Integrity review of Liu et al., First Chinese ultraviolet–visible hyperspectral satellite instrument implicating global air quality during the COVID-19 pandemic in early 2020 (Light: Sci. & Appl., 2022)

Academic fraud report · Geng Detector

Summary

Verdict: Clear. The original Chinese-language integrity review concludes the paper shows no substantive signs of academic fraud. Concerns raised are limited to minor typographical errors: 'Seoul' is misspelled as 'Soul' in Figure 3, and 'Tokyo' is misspelled as 'Tokoy' on the axis labels of Figures 5 and 6. Statistical logic, including the reported regression equations such as NO2 (y = 1.23x + 9.09×10^14, R^2 = 0.63) and HCHO (y = 1.14x − 9.86×10^14, R^2 = 0.92), is internally consistent and consistent with prior TROPOMI-based literature (e.g., a ~56% NO2 reduction in Wuhan). The authors openly acknowledge a key limitation: EMI data on GaoFen-5 became unavailable after April 2020 due to a solar battery failure, after which TROPOMI data were used. No image-manipulation concerns apply because the figures are heatmaps, scatter plots, and bar charts, not Western blots or gels. Overall confidence in the review is moderate-high, though limitations include lack of raw image data for pixel-level analysis.

Verdict

Clear. No substantive evidence of academic fraud. Only minor typographical errors were identified. The paper's statistics, timeline, and methodological transparency are consistent with legitimate research.

Key findings

Typographical errors (minor): "Seoul" rendered as "Soul" in Figure 3 caption; "Tokyo" rendered as "Tokoy" on axis labels of Figures 5 and 6. Severity: 🟡 low.
Statistical logic (normal): Reported regression equations (e.g., NO2: y = 1.23x + 9.09×10^14, R^2 = 0.63; HCHO: y = 1.14x − 9.86×10^14, R^2 = 0.92) follow standard satellite-retrieval methodology; random error calculation based on pixel count npix is conventional.
Reported effect sizes align with independent literature: A Wuhan NO2 reduction of ~56% is consistent with other TROPOMI-based COVID-era studies.
Transparent disclosure of data limitation (positive integrity indicator): Authors openly state that EMI data on GaoFen-5 became unavailable after April 2020 due to a solar battery failure, after which TROPOMI data were substituted.
Image-manipulation checks not applicable: All figures are geospatial heatmaps, scatter plots, or bar charts; no Western blots, gels, or microscopy images are present.

Evidence highlights

DOI: 10.1038/s41377-022-00722-x
Figure 3 caption: "Guangzhou, Seoul, Tokyo" but on-image labels display "Soul" and "Tokoy".
Figures 5 and 6 x-axis labels: "Tokoy" instead of "Tokyo".
Equation examples from Figure 3: NO2 (y = 1.23x + 9.09×10^14, R^2 = 0.63); HCHO (y = 1.14x − 9.86×10^14, R^2 = 0.92).
Methods note (verbatim from review): "EMI data was not available from April 2020, because of the failure in the solar battery of the Gaofen-5 satellite."

Notes

Confidence: moderate-high. Limitations: the review did not have access to high-resolution raw figures for pixel-level image-forensic analysis; reproducibility of the regression code and raw satellite L1/L2 products was not independently verified.
No contact with the authors or PubPeer comment is recommended for fraud concerns. An optional courteous email to the authors flagging the typographical errors may improve reader experience.
This report is AI-assisted and intended for academic discussion only; final determinations of misconduct require investigation by competent institutions.