Pith. sign in

REVIEW 3 major objections 6 minor 84 references

A candidate field for deep imaging of the Epoch of Reionization observed with MWA

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper reports that after removing the three strongest PCA foreground components from MWA sub-band images of a candidate Epoch of Reionization field, the residual angular power spectrum remains more than an order of magnitude above…

desk verdict Useful new deep MWA catalogue and noise estimates for a quiet EoR field; the PCA residual claim needs a noise floor before it can be taken seriously. read the letter →

arxiv 2507.08048 v1 pith:EDHIT2DX submitted 2025-07-10 astro-ph.IM astro-ph.CO

Xueying Zhang , Qian Zheng , Linhui Wu , Quan Guo , Stefan W.Duchesne , Mengfan He , Huanyuan Shan , Xiang-ping Wu
show 3 more authors
Melanie Johnston-Hollitt Feiyu Zhao Qingyuan Ma
This is my paper · ORCID
classification astro-ph.IMastro-ph.CO
keywords epochofreionization21cmcosmologyforegroundsubtractionprincipalcomponentanalysisangularpowerspectrumMurchisonWidefieldArrayradiosourcecatalogueSKApathfinder
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tests whether a sky field selected as 'quiet' for future Square Kilometre Array Epoch of Reionization imaging can be cleaned of radio foregrounds by Principal Component Analysis. From 4.43 hours of Murchison Widefield Array Phase II observations at 88 and 216 MHz, it builds deep images, detects up to 2,576 radio sources within 5 degrees of the field centre, and removes the three strongest PCA components from the four 7.68 MHz sub-band images. The central result is that the angular power spectrum of the residual maps remains more than an order of magnitude above the theoretical 21 cm CD/EoR signal predicted by 21cmFAST at every sub-band. The paper concludes that standard PCA foreground subtraction, together with this data reduction path, is not yet sufficient for deep imaging of reionization structures in this field.

What carries the argument

The mechanism that carries the argument is a three-component PCA foreground removal applied to four sub-band images, followed by an angular power-spectrum comparison. Principal Component Analysis is a linear projection onto the directions of largest variance across frequency; because foregrounds are spectrally smooth, the first components are expected to contain the foregrounds while the 21 cm signal, which varies rapidly with frequency, is expected to survive. The paper computes the brightness-temperature maps of the four 7.68 MHz sub-bands, subtracts the first three PCA components (denoted PCA-3), and measures the angular power spectrum $C_\ell$ of the residuals via spherical-harmonic decomposition. The target for comparison is the CD/EoR angular power spectrum extracted from the 21cmFAST light-cone simulation, rescaled to the same sky coverage and pixel size.

What would settle it

Run PCA-3 on the same four sub-band images after injecting a known synthetic 21 cm signal into a realistic foreground simulation: if the recovered residual power still sits an order of magnitude above the injected signal, the gap is caused by the subtraction procedure itself; if it tracks the injected signal, the observed excess is due to real unmodelled foregrounds. A simpler check is to repeat the subtraction with one, two, and four PCA components and see whether the residual power drops monotonically toward the 21cmFAST curve or saturates above it.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the G0044 field—selected as a low-brightness, low-variance candidate—still leaves a foreground residual that swamps the cosmological signal after aggressive PCA cleaning. After removing the first three principal components, over 98% of the resolved radio sources are captured by the subtracted components, and the residual angular power spectrum $C_\ell$ drops in all $\ell$-modes relative to the input temperature maps. Yet at both redshift windows ($z \approx 13$\textendash$18$ at 88 MHz and $z \approx 5$\textendash$6$ at 216 MHz) the residual power sits more than an order of magnitude above the theoretically predicted 21 cm signal from the 21cmFAST simulation. The paper reads this as evidence that the foregrounds in this field are not fully described by three spectral modes across a 7.68 MHz band, and that further improvements in data reduction and foreground subtraction are required before the field can deliver tomographic EoR images.

Load-bearing premise

The load-bearing premise is that radio foregrounds occupy no more than three spectral components across the 7.68 MHz sub-band, so subtracting the three largest PCA components removes foregrounds without projecting away the 21 cm signal; with only four frequency channels this premise is not tested in the paper.

Editorial extensions

If this is right

  • The G0044 field cannot yet be used for deep EoR imaging; the residual foreground power exceeds the expected 21 cm signal by more than an order of magnitude after PCA-3 subtraction.
  • The 216 MHz deep image, with 2,576 detected sources, 90% completeness at 10.4 mJy and an average RMS of 1.80 mJy, provides a foreground source catalogue that can feed future calibration and sky-model construction.
  • Confusion noise, not thermal noise, will be the limiting noise floor for SKA1-Low deep imaging of fields similar to G0044, given the estimated confusion levels of 1.47 mJy at 88 MHz and 0.17 mJy at 216 MHz for the MWA.
  • Because nearly all resolved sources are removed by the first three PCA components, the remaining excess power must come from diffuse or unresolved foreground structure, or from signal loss in the projection, rather than from bright point sources.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The three-mode PCA limit may be an artefact of using only four frequency channels: a wider band or finer frequency sampling would allow more foreground modes to be modelled and could close part of the gap.
  • The local minima seen in the residual spectra (around $\ell \approx 178$ at 88 MHz and $\ell \approx 116$ at 216 MHz) suggest a characteristic angular scale of residual diffuse foregrounds; if that scale is real, it may mark a window where EoR extraction is less contaminated.
  • A decisive test of whether the excess is method-inherent would be to apply an independent foreground filter, such as a Gaussian process or a trained denoiser, to the same four sub-band images and check whether the residual power approaches the 21cmFAST prediction.
  • The field's 'quietness' in source counts does not guarantee spectral cleanliness: source-count quietness and foreground-mode simplicity are different properties, and future field selection may need to rank candidates by spectral mode occupation rather than by source density.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. This manuscript presents MWA Phase II extended-array observations of a candidate SKA EoR field (G0044; center RA 8h, Dec +5 deg), producing deep images at 88 and 216 MHz with 4.43 h total integration. The authors construct a 2,576-source catalogue at 216 MHz, measure source counts and completeness, estimate thermal and confusion noise, and use PCA to remove foregrounds from four 7.68 MHz sub-band images per band. They compute angular power spectra of the PCA residuals and compare them to 21cmFAST predictions, finding that the residuals remain more than an order of magnitude above the expected EoR signal, and conclude that standard PCA foreground subtraction is insufficient for deep EoR imaging in this field.

Significance. The observational products -- source catalogue, completeness simulations, noise estimates, and field characterization -- are useful inputs for SKA1-Low field selection and for testing calibration and imaging pipelines. The source counts and confusion-noise estimates, if carefully propagated with uncertainties, are of interest to the low-frequency community. The PCA residual power-spectrum comparison, however, is currently the weakest part of the paper: the absence of a noise power spectrum means the central conclusion about foreground subtraction is not yet supported. With the addition of a noise comparison and a more careful treatment of the PCA-mode test, the paper could become a solid observational reference for this candidate EoR field.

major comments (3)
  1. [5.2 (Figs. 14-15)] The residual angular power spectra are compared only to the 21cmFAST cosmological signal; no instrumental noise power spectrum or noise-only realization is shown. With the stated thermal noise of 0.51 mJy/beam at 216 MHz and a measured RMS of 1.80 mJy/beam (Table 2), the expected noise C_l can lie many orders of magnitude above the EoR signal, so the observed factor-of-ten excess may be entirely consistent with the observation being sensitivity-limited rather than foreground-limited. The conclusion that 'standard PCA foreground subtraction is insufficient' therefore requires a demonstration that the residual power is above the noise floor; please add a noise power spectrum (analytic or from a noise-only simulation) and error bars that include the noise contribution.
  2. [5.1, Table 9] The statement that 'nearly 100% of the radio sources in observation have been extracted by PCA-3' is by construction and does not validate foreground removal. With four sub-band images, removing three principal components leaves a one-dimensional residual subspace, so any source whose spectral signature is correlated across sub-bands will be assigned to the PCA-reconstructed maps. The source-matching statistic therefore only shows that the reconstructed images retain the positions of detected sources, not that the residual map is free of foreground contamination or that the cosmological signal survives the projection. A test using injected simulated signals (for example, adding a mock EoR signal at the image level and checking its recovery rate) is needed before this claim can be assessed.
  3. [3.3.3, Eq. (3)] The SKA1-Low confusion-noise numbers are obtained by integrating the differential source-count fit (Eq. 1) that was derived from the same MWA image. While the text notes this at the end of Section 3.3.3, the quantitative statement that 'confusion noise will be the primary factor' for SKA1-Low should be accompanied by a propagation of the fit uncertainties (k and gamma) and by an explicit statement that the prediction is an extrapolation of the MWA counts, not an independent estimate. This matters because S_lim = 5 sigma is also set from the MWA RMS, and the resulting confusion noise may carry a large systematic uncertainty.
minor comments (6)
  1. [5.2, Eq. (10)] The variance expression uses only cosmic variance and f_sky; please add a noise term or note explicitly that the plotted error bars are cosmic-variance-only, otherwise the error bars in Figures 14 and 15 are incomplete.
  2. [3.1 and Figs. 5-6] The text says the integration times correspond to stacking 5, 10, 40, and all snapshots, but the figure legends label them as 10, 20, 80, and 338/266 minutes; please make the integration-time notation consistent.
  3. [Table 2] The column headers repeat the frequency labels in a way that is difficult to parse; please reformat the header so that it is clear which columns correspond to measured RMS, thermal noise, and confusion noise at 88 MHz and 216 MHz.
  4. [Table 8] Completeness corrections are listed as '——' for several high-flux bins; please either define this symbol (presumably unity) or list the actual values.
  5. [Abstract and Section 6] The phrasing 'nearly all resolved radio sources can be successfully removed using PCA' overstates what is demonstrated; suggest replacing 'removed' with 'identified' or 'reconstructed in the PCA foreground model.'
  6. [References] There are a few citation style inconsistencies, for example '(Offringa et al. 2015)' appears parenthetically in one place while 'Sokolowski et al. 2017' does not; please ensure uniform citation formatting.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the source counts, noise estimates, and PCA residual power spectra are direct measurements; the only self-citations are contextual and not load-bearing.

full rationale

The paper's central quantitative results are direct measurements from MWA data: source counts and completeness are derived from the images and compared against external catalogues (GLEAM, NVSS, LoBES), and the residual angular power spectra after PCA-3 are compared against the external 21cmFAST simulation. The confusion-noise calculation for SKA1-Low (Eq. 3) does use the source-count fit (Eq. 1) derived from the same G0044 field, but the paper explicitly states this limitation in Section 3.3.3 ('when calculating the confusion noise of SKA1-Low, we use the source count derived from the MWA observation'), making it a transparent extrapolation rather than a hidden circular prediction. Table 9's demonstration that PCA-3 components contain ~100% of the observed sources is an internal consistency check of PCA reconstruction—the components are projections of the input maps, so high overlap is expected, but this is not used to derive the paper's main conclusion. The claim that the PCA-3 residual power exceeds the 21cmFAST signal is a direct measurement; the absence of an explicit noise power spectrum in Figures 14–15 is a potential correctness/completeness issue (the residual could be noise-dominated), not a circularity. Self-citations, such as Zheng et al. (2020) for field selection and He et al. (2024) for spectral-index fitting, are contextual and do not carry the derivation of the paper's quantitative claims. No step reduces, by construction, to its own input. Score 2 reflects the presence of minor non-load-bearing self-citation but no circular derivation.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new entities. The main parameters are the fitted source-count power laws and the hand-chosen PCA component number. The analysis relies on standard astrophysical assumptions about foreground spectral smoothness, calibration accuracy, and the 21cmFAST theoretical signal.

free parameters (5)
  • dN/dS power-law normalization k at 88 MHz = 2866 ± 442
    Fitted to 88 MHz source counts above 49 mJy (Eq. 1, Section 3.2.2); used to compute confusion noise via Eq. 3.
  • dN/dS power-law index gamma at 88 MHz = 1.96 ± 0.10
    Fitted to 88 MHz source counts above 49 mJy; describes the source count slope used for confusion noise.
  • dN/dS power-law normalization k at 216 MHz = 2907 ± 219
    Fitted to 216 MHz source counts above 10 mJy (Eq. 1, Section 3.2.2); used for confusion noise and SKA forecasts.
  • dN/dS power-law index gamma at 216 MHz = 1.67 ± 0.03
    Fitted to 216 MHz source counts above 10 mJy; used for confusion noise and SKA forecasts.
  • Number of PCA foreground components removed = 3
    Chosen by hand in Section 5.1; the paper does not justify why 3 components is optimal over 2 or 4, and the result depends on this choice.
assumptions (6)
  • domain assumption Foregrounds are spectrally smooth and dominated by the first three PCA modes across a 7.68 MHz band.
    Load-bearing for the PCA-3 foreground removal and the interpretation of the residual power spectrum (Section 5.1).
  • domain assumption The GLEAM-based sky model provides an accurate absolute flux and astrometric reference, with flux errors of approximately 10%.
    Used for calibration and flux warping in the data reduction pipeline (Section 2.2).
  • domain assumption Source counts scale between 216 MHz and 154 MHz with spectral index -0.8.
    Used to scale source counts to 154 MHz for comparison with GLEAM and for completeness simulations (Section 4.4.3).
  • domain assumption The thermal noise model uses T_sky = 60 K (nu/300 MHz)^-2.25 and T_rec = 28 K.
    Standard values from Tingay et al. (2013) and Wayth et al. (2018), adopted in Section 3.3.2 for thermal noise predictions.
  • domain assumption The average synthesized beam across each sub-band is used to convert Jy/beam to brightness temperature.
    Used in Section 5.1 for temperature map generation; ignores beam variation across the sub-band.
  • domain assumption The 21cmFAST/EOS light-cone provides a realistic theoretical EoR signal for comparison.
    The theoretical benchmark in Figures 14-15 depends on unspecified astrophysical parameters (e.g., X-ray heating, reionization history), which are not stated in the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A candidate field for deep imaging of the Epoch of Reionization observed with MWA." pith.science (2026). https://pith.science/paper/EDHIT2DX

@misc{pith2026250708048,
  author       = {Pith},
  title        = {Pith review of: A candidate field for deep imaging of the Epoch of Reionization observed with MWA},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EDHIT2DX}},
  note         = {Machine review of arXiv:2507.08048}
}
abstract

Deep imaging of structures from the Cosmic Dawn (CD) and the Epoch of Reionization (EoR) in five targeted fields is one of the highest priority scientific objectives for the Square Kilometre Array (SKA). Selecting 'quiet' fields, which allow deep imaging, is critical for future SKA CD/EoR observations. Pre-observations using existing radio facilities will help estimate the computational capabilities required for optimal data quality and refine data reduction techniques. In this study, we utilize data from the Murchison Widefield Array (MWA) Phase II extended array for a selected field to study the properties of foregrounds. We conduct deep imaging across two frequency bands: 72-103 MHz and 200-231 MHz. We identify up to 2,576 radio sources within a 5-degree radius of the image center (at RA (J2000) $8^h$ , Dec (J2000) 5{\deg}), achieving approximately 80% completeness at 7.7 mJy and 90% at 10.4 mJy for 216 MHz, with a total integration time of 4.43 hours and an average RMS of 1.80 mJy. Additionally, we apply a foreground removal algorithm using Principal Component Analysis (PCA) and calculate the angular power spectra of the residual images. Our results indicate that nearly all resolved radio sources can be successfully removed using PCA, leading to a reduction in foreground power. However, the angular power spectra of the residual map remains over an order of magnitude higher than the theoretically predicted CD/EoR 21 cm signal. Further improvements in data reduction and foreground subtraction techniques will be necessary to enhance these results.

Figures

Figures reproduced from arXiv: 2507.08048 by the authors.

Figure 1
Figure 1. Distribution of spectral peak (SP) radio sources (He et al. 2024), especially showing bright SP sources with flux densities above 1 Jy. G0044 field (a purple circle covering a radius of 5 degrees) and 7 selected fields (black circles each covering 40 square degrees) are also shown with right ascension and declination listed in [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 3
Figure 3. We first run MIMAS tool from the AEGEAN (Hancock [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 2
Figure 2. Illustration of the data reduction pipeline used in this work. Showing all the different steps from data conversion to calibration and deep imaging [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figures from the paper (12 more)
Figure 3
Figure 3. Figure 3: MWA G0044 field images. Left: combining 169 observation snapshots (338 minutes) at 87.675 MHz (z = 15.2); Right: combining 133 observation snapshots (266 minutes) at 215.675 MHz (z = 5.6), both covering regions of 100 degrees2 . The radius of dashed circle is 5 degrees…
Figure 4
Figure 4. Figure 4: Integrated source counts (N) as a function of flux density (S ). Left: 5 (red), 10 (green), 40 (magenta), 169 (cyan) observation snapshots results at 88 MHz; Right: 5 (red), 10 (green), 40 (magenta), 133 (cyan) observation snapshots results at 216 MHz. the Poisson erro…
Figure 5
Figure 5. Figure 5: Differential source counts (dN / dS ) (left) and Euclidean-normalized differential source counts (S 2.5 dN / dS ) (right) as a function of flux density (S ), with different integration time of 10 (cyan), 20 (magenta), 80 (green), 338 (red) minutes at 88 MHz. Euclidean-…
Figure 6
Figure 6. Figure 6: Differential source counts (dN / dS ) (left) and Euclidean-normalized differential source counts (S 2.5 dN / dS ) (right) as a function of flux density (S ), with different integration time of 10 (cyan), 20 (magenta), 80 (green), 266 (red) minutes at 216 MHz. Euclidean…
Figure 7
Figure 7. Figure 7: Root-Mean-Squared images. Left: 169 observation snapshots (338 minutes) at 87.675 MHz (z = 15.2); Right: 133 observation snapshots (266 minutes) at 215.675 MHz (z = 5.6), both covering region of 100 degrees2 . The radius of dashed circle is 5 degrees, and the length of…
Figure 8
Figure 8. Figure 8: The distribution between the flux ratio and RA (top row) or Dec (bottom row), where the ratio is the flux density ratio of all matched sources in (left column) GLEAM at 84 MHz and our 88 MHz results; (right column) GLEAM at 212 MHz and our 216 MHz results [PITH_FULL_I…
Figure 9
Figure 9. Figure 9: Source position offsets in RA and Dec when matching with NVSS. We also show the histograms of the differences in RA (top) and Dec (right), and the dashed lines are the mean value of these differences, which are closing to zero. the same parameters employed during the c…
Figure 11
Figure 11. Figure 11: Ratios of the integrated to peak intensity as a function of the signal-to-noise ratio of the detected source within 5-degree radius. The dark green circles represent extended sources while the light green crosses are point-like sources. Extended sources account for 15…
Figure 12
Figure 12. Figure 12: MWA G0044 project observation temperature map (left), PCA-3 components (middle) and the residual map after removing PCA-3 components (right) at redshift z = 17.7 (top) and z = 14.5 (bottom) at center frequency of 88 MHz. PCA-3 results are derived from the analysis of …
Figure 13
Figure 13. Figure 13: MWA G0044 project observation temperature map (left), PCA-3 components (middle) and the residual map after removing PCA-3 components (right) at redshift z = 6.0 (top) and z = 5.5 (bottom) at center frequency of 216 MHz. PCA-3 results are derived from the analysis of f…
Figure 14
Figure 14. Figure 14: Angular power spectrum Cℓ(0) at redshift 17.7 (top left), 15.9 (top right), 14.5 (bottom left), 13.3 (bottom right) for observation temperature map (blue), the residual map after removing 2 (orange) and 3 (green) PCA components and cosmological signal generated by 21c…
Figure 15
Figure 15. Figure 15: Angular power spectrum Cℓ(0) at redshift 6.0 (top left), 5.7 (top right), 5.5 (bottom left), 5.3 (bottom right) for observation temperature map (blue), the residual map after removing 3 (green) PCA components and cosmological signal generated by 21cmFAST for compariso…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

84 extracted references · 72 canonical work pages

  1. [1]

    Adam R. et al. , 2016, A&A, 594, A1

    2016

  2. [2]

    Aghanim N. et al. , 2020, A&A, 641, A6

    2020

  3. [3]

    G., Santos M

    Alonso D., Bull P., Ferreira P. G., Santos M. G., 2015, , 447, 400

    2015

  4. [4]

    Barry N., Beardsley A., Byrne R., Hazelton B., Morales M., Pober J., Sullivan I., 2019 a , , 36, e026

    2019

  5. [5]

    Barry N. et al. , 2019 b , , 884, 1

    work page 2019

  6. [6]

    Beardsley A. P. et al. , 2019, , 36, e050

    work page 2019

  7. [7]

    Bernardi G., McQuinn M., Greenhill L., 2015, , 799, 90

    work page 2015

  8. [8]

    Bock D. C. J., Large M. I., Sadler E. M., 1999, , 117, 1578

    work page 1999

Show all 84 references
  1. [9]

    Bowman J. D. et al. , 2013, , 30, e031

    2013

  2. [10]

    D., Rogers A

    Bowman J. D., Rogers A. E., Hewitt J. N., 2008, , 676, 1

    2008

  3. [11]

    D., Rogers A

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

    2018

  4. [12]

    S., 1995, in American Astronomical Society Meeting Abstracts, Vol

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

    1995

  5. [13]

    B., 2010, arXiv preprint arXiv:1007.3709

    Chang T.-C., Pen U.-L., Bandura K., Peterson J. B., 2010, arXiv preprint arXiv:1007.3709

    2010 arXiv

  6. [14]

    Chapman E. et al. , 2012, , 423, 2518

    2012

  7. [15]

    Condon J., 1974, , 188, 279

    1974

  8. [16]

    Condon J. et al. , 2012, , 758, 23

    2012

  9. [17]

    J., Cotton W

    Condon J. J., Cotton W. D., Greisen E. W., Yin Q. F., Perley R. A., Taylor G. B., Broderick J. J., 1998, , 115, 1693

    1998

  10. [18]

    de Oliveira-Costa A., Tegmark M., Gaensler B., Jonas J., Landecker T., Reich P., 2008, , 388, 247

    2008

  11. [19]

    DeBoer D. R. et al. , 2017, , 129, 045001

    2017

  12. [20]

    J., 2002, , 564, 576

    Di Matteo T., Perna R., Abel T., Rees M. J., 2002, , 564, 576

    2002

  13. [21]

    W., Johnston-Hollitt M., Bartalucci I., 2021, , 38, e053

    Duchesne S. W., Johnston-Hollitt M., Bartalucci I., 2021, , 38, e053

    2021

  14. [22]

    W., Johnston-Hollitt M., Zhu Z., Wayth R

    Duchesne S. W., Johnston-Hollitt M., Zhu Z., Wayth R. B., Line J. L. B., 2020, , 37, e037

    2020

  15. [23]

    Franzen T. et al. , 2015, , 453, 4020

    2015

  16. [24]

    Franzen T., Vernstrom T., Jackson C., Hurley-Walker N., Ekers R., Heald G., Seymour N., White S., 2019, , 36, e004

    2019

  17. [25]

    Franzen T. M. et al. , 2016, , 459, 3314

    2016

  18. [26]

    Hale C. L. et al. , 2021, arXiv preprint arXiv:2109.00956

    2021 arXiv

  19. [27]

    J., Murphy T., Gaensler B

    Hancock P. J., Murphy T., Gaensler B. M., Hopkins A., Curran J. R., 2012, , 422, 1812

    2012

  20. [28]

    J., Trott C

    Hancock P. J., Trott C. M., Hurley-Walker N., 2018, , 35, e011

    2018

  21. [29]

    He M., Zheng Q., Guo Q., Shan H., Zhu Z., Xie Y., Huang Y., Zhao F., 2024, , 529, 3140

    2024

  22. [30]

    Hurley-Walker N. et al. , 2017, , 464, 1146

    2017

  23. [31]

    Hurley-Walker N. et al. , 2022, , 39, e035

    2022

  24. [32]

    Intema H., van Weeren R., R \"o ttgering H., Lal D., 2011, A&A, 535, A38

    2011

  25. [33]

    Li W. et al. , 2019, , 485, 2628

    2019

  26. [34]

    Line J. L. B., Trott C., Barry N., Null D., Jordan C. H., 2025, , 42, e024

    2025

  27. [35]

    R., 2020, , 132, 062001

    Liu A., Shaw J. R., 2020, , 132, 062001

    2020

  28. [36]

    Liu A., Tegmark M., 2012, , 419, 3491

    2012

  29. [37]

    Lynch C. R. et al. , 2021, , 38, e057

    2021

  30. [38]

    L., Lancaster L., Villaescusa-Navarro F., Melchior P., Ho S., Perreault-Levasseur L., Spergel D

    Makinen T. L., Lancaster L., Villaescusa-Navarro F., Melchior P., Ho S., Perreault-Levasseur L., Spergel D. N., 2021, J. Cosmology Astropart. Phys., 2021, 081

    2021

  31. [39]

    Masui K. et al. , 2013, , 763, L20

    2013

  32. [40]

    J., Curran J., Hunstead R

    Mauch T., Murphy T., Buttery H. J., Curran J., Hunstead R. W., Piestrzynski B., Robertson J. G., Sadler E. M., 2003, , 342, 1117

    2003

  33. [41]

    Mertens F. G. et al. , 2020, , 493, 1662

    2020

  34. [42]

    Mesinger A., Furlanetto S., 2007, , 669, 663

    2007

  35. [43]

    Mesinger A., Furlanetto S., Cen R., 2011, , 411, 955

    2011

  36. [44]

    Mesinger A., Greig B., Sobacchi E., 2016, , 459, 2342

    2016

  37. [45]

    J., Dulwich F., Salvini S., Adami K

    Mort B. J., Dulwich F., Salvini S., Adami K. Z., Jones M. E., 2010, in 2010 IEEE International Symposium on Phased Array Systems and Technology, IEEE, pp. 690--694

    2010

  38. [46]

    W., Piestrzynska B., Kels A

    Murphy T., Mauch T., Green A., Hunstead R. W., Piestrzynska B., Kels A. P., Sztajer P., 2007, , 382, 382

    2007

  39. [47]

    G., Trott C

    Murray S. G., Trott C. M., Jordan C. H., 2017, , 845, 7

    2017

  40. [48]

    C., Power C., 2020, , 893, 118

    Nasirudin A., Murray S., Trott C., Greig B., Joseph R. C., Power C., 2020, , 893, 118

    2020

  41. [49]

    G., Mesinger A., Bernardi G., 2022, , 514, 4655

    Nasirudin A., Prelogovic D., Murray S. G., Mesinger A., Bernardi G., 2022, , 514, 4655

    2022

  42. [50]

    Ni S., Li Y., Gao L.-Y., Zhang X., 2022, , 934, 83

    2022

  43. [51]

    Offringa A., De Bruyn A., Biehl M., Zaroubi S., Bernardi G., Pandey V., 2010, , 405, 155

    2010

  44. [52]

    Offringa A., Van De Gronde J., Roerdink J., 2012, A&A, 539, A95

    2012

  45. [53]

    Offringa A. R. et al. , 2014, , 444, 606

    2014

  46. [54]

    R., Smirnov O., 2017, , 471, 301

    Offringa A. R., Smirnov O., 2017, , 471, 301

    2017

  47. [55]

    Offringa A. R. et al. , 2016, , 458, 1057

    2016

  48. [56]

    Offringa A. R. et al. , 2015, , 32, e008

    2015

  49. [57]

    Paciga G. et al. , 2013, , 433, 639

    2013

  50. [58]

    K., Datta A., Mazumder A., 2024, arXiv preprint arXiv:2407.17573

    Pal S. K., Datta A., Mazumder A., 2024, arXiv preprint arXiv:2407.17573

    2024 arXiv

  51. [59]

    Parsons A. R. et al. , 2010, , 139, 1468

    2010

  52. [60]

    Patra N., Subrahmanyan R., Raghunathan A., Udaya Shankar N., 2013, Experimental Astronomy, 36, 319

    2013

  53. [61]

    Pedregosa F. et al. , 2011, the Journal of machine Learning research, 12, 2825

    2011

  54. [62]

    R., Loeb A., 2012, Reports on Progress in Physics, 75, 086901

    Pritchard J. R., Loeb A., 2012, Reports on Progress in Physics, 75, 086901

    2012

  55. [63]

    Procopio P. et al. , 2017, , 34, e033

    2017

  56. [64]

    Rahimi M. et al. , 2021, , 508, 5954

    2021

  57. [65]

    Scheuer P. A. G., 1957, Proceedings of the Cambridge Philosophical Society, 53, 764

    1957

  58. [66]

    Sokolowski M. et al. , 2017, , 34, e062

    2017

  59. [67]

    Sullivan I. S. et al. , 2012, , 759, 17

    2012

  60. [68]

    R., Chang T.-C., Masui K

    Switzer E. R., Chang T.-C., Masui K. W., Pen U.-L., Voytek T. C., 2015, , 815, 51

    2015

  61. [69]

    J., Hu W., de Oliveira-Costa A., 2000, , 530, 133

    Tegmark M., Eisenstein D. J., Hu W., de Oliveira-Costa A., 2000, , 530, 133

    2000

  62. [70]

    Tingay S. J. et al. , 2013, , 30, e007

    2013

  63. [71]

    Trott C. M. et al. , 2020, , 493, 4711

    2020

  64. [72]

    van Haarlem M. P. et al. , 2013, A&A, 556, A2

    2013

  65. [73]

    C., Natarajan A., Garc \' a J

    Voytek T. C., Natarajan A., Garc \' a J. M. J., Peterson J. B., L \'o pez-Cruz O., 2014, , 782, L9

    2014

  66. [74]

    Wayth R. et al. , 2015, , 32, e025

    2015

  67. [75]

    Wayth R. B. et al. , 2018, , 35, e033

    2018

  68. [76]

    Williams W., Intema H., R \"o ttgering H., 2013, A&A, 549, A55

    2013

  69. [77]

    Yatawatta S. et al. , 2013, A&A, 550, A136

    2013

  70. [78]

    Zhao B.-X., Zheng Q., Shan H.-Y., Guo Q., Li K.-J., 2022, RAA, 22, 015012

    2022

  71. [79]

    W., Li W., 2020, , 499, 3434

    Zheng Q., Wu X.-P., Guo Q., Johnston-Hollitt M., Shan H., Duchesne S. W., Li W., 2020, , 499, 3434

    2020

  72. [80]

    Zheng Q., Wu X.-P., Johnston-Hollitt M., Gu J.-h., Xu H., 2016, , 832, 190

    2016

  73. [81]

    Zonca A., Singer L., Lenz D., Reinecke M., Rosset C., Hivon E., Gorski K., 2019, Journal of Open Source Software, 4, 1298

    2019

  74. [82]

    , " * write output.state after.block = add.period write newline

    ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence a...

  75. [83]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

  76. [84]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.stat...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.