Machine-Learned Hydraulic Conductivity Beyond Random Splits

Abstract

What evidence is needed to determine how machine learning can predict hydraulic conductivity without hiding literature heterogeneity or unsupported chemical conditions? This review answers by treating extrapolation-aware geotechnical prediction as a property of a sociotechnical workflow rather than a feature that can be read from average accuracy. The focal setting is sodium-bentonite geosynthetic clay liners exposed to saline or leachate solutions, where a simplified or learned representation can look accurate while omitting the physical regime that determines transfer. Evidence from the assigned publications is synthesized with foundational studies of calibration, distribution shift, causal structure, and responsible deployment. Four requirements follow: preserve the lineage of instrument observations, constitutive parameters, geometry, loading history, simulation output, and validation measurements; measure stability across relevant perturbations; connect confidence to a specific action; and maintain a route for human challenge and correction. The framework distinguishes descriptive performance from decision utility and separates uncertainty about the world from uncertainty created by the model and its evaluator. It also shows why faster inference or richer reasoning is valuable only when it improves a defined decision under a transparent resource budget. The article is a literature review and research agenda, not a report of a newly completed trial.