Pan et al. (2025) Identifying Optimal Reanalysis and Remote Sensing Data Combinations for Multi-Scale SPEI-Based Drought Assessment in Zhejiang Province, China

Identification

• Journal: Atmosphere
• Year: 2025
• Date: 2025-09-12
• Authors: Suli Pan, Di Ma, Haiting Gu, Chao Xu, Xiaojie Zhou, Qiang Zhu
• DOI: 10.3390/atmos16091078

Research Groups

• Nanxun Innovation Institute, Zhejiang University of Water Resources and Electric Power, Hangzhou, China
• School of Civil Engineering, NingboTech University, Ningbo, China
• Institute of Water Science and Engineering, College of Civil Engineering and Architecture, Zhejiang University, Hangzhou, China
• School of Civil Engineering and Architecture, Zhejiang University of Science and Technology, Hangzhou, China

Short Summary

This study evaluates nine reanalysis and remote sensing data combinations for multi-scale Standardized Precipitation Evapotranspiration Index (SPEI) estimation in Zhejiang Province, China, identifying the optimal combination and subsequently analyzing spatiotemporal drought variations from 1980–2020. The research found that the CMFD V2.0 precipitation and GLEAM v4.2a evapotranspiration combination is most reliable, revealing a significant "wetter winters, drier springs" pattern and distinct spatial drying trends in southern/southeastern regions.

Objective

• To comprehensively evaluate the performance of nine precipitation and evapotranspiration data combinations in capturing drought conditions across multiple timescales (SPEI1, SPEI3, SPEI6, and SPEI12) and identify the optimal data combination for Zhejiang Province.
• To utilize the identified optimal data combination to construct long-term, high-resolution SPEI datasets and investigate the spatiotemporal variations in drought in Zhejiang Province from 1980–2020.

Study Configuration

• Spatial Scale: Zhejiang Province, China (approximately 105,500 km²), with all reanalysis and remote sensing datasets resampled to a 0.1° spatial resolution.
• Temporal Scale:
  • Evaluation Period: 2003–2018 (for comparison with ground-based observations).
  • Drought Analysis Period: 1980–2020 (for long-term spatiotemporal variations).
  • Drought Timescales: 1-month (SPEI1), 3-month (SPEI3), 6-month (SPEI6), and 12-month (SPEI12).

Methodology and Data

• Models used:
  • Standardized Precipitation Evapotranspiration Index (SPEI) calculation using the three-parameter log-logistic distribution and L-moment method.
  • Food and Agriculture Organization of the United Nations (FAO) Penman–Monteith (PM) method for ground-based potential evapotranspiration (PET) estimation.
  • Mann–Kendall (MK) statistical test for trend analysis.
  • Multiple Linear Regression (MLR) model for contribution analysis of climatic drivers.
• Data sources:
  • Ground-based Meteorological Observations: Daily precipitation, maximum/minimum air temperatures, relative humidity, wind speed, and solar radiation from 12 stations in Zhejiang Province (2003–2018), obtained from the National Climate Center, China Meteorological Administration.
  • Reanalysis and Remote Sensing Precipitation Datasets:
    • CMFD V2.0 (China Meteorological Forcing Dataset Version 2.0): 0.1° spatial resolution, 3-hourly temporal resolution (1951–2020).
    • IMERG V07B (Integrated Multi-satellitE Retrievals for GPM Version 07B) Final Run: 0.1° spatial resolution, 30 min/daily/monthly temporal resolution (1998–present).
    • TMPA 3B42V7 (TRMM Multi-satellite Precipitation Analysis Version 7): 0.25° spatial resolution, 3-hourly/daily/monthly temporal resolution (1998–2019).
  • Reanalysis and Remote Sensing Evapotranspiration Datasets:
    • GLDAS-2.2 (Global Land Data Assimilation System Version 2.2): 0.25° spatial resolution, daily temporal resolution (2003–present).
    • GLEAM v4.2a (Global Land Evaporation Amsterdam Model version 4.2a): 0.1° spatial resolution, daily temporal resolution (1980–2023).
    • PML-V2 (Penman–Monteith–Leuning Version 2): 0.05° spatial resolution, 8-day temporal resolution (2002–2019).

Main Results

• The choice of precipitation product is the dominant factor influencing SPEI accuracy, with evapotranspiration products having a relatively minor impact.
• The combination of CMFD V2.0 precipitation and GLEAM v4.2a evapotranspiration (PIEII) is identified as the most reliable for SPEI estimation across multiple timescales (SPEI1/3/6/12). For SPEI12, this combination achieved a Nash–Sutcliffe Efficiency (NSE) of 0.90, a Root Mean Square Error (RMSE) of 0.30, and a Pearson correlation coefficient (R) of 0.95.
• SPEI estimation accuracy generally improves with increasing timescale, with SPEI12 showing the highest accuracy.
• All data combinations exhibit substantial errors in estimating extreme droughts (SPEI ≤ -2.0) at shorter timescales (SPEI1 and SPEI3), which diminish at longer timescales (SPEI6 and SPEI12).
• Spatiotemporal Variations (1980–2020):
  • A pronounced multi-year drought period was observed from approximately 2003 to 2009, particularly evident in the SPEI12 series.
  • A significant seasonal asymmetry in SPEI trends emerged: "wetter winters, drier springs," with intensified spring drying posing a substantial threat to agricultural water security.
  • Spatially, central and northeastern Zhejiang exhibited wetting trends, while southern and southeastern regions showed a significant drying tendency, especially for long-term hydrological drought (SPEI12).
  • Light droughts (−1.0 < SPEI ≤ −0.5) are widespread across the province, with occurrence rates often exceeding 15%, suggesting a sustained baseline of water stress.
• Climatic Drivers: Precipitation is the dominant driver of SPEI variations across all timescales, contributing 57–86% to the variations. Potential evapotranspiration (PET) co-regulates short- to medium-term droughts, contributing 14–47%.

Contributions

• Provides a rigorously validated, high-resolution data foundation (CMFD V2.0 precipitation and GLEAM v4.2a evapotranspiration) for regional drought assessment in Zhejiang Province.
• Offers a scientific basis for developing targeted, season-specific, and spatially differentiated water resource management and drought adaptation strategies in a traditionally humid region facing escalating drought risks.
• Advances a more robust and multidimensional evaluation framework for selecting optimal reanalysis and remote sensing data combinations for SPEI-based drought assessment.
• Unveils complex, scale-dependent, and seasonally structured characteristics of drought evolution in Zhejiang Province, including a critical "wetter winters, drier springs" pattern and distinct regional drying trends.

Funding

• Natural Science Foundation of Zhejiang Province (LQN25E090007)
• Nanxun scholars program of ZJWEU (RC2023010969)
• National Natural Science Foundation of China (52209036 and 51909233)

Citation

@article{Pan2025Identifying,
  author = {Pan, Suli and Ma, Di and Gu, Haiting and Xu, Chao and Zhou, Xiaojie and Zhu, Qiang},
  title = {Identifying Optimal Reanalysis and Remote Sensing Data Combinations for Multi-Scale SPEI-Based Drought Assessment in Zhejiang Province, China},
  journal = {Atmosphere},
  year = {2025},
  doi = {10.3390/atmos16091078},
  url = {https://doi.org/10.3390/atmos16091078}
}