The digital twin concept has found widespread application across diverse industries. Herein, we present a comprehensive conceptual framework for the cognitive soil digital twin, which embodies the intricate physical reality of the soil ecosystem, aiding in its holistic monitoring and comprehension. The digital twin can seamlessly integrate a multitude of sensor data sources, including field Internet of Things sensors, remote sensing data, field measurements, digital cartography, surveys, and other Earth observation datasets. By virtue of its duality, this digital counterpart facilitates data organisation and rigorous analytical exploration, unravelling the intricacies of physical, chemical, and biological soil constituents while discerning their intricate interrelationships and their impact on ecosystem services. Its potential extends beyond mere data representation, acting as a versatile tool for scenario analysis and enabling the visualisation of diverse environmental impacts, including the effects of climate change and transformations in land use or management practices. Beyond academic circles, the digital twin’s utility extends to a broad spectrum of stakeholders across the entire quadruple helix, encompassing farmers and agronomists, soil researchers, the agro-industry, and policy-makers. By fostering collaboration among these stakeholders, the digital twin catalyses informed decision-making, underpinned by data-driven insights. Moreover, it acts as a testbed for the development of innovative sensors and monitoring frameworks, in addition to providing a platform that can educate users and the broader public using immersive and innovative visualisation tools, such as augmented reality. This innovative framework underscores the imperative of a holistic approach to soil ecosystem monitoring and management, propelling the soil science discipline into an era of unprecedented data integration and predictive modelling, by harnessing the effects of climate change towards the development of efficient decision-making.

Spectroscopy is a widely used technique that can contribute to food quality assessment in a simple and inexpensive way. Especially in grape production, the visible and near infrared (VNIR) and the short-wave infrared (SWIR) regions are of great interest, and they may be utilized for both fruit monitoring and quality control at all stages of maturity. The aim of this work was the quantitative estimation of the wine grape ripeness, for four different grape varieties, by using a highly accurate contact probe spectrometer that covers the entire VNIR–SWIR spectrum (350–2500 nm). The four varieties under examination were Chardonnay, Malagouzia, Sauvignon-Blanc, and Syrah and all the samples were collected over the 2020 and 2021 harvest and pre-harvest phenological stages (corresponding to stages 81 through 89 of the BBCH scale) from the vineyard of Ktima Gerovassiliou located in Northern Greece. All measurements were performed in situ and a refractometer was used to measure the total soluble solids content (°Brix) of the grapes, providing the ground truth data. After the development of the grape spectra library, four different machine learning algorithms, namely Partial Least Squares regression (PLS), Random Forest regression, Support Vector Regression (SVR), and Convolutional Neural Networks (CNN), coupled with several pre-treatment methods were applied for the prediction of the °Brix content from the VNIR–SWIR hyperspectral data. The performance of the different models was evaluated using a cross-validation strategy with three metrics, namely the coefficient of the determination ($R^2$), the root mean square error (RMSE), and the ratio of performance to interquartile distance (RPIQ). High accuracy was achieved for Malagouzia, Sauvignon-Blanc, and Syrah from the best models developed using the CNN learning algorithm ($R^2>0.8$, $\mathrm{RPIQ}\ge 4$), while a good fit was attained for the Chardonnay variety from SVR ($R^2=0.63$, $\mathrm{RMSE}=2.10$, $\mathrm{RPIQ}=2.24$), proving that by using a portable spectrometer the in situ estimation of the wine grape maturity could be provided. The proposed methodology could be a valuable tool for wine producers making real-time decisions on harvest time and with a non-destructive way.

The existing digital soil maps are mainly characterized by coarse spatial resolution and are not up to date; thus, they are unable to support the physical process-based models for improved predictions. The overarching objective of this work is oriented toward a data-driven approach and datacube-based tools (Soil Data Cube), leveraging Sentinel-2 imagery data, open access databases, ground truth soil data and Artificial Intelligence (AI) architectures to provide enhanced geospatial layers into the Revised Universal Soil Loss Equation (RUSLE) model, improving both the reliability and the spatial resolution of the final map. The proposed methodology was implemented in the agricultural area of the Imathia Regional Unit (northern Greece), which consists of both mountainous areas and lowlands. Enhanced soil maps of Soil Organic Carbon (SOC) and soil texture were generated at 10 m resolution through a time-series analysis of satellite data and an XGBoost (eXtrene Gradinent Boosting) model. The model was trained by 84 ground truth soil samples (collected from agricultural fields) taking into account also additional environmental covariates (including the digital elevation model and climatic data) and following a Digital Soil Mapping (DSM) approach. The enhanced layers were introduced into the RUSLE’s soil erodibility factor (K-factor), producing a soil erosion layer with high spatial resolution. Notable prediction accuracy was achieved by the AI model with $R^2$ 0.61 for SOC and 0.73, 0.67 and 0.63 for clay, sand, and silt, respectively. The average annual soil loss of the unit was found to be 1.76 ton/ha/yr with 6% of the total agricultural area suffering from severe erosion (>11 ton/ha/yr), which was mainly found in the mountainous border regions, showing the strong influence of the mountains in the agricultural fields. The overall methodology could strongly support regional decision making and planning and environmental policies such as the European Common Agricultural Policy (CAP) and the Sustainable Development Goals (SDGs).

In the pursuit of enhancing the wine production process through the utilization of new technologies in viticulture, this study presents a novel approach for the rapid assessment of wine grape maturity levels using non-destructive, in situ infrared spectroscopy and artificial intelligence techniques. Building upon our previous work focused on estimating sugar content (${}^{\circ }$Brix) from the visible and near-infrared (VNIR) and short-wave infrared (SWIR) regions, this research expands its scope to encompass pH and titratable acidity, critical parameters determining the grape maturity degree, and in turn, wine quality, offering a more representative estimation pathway. Data were collected from four grape varieties—Chardonnay, Malagouzia, Sauvignon Blanc, and Syrah—during the 2023 harvest and pre-harvest phenological stages in the vineyards of Ktima Gerovassiliou, northern Greece. A comprehensive spectral library was developed, covering the VNIR–SWIR spectrum (350–2500 nm), with measurements performed in situ. Ground truth data for pH, titratable acidity, and sugar content were obtained using conventional laboratory methods: total soluble solids (TSS) (${}^{\circ }$Brix) by refractometry, titratable acidity by titration (expressed as mg tartaric acid per liter of must) and pH by a pH meter, analyzed at different maturation stages in the must samples. The maturity indicators were predicted from the point hyperspectral data by employing machine learning algorithms, including Partial Least Squares regression (PLS), Random Forest regression (RF), Support Vector Regression (SVR), and Convolutional Neural Networks (CNN), in conjunction with various pre-processing techniques. Multi-output models were also considered to simultaneously predict all three indicators to exploit their intercorrelations. A novel multi-input–multi-output CNN model was also proposed, incorporating a multi-head attention mechanism and enabling the identification of the spectral regions it focuses on, and thus having a higher interpretability degree. Our results indicate high accuracy in the estimation of sugar content, pH, and titratable acidity, with the best models yielding mean $R^2$ values of 0.84, 0.76, and 0.79, respectively, across all properties. The multi-output models did not improve the prediction results compared to the best single-output models, and the proposed CNN model was on par with the next best model. The interpretability analysis highlighted that the CNN model focused on spectral regions associated with the presence of sugars (i.e., glucose and fructose) and of the carboxylic acid group. This study underscores the potential of portable spectrometry for real-time, non-destructive assessments of wine grape maturity, thereby providing valuable tools for informed decision making in the wine production industry. By integrating pH and titratable acidity into the analysis, our approach offers a holistic view of grape quality, facilitating more comprehensive and efficient viticultural practices.

This study presents a methodological framework for predicting soil organic carbon (SOC) using laboratory spectral recordings from a handheld near-infrared (NIR, 1350–2550 nm) device combined with open geospatial data derived from remote sensing sensors related to landform, climate, and vegetation. Initial experiments proved the superiority of convolutional neural networks (CNNs) using only spectral data captured by the low-cost spectral devices reaching an $R^2$ of 0.62, RMSE of 0.31 log-SOC, and an RPIQ of 1.87. Furthermore, the incorporation of geo-covariates with Neo-Spectra data substantially enhanced predictive capabilities, outperforming existing approaches. Although the CNN-derived spectral features had the greatest contribution to the model, the geo-covariates that were most informative to the model were primarily the rainfall data, the valley bottom flatness, and the snow probability. The results demonstrate that hybrid modeling approaches, particularly using CNNs to preprocess all features and fit prediction models with Extreme Gradient Boosting trees, CNN-XGBoost, significantly outperformed traditional machine learning methods, with a notable RMSE reduction, reaching an $R^2$ of 0.72, and an RPIQ of 2.17. The findings of this study highlight the effectiveness of multimodal data integration and hybrid models in enhancing predictive accuracy for SOC assessments. Finally, the application of interpretable techniques elucidated the contributions of various climatic and topographical factors to predictions, as well as spectral information, underscoring the complex interactions affecting SOC variability.