Overview
Water movement through unsaturated soils is one of the most fundamental processes in environmental science, controlling everything from agricultural productivity to contaminant transport and groundwater recharge. Unlike saturated flow where all pores are filled with water, unsaturated flow involves complex interactions between water, air, and the soil matrix.
Our research combines advanced measurement techniques, theoretical analysis, and numerical modeling to understand and predict water flow under diverse conditions—from desert soils experiencing extreme dryness to agricultural fields under irrigation.
The Richardson-Richards Equation
At the heart of unsaturated flow modeling is the Richardson-Richards equation (RRE), a partial differential equation that describes water movement based on gradients in water potential. The RRE accounts for:
- Capillary forces: Water retention in small pores
- Gravity: Downward water movement
- Hydraulic conductivity: Soil's ability to transmit water, which varies dramatically with water content
The nonlinear nature of the RRE makes it challenging to solve, especially for heterogeneous soils with contrasting layers or under dynamic boundary conditions. Our work develops novel solution approaches and evaluates their accuracy across diverse soil conditions.
Key Research Areas
1. Hydraulic Property Characterization
Accurate modeling requires knowing two key soil functions:
- Water retention curve (WRC): Relationship between water content and matric potential
- Hydraulic conductivity function (HCF): How water conductivity changes with water content or potential
We use advanced laboratory methods including:
- Evaporation method (HYPROP) for simultaneous WRC and HCF measurement
- Tempe cells for water retention at multiple pressures
- Infiltration experiments for field-scale hydraulic properties
2. Scale Mismatch Problems
A persistent challenge is that laboratory-measured properties often don't represent field-scale behavior. We address this through:
- Inverse modeling: Estimating field-scale properties from in situ moisture measurements
- Physics-informed machine learning: Learning hydraulic functions directly from field data
- Upscaling methods: Mathematical approaches to translate small-scale measurements to larger scales
3. Infiltration and Redistribution
We investigate how water enters soil (infiltration) and subsequently redistributes after rainfall or irrigation ceases. Key questions include:
- How do soil structure and macropores affect infiltration rates?
- What controls the depth and rate of water penetration?
- How long does water remain available to plants versus draining to groundwater?
4. Evaporation from Bare Soils
Desert soils and agricultural fields between crops lose significant water to evaporation. Our research examines:
- Stage transitions in evaporation (energy-limited vs. diffusion-limited)
- Effects of soil texture and structure on evaporative losses
- Improved hydraulic functions for very dry conditions (beyond pF 3.8)
Improved Hydraulic Functions
Peters-Durner-Iden (PDI) Model
Traditional van Genuchten functions often fail for very dry soils. We've demonstrated that the PDI model significantly improves predictions of:
- Moisture redistribution in desert environments
- Evaporation rates from drying soils
- Water retention between pF 2 and pF 3.8 (field capacity to near wilting point)
Applications
Agricultural Water Management
Understanding infiltration and redistribution helps optimize irrigation scheduling, reducing water waste while maintaining crop productivity. Our models predict:
- How much water infiltrates versus runs off
- Water availability in the root zone over time
- Deep drainage losses below the root zone
Wastewater Irrigation
Treated wastewater is increasingly used for irrigation in water-scarce regions. We've shown that long-term wastewater application:
- Reduces soil hydraulic conductivity and infiltration capacity
- Increases runoff and erosion risk
- Requires adjusted irrigation management strategies
Groundwater Recharge
Predicting deep percolation is essential for groundwater management. Our models help quantify recharge under different:
- Climate scenarios (rainfall intensity and distribution)
- Land use practices (tillage, cover crops)
- Soil types and layering
Numerical Modeling Approaches
We employ multiple modeling platforms depending on the problem:
- HYDRUS-1D/2D: Established finite element solver for the Richards equation
- Physics-informed neural networks: Novel AI approach that learns solutions from sparse data
- Analytical solutions: Simplified cases for validation and insight
Critical to all modeling is validation. We rigorously compare simulations against:
- Analytical solutions where available
- Laboratory experiments with known boundary conditions
- Field measurements from instrumented plots
Recent Publications
- Jensen’s Inequality Quantifies How Temporal Averaging of Moisture Inputs Affects Modeled Soil Respiration Across Continental Scales.Rojas, Y. T. P., & Ghezzehei, T. A.JGR Biogeosciences. 2026.
Abstract
Understanding how temporal patterns of moisture variability control biogeochemical responses remains a fundamental challenge in Earth system science. Jensen’s inequality provides a mathematical framework for quantifying when episodic environmental events dominate over mean conditions. We applied this framework to continental-scale AmeriFlux data (134.5 million hourly observations from 2,004 soil moisture sensors) to quantify how moisture distribution patterns control soil respiration responses across environmental gradients. Sensors in dry regions show large Jensen’s inequality effects (median temporal averaging difference of -63.6%) because they experience highly skewed moisture distributions where brief wet periods drive disproportionate respiratory responses. Wet regions show minimal effects (median -27.1%) because they have more uniform moisture distributions. Data density analysis reveals that sensors operating at θ≈0.05 exhibit severe temporal averaging effects, while sensors at θ≈0.45 show minimal effects, demonstrating the mechanistic basis for where ecosystems operate on the moisture-respiration relationship. Climate gradient analysis shows systematic transitions from severe effects in arid systems to moderate effects in humid systems. Depth analysis reveals that surface soils experience maximum episodic event importance while deeper soils show reduced effects due to environmental buffering. Moisture-temperature coupling demonstrates systematic negative correlations in water-limited systems, indicating that environmental co-variation modulates biogeochemical responses. Jensen’s inequality emerges as a diagnostic tool for identifying when moisture variability patterns dominate biogeochemical processes, with continental-scale patterns revealing fundamental controls on episodic event importance across ecosystems.BibTeX
@article{2026-PerezRojas, title = {Jensen's Inequality Quantifies How Temporal Averaging of Moisture Inputs Affects Modeled Soil Respiration Across Continental Scales}, language = {en}, journal = {JGR Biogeosciences}, author = {Rojas, Yulissa T. Perez and Ghezzehei, Teamrat Afewerki}, year = {2026}, month = jun, doi = {10.1029/2025JG009613}, pdf = {https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2025JG009613}, publisher = {American Geophysical Union (AGU)}, research-theme = {water-flow, soil-structure} } - Destabilization of Buried Carbon Under Changing Moisture Regimes.Dolui, M., Nel, T., McMurtry, A. R., Chacon, S., Mason, J. A., Phillips, L. M., … Ghezzehei, T. A.SOIL, 12, 561–580. 2026.
Abstract
Paleosols formed by the burial of topsoil during landscape evolution can sequester substantial amounts of soil organic carbon (SOC) over millennia due to protection from surface disturbances. We investigated the moisture sensitivity of buried SOC storage in the Brady paleosol, a loess-derived soil in Nebraska, USA, where historical aeolian deposition during the Pleistocene–Holocene transition buried soils up to 6 m deep. Topsoils from erosional (up to 1.8 m depth) and depositional (up to 5.8 m depth) transects were incubated under two moisture regimes – continuous wetting (60 % water-holding capacity) and repeated drying–rewetting – to assess SOM vulnerability to changing hydrologic conditions. SOC decomposition rates modeled from CO2 fluxes were consistently higher in erosional than depositional settings, with surface re-exposure of Brady soils enhancing microbial accessibility and destabilization. A two-pool model showed that >96 % of SOC was stored in a slow-cycling pool, particularly in deeply buried soils where stabilization was linked to mineral association, fine particles, and Ca-mediated flocculation. However, this pool decomposed more rapidly in shallower Brady soils (higher turnover rate relative to buried soil), reflecting increased microbial responsiveness to surface-driven processes. Drying–rewetting cycles caused greater SOC losses from Brady soils than continuous wetting, despite the dominance of the slow pool and depletion of labile SOC. These cycles also accelerated fast pool decay in modern soils and erosional transects, whereas burial dampened variability in Brady soils. Although continuous wetting increased overall decay in burial transects during the incubation period, wet–dry cycles destabilized the slow pool, which may result in greater long-term SOC loss. Together, these results underscore the importance of burial depth, geomorphic context, and moisture regime in shaping the long-term vulnerability of ancient SOC under climate change.BibTeX
@article{p2025-Dolui-SOIL, title = {Destabilization of Buried Carbon Under Changing Moisture Regimes}, language = {en}, journal = {SOIL}, volume = {12}, pages = {561--580}, doi = {10.5194/soil-12-561-2026}, pdf = {https://soil.copernicus.org/articles/12/561/2026/soil-12-561-2026.pdf}, author = {Dolui, Manisha and Nel, Teneille and McMurtry, Abbygail R. and Chacon, Stephanie and Mason, Joseph A. and Phillips, Laura M. and Marin-Spiotta, Erika and de Graaff, Marie-Anne and Berhe, Asmeret A. and Ghezzehei, Teamrat A.}, year = {2026}, research-theme = {soil-structure, water-flow} } - Impact of almond shell biochar properties and application rate on soil physical and hydraulic characteristics.Thao, T., Lopez, V. D., Gonzales, M., Berhe, A. A., Diaz, G., & Ghezzehei, T. A.Sustainable Environment, 11(1), 2485688. 2025.
Abstract
We conducted two 64-day incubation experiments to assess how locally produced almond-shell biochar influences soil physical and hydraulic properties. Biochar was created using slow pyrolysis at different temperatures (350 °C or 700 °C), separated into different particle sizes (<250 μm or 1–2 mm), and applied at 10 ton/ha or 60 ton/ha to a coarse-textured soil. While our analysis shows that biochar yielded greater cation exchange capacity (CEC) and specific surface area (SSA) with increasing pyrolysis temperature and finer particle size, its contributions to improving soil hydraulic properties were marginal. In the first experiment, the addition of biochar at high rates slightly improved water stable aggregate (WSA) (3.8%–5.3% increase) but has no effect on saturated hydraulic conductivity (Ksat). Soil respiration measured throughout the experiment were not significantly different among treatments. In the second experiment, the addition of biochar increased soil infiltration rate at the initial stage (8.18E–4 cm/s), but this effect diminished over time. WSA was lower for biochar amended soil and lowest at high application rates (5%–21% reduction). Cumulative carbon dioxide (CO2) flux varied between biochar particle sizes and rates. Additionally, a significant difference between the two experiments was also observed, with cumulative CO2 (38%–56% greater) and WSA (11%–40%) being inversely correlated. Our findings suggest that almond-shell derived biochar has a limited impact on arable loamy sand soil properties, specifically for water retention under short-term conditions.BibTeX
@article{Thao31122025, author = {Thao, Touyee and Lopez, Vivian D. and Gonzales, Melinda and Berhe, Asmeret A. and Diaz, Gerardo and Ghezzehei, Teamrat A.}, title = {Impact of almond shell biochar properties and application rate on soil physical and hydraulic characteristics}, journal = {Sustainable Environment}, volume = {11}, number = {1}, pages = {2485688}, year = {2025}, publisher = {Taylor \& Francis}, doi = {10.1080/27658511.2025.2485688}, pdf = {https://doi.org/10.1080/27658511.2025.2485688}, research-theme = {water-flow, soil-structure, sustainable-agriculture} }
- Two Decades of Conservation Agriculture Enhances Soil Structure, Carbon Sequestration, and Water Retention in Mediterranean Soils.Alvarez-Sagrero, J., Berhe, A. A., Chacon, S. S., Mitchell, J. P., & Ghezzehei, T. A.Soils, EGUsphere [Revision Resubmitted].
Abstract
Conservation agriculture offers a pathway for enhancing soil health with climate co-benefits in Mediterranean agricultural systems. This study examined long-term impacts of combining no-till management with cover cropping over 20 years in California’s Central Valley, providing rare insights into soil system equilibrium under sustained conservation management. We assessed soil physical, chemical, and structural properties comparing reduced tillage with cover crops to standard tillage without cover crops, employing density fractionation and spectroscopic analysis to understand carbon protection mechanisms. After two decades, conservation agriculture achieved dynamic equilibrium characterized by fundamental shifts in carbon stabilization pathways. Water-stable aggregate analysis revealed the most pronounced management effects, with conservation practices exhibiting 136% greater stability, indicating substantial improvements in soil structural integrity. These structural enhancements corresponded with a reorganization of carbon protection mechanisms, demonstrating that physical protection within aggregates becomes a dominant carbon stabilization pathway under long-term conservation management. Mineral-associated organic carbon saturation analysis revealed that both management systems remained well below theoretical maximum capacity, indicating substantial remaining potential for carbon sequestration even after reaching equilibrium. Physical property improvements included 15% lower bulk density and 13% greater water retention at field capacity. Our findings demonstrate that two decades of conservation agriculture fundamentally transforms soil functioning through aggregate-mediated physical protection.BibTeX
@article{2025-AlvarezSagrero, title = {Two Decades of Conservation Agriculture Enhances Soil Structure, Carbon Sequestration, and Water Retention in Mediterranean Soils}, language = {en}, journal = {Soils, EGUsphere [revision resubmitted]}, author = {Alvarez-Sagrero, Jennifer and Berhe, Asmeret Asefaw and Chacon, Stephany S. and Mitchell, Jeffrey P. and Ghezzehei, Teamrat A.}, doi = {10.5194/egusphere-2025-6047}, pages = {}, research-theme = {sustainable-agriculture, soil-structure, water-flow} } - Interpretable Soil Water Retention Prediction Using Hierarchical Attention Networks with Uncertainty Quantification.Ghezzehei, T. A.Water Resources Research [Revision Resubmitted].
Abstract
Understanding which soil properties control water retention at different moisture states remains a critical challenge for predicting how soils respond to management and environmental change. Traditional pedotransfer functions predict parameters of equations like van Genuchten’s model, but parameter correlations create ill-posed inverse problems that propagate errors and reduce prediction accuracy. Fundamentally, these parameter-based approaches cannot reveal which properties matter most at different water potentials or how texture-structure interactions vary across the moisture spectrum—knowledge essential for mechanistic understanding and management applications. We present Hierarchical Attention-Based Pedotransfer Function (HABIT) to address these limitations through two objectives: (1) achieve superior prediction accuracy through direct moisture prediction that circumvents parameter correlation errors, and (2) quantify property importance and interactions across moisture regimes through interpretable attention mechanisms. The model employs neural network attention architectures that dynamically weight soil properties based on water potential, enforces physical constraints including monotonicity, and accommodates incomplete soil characterization through hierarchical training. We evaluate HABIT against traditional parameter-based approaches using comprehensive international soil databases spanning global textural and structural variability. Attention weight analysis reveals moisture-dependent property interactions, including asymmetric texture-structure information flow, organic carbon’s dual structural and sorptive roles, and saturated hydraulic conductivity’s diagnostic function in integrating pore network information. These patterns generate testable predictions while explaining how direct prediction with adaptive property weighting captures soil-specific retention behaviors that parametric forms miss. HABIT demonstrates that properly designed machine learning architectures serve as prediction tools and scientific discovery platforms, providing frameworks for understanding texture-structure interactions that remain qualitatively understood but quantitatively elusive.BibTeX
@article{2025-Ghezzehei, title = {Interpretable Soil Water Retention Prediction Using Hierarchical Attention Networks with Uncertainty Quantification}, language = {en}, journal = {Water Resources Research [revision resubmitted]}, author = {Ghezzehei, Teamrat Afewerki}, pages = {}, research-theme = {machine-learning, water-flow} }
Related Research
Key Equipment
- HYPROP system for evaporation method
- Tempe pressure cells
- Hood infiltrometer
- Tensiometers and TDR probes
- Soil moisture sensors
Collaborate With Us
Interested in hydraulic property measurement or water flow modeling? Contact us to discuss research collaborations.
Get in Touch