Pith. sign in

REVIEW 3 major objections 4 minor 290 references

A null detection of the 21-cm forest already disfavours cold, largely neutral intergalactic gas at z≈5.6, including mean temperatures below about 27 K at 10% neutral fraction.

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 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 23:37 UTC pith:T3GJGOWS

load-bearing objection First real-data 21-cm forest power spectrum; null detection is solid, but the exclusion contours rest on model choices and unquantified proximity effects, so treat them as preliminary rather than decisive. the 3 major comments →

arxiv 2607.15341 v1 pith:T3GJGOWS submitted 2026-07-16 astro-ph.CO

Forest without Trees is still Fruitful: Constraints on the thermal state of the neutral IGM at zapprox5.6 with the 21-cm forest power spectrum

Tom\'a\v{s} \v{S}oltinsk\'y (INAF-OATs) , Arnab Chakraborty (McGill University) , Girish Kulkarni (TIFR) , Matteo Viel (SISSA) , Cathryn M. Trott (Curtin University) , Rashmi Sagar (IIT Indore) , Nithyanandan Thyagarajan (CSIRO) , James S. Bolton (University of Nottingham)
show 5 more authors
Benedetta Ciardi (MPA) Emma V. Ryan-Weber (Swinburne University of Technology) Soumak Maitra (TIFR) Abhirup Datta (IIT Indore) Nirupam Roy (IISc Bangalore)
This is my paper
classification astro-ph.CO
keywords 21-cm forestEpoch of Reionizationintergalactic mediumone-dimensional power spectrumneutral hydrogenquasar absorptionBayesian inferenceradio interferometry
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 tries to establish that the 21-cm forest — the narrow absorption features neutral hydrogen imprints on radio-loud quasar spectra before reionization ends — has become an observationally informative probe even when no individual absorption line is detected. Using archival uGMRT observations of the quasar J352-15 at z=5.82, the authors measure, for the first time, the one-dimensional power spectrum of 21-cm forest flux fluctuations. A Bayesian comparison of this spectrum against forward-modelled spectra from 534 cosmological simulations yields no statistical detection, but the null result nevertheless constrains the joint plane of mean neutral fraction ⟨x_HI⟩ and mean neutral-IGM temperature ⟨T_HI⟩. At the 68% credible level, models with ⟨T_HI⟩≲27 K at ⟨x_HI⟩=0.1 are disfavoured, along with the coldest, most neutral models that existing Lyα and 21-cm tomography data still allow. If the paper is right, current-generation radio telescopes can extract real astrophysical constraints from a non-detection — evidence that the neutral IGM has been substantially pre-heated above the adiabatic cooling floor.

Core claim

On the paper's own terms, the central discovery is a null-detection exclusion. The first observational estimate of the 21-cm forest 1D power spectrum, measured from 17.5 hours of archival uGMRT observations (11.73 effective hours after flagging, at 3.62 mJy beam^-1 per 6.1 kHz channel) of the z=5.82 quasar J352-15, is statistically indistinguishable from noise (KS p=0.353). Yet a Gaussian likelihood with a model-dependent covariance matrix, comparing the noise-subtracted spectrum against forward-modelled spectra from simulations spanning ⟨x_HI⟩∈[0,1] and ⟨T_HI⟩ from ~14 K to ~10^4 K, excludes cold and substantially neutral IGM models at z≈5.6. At the 68% credible level, the analysis disfavou

What carries the argument

The load-bearing object is the one-dimensional power spectrum P21(κν) of the 21-cm forest, defined from the flux fluctuation δF21 = F21 − 1 = S/S_cont − 1 and computed with a Lomb–Scargle periodogram over 13 logarithmically spaced bins in κν from 0.316 to 562 MHz^−1, with wavenumbers below 1 MHz^−1 discarded as sidelobe-contaminated. Because the 21-cm optical depth scales roughly as τ21 ∝ x_HI/T_S, colder or more neutral gas produces stronger absorption fluctuations and hence a larger power spectrum; this is why a noise-level measurement can still bound the x_HI–T_HI combination. The inference machinery is a Gaussian likelihood whose model-dependent covariance matrix combines empirically mea

Load-bearing premise

The load-bearing premise is that the likelihood's covariance matrix — built from off-source sightline noise and 1000 simulated sightlines — fully represents the variance of the measured power spectrum; if unmodelled systematics (such as the residual excess seen at 3–20 MHz^−1) add correlated power, the 68% exclusion curves shift.

What would settle it

A deeper re-observation of J352-15 settles it: if the excess power at 3–20 MHz^−1 persists as thermal noise drops, it is a systematic artifact and the exclusions are biased; conversely, if synthetic 21-cm forest signals at the level of the excluded cold models are injected into the off-source noise and recovered by the same pipeline, the null-detection exclusion is genuine.

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

If this is right

  • The 21-cm forest has entered the regime of observationally informative statistics: a null detection now places meaningful constraints on the thermal and ionization state of the neutral IGM.
  • The excluded cold models imply the neutral IGM at z≈5.6 has been substantially pre-heated above the adiabatic cooling floor, in a region of parameter space that Lyα and 21-cm tomography observations still allow.
  • Deeper observations of J352-15 (roughly 100 total hours) would either make a statistical detection likely if the IGM is cold and neutral, or tighten the lower limits on ⟨T_HI⟩ to 35–160 K across ⟨x_HI⟩=0.1–0.8.
  • The independent MWA measurement, though less sensitive, is consistent with noise-dominated fluctuations, showing that 21-cm forest power spectra can be pursued with multiple low-frequency radio facilities.
  • Because the forest power spectrum is sensitive to both neutral fraction and temperature simultaneously, it complements Lyα (ionization) and 21-cm tomography (temperature) probes, jointly constraining the UV and X-ray source populations driving reionization.

Where Pith is reading between the lines

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

  • If the exclusions hold, they imply the first galaxies or black holes produced enough X-rays to heat the neutral IGM to tens of kelvin by z≈5.6, sharpening the allowed X-ray efficiency of early sources beyond what Lyα and 21-cm tomography alone imply.
  • Extending the same power-spectrum analysis to the ~34 known radio-loud quasars at z>5.5 would build a multi-sightline sample, reducing sample variance and testing whether the inferred pre-heating is uniform across different lines of sight.
  • The measured excess power at intermediate scales (3≲κν≲20 MHz^−1) over idealized noise is itself a candidate signal; a targeted search for its coherence across sightlines — or its persistence in deeper data — could convert it into a detection or reveal an unmodelled bandpass artifact.
  • The statistical framework is portable to other narrow absorption structures in quasar spectra, such as minihalos or metal lines, since the same power-spectrum statistic responds to any small-scale absorption feature.

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

3 major / 4 minor

Summary. The paper presents the first observational estimate of the 1D power spectrum of the 21-cm forest, using archival uGMRT observations of the z=5.82 quasar J352–15. After direction-dependent calibration, continuum subtraction, wavelet-based bandpass removal, and channel flagging, the authors obtain a noise level of 3.62 mJy/beam per 6.1 kHz channel and a spectrum over 209–222.5 MHz. The measured power spectrum is statistically indistinguishable from off-source sightlines (KS and AD tests), so no 21-cm forest detection is claimed. The authors then compare the noise-subtracted power spectrum with forward models from 21cmFAST-generated simulations spanning a grid of ⟨x_HI⟩ and ⟨T_HI⟩, using a Gaussian likelihood with a parameter-dependent covariance matrix. The resulting Bayesian constraints are presented as 68% credible exclusions in the ⟨T_HI⟩–⟨x_HI⟩ plane, including disfavouring ⟨T_HI⟩ ≲ 27 K at ⟨x_HI⟩ = 0.1. An independent MWA measurement is presented but is not used for inference. Forecasts for a deeper campaign are also given. The central claim is that even a null detection can exclude cold, substantially neutral IGM states that are still allowed by Lyα and 21-cm tomography observations.

Significance. If the result holds, this is a milestone: it would be the first observational measurement of the 21-cm forest 1D power spectrum and the first demonstration that current-generation data can constrain the thermal state of the neutral IGM at z≈5.6. The paper has real strengths: careful use of archival data with direction-dependent calibration, an empirical noise model rather than a purely radiometric one, explicit null tests using off-source sightlines, and a forward-model grid from previously validated simulations. The claimed constraints are complementary to Lyα and global/tomographic 21-cm limits and, if robust, would indicate substantial pre-heating of the neutral IGM. However, the quantitative exclusion contours rest on modelling assumptions that are acknowledged but not quantified, so the strength of the headline limits is not yet established.

major comments (3)
  1. [Sec. 2.1, Sec. 2.6, Eq. (8)] The main quantitative result, Fig. 11, is an exclusion boundary in the ⟨T_HI⟩–⟨x_HI⟩ plane. But most of the quoted quasar proximity region is included in the analysed band. Sec. 2.1 states the proximity zone can extend to ~66 cMpc, which is ~5 MHz at 208 MHz, yet only the 208–209 MHz interval is masked while the band 209–222.5 MHz is fit with forward models that assume a globally neutral, coeval IGM with no quasar feedback. Since τ21 ∝ x_HI/T_S (Eq. 8), photoionized or X-ray-heated gas near the quasar suppresses the predicted 1D power spectrum, which is exactly the direction that disfavours cold, neutral models. This is listed as the first limitation in Sec. 6, but its impact on the exclusion boundary is not quantified. I ask for a quantitative assessment: either mask the full expected proximity band, include a simple radiative-transfer treatment, or demonstrate that the contours in Fig.
  2. [Sec. 3.2, Sec. 4, Eq. (9), Fig. 10] The likelihood in Eq. (9) assumes that the covariance matrix C(θ) fully describes the variance of the measured power spectrum. This covariance is built from off-source sightline noise and simulated sample variance. However, Sec. 3.2 explicitly reports 'excess power at intermediate scales, 3 MHz−1 ≲ κν ≲ 20 MHz−1, likely reflecting residual systematic effects beyond idealized thermal noise.' No signal-injection or null tests are presented; Sec. 6 defers them to future work. If the observed excess is due to unmodeled systematics that inflate or correlate the noise, the 68% contours in Fig. 11 will be too aggressive. I recommend validating the likelihood by comparing the target χ² against the distribution of off-source χ² values under the same covariance model, or by adding a systematic covariance term and rerunning the inference. This is necessary to secure the quoted exclusion levels.
  3. [Sec. 3.1, Sec. 6] The forward models are generated at z=6, while the observations span z≈5.4–5.8. The text in Sec. 3.1 argues that the redshift offset is modest, but Sec. 6 later acknowledges that both the neutral fraction and thermal state may evolve significantly along the 200 cMpc sightline during the final stages of reionization. A coeval z=6 model will generally overpredict absorption at the lower-redshift end of the band, again suppressing the model power spectrum relative to the data and biasing the exclusion in the same direction as the proximity effect. I ask the authors to quantify this: for example, run the same inference with lightcone-based forward models or with the band split into two redshift intervals to show that the Fig. 11 contours do not shift by more than the quoted 68% uncertainties.
minor comments (4)
  1. [Eq. (1)] The typesetting of the radiometer equation is garbled; the factor (A_eff/T_sys)^{-1} and the N_d dependence are difficult to parse. Please rewrite cleanly.
  2. [Fig. 7 caption] The caption says the vertical line marks the 'highest κν-bin expected to be dominated by sidelobe contamination,' but the line marks the κν,SL = 1 MHz−1 cutoff and the excluded modes are below this value. Please reword to avoid confusion.
  3. [Sec. 2.6] The sentence 'The signal κνP21 aligns with the spread of the noise-only cases' should read 'The measured κνP21' for clarity.
  4. [References] Some references are formatted inconsistently, e.g. 'Villanueva-Domingo & Ichiki 2022, PASJ[arXiv:2104.10695]' lacks a volume/page or DOI. Please unify the reference list.

Circularity Check

0 steps flagged

No significant circularity: the null power spectrum is compared against independent forward simulations; self-citations are methodological, not load-bearing.

full rationale

The derivation chain is externally anchored at each link. The measured 1D power spectrum comes from uGMRT visibilities through calibration, continuum subtraction, wavelet bandpass removal, and a Lomb–Scargle periodogram; the noise model is empirical, built from 1000 off-source sightlines. The signal templates are forward-modelled spectra from 21cmFAST simulations spanning a grid of ionizing and X-ray efficiencies, with the likelihood (Eqs. 9–10) forming residuals between the observed noise-subtracted spectrum and simulated signal-only spectra. No parameter is fitted to the measured power spectrum and then renamed a prediction: the inferred ⟨T_HI⟩ and ⟨x_HI⟩ are MCMC outputs, not inputs to the data calibration or noise characterization. The substantial self-citations (Šoltinský et al. 2025 for the simulation/estimator framework; Patil et al. 2026 for method A2; Šoltinský et al. 2021 for T_S=T_K) are methodological reuse of published, code-based work, not an unverified authority invoked to forbid alternatives; no uniqueness theorem or ansatz is smuggled in via citation. The paper's own stated limitations (Sec. 6: quasar radiative feedback, missing minihalos, coeval IGM, excursion-set ionization; Sec. 3.2: excess power at intermediate scales likely from residual systematics) are external-validity and correctness risks that could shift the exclusion boundary, but they do not make any equation reduce to its own input. Therefore no circular step is present.

Axiom & Free-Parameter Ledger

5 free parameters · 6 axioms · 0 invented entities

The central inference rests on forward models from 21cmFAST generated in prior same-group work; the paper contributes the data reduction, power-spectrum measurement, and likelihood framework. No new physical entities are introduced; the inferred parameters are established IGM properties, not ad hoc constructs.

free parameters (5)
  • mean neutral hydrogen fraction ⟨x_HI⟩ = posterior; uniform prior [0,1]
    Inferred parameter; the main ionization-state output constrained by the null detection.
  • mean neutral IGM temperature ⟨T_HI⟩ = posterior; uniform prior in log10(⟨T_HI⟩/K) over [log10(15),4]
    Inferred parameter; the main thermal-state output constrained by the null detection.
  • X-ray heating efficiency f_X = log10 f_X ∈ [-4,1] grid over 534 models
    Simulation input controlling the thermal state of the neutral IGM; not fitted in the MCMC but defines the ⟨T_HI⟩ grid.
  • ionizing efficiency HII_EFF_FACTOR = varied to set ⟨x_HI⟩ across models
    Simulation input controlling the neutral fraction grid.
  • empirical noise-scaling exponent (Eq. 2) = -0.45
    Fitted to five archival integration-time points; used for the 100-hour forecast, not for the measured constraints.
axioms (6)
  • standard math 21-cm optical depth formalism of Furlanetto & Loeb (2002), Eq. 8
    Standard radiative-transfer expression for 21-cm absorption; assumed without derivation.
  • domain assumption Saturated spin-temperature coupling, T_S = T_K
    Invoked in Sec. 3.1; supported by Šoltinský et al. (2021), but if Wouthuysen-Field coupling were unsaturated the signal would be weaker and the exclusions weaker.
  • domain assumption 21cmFAST excursion-set ionization field at z=6 approximates the z=5.4–5.8 lightcone probed by the observations
    Acknowledged in Sec. 6 as a limitation; the simulations are coeval and do not evolve along the line of sight.
  • domain assumption Background quasar is a point source with no radiative feedback on the surrounding IGM
    Sec. 2.1 excludes 208–209 MHz for host-galaxy and proximity effects, but the paper notes the proximity zone may extend to tens of cMpc and is not modeled.
  • domain assumption Unresolved small-scale absorbers (minihalos, dense structures) do not significantly affect the 1D power spectrum
    Acknowledged in Sec. 6; their abundance and impact remain uncertain, and they are not included in the forward models.
  • domain assumption Gaussian likelihood with model-dependent covariance is adequate
    Sec. 4; Patil et al. (2026) report non-Gaussianity in the forest, and the paper relies on robustness rather than a full non-Gaussian likelihood.

pith-pipeline@v1.3.0-alltime-deepseek · 30732 in / 13973 out tokens · 148440 ms · 2026-08-01T23:37:56.911858+00:00 · methodology

0 comments
read the original abstract

Neutral regions of the intergalactic medium (IGM) during the Epoch of Reionization (EoR) remain largely unexplored due to the limitations of existing probes. Owing to discoveries of numerous high-redshift radio-bright sources, the 21-cm forest, a series of absorption features imprinted by the neutral IGM in the spectra of such sources, now offers an attractive probe of the thermal and ionization state of the predominantly neutral IGM at $z\gtrsim5.5$. We analyse archival upgraded Giant Metrewave Radio Telescope (uGMRT) observations of J352-15, the brightest known radio-loud quasar in the EoR ($z=5.82$), to measure the one-dimensional (1D) power spectrum of the 21-cm forest. By comparing the observed power spectrum with forward-modelled synthetic spectra generated from cosmological simulations spanning a wide range of ionization and X-ray pre-heating scenarios, we perform Bayesian inference even in the absence of a statistical detection. We also present an independent Murchison Widefield Array measurement, although its lower sensitivity prevents competitive constraints. Using uGMRT, we achieve a sensitivity of $3.62\,\rm mJy\,beam^{-1}$ per $6.1\,\rm kHz$ channel. While we do not detect the 21-cm forest statistically, the null detection jointly constrains the mean neutral hydrogen fraction, $\langle x_{\rm HI}\rangle$, and the mean temperature of the neutral IGM, $\langle T_{\rm HI}\rangle$. At the $68\%$ credible level, our analysis disfavours cold and substantially neutral IGM models at $z\approx5.6$, including models with $\langle T_{\rm HI}\rangle \lesssim 27\,\rm K$ for $\langle x_{\rm HI}\rangle=0.1$. These limits probe parameter space allowed by existing Ly$\alpha$ and 21-cm observations, indicating substantial pre-heating of the neutral IGM above the adiabatic cooling floor. This demonstrates that the 21-cm forest has entered the regime of observationally informative statistics.

Figures

Figures reproduced from arXiv: 2607.15341 by Abhirup Datta (IIT Indore), Arnab Chakraborty (McGill University), Benedetta Ciardi (MPA), Cathryn M. Trott (Curtin University), Emma V. Ryan-Weber (Swinburne University of Technology), Girish Kulkarni (TIFR), James S. Bolton (University of Nottingham), Matteo Viel (SISSA), Nirupam Roy (IISc Bangalore), Nithyanandan Thyagarajan (CSIRO), Rashmi Sagar (IIT Indore), Soumak Maitra (TIFR), Tom\'a\v{s} \v{S}oltinsk\'y (INAF-OATs).

Figure 1
Figure 1. Figure 1: The continuum image of the target field at 212 [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: The spectra along 1000 arbitrary LOS of the continuum [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The noise RMS measured for various combined nights [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Residual flux-density spectrum of J352–15 after the calibration and continuum subtraction. Note that the strong absorption [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Distribution of frequency-channel flux densities of the [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Measured κνP21 (black curve) from the J352–15 spectrum ( [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Measured κνP21 from the MWA observations of J352– 15 (black) compared to the empirically estimated noise power spectrum (pink). The vertical dotted grey line indicates the κν-bin up to which the sidelobe contamination is expected to dominate. numerical approach which allows us to explore many thermal and ionization models of the IGM. Our modelling of the IGM and the forward construction of synthetic 21-cm … view at source ↗
Figure 9
Figure 9. Figure 9: also shows the synthetic 1D power spectra of the 21- cm forest signal only, κνP S 21 where the solid orange curve is the mean over 1000 synthetic LOS and the orange shaded re￾gion is 68% scatter around it. This is computed considering the same IGM model as in the top panel. The noise limit, ⟨κνP N 21⟩, both small compared to the timescales over which the thermal and ion￾ization state of the IGM evolve duri… view at source ↗
Figure 10
Figure 10. Figure 10: Covariance matrix for the noise variance (measured, left panel), sample variance (simulated assuming an IGM model of [PITH_FULL_IMAGE:figures/full_fig_p011_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Constraints on the ⟨THI⟩–⟨xHI⟩ plane based on a null detection of the 21-cm forest 1D power spectrum from the archival observations of teff = 11.73 hr and the rms noise level of σN = 3.62 mJy beam−1 targeting J352–15, a RLQSO at z = 5.82. The blue shaded region indicates the 68% credible interval favoured parameter space. For comparison, the grey vertical band indicates current Lyα-based constraints on ⟨x… view at source ↗
Figure 12
Figure 12. Figure 12: Similar to Fig [PITH_FULL_IMAGE:figures/full_fig_p013_12.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

290 extracted references · 2 canonical work pages

  1. [1]

    P., Tollerud , E

    Astropy Collaboration , Robitaille , T. P., Tollerud , E. J., et al. 2013, , 558, A33

    2013

  2. [2]

    2018, , 861, L14

    Ba \ n ados , E., Carilli , C., Walter , F., et al. 2018, , 861, L14

    2018

  3. [3]

    2021, , 909, 80

    Ba \ n ados , E., Mazzucchelli , C., Momjian , E., et al. 2021, , 909, 80

    2021

  4. [4]

    2025, Nature Astronomy, 9, 293

    Ba \ n ados , E., Momjian , E., Connor , T., et al. 2025, Nature Astronomy, 9, 293

    2025

  5. [5]

    P., et al

    Ba \ n ados , E., Schindler , J.-T., Venemans , B. P., et al. 2023, , 265, 29

    2023

  6. [6]

    D., Bolton, J

    Becker, G. D., Bolton, J. S., Madau, P., et al. 2015, , 447, 3402

    2015

  7. [7]

    D., Bolton , J

    Becker , G. D., Bolton , J. S., Zhu , Y., & Hashemi , S. 2024, , 533, 1525

    2024

  8. [8]

    D., D'Aloisio , A., Christenson , H

    Becker , G. D., D'Aloisio , A., Christenson , H. M., et al. 2021, , 508, 1853

    2021

  9. [9]

    D., Pettini , M., Rafelski , M., et al

    Becker , G. D., Pettini , M., Rafelski , M., et al. 2019, , 883, 163

    2019

  10. [10]

    2020, , 635, L7

    Belladitta , S., Moretti , A., Caccianiga , A., et al. 2020, , 635, L7

    2020

  11. [11]

    2022, , 517, 2331

    Bhagwat , A., Ciardi , B., Zackrisson , E., & Schaye , J. 2022, , 517, 2331

    2022

  12. [12]

    Bosman , S. E. I., Davies , F. B., Becker , G. D., et al. 2022, , 514, 55

    2022

  13. [13]

    D., Cairns , I., Kaplan , D

    Bowman , J. D., Cairns , I., Kaplan , D. L., et al. 2013, , 30, e031

    2013

  14. [14]

    D., Rogers, A

    Bowman, J. D., Rogers, A. E. E., Monsalve, R. A., Mozdzen, T. J., & Mahesh, N. 2018, Nature, 555, 67–70

    2018

  15. [15]

    Briggs , D. S. 1995, in American Astronomical Society Meeting Abstracts, Vol. 187, American Astronomical Society Meeting Abstracts, 112.02

    1995

  16. [16]

    Cain , C., D'Aloisio , A., Gangolli , N., & Becker , G. D. 2021, , 917, L37

    2021

  17. [17]

    2026, arXiv e-prints, arXiv:2606.24665

    Cang , J., Ciardi , B., Maity , B., et al. 2026, arXiv e-prints, arXiv:2606.24665

    Pith/arXiv arXiv 2026

  18. [18]

    C., Rogers , A

    Cappallo , R. C., Rogers , A. E. E., Lonsdale , C. J., et al. 2025, , 137, 125002

    2025

  19. [19]

    L., Gnedin, N

    Carilli, C. L., Gnedin, N. Y., & Owen, F. 2002, , 577, 22

    2002

  20. [20]

    L., Wang , R., van Hoven , M

    Carilli , C. L., Wang , R., van Hoven , M. B., et al. 2007, , 133, 2841

    2007

  21. [21]

    2026, arXiv e-prints, arXiv:2606.26435

    CD Science Working Group , Bernardi , G., Breitman , D., et al. 2026, arXiv e-prints, arXiv:2606.26435

    Pith/arXiv arXiv 2026

  22. [22]

    K., Ben \' tez-Llambay , A., Theuns , T., Frenk , C., & Bower , R

    Chan , T. K., Ben \' tez-Llambay , A., Theuns , T., Frenk , C., & Bower , R. 2024, , 528, 1296

    2024

  23. [23]

    M., Becker , G

    Christenson , H. M., Becker , G. D., Furlanetto , S. R., et al. 2021, , 923, 87

    2021

  24. [24]

    B., et al

    Ciardi , B., Inoue , S., Abdalla , F. B., et al. 2015, , 453, 101

    2015

  25. [25]

    2013, , 428, 1755

    Ciardi, B., Labropoulos, P., Maselli, A., et al. 2013, , 428, 1755

    2013

  26. [26]

    J., Cotton , W

    Condon , J. J., Cotton , W. D., Greisen , E. W., et al. 1998, , 115, 1693

    1998

  27. [27]

    2021, , 911, 120

    Connor , T., Ba \ n ados , E., Stern , D., et al. 2021, , 911, 120

    2021

  28. [28]

    K., Bharadwaj , S., & Choudhury , T

    Datta , K. K., Bharadwaj , S., & Choudhury , T. R. 2007, , 382, 809

    2007

  29. [29]

    B., Bosman , S

    Davies , F. B., Bosman , S. E. I., D'Odorico , V., et al. 2026, , 545, staf1862

    2026

  30. [30]

    de Lera Acedo , E., de Villiers , D. I. L., Razavi-Ghods , N., et al. 2022, Nature Astronomy, 6, 984

    2022

  31. [31]

    R., Parsons , A

    DeBoer , D. R., Parsons , A. R., Aguirre , J. E., et al. 2017, , 129, 045001

    2017

  32. [32]

    E., Hall , P

    Dewdney , P. E., Hall , P. J., Schilizzi , R. T., & Lazio , T. J. L. W. 2009, IEEE Proceedings, 97, 1482

    2009

  33. [33]

    2025, , 544, 1608

    Dhandha , J., Fialkov , A., Gessey-Jones , T., et al. 2025, , 544, 1608

    2025

  34. [34]

    P., Lyu , J., et al

    Endsley , R., Stark , D. P., Lyu , J., et al. 2023, , 520, 4609

    2023

  35. [35]

    J., et al

    Euclid Collaboration , Barnett , R., Warren , S. J., et al. 2019, , 631, A85

    2019

  36. [36]

    S., Mesinger, A., & Hewitt, J

    Ewall-Wice, A., Dillon, J. S., Mesinger, A., & Hewitt, J. 2014, , 441, 2476

    2014

  37. [37]

    A., Becker , R

    Fan , X., Strauss , M. A., Becker , R. H., et al. 2006, , 132, 117

    2006

  38. [38]

    & Solanki , S

    Fligge , M. & Solanki , S. K. 1997, , 124, 579

    1997

  39. [39]

    2016, The Journal of Open Source Software, 1, 24

    Foreman-Mackey, D. 2016, The Journal of Open Source Software, 1, 24

    2016

  40. [40]

    W., Lang , D., & Goodman , J

    Foreman-Mackey , D., Hogg , D. W., Lang , D., & Goodman , J. 2013, , 125, 306

    2013

  41. [41]

    Furlanetto, S. R. 2006, , 371, 867

    2006

  42. [42]

    Furlanetto, S. R. & Loeb, A. 2002, , 579, 1

    2002

  43. [43]

    R., Peng Oh, S., & Briggs, F

    Furlanetto, S. R., Peng Oh, S., & Briggs, F. H. 2006, , 433, 181

    2006

  44. [44]

    G., Davies , F

    Gaikwad , P., Haehnelt , M. G., Davies , F. B., et al. 2023, , 525, 4093

    2023

  45. [45]

    G., et al

    Gaikwad, P., Rauch, M., Haehnelt, M. G., et al. 2020, , 494, 5091

    2020

  46. [46]

    2025, , 699, A109

    Ghara , R., Zaroubi , S., Ciardi , B., et al. 2025, , 699, A109

    2025

  47. [47]

    J., Duncan , K

    Gloudemans , A. J., Duncan , K. J., Saxena , A., et al. 2022, , 668, A27

    2022

  48. [48]

    J., Saxena , A., Intema , H., et al

    Gloudemans , A. J., Saxena , A., Intema , H., et al. 2023, , 678, A128

    2023

  49. [49]

    & Weare , J

    Goodman , J. & Weare , J. 2010, Communications in Applied Mathematics and Computational Science, 5, 65

    2010

  50. [50]

    Greig , B., Mesinger , A., Koopmans , L. V. E., et al. 2021 a , , 501, 1

    2021

  51. [51]

    M., Barry , N., et al

    Greig , B., Trott , C. M., Barry , N., et al. 2021 b , , 500, 5322

    2021

  52. [52]

    2021, , 255, 28

    Gupta , N., Srianand , R., Shukla , G., et al. 2021, , 255, 28

    2021

  53. [53]

    S., et al

    Gupta , Y., Ajithkumar , B., Kale , H. S., et al. 2017, Current Science, 113, 707

    2017

  54. [54]

    R., Jarrod Millman , K., van der Walt , S

    Harris , C. R., Jarrod Millman , K., van der Walt , S. J., et al. 2020, Nature, 585, 357

    2020

  55. [55]

    F., Davies , F

    Hennawi , J. F., Davies , F. B., Wang , F., & O \ n orbe , J. 2021, , 506, 2963

    2021

  56. [56]

    2023, , 945, 124

    HERA Collaboration . 2023, , 945, 124

    2023

  57. [57]

    L., Jarvis , M

    Heywood , I., Hale , C. L., Jarvis , M. J., et al. 2020, , 496, 3469

    2020

  58. [58]

    J., Hale , C

    Heywood , I., Jarvis , M. J., Hale , C. L., et al. 2022, , 509, 2150

    2022

  59. [59]

    A., Maddox , N., et al

    Heywood , I., Ponomareva , A. A., Maddox , N., et al. 2024, , 534, 76

    2024

  60. [60]

    Hunter , J. D. 2007, Computing in Science and Engineering, 9, 90

    2007

  61. [61]

    2021, , 647, L11

    Ighina , L., Belladitta , S., Caccianiga , A., et al. 2021, , 647, L11

    2021

  62. [62]

    2023, , 519, 2060

    Ighina , L., Caccianiga , A., Moretti , A., et al. 2023, , 519, 2060

    2023

  63. [63]

    2024, , 692, A241

    Ighina , L., Caccianiga , A., Moretti , A., et al. 2024, , 692, A241

    2024

  64. [64]

    T., Jagannathan , P., Mooley , K

    Intema , H. T., Jagannathan , P., Mooley , K. P., & Frail , D. A. 2017, , 598, A78

    2017

  65. [65]

    & M \'e sz \'a ros , P

    Ioka , K. & M \'e sz \'a ros , P. 2005, , 619, 684

    2005

  66. [66]

    A., McLaughlin , M

    Jiang , M., Cui , B.-Y., Schmid , N. A., McLaughlin , M. A., & Cao , Z.-C. 2017, , 847, 75

    2017

  67. [67]

    Jordan, C., Null, D., Trott, C., et al. 2025, in 2025 URSI Asia-Pacific Radio Science Meeting, AP-RASC 2025, 2025 URSI Asia-Pacific Radio Science Meeting, AP-RASC 2025 (Institute of Electrical and Electronics Engineers Inc.), publisher Copyright: 2025 URSI.; 2025 URSI Asia-Pacific Radio Science Meeting, AP-RASC 2025 ; Conference date: 17-08-2025 Through 2...

    2025

  68. [68]

    2023, , 2023, 017

    Kadota , K., Villanueva-Domingo , P., Ichiki , K., Hasegawa , K., & Naruse , G. 2023, , 2023, 017

    2023

  69. [69]

    2025, arXiv e-prints, arXiv:2503.07074

    Kakiichi , K., Jin , X., Wang , F., et al. 2025, arXiv e-prints, arXiv:2503.07074

    Pith/arXiv arXiv 2025

  70. [70]

    2022, , 511, 4005

    Kannan , R., Garaldi , E., Smith , A., et al. 2022, , 511, 4005

    2022

  71. [71]

    J., Shibuya , T., Ouchi , M., & Kashikawa , N

    Kashino , D., Lilly , S. J., Shibuya , T., Ouchi , M., & Kashikawa , N. 2020, , 888, 6

    2020

  72. [72]

    C., Haehnelt , M

    Keating , L. C., Haehnelt , M. G., Becker , G. D., & Bolton , J. S. 2014, , 438, 1820

    2014

  73. [73]

    C., Weinberger, L

    Keating, L. C., Weinberger, L. H., Kulkarni, G., et al. 2020, , 491, 1736

    2020

  74. [74]

    S., Smirnov , O

    Kenyon , J. S., Smirnov , O. M., Grobler , T. L., & Perkins , S. J. 2018, , 478, 2399

    2018

  75. [75]

    F., & Davies , F

    Kist , T., Hennawi , J. F., & Davies , F. B. 2025, arXiv e-prints, arXiv:2508.21812

    Pith/arXiv arXiv 2025

  76. [76]

    N., Hardcastle , M

    Kondapally , R., Best , P. N., Hardcastle , M. J., et al. 2021, , 648, A3

    2021

  77. [77]

    C., Haehnelt, M

    Kulkarni, G., Keating, L. C., Haehnelt, M. G., et al. 2019, , 485, L24

    2019

  78. [78]

    2021, , 908, 124

    Liu , Y., Wang , R., Momjian , E., et al. 2021, , 908, 124

    2021

  79. [79]

    Lomb , N. R. 1976, , 39, 447

    1976

  80. [80]

    P., Leonard , A., Starck , J.-L., Abdalla , F

    Machado , D. P., Leonard , A., Starck , J.-L., Abdalla , F. B., & Jouvel , S. 2013, , 560, A83

    2013

Showing first 80 references.