Arivoli et al. (2026) Application of soil moisture probe in optimizing the parameters of a land surface model

Identification

• Journal: Discover Geoscience
• Year: 2026
• Authors: E. Arivoli, Subbarayan Saravanan, K. Chandrasekar
• DOI: 10.1007/s44288-026-00389-6

Short Summary

This study optimizes the Variable Infiltration Capacity (VIC) land surface model parameters using site-specific volumetric soil moisture probe data in Pantnagar, India, concluding that calibration based on root fraction and Leaf Area Index significantly improves the accuracy of subsurface soil moisture simulation (R up to 0.89, KGE up to 0.80).

Objective

• To develop a calibration framework leveraging local soil profile measurements and crop phenological data to optimize soil and vegetation parameters within the Variable Infiltration Capacity (VIC) model.
• To identify the sensitive parameters (e.g., root fraction, Leaf Area Index) that critically control the dynamics of surface and subsurface soil moisture in the modeling framework for a site with homogenous crop and soil.

Study Configuration

• Spatial Scale: Gridded framework at 5 km spatial resolution (approximately 3 arcseconds). The in-situ measurements were taken from a homogenous agricultural area of 70,000 square meters (0.07 square kilometers).
• Temporal Scale: Daily timestep for meteorological forcing and simulation output; Monthly updates for Leaf Area Index (LAI) and Albedo; Analysis period spanning February 2022 to December 2022, following a 2-year warmup period (2020–2021).

Methodology and Data

• Models used: Variable Infiltration Capacity (VIC) hydrological model, configured with a three-layer soil scheme (total depth 1.5 meters, stratified layers at 0.15 m, 0.35 m, and 1.0 m).
• Data sources:
  • In-situ Observations: Volumetric soil moisture measured by a Gro-point profile sensor (Time Domain Transmissometer, TDT) installed at G. B. Pant University of Agriculture and Technology (GBPUAT), Pantnagar, India, measuring depths up to 90 centimeters.
  • Meteorological Forcing: Daily gridded Rainfall (0.25° resolution), Minimum and Maximum Temperature (0.5° resolution) from the India Meteorological Department (IMD). Wind speed data accessed via the Bhuvan portal.
  • Land Surface Parameters: Monthly Leaf Area Index (LAI) from MODIS-MOD 15; Monthly Albedo from MODIS-MOD 43; Soil texture and hydraulic properties (Field Capacity, Wilting Point, Porosity) derived from the National Bureau of Soil Survey and Land Use Planning (NBSSLUP).

Main Results

• The site-specific calibration, achieved by updating root fraction distribution and temporal Leaf Area Index (LAI) according to crop stage (Green Peas/Tall Grass), significantly improved the agreement between simulated and observed soil moisture.
• Parameter Sensitivity: Root fraction distribution, temporal LAI, and local meteorological conditions were identified as the most sensitive and crucial components for estimating water balance components and soil moisture dynamics.
• Model Performance (Calibrated Scenario):
  • Correlation Coefficient (R) ranged from 0.81 to 0.89 across the two analyzed layers (0–15 cm and 16–30 cm).
  • Kling Gupta Efficiency (KGE) ranged from 0.77 to 0.80, indicating good overall model accuracy.
  • Nash Sutcliffe Efficiency (NSE) ranged from 0.10 to 0.70, falling within the satisfactory range for Layer 1 but showing limitations in Layer 2.
  • Percentage Bias (P-Bias) ranged from 9.02% to 9.38% in the respective layers, indicating a slight tendency for the model to overestimate soil moisture.

Contributions

• Provides a novel calibration framework for the VIC model that effectively integrates high-resolution, site-specific soil moisture probe data to optimize vegetation-related parameters (root fraction and LAI).
• Quantitatively demonstrates that updating crop phenological parameters is more critical than relying solely on fixed soil parameters for accurate subsurface soil moisture simulation in agricultural settings.
• Establishes a calibrated root fraction distribution that can be utilized for simulating similar land use and soil texture types in regional hydrological models.

Funding

Not applicable.

Citation

@article{Arivoli2026Application,
  author = {Arivoli, E. and Saravanan, Subbarayan and Chandrasekar, K.},
  title = {Application of soil moisture probe in optimizing the parameters of a land surface model},
  journal = {Discover Geoscience},
  year = {2026},
  doi = {10.1007/s44288-026-00389-6},
  url = {https://doi.org/10.1007/s44288-026-00389-6}
}

Generated by BiblioAssistant using gemini-flash-latest (Google API)