Dai et al. (2026) Analysis of Seasonal Propagation Dynamics and the Potential Driving Factors from Meteorological to Soil Moisture Drought

Identification

• Journal: Water Resources Management
• Year: 2026
• Date: 2026-02-21
• Authors: Meng Dai, Ping Feng, Jie Li, Renjie Tao
• DOI: 10.1007/s11269-026-04497-3

Research Groups

State Key Laboratory of Hydraulic Engineering Intelligent Construction and Operation, Tianjin University, Tianjin, 300072, China

Short Summary

This study investigates the seasonal propagation dynamics from meteorological drought to soil moisture drought (SMD) and their driving factors in the Luanhe River Basin, China, using the Soil and Water Assessment Tool (SWAT) model and copula functions. It reveals distinct seasonal propagation times, with the shortest in summer and longest in spring, and identifies hydrometeorological, teleconnection, and land use factors as significant drivers of these dynamics.

Objective

• To explore the propagation law from meteorological drought to soil moisture drought (SMD) based on the SWAT model in the Luanhe River Basin (LRB).

Study Configuration

• Spatial Scale: Panjiakou Reservoir Basin within the Luanhe River Basin, China (total drainage area 44,750 km²), analyzed at the subbasin level.
• Temporal Scale: Data from 1960–2018 (59 years); analysis period 1962–2018 (57 years); dynamic analysis using a 29-year moving window; seasonal analysis (spring: March–May, summer: June–August, autumn: September–November).

Methodology and Data

• Models used:
  • Soil and Water Assessment Tool (SWAT) model for hydrological simulation and soil moisture data generation.
  • Mann–Kendall (MK) trend test for trend analysis of hydro-meteorological data and drought propagation time.
  • Copula function (Clayton copula) for calculating conditional probability and drought propagation time (PT).
  • Random Forest for exploring driving factors of drought propagation time.
• Data sources:
  • Daily precipitation data from 10 rain gauges (1960–2018).
  • Daily relative humidity, daily relative wind speed, daily maximum and minimum temperatures from 4 meteorological stations (1960–2018) (National Meteorological Information Center of China).
  • Soil data from the Harmonized World Soil Database (HWSD).
  • Land use data (1 km spatial resolution for 2015) from the Resource and Environment Science and Data Center, Chinese Academy of Sciences.
  • Daily runoff data from hydrological stations (Sandaohezi, Hanjiaying, Chengde) (1962–2018) (Haihe Basin Hydrological Yearbook).
  • Sunspots data (1962–2018) from SIDC.
  • Niño 3.4 Index (ENSO) from PSL NOAA.
  • Pacific Decadal Oscillation (PDO) from JMA.
  • Arctic Oscillation (AO) Index from Lijianping.cn.
  • Drought indices: Standardized Precipitation Index (SPI) for meteorological drought, Standardized Soil Moisture Index (SSMI) for soil moisture drought.

Main Results

• The SWAT model showed good performance in simulating runoff in the LRB, with NSE, R², and KGE consistently above 0.58, 0.66, and 0.51, respectively, across three hydrological stations.
• Soil moisture (SM) simulated by the SWAT model exhibited an insignificant upward trend in spring and autumn, and an insignificant downward trend in summer.
• Meteorological drought (SPI) and soil moisture drought (SSMI) showed a strong correlation (Pearson correlation coefficients > 0.5), with the optimal SPI time scale for correlation concentrated in 1–12 months.
• The static drought propagation time (DPT) from meteorological drought to SMD was longest in spring (average 9.8 months), shortest in summer (average 2.9 months), and intermediate in autumn (average 6.3 months).
• Dynamic DPT analysis using a moving window revealed a significant upward trend in summer and autumn overall, but a significant downward trend in the middle region of the Panjiakou Reservoir Basin in spring.
• Hydrometeorological factors: SM, evapotranspiration (potential and actual), and water balance were identified as the main driving factors of DPT. Snowmelt and evapotranspiration were key in spring; water supply (precipitation, surface runoff, SM, water balance) and potential evapotranspiration in summer; and water supply and actual evapotranspiration in autumn.
• Teleconnection factors: Sunspots and ENSO significantly influenced drought propagation in summer and autumn, while sunspots, PDO, and the coupling effects of teleconnection factors were main drivers in spring.
• Land use type: A significant positive correlation (r = 0.469, p < 0.05) was found between the proportion of forest area and spring DPT, suggesting longer DPT with increased forest cover. A significant negative correlation (r = -0.532, p < 0.05) was observed for grassland area, indicating shorter DPT with increased grassland. Agricultural land showed an insignificant positive correlation.

Contributions

• Provides a comprehensive analysis of seasonal propagation dynamics from meteorological drought to soil moisture drought (SMD) in the Luanhe River Basin, a region where SMD studies are less common.
• Utilizes the physically-based SWAT model to generate spatially continuous, long-term soil moisture data, addressing the scarcity of large-scale observational SM data for drought propagation studies.
• Employs a copula function-based conditional probability approach to capture both linear and nonlinear relationships in drought propagation time, improving upon traditional linear methods.
• Systematically explores the driving factors of drought propagation time, including hydrometeorological variables, large-scale teleconnection patterns (ENSO, PDO, AO, sunspots, and their coupling effects), and land use types, using the random forest method.
• Offers valuable insights for understanding drought propagation laws and informing rational water resource planning and drought prevention strategies in the study area and similar regions.

Funding

National Natural Science Foundation of China (No. 52079086)

Citation

@article{Dai2026Analysis,
  author = {Dai, Meng and Feng, Ping and Li, Jie and Tao, Renjie},
  title = {Analysis of Seasonal Propagation Dynamics and the Potential Driving Factors from Meteorological to Soil Moisture Drought},
  journal = {Water Resources Management},
  year = {2026},
  doi = {10.1007/s11269-026-04497-3},
  url = {https://doi.org/10.1007/s11269-026-04497-3}
}