Liu et al. (2025) Soil hydrological processes and drought response of typical vegetation in arid regions under long-term climate patterns

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2025
• Date: 2025-11-12
• Authors: Jing Liu, Yunfei Chen, Xiuhua Liu, Yudong Lu, Ce Zheng, Changchun Shi, Zuyu Liu, Yaqin Gao, Leyi Zhang
• DOI: 10.1016/j.ejrh.2025.102919

Research Groups

• School of Water and Environment / Key Laboratory of Subsurface Hydrology and Ecological Effect in Arid Region of Ministry of Education, Chang’an University, Xi’an, China
• Key Laboratory of Eco-hydrology and Water Security in Arid and Semi-arid Regions of Ministry of Water resources, Chang’an University, Xi’an, China
• State Long-term Observation and Research Station for Mu Us Sand Land Ecosystem in Yulin of Shaanxi, Yulin, China

Short Summary

This study integrated field data and Hydrus-1D modeling to analyze long-term soil water dynamics and water-use strategies of five land types in China's arid Mu Us Sandy Land (1980–2022), revealing that Caragana korshinskii (CK) exhibits the strongest drought resistance and water conservation, while bare land (BL) is crucial for groundwater recharge.

Objective

• To elucidate the differences in water regulation mechanisms among vegetation types in terms of drought resistance, water use efficiency, and water conservation capacity.
• To quantify the drought response thresholds of each vegetation type.
• To propose optimized vegetation structure adjustment strategies based on the findings.

Study Configuration

• Spatial Scale: Arid desert region in northwest China, specifically the southeastern edge of the Mu Us Sandy Land. Experimental sites include the State Forestry Administration Yulin Desert Ecosystem Research Station and the Shaanxi Yulin Rare Sandy-plants Conservation Field.
• Temporal Scale: 43 years (1980–2022) for long-term modeling; field data collected from 2018 to 2023 for model calibration and validation.

Methodology and Data

• Models used: Hydrus-1D model (for one-dimensional vertical water flow and root water uptake), Standardized Precipitation Evapotranspiration Index (SPEI) (for drought assessment), Structural Equation Modeling (SEM) (for analyzing meteorological factor effects), Random Forest (for LAI inversion), moving average time series model (for LAI prediction).
• Data sources:
  • Field observations: TDR315 probes for soil moisture content at multiple depths, undisturbed soil samples for soil hydraulic parameters (particle size distribution, saturated water content, saturated hydraulic conductivity, dry bulk density), ground-penetrating radar (EKKO Conquest 100) for root distribution data.
  • Meteorological data: Micro-meteorological stations (2018–2023) for atmospheric temperature (T), relative humidity (RH), precipitation (P), solar radiation (Rn), and wind speed (U); National Automatic Weather Monitoring Station (1980–2022) for long-term meteorological data.
  • Satellite data: MODIS LAI product (MOD15A2H, 2002-present) and LANDSAT NDVI data (1984-present) from Google Earth Engine for Leaf Area Index (LAI) inversion and construction.

Main Results

• The Hydrus-1D model accurately simulated soil moisture dynamics across five land types, with mean index of agreement (d) of 0.766 and Root Mean Square Error (RMSE) of 0.013.
• Caragana korshinskii (CK) demonstrated the strongest drought resistance and water conservation capacity, maintaining 0.028 cm³⋅cm⁻³ of available water space and a 7 % deep rainfall-percolation rate even under continuous drought conditions.
• Bare land (BL) exhibited the highest soil moisture content (above 0.08 cm³⋅cm⁻³) and deep percolation rates (ranging from 14 % to 50 %, averaging ~20 % over 43 years), playing a critical role in groundwater recharge. A 10 % reduction in BL area was estimated to decrease deep percolation by approximately 5.28 %.
• Other vegetation types (Salix psammophila, Pinus sylvestris var. mongolica, Grassland) showed near-zero deep percolation and significant soil moisture deficits at root depths during drought periods.
• Vegetation transpiration (Tc) and soil water storage (SWS) exhibited a nonlinear response to Vapor Pressure Deficit (VPD). CK had the highest VPD threshold (1.61 kPa), followed by Grassland (GL) (1.32 kPa) and Salix psammophila (SP) (1.30 kPa), with Pinus sylvestris var. mongolica (PSVM) showing the lowest (1.07 kPa), indicating varying drought tolerance.
• Grassland (GL) displayed the highest water use efficiency, with a transpiration to total evapotranspiration (Tc/ET) ratio of approximately 75 %, compared to 50 %–60 % for other vegetation types.
• Meteorological factors (solar radiation, air temperature, VPD, precipitation) were key drivers of evapotranspiration and subsurface water dynamics, but plant self-regulation mechanisms became dominant under extreme drought conditions.

Contributions

• Provided a comprehensive, long-term (43-year) quantitative water balance analysis for five distinct land types in an arid sandy region, integrating field observations and numerical modeling.
• Elucidated the differentiated water regulation mechanisms (drought resistance, water use efficiency, water conservation capacity) and water utilization strategies among typical vegetation types.
• Quantified critical drought response thresholds (soil moisture and VPD) for various vegetation types, offering a robust framework for assessing vegetation adaptability to water stress.
• Highlighted the significant role of bare land in maintaining groundwater recharge and provided a mechanistic explanation for the observed decline in groundwater storage during regional greening efforts.
• Proposed optimized vegetation structure adjustment strategies, including increasing Caragana korshinskii planting, maintaining an appropriate proportion of bare land, and adopting mixed strip planting, to enhance ecological resilience and sustainable water resource management in sandy ecosystems.

Funding

• National Natural Science Foundation of China [grant number 42372288, 42202279, 41877179, U2243204]
• Fundamental Research Funds for the Central Universities, CHD [grant number 300102294903, 300102294723]

Citation

@article{Liu2025Soil,
  author = {Liu, Jing and Chen, Yunfei and Liu, Xiuhua and Lu, Yudong and Zheng, Ce and Shi, Changchun and Liu, Zuyu and Gao, Yaqin and Zhang, Leyi},
  title = {Soil hydrological processes and drought response of typical vegetation in arid regions under long-term climate patterns},
  journal = {Journal of Hydrology Regional Studies},
  year = {2025},
  doi = {10.1016/j.ejrh.2025.102919},
  url = {https://doi.org/10.1016/j.ejrh.2025.102919}
}