Abbaszadeh et al. (2025) Coupling the ParFlow Integrated Hydrology Model within the NASA Land Information System: a case study over the Upper Colorado River Basin

Identification

• Journal: Hydrology and earth system sciences
• Year: 2025
• Date: 2025-10-21
• Authors: Peyman Abbaszadeh, Fadji Zaouna Maina, Chen Yang, Dan Rosen, Sujay V. Kumar, Matthew Rodell, R. M. Maxwell
• DOI: 10.5194/hess-29-5429-2025

Research Groups

• Department of Civil and Environmental Engineering, Princeton University, Princeton, NJ, USA
• High Meadows Environmental Institute, Princeton University, Princeton, NJ, USA
• Hydrological Sciences Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD, USA
• Integrated GroundWater Modeling Center, Princeton University, Princeton, NJ, USA
• Climate & Global Dynamics Lab, The National Center for Atmospheric Research, Boulder, Colorado, USA
• Earth Sciences Division, NASA Goddard Space Flight Center, Greenbelt, MD, USA
• Department of Civil and Environmental Engineering, Hydrologic Modeling and Assimilation Lab, Portland State University, Portland, OR, USA

Short Summary

This study evaluates the newly coupled ParFlow-LIS/Noah-MP model in the Upper Colorado River Basin, demonstrating its enhanced capability to simulate three-dimensional groundwater flow and improve soil moisture representation in complex topography compared to standalone LIS/Noah-MP. The coupled model provides new hydrologic prediction capabilities for both surface and subsurface processes.

Objective

• To evaluate the effectiveness and usefulness of the newly coupled ParFlow-LIS/Noah-MP model in representing surface and subsurface hydrological processes, particularly soil moisture, streamflow, water table depth, and terrestrial water storage, over the Upper Colorado River Basin.

Study Configuration

• Spatial Scale: Upper Colorado River Basin (UCRB), approximately 280 000 km². Model extent: 608 km (east-west) by 896 km (south-north) with a 1 km horizontal resolution and 392 m vertical depth (10 variable layers).
• Temporal Scale: 20-year simulation period from January 2002 to December 2022. Model spin-up included 20 years for LIS/Noah-MP and six repetitions of water year 2005 for PF-LIS/Noah-MP.

Methodology and Data

• Models used:
  • ParFlow (integrated, parallel model solving variably saturated three-dimensional Richards' equation and overland flow via kinematic wave equation).
  • NASA Land Information System (LIS).
  • Noah-MP (Multi-Parameterization) land surface model (within LIS, uses a simplified bucket-type linear reservoir for groundwater).
  • Coupled model: ParFlow-LIS/Noah-MP (PF-LIS/Noah-MP).
• Data sources:
  • Atmospheric Forcing: North American Land Data Assimilation System phase-2 (NLDAS-2) product (12.5 km spatial, hourly temporal resolution).
  • Topography: Hydrological data and maps derived from Shuttle Elevation Derivatives at multiple Scales (HydroSHEDS) elevation data.
  • Land Cover: National Land Cover Database (NLCD) at 30 m resolution, resampled to 1 km.
  • Subsurface Parameters: Three-dimensional subsurface datasets (permeability, porosity, unconsolidated, semi-confining layer, bedrock aquifers).
  • In-situ Observations:
    • Soil moisture: International Soil Moisture Network (ISMN) compiling data from ARM, PBO_H2O, SCAN, SNOTEL, USCRN, and iRON (238 stations, multiple depths).
    • Streamflow: USGS streamflow stations (374 stations).
    • Groundwater: USGS groundwater monitoring wells (18 wells for water table depth).
  • Satellite Products:
    • Soil moisture: THySM (Thermal Hydraulic disaggregation of Soil Moisture), a downscaled SMAP product (1 km spatial, daily temporal resolution).
    • Terrestrial Water Storage (TWS) anomalies: GRACE and GRACE Follow-On (GRACE-FO) (monthly, global maps, CSR Release-06 GRACE Mascon Solutions).

Main Results

• PF-LIS/Noah-MP produces soil moisture simulations comparable to LIS/Noah-MP across the UCRB, with average root mean squared error of 0.036 m³ m⁻³ and correlation coefficient of 0.608 against SMAP observations.
• In regions with high elevation gradients and complex topography, PF-LIS/Noah-MP slightly outperforms standalone LIS/Noah-MP in soil moisture simulation (e.g., in Region 2, ubRMSE improved from 0.0388 m³ m⁻³ to 0.0330 m³ m⁻³ and correlation from 0.482 to 0.539).
• The coupled system provides soil moisture data with higher spatial specificity and better identifies areas of high soil moisture along river corridors due to explicit simulation of lateral subsurface flow.
• PF-LIS/Noah-MP demonstrates reasonable skill in simulating streamflow, adequately capturing both magnitude and timing (e.g., 197 out of 378 stations show good timing and low flow bias).
• Water table depth estimates from PF-LIS/Noah-MP generally agree with USGS well observations, though performance is limited in some complex topographic and geological locations.
• PF-LIS/Noah-MP simulated terrestrial water storage (TWS) anomalies show strong agreement with GRACE/GRACE-FO observations from 2002 to 2012.
• A noticeable decline in TWS agreement between the model and GRACE/GRACE-FO occurs from 2013 onwards, becoming more pronounced in 2020-2022, likely attributed to increased anthropogenic groundwater withdrawals not explicitly simulated by the model.

Contributions

• Demonstrates the capability and effectiveness of the newly developed coupled ParFlow-LIS/Noah-MP model for simulating integrated surface and subsurface hydrologic processes.
• Expands the physical processes represented by the LIS/Noah-MP model by integrating three-dimensional variably saturated groundwater flow and lateral subsurface flow, which was not possible with standalone LIS/Noah-MP.
• Shows improved soil moisture representation in regions with complex topography compared to standalone LIS/Noah-MP.
• Enables accurate estimation of subsurface hydrologic processes like groundwater storage and water table depth, providing new hydrologic prediction capabilities.
• Serves as a foundational step towards incorporating human impacts (e.g., groundwater pumping) and leveraging advanced data assimilation techniques (e.g., GRACE TWS) in integrated hydrological models.

Funding

• NASA MAP (80NSSC20K1714)
• NASA High-End Computing (HEC) Program through the NASA Center for Climate Simulation (NCCS)
• GRACE and GRACE-FO were jointly developed and operated by NASA, DLR, and the GFZ German Research Centre for Geosciences.

Citation

@article{Abbaszadeh2025Coupling,
  author = {Abbaszadeh, Peyman and Maina, Fadji Zaouna and Yang, Chen and Rosen, Dan and Kumar, Sujay V. and Rodell, Matthew and Maxwell, R. M.},
  title = {Coupling the ParFlow Integrated Hydrology Model within the NASA Land Information System: a case study over the Upper Colorado River Basin},
  journal = {Hydrology and earth system sciences},
  year = {2025},
  doi = {10.5194/hess-29-5429-2025},
  url = {https://doi.org/10.5194/hess-29-5429-2025}
}