This is a systematic literature review of the application of machine learning (ML) algorithms in geosciences, with a focus on environmental monitoring applications. ML algorithms, with their ability to analyze vast quantities of data, decipher complex relationships, and predict future events, and they offer promising capabilities to implement technologies based on more precise and reliable data processing. This review considers several vulnerable and particularly at-risk themes as landfills, mining activities, the protection of coastal dunes, illegal discharges into water bodies, and the pollution and degradation of soil and water matrices in large industrial complexes. These case studies about environmental monitoring provide an opportunity to better examine the impact of human activities on the environment, with a specific focus on water and soil matrices. The recent literature underscores the increasing importance of ML in these contexts, highlighting a preference for adapted classic models: random forest (RF) (the most widely used), decision trees (DTs), support vector machines (SVMs), artificial neural networks (ANNs), convolutional neural networks (CNNs), principal component analysis (PCA), and much more. In the field of environmental management, the following methodologies offer invaluable insights that can steer strategic planning and decision-making based on more accurate image classification, prediction models, object detection and recognition, map classification, data classification, and environmental variable predictions.

The climate in recent decades has aroused interest in the scientific community, prompting us to analyse the mechanisms that regulate it, to understand the climate change responsible for an increase in extreme phenomena. Consequently, the increase in hydrogeological instability in the Italian territory has led to an in-depth study of atmospheric parameters to understand the variations of the atmospheric system. One tool capable of detecting such variations is the weather radar. The weather radar data available in the area provided by the National Radar Network of the Department of Civil Protection allow the evaluation of variations on a national scale for hydro-meteorological-climatic monitoring as well as the disasters that have occurred. Using open-source programming software, the servers can be queried and data retrieved from a source to perform processing for specific purposes through data extraction techniques.

This paper examines land management technologies to enhance environmental monitoring more efficiently. The study highlights the interactions between human activities and environmental systems with a data-driven environmental monitoring approach. There are many human pressures, such as pollution, land degradation, and habitat loss, negatively impacting soil health. The methodology proposed improves soil status assessments in response to evolving environmental pressures by utilizing satellite imagery and predictive modeling. The integration of Sentinel-2 imagery, the calculation of various spectral indices (NDVI, NBR, NDMI, EVI, SAVI) at different time intervals, and the application of the Isolation Forest algorithm are employed in this study to determine the specific area that is affected by the environmental issue. The chosen algorithm was favored due to its superior performance in handling high-dimensionality data, enhanced computational efficiency, provision of interpretable results, and insensitivity to disparities in class distribution. This study analyzes two separate study cases at different scales. The first involves wildfire identification achieving an overall accuracy of 98%. The second focuses on the expansion areas to pre-existing quarries with an overall accuracy of 95%. The NBR proved most effective in delineating burned areas, whereas the EVI generated the most remarkable results in the quarry case study. This approach provides an effective and scalable tool for environmental monitoring, supporting sustainable management policies, and strengthening ecosystem resilience.

In highly urbanised and industrialised settings, managing environmental pressures and enhancing urban resilience demand integrated, spatially explicit approaches. This study presents a methodological framework that integrates topographic data, land cover information, and open geodata to produce a high-resolution vulnerability map. A multi-criteria analysis was performed using indicators such as land use, population density, proximity to emission sources, vegetation cover, and sensitive services (e.g., schools and hospitals). The result is a high-resolution vulnerability map that classifies the urban, peri-urban, and coastal zones into five levels of environmental risk. These evaluation levels are derived from geospatial analyses combining pollutant dispersion modelling with land-use classification, enabling the identification of the most vulnerable urban zones. These findings support evidence-based planning and can guide local governments and environmental agencies in prioritising Nature-based Solutions (NBSs), enhancing ecological connectivity, and reducing exposure for vulnerable populations.

This study addresses the global challenge of superficial soil contamination by heavy metals, focusing on differentiating natural geogenic sources from anthropogenic contributions in complex industrial–urban environments. We develop an integrated geostatistical and multivariate framework combining soil metal concentration analysis with AERMOD atmospheric dispersion modeling using a comparative multi-model machine learning approach (including Extreme Gradient Boosting, Random Forest, and Ridge Regression). Applied to the industrialized area of Taranto, Southern Italy, this approach incorporates spatial autocorrelation and multiple environmental predictors to identify contamination patterns and sources. The results reveal variable predictive accuracy across metals, with RF generally outperforming the other algorithms. The model achieved its highest performance for copper (R² = 0.58, RMSE = 25.82), Tin (R² = 0.53, RMSE = 5.95), and chromium, while showing instability for others. These disparities highlight the differential influence of remote sensing data on contamination mapping. The framework advances the quantitative assessment of soil pollution by linking atmospheric deposition and spatial processes with causal interpretability.

The topic of diffuse pollution is of particular interest from technical, scientific, and administrative management points of view. Diffuse pollution is defined as the contamination or chemical, physical, or biological alterations of environmental matrices caused by diffuse sources and not attributable to a single origin. In this study, various sources of diffuse pollution such as nitrates, pesticides, metals, and plastics were analysed. This was aimed at the implementation of a decision support system able to represent the state of environmental matrices degradation, with particular attention to water resources, and to make decisions evaluating similar environmental contexts. The potential of the developed system makes it possible to identify areas with the same environmental characteristics, referring to the various activities that create diffuse pollution and areas with the same pressure values on the environmental matrices. The system provides the political decision-maker with greater awareness of the environmental state, thus enabling him to apply more accurate land management policies. The created system, based on open-source software, which can be implemented with additional available data sources, is characterised by a data processing workflow that provides output information at the municipal level, so that it can be managed both by mayors and regional managers who are able to share the same information with all.