.01 would support effective field theory descriptions.  As a core analysis member of the PolarBear collaboration I am interested in optimizing our observation strategy as well as statistical techniques to eventually detect the gravity wave signature.

More accurate observations of large scale structure and in particular galaxy clusters require more accurate models of the co-evolution of matter and gas in the universe. I’m currently studying the interaction of dark matter and gas with the largest hydrodynamical simulations ever produced, focusing on predictions for the CMB secondary anisotropies (an animation from one of my simulations is shown on the right). An enormous dynamic range is necessary to reliably model this problem, as massive clusters contributing to the thermal SZ effect (inverse Compton-scattering of CMB off hot electrons) are rare and velocity st photons reams sourcing the kinetic SZ (Doppler scattering of CMB photons into the line of sight) extend to scales as large as a billion light years, and at the same time very small structures down to group scale of 10^12 solar masses contribute.

Redshifted 21 cm radiation

I have also been involved in promoting redshifted 21 cm radiation as a potentially revolutionary probe of the epoch of reionization and “dark ages” of the universe. With a rest-frame frequency of 1.4 GHz, the hyperfine transition of neutral hydrogen at these redshifts can be observed at radio wavelengths today. Three-dimensional tomographical surveys will be able to trace the ionization morphology as a function of angular position and time, hence allowing inferences on the sources and sinks of reionization. While foregrounds are expected to be orders of magnitude brighter than the signal at these frequencies, they are also thought to be smooth functions of frequency, in contrast to the rapidly varying reionization signal (which will be zero inside an ionized region and modulated by fluctuations in the baryonic density along outside). The first pathfinder experiments including MWA, PAPER, LOFAR, and the GMRT have begun observations and are getting closer to a first detection. A future experiment, the SKA, should yield unprecedented constraints on the EoR. In addition to reionization studies, as a three-dimensional probe of approximately half the Hubble volume, redshifted 21 cm also has the potential to revolutionize other cosmological fields by sampling orders of magnitude more modes than the CMB or even futuristic galaxy surveys. As I have predicted in my papers with collaborators, this should yield improvements of inflationary, dark energy, and astroparticle physics constraints.

Galaxy clusters and large scale structure