Corpus 1,272 assessed · 1,173 scored · 643 reproduced ≥75 · 168 flagged ·∅ 74.1/100
← New search

Direct observations of cross-scale wave-particle energy transfer in space plasmas.

Sci Adv · 2025
94/100 3/4
Why this verdict

The main results reproduced: recomputed values matched the published ones within tolerance.

Reproduced on the brainbox compute brainarbeit.com
How its reproducibility compares
94/100
Reproducibility score
1.1 SD above mean
vs. all fields · 1173 studies
🎯 Scores higher than 87% of all assessed papers rank 133 of 1173 scored

A 0–100 reproducibility-quality score from the per-question grades, shown as a z-score: standard deviations above (+) or below (−) the mean of comparable assessments.

Reproduction agent’s raw note

DESCRIBED WELL ENOUGH; in-scope pipeline reproduced 1:1 (EXACT), in FULL, on a fresh re-run. Space-plasma physics paper (Li et al. 2025, Sci Adv, MMS Earth-foreshock event 7 Jan 2021; PMID 39919183, PMCID PMC11804927, DOI 10.1126/sciadv.adr8227). The only self-contained, reproducible compute pipeline in the deposit (Zenodo 10.5281/zenodo.13947440 = github lijinghuan1997/cross-scale-foreshock @ 6ffb02f) is two deterministic Fortran test-particle simulations: ion-ULF and electron-whistler (analytic wave fields, COMMON /GEOPACK/ declared-but-never-called, all stdin reads commented out, no external libs/inputs). This pass rebuilt the conda env (gfortran 15.2.0), refetched the repo at the pinned commit, recompiled both UNMODIFIED (only 2 diagnostic ID columns added to the FSD write), and re-ran the FULL distributions inside «our HPC» SLURM «job» (compute node n096, partition std, 16 cores, elapsed 1:49:33, COMPLETED exit 0). Per-particle final velocity VECTORS match the authors' shipped .mat to MACHINE PRECISION: ion FULL 1812/1812 max dv_rel 1.60e-10, electron FULL 10800/10800 max dv_rel 2.24e-10 -- reproducing the earlier full run's numbers. Agreement SURVIVES on the long-integrated particles (ion ~1e7 RK4 steps, n=552, still 1.60e-10), which is the non-trivial test the original 'bogus' partial run (750/1812 short particles) skipped. Energy agrees within-tol (small constant relativistic-vs-MATLAB-nonrel convention offset, not a dynamical difference). Source SHAs (240fc55f ion / b1beac97 electron) and shipped .mat SHAs (2942e301 / 5f9a32a9) match recorded values; nothing fabricated -- every value is regenerable from the shipped code+params. NOT ATTEMPTED (out of scope, the paper's observational core): all MMS-measurement panels (wavelet PSD, FPI spectrograms, PDRK dispersion, measured Ji.E1 energy transfer) need irfu-matlab + MATLAB license + live MMS L2 data -- not a self-contained pipeline. Hence room-status 'partial': the in-scope simulation pipeline reproduced exactly and in full, but it is a supporting minority of the paper's results. All grades PROVISIONAL -- a human reviewer signs off.

💻 Code ↗ 🗄 Data: 10.5281/zenodo.13947440

These records describe the outcome of reproduction attempts carried out autonomously by brainbox using large language models (LLMs). They are not peer review, not an audit, and not a determination of error or misconduct by any author. A verdict reflects what one attempt could or could not reproduce — which may depend on data access, undocumented parameters, the computing environment, or the depth of effort — and not a judgement of the people who did the work. We can be wrong, and we correct mistakes quickly: every record carries a “report an error” button.

Assessment versions

Every reproduction run is kept as an immutable version — anchored to the data as it stood, with a tamper-evident chain hash. A rerun (e.g. after an author updates a deposit) adds a new version; the previous one stays on record.

  1. v1 current initial assessment Score 94
    assessed: 2026-06-19 ⛓ 372e1f82f957
✎ I am an author of this paper

Updated or fixed a deposit, or is there an erratum? Ask us to re-run the metrics. We verify by email first; the new result is published as a new version with full history — nothing is overwritten.

Reason for the rerun

We email you a confirmation link first. The rerun is an objective re-measurement — it cannot change the verdict in your favour, only ask us to look again.

Provenance — full disclosure

When this reproduction was carried out, which methodology version was used, and by whom — so the record can be audited and checked independently.

Reproduced
2026-06-22
Rubric version
not recorded
Assessed by
Last updated
2026-08-05

Provisional, curator- or AI-assessed, and independently checkable. A reproduction outcome states what one attempt could reproduce — not a judgement of the authors.

Deep full-text extraction

Model: sonnet
Founding hypothesis

The paper tests whether a chain of coherent wave-particle resonances in Earth's foreshock can transfer energy sequentially across scales, from fluid-scale ULF waves to ion-scale magnetosonic-whistler waves to electron-scale whistler waves.

Core claims
  • Fluid-scale ULF waves resonate with reflected ions, modifying their velocity distributions and driving growth of ion-scale magnetosonic-whistler waves finding
  • Magnetosonic-whistler waves resonate with electrons, and the accelerated electrons contribute to excitation of electron-scale, higher-frequency whistler waves finding
  • The observed chain of wave-particle resonances constitutes an efficient mechanism for cross-scale energy transfer from fluid to ion to electron scales finding
  • Anomalous resonance, a nonlinear resonance occurring when wave amplitude is large enough to modify the classical resonance condition, plays an important role in cross-scale energy transfer mechanism
  • Reflected ions in the foreshock excite ULF waves via ion/ion cyclotron resonant instability mechanism
  • Classical cyclotron resonance theory cannot explain the observed phase-bunched ion signatures at pitch angles 40°-90° during large-amplitude wave intervals finding
  • Nonlinear steepening of ULF waves produces magnetosonic-whistler wave packets (shocklets) radiating sunward in the plasma frame mechanism
Experimental setups
Assay System Perturbation Readout Platform
Magnetic field measurement Terrestrial foreshock plasma (in situ space observation) none Magnetic field components, wave power spectral density Flux-gate magnetometer (MMS)
Electric field measurement Terrestrial foreshock plasma none Electric field for SVD wave analysis Electric field double probes (MMS)
Ion velocity distribution measurement Terrestrial foreshock protons (solar wind and reflected populations) none Ion phase-space density, pitch angle and gyro-phase spectra Fast Plasma Investigation (FPI)
Energetic particle spectrometry Terrestrial foreshock plasma none Higher-energy ion/electron flux spectra Fly's eyes energetic particle spectrometer (FEEPS)
Wavelet spectral and coherence analysis MMS magnetic field and ion flux time series none Wave power spectrum, wavelet coherence and phase difference between ion flux and Bz
Singular value decomposition (SVD) wave analysis MMS electromagnetic field data none Wave phase speed, propagation direction, polarization
Linear dispersion relation / growth rate solver Observed foreshock ion velocity distributions none Wave dispersion relation and growth rate Dispersion relation solver
Test-particle / kinetic simulation Simulated protons in ULF and magnetosonic-whistler wave fields Applied simulated wave electromagnetic field Ion gyro-phase (ζ) and kinetic energy time evolution
Key results
  • ULF waves observed with period ~45 s (ω_ULF ~0.14 rad/s), propagating quasi-parallel to bulk flow at v_ULF = 433 km/s, slightly below bulk velocity (465 km/s), with small parallel wavenumber indicating fluid-scale waves 433 km/s; k_ULF ~0.04/r_i
  • Magnetosonic-whistler wave packet phase speed determined as v_mw = 378 km/s at ω_mw = 2.83 rad/s, with parallel wavenumber indicating ion-scale waves 378 km/s; k_mw ~0.91/r_i
  • Observed ULF wavenumbers fall in a region of positive linear growth rate from the dispersion relation solver, confirming cyclotron-resonant excitation of fluid-scale ULF waves by reflected ions
  • Classical resonance velocity (Vr = -458 km/s) predicts phase-bunching at pitch angle 130°, but observed phase-bunched (gyro-phase locked) ions occur instead at pitch angles 40°-90°, a discrepancy attributed to anomalous resonance under large wave amplitude (B1,ULF/B0 ~ 2/3) Vr = -458 km/s; B1,ULF/B0 ~2/3
  • Wavelet coherence between 3-keV ion flux (moving in -z direction) and Bz shows high coherence at the ~45 s ULF wave period with an in-phase relationship, indicating wave-particle resonance
  • Reflected proton beam population near pitch angle ~135° (visible in first ~2 min) gradually disappears as ULF waves grow, while a new ion population emerges at pitch angles 40°-90°
Key statistics
  • other ω_ULF ~ 0.14 rad/s (ULF wave angular frequency)
  • other v_ULF = 433 km/s (ULF wave phase speed from SVD analysis)
  • other v_b = 465 km/s (Plasma bulk velocity along -x)
  • other k_ULF ~ 0.04/r_i (r_i = 121 km) (ULF wave parallel wavenumber normalized to ion inertial length)
  • other v_mw = 378 km/s; ω_mw = 2.83 rad/s; k_mw ~ 0.91/r_i (Magnetosonic-whistler wave phase speed, frequency, and wavenumber)
  • other Vr = -458 km/s (Classical cyclotron resonance velocity for 3-keV protons with ULF waves)
  • other B1,ULF/B0 ~ 2/3 (Ratio of ULF wave amplitude to background field, indicating large-amplitude regime enabling anomalous resonance)
  • other Ωp ~ 0.29 rad/s; background field ~3 nT (Proton gyrofrequency and background magnetic field magnitude)

Statistical methods review

Model: sonnet

A neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.

This is an observational space-physics case-study paper using high-resolution in situ measurements from the four-spacecraft MMS constellation. The analytical approach is dominated by signal-processing techniques (wavelet spectral analysis, singular-value decomposition, wavelet coherence) and physics-based methods (linear dispersion-relation solving, resonance-condition calculations, gyro-phase spectral averaging across spacecraft). Classical inferential statistics are not employed; conclusions are drawn from qualitative and semi-quantitative agreement between observed wave-particle signatures and theoretical resonance conditions, supplemented by test-particle simulations.

Replicationtechnical Sample sizeSingle event (7 January 2021); four-spacecraft averaging used for gyro-phase spectra to improve statistical significance; no formal sample-size or power justification provided GroupsSingle observational event; different time intervals (wave-growth phases, shadowed intervals) compared qualitatively against each other and against theoretical predictions Pairingna Randomization/blindingna Dispersionnone Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionnone stated
Statistical tests used
Test Applied to n Assumptions
Wavelet coherence (time-frequency coherence with phase arrows) Correlation between 3-keV ion flux and magnetic field Bz component to demonstrate ULF wave-particle resonance (Fig. 1F) not stated
Singular Value Decomposition (SVD) of electromagnetic fields Estimation of ULF wave phase speed and direction (fig. S1) and magnetosonic-whistler wave properties (fig. S3) not stated
Linear plasma instability / dispersion-relation solver Confirmation that observed ion distribution drives fast/magnetosonic wave growth at observed wave numbers (Fig. 2B) not stated
Gyro-phase spectral averaging across four MMS spacecraft Gyro-phase spectra for 3-keV ions at pitch angles 40°, 90°, 130° to detect phase-bunching signatures (Fig. 3B–D) 4 spacecraft not stated
Wavelet power spectral density Characterization of ULF and magnetosonic-whistler wave power as a function of time and frequency (Fig. 1C) not stated
Approaches that could also have been used
  • Wave properties (phase speed, propagation direction) were estimated via Singular Value Decomposition (SVD) of the electromagnetic field matrix from a single spacecraft interval
    Could also: Minimum Variance Analysis (MVA) of the magnetic field, or multi-spacecraft timing analysis using all four MMS spacecraft simultaneously, could also estimate propagation direction and phase speed — Multi-spacecraft timing (k-filtering / phase-difference method) uses spatial baselines between all four spacecraft and yields an independent velocity estimate; reporting both SVD and timing results would allow cross-validation of the wave property determination
  • Gyro-phase spectra were averaged across the four MMS spacecraft to improve statistical significance without formal uncertainty quantification
    Could also: Bootstrapped confidence intervals or inter-spacecraft standard deviation could also be reported to quantify how consistently each spacecraft sees the same phase-bunching pattern — Explicit uncertainty bounds on the averaged spectra would allow readers to assess how much of the observed stripe pattern persists above measurement noise, and whether the four spacecraft agree quantitatively
  • The study is a single carefully selected event chosen to illustrate the full chain of wave-particle energy transfer
    Could also: A superposed-epoch or multi-event statistical analysis of foreshock intervals could also be conducted — A multi-event study would allow estimation of how frequently this complete energy-transfer chain occurs under varying upstream conditions, supporting generalization of the mechanism beyond the single case
  • Wavelet coherence between ion flux and Bz was used to demonstrate wave-particle correlation; phase relationships were shown qualitatively with arrows
    Could also: Cross-spectral analysis with a formal coherence significance threshold (e.g., against a red-noise background via Monte Carlo, as in Torrence & Compo 1998) could also be applied — Reporting a significance level on the coherence estimate would distinguish regions of genuine coherence from those that could arise by chance, making the wave-particle coupling inference more quantitatively defensible
  • Linear growth rates were computed from the observed particle distributions using a dispersion-relation solver and compared with observed wave numbers qualitatively (shaded region in Fig. 2B)
    Could also: Sensitivity analysis varying the assumed distribution parameters within observational uncertainty could also be reported — Since the input distribution is itself derived from limited phase-space measurements, propagating that uncertainty into the modeled growth rate would show how robustly the observed wave number falls within the unstable region
  • Phase-bunching signatures in gyro-phase spectra were identified by visual inspection of stripe patterns relative to the wave phase reference
    Could also: A quantitative phase-locking value (PLV) or mean resultant length of the phase distribution could also be computed as a scalar summary of the degree of phase-bunching — A scalar metric of phase coherence would allow the strength of wave-particle locking to be compared across different pitch-angle bins and across different events in a standardized way
Software: Dispersion relation solver (cited as ref. 45) · Wavelet coherence analysis (cited as ref. 47, likely Torrence & Compo 1998 method)

What was reproduced

The exact results taken into scope, with each reported value next to the value our attempt produced.

Scope — pmid-39919183 (honest re-do, 2026-06-16)

Paper: Li et al. 2025, Direct observations of cross-scale wave-particle energy transfer in space plasmas, Sci Adv 11, eadr8227. DOI 10.1126/sciadv.adr8227. PMID 39919183 · PMCID PMC11804927.

Event: MMS in Earth's foreshock, 7 Jan 2021, ~0816–0834 UT.

NB — this RU is space-plasma physics, not bioinformatics; the generic brief template still applies. A prior automated reproduction was renamed reproduction.bogus-* by a reviewer because it finalized "reproduced" on a partial run (750/1812 ion particles) consisting almost entirely of the short-trajectory particles which match trivially (see below). This re-do runs the FULL distributions and reports agreement broken down by trajectory length so the conclusion is honest.

Artifacts (Zenodo 10.5281/zenodo.13947440 = repo lijinghuan1997/cross-scale-foreshock)

Pinned commit 6ffb02fbddd79416fba7ea1898fe9d5947d94d5c (2024-10-17):

  • ion_ULF_interaction.f90 — Fortran test-particle sim, ion–ULF wave interaction
  • electron_whistler_interaction.f90 — Fortran, electron–whistler interaction
  • gyro_ion_ULF.m, gyro_electron_whistler.m — MATLAB: bin FSD.txt → gyrophase spectra
  • datasave3keV_..._tp_151_..._GYRO_12_..._vphase450.mat (869 KB) — authors' ion-sim product, datasave50eV = (12 gyro × 151 tp) cell grid
  • datasave39eV_PA140_gyro36_..._tp300_....mat (7.8 MB) — electron-sim product, datasave50eV = (36 gyro × 300 tp)
  • B_modify_gaussian_phasechanged.mat (5 KB) — MATLAB plotting helper; NOT read by the Fortran

Both Fortran programs are self-contained & deterministic: Bfield/Efield are purely analytic wave models; the declared COMMON /GEOPACK/ variables are never used in the field math; every read * is commented out (inputs hardcoded); no input files / stdin. Output FSD.txt = each particle's final back-traced state. Recompiling with gfortran and re-running = the reproducible pipeline. No MATLAB license needed for the simulation (the .m scripts only bin/plot).

In scope (pipeline-derived, attempted)

R1 — ion test-particle sim (ion_ULF_interaction.f90). 3 keV protons, pitch 80°, B0z=3 nT, wave-B 2 nT, E 0.9 mV/m, v_phase 450 km/s, ω=2π/45, 12 gyrophases (15–345° step 30), 151 timepoints with T_end=(wnum−1)·3 s → 0–450 s. Backward trace, RK4, fixed step h≈4.2e−5 s ⇒ the longest (wnum=151) particle integrates ~1.07×10⁷ steps. The honest test: short-trajectory (low-wnum) particles integrate only tens of steps and match the shipped product trivially; the question is whether agreement survives for the long-trajectory particles under a different compiler. We compare the FINAL velocity VECTOR per particle (the physics) and break agreement down by wnum.

R2 — electron–whistler sim (electron_whistler_interaction.f90). 39 eV electrons, pitch 140°, B0z=2 nT, wave-B 1.5 nT, E 0.6 mV/m, v_phase 400 km/s, ω=2π·0.1·exp(...) (spatially varying whistler), 36 gyrophases (5–355° step 10), 300 timepoints T_end=(wnum−1)·0.06 s → 0–17.94 s, step h≈2e−6 s ⇒ ~9×10⁶ steps for the longest particle, ~10800 particles total (~5× the ion cost). Parallelized by striding the independent wnum loop across cores (physics-identical).

Out of scope (not attempted — observational / needs MATLAB + live MMS)

All MMS-observation panels (wavelet PSD, FPI pitch-angle/gyrophase spectrograms, PDRK dispersion, measured Jᵢ·E₁ energy transfer) require irfu-matlab + live MMS L2 data + a MATLAB license — observational, not a self-contained compute pipeline. The MMS-overlay line in the plotting scripts (mms.get_data) likewise needs live data; the simulation panels do not depend on it.

Comparison method

compare_v2.py: loads the shipped .mat, auto-calibrates which cell column equals which physical quantity using the trivially-matching wnum=1 particles (sign- and unit-aware), then for all particles compares the final velocity vector |Δv|/|v|, plus per-particle energy. Reports aggre

Figures / tables: Fig 3BFig 3Fig 4
R1_velspace
Reported
authors' shipped datasave50eV (12 gyrophase x 151 timepoint = 1812 cells); per-particle back-traced final velocity-space state of 3 keV / pitch-80deg protons through the analytic ULF wave field (Fig 3 ion gyrophase spectrum input)
Reproduced
FULL 1812/1812 particles, 0 missing; final velocity-vector dv_rel median=0, mean 2.71e-13, max 1.60e-10; agreement SURVIVES the longest-integrated bin (wnum 106-151, Tend 315-450 s, ~1e7 RK4 steps, n=552, max dv_rel 1.60e-10) -- not just the trivially-matching short particles
exact
R1_energy
Reported
ion per-particle final kinetic energy, datasave50eV energy column (~3 keV nominal)
Reproduced
energy residual median 0.59 eV (0.020% of 3 keV), max 1.53 eV (0.051%); n=1812/1812. Small near-constant offset (MATLAB recomputes energy non-relativistically vs Fortran relativistic enem), NOT a trajectory divergence -- the 1e-10 velocity match confirms identical dynamics
within tolerance
R2_velspace_electron
Reported
authors' shipped datasave50eV (36 gyrophase x 300 timepoint = 10800 cells); per-particle back-traced final velocity-space state of 39 eV / pitch-140deg electrons through the analytic spatially-varying whistler field (Fig 4 electron gyrophase spectrum input)
Reproduced
FULL 10800/10800 particles, 0 missing (distinct wnum=300, alpha=36 confirmed); final velocity-vector dv_rel median=0, mean 1.13e-12, max 2.24e-10; machine precision across ALL trajectory-length bins incl. the longest
exact
R2_energy_electron
Reported
electron per-particle final kinetic energy, datasave50eV energy column (~39 eV nominal)
Reproduced
energy residual median 0.0080 eV, max 0.0082 eV; n=10800/10800. Same constant relativistic-vs-MATLAB-nonrel convention offset as the ion case
within tolerance
R_config
Reported
ion: 3 keV / pitch 80 / B0z 3nT / Bw 2nT / E 0.9mV/m / vphase 450km/s / omega 2pi/45 / 12 gyro (15-345 step30) / 151 tp (3s) / backward trace. electron: 39 eV / pitch 140 / B0z 2nT / Bw 1.5nT / E 0.6mV/m / vphase 400km/s / spatially-varying whistler / 36 gyro (5-355 step10) / 300 tp (0.06s)
Reproduced
all params verified hardcoded in the unmodified Fortran (original SHAs 240fc55f / b1beac97 match recorded values) and consistent with the shipped .mat filenames + Methods; only the FSD write was instrumented with particle-ID columns (pure diagnostic, guarded by a patch-count assertion)
exact

Assessments & scoring basis

Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.

No assessment has been recorded yet.
🤝
Reproduced automatically — and fairly

Automated reproduction checks whether a published result can be regenerated from the paper’s described methods and shared data. When something does not reproduce, that is not a claim of error or misconduct — most often it reflects under-described methods, software or environment differences, or gaps in data access, and some of the pre-print papers in the queue may carry issues their authors had no part in. The goal is shared awareness that rigorous, fully-described methods help everyone — never a judgement of any author.

Are you an author? We would genuinely like to hear from you — to clarify the record, add data or code, re-run the pipeline after an accession update, and publish your response right next to the assessment. Everything here is open and auditable.

🚩 Report an error in this record

Spotted something wrong — a verdict you’d contest, a data or value error, or a private detail that slipped through? Tell us, with a short justification. Authors and readers are equally welcome to write in; we review every report.

Prefer email, or the form below not working? Contact us at support@doesitreproduce.com.

Reproduction footprint

claude-opus-4-8

Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.

2 M
tokens (I/O) · 129.3 M incl. cache
665 min
runtime · 2.44 CPU-h
2.5 GB
peak RAM
1
HPC jobs
hummel
machine