We study the evolution of linear perturbations in a Lema\^itre-Tolman-Bondi (LTB) void model with realistic cosmological initial conditions. Linear perturbation theory in LTB models is substantially more complicated than in standard Friedmann universes as the inhomogeneous background causes gauge-invariant perturbations to couple at first order. As shown by Clarkson et al. (2009), the evolution is constrained by a system of linear partial differential equations which need to be integrated numerically. We present a new numerical scheme using finite element methods to solve this equation system and generate scalar initial conditions based on Gaussian random fields with an underlying power spectrum for the Bardeen potential. After spherical harmonic decomposition, the initial fluctuations are propagated in time and estimates of angular power spectra of each gauge invariant variable are computed as functions of redshift. This allows to analyse the coupling strength in a statistical way. We find significant couplings up to $25\%$ for large and deep voids of Gpc scale as required to fit the distance redshift relations of SNe.

Based on techniques developed in the previous papers of this series, we investigate the impact of galaxy-cluster mergers on the order statistics of the largest Einstein radii. We show that the inclusion of mergers significantly shifts the extreme value distribution of the largest Einstein radius to higher values, typically increasing the expected value by ${\sim}10\%$. A comparison with current data reveals that the largest observed Einstein radius agrees excellently well with the theoretical predictions of the $\Lambda$CDM model at redshifts $z > 0.5$. At redshifts $z < 0.5$, our results are somewhat more controversial. Although cluster mergers also increase the expected values of the order statistics of the $n$ largest Einstein radii by ${\sim}10\%$, the theoretically expected values are notably lower (${\sim}3\sigma$ deviation for $n = 12$) than the largest Einstein radii of a selected sample of SDSS clusters in the redshift range $0.1 \leq z \leq 0.55$. The uncertainties of the observed Einstein radii are still large, however, and thus the measurements need to be carefully revised in future works. Therefore, given the premature state of current observational data, overall, there is still no reliable statistical evidence for observed Einstein radii to exceed the theoretical expectations of the standard cosmological model.

Do current observational data confirm the assumptions of the cosmological principle, or is there statistical evidence for deviations from spatial homogeneity on large scales? To address these questions, we developed a flexible framework based on spherically symmetric, but radially inhomogeneous Lemaitre-Tolman-Bondi (LTB) models with synchronous Big Bang. We expanded the (local) matter density profile in terms of flexible interpolation schemes and orthonormal polynomials. A Monte Carlo technique in combination with recent observational data was used to systematically vary the shape of these profiles. In the first part of this article, we reconsider giant LTB voids without dark energy to investigate whether extremely fine-tuned mass profiles can reconcile these models with current data. While the local Hubble rate and supernovae can easily be fitted without dark energy, however, model-independent constraints from the Planck 2013 data require an unrealistically low local Hubble rate, which is strongly inconsistent with the observed value; this result agrees well with previous studies. In the second part, we explain why it seems natural to extend our framework by a non-zero cosmological constant, which then allows us to perform general tests of the cosmological principle. Moreover, these extended models facilitate explorating whether fluctuations in the local matter density profile might potentially alleviate the tension between local and global measurements of the Hubble rate, as derived from Cepheid-calibrated type Ia supernovae and CMB experiments, respectively. We show that current data provide no evidence for deviations from spatial homogeneity on large scales. More accurate constraints are required to ultimately confirm the validity of the cosmological principle, however.

We discuss how Type Ia supernovae (SNe) strongly magnified by foreground galaxy clusters should be self-consistently treated when used in samples fitted for the cosmological parameters. While the cluster lens magnification of a SN can be well constrained from sets of multiple images of various background galaxies with measured redshifts, its value is typically dependent on the fiducial set of cosmological parameters used to construct the mass model to begin with. In such cases, one should not naively demagnify the observed SN luminosity by the model magnification into the expected Hubble diagram, which would then create a bias, but take into account the cosmological parameters a-priori chosen to construct the mass model. We quantify the effect and find that a systematic error of typically a few percent, up to a few-dozen percent, per magnified SN, may be propagated onto a cosmological parameter fit, unless the cosmology assumed for the mass model is taken into account (the bias can be even larger if the SN is lying very near the critical curves). We also simulate how such a bias propagates onto the cosmological parameter fit using the Union2.1 sample, supplemented with strongly magnified SNe. The resulting bias on the deduced cosmological parameters is generally at the few percent level, if only few biased SNe are included, and increasing with the number of lensed SNe and their redshift. Samples containing magnified Type Ia SNe, e.g. from ongoing cluster surveys, should readily account for this possible bias.

The Einstein radius (ER) of a gravitational lens encodes information about decisive quantities such as halo mass, concentration, triaxiality, and orientation with respect to the observer. Thus, the largest Einstein radii can potentially be utilised to test the predictions of the LCDM model. Hitherto, studies have focussed on the single largest observed ER. We extend those studies by employing order statistics to formulate exclusion criteria based on the n largest Einstein radii and apply these criteria to the strong lensing analysis of 12 MACS clusters at z>0.5. We obtain the order statistics of Einstein radii by a MC approach, based on the semi-analytic modelling of the halo population on the past lightcone. After sampling the order statistics, we fit a GEV distribution to the first-order distribution, which allows us to derive analytic relations for the order statistics of the Einstein radii. We find that the Einstein radii of the 12 MACS clusters are not in conflict with the LCDM expectations. Our exclusion criteria indicate that, in order to exhibit tension with the concordance model, one would need to observe approximately twenty Einstein radii >30", ten >35" or five >42" in the range of 0.5<z<1.0 on the full sky. Furthermore, we find that, with increasing order, the haloes with the largest Einstein radii are on average less aligned along the line-of-sight and less triaxial. In general, the cumulative distribution functions steepen for higher orders, giving them better constraining power. (abridged)

For more than a decade now, it has been controversial whether or not the high rate of giant gravitational arcs and the largest observed Einstein radii are consistent with the standard cosmological model. Recent studies indicate that mergers provide an efficient mechanism to substantially increase the strong-lensing efficiency of individual clusters. Based on purely semi-analytic methods, we investigated the statistical impact of cluster mergers on the distribution of the largest Einstein radii and the optical depth for giant gravitational arcs of selected cluster samples. Analysing representative all-sky realizations of clusters at redshifts z < 1 and assuming a constant source redshift of z_s = 2.0, we find that mergers increase the number of Einstein radii above 10 arcsec (20 arcsec) by ~ 35 % (~ 55 %). Exploiting the tight correlation between Einstein radii and lensing cross sections, we infer that the optical depth for giant gravitational arcs with a length-to-width ratio > 7.5 of those clusters with Einstein radii above 10 arcsec (20 arcsec) increases by ~ 45 % (85 %). Our findings suggest that cluster mergers significantly influence in particular the statistical lensing properties of the strongest gravitational lenses. We conclude that semi-analytic studies must inevitably take these events into account before questioning the standard cosmological model on the basis of the largest observed Einstein radii and the statistics of giant gravitational arcs.

Can the standard cosmological model be questioned on the basis of a single observed extreme galaxy cluster? Usually, the word extreme refers directly to cluster mass, which is not a direct observable and thus subject to substantial uncertainty. Hence, it is desirable to extend studies of extreme clusters to direct observables, such as the Einstein radius (ER). We aim to evaluate the occurrence probability of the large observed ER of MACS J0717.5 within the standard LCDM cosmology. In particular, we want to model the distribution function of the single largest ER in a given cosmological volume and to study which underlying assumptions and effects have the strongest impact on the results. We obtain this distribution by a Monte Carlo approach, based on the semi-analytic modelling of the halo population on the past lightcone. After sampling the distribution, we fit the results with the general extreme value (GEV) distribution which we use for the subsequent analysis. We find that the distribution of the maximum ER is particularly sensitive to the precise choice of the halo mass function, lens triaxiality, the inner slope of the halo density profile and the mass-concentration relation. Using the distributions so obtained,we study the occurrence probability of the large ER of MACS J0717.5, finding that this system is not in tension with LCDM. We also find that the GEV distribution can be used to fit very accurately the sampled distributions and that all of them can be described by a Frechet distribution. With a multitude of effects that strongly influence the distribution of the single largest ER, it is more than doubtful that the standard LCDM cosmology can be ruled out on the basis of a single observation. If, despite the large uncertainties in the underlying assumptions, one wanted to do so, a much larger ER (> 100 arcsec) than that of MACS J0717.5 would have to be observed.