Liu et al. (2026) Layer-specific soil water dynamics and root-zone partitioning in alfalfa under regulated deficit irrigation
Identification
- Journal: Agricultural Water Management
- Year: 2026
- Date: 2026-11-14
- Authors: Minguo Liu, Xiuping Zhang, Yihao Zhao, Zhixin Zhang, Jing Zhang, Huimin Yang, Peizhi Yang
- DOI: 10.1016/j.agwat.2026.110942
Research Groups
- College of Grassland Agriculture, Northwest A&F University, Yangling, China
- State Key Laboratory of Herbage Improvement and Grassland Agro-ecosystems, College of Pastoral Agriculture Science and Technology, Lanzhou University, Lanzhou, China
- College of Grassland Science, Gansu Agricultural University, Lanzhou, China
Short Summary
This study developed and evaluated a one-dimensional layered soil water balance model to quantify depth-specific soil water dynamics and root-zone partitioning in alfalfa under regulated deficit irrigation (RDI) strategies over a three-year field trial in arid northwestern China, revealing an increasing dependence on subsoil water with stand maturation.
Objective
- To evaluate the performance and parameter sensitivity of a developed layered soil water balance model across years and irrigation treatments.
- To quantify depth-specific soil-water dynamics and layer water-balance components in alfalfa under various regulated deficit irrigation treatments.
Study Configuration
- Spatial Scale: A three-year field trial (2017–2019) in Suzhou District, Jiuquan City, Gansu Province, China (39.67°N, 98.80°E; 1368 m a.s.l.). The model simulated a 0–3.0 m soil profile, discretized into 30 computational layers of 0.10 m thickness, with observations aggregated into 0–0.6 m, 0.6–1.2 m, and 1.2–1.8 m layers.
- Temporal Scale: Three-year growing seasons (2017–2019). Daily weather data, soil water content monitored every 3 days, plant height and leaf area index (LAI) every 6–7 days, and root depth assessed annually.
Methodology and Data
- Models used:
- A simplified one-dimensional layered soil water balance model (SoilWatPy, developed in this study).
- van Genuchten-Mualem functions for soil-water retention and unsaturated hydraulic conductivity.
- Darcy-Buckingham fluxes for vertical water movement.
- FAO-56 Penman–Monteith equation for reference evapotranspiration (ET₀).
- FAO-56 two-stage surface depletion approach for actual soil evaporation.
- Feddes pressure-head response function for root water uptake.
- Rosetta H3 for pedotransfer estimates of soil hydraulic parameters.
- Logistic model for fitting daily LAI values.
- Data sources:
- Field trial observations (2017–2019) in arid northwestern China, including seven irrigation treatments (full irrigation and six RDI strategies).
- Daily weather data from a nearby meteorological station.
- Soil water content measured using a manual time-domain reflectometry (TDR) system at 0.2 m depth intervals down to 2.0 m.
- Alfalfa plant height and LAI measured using a measuring tape and a Sunscan canopy analyzer.
- Root depth assessed annually using a root drill.
- Soil properties: Measured texture, bulk density, field capacity, and physically screened saturated water content.
- Published parameter values for alfalfa (e.g., Feddes thresholds).
Main Results
- The model demonstrated satisfactory overall performance with an R² of 0.73, a Root Mean Square Error (RMSE) of 3.15 percentage points, a Mean Absolute Error (MAE) of 2.45 percentage points, and a mean bias of +0.13 percentage points across 1302 observations, although weak temporal agreement was noted in the deepest layer (1.2–1.8 m).
- Sensitivity analysis indicated that soil hydraulic properties, particularly the van Genuchten shape parameter (n), field-capacity water content, van Genuchten pore-connectivity parameter (l), unsaturated-conductivity scale (K0), and saturated water content, were the dominant controls on model response.
- Simulated soil water dynamics exhibited clear vertical stratification: the 0–0.6 m layer showed the largest seasonal fluctuations and strongest response to surface water inputs, while deeper layers (0.6–1.2 m and 1.2–1.8 m) exhibited attenuated temporal variations and progressive seasonal depletion.
- Root water uptake below 0.6 m significantly increased with stand age, contributing 2.1% in 2017, 22.6% in 2018, and 30.0% in 2019. Uptake from the 1.2–1.8 m layer was only detected in 2019, accounting for approximately 6.2% of total uptake, indicating a growing dependence on subsoil water during stand maturation.
- The model achieved negligible numerical water-balance closure, with a maximum absolute cumulative seasonal residual of 8.0 × 10⁻¹³ mm for any 0.60 m reporting layer.
Contributions
- Development of a transparent, mass-conservative, and computationally parsimonious process-based one-dimensional layered soil-water model (SoilWatPy) for depth-resolved water-balance accounting using commonly available field observations.
- Quantification of depth-specific soil water dynamics and root-zone partitioning in deep-rooted alfalfa under regulated deficit irrigation, addressing a gap due to limited stratified soil-water data.
- Elucidation of the distinct hydrological roles of different soil layers in alfalfa's root zone, demonstrating how deeper layers transition from passive storage to active water sources as roots extend.
- Provision of valuable insights for arid irrigation management, emphasizing the importance of depth-resolved root-zone water status assessment over sole reliance on surface measurements for optimizing deficit irrigation strategies.
Funding
- National Natural Science Foundation of China (32301495 and 32503285)
Citation
@article{Liu2026Layerspecific,
author = {Liu, Minguo and Zhang, Xiuping and Zhao, Yihao and Zhang, Zhixin and Zhang, Jing and Yang, Huimin and Yang, Peizhi},
title = {Layer-specific soil water dynamics and root-zone partitioning in alfalfa under regulated deficit irrigation},
journal = {Agricultural Water Management},
year = {2026},
doi = {10.1016/j.agwat.2026.110942},
url = {https://doi.org/10.1016/j.agwat.2026.110942}
}
Original Source: https://doi.org/10.1016/j.agwat.2026.110942