Since the advent of the accelerated expanding homogeneous universe model, some other explanations for the supernova Ia dimming have been explored, among which there are inhomogeneous models constructed with exact $\Lambda = 0$ solutions of Einstein's equations. They have been used either as one patch or to build Swiss-cheese models. The most studied ones have been the Lema\^itre-Tolman-Bondi (LTB) models. However, these models being spatially spherical, they are not well designed to reproduce the large scale structures which exhibit clusters, filaments and non spherical voids. This is the reason why Szekeres models, which are devoid of any symmetry, have recently come into play. In this paper, we give the equations and an algorithm to compute the redshift-drift for the most general quasi-spherical Szekeres (QSS) models with no dark energy. We apply it to a QSS model recently proposed by Bolejko and Sussman (BSQSS model) who averaged their model to reproduce the density distribution of the Alexander and collaborators' LTB model which is able to fit a large set of cosmological data without dark energy. They concluded that their model represents a significant improvement over the observed cosmic structure description by spherical LTB models. We show here that this QSS model is ruled out by a negative cosmological redshift, i.e. a blueshift, which is not observed in the Universe. We also compute a positive redshift and the redshift-drift for the Alexander et al.'s model and compare this redshift-drift to that of the $\Lambda$CDM model. We conclude that the process of averaging an unphysical QSS model can lead to obtain a physical model able to reproduce our observed local Universe with no dark energy need and that the redshift-drift can discriminate between this model and the $\Lambda$CDM model. For completeness, we also compute the blueshift-drift of the BSQSS model.

Exact inhomogeneous solutions of Einstein's equations have been used in the literature to build models reproducing the cosmological data without dark energy. However, owing to the degrees of freedom pertaining to these models, it is necessary to get rid of the degeneracy often exhibited by the problem of distinguishing between them and accelerating universe models. We give an overview of redshift drift in inhomogeneous cosmologies, and explain how it serves to this purpose. One class of models which fits the data is the Szekeres Swiss-cheese class where non-spherically symmetric voids exhibit a typical size of about 400 Mpc. We present our calculation of the redshift drift in this model, and compare it with the results obtained by other authors for alternate scenarios.

Inhomogeneous exact solutions of General Relativity with zero cosmological constant have been used in the literature to challenge the \Lambda CDM model. From one patch Lema\^itre-Tolman-Bondi (LTB) models to axially symmetric quasi-spherical Szekeres (QSS) Swiss-cheese models, some of them are able to reproduce to a good accuracy the cosmological data. It has been shown in the literature that a zero-\Lambda -LTB model with a central observer can be fully determined by two data sets. We demonstrate that an axially symmetric zero-\Lambda -QSS model with an observer located at the origin can be fully reconstructed from three data sets, number counts, luminosity distance and redshift drift. This is a first step towards a future demonstration involving five data sets and the most general Szekeres model.

Models of inhomogeneous universes constructed with exact solutions of Einstein's General Relativity have been proposed in the literature with the aim of reproducing the cosmological data without any need for a dark energy component. Besides large scale inhomogeneity models spherically symmetric around the observer, Swiss-cheese models have also been studied. Among them, Swiss-cheeses where the inhomogeneous patches are modeled by different particular Szekeres solutions have been used for reproducing the apparent dimming of the type Ia supernovae (SNIa). However, the problem of fitting such models to the SNIa data is completely degenerate and we need other constraints to fully characterize them. One of the tests which is known to be able to discriminate between different cosmological models is the redshift-drift. This drift has already been calculated by different authors for Lema\^itre-Tolman-Bondi (LTB) models. We compute it here for one particular axially symmetric quasi-spherical Szekeres (QSS) Swiss-cheese which has previously been shown to reproduce to a good accuracy the SNIa data, and we compare the results to the drift in the $\Lambda$CDM model and in some LTB models that can be found in the literature. We show that it is a good discriminator between them. Then, we discuss our model's remaining degrees of freedom and propose a recipe to fully constrain them.

During the past fifteen years, inhomogeneous cosmological models have been put forward to explain the observed dimming of the SNIa luminosity without resorting to dark energy. The simplest models are the spherically symmetric Lema\^itre-Tolman (LT) solutions with a central observer. Their use must be considered as a mere first step towards more sophisticated models. Spherical symmetry is but a mathematical simplification and one must consider spherical symmetric models as exhibiting an energy density smoothed out over angles around us. However, they have been taken at face value by some authors who tried to use them for either irrelevant purposes or to put them to the test as if they were robust models of our Universe. We wish to clarify how these models must be used in cosmology. We first use the results obtained by Iguchi and collaborators to derive the density profiles of the pure growing and decaying mode LT models. We then discuss the relevance of the different test proposals in the light of the interpretation given above. We show that decaying-mode (parabolic) LT models always exhibit an overdensity near their centre and growing-mode (elliptic or hyperbolic) LT models, a void. This is at variance with some statements in the literature. We dismiss all previous proposals merely designed to test the spherical symmetry of the LT models, and we agree that the value of $H_0$ and the measurement of the redshift drift are valid tests of the models. However, we suspect that this last test, which is the best in principle, will be more complicated to implement than usually claimed.

The current standard model of cosmology, the LambdaCDM model, is based on the homogeneous FLRW solutions of the Einstein equations to which some perturbations are added to account for the CMB features and structure formation at large scales. This model fits rather well the observations provided 95% of the energy density budget of the Universe should be of an unknown physical nature, i.e. dark matter and dark energy. Now, the aim of a cosmological model is not merely to reproduce the observations, but also to give a physical understanding of the Universe we live in. Moreover, even if the assumption of homogeneity seems to be more or less valid at large scales, it appears to be in contradiction with observations at intermediate scales (between the scale of non linear structure formation and that where structures virialize). This is the reason why, during the last decade, a community of researchers formed whose aim was to look for the best way to take into account the influence of the inhomogeneities seen in the Universe and to construct accurate cosmological models which could possibly get rid of the dark components. This task, which is still in its infancy, is currently progressing towards promising results. Two types of methods can be found in the literature: spatial averaging of scalar quantities and use of exact inhomogeneous solutions of General Relativity. We will give here a brief report of the second one.

Recently, inhomogeneous generalisations of the Friedmann-Lemaitre-Robertson-Walker cosmological models have gained interest in the astrophysical community and are more often employed to study cosmological phenomena. However, in many papers the inhomogeneous cosmological models are treated as an alternative to the FLRW models. In fact, they are not an alternative, but an exact perturbation of the latter, and are gradually becoming a necessity in modern cosmology. The assumption of homogeneity is just a first approximation introduced to simplify equations. So far this assumption is commonly believed to have worked well, but future and more precise observations will not be properly analysed unless inhomogeneities are taken into account. This paper reviews recent developments in the field and shows the importance of an inhomogeneous framework in the analysis of cosmological observations.

We use different particular classes of axially symmetric Szekeres Swiss-cheese models for the study of the apparent dimming of the supernovae of type Ia. We compare the results with those obtained in the corresponding Lemaitre-Tolman Swiss-cheese models. Although the quantitative picture is different the qualitative results are comparable, i.e, one cannot fully explain the dimming of the supernovae using small scale ~50 Mpc inhomogeneities. To fit successfully the data we need structures of order of 500 Mpc size or larger. However, this result might be an artifact due to the use of axial light rays in axially symmetric models. Anyhow, this work is a first step in trying to use Szekeres Swiss-cheese models in cosmology and it will be followed by the study of more physical models with still less symmetry.

Lema\^itre - Tolman (L--T) toy models with a central observer have been used to study the effect of large scale inhomogeneities on the SN Ia dimming. Claims that a giant void is mandatory to explain away dark energy in this framework are currently dominating. Our aim is to show that L-T models exist that reproduce a few features of the $\Lambda$CDM model, but do not contain the giant cosmic void. We propose to use two sets of data - the angular diameter distance together with the redshift-space mass-density and the angular diameter distance together with the expansion rate - both defined on the past null cone as functions of the redshift. We assume that these functions are of the same form as in the $\Lambda$CDM model. Using the Mustapha-Hellaby-Ellis algorithm, we numerically transform these initial data into the usual two L-T arbitrary functions and solve the evolution equation to calculate the mass distribution in spacetime. For both models, we find that the current density profile does not exhibit a giant void, but rather a giant hump. However, this hump is not directly observable, since it is in a spacelike relation to a present observer. The alleged existence of the giant void was a consequence of the L-T models used earlier because their generality was limited a priori by needless simplifying assumptions, like, for example, the bang-time function being constant. Instead, one can feed any mass distribution or expansion rate history on the past light cone as initial data to the L-T evolution equation. When a fully general L-T metric is used, the giant void is not implied.

It is commonly stated that we have entered the era of precision cosmology in which a number of important observations have reached a degree of precision, and a level of agreement with theory, that is comparable with many Earth-based physics experiments. One of the consequences is the need to examine at what point our usual, well-worn assumption of homogeneity associated to the use of perturbation theory begins to compromise the accuracy of our models. It is now a widely accepted fact that the effect of the inhomogeneities observed in the Universe cannot be ignored when one wants to construct an accurate cosmological model. Well-established physics can explain several of the observed phenomena without introducing highly speculative elements, like dark matter, dark energy, exponential expansion at densities never attained in any experiment (i.e. inflation), and the like. Two main classes of methods are currently used to deal with these issues. Averaging, sometimes linked to fitting procedures a la Stoegger and Ellis, provide us with one promising way of solving the problem. Another approach is the use of exact inhomogeneous solutions of General Relativity. This will be developed here.

Standard cosmology is based on the assumption that the universe is spatially homogeneous. However the consensus on a homogeneous matter structure, even on very large scales, has never been complete. The advantage of correlation dimension calculations is that they enable one to obtain the transition scale to a homogeneous distribution, whereas other methods, such as those using the two-point correlation function, make it hard to exhibit the possible fractal properties of the Universe matter content. Our purpose is to calculate the correlation dimension $D_2$, looking for a possible transition to homogeneity, which would imply $D_2 = 3$. We apply the correlation integral method to the three dimensional sample composed of 332,876 galaxies which we extract from the Fifth Data Release of the Sloan Digital Sky Survey. We analyze the raw data up to the scale $d_{max} = 160$ Mpc, assuming $H_0 = 70$ km s$^{-1}$ Mpc$^{-1}$ and considering three cosmological models in order to test the model dependence of our method. Using volume limited samples for this range leaves us with about 20,000 galaxies. Applying our method to random maps helps us to calibrate our results. We obtain a correlation dimension of the galaxy distribution which seems to increase with scales up to $D_2=3$ reached around 70 Mpc. The results of our analysis, performed on the largest volume limited sample which can be extracted from the SDSS catalog, are compatible with those formerly obtained by other authors. However, to get a more reliable description of the structures at various scales, we think it will be mandatory to use still larger catalogs than those which are currently available.

The magnitude-redshift relation is one of the tools for a direct observational approach to cosmology. The discovery of high redshift Type Ia supernovae (SNIa) and their use as ``standard candles'' has resurrected interest in this approach. Recently collected data have been used to address the problem of measuring the cosmological parameters of the universe. Analysed in the framework of homogeneous models, they have yielded, as a primary result, a strictly positive cosmological constant. However, a straight reading of the published measurements, conducted with no a priori idea of which model would best describe our universe at least up to redshifts $z\sim 1$, does not exclude the possibility of ruling out the Cosmological Principle - and cosmological constant - hypotheses. It is therefore shown here how the large scale homogeneity of this part of the universe can be tested on our past light cone, using the magnitude-redshift relation, provided sufficiently accurate data from sources at redshifts approaching $z=1$ would be available. An exemple of an inhomogeneous model with zero cosmological constant reproducing the current observations is given. The presently published SNIa data can thus be interpreted as implying either a strictly positive cosmological constant in a homogeneous universe or large scale inhomogeneity with no constraint on $\Lambda$. An increase in the number and measurement accuracy of the candidate ``standard candles'' at very high redshift is therefore urgently needed, for progress in both fundamental issues of the Cosmological Principle and cosmological constant.

Observational cosmology provides us with a large number of high precision data which are used to derive models trying to reproduce ``on the mean'' our observable patch of the Universe. Most of these attempts are achieved in the framework of a Friedmann-Lema\^itre cosmology where large scale homogeneity is assumed. However, we know, from the observation of structures at increasing scales, that these models are only approximations of a smoothed or averaged inhomogeneous underlying patern. Anyhow, when modelling the Universe, the usual method is to use continuous functions representing the kinematical scalars of the velocity field, implicitly assuming that they represent volume averages of the corresponding fine-scale inhomogeneous quantities, then put them into the Einstein equations which are solved to give the model and its dependance upon a number of parameters arbitrarily defined. In General Relativity, such a method is very much involved since the equations which determine the metric tensor and the quantities calculated from it are highly nonlinear. The question raised by the method consisting of determining the parameters of an a priori assumed FLRW model from observational data is the ``fitting problem'' brought to general attention by Ellis and Stoeger in the 80's. This problem has recently experienced a reniewed attention due to the amount of available data and the increase of the minimum scale at which homogeneity can be assumed. We propose a discussion of this issue in the light of the latest developments of observational and theoretical cosmology.

Since its decovery during the late 90's, the dimming of distant SN Ia apparent luminosity has been mostly ascribed to the influence of a mysterious dark energy component. Formulated in a Friedmannian cosmological modelling framework based upon the cosmological ``principle'' hypothesis, this interpretation has given rise to the ``concordance'' model. However, a caveat of such a reasoning is that the cosmological ``principle'' derives from a philosophical Copernican assumption and has never been tested. Furthermore, a weakness of its conclusion, i. e., the existence of a negative-pressure fluid or a cosmological constant, is that it would have profound implications for the current theories of physics. This is why we have proposed a more conservative explanation, ascribing the departure of the observed universe from an Einstein-de Sitter model to the influence of inhomogeneities. This idea has been independently developed by other authors and further enlarged to the reproduction of different cosmological data. We review here the main proposals which have been put forward to deal with this purpose and present some prospects for future developments.

Since its decovery during the late 90's, the dimming of distant SN Ia apparent luminosity has been mostly ascribed to the influence of a mysterious dark energy component. Based upon the cosmological "principle" hypothesis, this interpretation has given rise to the "concordance" model, developed in the context of a Friedmann-Lemaitre cosmology. However, a caveat of this reasoning is that the cosmological "principle" derives from a philosophical Copernican assumption and has never been tested. Furthermore, a weakness of its conclusion, i. e., the existence of a negative-pressure fluid or a cosmological constant, is that it would have profound implications for the current theories of physics. This is why we have proposed a more conservative explanation, ascribing the departure of the observed universe from an Einstein-de Sitter model to the influence of inhomogeneities. This idea has been independently developed by other authors and further enlarged to the reproduction of different cosmological data. We review here the main proposals which has been made along these lines of though and present some prospects for future developments.

The effect of curvature on the results of fractal analyses of the galaxy distribution is investigated. We show that, if the universe satisfies the criteria of a wide class of parabolic homogeneous models, the observers measuring the fractal index with the integrated conditional density procedure may use the Hubble formula, without having to allow for curvature, out to distances of 600 Mpc, and possibly far beyond. This contradicts a previous claim by Ribeiro (1995) that, in the Einstein-de Sitter case, relativistic corrections should be taken into account at much smaller scales. We state for the class of cosmological models under study, and give grounds for conjecture for others, that the averaging procedure has a smoothing effect and that, therefore, the redshift-distance relation provides an upper limit to the relativistic corrections involved in such analyses.

This paper is the third of a series dedicated to the study of the Delayed Big-Bang (DBB) class of inhomogeneous cosmological models of Lema\^itre-Tolman-Bondi type. In the first work, it was shown that the geometrical properties of the DBB model are such that the horizon problem can be solved, without need for any inflationary phase, for an observer situated sufficiently near the symmetry center of the model to justify the ``centered earth'' approximation. In the second work, we studied, in a peculiar subclass of the DBB models, the extent to which the values of the dipole and quadrupole moments measured in the cosmic microwave background radiation (CMBR) temperature anisotropies can support a cosmological origin. This implies a relation between the location of the observer in the universe and the model parameter value: the farther the observer from the symmetry center, the closer our current universe to a local homogeneous pattern. However, in this case, the centered earth approximation is no longer valid and the results of the first work do not apply. We show here that the horizon problem can be solved, in the DBB model, also for an off-center observer, which improves the consistency of this model regarding the assumption of a CMBR large scale anisotropy cosmological origin.

Recently collected SNIa data have been used to address the problem of measuring the cosmological parameters of the universe. Analysed in the framework of homogeneous models, they have yielded, as a primary result, a strictly positive cosmological constant. However, a straight reading of the published measurements, conducted with no a priori idea of which model would best describe our universe, at least up to redshifts z=1, does not exclude the possibility of ruling out the Cosmological Principle - and cosmological constant - hypotheses. It is here shown how the large scale (in)homogeneity of this part of the universe can be tested on our past light cone, using the magnitude-redshift relation, provided sufficiently accurate data from sources at redshifts approaching z=1 would be available.

Analysed in the framework of homogeneous FLRW models, the magnitude-redshift data from high redshift supernovae yield, as a primary result, a strictly positive cosmological constant. Another reading of the currently published measurements does not exclude a possible ruling out of the Cosmological Principle and, thus, also, of the cosmological constant hypothesis. It is shown how shortly coming data can be used to settle this fondamental issue, pertaining to both cosmology and particle physics.

The hot Big-Bang standard model for the evolution of the universe, despite strong successes, lets unresolved a number of problems. One of its main drawbacks, known as the horizon problem, was until now thought to be only solvable by an inflationary scenario. Here is proposed a class of inhomogeneous models of universe, getting rid of some of the worst drawbacks of standard cosmology. The horizon problem is solved by means of an initial singularity of "delayed" type and without need for any inflationary phase. The flatness and cosmological constant problems disappear.