This document is made available in accordance with publisher policies and may differ from the published version or from the version of record. If you wish to cite this item you are advised to consult the publisher's version. Please see the URL above for details on accessing the published version.

The large-scale structure of the universe can only be observed via luminous tracers of the dark matter. However, the clustering statistics of tracers are biased and depend on various properties, such as their host-halo mass and assembly history. On very large scales this tracer bias results in a constant offset in the clustering amplitude, known as linear bias. Towards smaller non-linear scales, this is no longer the case and tracer bias becomes a complicated function of scale and time. We focus on tracer bias centred on cosmic voids, depressions of the density field that spatially dominate the universe. We consider three types of tracers: galaxies, galaxy clusters and AGN, extracted from the hydrodynamical simulation Magneticum Pathfinder. In contrast to common clustering statistics that focus on auto-correlations of tracers, we find that void-tracer cross-correlations are successfully described by a linear-bias relation. The tracerdensity profile of voids can thus be related to their matter-density profile by a single number. We show that it coincides with the linear tracer bias extracted from the large-scale autocorrelation function and expectations from theory, if sufficiently large voids are considered. For smaller voids we observe a shift towards higher values. This has important consequences on cosmological parameter inference, as the problem of unknown tracer bias is alleviated up to a constant number. The smallest scales in existing datasets become accessible to simpler models, providing numerous modes of the density field that have been disregarded so far, but may help to further reduce statistical errors in constraining cosmology.

In this work we analyse the properties of cosmic voids in standard and coupled dark energy cosmologies. Using large numerical simulations, we investigate the effects produced by the dark energy coupling on three statistics: the filling factor, the size distribution and the stacked profiles of cosmic voids. We find that the bias of the tracers of the density field used to identify the voids strongly influences the properties of the void catalogues, and, consequently, the possibility of using the identified voids as a probe to distinguish coupled dark energy models from the standard ΛCDM cosmology. In fact, on one hand coupled dark energy models are characterised by an excess of large voids in the cold dark matter distribution as compared to the reference standard cosmology, due to their higher normalisation of linear perturbations at low redshifts. Specifically, these models present an excess of large voids with R eff > 20, 15, 12 h −1 Mpc , at z = 0, 0.55, 1, respectively. On the other hand, we do not find any significant difference in the properties of the voids detected in the distribution of collapsed dark matter halos. These results imply that the tracer bias has a significant impact on the possibility of using cosmic void catalogues to probe cosmology.