From soil pollution mapping to sustainable crop production: Advances in geostatistical and hybrid modelling techniques

Soil is a critical component of terrestrial ecosystems and is increasingly exposed to a wide range of pollutants, particularly heavy metals and potentially toxic elements (PTEs), which are intensified by rapid industrialisation, urbanisation and agricultural intensification. Heavy metal pollution is a significant concern in many regions worldwide that requires systematic and regular monitoring of soils to track contamination levels and prevent associated health risks. Traditional soil pollution monitoring methods, while accurate at localised scales, are limited in their ability to capture the spatial complexity and heterogeneity of heavy metal distribution. Thus, recent advances in geostatistical and modelling approaches have enhanced the accuracy, efficiency and scalability of soil pollution assessment. Geostatistical methods such as kriging, cokriging and empirical Bayesian kriging are known for their ability to generate high-resolution spatial maps and quantify uncertainty in unsampled areas. The integration of geostatistics with remote sensing, machine learning and proximal sensing technologies has significantly improved predictive performance and enabled large-scale monitoring. Advanced hybrid models, such as random forest-ordinary kriging (RF-OK) combinations, artificial neural networks (ANN) and empirical Bayesian kriging regression prediction (EBKRP), have demonstrated superior performance in spatial interpolation, with consistent improvements of 20–40 % in accuracy metrics compared to traditional methods. Despite notable advancements, challenges remain regarding computational complexity, data scarcity, implementation in developing regions and the integration of uncertainty quantification. This review synthesises state-of-the-art geostatistical advancements, hybrid modelling approaches and emerging technologies to provide a comprehensive framework for soil pollution monitoring and risk assessment, aligning with global sustainability and environmental protection goals.