Wang et al. (2025) Drought onsets and their driving factors for multiple drought types in the Yangtze River Basin

Identification

• Journal: Climate Dynamics
• Year: 2025
• Date: 2025-12-09
• Authors: Caiyuan Wang, Peng Yang, Jun Xia, Heqing Huang, Lu Chen, Kaiya Sun, Xixi Lu
• DOI: 10.1007/s00382-025-07953-9

Research Groups

• Hubei Key Laboratory of Regional Ecology and Environmental Change, School of Geography and Information Engineering, China University of Geosciences, Wuhan, China
• State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan, China
• Department of Geography, National University of Singapore, Singapore

Short Summary

This study analyzed the spatiotemporal evolution and driving mechanisms of meteorological, agricultural, and hydrological drought onset characteristics in the Yangtze River Basin (YRB), revealing that temperature is the dominant driver with complex nonlinear interactions among factors, especially under extreme conditions, which provides a scientific basis for drought risk prevention.

Objective

• To analyze the spatiotemporal evolution patterns of drought onset characteristics (Drought Onset Time and Drought Onset Speed) for multiple drought types in the Yangtze River Basin.
• To identify and quantify the driving factors and their interactive mechanisms for conventional drought onset characteristics in the YRB.
• To investigate how the driving mechanisms of drought onset characteristics vary under extreme climate conditions, specifically during the 2022 extreme drought event.

Study Configuration

• Spatial Scale: Yangtze River Basin (YRB), central China (90.5° E–122.5° E, 24.5° N–35.75° N), covering 1.8 million km², divided into 12 secondary sub-basins. Data used has a 0.25° spatial resolution.
• Temporal Scale: Daily-scale data spanning from 1940 to 2024. Drought indices were computed at a 30-day timescale. Analysis included the 2022 extreme drought event.

Methodology and Data

• Models used: Extreme Gradient Boosting (XGB) model, Shapley additive explanation (SHAP) for driving mechanism analysis, Mann–Kendall (MK) trend test for trend analysis.
• Data sources: ERA5 reanalysis data (European Centre for Medium-Range Weather Forecasts; ECMWF) including precipitation (PRE), 0–7 cm soil moisture, total runoff, 2 m dew point temperature, 2 m air temperature (TEM), 10 m wind speed (WS), net surface solar radiation (NSSR), surface solar radiation, and surface pressure (SP). Potential evapotranspiration (PET) was calculated using the FAO-56 Penman–Monteith method, and vapor pressure deficit (VPD) was derived from temperature and dew-point temperature. Drought types were characterized by the standardized precipitation evapotranspiration index (SPEI), standardized soil moisture index (SSI), and standardized runoff index (SRI).

Main Results

• Drought Onset Time (DOT) in the YRB ranged from 10 days for mild meteorological drought to 112 days for extreme hydrological drought, generally increasing with drought severity and type propagation (meteorological to hydrological). Drought Onset Speed (DOS) ranged from 0.14 day⁻¹ for mild meteorological drought to 0.02 day⁻¹ for severe hydrological drought, decreasing with severity and type propagation. The middle reaches typically exhibited shorter DOT and faster DOS.
• Temperature (TEM) was identified as the primary driver for both DOT and DOS in the YRB, exerting negative effects on DOT and positive effects on DOS. Potential Evapotranspiration (PET) showed similar effects to TEM, while Precipitation (PRE) delayed onset. Vapor Pressure Deficit (VPD), Net Surface Solar Radiation (NSSR), and Surface Pressure (SP) exhibited nonlinear influences.
• During the 2022 extreme drought, TEM's role as the primary driver diminished, with other meteorological factors (VPD, PET, SP, NSSR) becoming dominant, particularly for hydrological droughts. When TEM exceeded 20 °C, elevated PET, SP, and NSSR mitigated TEM's impact on DOT and DOS, while increased VPD amplified this influence.

Contributions

• Provided a comprehensive analysis of drought onset characteristics (DOT and DOS) across multiple drought types (meteorological, agricultural, hydrological) and severity levels (mild to extreme) in the Yangtze River Basin.
• Utilized an advanced, explainable machine learning framework (XGB-SHAP) to quantify the main and interactive effects of multivariate meteorological drivers on drought onset, moving beyond studies focused on isolated drivers and linear relationships.
• Revealed the dynamic shifts in dominant driving mechanisms under extreme climate conditions, specifically during the unprecedented 2022 YRB drought, offering crucial insights for enhancing regional drought monitoring and early warning systems.

Funding

• National Natural Science Foundation of China (42442105 and 42207078)
• CUG Scholar- Scientific Research Funds at China University of Geosciences (Wuhan) (2022166)

Citation

@article{Wang2025Drought,
  author = {Wang, Caiyuan and Yang, Peng and Xia, Jun and Huang, Heqing and Chen, Lu and Sun, Kaiya and Lu, Xixi},
  title = {Drought onsets and their driving factors for multiple drought types in the Yangtze River Basin},
  journal = {Climate Dynamics},
  year = {2025},
  doi = {10.1007/s00382-025-07953-9},
  url = {https://doi.org/10.1007/s00382-025-07953-9}
}