DATA-SOURCE VERIFICATION OPTIONS — DO NOT CITE UNTIL MATCHED TO THE ACTUAL INPUT FILES
Checked: 2026-09-02

1. CHINA LAND COVER DATASET (CLCD), 2023 LAYER

Current manuscript problem:
- The manuscript currently cites Zenodo DOI 10.5281/zenodo.8176941, whose record title and file set document CLCD through 2022.

Candidate that can support 2023, only if it matches the file actually analysed:
- Yang J, Huang X (2024) The 30 m annual land cover datasets and its dynamics in China from 1985 to 2023. Zenodo, version 1.0.3. https://doi.org/10.5281/zenodo.12779975
- The record contains CLCD_v01_2023_albert.tif and CLCD_v01_2023_albert_province.zip.

If confirmed, update all three places together:
- Table 1 temporal coverage and source/version;
- Data availability statement;
- Reference list/dataset citation.

2. NIGHT-TIME LIGHT (NTL), 2019–2023

Current manuscript problem:
- Li et al. (2020), https://doi.org/10.1038/s41597-020-0510-y, describes the harmonised 1992–2018 series. A later Figshare version extends that product only through 2021. Neither citation alone documents values through 2023.

Route A — use a published 1992–2023 product, only if this is the actual product analysed:
- Chen X, Wang Z, Zhang F et al (2024) A global annual simulated VIIRS nighttime light dataset from 1992 to 2023. Scientific Data 11:1380. https://doi.org/10.1038/s41597-024-04228-6
- Data record: https://doi.org/10.6084/m9.figshare.22262545.v8
- Product: simulated VIIRS data for 1992–2011 plus annual VIIRS composition for 2012–2023, 15 arc-second (approximately 500 m) resolution.

Route B — retain Li et al. (2020) and document an author-created extension:
- State the exact VIIRS product used for 2019–2023, monthly-to-annual compositing rules, masking, unit/DN conversion, cross-sensor calibration, resampling, clipping, and validation.
- Cite the VIIRS source and any calibration method in addition to Li et al. (2020).
- Archive the actual processing script and parameters in the replication package.

Route C — if no defensible 2022–2023 NTL input exists:
- Do not relabel an older product as 2023.
- Shorten the analysis endpoint and recalculate all affected tables, figures, EVI/SRP values, trends, Moran/LISA, GeoDetector and border-distance results.

AUTHOR ACTION
- Record the original filenames, checksums, download URLs, version/release dates and access dates for every input actually used.
- Select the route that matches the completed analysis; do not select a newer dataset merely because it covers the desired years.
