pith. sign in

arxiv: 2606.07966 · v1 · pith:TFAO2YA7new · submitted 2026-06-06 · 🌌 astro-ph.SR · astro-ph.IM

Modern Time-Series and Spectral Methods for Analyzing Solar and Stellar Oscillatory Signals

Pith reviewed 2026-06-27 19:44 UTC · model grok-4.3

classification 🌌 astro-ph.SR astro-ph.IM
keywords time-series analysissolar oscillationsstellar oscillationsLomb-Scargle periodogramwavelet transformsempirical mode decompositionspectral methods
0
0 comments X

The pith

Synthetic benchmarks of spectral methods yield guidelines for choosing analysis techniques suited to solar and stellar oscillation signals.

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

This review surveys the main techniques used to detect and characterize periodicities in solar and stellar time series that are corrupted by noise and irregular sampling. It covers Fourier-based transforms, the Lomb-Scargle periodogram, wavelet and synchrosqueezed transforms, and empirical mode decomposition, along with statistical tests for significance and their common pitfalls. The central contribution is a set of selection guidelines derived from comparative tests on synthetic data that vary in stationarity, sampling regularity, and noise properties. The paper also sketches future work that would combine Bayesian inference with time-frequency methods to improve handling of non-stationary signals.

Core claim

Comparative tests on synthetic benchmarks demonstrate that method performance depends on signal stationarity, sampling regularity, and noise characteristics, and these results are distilled into practical guidelines for selecting appropriate techniques when analyzing solar and stellar oscillatory signals.

What carries the argument

Comparative analysis of methods on synthetic benchmarks that vary stationarity, sampling regularity, and noise characteristics.

Load-bearing premise

Performance differences measured on the chosen synthetic benchmarks will carry over directly to real solar and stellar observations that contain additional unmodeled effects such as instrumental artifacts.

What would settle it

Apply the recommended method-selection rules to a collection of real solar or stellar time series whose oscillation frequencies and amplitudes are already known from independent measurements, and verify whether the guidelines produce more accurate recoveries than the non-recommended alternatives.

Figures

Figures reproduced from arXiv: 2606.07966 by Feng Song, Yuan Ding.

Figure 1
Figure 1. Figure 1: Examples of rotationally variable stars. The horizontal axis represents frequency (in cycles per day, c/d), while the vertical axis shows brightness amplitude (in millimagnitudes, mmag). The TIC numbers in each panel correspond to the TESS Input Catalog identifiers. The main panels display the frequency spectra, and the insets illustrate the light curves over 5 days (upper four panels) and 10 days (bottom … view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of the Lomb–Scargle periodogram (LSP) analysis applied to solar coronal loop oscillations. Panels (a), (c), and (e) show the time series of loop displacement (ξ), normalized emission flux (F/⟨F⟩), and loop width (w), respectively, measured at s = 0.5L0 in Event V. Panels (b), (d), and (f) display the corresponding Lomb– Scargle power spectra. The dashed lines indicate the 0.05 false alarm prob… view at source ↗
Figure 3
Figure 3. Figure 3: Demonstration of the continuous wavelet transform (CWT) applied to a stochastic finite-lifetime transient model. Panel (a) shows the synthetic time series, while panel (b) presents the corresponding wavelet power spectrum. The wavelet spectrum reveals quasi-periodic oscillations in the 10–30 minute range, enclosed by the green dashed lines. The cross-hatched region marks the cone of influence, within which… view at source ↗
Figure 4
Figure 4. Figure 4: Left column: Synchrosqueezed Wavelet Transform (SWT) power spectra for different wavelength channels (171 Å, 304 Å, Hα, 1600 Å, 1700 Å, and TiO). The dashed black curves indicate the cone of influence (COI) where edge effects become significant. Right column: corresponding reconstructed temporal components obtained from inverse SWT, revealing coherent oscillatory behavior at multiple atmospheric heights. I… view at source ↗
Figure 5
Figure 5. Figure 5: Empirical Mode Decomposition (EMD) of the mean galactic cosmic-ray signal. The top panel (x0) shows the original signal, while the subsequent panels display intrinsic mode functions (IMFs) from IMF1 to IMF5, representing progressively lower-frequency oscillatory components. The bottom panel illustrates the final residual trend (Res.). The EMD adaptively separates the signal into physically interpretable mo… view at source ↗
Figure 6
Figure 6. Figure 6: Lomb–Scargle periodograms (LSPs) illustrating the effects of iterative peak suppression at different frequencies. Columns from left to right show the results for Nobeyama Radioheliograph data at 17 GHz and Nobeyama Radiopolarimeter data at 9 GHz, 17 GHz, and 35 GHz, respectively. In all cases, the same three significant peaks are consistently detected above the 99% confidence level, confirming the robustne… view at source ↗
Figure 7
Figure 7. Figure 7: Illustration of spurious periodicities introduced by detrending a red-noise time series. The top panel shows the original synthetic data (gray) that follow a f−2 power-law spectrum, and the detrended version (magenta) obtained with a running boxcar filter of width δt. The middle and bottom panels display the corresponding wavelet and Fourier power spectra, respectively, demonstrating that detrending artifi… view at source ↗
Figure 8
Figure 8. Figure 8: Illustration of power-law noise modeling using a Bayesian MCMC framework. Panel (a) shows a synthetic signal composed of colored and white noise with a short-lived oscillation burst. Panels (b) and (c) display the power spectral density (PSD) fitted by single-component (M0) and double-component (M1) power-law models, respectively. The dashed red line marks the 95% confidence level, while the black dashed l… view at source ↗
Figure 9
Figure 9. Figure 9: A synthetic solar-like signal: (a) A 50 s oscillation emerging between 600 and 1600 s; (b) An 80 s oscillation emerging between 200 and 1200 s; (c) Red and white noise; (d) The synthetic signal containing both oscillatory components and mixed noise. This synthetic signal mimics the non-stationary, multi-periodic, and noise-contaminated nature of solar and stellar oscillatory signals. It serves as a benchma… view at source ↗
Figure 10
Figure 10. Figure 10: FFT spectra of the synthetic signal: (a) without windowing; (b) with Hann window; (c) Bayesian MCMC estimation of the FFT spectrum and red-noise background [27]. The FFT reveals the global frequency content of the signal. Windowing reduces spectral leakage at the cost of frequency resolution. The Bayesian MCMC approach provides a statistically rigorous estimation of the red-noise background and the signif… view at source ↗
Figure 11
Figure 11. Figure 11: Continuous wavelet transform (CWT) analysis of the synthetic signal: (a) CWT spectrum without detrending; (b) CWT significance levels estimated with an AR(1) red-noise background following [14]. The CWT reveals the time￾frequency evolution of oscillatory power. Without detrending, the red-noise background dominates the low-frequency power, masking the oscillations. Detrending enhances the visibility of os… view at source ↗
Figure 12
Figure 12. Figure 12: Adaptive decomposition of the synthetic signal: (a) HHT based on EMD showing three intrinsic mode functions (IMFs); (b) SWT decomposition with two reconstructed oscillatory components showing higher spectral concentration. EMD adaptively decomposes the signal into oscillatory modes without a predefined basis, but it is sensitive to noise and may suffer from mode mixing. SWT sharpens the time-frequency rep… view at source ↗
read the original abstract

Time-series analysis plays a central role in understanding oscillatory and wave phenomena in solar and stellar atmospheres. However, astrophysical observations are inherently affected by instrumental noise, non-stationary dynamics, and uneven sampling. This review provides a comprehensive overview and comparative analysis of principal methods for detecting and characterizing periodicities in solar and stellar signals. We cover Fourier-transform-based transforms, nonlinear-fitting-based methods (Lomb--Scargle periodogram), time-frequency methods (wavelet and synchrosqueezed transforms), and adaptive decomposition techniques (Empirical Mode Decomposition). Advanced statistical significance tests, including false-alarm probability, autoregressive models, and Bayesian Markov Chain Monte Carlo (MCMC) approaches, are discussed their practical limitations and misuse risks. Through comparative analysis using synthetic benchmarks, we provide guidelines for selecting methods based on signal stationarity, sampling regularity, and noise characteristics. Finally, we outline future directions that integrate Bayesian inference with time-frequency analysis to achieve both statistical rigor and temporal localization in studying non-stationary solar and stellar oscillations.

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 / 1 minor

Summary. The manuscript reviews principal methods for detecting and characterizing periodicities in solar and stellar oscillatory signals, including Fourier-transform-based approaches, the Lomb-Scargle periodogram, wavelet and synchrosqueezed transforms, Empirical Mode Decomposition, and statistical significance tests (false-alarm probability, autoregressive models, Bayesian MCMC). It presents a comparative analysis on synthetic benchmarks to derive guidelines for method selection according to signal stationarity, sampling regularity, and noise characteristics, and outlines future directions integrating Bayesian inference with time-frequency methods.

Significance. A rigorously supported set of selection guidelines grounded in quantitative synthetic benchmarks could offer practical utility to the solar/stellar community for handling non-stationary and unevenly sampled data. The review nature of the work limits its novelty, but clear documentation of benchmark performance metrics and explicit discussion of generalization limits would strengthen its value as a reference.

major comments (2)
  1. [Abstract / comparative analysis] Abstract and the comparative-analysis section: the central claim that actionable guidelines are provided 'through comparative analysis using synthetic benchmarks' is load-bearing for the paper's contribution, yet no quantitative results (detection rates, false-alarm rates, error metrics), specific benchmark parameters, or tables of performance comparisons are described; without these the guidelines cannot be evaluated.
  2. [Guidelines / future directions] Guidelines section: the assumption that relative performance differences observed on the synthetic suite will hold for real solar/stellar observations is not tested or qualified; real data routinely contain additional effects (instrumental systematics, irregular gaps, correlated noise, multi-scale non-stationarity) whose impact on method ranking is left unaddressed.
minor comments (1)
  1. The abstract states that statistical tests are discussed 'their practical limitations and misuse risks' but the corresponding section should explicitly balance limitations for every method covered, not only the significance tests.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. The comments highlight opportunities to strengthen the quantitative presentation of our benchmarks and to qualify the scope of the derived guidelines. We address each point below and will incorporate revisions in the next version of the manuscript.

read point-by-point responses
  1. Referee: [Abstract / comparative analysis] Abstract and the comparative-analysis section: the central claim that actionable guidelines are provided 'through comparative analysis using synthetic benchmarks' is load-bearing for the paper's contribution, yet no quantitative results (detection rates, false-alarm rates, error metrics), specific benchmark parameters, or tables of performance comparisons are described; without these the guidelines cannot be evaluated.

    Authors: We acknowledge that the current manuscript presents the comparative analysis and resulting guidelines in descriptive form without tabulating explicit performance metrics or benchmark parameters. This limits the ability of readers to independently evaluate the strength of the claims. In revision we will add a new table (and accompanying text) that reports the specific synthetic benchmark parameters together with quantitative metrics such as detection rates, false-alarm rates, and reconstruction error for each method under the tested conditions of stationarity, sampling regularity, and noise level. revision: yes

  2. Referee: [Guidelines / future directions] Guidelines section: the assumption that relative performance differences observed on the synthetic suite will hold for real solar/stellar observations is not tested or qualified; real data routinely contain additional effects (instrumental systematics, irregular gaps, correlated noise, multi-scale non-stationarity) whose impact on method ranking is left unaddressed.

    Authors: The referee is correct that the manuscript does not explicitly test or discuss how additional real-data complications (instrumental systematics, irregular gaps beyond the synthetic sampling, correlated noise, or multi-scale non-stationarity) might alter the relative performance rankings obtained from the controlled synthetic suite. We will revise the guidelines section to include a dedicated paragraph that qualifies the applicability of the synthetic results, notes these untested effects, and recommends case-by-case validation when applying the guidelines to actual observations. revision: yes

Circularity Check

0 steps flagged

No circularity; review and benchmark comparison with no load-bearing derivations or self-referential predictions

full rationale

The paper is explicitly a review synthesizing existing methods (Fourier, Lomb-Scargle, wavelets, EMD, statistical tests) and reporting comparative results on synthetic benchmarks to produce selection guidelines. No equations, parameter fits, uniqueness theorems, or predictions are derived that reduce to the paper's own inputs by construction. The benchmark-based guidelines are an empirical output, not a closed-loop renaming or self-definition. Self-citations, if present, are not load-bearing for any central claim. This matches the default non-circular case for a synthesis paper.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

As a review paper the contribution is synthesis of existing techniques rather than new postulates or derivations; no free parameters, axioms, or invented entities are introduced in the abstract.

pith-pipeline@v0.9.1-grok · 5704 in / 1106 out tokens · 13462 ms · 2026-06-27T19:44:02.175929+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

67 extracted references · 53 canonical work pages · 16 internal anchors

  1. [1]

    Stochastic transients as a source of quasi-periodic processes in the solar atmosphere

    doi:10.3847/0067-0049/224/2/30 , eid =. arXiv , author =:1603.08977 , journal =

  2. [2]

    Philosophical Transactions of the Royal Society of London Series A , keywords =

    Philosophical Transactions of the Royal Society of London Series A , keywords =. doi:10.1098/rsta.2011.0640 , eprint =

  3. [3]

    doi:10.1007/s11207-015-0691-2 , eprint =

    , keywords =. doi:10.1007/s11207-015-0691-2 , eprint =

  4. [4]

    Significance testing for quasi-periodic pulsations in solar and stellar flares

    doi:10.1051/0004-6361/201730595 , eid =. arXiv , author =:1703.07294 , journal =

  5. [5]

    Ensemble empirical mode decomposition: a noise-assisted data analysis method , volume =

    Wu, Zhaohua and Huang, Norden E , journal =. Ensemble empirical mode decomposition: a noise-assisted data analysis method , volume =

  6. [6]

    Multivariate empirical mode decomposition , volume =

    Rehman, Naveed and Mandic, Danilo P , journal =. Multivariate empirical mode decomposition , volume =

  7. [7]

    A complete ensemble empirical mode decomposition with adaptive noise , year =

    Torres, Mar. A complete ensemble empirical mode decomposition with adaptive noise , year =. 2011 IEEE international conference on acoustics, speech and signal processing (ICASSP) , organization =

  8. [8]

    arXiv , author =:2406.12578 , journal =

    doi:10.1051/0004-6361/202450711 , eid =. arXiv , author =:2406.12578 , journal =

  9. [9]

    Forward Modelling of Standing Kink Modes in Coronal Loops II. Applications

    doi:10.3847/0067-0049/223/2/24 , eid =. arXiv , author =:1602.07598 , journal =

  10. [10]

    arXiv , author =:2403.10392 , journal =

    doi:10.3847/1538-4357/ad36ca , eid =. arXiv , author =:2403.10392 , journal =

  11. [11]

    doi:10.1007/s11214-009-9536-3 , journal =

  12. [12]

    doi:10.1007/BF00733425 , journal =

  13. [13]

    doi:10.1088/2041-8205/713/2/L79 , eprint =

    , keywords =. doi:10.1088/2041-8205/713/2/L79 , eprint =

  14. [14]

    doi:10.1051/0004-6361:200810473 , journal =

  15. [15]

    An algorithm for the machine calculation of complex Fourier series , volume =

    Cooley, James W and Tukey, John W , journal =. An algorithm for the machine calculation of complex Fourier series , volume =

  16. [16]

    doi:10.1086/160554 , journal =

  17. [17]

    , eprint =

    , keywords =. doi:10.1051/0004-6361:20041453 , eprint =

  18. [18]

    , keywords =

    , keywords =. doi:10.1111/j.1365-2966.2009.15868.x , eprint =

  19. [19]

    arXiv , author =:2208.02493 , journal =

    doi:10.3847/1538-4357/ac85ba , eid =. arXiv , author =:2208.02493 , journal =

  20. [20]

    doi:10.1007/BF00648343 , journal =

  21. [21]

    doi:10.1086/112924 , journal =

  22. [22]
  23. [23]

    doi:10.1093/mnras/sty2469 , eprint =

    , keywords =. doi:10.1093/mnras/sty2469 , eprint =

  24. [24]

    Oscillations and waves in sunspots , volume =

    Khomenko, Elena and Collados, Manuel , journal =. Oscillations and waves in sunspots , volume =

  25. [25]

    Pulsation and Mass Loss in Stars , doi =

  26. [26]

    Publications de l'Observatoire Astronomique de Beograd , keywords =

  27. [27]

    doi:10.1007/s11207-016-0977-z , eprint =

    , keywords =. doi:10.1007/s11207-016-0977-z , eprint =

  28. [28]

    doi:10.1038/304689a0 , journal =

  29. [29]

    doi:10.1146/annurev.aa.22.090184.003113 , journal =

  30. [30]

    Challenges to Theories of the Structure of Moderate-Mass Stars , doi =

  31. [31]

    Global Seismology of the Sun

    doi:10.1007/s41116-016-0003-4 , eid =. arXiv , author =:1606.07071 , journal =

  32. [32]

    QPOs in the Time Domain: An Autocorrelation Analysis

    doi:10.1051/0004-6361/201014736 , eid =. arXiv , author =:1009.2465 , journal =

  33. [33]

    Automatic Autocorrelation and Spectral Analysis , doi =

    Estimation of Time Series Models , url =. Automatic Autocorrelation and Spectral Analysis , doi =. 2006 , bdsk-url-1 =

  34. [34]

    doi:10.3847/1538-4357/acb34f , eid =

    , keywords =. doi:10.3847/1538-4357/acb34f , eid =

  35. [35]

    doi:10.3847/1538-4357/ad4a58 , eid =

    , keywords =. doi:10.3847/1538-4357/ad4a58 , eid =

  36. [36]

    arXiv , author =:2002.09816 , journal =

    doi:10.1007/s10509-020-3744-4 , eid =. arXiv , author =:2002.09816 , journal =

  37. [37]

    arXiv , author =:1911.05217 , journal =

    doi:10.3847/2041-8213/ab5648 , eid =. arXiv , author =:1911.05217 , journal =

  38. [38]

    High-frequency Oscillations in the Atmosphere above a Sunspot Umbra

    doi:10.3847/2041-8213/aab4f3 , eid =. arXiv , author =:1803.09046 , journal =

  39. [39]
  40. [40]

    doi:10.1175/1520-0477(1998)079<0061:APGTWA>2.0.CO;2 , journal =

  41. [41]

    doi:10.1098/rspa.1998.0193 , journal =

  42. [42]

    Synchrosqueezed wavelet transforms: An empirical mode decomposition-like tool , volume =

    Daubechies, Ingrid and Lu, Jianfeng and Wu, Hau-Tieng , journal =. Synchrosqueezed wavelet transforms: An empirical mode decomposition-like tool , volume =

  43. [43]

    arXiv , author =:2101.11998 , journal =

    doi:10.1088/1674-4527/21/7/179 , eid =. arXiv , author =:2101.11998 , journal =

  44. [44]

    AAS/Solar Physics Division Abstracts \#48 , eid =

  45. [45]
  46. [46]

    arXiv , author =:2110.14257 , journal =

    doi:10.1051/0004-6361/202142028 , eid =. arXiv , author =:2110.14257 , journal =

  47. [47]

    arXiv , author =:2010.10851 , journal =

    doi:10.3847/1538-4357/abb5b7 , eid =. arXiv , author =:2010.10851 , journal =

  48. [48]

    doi:10.1093/mnras/staa1061 , eprint =

    , keywords =. doi:10.1093/mnras/staa1061 , eprint =

  49. [49]

    ConceFT: Concentration of frequency and time via a multitapered synchrosqueezed transform , volume =

    Daubechies, Ingrid and Wang, Yi and Wu, Hau-tieng , journal =. ConceFT: Concentration of frequency and time via a multitapered synchrosqueezed transform , volume =

  50. [50]

    Oscillatory processes in solar flares

    doi:10.1088/0741-3335/52/12/124009 , eid =. arXiv , author =:1010.0063 , journal =

  51. [51]

    doi:10.1007/s11207-010-9701-6 , journal =

  52. [52]

    On standing sausage waves in photospheric magnetic waveguides

    doi:10.1051/0004-6361/201220542 , eid =. arXiv , author =:1210.6476 , journal =

  53. [53]

    The Astrophysical Journal , keywords =

    , keywords =. doi:10.3847/0004-637X/830/2/110 , eid =

  54. [54]

    A study of the characteristics of white noise using the empirical mode decomposition method , volume =

    Wu, Zhaohua and Huang, Norden E , journal =. A study of the characteristics of white noise using the empirical mode decomposition method , volume =

  55. [55]

    doi:10.3390/universe9020092 , eid =

    Universe , keywords =. doi:10.3390/universe9020092 , eid =

  56. [56]

    doi:10.1007/s11207-010-9533-4 , journal =

  57. [57]

    arXiv , author =:2108.03781 , journal =

    doi:10.3847/1538-4357/ac1c05 , eid =. arXiv , author =:2108.03781 , journal =

  58. [58]

    doi:10.1126/science.1168680 , eprint =

    Science , keywords =. doi:10.1126/science.1168680 , eprint =

  59. [59]

    doi:10.1051/0004-6361:20035666 , journal =

  60. [60]

    A New Analysis Procedure for Detecting Periodicities within Complex Solar Coronal Arcades

    doi:10.3847/1538-4357/ab264c , eid =. arXiv , author =:1902.06644 , journal =

  61. [61]

    doi:10.1111/j.1365-2966.2006.10396.x , journal =

  62. [62]

    The global oscillation network group (GONG) project , volume =

    Harvey, JW and Hill, F and Hubbard, RP and Kennedy, JR and Leibacher, JW and Pintar, JA and Gilman, PA and Noyes, RW and Title, AM and Toomre, J and others , journal =. The global oscillation network group (GONG) project , volume =

  63. [63]

    Determining global parameters of the oscillations of solar-like stars

    doi:10.1051/0004-6361/200913266 , eid =. arXiv , author =:0912.3367 , journal =

  64. [64]
  65. [65]

    Coronal Fourier power spectra: implications for coronal seismology and coronal heating

    doi:10.1088/0004-637X/798/1/1 , eid =. arXiv , author =:1410.2171 , journal =

  66. [66]

    Modelling Quasi-Periodic Pulsations in Solar and Stellar Flares

    doi:10.1007/s11214-018-0478-5 , eid =. arXiv , author =:1802.04180 , journal =

  67. [67]

    On The Fourier And Wavelet Analysis Of Coronal Time Series

    doi:10.3847/0004-637X/825/2/110 , eid =. arXiv , author =:1606.05251 , journal =