In this paper we present the first results from implementing two scalar-tensor modified gravity theories, the symmetron and the Hu-Sawicki f (R)-gravity model, into a hydrodynamic N-body code with dark matter particles and a baryonic ideal gas. The study is a continuation of previous work where the symmetron and f (R) have been successfully implemented in the RAMSES code, but for dark matter only. By running simulations, we show that the deviation from ΛCDM in these models for the gas density profiles are significantly lower than the dark matter equivalents. When it comes to the matter power-spectrum we find that hydrodynamic simulations agree very well with dark matter only simulations as long as we consider scales larger than k ∼ 0.5 h/Mpc. In general the effects of modified gravity on the baryonic gas is found to not always mirror the effects it has on the dark matter. The largest signature is found when considering temperature profiles. We find that the gas temperatures in the modified gravity model studied here show deviations, when compared to ΛCDM, that can be a factor of a few larger than the deviations found in density profiles and power spectra.

We propose that the mass-temperature relation of galaxy clusters is a prime candidate for testing gravity theories beyond Einstein's general relativity, for modified gravity models with universal coupling between matter and the scalar field. For non-universally coupled models we discover that the impact of modified gravity can remain hidden from the mass-temperature relation. Using cosmological simulations, we find that in modified gravity the mass-temperature relation varies significantly from the standard gravity prediction of M ∝ T 1.73 . To be specific, for symmetron models with a coupling factor of β = 1 we find a lower limit to the power law as M ∝ T 1.6 ; and for f(R) gravity we compute predictions based on the model parameters. We show that the mass-temperature relation, for screened modified gravities, is significantly different from that of standard gravity for the less massive and colder galaxy clusters, while being indistinguishable from Einstein's gravity at massive, hot galaxy clusters. We further investigate the mass-temperature relation for other mass estimates than the thermal mass estimate, and discover that the gas mass-temperature results show an even more significant deviations from Einstein's gravity than the thermal mass-temperature.Article number, page 1 of 9

We consider issues related to the conformal mapping between the Einstein and Jordan frames in f (R) cosmology. We consider the impact of the conformal transformation on the gauge of a perturbed system and show that unless the system is written in a restricted set of gauges the mapping could produce an inconsistent result in the target frame. Newtonian gauge lies within the restricted group but synchronous gauge does not. If this is not treated carefully it could in principle contaminate numerical calculations. I. INTRODUCTIONExtended gravity theories, where the Einstein-Hilbert Lagrangian density L EH = R is replaced by a more general function including terms of higher-order in derivatives of the metric (R 2 , R µν R µν , R αβµν R αβµν . . .) and couplings to new dynamical degrees of freedom, have long been of interest in relativity. The last few decades have seen increasing applications of these models to cosmology. See [1,2] and their references for an overview and further details on these models and their applications to cosmology.1 In the last decade attention has focused on exploiting extended gravity to model dark energy without the need to introduce exotic particle species. A model frequently employed in this context is f (R) gravity (see for example [4,5] for recent reviews), where the Einstein-Hilbert Lagrangian density is replaced with an arbitrary function of the Ricci scalar, L = f (R), while the matter couples minimally to the metric and follows its geodesics in free motion. This representation of an extended gravity model is known as the "Jordan frame".The action can be transformed to a variety of different forms. One of the most common transformations is into the so-called "Einstein frame" where the action is manipulated to isolate a Ricci scalar of a new metric. The residual terms can be interpreted as an effective scalar field to which matter is non-minimally coupled and deflected from the geodesics of the metric. The field equations are otherwise those of standard general relativity. We review the model in the Jordan and Einstein frames in §II.Transforming from the Jordan frame to the Einstein frame is extremely useful in the study of f (R) gravity. We employ the metric formulation, in which the action is varied with respect to the metric alone, and in which the field equations are fourth-order. In the Einstein frame, as in standard GR, the theory is second-order -a significant simplification. Aspects of the transformation have been controversial for some time (see for example [1,[6][7][8] and further references in §III). The discussion has centred upon the nature of the equivalence, and authors can be separated [1,6] into two camps: those who feel the equivalence is "physical" and observables can be calculated in either frame, and those who feel the equivalence is mathematical in nature and that observables should be calculated in a chosen "physical frame". We briefly discuss this issue in §III.Modern cosmology is the study of Friedmann-Lemaître-Robertson-Walker (FLRW) spacetimes, perturbed to...

We use cosmological hydrodynamical simulations to study the effect of screened modified gravity models on the mass estimates of galaxy clusters. In particular, we focus on two novel aspects: (i) we study modified gravity models in which baryons and dark matter are coupled with different strengths to the scalar field, and, (ii) we put the simulation results into the greater context of a general screened-modified gravity parametrization. We have compared the mass of clusters inferred via lensing versus the mass inferred via kinematical measurements as a probe of violations of the equivalence principle at Mpc scales. We find that estimates of cluster masses via X-ray observations is mainly sensitive to the coupling between the scalar degree of freedom and baryons -while the kinematical mass is mainly sensitive to the coupling to dark matter. Therefore, the relation between the two mass estimates is a probe of a possible non-universal coupling between the scalar field, the standard model fields, and dark matter. Finally, we used observational data of kinetic, thermal and lensing masses to place constraints on deviations from general relativity on cluster scales for a general parametrization of screened modified gravity theories which contains f (R) and Symmetron models. We find that while the kinematic mass can be used to place competitive constraints, using thermal measurements is challenging as a potential non-thermal contribution is degenerate with the imprint of modified gravity.

Aims. We study the effects of letting dark matter and gas in the Universe couple to the scalar field of the symmetron model, a modified gravity theory, with varying coupling strength. We also search for a way to distinguish between universal and non-universal couplings in observations. Methods. The research is performed utilising a series of hydrodynamic, cosmological N-Body simulations, studying the resulting power spectra and galaxy halo properties, such as density and temperature profiles. Results. In the cases of universal couplings, the deviations in the baryon fraction from ΛCDM are smaller than in the cases of nonuniversal couplings throughout the halos. The same is apparent in the power spectrum baryon bias, defined as the ratio of gas to dark matter power spectrum. Deviations of the density profiles and power spectra from the ΛCDM reference values can differ significantly between dark matter and gas because the dark matter deviations are mostly larger than the deviations in the gas.