I use correlations between cosmic microwave background data and optical wide-field surveys as probes of cosmology and astrophysics, mostly focusing on the Sunyaev-Zel'dovich (SZ) effect. This signal, which was theoretically predicted in the 1970s, is caused by the scattering of cosmic microwave background photons in the hot, ionized gas in groups or clusters of galaxies (see the carton on the left). Both the random motion of the electrons in the hot gas and the bulk velocity of the entire cluster contribute to the SZ effect. These two contributions are often called the thermal and kinematic SZ effect, respectively.

I am part of the Dark Energy Survey (DES) collaboration, and have also worked with data from the South Pole Telescope and the Planck and AKARI satellites. Furthermore, I work with the output of large cosmological simulations, such as the Magneticum simulation suite, and the SZ simulations from the Titan-Mira Universe.

Please see below for a more detailed description of the research projects I have led. In addition to these main projects, I have also contributed to several other projects within the Dark Energy Survey collaboration. A full list of my publications is available from ADS , the arXiv or HEP-INSPIRE .

A new way of probing AGN feedback?

Active galactic nuclei (AGN), powered by accretion onto a supermassive black hole, are amongst the most powerful sources of radiation in the Universe. The energy they deposit into their surroundings has important consequences for galaxy formation, and even affects structure formation on scales that are relevant for measuring cosmological parameters using weak galaxy lensing. Accurate modelling of the energetics of AGN feedback is also a crucial ingredient of modern cosmological simulations.

Recently, a few studies have claimed that the signature of AGN feedback can also be detected via the thermal Sunyaev-Zel'dovich (tSZ) effect - the scattering of cosmic microwave background photons on hot, ionized gas. Using quasars (a subclass of AGN) detected in the Sloan Digital Sky Survey and all-sky cosmic microwave background and far-infrared maps from the Planck and AKARI satellites, we have investigated this signal in detail. Thanks to the large spectral range of our maps (covering almost a factor of 50 in frequency), we were able to show that a significant part of this signal is in fact caused by dust emission.

To back up our estimates from the observational data, we have also compared our results to high-resolution hydrodynamical zoom-in simulations of AGN host haloes. Crucially, the simulations were run twice, once with and once without AGN feedback. The simulations show that AGN feedback affects the distribution of the hot gas, but it has only a relatively mild impact on the total thermal energy of the simulated objects. The latter quantity is also what our observations effectively measure, and the results from the observational data agree very well with the simulations.

The figures are taken from our recent paper: B. Soergel et al. (2017), Monthly Notices of the Royal Astronomical Society, 468, 1577.

Detecting the kinematic Sunyaev-Zel'dovich effect

The kinematic Sunyaev-Zel'dovich (kSZ) effect is a secondary anisotropy in the cosmic microwave background. As CMB photons scatter on the electrons in the ionized gas in clusters of galaxies, the bulk motion of the cluster imparts a Doppler shift on the photons. Therefore, the kSZ signal allows us to measure the velocities of clusters of galaxies and therefore to study the formation of structures in the Universe. This, in turn, can be used as a probe for dark energy or modified gravity. However, it has only become possible to measure the kSZ signal with the latest generation of cosmic microwave background data and optical wide-field surveys.

Using data from the Dark Energy Survey (DES) and the South Pole Telescope (SPT), my collaborators and I have detected this effect in a statistical way by exploiting the pairwise motion of galaxy clusters (see cartoon on the left). On average, relatively close pairs of clusters will fall towards each other because of their mutual gravitational attraction. At large separations they move almost independently, so the pairwise velocity is zero.

For our measurement we used galaxy clusters identified in the first year of DES data. Because DES is a photometric and not a spectroscopic survey, we could, however, only estimate the redshift (i.e. the distance to the cluster) with a significant uncertainty. This meant that the signal was almost completely erased for pairs that were closer together than this typical distance uncertainty. We were, however, able to incorporate this effect into our modelling of the signal.

The second crucial ingredient for this measurement was CMB temperature data from the SPT-SZ survey. To separate the kSZ signal from fluctuations on larger scales, we filtered the data with a matched spatial filter that optimally extracts the signal of galaxy clusters. Putting all of these ingredients together, we were able to obtain the first detection of the kSZ effect from a photometric cluster sample.

The figure on the left is taken from our paper: B. Soergel, K. Story, S. Flender, et al. (The DES and SPT collaborations), 2016, Monthly Notices of the Royal Astronomical Society, 461, 3172 .

Constraining perturbations in dark energy

The accelerating expansion of the Universe is arguably the greatest puzzle of modern cosmology. While the predictions provided by the standard model including Einstein's cosmological constant agree remarkably well with a variety of different observations, there are well-founded theoretical objections against a cosmological constant. Most notably, the expectation for its value from quantum field theory is many orders of magnitude larger than its measured value. Such a large deviation from a "natural" value would mean that our Universe would have to be fine-tuned with incredible precision.

To remedy this situation, a lot of alternative explanations for cosmic acceleration have been proposed. Broadly, they fall into two categories: dark energy (an additional field driving the acceleration), or modified gravity (a modification of the field equations of Einstein's general relativity). On the other hand, most attempts to constrain these models from observational data use purely phenomenological parametrisations, such as a smooth dark energy with a constant equation of state (ratio of pressure to density). While these simple parametrisations are certainly useful, they do not account for perturbations in this component, although the latter should clearly be present from a fundamental-physics perspective.

To aid with the mapping between fundamental theories and observables, I have implemented a generic parametrisation for the evolution of perturbations in dark energy or modified gravity into the numerical codes CAMB and CosmoMC. This allowed us to compute predictions for cosmological observables in models which included the aforementioned perturbations.

Most notably, the details of the evolution of perturbations affects the strength of the integrated Sachs-Wolfe effect (see cartoon on the left) and the lensing of the cosmic microwave background (CMB) by intervening foreground structures. Using observations of these effects from the Planck satellite and various galaxy surveys, we were able to constrain the theoretical parameters using a Markov Chain Monte Carlo sampler. In a nutshell, we found that the impact of the perturbations can be significant if there are deviations from the cosmological-constant paradigm.

The figure on the left is taken from our paper: B. Soergel et al., 2015, Journal of Cosmology and Astroparticle Physics, 1502 (2015) 037 . This paper originated from my MSc thesis research at LMU Munich, which was supervised by Jochen Weller, Tommaso Giannantonio, and Richard Battye (University of Manchester).