Estimating properties of a homogeneous bounded soil using machine learning models

2025/06/02 by Konstantinos Kalimeris, Kalimeris, Konstantinos, Leonidas Mindrinos +3 · 2 citations
Engineering · Environmental Science · #Soil and Unsaturated Flow #Soil Moisture and Remote Sensing #Dam Engineering and Safety

paper · pdf · doi:10.48550/arxiv.2506.04256

Abstract

This work focuses on estimating soil properties from water moisture measurements. We consider simulated data generated by solving the initial-boundary value problem governing vertical infiltration in a homogeneous, bounded soil profile, with the usage of the Fokas method. To address the parameter identification problem, which is formulated as a two-output regression task, we explore various machine learning models. The performance of each model is assessed under different data conditions: full, noisy, and limited. Overall, the prediction of diffusivity D tends to be more accurate than that of hydraulic conductivity K. Among the models considered, Support Vector Machines (SVMs) and Neural Networks (NNs) demonstrate the highest robustness, achieving near-perfect accuracy and minimal errors.

Citations

• Determining anomalies in a semilinear elliptic equation by a minimal number of measurements
• Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

Cited by

• Estimating parameters of the diffusion model via asymptotic expansions
• Fokas method for linear convection-diffusion equation with time-dependent coefficients and its extension to other evolution equations

Related

• Estimating parameters of empirical infiltration models from the global dataset using machine learning
• Forward and inverse modeling of water flow in unsaturated soils with discontinuous hydraulic conductivities using physics-informed neural networks with domain decomposition
• A Machine Learning Data Fusion Model for Soil Moisture Retrieval
• The Effect of Different Optimization Strategies to Physics-Constrained Deep Learning for Soil Moisture Estimation
• New pedotransfer function (“CRC”) for the prediction of unsaturated soil hydraulic conductivity using soil water retention data