Academic Integrity Report: "Retrieval of Temperature Profiles in the MLT Region Using the O2(b1Σ+g) Airglow With Spectral Contamination Correction of N2 1PG Emissions"

Academic fraud report · Geng Detector

Summary

This report assesses a 2024 IEEE Transactions on Geoscience and Remote Sensing article (DOI: 10.1109/TGRS.2024.3492178) by Kuijun Wu et al. for potential academic integrity issues. The verdict is highly suspicious (orange tier). Six issues were identified across textual, statistical, and editorial dimensions. The most severe finding is a direct numerical contradiction: the Abstract and Section IV-A report a 6 K average deviation (120–140 km) with 80% reduction, whereas Section V reports 9 K with 80% improvement for the same procedure. A second critical concern is Equation (9), which defines RMSE by subtracting each MIGHTI measurement (X_i) from F, defined as the average of all ACE-FTS temperatures at that altitude, rather than matched point-to-point values, an invalid formulation for cross-satellite validation. Additional issues include a duplicated sentence in Section II-B suggesting automated translation without proofreading, implausibly uniform improvement figures (42%, 63%, 79%, 80%), and corrupted reference entries missing authors. Timeline checks were clean. Confidence is moderate; the report is based on text and equations only, without raw data or images. Final determination requires institutional investigation.

Verdict

Highly suspicious (🟠). Multiple textual contradictions, a statistically invalid validation formula, duplicated prose, and implausibly uniform performance gains raise serious concerns about scientific rigor. No raw images were available for image-manipulation analysis. The findings should be treated as allegations pending institutional review.

Key findings

Self-contradictory core results (Critical): Abstract/Section IV-A state an average deviation of 6 K with 80% reduction at 120–140 km; Section V states 9 K with 80% improvement for the same altitude range and procedure.
Invalid RMSE definition in Equation (9) (Critical): $U_{rmse}=\sqrt{\frac{\sum (X_i - F)^2}{N}}$ uses $F$ = the average of all ACE-FTS temperatures at each altitude, not the matched co-located value. This conflates ACE-FTS internal variance with MIGHTI bias and is not a standard point-to-point RMSE.
Duplicated sentence in Section II-B (Moderate): The clause "Meanwhile, the ratios of the B and C channels show a linear functional relationship with temperature, while those of the D and C channels exhibit/exponential variation with temperature" appears twice consecutively with minor wording shifts.
Implausibly uniform improvements (Moderate): Reported gains of 80% (self-consistency), 79% (mean deviation reduction), 42% (vs. ACE-FTS), and 63% (vs. ICON official algorithm) appear across independent metrics and are unusually large and consistent for an atmospheric retrieval correction.
Corrupted reference list (Low–Moderate): Reference [1] begins with the journal name with no author listed; reference [2] begins with the word "using," indicating incomplete metadata export.
Timeline check: no anomaly. Paper uses ICON (launched 2019), 2021–2022 MIGHTI data, HITRAN2020 (released 2021), submitted September 2024, published November 2024, consistent with normal scientific workflow.

Evidence highlights

Direct quotation from Abstract/Section IV-A: "The average deviation in the altitude range of 120–140 km is only 6 K. The average deviation... is reduced by 80%."
Direct quotation from Section V (Conclusion): "In the state after eliminating N2 SC, the average temperature deviation of the two channel combinations is only 9 K, which increases the temperature self-consistency... by 80%."
Equation (9): $U_{rmse}=\sqrt{\frac{\sum (X_i - F)^2}{N}}$, where $F$ is defined as "the average of all ACE-FTS temperature data at each altitude."
Duplicated text in Section II-B: identical clause about the B/C and D/C channel ratios appears twice in succession with only "exhibit" → "demonstrate" and "variation" → "variation;" substituted.
Reported improvements: 80% self-consistency gain (Sec. IV-A), 79% mean-deviation reduction (Sec. IV-A), 42% agreement gain vs. ACE-FTS (Sec. IV-B), 63% improvement vs. ICON official algorithm (Sec. IV-C).
Reference anomalies: [1] starts with "Surv. Geophys., vol. 33…" (no author); [2] starts with "using an all-fiber-coupled…" (no author).
DOI: 10.1109/TGRS.2024.3492178.

Notes

Image-based forensics were not applicable because no original figures were provided; this report is limited to textual, mathematical, and editorial evidence.
The discrepancy between 6 K and 9 K may conceivably reflect a last-minute revision that was not propagated through the manuscript, but the magnitude and the accompanying "80%" figure being repeated on both sides weaken the typographical-error explanation.
The Equation (9) issue would invalidate quantitative claims of cross-mission agreement and likely any conclusions tied to the "42% improvement" figure unless a corrected point-to-point RMSE is supplied.
Authors are affiliated with Yantai University per the report; any formal complaint to the editorial office or institutional research integrity committee should include both the numerical contradiction and the RMSE formulation as primary exhibits.
This report is AI-assisted and intended for academic discussion only; final determinations of misconduct rest with the publisher, IEEE, and the authors' institution.