Huang et al. (2025) Drought propagation in china: Uncertainties originate more from dataset choice than drought index selection

Identification

• Journal: Atmospheric Research
• Year: 2025
• Date: 2025-10-11
• Authors: Kesheng Huang, Haicheng Zhang, Guotao Cui, Yijia Wang, Mijia Yin, Jianhui Du
• DOI: 10.1016/j.atmosres.2025.108555

Research Groups

• School of Geography and Planning, Sun Yat-Sen University, Guangzhou, China
• Carbon-Water Research Station in Karst Regions of Northern Guangdong, Guangzhou, China
• School of Geography, South China Normal University, Guangzhou, China
• Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China
• College of Resources and Environment, University of Chinese Academy of Sciences, Beijing, China

Short Summary

This study quantifies the uncertainties in drought propagation time and probability from meteorological to soil moisture drought across China, revealing that dataset choice contributes more to uncertainty than drought index selection, and identifies SPEI as optimal for minimizing these uncertainties.

Objective

• To evaluate multiple meteorological and soil moisture datasets against in-situ observations to establish a benchmark for drought propagation assessment in China.
• To estimate drought propagation time (PT) and probability (PB) using various drought indices (SPI, SPEI, PA, MI).
• To quantify uncertainties in PT and PB arising from the choice of datasets versus drought indices.
• To propose a framework for selecting optimal drought indices under varying dataset scenarios to support reliable drought early warning.

Study Configuration

• Spatial Scale: China (national scale)
• Temporal Scale: Monthly (analysis of propagation time in months, monthly bias)

Methodology and Data

• Models used:
  • Maximum correlation coefficient for estimating PT.
  • Copula function for estimating PB.
  • Drought Indices: Standardized Precipitation Index (SPI), Standardized Precipitation Evapotranspiration Index (SPEI), Palmer Drought Index (PA), and Moisture Index (MI).
• Data sources:
  • In-situ observations (benchmark).
  • Reanalysis datasets: ERA5-Land, MERRA-2.
  • Land surface model outputs: GLDAS-2.
  • Multiple meteorological and soil moisture datasets.

Main Results

• Benchmark-based drought propagation time (PT) ranges from 3.10 months (for SPEI) to 5.37 months (for MI), and drought propagation probability (PB) increases with regional humidity.
• ERA5-Land shows the highest spatial consistency, matching benchmark PT in 48.47 % (MI) to 61.62 % (SPEI) of pixels with no significant monthly bias. In contrast, MERRA-2 overestimates PT by 4–10 months in humid regions and yields prolonged PT in over 80 % of pixels.
• More than 80 % of pixels from ERA5-Land, GLDAS-2, and MERRA-2 show significant overestimations for PB.
• On average, 67.63 % of pixels exhibit greater PT uncertainty from dataset differences than from drought index differences, and 68.77 % show higher PB uncertainty from datasets.
• SPEI minimizes assessment uncertainties in drought PT and PB across different dataset scenarios compared to other indices.

Contributions

• Provides a quantitative basis for understanding and reducing uncertainties in drought propagation assessment in China.
• Offers guidance for selecting appropriate drought indices and datasets for more reliable drought early warning systems.
• Highlights that dataset choice is a more significant source of uncertainty than drought index selection for drought propagation time and probability.

Funding

Not explicitly stated in the provided text.

Citation

@article{Huang2025Drought,
  author = {Huang, Kesheng and Zhang, Haicheng and Cui, Guotao and Wang, Yijia and Yin, Mijia and Du, Jianhui},
  title = {Drought propagation in china: Uncertainties originate more from dataset choice than drought index selection},
  journal = {Atmospheric Research},
  year = {2025},
  doi = {10.1016/j.atmosres.2025.108555},
  url = {https://doi.org/10.1016/j.atmosres.2025.108555}
}