Pith. sign in

REVIEW 4 major objections 5 minor 42 references

Joint cosmological data pin the effective fractal dimension of spacetime to d = 2.0004, leaving almost no room for fractional deviations from ΛCDM.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

T0 review · grok-4.5

2026-07-30 13:52 UTC pith:XRIG5HOG

load-bearing objection Competent null constraint on their prior EFF model: d pinned to 2 within ~10^{-4} once CMB priors enter, but the headline width rests on underspecified early-universe sound-horizon handling. the 4 major comments →

arxiv 2607.26084 v1 pith:XRIG5HOG submitted 2026-07-26 gr-qc astro-ph.CO

Observational Constraints on Emergent Fractional Fractal Cosmology

classification gr-qc astro-ph.CO
keywords emergent fractional fractal cosmologyeffective fractal dimensionobservational constraintsΛCDMDESI BAOCMB distance priorsgrowth rate fσ8model comparison
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper tests whether spacetime can carry a mild fractal structure that changes how matter and radiation thin out as the Universe expands. The authors take an emergent-cosmology model with one extra parameter, an effective fractal dimension d, and fit it to supernovae, Hubble-rate data, structure-growth measurements, DESI baryon acoustic oscillations, big-bang nucleosynthesis, and CMB distance priors. When the CMB information is included, d is forced to 2.0004 with uncertainties of a few parts in ten thousand—statistically indistinguishable from the ordinary non-fractal value d = 2. The model therefore remains observationally viable, but only by sitting extremely close to standard ΛCDM. A sympathetic reader cares because the result shows how tightly present data already police quantum-gravity-inspired modifications that would otherwise look attractive on theoretical grounds.

Core claim

With late-time probes plus DESI DR2 BAO and Planck CMB distance priors, the effective fractal dimension of the Emergent Fractional Fractal model is constrained to d = 2.0004^{+0.0006}_{-0.0003} at 1σ. AIC finds the model and ΛCDM statistically comparable, while BIC prefers the simpler ΛCDM model; any fractional deviation from standard cosmology is limited to O(10^{-4}).

What carries the argument

The effective fractal dimension d, which enters the modified continuity equation ρ̇_i = −(3d/2)(ρ_i + p_i)H and the fractional Friedmann equation; when d = 2 the equations collapse exactly to ΛCDM.

Load-bearing premise

The analysis re-uses CMB acoustic-peak distance priors and a standard big-bang-nucleosynthesis grid that were both computed assuming ordinary ΛCDM expansion, without recomputing them on the modified fractal background.

What would settle it

A full re-analysis that recompresses the CMB peaks and recomputes primordial helium on the EFF expansion history itself, then re-runs the same MCMC; if the new posterior on d moves several sigma away from 2, the present claim is overturned.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Any constant fractal correction to cosmic expansion is already ruled out above the 10^{-4} level once CMB information is used.
  • The Hubble-constant and σ_8 tensions are not relieved by the EFF extension; the model tracks ΛCDM once early-universe anchors are added.
  • Dark-matter density remains non-zero and consistent with standard values, so fractal geometry cannot replace cold dark matter.
  • Future work that allows a redshift-dependent d would have to confront the same tight late-time and CMB limits at each epoch.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Because the growth equation is also modified by d, next-generation weak-lensing and redshift-space-distortion surveys could push the bound below 10^{-5} without needing new CMB data.
  • The near-identity of EFF and ΛCDM once d is fixed near 2 suggests that other single-parameter fractal or fractional-calculus cosmologies will face the same compression once sound-horizon anchors are included.
  • A self-consistent early-universe pipeline for EFF would be the natural next numerical step; any residual shift in the sound horizon would translate almost linearly into a shift in the allowed window for d.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper performs an MCMC constraint of the Emergent Fractional Fractal (EFF) cosmology — a model in which the apparent horizon carries a fractal entropy S ∝ S_BH^{d/2}, leading to a modified Friedmann equation H^d ∝ Σρ and matter/radiation scalings ρ_r ∝ a^{-2d}, ρ_m ∝ a^{-3d/2}, reducing exactly to ΛCDM at d=2. Using three dataset combinations (Pantheon+SH0ES + H(z) + fσ8; plus DESI DR2 BAO and a PRIMAT-based BBN prior on Y_p; plus DESI DR2 BAO and Planck 2018 CMB distance priors R, ℓ_A, ω_b), the authors find late-time data allow d ≃ 2.004, while adding CMB distance priors collapses the posterior to d = 2.0004^{+0.0006}_{-0.0003}. AIC finds EFF and ΛCDM statistically comparable; BIC penalizes the extra parameter and favors ΛCDM. The MCMC setup (emcee, Gelman–Rubin, ESS diagnostics in Table 2) and likelihood construction are conventional and clearly reported, and the growth-ode and background implementation appear internally consistent. The central concern is that the 10^{-4} bound on d is generated by the CMB distance priors, while the manuscript does not document how the sound horizon r_s(z*) — which shifts at O(10^{-3}) under the EFF-modified pre-recombination scaling for the best-fit d−2 — is computed, and it applies ΛCDM-compressed priors to a non-ΛCDM model whose deviations grow with redshift.

Significance. If the analysis is internally consistent, the result usefully closes off a class of fractal/emergent-gravity cosmologies: deviations of the horizon fractal dimension from d=2 are limited to O(10^{-4}), and the model contributes no improvement in χ² (ΔAIC = 2.00 is exactly the parameter-count penalty). Strengths worth crediting: the likelihoods, covariances, priors, and convergence diagnostics (Gelman–Rubin, autocorrelation, ESS in Table 2) are reported with unusual completeness; the three staged dataset combinations cleanly show which probes drive the constraint; and the AIC/BIC comparison is honest — the paper does not oversell a null result. The work is incremental (constraining the authors' own earlier model) but provides a concrete, falsifiable bound that the community working on fractional/fractal cosmology will need to respect. However, the precision of the headline number is only as good as the early-universe treatment inside Eqs. (16)–(17), which is currently undocumented — this must be resolved before the result can be trusted at the quoted digits.

major comments (4)
  1. [Sec. 3.1, Eqs. (16)-(17)] The headline bound d = 2.0004^{+0.0006}_{-0.0003} is produced almost entirely by Eqs. (16)-(18), yet the manuscript never states how r_s(z*) or z* itself is computed. In the EFF model radiation scales as a^{-2d} and matter as a^{-3d/2} (Eq. 8), so at recombination (a~10^{-3}) the radiation density is shifted by ~exp[2(d-2)ln(10^3)]-1 ~ 5x10^{-3} for d-2 = 4x10^{-4}. Propagated through the sound-horizon integral this is an O(10^{-3}) effect on r_s, comparable to or larger than the precision of the priors (sigma(l_A)~0.09, i.e. 3x10^{-4} fractional) that sets the 10^{-4} posterior width on d. If a standard LCDM fitting formula for r_s(z*) was used while D_A is evaluated in the EFF background, the quoted bound is partly an artifact of inconsistent early-universe treatment. The modified redshift definition (a0/a)^{d/2}=1+z must also be applied consistently to z* and D_A. Sec. 3.3 says r_d is
  2. [Sec. 3.1, Eq. (15)] The (R, l_A, omega_b) values and covariance were compressed from the Planck 2018 baseline LCDM analysis. Applying them unchanged to EFF assumes the compression remains sufficient under a model whose deviations from LCDM grow toward early times -- precisely the regime the priors encode. This is a correctness risk, not a circularity claim: a concrete test is available. Ref. [32] (Chen, Huang & Wang 2019) itself provides distance priors and covariances for extended models beyond LCDM; the authors should either recompute the constraint using such an extended-model compression, or demonstrate numerically (e.g., by scanning d over the prior range) that the compressed likelihood faithfully reproduces the full CMB response for EFF backgrounds. Without this check, the O(10^{-4}) bound should be presented with an explicit robustness caveat.
  3. [Secs. 3.3-3.4, Tables 3-4] The OHD sample (Table 3) is described as cosmic chronometers plus 'radial BAO analyses', and it is combined in a single chi^2 with the DESI DR2 BAO likelihood (Eq. 31). Radial BAO H(z) points (e.g. the z=2.30 point, H=224+/-8, which is a Lyman-alpha BAO measurement) are not independent of the DESI DR2 D_H/r_d data at overlapping redshifts. The manuscript assumes all datasets are statistically independent (Eq. 12). The authors should either restrict the H(z) sample to chronometer-only points for the BAO combinations, or quantify the covariance/double-counting and show the posteriors are unaffected.
  4. [Secs. 3.2 and 4, Eqs. (23)-(25), Tables 5-6] In Tables 5-6, M_B moves from -19.33 (LT, H0~71) to -19.438 +/- 0.008 (LT+BAO+CMB, H0=67.3), tracking M_B ~ -19.253 + 5 log10(H0/73.04). This is the behavior expected when the supernova data constrain only the M_B-H0 combination, i.e., when the Cepheid calibrator likelihood (Eqs. 23-24) does not anchor M_B absolutely. In the Pantheon+SH0ES construction the 42 Cepheid calibrators are precisely what fixes M_B ~ -19.25; if they were implemented as described, M_B should not drift by 0.18 mag (many sigma given +/-0.008) to accommodate a Planck-like H0. Please verify the calibrator implementation (covariance, per-host anchoring) or clarify that the SN sample is effectively used uncalibrated. As it stands the Cepheid terms appear to contribute nothing, and the H0-tension discussion in Sec. 4 depends on this.
minor comments (5)
  1. [Sec. 3, Table 1] The prior d >= 2 (Table 1) truncates the parameter space at the LCDM limit, and the LT/BBN posteriors sit close to the boundary. The d constraint should also be quoted as a one-sided upper limit, and the effect of the boundary on the reported asymmetric intervals discussed.
  2. [Sec. 3.5] The manuscript acknowledges that Y_p is read from a standard PRIMAT grid without recomputing BBN on the EFF expansion history. Since the LT+BAO+BBN column of Table 5 quotes d = 2.0039 +/- 0.003, at which the expansion rate at BBN temperatures deviates from standard, a quantitative estimate of the induced shift in Y_p (or a caveat propagated to the table) is needed.
  3. [Sec. 3.6, Table 4] Table 4 lists two independent f sigma_8 entries at z = 0.38 and two at z = 0.60; please state whether these are distinct measurements and whether they are treated as independent in Eq. (14).
  4. [Sec. 4] Delta AIC = 2.00 for the CMB combination is exactly the parameter-count penalty: chi^2_min for EFF (1632.92053) is only 0.007 below LCDM (1632.92754). The text should state plainly that the extra parameter yields no fit improvement, rather than only 'fit equally well'.
  5. [Figs. 1-6] Figs. 3-6: axis labels and legends are very small; the lower 'relative difference' panels lack explicit labels identifying which dataset each curve belongs to. Fig. 1's Omega_r0 axis formatting ('x10^5' offset) is confusing. A public release of the likelihood code and chains would strengthen the paper.

Circularity Check

1 steps flagged

Standard observational constraints paper: d→2 is data-driven collapse to the model's built-in ΛCDM limit, not a circular identity; only mild non-load-bearing self-citation of the EFF definition.

specific steps
  1. self citation load bearing [Sec. 2 opening; Ref. [20]; Eqs. 1–10]
    "In this section, we provide a brief overview of the EFF model originally proposed in [20]. This framework is based on fractional quantum gravity... these equations reduce exactly to the standard Friedmann equations when d=2."

    The entire EFF setup (fractional WDW, fractal horizon entropy, modified continuity and Friedmann equations) is imported from a prior paper with overlapping authorship. That is normal for a constraints follow-up and is not load-bearing for the numerical bound on d—the bound comes from external likelihoods—but it is the only self-citation link in the chain, so it is recorded at minimal weight.

full rationale

The EFF framework is defined so that d=2 recovers standard Friedmann dynamics and density scalings (Eqs. 6–10, 8). That reduction is an honest model property, not a hidden identity used to manufacture the result. The paper’s actual claim is an external MCMC constraint: with LT+DESI DR2 BAO+CMB, d=2.0004^{+0.0006}_{-0.0003}, while LT alone still allows mild d>2 degeneracies with H0 and σ8. Likelihoods (Pantheon+SH0ES, H(z), fσ8, DESI BAO, Planck distance priors, BBN Yp) are independent datasets; AIC/BIC comparisons are ordinary model-selection arithmetic. The sole mild circularity-adjacent element is that the EFF construction and fractal entropy are taken from overlapping-author theory work [20], which enters only as the hypothesis under test, not as a uniqueness theorem or fitted answer. Methodological caveats (ΛCDM-compressed CMB priors; BBN grid not recomputed on EFF expansion) affect correctness risk, not derivation circularity. No fitted input is relabeled a prediction, and no step reduces Eq. X to Eq. Y by construction beyond the openly stated d=2 limit.

Axiom & Free-Parameter Ledger

2 free parameters · 5 axioms · 2 invented entities

The observational claim rests on the pre-existing EFF construction (fractional WDW → fractal BH entropy → Padmanabhan emergent-space balance → modified Friedmann and continuity equations) plus standard FLRW+perturbation assumptions and external compressed likelihoods. The only genuinely new degree of freedom fitted here is the constant fractal dimension d; other free parameters are the usual cosmological set. Invented content is inherited from the theory paper, not re-derived.

free parameters (2)
  • d (effective fractal dimension) = 2.0004^{+0.0006}_{-0.0003} (LT+BAO+CMB)
    Single extra model parameter, prior U(2,3), fitted by MCMC; central quantity whose posterior is the paper's main result.
  • Ω_b0, Ω_dm0, H0, σ8, M_B = e.g. H0≈67.32, Ω_dm0≈0.2722, σ8≈0.816 (LT+BAO+CMB)
    Standard cosmological and nuisance parameters varied jointly with d; values shift with dataset stack in the usual way.
axioms (5)
  • ad hoc to paper Apparent cosmological horizon carries the same fractal entropy scaling S ∝ S_BH^{d/2} as a fractional black-hole horizon, with 2≤d<3.
    Imported from Ref. [20] and Sec. 2; not independently derived or tested here; sets the whole modified dynamics.
  • domain assumption Padmanabhan emergent-space law dV/dt = L_P²(N_sur−N_bulk) remains valid after fractal redefinition of area/volume.
    Sec. 2, Eq. (1); standard within emergent-gravity literature but not established physics.
  • ad hoc to paper Modified continuity equation ρ̇_i = −(3d/2)(ρ_i+p_i)H and the associated growth ODE govern background and linear perturbations.
    Eqs. (8), (37); follows from the fractal scaling choice a∝(1+z)^{−2/d} built into the model.
  • domain assumption Planck 2018 ΛCDM-compressed CMB distance priors and DESI/Pantheon/OHD/fσ8 likelihoods are statistically adequate for EFF.
    Sec. 3; common practice, but compression assumes near-ΛCDM acoustics.
  • standard math Flat FLRW, standard radiation content N_eff≈3.04–3.046, Gaussian independent likelihoods.
    Implicit throughout Secs. 2–3; conventional cosmology toolkit.
invented entities (2)
  • Effective fractal dimension d of the cosmological apparent horizon no independent evidence
    purpose: Single parameter controlling fractional deviations of Friedmann, continuity, and growth equations from ΛCDM.
    Defined via d=2/α+1 from Lévy/Riesz fractional WDW and extended from black-hole horizons to cosmology in prior EFF work; this paper only constrains it.
  • Fractal effective horizon radius R_eff and fractal density parameters Ω^{(i,frac)}_0 no independent evidence
    purpose: Rewrite the modified Friedmann equation in observationally usable form (Eqs. 5, 9–10).
    Auxiliary constructs of the EFF framework; no separate empirical handle outside fitting d.

pith-pipeline@v1.2.0-grok45-kimik3 · 20751 in / 3582 out tokens · 76924 ms · 2026-07-30T13:52:02.948741+00:00 · methodology

0 comments
read the original abstract

We constrain the Emergent Fractional Fractal (EFF) cosmological model through a joint likelihood analysis of recent cosmological observations at the background and perturbation levels. In this framework, an effective fractal dimension $d$ is introduced to parameterize possible fractional deviations from the standard cosmological model. We consider three combinations of datasets: (i) late-time (LT) observations including PantheonPlus Type Ia supernovae, $H(z)$ measurements, and growth-rate measurements $f\sigma_8$; (ii) LT combined with DESI DR2 BAO and Big Bang nucleosynthesis (BBN); and (iii) LT combined with DESI DR2 BAO and CMB distance priors. With the inclusion of CMB distance priors, the fractal dimension is constrained to $d = 2.0004^{+0.0006}_{-0.0003}$ at the $1\sigma$ confidence level. Model comparison using the Akaike Information Criterion (AIC) shows that the EFF and $\Lambda$CDM models fit the observational data equally well, while the Bayesian Information Criterion (BIC) favors the simpler $\Lambda$CDM model because of its smaller parameter space. These results show that current cosmological observations place strong constraints on fractal extensions of the standard cosmological framework and possible deviations from the $\Lambda$CDM model.

Figures

Figures reproduced from arXiv: 2607.26084 by Hooman Moradpour, Raheleh Jalalzadeh, Shahram Jalalzadeh.

Figure 1
Figure 1. Figure 1: The posterior distributions and corresponding 1 [PITH_FULL_IMAGE:figures/full_fig_p014_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The posterior distributions and corresponding 1 [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Hubble parameter H(z) as a function of redshift for the ΛCDM (right panel) and EFF (left panel) models, compared with observational data points for the three dataset combinations. The lower panel shows the relative difference. In the [PITH_FULL_IMAGE:figures/full_fig_p017_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The deceleration parameter q(z) as a function of redshift for the ΛCDM (right panel) and EFF (left panel) cosmological models for the three dataset combinations. The lower panel shows the relative difference. The redshift evolution of the deceleration parameter q(z) is shown in [PITH_FULL_IMAGE:figures/full_fig_p018_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Distance modulus µ(z) as a function of redshift for ΛCDM (right panel) and EFF (left panel), compared to PantheonPlus data points for the three dataset combinations. The lower panel shows the relative difference [PITH_FULL_IMAGE:figures/full_fig_p019_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Evolution of fσ8(z) as a function of redshift for the EFF model (left panel) and the ΛCDM model (right panel), compared with observational measurements (black points). The lower panel shows the relative difference [PITH_FULL_IMAGE:figures/full_fig_p020_6.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

42 extracted references · 26 linked inside Pith

  1. [1]

    Reuter M 1998Phys. Rev. D57971–985 (Preprinthep-th/9605030)

  2. [2]

    Rovelli C 2008Living Rev. Rel.115

  3. [3]

    Zwiebach B 2006A first course in string theory(Cambridge University Press) ISBN 978-0-521- 83143-7, 978-0-511-20757-0

  4. [4]

    Jalalzadeh R, Jalalzadeh S and Heydarzade Y 2025Nucl. Phys. B1017116945 (Preprint 2504.04502)

  5. [5]

    Calcagni G 2010JHEP03120 (Preprint1001.0571)

  6. [6]

    Hawking S W 1976Phys. Rev. D142460–2473

  7. [7]

    Moniz P V and Jalalzadeh S 2020Mathematics8313 (Preprint2003.01070)

  8. [8]

    Jalalzadeh S 2022Phys. Lett. B829137058 (Preprint2203.09968)

  9. [9]

    Chen D M and Wang L 2024Universe10333 (Preprint2409.02954) Observational Constraints on Emergent Fractional Fractal Cosmology22

  10. [10]

    Jalalzadeh S and Vargas Moniz P 2022Challenging Routes in Quantum Cosmology(World Scientific) ISBN 978-981-4415-06-4

  11. [11]

    Jalalzadeh S, Costa E W O and Moniz P V 2022Phys. Rev. D105L121901 (Preprint2206.07818)

  12. [12]

    Calcagni G 2017Phys. Rev. D96046001 (Preprint1705.01619)

  13. [13]

    Costa E W d O, Jalalzadeh R, da Silva J´ unior P F, Rasouli S M M and Jalalzadeh S 2023Fractal and Fractional7854

  14. [14]

    Dark Univ.44101498 (Preprint2404.06986)

    Jalalzadeh R, Jalalzadeh S, Jahromi A S and Moradpour H 2024Phys. Dark Univ.44101498 (Preprint2404.06986)

  15. [15]

    Phys.87835–840

    El-Nabulsi A R 2013Indian J. Phys.87835–840

  16. [16]

    Rasouli S M M, Costa E W O, Moniz P V and Jalalzadeh S 2022Fractal Fract.6655 (Preprint 2210.00909)

  17. [17]

    Benetti F, Lapi A, Gandolfi G and Liberati S 2024Class. Quant. Grav.41175010 (Preprint 2407.16787)

  18. [18]

    Landim R G 2021Phys. Rev. D103083511 (Preprint2101.05072)

  19. [19]

    Giusti A 2020Phys. Rev. D101124029 (Preprint2002.07133)

  20. [20]

    da Silva J´ unior P, de Oliveira Costa E and Jalalzadeh S 2023Eur. Phys. J. Plus1381–16 (Preprint 2309.12478)

  21. [21]

    Padmanabhan T 2012 (Preprint1206.4916)

  22. [22]

    Cai R G 2012JHEP11016 (Preprint1207.0622)

  23. [23]

    Tu F Q and Chen Y X 2013JCAP05024 (Preprint1303.5813)

  24. [24]

    Hashemi M, Jalalzadeh S and Vasheghani Farahani S 2015Gen. Rel. Grav.4753 (Preprint 1308.2383)

  25. [25]

    Yuan F F and Huang Y C 2013 (Preprint1304.7949)

  26. [26]

    Moradpour H 2016Int. J. Theor. Phys.554176–4184 (Preprint1601.05014)

  27. [27]

    Chen G R 2022Eur. Phys. J. C82532

  28. [28]

    Abdul Karim M, Aguilar J, Ahlen S, Alam S, Allen L, Prieto C A, Alves O, Anand A, Andrade U, Armengaud Eet al.2025Phys. Rev. D112083515

  29. [29]

    J.938113 (Preprint2112.03863)

    Scolnic Det al.2022Astrophys. J.938113 (Preprint2112.03863)

  30. [30]

    Tamri Z, Aghamohammadi A, Golanbari T and Khodam-Mohammadi A 2026Eur. Phys. J. C86 96

  31. [31]

    Mohebi R, Saaidi K, Golanbari T and Karami K 2026JHEAp53100648 (Preprint2508.20129)

  32. [32]

    Chen L, Huang Q G and Wang K 2019JCAP02028 (Preprint1808.05724)

  33. [33]

    Rept7541–66

    Pitrou C, Coc A, Uzan J P and Vangioni E 2018Phys. Rept7541–66

  34. [34]

    Riesz M 1949Acta Mathematica811 – 222

  35. [35]

    High Energy Phys.20187612490 (Preprint1805.08566)

    Tarasov V E 2018Adv. High Energy Phys.20187612490 (Preprint1805.08566)

  36. [36]

    Laskin N 2002Phys. Rev. E66056108 (Preprintquant-ph/0206098)

  37. [37]

    Foreman-Mackey D, Hogg D W, Lang D and Goodman J 2013Publ. Astron. Soc. Pac.125306–312

  38. [38]

    Sci.7457–472

    Gelman A and Rubin D B 1992Statist. Sci.7457–472

  39. [39]

    Lewis A 2019arXiv e-prints(Preprint1910.13970)

  40. [40]

    Aver E, Berg D A, Hirschauer A S, Olive K A, Pogge R W, Rogers N S, Salzer J J and Skillman E D 2022Mon. Not. Roy. Astron. Soc.510373–382

  41. [41]

    Mehrabi A, Basilakos S, Malekjani M and Davari Z 2015Phys. Rev. D92123513 (Preprint 1510.03996)

  42. [42]

    Batista R C and Pace F 2013JCAP06044 (Preprint1303.0414)