Jia et al. (2026) Advancing ecohydrological modelling: coupling LPJ-GUESS with ParFlow for integrated vegetation and surface-subsurface hydrology simulations

Identification

• Journal: Geoscientific model development
• Year: 2026
• Date: 2026-02-27
• Authors: Zitong Jia, Shouzhi Chen, Yongshuo H. Fu, David Martín Belda, David Wårlind, Stefan Olin, Chongyu Xu, Jing Tang
• DOI: 10.5194/gmd-19-1727-2026

Research Groups

• College of Water Sciences, Beijing Normal University, Beijing, China
• Plants and Ecosystems, Department of Biology, University of Antwerp, Antwerp, Belgium
• Institute of Meteorology and Climate Research Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology (KIT), Garmisch-Partenkirchen, Germany
• Department of Physical Geography and Ecosystem Science, University of Lund, Lund, Sweden
• Department of Geosciences, University of Oslo, Oslo, Norway
• Center for Volatile Interactions (VOLT), Department of Biology, Universitetsparken 15, Copenhagen, Denmark

Short Summary

Groundwater sustains vegetation and regulates land-atmosphere exchanges, yet most Earth system models oversimplify its dynamics. This study develops an integrated framework coupling a dynamic vegetation model with the three-dimensional hydrological model ParFlow to explicitly represent groundwater-vegetation interactions, demonstrating that groundwater flow strongly regulates water exchanges and improves simulations of water cycles in Earth system models.

Objective

• To develop and evaluate an integrated ecohydrological modeling framework (PF-LPJG) by coupling the three-dimensional surface-subsurface hydrological model ParFlow with the dynamic global vegetation model LPJ-GUESS, to explicitly investigate how lateral groundwater flow and vegetation dynamics jointly regulate hydrological fluxes and improve the representation of vegetation-groundwater interactions.

Study Configuration

• Spatial Scale: Danube River Basin, simulated at a horizontal resolution of 10 kilometers (0.1° x 0.1°), with a model grid of 161 columns x 82 rows and a total depth of 112 meters below the surface.
• Temporal Scale: 38-year simulation period (1980–2018) for hydrological processes, following extensive spin-up periods (500 years for vegetation, 40,000 years for soil organic matter, and 10 years for coupled system equilibrium). Daily time steps for water flux calculations.

Methodology and Data

• Models used:
  • ParFlow (v3.13.0): A fully distributed, physically based three-dimensional surface-subsurface hydrological model.
  • LPJ-GUESS: A process-based dynamic global vegetation model.
  • PF-LPJG: The newly developed fully coupled ParFlow-LPJ-GUESS model.
• Data sources:
  • Meteorological forcing: ERA5-LAND dataset (0.1° x 0.1°, 1980-2018) for precipitation, 10 m u-component of wind, relative humidity, surface downwelling shortwave radiation, surface air pressure, and 2 m temperature, aggregated to daily time scale.
  • Soil data: WISE database (FAO Soil Map and Harmonized World Soil Database) for sand, clay, silt content, organic carbon, pH, and bulk density, resampled to 10 km resolution.
  • Subsurface data: GLHYMPS 1.0 dataset (permeabilities from GLiM) refined with the hydrogeological map of the Danube River Basin (Duscher et al., 2015), and bedrock depth values from Shangguan et al. (2017), all resampled to 10 km.
  • Land Cover: MODIS Land Cover Type product (MCD12Q1, Version 6.1, IGBP scheme), resampled to 10 km.
  • Topography and rivers: MERIT Hydro IHU dataset (5 arcminute Digital Elevation Model, resampled to 10 km).
  • Evaluation data:
    • Streamflow: Global Runoff Data Centre (GRDC) observations.
    • Surface Soil Moisture (SM): European Space Agency Climate Change Initiative Soil Moisture (ESA CCI-SM) product (0.25° interpolated to 0.1°).
    • Evapotranspiration (ET): Global Land Evaporation Amsterdam Model (GLEAM) 4.2 datasets.
    • Water Table Depth (WTD): Global gridded WTD benchmark (Fan et al., 2013) and in-situ groundwater observations from monitoring wells.

Main Results

• Streamflow: The PF-LPJG model substantially improved streamflow simulations compared to stand-alone LPJ-GUESS. Over 80% of gauging stations achieved a Kling-Gupta Efficiency (KGE) > 0.5 and Spearman's ρ > 0.80. The coupled model mitigated the underestimation of summer low flows during dry years and increased the accuracy of peak flow timing in wet years.
• Evapotranspiration (ET): PF-LPJG successfully reproduced similar spatial patterns of ET across the basin as GLEAM. The mean difference between PF-LPJG and GLEAM (0.000101 m d⁻¹) was marginally smaller than that of LPJ-GUESS (0.000108 m d⁻¹). Seasonal ET dynamics were comparable, with larger positive deviations in spring (0.000226 m d⁻¹) and summer (0.000464 m d⁻¹).
• Soil Moisture (SM): PF-LPJG captured spatial heterogeneity effectively and showed good temporal consistency with ESA CCI-SM anomalies (Spearman's ρ = 0.51, RMSE = 0.73 cm³ cm⁻³). It significantly improved soil moisture simulation during dry periods with higher correlation and lower error compared to LPJ-GUESS.
• Water Table Depth (WTD): The PF-LPJG model achieved a residual-based Root Mean Square Error-observations Standard deviation Ratio (RSR) of 1.31, which is substantially lower than the RSR of 4.46 from a benchmark simulation (Fan et al., 2013). For most monitoring wells (48 points), the absolute error between simulated and observed WTD was within 0–5 meters.
• ET Partitioning: The coupled model improved the representation of bare-soil evaporation and reduced fluctuations in the transpiration-to-evaporation (T/E) ratio, aligning more closely with the GLEAM v4.2 product, particularly during summer months.

Contributions

• Introduces the first basin-scale, long-term implementation of a fully coupled ParFlow-LPJ-GUESS (PF-LPJG) modeling framework, physically integrating three-dimensional surface-subsurface hydrology with dynamic vegetation.
• Demonstrates substantial improvements in streamflow, soil moisture, and evapotranspiration partitioning simulations compared to stand-alone dynamic global vegetation models, without requiring parameter calibration.
• Provides a mechanistic framework for capturing bidirectional interactions among surface-subsurface water, vegetation dynamics, and ecosystem biogeochemical processes, addressing a critical limitation in existing Earth system models.
• Offers new mechanistic understanding of vegetation–groundwater responses to climate change and enhances the representation of low-flow dynamics, drought buffering, and baseflow-supported vegetation growth.

Funding

• National Natural Science Foundation of China for Distinguished Young Scholars (Grant No. 42025101)
• Key Program of the National Natural Science Foundation of China (Grant No. 42430504)
• National Key Research and Development Program of China (Grant No. 2023YFF0805604)
• Fundamental Research Funds for the Central Universities (Grant No. 2243300004)
• 111 Project (Grant No. B18006)
• Joint China–Sweden Mobility Program (Grant No. CH2020-8656)
• Villum Young Investigator grant (Grant No. VIL53048)
• Danish National Research Foundation (Center for Volatile Interactions – VOLT, Grant No. DNRF168)

Citation

@article{Jia2026Advancing,
  author = {Jia, Zitong and Chen, Shouzhi and Fu, Yongshuo H. and Belda, David Martín and Wårlind, David and Olin, Stefan and Xu, Chongyu and Tang, Jing},
  title = {Advancing ecohydrological modelling: coupling LPJ-GUESS with ParFlow for integrated vegetation and surface-subsurface hydrology simulations},
  journal = {Geoscientific model development},
  year = {2026},
  doi = {10.5194/gmd-19-1727-2026},
  url = {https://doi.org/10.5194/gmd-19-1727-2026}
}