Research

I am a cosmologist, in other words I use physics to study the Universe, how it started and evolved into the structure that we observe around us. It is now an exceptional time for modern cosmology, when we can observe the Universe and connect high precision cosmological measurements with theory.  With the evolution of the universe spanning a vast range of energies and scales, cosmological observables can shed light on virtually any particle physics model as well as on any theory of gravity. There is in fact an interplay between particle physics and cosmology, and the deepest questions of these two fields of research are the same. My real passion and interest are in using the array of cosmological data available to us to test fundamental physics. Through the past years I have been involved both in theoretical and observational aspects of this endeavor, with a particular focus on the dynamics of the late-time universe and tests of gravity on cosmological scales. You can find a more detailed overview of the selected research projects in which I have been actively involved over the past years, including the links to the relevant publications.


Interested in working on these topics? Send me an email to inquire about openings in our group!

Overview of selected topics

Exploring the dark universe with Effective Field Theory and CAMB

with B. Hu, M. Raveri and N. Frusciante

We have performed a careful and comprehensive implementation of the effective field theory (EFT) of cosmic acceleration into CAMB.  The result, EFTCAMB, is a powerful tool which can be used to investigate the implications of the different EFT operators on linear perturbations, in an agnostic approach to tests of gravity, as well as to study perturbations in any specific single field dark energy or modified gravity model. This framework has several crucial virtues (e.g. it evolves the full dynamics of perturbations on all linear scales) and overcomes some of the limitations of previous approaches, offering a significantly extended applicability as well as a more accurate modeling which will be crucial in view of the percent level accuracy of future measurements. In a series of three papers we present the numerical implementation, illustrate its use to extract theoretical predictions for power spectra in a very broad range of models and put it at work with current data to derive new constraints on f(R) and pure EFT models in massive neutrinos cosmologies.

A principal component analysis for cosmological tests of gravity

with A. Hojjati, L. Pogosian, G.-B. Zhao, J. Zylberberg, K. Koyama, R. Crittenden

Ref.: Phys. Rev. D 89 (2014) 10, 103530, Phys. Rev. D 90 (2014) 4, 043513 , astro-ph.CO: 1410.5807

Ref.: Phys. Rev. Lett. 103, 241301 (2009), Phys. Rev. D 81, 103510 (2010), Phys. Rev. D 85, 043508 (2012)

Investigating the scale dependence of the dark sector

with  L. Pogosian, R. Buniy, A. Hojjati and G.-B. Zhao

Two generic functions of time and scale, μ(a,k) and γ(a,k), are enough to parametrize deviations from the standard cosmological model in the dynamics of linear scalar perturbations. However, an arbitrary k-dependence allows for unphysical possibilities. We argue that in local theories of gravity, in the quasi-static limit, these functions must be ratios of polynomials in k, with the numerator of μ being equal to the denominator of γ; moreover the polynomials are even and of second degree in all viable models of single (spin-0) field dark energy/modified gravity considered today. This means that, without significant loos of information, one can use data to constrain only five functions of a single variable (time). In a companion paper we investigate future constraints on these five functions, as well as detectability of this scale-dependent pattern, by means of a one dimensional principal component analysis which allows us to identify the observable physical modes of modified gravity.

Ref.: Phys. Rev. D 87, 104015 (2013), Phys. Rev. D 89 (2014) 8, 083505

Measuring the speed of cosmological GWs with CMB polarization

with M.Raveri, C. Baccigalupi, Y.-S. Zhou

Ref.: Phys. Rev. D 91 (2015) 6, 061501

Scalar radiation from Chameleon shielded regions

In general relativity the metric outside a compact source will be static no matter what the dynamics of the source is as long as it obeys spherical symmetry. This may not be the case any more in models of gravity where the Birkhoff’s theorem ceases to hold. I have investigated one of the consequences of the violation of the latter theorem, namely scalar radiation from chameleon shielded regions. Chameleons are scalar fields characterized by a profile that depends on the local matter density, as a consequence of their coupling to matter fields, and can screen themselves from local tests of gravity. The very same coupling allows a radially pulsating mass to directly propagate a disturbance into the space surrounding it with the radiation spectrum carrying characteristic imprints depending on whether the source is successfully screened or not. There are several interesting related scenarios; for instance, energy could be drained from binary black holes faster than in GR; the ripples in the Chameleon profile would induce time-variations of masses and fundamental coupling constants;  the hydrodynamics of compact objects, in particular the core collapse, could be significantly modified.

Ref.: Phys. Rev. Lett. 106, 251101 (2011)

  

  

 

The dynamics of linear scalar perturbations in generalized theories  of gravity can be parametrized in terms of two functions of time and scale, e.g. μ(a,k) and γ(a,k), that we can define modified growth (MG) functions. How to treat these free functions when performing forecasts or fits to data? We have investigated this in a series of papers where we have pioneered the application of the two-dimensional Principal Component Analysis (PCA) method to cosmological linear perturbations. As shown in the papers, the latter has proven very promising in performing model-independent forecasts for upcoming surveys as well as informative fits to currently available data, minimizing the assumptions on the time and scale dependence of μ and γ. It tells us which observables are more likely to be sensitive to the MG functions; or, inversely, given a survey (and therefore a set of observables), which features of MG will be better constrained, at which scales/times, etc. All this while taking into account degeneracies among the functions used to describe MG and cosmological parameters.

The growth of structure as mapped by weak lensing and galaxy clustering will certainly provide us with a rich testing ground for gravity. But are there more cosmological probes that we can use to  constrain modified gravity? We identified an interesting one in the dynamics of tensor modes. In general relativity gravitational waves propagate at the speed of light, however in alternative theories of gravity that might not be the case. We investigated the effects of a modified speed of gravity on the B-modes of the Cosmic Microwave Background (CMB) anisotropy in polarization, identifying the unique corresponding imprint: a shift in the angular scale of its peaks which allows to constrain c_T without any significant degeneracy with other cosmological parameters. We forecast the accuracy with which c_T will be measured by the next generation CMB satellites. We find future CMB satellites capable of constraining c_T^2 at percent level, comparable with bounds from binary pulsar measurements, largely due to the absence of degeneracy with other cosmological parameters.

Exploring consequences of modified growth with CAMB

with A.Hojjati, L.Pogosian, G.-B. Zhao, J. Zylberberg

We modified the publicly available Einstein-Boltzmann solver CAMB, to study the dynamics of linear scalar perturbations in general models of modified gravity and dark energy. The result is MGCAMB, a code based on the parametrization of deviations from the standard cosmological model in terms of two free functions of time and scale, μ(a,k) and γ(a,k). The latter provide a theoretically consistent and complete framework to evolve linear scalar perturbations in full generality. The code can be used in several ways; minimizing assumptions on the underlying theory and evolving the full dynamics of the system, one can extract predictions for observables for instance binning μ and γ in time and/or space, or choosing some motivated ansatzae for their time and/or scale dependence. Alternatively, reducing to the quasi-static regime (good on sub-horizon scales), the functions can be customized to parametrizations that are representative of specific classes of models. In the latter case, MGCAMB has been widely used by the community to explore constraints from current and future data on   f(R) models of gravity.

Ref.: Phys. Rev. D 79, 083513 (2009)