pith. machine review for the scientific record. sign in

arxiv: 2604.12200 · v1 · submitted 2026-04-14 · 🌌 astro-ph.IM · astro-ph.CO

Recognition: unknown

21 cm Power Spectrum Analysis of North Celestial Pole Observations with the Tianlai Dish Pathfinder Array

Guangzhi He , Shifan Zuo , Jixia Li , Yichao Li , Furen Deng , Shijie Sun , Reza Ansari , Olivier Perdereau
show 6 more authors
Peter Timbie Albert Stebbins Ayodeji Ibitoye Fengquan Wu Yougang Wang Xuelei Chen
Authors on Pith no claims yet

Pith reviewed 2026-05-10 16:13 UTC · model grok-4.3

classification 🌌 astro-ph.IM astro-ph.CO
keywords 21 cm intensity mappingradio interferometrypower spectrumforeground removalTianlai arrayNorth Celestial Polecosmological structurepost-reionization
0
0 comments X

The pith

The Tianlai Dish Pathfinder Array has produced a 21 cm power spectrum from 107 hours of North Celestial Pole observations after calibration, imaging, and foreground subtraction.

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

The paper processes nine nights of radio interferometer data from the Tianlai Dish Pathfinder Array in the 700-800 MHz range to extract the spherically averaged 21 cm power spectrum. It applies steps including RFI flagging, calibration, point-source subtraction, and singular value decomposition to remove dominant foregrounds. This establishes that the instrument and pipeline can deliver usable power spectrum measurements at redshift around 0.9. A sympathetic reader would care because such measurements offer a way to map large-scale cosmic structure through neutral hydrogen without resolving individual galaxies.

Core claim

The central claim is that the TDPA observations have been successfully reduced to obtain the spherically averaged 21 cm power spectrum Δ²(k), demonstrating that the full analysis framework from raw data to final power spectrum is validated and ready for future use in auto-correlation and cross-correlation studies of the post-reionization universe.

What carries the argument

The singular value decomposition (SVD) foreground removal step applied after point-source subtraction, which isolates residual fluctuations for power spectrum estimation in the data reduction pipeline.

If this is right

  • Sky model refinements will reduce residuals and improve the accuracy of future power spectrum measurements.
  • Increased integration time beyond 107 hours will raise sensitivity enough to enable detection of the 21 cm signal.
  • Pipeline optimizations will allow reliable cross-correlation of the 21 cm data with optical galaxy surveys.
  • The validated framework supports extending observations to map large-scale structure at redshift near 0.9.

Where Pith is reading between the lines

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

  • The same data reduction steps could be tested on other dish arrays to check whether the SVD approach generalizes across instruments.
  • If the measured power spectrum matches expectations from simulations, it would strengthen the case for using pathfinder arrays to develop techniques ahead of larger intensity mapping surveys.
  • Residual foregrounds after SVD could be studied separately to identify which frequency modes are most contaminated and how to mitigate them.

Load-bearing premise

The assumption that singular value decomposition foreground removal combined with point-source subtraction has not removed or biased the underlying 21 cm fluctuations in a way that invalidates the reported power spectrum.

What would settle it

A test in which a simulated 21 cm signal with known amplitude is injected into the raw visibility data before processing and then fails to appear at the expected level in the final Δ²(k) would falsify the claim that the pipeline preserves the cosmological signal.

Figures

Figures reproduced from arXiv: 2604.12200 by Albert Stebbins, Ayodeji Ibitoye, Fengquan Wu, Furen Deng, Guangzhi He, Jixia Li, Olivier Perdereau, Peter Timbie, Reza Ansari, Shifan Zuo, Shijie Sun, Xuelei Chen, Yichao Li, Yougang Wang.

Figure 1
Figure 1. Figure 1: Tianlai Dish Pathfinder Array layout with antenna number and ground coordinates with respect [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Schematic diagram of the TDPA processing pipeline. Blue blocks indicate the data-processing [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: uv coverage constructed from the full TDPA dataset after data format conversion from HDF5 to CASA Measurement Set. The plot shows the distribution of baseline sampling in the uv plane, where each sampled point represents the (u, v) coordinate of a baseline at a given time stamp. 3.2 Calibration The calibration begins with a self-calibration (Cornwell & Wilkinson 1981; Pearson & Readhead 1984) and the main … view at source ↗
Figure 4
Figure 4. Figure 4: Percentage of flagged data as a function of frequency channel and antenna. Edge channels show [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Amplitude and phase of a representative baseline as a function of time and frequency, before [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Stokes I continuum images of the NCP field over 700–800 MHz. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: The left panel exhibits a clear transition around [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 7
Figure 7. Figure 7: Left: singular-value spectrum of the SVD foreground cleaning. The singular values decrease [PITH_FULL_IMAGE:figures/full_fig_p010_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Spectra extracted from the three representative pixels marked in the panel (b) of Fig. 6: P1 [PITH_FULL_IMAGE:figures/full_fig_p011_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Cylindrical power spectrum P(k⊥, k∥) at different processing stages. Left: After calibration; Middle: After point source subtraction; Right: After SVD-based foreground cleaning with Nfg = 30 modes removed. The diagonal dashed line indicates the horizon limit (k∥ = 0.56k⊥ at z ≈ 0.9), below which foreground contamination dominates the wedge-shaped region. strategy and compute P(k) using only those (k⊥, k∥) … view at source ↗
Figure 10
Figure 10. Figure 10: Spherically averaged dimensionless power spectrum [PITH_FULL_IMAGE:figures/full_fig_p014_10.png] view at source ↗
read the original abstract

The Tianlai Dish Pathfinder Array (TDPA) is a radio interferometer designed to test techniques for 21 cm intensity mapping in the post-reionization universe as a means of measuring large-scale cosmic structure. Using 9 nights of observations targeting the North Celestial Pole (NCP) field, totaling approximately 107 hours of integration time, we analyze data in the frequency range 700-800 MHz (corresponding to redshift $z \sim 0.9$). We do the data format conversion, radio frequency interference (RFI) flagging, calibration, imaging and point source subtraction, and foreground removal via Singular Value Decomposition (SVD). The spherically averaged power spectrum $\Delta^2(k)$ is obtained. This work successfully establishes and validates a comprehensive data analysis framework for the TDPA. We identify key improvements including sky model refinement, increased integration time, and pipeline optimization that will enable future detection of the 21 cm signal through auto-correlation and cross-correlation with optical galaxy surveys.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript describes the data analysis pipeline for 21 cm intensity mapping observations using the Tianlai Dish Pathfinder Array (TDPA) at the North Celestial Pole. With 9 nights of data (~107 hours integration) in the 700-800 MHz range (z ≈ 0.9), the authors perform RFI flagging, calibration, imaging, point-source subtraction, and singular value decomposition (SVD) for foreground removal to derive the spherically averaged 21 cm power spectrum Δ²(k). The central claim is that this establishes and validates a comprehensive analysis framework for the TDPA, highlighting improvements needed for future 21 cm signal detection via auto- and cross-correlations.

Significance. If the reported power spectrum is unbiased and the pipeline validated without significant signal loss, this work represents a valuable demonstration of the TDPA's capabilities for post-reionization 21 cm cosmology. It provides a practical end-to-end analysis framework that can serve as a foundation for more sensitive observations, particularly in identifying the benefits of longer integration times and refined sky models. The data-driven approach with external calibration is a strength.

major comments (2)
  1. [Abstract and foreground removal description] The SVD foreground removal step is central to the pipeline (as described in the abstract), yet the manuscript provides no details on the number of eigenmodes subtracted, nor any signal-injection tests, transfer-function analysis, or before/after power spectrum comparisons on simulated skies to quantify potential loss of 21 cm power. Without these, it is unclear whether the obtained Δ²(k) represents an unbiased measurement or is suppressed by the cleaning process, which directly impacts the claim of pipeline validation for future detections.
  2. [Results section (power spectrum)] The abstract states that the spherically averaged power spectrum Δ²(k) is obtained, but supplies no quantitative values, error bars, null tests, or explicit checks against signal loss. This omission makes it difficult to assess the significance of the result and its consistency with expectations or simulations.
minor comments (2)
  1. [Abstract] The abstract could be clarified by specifying the exact frequency coverage and integration details more quantitatively to aid readers in assessing the dataset scale.
  2. [Abstract] Notation for the power spectrum Δ²(k) is introduced without an explicit definition or reference to the standard cosmological convention used in 21 cm literature.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful and constructive review of our manuscript on the 21 cm power spectrum analysis with the Tianlai Dish Pathfinder Array. We agree that additional details on the SVD foreground removal and more explicit quantitative presentation of the power spectrum results are needed to strengthen the validation of our analysis pipeline. We will revise the manuscript accordingly. Our point-by-point responses to the major comments follow.

read point-by-point responses

  1. Referee: [Abstract and foreground removal description] The SVD foreground removal step is central to the pipeline (as described in the abstract), yet the manuscript provides no details on the number of eigenmodes subtracted, nor any signal-injection tests, transfer-function analysis, or before/after power spectrum comparisons on simulated skies to quantify potential loss of 21 cm power. Without these, it is unclear whether the obtained Δ²(k) represents an unbiased measurement or is suppressed by the cleaning process, which directly impacts the claim of pipeline validation for future detections.

    Authors: We agree that the manuscript would benefit from more explicit details on the SVD foreground removal to demonstrate lack of significant 21 cm signal loss. In the revised manuscript we will report the number of eigenmodes subtracted (selected from the eigenvalue spectrum as the dominant foreground modes) and include signal-injection tests in which simulated 21 cm fields are added to the real data before SVD cleaning, with the recovered power spectrum compared to the input. We will also add transfer-function analysis and before/after power-spectrum comparisons on simulated skies that include realistic foregrounds and noise. These additions will directly address the concern about potential suppression and support the pipeline-validation claim. revision: yes

  2. Referee: [Results section (power spectrum)] The abstract states that the spherically averaged power spectrum Δ²(k) is obtained, but supplies no quantitative values, error bars, null tests, or explicit checks against signal loss. This omission makes it difficult to assess the significance of the result and its consistency with expectations or simulations.

    Authors: We thank the referee for highlighting this presentational issue. While the results section shows the measured power spectrum, we acknowledge that numerical values, error bars, and null-test descriptions should be stated more clearly and prominently. In the revision we will tabulate or plot the specific Δ²(k) values with 1σ uncertainties across the measured k range, describe the null tests performed (including data splits by night and frequency shuffling), and provide explicit checks for signal loss through the full pipeline. These changes will make the significance and consistency of the result easier to evaluate. revision: yes

Circularity Check

0 steps flagged

No significant circularity; pipeline is data-driven processing

full rationale

The paper describes a sequential observational data pipeline (RFI flagging, calibration, imaging, point-source subtraction, SVD foreground removal) applied to 107 hours of TDPA observations to compute the spherically averaged power spectrum Δ²(k). No equations or steps reduce the reported power spectrum to a fitted parameter or self-defined quantity from the same dataset. No load-bearing self-citations, uniqueness theorems, or ansatzes imported from prior author work are present in the abstract or described chain. The result is a direct measurement after standard cleaning steps, with external calibration and sky models; the validation claim rests on successful application rather than tautological re-derivation.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Paper relies on standard radio-astronomy assumptions without introducing new free parameters, axioms beyond domain conventions, or invented entities.

axioms (1)
  • domain assumption Standard assumptions that RFI flagging, calibration, and SVD foreground removal preserve the 21 cm signal statistics.
    Invoked throughout the described pipeline.

pith-pipeline@v0.9.0 · 5528 in / 1197 out tokens · 53735 ms · 2026-05-10T16:13:13.120156+00:00 · methodology

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

57 extracted references · 6 canonical work pages

  1. [1]

    J., Luciw, N

    Anderson, C. J., Luciw, N. J., Li, Y .-C., et al. 2018, Monthly Notices of the Royal Astronomical Society, 476, 3382 2, 8

    2018

  2. [2]

    2012, Astronomy and Astrophysics, 540, A129 13

    Ansari, R., et al. 2012, Astronomy and Astrophysics, 540, A129 13

    2012

  3. [3]

    A., Marriner, J

    Ansari, R., Marques, G. A., Marriner, J. P., et al. 2025, arXiv e-prints, arXiv:2512.23899 2

    work page arXiv 2025

  4. [4]

    M., et al

    Barry, N., Wilensky, M., Trott, C. M., et al. 2019, Astrophysical Journal, 884, 1 14

    2019

  5. [5]

    A., Brown, M

    Battye, R. A., Brown, M. L., Browne, I. W. A., et al. 2012, arXiv:1209.1041 2

    work page arXiv 2012

  6. [6]

    A., Davies, R

    Battye, R. A., Davies, R. D., & Weller, J. 2004, Monthly Notices of the Royal Astronomical Society, 355, 1339 2

    2004

  7. [7]

    B., & Sethi, S

    Bharadwaj, S., Nath, B. B., & Sethi, S. K. 2001, Journal of Astrophysics and Astronomy, 22, 21 2

    2001

  8. [8]

    B., et al

    Chang, T.-C., Pen, U.-L., Peterson, J. B., et al. 2008, Physical Review Letters, 100, 091303 2

    2008

  9. [9]

    2012, International Journal of Modern Physics Conference Series, 12, 256 2

    Chen, X. 2012, International Journal of Modern Physics Conference Series, 12, 256 2

    2012

  10. [10]

    2025, arXiv:2511.19620 2

    Collaboration, C., Amiri, M., Bandura, K., et al. 2025, arXiv:2511.19620 2

    work page arXiv 2025

  11. [11]

    J., & Wilkinson, P

    Cornwell, T. J., & Wilkinson, P. N. 1981, Monthly Notices of the Royal Astronomical Society, 196, 1067 5

    1981

  12. [12]

    L., et al

    Cunnington, S., Barberi-Squarotti, M., Bernal, J. L., et al. 2026, Ap&SS, 371, 16 9

    2026

  13. [13]

    D., & Carilli, C

    Datta, A., Bowman, J. D., & Carilli, C. L. 2010, Astrophysical Journal, 724, 526 12 16 Guangzhi He et al

    2010

  14. [14]

    N., Santos, M

    Engelbrecht, B. N., Santos, M. G., Fonseca, J., et al. 2025, Monthly Notices of the Royal Astronomical Society, 536, 1035 2

    2025

  15. [15]

    S., Liu, A., et al

    Ewall-Wice, A., Dillon, J. S., Liu, A., et al. 2017, Monthly Notices of the Royal Astronomical Society, 470, 1849 14

    2017

  16. [16]

    R., Oh, S

    Furlanetto, S. R., Oh, S. P., & Briggs, F. H. 2006, Physics Reports, 433, 181 7

    2006

  17. [17]

    2020, Journal of Open Source Software, 5, 2363 12

    Giri, S., Mellema, G., & Jensen, H. 2020, Journal of Open Source Software, 5, 2363 12

    2020

  18. [18]

    E., & Dickinson, C

    Harper, S. E., & Dickinson, C. 2018, Monthly Notices of the Royal Astronomical Society, 479, 2024 2

    2018

  19. [19]

    O., Bull, P., Santos, M

    Irfan, M. O., Bull, P., Santos, M. G., et al. 2022, Monthly Notices of the Royal Astronomical Society, 509, 4923 2 Jeli´c, V ., Zaroubi, S., Labropoulos, P., et al. 2010, MNRAS, 409, 1647 15

    2022

  20. [20]

    2024, Journal of Astronomical Instrumentation, 13, 2450002 2

    Kwak, J., Podczerwinski, J., Timbie, P., et al. 2024, Journal of Astronomical Instrumentation, 13, 2450002 2

    2024

  21. [21]

    2020, Science China Physics, Mechanics, and Astronomy, 63, 129862 2

    Li, J., Zuo, S., Wu, F., et al. 2020, Science China Physics, Mechanics, and Astronomy, 63, 129862 2

    2020

  22. [22]

    2023, ApJ, 954, 139 2

    Li, Y ., Wang, Y ., Deng, F., et al. 2023, ApJ, 954, 139 2

    2023

  23. [23]

    R., & Trott, C

    Liu, A., Parsons, A. R., & Trott, C. M. 2014, Physical Review D, 90, 023018 12

    2014

  24. [24]

    Liu, A., & Shaw, J. R. 2020, Publications of the Astronomical Society of the Pacific, 132, 062001 7

    2020

  25. [25]

    W., Switzer, E

    Masui, K. W., Switzer, E. R., Banavar, N., et al. 2013, Astrophysical Journal Letters, 763, L20 2

    2013

  26. [26]

    D., Spinelli, M., Santos, M

    Matshawule, S. D., Spinelli, M., Santos, M. G., et al. 2021, Monthly Notices of the Royal Astronomical Society, 506, 5075 2

    2021

  27. [27]

    G., et al

    Mertens, F. G., et al. 2020, Monthly Notices of the Royal Astronomical Society, 493, 1662 3

    2020

  28. [28]

    2015, Astrophysics Source Code Library, ascl:1502.007 6

    Mohan, N., & Rafferty, D. 2015, Astrophysics Source Code Library, ascl:1502.007 6

    2015

  29. [29]

    F., Aguirre, J

    Moore, D. F., Aguirre, J. E., Parsons, A. R., Jacobs, D. C., & Pober, J. C. 2013, ApJ, 769, 154 15

    2013

  30. [30]

    J., Mahesh, N., Monsalve, R

    Mozdzen, T. J., Mahesh, N., Monsalve, R. A., et al. 2019, Monthly Notices of the Royal Astronomical Society, 483, 4411 2

    2019

  31. [31]

    B., Bandura, K., Bucher, M

    Newburgh, L. B., Bandura, K., Bucher, M. A., et al. 2016, in Proceedings of SPIE, V ol. 9906, 99065X 2

    2016

  32. [32]

    R., van de Gronde, J

    Offringa, A. R., van de Gronde, J. J., & Roerdink, J. B. T. M. 2012, Astronomy and Astrophysics, 539, A95 4

    2012

  33. [33]

    R., et al

    Offringa, A. R., et al. 2014, Monthly Notices of the Royal Astronomical Society, 444, 606 4

    2014

  34. [34]

    H., Yatawatta, S., Zaroubi, S., et al

    Patil, A. H., Yatawatta, S., Zaroubi, S., et al. 2016, Monthly Notices of the Royal Astronomical Society, 463, 4317 14

    2016

  35. [35]

    H., et al

    Patil, A. H., et al. 2017, Astrophysical Journal, 838, 65 3

    2017

  36. [36]

    G., Chen, Z., et al

    Paul, S., Santos, M. G., Chen, Z., et al. 2023, arXiv:2301.11943 2

    work page arXiv 2023

  37. [37]

    J., & Readhead, A

    Pearson, T. J., & Readhead, A. C. S. 1984, Annual Review of Astronomy and Astrophysics, 22, 97 5

    1984

  38. [38]

    2022, MNRAS, 517, 4637 2, 3

    Perdereau, O., Ansari, R., Stebbins, A., et al. 2022, MNRAS, 517, 4637 2, 3

    2022

  39. [39]

    C., et al

    Pober, J. C., et al. 2013, Astronomical Journal, 145, 65 13

    2013

  40. [40]

    G., Cooray, A., & Knox, L

    Santos, M. G., Cooray, A., & Knox, L. 2005, Astrophysical Journal, 625, 575 7

    2005

  41. [41]

    R., Sigurdson, K., Sitwell, M., et al

    Shaw, J. R., Sigurdson, K., Sitwell, M., et al. 2015, Physical Review D, 91, 083514 2

    2015

  42. [42]

    A Tutorial on Principal Component Analysis

    Shlens, J. 2014, arXiv e-prints, arXiv:1404.1100 8

    work page Pith review arXiv 2014

  43. [43]

    2021, Monthly Notices of the Royal Astronomical Society, 505, 1575 2

    Spinelli, M., Bernardi, G., Garsden, H., et al. 2021, Monthly Notices of the Royal Astronomical Society, 505, 1575 2

    2021

  44. [44]

    P., Cunnington, S., et al

    Spinelli, M., Carucci, I. P., Cunnington, S., et al. 2022, Monthly Notices of the Royal Astronomical Society, 509, 2048 2

    2022

  45. [45]

    2022, Publications of the Astronomical Society of the Pacific, 134, 114501 4

    Team, C. 2022, Publications of the Astronomical Society of the Pacific, 134, 114501 4

    2022

  46. [46]

    G., Chatterjee, S., et al

    Townsend, J., Santos, M. G., Chatterjee, S., et al. 2026, arXiv e-prints, arXiv:2602.23055 2

    work page arXiv 2026

  47. [47]

    2018, Astrophysical Journal, 866, 135 2

    Villaescusa-Navarro, F., Genel, S., Castorina, E., et al. 2018, Astrophysical Journal, 866, 135 2

    2018

  48. [48]

    J., Morales, M

    Wilensky, M. J., Morales, M. F., Hazelton, B. J., et al. 2019, Publications of the Astronomical Society of the Pacific, 131, 114507 15

    2019

  49. [49]

    B., & Gnanadesikan, R

    Wilk, M. B., & Gnanadesikan, R. 1968, Biometrics, 55, 1 9

    1968

  50. [50]

    B., et al

    Wolz, L., Blake, C., Abdalla, F. B., et al. 2017, Monthly Notices of the Royal Astronomical Society, 464, 4938 8 21 cm Power Spectrum Analysis of NCP Observations with the TDPA 17

    2017

  51. [51]

    2021, Monthly Notices of the Royal Astronomical Society, 506, 3455 2, 3, 6

    Wu, F., Li, J., Zuo, S., et al. 2021, Monthly Notices of the Royal Astronomical Society, 506, 3455 2, 3, 6

    2021

  52. [52]

    Wyithe, J. S. B., Loeb, A., & Geil, P. M. 2008, Monthly Notices of the Royal Astronomical Society, 383, 1195 2

    2008

  53. [53]

    2025, ApJS, 279, 32 2

    Yang, W., Wolz, L., Li, Y ., et al. 2025, ApJS, 279, 32 2

    2025

  54. [54]

    2022, Research in Astronomy and Astrophysics, 22, 125007 3

    Yu, Z., Deng, F., Sun, S., et al. 2022, Research in Astronomy and Astrophysics, 22, 125007 3

    2022

  55. [55]

    2016, MNRAS, 461, 1950 2

    Zhang, J., Ansari, R., Chen, X., et al. 2016, MNRAS, 461, 1950 2

    2016

  56. [56]

    2021, IEEE Antennas and Propagation Magazine, 63, 98 2

    Zhang, J., Liu, J., Wu, F., et al. 2021, IEEE Antennas and Propagation Magazine, 63, 98 2

    2021

  57. [57]

    2021, Astronomy and Computing, 34, 100439 4

    Zuo, S., Li, J., Li, Y ., et al. 2021, Astronomy and Computing, 34, 100439 4

    2021