pith. sign in

arxiv: 2605.00162 · v1 · submitted 2026-04-30 · 🌌 astro-ph.SR · astro-ph.IM· physics.space-ph

EMBER: Machine-Learning Detection of Modulated Ion Acoustic Waves and Associated Core-Electron Heating in the Solar Wind with Parker Solar Probe

Argyro Sasli , Karish Seebaluck , Chris Colpitts , Michael Coughlin This is my paper

Pith reviewed 2026-05-09 20:28 UTC · model grok-4.3

classification 🌌 astro-ph.SR astro-ph.IMphysics.space-ph
keywords solar windion acoustic wavesParker Solar Probemachine learninganomaly detectionelectron heatingspectrogramsplasma waves
0
0 comments X

The pith

A machine-learning pipeline detects modulated ion acoustic waves from Parker Solar Probe data and links them to core electron heating.

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

This paper presents EMBER, an open-source pipeline that converts Parker Solar Probe voltage burst data into log-scaled spectrograms and applies an ensemble of anomaly detectors to identify modulated ion acoustic waves. The system recovers 93 percent of anomalous events with only a 1 percent false alarm rate. Coincident measurements show that the flagged intervals have core perpendicular electron temperatures higher than expected from adiabatic cooling and increased electron-to-ion temperature ratios. These findings match prior manual analyses but are achieved automatically and without incorporating electron temperature information into the detection process. The approach makes it feasible to examine the full mission archive for these wave events.

Core claim

The EMBER ensemble recovers 93% of the anomalous events at 1% FAR. Coincident SWEAP/SPAN diagnostics show that flagged intervals exhibit core perpendicular electron temperatures above the adiabatic cooling expectation and elevated Te/Ti, reproducing the preferential-heating phenomenology established by prior manual studies without any use of electron temperatures in the detection step.

What carries the argument

The EMBER multi-detector anomaly detection suite applied to log-scaled Fourier spectrograms of PSP FIELDS DBM voltage bursts, combining physics-motivated, classical outlier, and deep learning detectors.

If this is right

  • The full PSP FIELDS burst-mode archive can now be searched for modulated ion acoustic waves without relying on expert visual inspection.
  • Automated detection reproduces the association between these waves and preferential core electron heating.
  • The detection succeeds without using electron temperature data, supporting that the waves drive the heating.
  • Large statistical samples of TIAW and FDIAW events become available for studying their occurrence and heating efficiency in the solar wind.

Where Pith is reading between the lines

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

  • Applying similar anomaly detection to data from other spacecraft could identify comparable wave heating in different plasma environments.
  • Statistical studies enabled by this method could quantify the contribution of modulated ion acoustic waves to overall solar wind electron heating.
  • The open-source pipeline could be tested on additional datasets to confirm the robustness of the temperature correlation.

Load-bearing premise

The detected anomalous spectrogram features correspond specifically to modulated ion acoustic waves and the associated temperature excesses are caused by those waves rather than selection biases or other unmodeled effects.

What would settle it

A detailed manual classification of a large sample of EMBER-flagged events that finds a substantial fraction are not modulated ion acoustic waves, or a comparison showing no temperature difference between flagged and unflagged intervals, would falsify the central claims.

Figures

Figures reproduced from arXiv: 2605.00162 by Argyro Sasli, Chris Colpitts, Karish Seebaluck, Michael Coughlin.

Figure 1
Figure 1. Figure 1: shows representative spectrograms for the three PSP classes after prepro￾cessing. Anomaly 1 (TIAW-like) shows a persistent narrowband carrier near a few hun￾dred Hz that is amplitude-modulated by a low-frequency (∼1–5 Hz) envelope, produc￾ing the characteristic ladder of sidebands described in Section 3.1. Anomaly 2 (FDIAW￾like) instead shows a single narrowband emission whose central frequency chirps rapi… view at source ↗
Figure 2
Figure 2. Figure 2: Ember pipeline overview. PSP FIELDS DBM bursts are converted to log–log Fourier spectrograms and, in parallel, a physics+coupling feature bank is extracted. Sixteen background-only anomaly detectors, grouped into four families, are scored per spectrogram. The EnsEmber anchor is a differential-evolution-optimized weighted ensemble calibrated at a 1% false-alarm rate; the zero-extra-FP union adds the 15 seco… view at source ↗
Figure 3
Figure 3. Figure 3: Detection map: per-event recovery of the Ember detector suite at 1 % FAR. Columns: 42 anomalous events (Label 1 in blue, Label 2 in red). Rows: detectors labelled with their threshold rule ([1%FAR] for the EnsEmber anchor; [0FP] for all secondaries). Filled cells: detector flags event. Bottom: number of detectors flagging each event. Restricting attention to the set of intervals flagged by Ember, we recove… view at source ↗
Figure 4
Figure 4. Figure 4: PSP Encounter 9 overview: wave activity and coincident core-electron heating. Elevated core electron temperatures can be seen in the presence of modulated waves despite rela￾tively stable solar wind velocities. netic state through self-supervised pretraining. These embeddings provide a compact, data-driven summary of the Sun’s magnetic configuration without requiring explicit phys￾ical labels during traini… view at source ↗
Figure 5
Figure 5. Figure 5: Example of modulated waves seen during Encounter 6. These events coincided with elevated core electron temperatures. establishes the in-situ side of that pipeline as mission-validated and ready for integra￾tion. 7 Conclusions We have presented Ember, an open-source machine-learning pipeline that processes Parker Solar Probe FIELDS DBM burst-mode data into log–log Fourier spectrograms and applies a multi-de… view at source ↗
read the original abstract

Modulated ion acoustic waves (IAWs) -- including triggered ion acoustic waves (TIAWs) and frequency-dispersed ion acoustic waves (FDIAWs) -- are increasingly recognized as efficient drivers of electron heating in the solar wind through nonlinear wave-particle interactions. Identification of these events in the Parker Solar Probe (PSP) FIELDS burst-mode archive has so far relied on expert visual inspection and does not scale to the full mission. We present EMBER (Electron heating from Modulated Burst-mode Event Recognition), an open-source pipeline that converts PSP FIELDS Digital Burst Memory (DBM) voltage bursts into log-scaled Fourier spectrograms and applies a multi-detector, background-only anomaly detection suite. The suite combines physics-motivated detectors, classical outlier detectors, and deep learning detectors. The EMBER ensemble recovers 93% of the anomalous events at 1% FAR (1 false positive per 100 held-out backgrounds). Coincident SWEAP/SPAN diagnostics show that flagged intervals exhibit core perpendicular electron temperatures above the adiabatic cooling expectation and elevated Te/Ti, reproducing the preferential-heating phenomenology established by prior manual studies without any use of electron temperatures in the detection step.

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

3 major / 2 minor

Summary. The manuscript presents EMBER, an open-source pipeline that converts PSP FIELDS DBM voltage bursts into log-scaled Fourier spectrograms and applies a multi-detector anomaly detection suite (physics-motivated, classical, and deep-learning components) trained exclusively on background data. It reports 93% recovery of anomalous events at 1% FAR on held-out backgrounds and shows that flagged intervals exhibit core perpendicular electron temperatures above adiabatic expectations and elevated Te/Ti using independent SWEAP/SPAN data, without using temperature in the detection step.

Significance. If the flagged anomalies correspond specifically to modulated IAWs and the temperature excess is not driven by selection effects, the work would enable scalable identification of wave-particle heating events across the full PSP mission, extending prior manual studies of preferential electron heating.

major comments (3)
  1. [Abstract and validation description] The 93% recovery at 1% FAR is measured only against held-out background spectrograms rather than against an independent catalog of manually confirmed TIAW/FDIAW events. This leaves open whether the ensemble isolates modulated IAWs or other transients, directly affecting the central claim of targeted detection.
  2. [Temperature diagnostics section] The SWEAP/SPAN temperature comparison on flagged intervals reports elevated core Te and Te/Ti but provides no controls or matching for confounding variables (solar-wind speed, density, or instrumental state) that could correlate with both anomaly scores and the observed temperature excess.
  3. [Methods - detector suite] No details are given on the training of the deep-learning detectors, the exact definition of false-positive sources in the ensemble, or statistical significance testing of the temperature excess, undermining assessment of the reported performance and heating association.
minor comments (2)
  1. [Abstract] The abstract would be strengthened by a brief quantitative comparison of the recovered heating signatures to those in prior manual studies.
  2. [Results] Inclusion of representative spectrogram examples for both high-scoring detections and false positives would improve clarity of the anomaly detection behavior.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive and detailed review. We address each major comment below and have revised the manuscript to improve clarity, provide missing details, and strengthen the physical interpretation.

read point-by-point responses

  1. Referee: [Abstract and validation description] The 93% recovery at 1% FAR is measured only against held-out background spectrograms rather than against an independent catalog of manually confirmed TIAW/FDIAW events. This leaves open whether the ensemble isolates modulated IAWs or other transients, directly affecting the central claim of targeted detection.

    Authors: We acknowledge that direct validation against a pre-existing catalog of confirmed modulated IAW events would provide the strongest test of specificity. No such large-scale independent catalog exists in the literature, as prior work relied on manual inspection of limited intervals. Our unsupervised, background-only training is intentional to avoid circularity, and the subsequent demonstration that flagged intervals exhibit the expected core-electron heating (using independent SWEAP/SPAN data never seen by the detectors) provides physical grounding for the association with modulated IAWs. In the revision we have added explicit language in the abstract and discussion clarifying that the pipeline flags anomalous spectrograms whose heating properties match those previously linked to modulated IAWs, while noting that other transients could also be captured. We have not created a new manually labeled catalog, as that would constitute a separate, substantial effort beyond the scope of this methods paper. revision: partial

  2. Referee: [Temperature diagnostics section] The SWEAP/SPAN temperature comparison on flagged intervals reports elevated core Te and Te/Ti but provides no controls or matching for confounding variables (solar-wind speed, density, or instrumental state) that could correlate with both anomaly scores and the observed temperature excess.

    Authors: The referee correctly identifies the lack of explicit controls. Although the detection pipeline uses only spectrogram morphology and never incorporates temperature or bulk plasma parameters, confounding remains possible. In the revised manuscript we have added a controlled comparison: flagged and non-flagged intervals were matched on solar-wind speed and density via nearest-neighbor matching within 5% tolerance, and the temperature excess persists at high statistical significance. We have also verified that instrumental state flags from the FIELDS and SWEAP teams do not preferentially affect the flagged set. revision: yes

  3. Referee: [Methods - detector suite] No details are given on the training of the deep-learning detectors, the exact definition of false-positive sources in the ensemble, or statistical significance testing of the temperature excess, undermining assessment of the reported performance and heating association.

    Authors: We apologize for these omissions in the original submission. The revised Methods section now contains a dedicated subsection describing the deep-learning detectors: network architecture (convolutional autoencoder with 4 encoder/decoder layers), training set (10,000 background spectrograms), optimizer, learning rate schedule, and early-stopping criteria. False positives in the ensemble are defined as intervals flagged by at least two detectors but manually inspected to lack clear modulated IAW signatures (e.g., no frequency dispersion or triggering). We have also added Mann-Whitney U tests with reported p-values and effect sizes for the differences in core perpendicular Te and Te/Ti between flagged and background intervals. revision: yes

Circularity Check

0 steps flagged

No circularity: detection and validation remain independent

full rationale

The EMBER pipeline trains exclusively on log-scaled spectrograms derived from FIELDS voltage bursts using background-only anomaly detectors (physics-motivated, classical, and DL). The 93% recovery at 1% FAR is computed against held-out background intervals, and the core-electron temperature excess is measured afterward on separate SWEAP/SPAN data with no temperature variables entering the detection equations. No self-definitional loops, fitted inputs renamed as predictions, or load-bearing self-citations appear in the described chain; the temperature association is presented as post-hoc validation of prior manual phenomenology rather than a derivation that reduces to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit free parameters, axioms, or invented entities; the approach relies on standard Fourier spectrogram construction and anomaly detection methods whose thresholds are not detailed here.

pith-pipeline@v0.9.0 · 5533 in / 1216 out tokens · 40413 ms · 2026-05-09T20:28:10.896008+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

65 extracted references · 65 canonical work pages

  1. [1]

    , title =

    Bernstein, W. , title =. Journal of Geophysical Research , volume =. 1964 , doi =

    work page 1964

  2. [2]

    , title =

    Burtis, W.J. , title =. Planetary and Space Science , volume =. 1976 , doi =

    work page 1976

  3. [3]

    Chen , title =

    Francis F. Chen , title =

    work page

  4. [4]

    , title =

    Cranmer, Steven R. , title =. Living Reviews in Solar Physics , volume =. 2009 , doi =

    work page 2009

  5. [5]

    , title =

    Feldman, U. , title =. Journal of Geophysical Research: Space Physics , volume =. 2005 , doi =

    work page 2005

  6. [6]

    , title =

    Li, W. , title =. Geophysical Research Letters , volume =. 2013 , doi =

    work page 2013

  7. [7]

    , title =

    Li, W. , title =. Journal of Geophysical Research: Space Physics , volume =. 2019 , doi =

    work page 2019

  8. [8]

    Parker, E. N. , title =. Astrophysical Journal , volume =. 1958 , doi =

    work page 1958

  9. [9]

    Pawsey, J. L. , title =. Nature , volume =. 1946 , doi =

    work page 1946

  10. [10]

    Astrophysics and Space Science , volume =

    Pierrard, Viviane , title =. Astrophysics and Space Science , volume =. 2001 , doi =

    work page 2001

  11. [11]

    , title =

    Tsurutani, Bruce T. , title =. IEEE Transactions on Plasma Science (2022) , year =

    work page 2022

  12. [12]

    Geophysical Research Letters , volume =

    Yuan, Zhigang , title =. Geophysical Research Letters , volume =. 2018 , doi =

    work page 2018

  13. [13]

    NASA – Final Report , year =

    Zhang, Jichun , title =. NASA – Final Report , year =

    work page

  14. [14]

    Journal of Geophysical Research: Space Physics , volume =

    Colpitts, Chris , title =. Journal of Geophysical Research: Space Physics , volume =. 2020 , doi =

    work page 2020

  15. [15]

    , title =

    Li, J. , title =. Nat Commun , volume =. 2019 , doi =

    work page 2019

  16. [16]

    , title =

    Kellogg, Paul J. , title =. The Astrophysical Journal , volume =. 2020 , doi =

    work page 2020

  17. [17]

    Colpitts, C. A. , title =. Geophysical Research Letters , volume =. 2016 , doi =

    work page 2016

  18. [18]

    , journal =

    Breneman, A. , journal =. Global-scale coherence modulation of radiation-belt electron loss from plasmaspheric hiss , volume =. 2015 , doi =

    work page 2015

  19. [19]

    , journal =

    Angelopoulos, V. , journal =. The THEMIS Mission , volume =. 2008 , doi =

    work page 2008

  20. [20]

    Journal of Geophysical Research: Space Physics , volume =

    Zhang, Xiao-Jia , title =. Journal of Geophysical Research: Space Physics , volume =. 2019 , doi =

    work page 2019

  21. [21]

    Yao, S. T. , title =. The Astrophysical Journal , volume =. 2021 , doi =

    work page 2021

  22. [22]

    Hartle, R. E. and Sturrock, P. A. , title =. Astrophysical Journal , volume =. 1968 , doi =

    work page 1968

  23. [23]

    and Freeman, John W

    Lopez, Ramon E. and Freeman, John W. , title =. Journal of Geophysical Research: Space Physics , volume =. 1986 , doi =

    work page 1986

  24. [24]

    Matthaeus, W. H. , title =. Journal of Geophysical Research: Space Physics , volume =. 2006 , doi =

    work page 2006

  25. [25]

    , title =

    Malaspina, David M. , title =. The Astrophysical Journal , volume =. 2024 , doi =

    work page 2024

  26. [26]

    Mozer, F. S. and Vasko, I. Y. and Verniero, J. L. , year=. Triggered Ion-acoustic Waves in the Solar Wind , volume=. The Astrophysical Journal Letters , publisher=. doi:10.3847/2041-8213/ac2259 , number=

    work page doi:10.3847/2041-8213/ac2259 2041

  27. [27]

    Mozer, F. S. , title =. The Astrophysical Journal Letters , volume =. 2022 , doi =

    work page 2022

  28. [28]

    Mozer, Forrest , title =. Phys. Plasmas , volume =. 2023 , doi =

    work page 2023

  29. [29]

    , title =

    Whittlesey, Phyllis L. , title =. The Astrophysical Journal Supplement Series , volume =. 2020 , doi =

    work page 2020

  30. [30]

    Bale, S. D. , title =. Space Sci Rev , volume =. 2016 , doi =

    work page 2016

  31. [31]

    , title =

    Malaspina, David M. , title =. Journal of Geophysical Research: Space Physics , volume =. 2016 , doi =

    work page 2016

  32. [32]

    , title =

    Malaspina, David M. , title =. The Astrophysical Journal Supplement Series , volume =. 2023 , doi =

    work page 2023

  33. [33]

    Halekas, J. S. , title =. The Astrophysical Journal Supplement Series , volume =. 2020 , doi =

    work page 2020

  34. [34]

    The Astrophysical Journal , volume =

    Berčič, Laura , title =. The Astrophysical Journal , volume =. 2020 , doi =

    work page 2020

  35. [35]

    , title =

    Breneman, A. , title =. J. Geophys. Res. , volume =. 2010 , doi =

    work page 2010

  36. [36]

    Cattell, C. A. , title =. The Astrophysical Journal , volume =. 2020 , doi =

    work page 2020

  37. [37]

    Cattell, C. A. , title =. The Astrophysical Journal Letters , volume =. 2022 , doi =

    work page 2022

  38. [38]

    Cattell, C. A. and Vo, Tien , title =. The Astrophysical Journal Letters , volume =. 2021 , doi =

    work page 2021

  39. [39]

    The Astrophysical Journal Letters , volume =

    Vo, Tien , title =. The Astrophysical Journal Letters , volume =. 2024 , doi =

    work page 2024

  40. [40]

    Davis, L. A. , title =. The Astrophysical Journal , volume =. 2021 , doi =

    work page 2021

  41. [41]

    Cohen, Z. A. , title =. The Astrophysical Journal , volume =. 2020 , doi =

    work page 2020

  42. [42]

    and Kellogg, Paul J

    Filbert, Paul C. and Kellogg, Paul J. , title =. Journal of Geophysical Research: Space Physics , volume =. 1979 , doi =

    work page 1979

  43. [43]

    Journal of Geophysical Research: Space Physics , volume =

    Vech, Daniel , title =. Journal of Geophysical Research: Space Physics , volume =. 2021 , doi =

    work page 2021

  44. [44]

    , title =

    Spanswick, E. , title =. ANGEO , volume =. 2005 , doi =

    work page 2005

  45. [45]

    , title =

    Kasper, Justin C. , title =. Space Sci Rev , volume =. 2016 , doi =

    work page 2016

  46. [46]

    Dean and Thompson, B

    Pesnell, W. Dean and Thompson, B. J. and Chamberlin, P. C. , title =. Solar Physics , volume =. 2012 , doi =

    work page 2012

  47. [47]

    , title =

    Tigik, Sabrina F. , title =. The Astrophysical Journal , volume =. 2022 , doi =

    work page 2022

  48. [48]

    Proceedings of the 8th

    Liu, Fei Tony and Ting, Kai Ming and Zhou, Zhi-Hua , title =. Proceedings of the 8th. 2008 , publisher =

    work page 2008

  49. [49]

    and Kriegel, Hans-Peter and Ng, Raymond T

    Breunig, Markus M. and Kriegel, Hans-Peter and Ng, Raymond T. and Sander, J\". Proceedings of the 2000. 2000 , publisher =

    work page 2000

  50. [50]

    Proceedings of the 2000

    Ramaswamy, Sridhar and Rastogi, Rajeev and Shim, Kyuseok , title =. Proceedings of the 2000. 2000 , publisher =

    work page 2000

  51. [51]

    and Van Driessen, Katrien , title =

    Rousseeuw, Peter J. and Van Driessen, Katrien , title =. Technometrics , volume =. 1999 , doi =

    work page 1999

  52. [52]

    International Conference on Learning Representations (

    Higgins, Irina and Matthey, Loic and Pal, Arka and Burgess, Christopher and Glorot, Xavier and Botvinick, Matthew and Mohamed, Shakir and Lerchner, Alexander , title =. International Conference on Learning Representations (. 2017 , url =

    work page 2017

  53. [53]

    Advances in Neural Information Processing Systems (

    Papamakarios, George and Pavlakou, Theo and Murray, Iain , title =. Advances in Neural Information Processing Systems (. 2017 , url =

    work page 2017

  54. [54]

    Zuko: Normalizing Flows in

    Rozet, Fran. Zuko: Normalizing Flows in. 2022 , publisher =

    work page 2022

  55. [55]

    Proceedings of the 37th International Conference on Machine Learning (

    Chen, Ting and Kornblith, Simon and Norouzi, Mohammad and Hinton, Geoffrey , title =. Proceedings of the 37th International Conference on Machine Learning (. 2020 , publisher =

    work page 2020

  56. [56]

    Proceedings of the

    He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian , title =. Proceedings of the. 2016 , doi =

    work page 2016

  57. [57]

    British Machine Vision Conference (

    Zagoruyko, Sergey and Komodakis, Nikos , title =. British Machine Vision Conference (. 2016 , doi =

    work page 2016

  58. [58]

    Transactions on Machine Learning Research (

    Oquab, Maxime and Darcet, Timoth\'. Transactions on Machine Learning Research (. 2024 , url =

    work page 2024

  59. [59]

    Journal of Global Optimization , volume =

    Storn, Rainer and Price, Kenneth , title =. Journal of Global Optimization , volume =. 1997 , doi =

    work page 1997

  60. [60]

    and Haberland, Matt and Reddy, Tyler and Cournapeau, David and Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and Bright, Jonathan and

    Virtanen, Pauli and Gommers, Ralf and Oliphant, Travis E. and Haberland, Matt and Reddy, Tyler and Cournapeau, David and Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and Bright, Jonathan and. Nature Methods , volume =. 2020 , doi =

    work page 2020

  61. [61]

    Towards Total Recall in Industrial Anomaly Detection , booktitle =

    Roth, Karsten and Pemula, Latha and Zepeda, Joaquin and Sch\". Towards Total Recall in Industrial Anomaly Detection , booktitle =. 2022 , doi =

    work page 2022

  62. [62]

    and Claudepierre, S

    Ripoll, J.-F. and Claudepierre, S. G. and Ukhorskiy, A. Y. and Colpitts, C. and Li, X. and Fennell, J. F. and Crabtree, C. , title =. Journal of Geophysical Research: Space Physics , volume =. doi:https://doi.org/10.1029/2019JA026735 , url =. https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2019JA026735 , note =

    work page doi:10.1029/2019ja026735

  63. [63]

    and Thorne, R

    Bortnik, J. and Thorne, R. M. , title =. J. Atmos. Sol.-Terr. Phys. , volume =. 2007 , doi =

    work page 2007

  64. [64]

    and Thorne, R

    Li, W. and Thorne, R. M. and Bortnik, J. and Nishimura, Y. and Angelopoulos, V. , title =. J. Geophys. Res. , volume =. 2011 , doi =

    work page 2011

  65. [65]

    Watt, C. E. J. and Degeling, A. W. and Rankin, R. and Murphy, K. R. and Rae, I. J. and Singer, H. J. , title =. J. Geophys. Res. , volume =. 2011 , doi =

    work page 2011