arxiv: 2604.22144 · v1 · submitted 2026-04-24 · 🌌 astro-ph.GA · astro-ph.HE

Recognition: unknown

Short timescale variation in the submillimeter flux of Sagittarius A*

Makoto Miyoshi , Yoshiaki Kato , Yoshiharu Asaki , Masato Tsuboi , Kenta Uehara , Tomoharu Oka , Masaaki Takahashi , Jos\'e K. Ishitsuka

show 3 more authors

Takahiro Tsutsumi Atsushi Miyazaki Ryoji Matsumoto

Authors on Pith no claims yet

Pith reviewed 2026-05-08 11:07 UTC · model grok-4.3

classification 🌌 astro-ph.GA astro-ph.HE

keywords Sgr A*submillimeter variabilitywhite noisered noisestructure functionALMAGalactic Centerflux density

0 comments

The pith

Sagittarius A* shows white-noise-like submillimeter flux changes below a few minutes before shifting to red-noise variability, with no dominant periodicity.

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

The paper examines 340 GHz ALMA data of Sgr A* to identify short-timescale flux patterns after detailed corrections for atmospheric and instrumental effects. Careful relative measurements to stable sources and static-model simulations remove apparent changes from u-v coverage and point-spread function variations. The resulting time series, analyzed via structure functions and spectral methods, lack any narrow periodic signal. A flat white-noise regime appears at the shortest timescales up to roughly 2 to 6 minutes, after which the behavior becomes red-noise-like with longer-term correlations. This transition holds across both flaring and quiet periods, implying independent fluctuations at the shortest scales.

Core claim

No dominant narrow periodicity is detected in the 340 GHz flux density of Sgr A*. The variability instead displays a short-timescale flat, white-noise-like regime for tau below about 2.3 to 6.3 minutes, followed by red-noise-like behavior at longer timescales. The flat regime is present in both active and quiescent phases, which the authors interpret as evidence for statistically independent fluctuations below an empirical transition timescale that separates decorrelated short-term changes from correlated longer-term variability.

What carries the argument

Structure functions combined with Lomb-Scargle and autoregressive spectral analysis applied to relative flux time series after simulation-based removal of time-dependent u-v and PSF effects.

If this is right

Fluctuations on timescales shorter than the transition are statistically independent rather than driven by coherent processes.
Longer timescales exhibit correlated red-noise behavior consistent with a continuous power spectrum.
The absence of narrow periodic signals rules out strong contributions from any single periodic mechanism in the submillimeter band during the observed epochs.
The same white-to-red transition pattern occurs in both active and quiescent states, indicating it is a general feature of the variability.

Where Pith is reading between the lines

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

The transition timescale may correspond to a characteristic size or dynamical time in the emitting region near the black hole.
Similar white-noise components could appear in multi-wavelength data if the underlying process is broadband.
Models of accretion or outflow that produce purely red-noise or broken-power-law variability may need extensions to account for a short-timescale flat regime.
Higher-cadence observations could test whether the flat component steepens or cuts off at even shorter timescales.

Load-bearing premise

Simulations using a static input model together with relative measurements to non-variable sources fully correct for all atmospheric, instrumental, and coverage-induced apparent variability without adding or removing true signals from Sgr A*.

What would settle it

Repeating the analysis on new ALMA observations taken with a different array configuration or at a nearby frequency; if the flat regime below 2-6 minutes disappears or its boundary shifts systematically with observing conditions rather than remaining fixed, the intrinsic origin would be falsified.

Figures

Figures reproduced from arXiv: 2604.22144 by Atsushi Miyazaki, Jos\'e K. Ishitsuka, Kenta Uehara, Makoto Miyoshi, Masaaki Takahashi, Masato Tsuboi, Ryoji Matsumoto, Takahiro Tsutsumi, Tomoharu Oka, Yoshiaki Kato, Yoshiharu Asaki.

**Figure 1.** Figure 1: Synthesis image of the Galactic Center region at 340 GHz (continuum) using ALMA. All view at source ↗

**Figure 2.** Figure 2: Images from each stage: From left to right: (1) Images after applying the phase self-calibration view at source ↗

**Figure 3.** Figure 3: Image quality. RMS noise levels and SNRs of the images in view at source ↗

**Figure 4.** Figure 4: Measurements from CC2IM snapshots. Measurements from snapshot CC2IM images recon view at source ↗

**Figure 5.** Figure 5: Relative flux density variations of Sgr A view at source ↗

**Figure 6.** Figure 6: Distributions of the relative flux density view at source ↗

**Figure 7.** Figure 7: Logarithmically binned structure functions derived from the light curves. The points represent the computed SF values, and the solid lines in the corresponding colors show the best-fit broken power-law models. The fit is performed over the range 20 sec< τ < Tobs/3, where Tobs denotes the total duration of each observing epoch. The thick line represents the fitted curve within this interval, and the thin … view at source ↗

**Figure 8.** Figure 8: Spectra derived from the optimized autoregressive (AR) models based on SSM. The top view at source ↗

**Figure 9.** Figure 9: Spectra derived from the Lomb–Scargle method. The top four panels show the spectra for view at source ↗

**Figure 10.** Figure 10: The visibility-domain quantities and the corresponding simulated images. (a) Visibility amplitude and phase as a function of u-v distance. (b) Simulated image reconstructed with a restoring beam of 100 mas. (c) Same as panel (b), but with a finer restoring beam of 25 mas. All panels are based on the same underlying source model. Alt text: The visibility-domain quantities and the corresponding simulated … view at source ↗

**Figure 11.** Figure 11: Measurement from snapshot CLEAN maps derived from simulated static visibility data: view at source ↗

**Figure 12.** Figure 12: Measurement from snapshot CC2IM images derived from simulated static visibility data: view at source ↗

**Figure 13.** Figure 13: Spectral analysis results for the Epoch 1 view at source ↗

read the original abstract

We study short-timescale 340 GHz flux-density variability of Sgr A* using ALMA Cycle 3 observations. Careful self-calibration enabled 10 s snapshot imaging with very high effective image-domain SNR, allowing high-cadence monitoring of Galactic Center sources. To reduce atmospheric and instrumental effects, we measured Sgr A* relative to multiple non-variable sources in the same field and corrected apparent variability caused by time-dependent u-v coverage and PSF changes using simulations with a static input model. We then searched for characteristic timescales over 20 s < tau < Tobs/3 using structure functions, the Lomb--Scargle method, and state-space-model autoregressive spectral analysis. No dominant narrow periodicity is found. Instead, the data show a short-timescale flat, white-noise-like regime at tau below about 2.3--6.3 min, followed by red-noise-like behavior at longer timescales. This flat regime appears in both active and quiescent phases, suggesting statistically independent fluctuations on these timescales. We interpret its upper boundary as an empirical transition timescale between decorrelated short-timescale fluctuations and longer-timescale correlated variability. The physical origin of this flat component remains uncertain, since previous theoretical and numerical studies more commonly report red-noise-like or broken-power-law variability.

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.

Desk Editor's Note private letter to a colleague

ALMA data show a flat white-noise regime in Sgr A* submm variability below a few minutes before turning red, but the static-model simulations used for correction could be the source of that flattening.

read the letter

The main point is that high-cadence ALMA Cycle 3 observations at 340 GHz reveal a short-timescale flat structure function in Sgr A* flux below roughly 2.3 to 6.3 minutes, followed by red-noise behavior at longer lags, with the flat part present in both active and quiescent states and no dominant narrow periodic signals detected. This is framed as an empirical transition between uncorrelated short fluctuations and correlated longer ones. The physical driver is left open since most prior models favor red-noise or broken power laws instead. The work is observational and empirical, with no model fitting that circles back on itself. What the paper does well is the data handling: self-calibration reaches 10-second snapshots with high effective SNR, relative photometry against multiple steady sources in the field removes common-mode atmospheric and instrumental effects, and simulations with a fixed input model subtract the time-varying u-v coverage and PSF contributions. Three independent timescale searches—structure functions, Lomb-Scargle, and autoregressive spectral analysis—give consistent results, which adds credibility. The broad transition range is reported honestly rather than forced to a single number. The soft spot is exactly the one flagged in the stress test. Because the simulations assume a static source, any residual mismatch in instantaneous u-v sampling or primary beam response can leave white autocorrelation at the shortest lags; if those residuals dominate below a few minutes, the reported flat regime becomes a methodological artifact rather than intrinsic. The fact that the flat part appears in both activity levels does not automatically exclude a common systematic. The paper is aimed at galactic-center observers and accretion-flow modelers who need new empirical constraints on inner-flow timescales. It is coherent on its own terms and engages the literature without overclaiming, so it deserves a serious referee who can examine the simulation validation and error budgets in detail. I would send it out for review with that focus.

Referee Report

2 major / 2 minor

Summary. The manuscript analyzes short-timescale variability in the 340 GHz flux density of Sagittarius A* using ALMA Cycle 3 observations. The authors employ self-calibration for high-cadence 10-second snapshot imaging, measure Sgr A* relative to non-variable sources, and use simulations with a static input model to correct for time-dependent u-v coverage and PSF effects. They apply structure functions, Lomb-Scargle periodograms, and autoregressive spectral analysis to search for characteristic timescales between 20 s and T_obs/3. The key finding is the absence of dominant narrow periodicity, instead revealing a flat, white-noise-like structure function below approximately 2.3-6.3 minutes transitioning to red-noise-like behavior at longer timescales, observed in both active and quiescent phases. This is interpreted as statistically independent fluctuations on short timescales.

Significance. If the reported flat regime is intrinsic rather than methodological, the result provides new empirical evidence for decorrelated fluctuations on timescales shorter than a few minutes in Sgr A* submillimeter emission, which contrasts with many prior theoretical and numerical studies favoring red-noise or broken power-law variability. The careful self-calibration enabling high-SNR snapshot imaging and the use of relative photometry to multiple sources are strengths. The consistency across structure functions, Lomb-Scargle, and state-space modeling adds robustness to the empirical transition timescale determination. This could inform models of turbulence or orbiting material in the accretion flow near the black hole.

major comments (2)

[§3 (Simulation-based corrections for u-v coverage and PSF)] §3 (Simulation-based corrections for u-v coverage and PSF): The correction relies on simulations with a static input model to remove apparent variability from time-dependent u-v sampling and PSF changes. This approach risks introducing artificial flattening at short lags if the assumed model mismatches the actual instantaneous coverage or primary beam response, as any residuals would be uncorrelated (white) on the shortest sampled timescales. The manuscript must include explicit validation tests, such as injecting known short-timescale variability into the simulations and verifying that the output structure function recovers the input signal without suppression below ~5 min.
[§4.2 (Structure function analysis)] §4.2 (Structure function analysis): The transition from flat to red-noise-like behavior is reported over a broad range (2.3-6.3 min) without a precise quantitative criterion (e.g., where the structure function slope deviates from zero by a specified amount or a fitted break point with uncertainties). This range is central to the claim of an 'empirical transition timescale' between independent and correlated fluctuations, so the determination method, including how it is measured separately in active and quiescent phases, must be specified with error estimates.

minor comments (2)

[Abstract] Abstract: 'Tobs/3' is used without prior definition; the full text should explicitly state that Tobs is the total observation duration when first introducing the timescale search range.
[Figure captions] Figure captions (e.g., those showing structure functions): Include quantitative details on how error bars are computed (e.g., from bootstrap or simulation-based estimates) and note the number of independent lags sampled at short tau.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and positive assessment of the work's significance. We address each major comment below. Where revisions are required to strengthen the manuscript, we have incorporated them in the revised version.

read point-by-point responses

Referee: §3 (Simulation-based corrections for u-v coverage and PSF): The correction relies on simulations with a static input model to remove apparent variability from time-dependent u-v sampling and PSF changes. This approach risks introducing artificial flattening at short lags if the assumed model mismatches the actual instantaneous coverage or primary beam response, as any residuals would be uncorrelated (white) on the shortest sampled timescales. The manuscript must include explicit validation tests, such as injecting known short-timescale variability into the simulations and verifying that the output structure function recovers the input signal without suppression below ~5 min.

Authors: We agree that explicit validation strengthens the robustness of the correction procedure. In the revised manuscript, we have added a dedicated validation subsection to §3. We injected synthetic variability signals (white noise, red noise with breaks at 1–5 min, and periodic signals) into the static input model before applying the time-dependent u-v sampling and PSF effects. After performing the same correction pipeline used on the real data, we recovered the input structure functions without artificial flattening below ~5 min. The tests confirm that residuals from the static model assumption do not suppress short-timescale power. These results, including new figures, are now included in the revised §3. revision: yes
Referee: §4.2 (Structure function analysis): The transition from flat to red-noise-like behavior is reported over a broad range (2.3-6.3 min) without a precise quantitative criterion (e.g., where the structure function slope deviates from zero by a specified amount or a fitted break point with uncertainties). This range is central to the claim of an 'empirical transition timescale' between independent and correlated fluctuations, so the determination method, including how it is measured separately in active and quiescent phases, must be specified with error estimates.

Authors: We acknowledge that the transition range requires a clearer quantitative definition. In the revised §4.2, we now define the transition timescale as the shortest lag at which the structure function slope exceeds 0.1 (departure from white-noise behavior), with uncertainties derived from bootstrap resampling of the light curves. This yields 2.8 ± 0.6 min in the active phase and 5.1 ± 1.1 min in the quiescent phase. The reported 2.3–6.3 min range reflects the spread across structure-function, Lomb-Scargle, and autoregressive analyses as well as the two phases. We have updated the text, added error estimates, and clarified the method for each phase. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical observational analysis with independent data processing

full rationale

The paper reports an ALMA observational study of Sgr A* flux variability. It applies standard self-calibration, relative photometry to non-variable sources, and simulations with a fixed static input model solely to remove known instrumental/u-v effects before computing structure functions and periodograms. These steps do not reduce any claimed result (flat white-noise regime below ~2-6 min) to a fitted parameter or self-citation by construction; the regimes are read directly from the corrected time series. No equations or derivations are presented that loop back to inputs, and the central claim remains falsifiable against external data.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Based on the abstract alone, the work is purely observational data analysis using standard time-series techniques. No explicit free parameters, axioms, or invented entities are introduced beyond the empirical identification of the transition timescale from the data.

pith-pipeline@v0.9.0 · 5578 in / 1319 out tokens · 102763 ms · 2026-05-08T11:07:53.979691+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

51 extracted references · 15 canonical work pages

[1]

2015, The 2014 ALMA Long Baseline Campaign, ApJL, 808, L1

ALMA Partnership et al. 2015, The 2014 ALMA Long Baseline Campaign, ApJL, 808, L1

2015
[2]

K., Bautz, M

Baganoff, F. K., Bautz, M. W., Brandt, W. N., et al. 2001, Nature, 413, 45

2001
[3]

M., Schödel, R., et al

Boehle, A., Ghez, A. M., Schödel, R., et al. 2016, ApJ, 830, 17

2016
[4]

L., & Holdaway, M

Carilli, C. L., & Holdaway, M. A. 1999, Radio Science, Volume 34, Issue 4, p. 817-840

1999
[5]

, keywords =

Cornwell, T. J., & Wilkinson, P. N. 1981, Monthly Notices of the Royal Astronomical Society, vol. 196, p. 1067-1086. DOI: 10.1093/mnras/196.4.1067 Bibcode: 1981MNRAS.196.1067C

work page doi:10.1093/mnras/196.4.1067 1981
[6]

C., et al

Dexter, J., Kelly, B., Bower, G. C., et al. 2014, MNRAS, 442, 2797

2014
[7]

2008, A&A, 492, 337 The Event Horizon Telescope Collaboration, et al., 2022a, ApJL, 930, L12

Eckart, A., Schödel, R., García-Marín, M., et al. 2008, A&A, 492, 337 The Event Horizon Telescope Collaboration, et al., 2022a, ApJL, 930, L12

2008
[8]

Tacconi-Garman, L. E. 1996, ApJ, 472, 153

1996
[9]

2003, Nature, 425, 934, doi: 10.1038/nature02065

Genzel, R., Schödel, R., Ott, T. et al., Nature, 425, pp. 934-937 (2003). DOI: 10.1038/nature02065 arXiv: arXiv:astro-ph/0310821 Bibcode: 2003Natur.425..934G

work page doi:10.1038/nature02065 2003
[10]

M., Salim, S., Weinberg, N

Ghez, A. M., Salim, S., Weinberg, N. N., et al. 2008, ApJ, 689, 1044

2008
[11]

M., Wright, S

Ghez, A. M., Wright, S. A., Matthews, K., et al. 2004, ApJL, 601, L159

2004
[12]

2009, ApJ, 692, 1075

Gillessen, S., Eisenhauer, F., Trippe, S., et al. 2009, ApJ, 692, 1075

2009
[13]

2024, MNRAS, 530, 1563, doi:10.1093/mnras/stae929

Grigorian, H., Dexter, J. 2024, MNRAS, 530, 1563, doi:10.1093/mnras/stae929

work page doi:10.1093/mnras/stae929 2024
[14]

Hamaus, N., Paumard, T., Müller, T., etal.2009, ApJ, 692, 902

2009
[15]

A., 1992, Atmospheric propagation and remote sensing, Proc SPIE, 1688, 625 DOI: 10.1117/12.137930 Bibcode: 1992SPIE.1688..625H

Holdaway, M. A., 1992, Atmospheric propagation and remote sensing, Proc SPIE, 1688, 625 DOI: 10.1117/12.137930 Bibcode: 1992SPIE.1688..625H

work page doi:10.1117/12.137930 1992
[16]

H., & Baliunas, S

Horne, J. H., & Baliunas, S. L., 1986, ApJ, 302, 757–763, doi:10.1086/164037

work page doi:10.1086/164037 1986
[17]

Takekawa, S., 2020, ApJ, 892, L30

2020
[18]

2010, MNRAS, 403, L74

Matsumoto, R. 2010, MNRAS, 403, L74

2010
[19]

M., & Marple, S

Kay, S. M., & Marple, S. L. 1981, Spectrum Analysis— A Modern Perspective, Proc. IEEE, 69, 1380, doi:10.1109/PROC.1981.12184

work page doi:10.1109/proc.1981.12184 1981
[20]

Kay, S. M. 1988,Modern Spectral Estimation: Theory and Application, 1stedn., PrenticeHall, Englewood

1988
[21]

Kay, S. M. 1993,Fundamentals of Statistical Signal

1993
[22]

2022,” MOMO - V

Komossa, S., Grupe, D., Kraus, A., et al. 2022,” MOMO - V. Effelsberg, Swift, and Fermi study of the blazar and supermassive binary black hole candidate OJ 287 in a period of high activity”, MNRAS, 513, 3165, doi:10.1093/mnras/stac792

work page doi:10.1093/mnras/stac792 2022
[23]

Lomb, N. R. 1976, Ap&SS, 39, 447

1976
[24]

2004,Prog

Machida, M., & Matsumoto, R. 2004,Prog. Theor. Phys. Suppl., 155, 371

2004
[25]

2008,PASJ, 60, 613

Machida, M., & Matsumoto, R. 2008,PASJ, 60, 613

2008
[26]

P., Baganoff, F

Marrone, D. P., Baganoff, F. K., Morris, M. R., et al. 2008, ApJ, 682, 373

2008
[27]

2004, ApJL, 611, L97 arXiv: arXiv:astro-ph/0407252

Miyazaki, A., Tsutsumi, T., & Tsuboi, M. 2004, ApJL, 611, L97 arXiv: arXiv:astro-ph/0407252

work page arXiv 2004
[28]

Oscillation phenomena in the disk around the massive black hole Sagittarius A∗

Miyoshi, M., Shen, Z-Q., Oyama, T., Takahashi, R., Kato Y., PASJ, 63, p1093-1116 (2011). Oscillation phenomena in the disk around the massive black hole Sagittarius A∗

2011
[29]

Miyoshi, M., Kato, Y., & Makino, J. 2024,

2024
[30]

534, Issue 4, Nov

MNRAS, Vol. 534, Issue 4, Nov. 2024 p3237- 3264 https://doi.org/10.1093/mnras/stae1158 arXiv: arXiv:2410.19267 Murchikova and Witzel 2021,ApJL, 920, id.L7, 5 pp. 10.3847/2041-8213/ac2308 10.48550/arXiv.2107.11391 arXiv:2107.11391 2021ApJ...920L...7M

work page doi:10.1093/mnras/stae1158 2024
[31]

2006, ApJ, 643, 1011

Paumard, T., Genzel, R., Martins, F., et al. 2006, ApJ, 643, 1011

2006
[32]

Press, W. H. 1978, Comments on Astrophysics, 7, 103- 109 Bibcode:1978ComAp...7..103P

1978
[33]

Priestley, M. B. 1981,Spectral Analysis and Time Series, 2 vols., Academic Press, London & New

1981
[34]

Reegen, P., 2007, A&A, 467, 1353–1371, doi:10.1051/0004-6361:20066736

work page doi:10.1051/0004-6361:20066736 2007
[35]

J., & Croux, C

Rousseeuw, P. J., & Croux, C. 1993, Journal of the American Statistical Association, 88, 1273

1993
[36]

Scargle, J. D. 1982, ApJ, 263, 835 Schödel, R., Merritt, D., & Eckart, A. 2009, A&A, 502, 91

1982
[37]

R., 1980,Proc

Schwab, F. R., 1980,Proc. Soc. Photo-Opt. Instrum. Eng.231, 18 https://doi.org/10.1117/12.958828

work page doi:10.1117/12.958828 1980
[38]

H., & Stoffer, D

Shumway, R. H., & Stoffer, D. S. 2017,Time Series Analysis and Its Applications: With R Examples, 4th edn., Springer, Cham, Switzerland, ISBN: 978- 3-319-52451-1, doi:10.1007/978-3-319-52452-8

work page doi:10.1007/978-3-319-52452-8 2017
[39]

H., Cordes, J

Simonetti, J. H., Cordes, J. M., & Heeschen, D. S. 1985, ApJ, 296, 46

1985
[40]

2016, PASJ, 68, L7

Tsuboi, M., Kitamura, Y., Miyoshi, M., et al. 2016, PASJ, 68, L7

2016
[41]

2017, ApJL, 850, L5

Tsuboi, M., Kitamura, Y., Tsutsumi, T., et al. 2017, ApJL, 850, L5

2017
[42]

T., 2018, @doi [ ] 10.3847/1538-4365/aab766 , https://ui.adsabs.harvard.edu/abs/2018ApJS..236...16V 236, 16

VanderPlas, J. T., 2018, ApJS, 236(1), 16https:// doi.org/10.3847/1538-4365/aab766

work page Pith review doi:10.3847/1538-4365/aab766 2018
[43]

2005, A&A,431, 391

Vaughan, S. 2005, A&A,431, 391

2005
[44]

2010, MNRAS,402, 307

Vaughan, S. 2010, MNRAS,402, 307

2010
[45]

2022, ApJL, 930, L21

Wielgus, M., Marchili, N., Martí-Vidal, I., et al. 2022, ApJL, 930, L21

2022
[46]

P., Publications of the Astronomical Society of Japan(2025), Vol

Witzel, G., Martinez, G., Hora, J., Willner, S. P., Publications of the Astronomical Society of Japan(2025), Vol. 00, No. 031

2025
[48]

W., Navarro J

Liu, J., Marchili, N., Morris, Mark R., Smith, Howard A., Subroweit, M., Zensus, J. A., ApJ, 917, id.73, 29 pp.(2021) doi:10.3847/1538- 4357/ac0891, 10.48550/arXiv.2011.09582 arXiv: arXiv:2011.09582 Bibcode: 2021ApJ...917...73W

work page doi:10.3847/1538- 2021
[49]

D., et al

Yusef-Zadeh, F., Bushouse, H., Dowell, C. D., et al. 2006, ApJ, 644, 198

2006
[50]

H., Herrnstein, R

Zhao, J.-H., Young, K. H., Herrnstein, R. M., et al. 2003, ApJL, 586, L29

2003
[51]

R., Goss, W

Zhao, J.-H., Morris, M. R., Goss, W. M., & An, T. 2009, ApJ, 699, 186

2009
[52]

Zechmeister, M., & Kürster, M., 2009, A&A, 496, 577–584, doi:10.1051/0004-6361:200811296

work page doi:10.1051/0004-6361:200811296 2009