Liu et al. (2026) Evidence of vertical soil hydraulic heterogeneity regulating hydrothermal simulations in Qinghai–Tibetan Plateau wetlands

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-10-15
• Authors: Rong Liu, Xin Wang, Zuoliang Wang, Yongjie Pan, Yan Song
• DOI: 10.5194/hess-30-6289-2026

Research Groups

• State Key Laboratory of Cryospheric Science and Frozen Soil Engineering, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, China
• Department of City and Regional Planning, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Short Summary

This study provides the first process-based and depth-resolved characterization of layered soil hydraulic properties in Qinghai–Tibetan Plateau wetlands using laboratory experiments. It demonstrates that implementing this vertical hydraulic heterogeneity in the Noah-MP land surface model significantly alters water and energy partitioning, highlighting its crucial role in regulating hydrothermal simulations.

Objective

• To characterize the depth-resolved soil hydraulic heterogeneity in Qinghai–Tibetan Plateau wetlands and quantify its impact on hydrothermal simulations in the Noah-MP land surface model.

Study Configuration

• Spatial Scale: Sanjiangyuan region, Qinghai–Tibetan Plateau, China. Field observations were conducted at the Sanjiangyuan Land Surface Process Observation Station in Maqu County (33.5° N, 101.9° E; elevation 3436 m). Soil cores were collected from 11 depth intervals, covering 0 to 160 cm.
• Temporal Scale: Mean annual precipitation and temperature data for 1967–2017. Model simulations were conducted for the summer season (June–August) of 2023, with specific analyses for a wet period (15–16 July 2023) and a dry period (12–13 August 2023).

Methodology and Data

• Models used:
  • Noah-MP land surface model (v5.0)
  • Clapp–Hornberger (CH) hydraulic scheme (a variant of the Brooks–Corey model)
  • Jarvis stomatal conductance scheme (for vegetation processes)
  • Niu and Yang (2006) parameterizations for frozen soil processes
  • Shuffled Complex Evolution (SCE-UA) global optimization algorithm for parameter inversion
  • Markov Chain Monte Carlo (MCMC) approach for uncertainty assessment
• Data sources:
  • Observation:
    • Undisturbed soil cores collected from 11 depth intervals (0–160 cm).
    • Eddy covariance (EC) system for latent heat flux (LHF) and sensible heat flux (SHF).
    • Automatic weather station for air temperature, relative humidity, wind speed, downward shortwave and longwave radiation, and precipitation.
    • Soil temperature and volumetric water content (VWC) profiles at 5, 10, 20, 40, and 80 cm depths.
    • Infrared surface temperature sensors, soil heat flux plates, and an evaporation lysimeter.
    • Gravimetric sampling for soil moisture sensor calibration.
  • Laboratory Experiments:
    • Potassium dichromate oxidation method for soil organic carbon (SOC).
    • Von Post method for fiber content (FC).
    • Ring core method for bulk density (BD).
    • Hydrometer method for particle size distribution (PSD).
    • Falling head method for saturated hydraulic conductivity (Ks).
    • Gravimetric procedure for saturated volumetric water content (θs).
    • HYPROP-2 system for transient evaporation experiments (matric potential and sample mass for near-saturated water retention and hydraulic conductivity curves).
    • WP4C dew point potentiometer for dry range water retention curves.
    • CT scan for visualizing soil pore structure.

Main Results

• Vertical Hydraulic Heterogeneity: Wetland soils exhibit a pronounced vertical gradient. The surface layer (0–8 cm) is characterized by high θs (0.64–0.70 m³ m⁻³) and Ks (4.2 × 10⁻⁶ – 9.9 × 10⁻⁶ m s⁻¹), low BD (630–810 kg m⁻³), and high SOC (45–99 g kg⁻¹), indicating macropore dominance and rapid drainage. With increasing depth, θs decreases (0.34–0.46 m³ m⁻³) and Ks decreases (1.6 × 10⁻⁸ – 5.6 × 10⁻⁷ m s⁻¹), while BD increases (1570–1740 kg m⁻³) and SOC decreases (0.88–8.97 g kg⁻¹), approaching mineral soil characteristics with stronger water retention.
• Clapp–Hornberger (CH) Parameters: Inverted CH parameters show clear vertical contrasts: the empirical exponent 'b' is approximately 3.2 in the surface layer, peaks at approximately 5.1 in the subsurface (10–20 cm), and decreases to approximately 1.0 in deeper layers. The air entry suction (ψs) also varies significantly, with larger absolute values near the surface (≈−0.378 m) and in deeper layers (≈−0.518 m), and a minimum in the subsurface (≈−0.071 m).
• Model Sensitivity to Parameters: The 'b' parameter exerts the strongest influence on LHF and SHF simulations. Using the surface layer 'b' (3.215) significantly improved LHF simulation accuracy during summer (RMSE reduced from 9.45 W m⁻² to 3.87 W m⁻²). Adjusting θs markedly reduced the RMSE of soil moisture at 20 cm (from 0.063 m³ m⁻³ to 0.047 m³ m⁻³), highlighting its role in storage capacity. Ks showed weak sensitivity under wet summer conditions.
• Wet vs. Dry Period Responses: During wet periods, a lower 'b' (representing more macropores) enhanced rapid drainage and improved LHF peak simulation (RMSE reduced from 42 W m⁻² to 28 W m⁻²). During dry periods, low 'b' schemes captured rapid rainfall-evapotranspiration events but intensified underestimation under prolonged dry conditions, while higher 'b' improved soil moisture retention.
• Limitations of CH Scheme: The single power-law structure of the CH model struggles to simultaneously represent rapid macropore drainage and micropore capillary retention, leading to structural biases in water retention and hydraulic conductivity curves, particularly near saturation.

Contributions

• Provides the first process-based and depth-resolved characterization of layered pore structures and hydraulic properties in Qinghai–Tibetan Plateau wetlands.
• Offers quantitative evidence that pore-scale structure and vertical hydraulic heterogeneity significantly regulate profile-scale hydrothermal responses in these unique ecosystems.
• Establishes a physically grounded basis for incorporating vertically heterogeneous hydraulic processes into land surface models, improving their representation of complex wetland hydrology.
• Quantifies the sensitivity of Noah-MP simulations to individual Clapp–Hornberger parameters under varying hydrological conditions, highlighting the critical roles of 'b' and 'ψs' in energy partitioning and 'θs' in soil moisture storage.
• Identifies the structural limitations of simplified empirical hydraulic functions (like the CH model) in representing the nonlinear release and multiscale pore responses of wetland soils, advocating for more physically based models (e.g., van Genuchten, VG-PDI).

Funding

• National Natural Science Foundation of China (grant nos. 42575094 and 52539002)
• Joint Research Program for Ecological Conservation and High Quality Development of the Yellow River Basin (Phase I) (grant no. 2022-YRUC-01-0601)
• Program of the State Key Laboratory of Cryospheric Science and Frozen Soil Engineering (grant no. CSFSE-ZZ-2407)

Citation

@article{Liu2026Evidence,
  author = {Liu, Rong and Wang, Xin and Wang, Zuoliang and Pan, Yongjie and Song, Yan},
  title = {Evidence of vertical soil hydraulic heterogeneity regulating hydrothermal simulations in Qinghai–Tibetan Plateau wetlands},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-6289-2026},
  url = {https://doi.org/10.5194/hess-30-6289-2026}
}