Monitoring and Modeling the Soil‐Plant System Toward Understanding Soil Health

Summary

This review traces the evolution of soil health assessment from a narrow focus on agricultural productivity and physicochemical properties towards an integrated evaluation of soil biota, microbial communities, and their ecological responses to climate and anthropogenic change. The authors identify a critical gap: despite abundant soil health indicators and monitoring data, mechanistic linkages between individual indicators and soil functions remain incompletely understood across spatial and temporal scales. The review proposes that a soil-plant digital twin approach—amalgamating process-based modelling, remote sensing, data assimilation, and physics-informed machine learning—offers a promising framework to systematically observe and model the soil-plant system and elucidate the intricate interplay between soil properties, hydrological processes, plant physiology, soil microbiome composition, and landscape genetics.

UK applicability

The review's framework for integrating soil monitoring indicators and digital twin modelling is methodologically relevant to UK farming systems and soil stewardship policy, particularly for devolved nations with evolving soil health metrics. However, the abstract does not specify UK-specific case studies or regional validation, so applicability to British pedological contexts and agricultural practices would require engagement with the full paper.

Key measures

Physical, chemical, and biological soil health indicators; soil microbiome properties; soil hydrological processes; soil-plant hydraulics; landscape genomics; Earth Observation data; machine learning model outputs

Outcomes reported

This review synthesises state-of-the-art approaches to soil health monitoring, emphasising how soil-microbiome-plant processes create feedback mechanisms affecting soil properties and functions. It evaluates opportunities for a soil-plant digital twin framework integrating process-based models, Earth Observation data, data assimilation, and physics-informed machine learning to advance soil health understanding across multiple spatiotemporal scales.

Topic tags