A reliable overview of the spatial distribution of soil properties is a straightforward approach in soil policies and decision-making. Soil organic carbon (SOC) content, SOC stock and bulk density (BD) directly affect soil quality and fertility. Therefore, an accurate assessment of these crucial soil parameters is required. To do this, we used machine learning algorithms (MLAs) including, multiple linear regression (MLR), random forest (RF), artificial neural network (ANN), and support vector machine (SVM) with the help of environmental covariates to predict SOC content, BD, and SOC stock. The study was conducted in two different areas, Látókép and Westsik (East Hungary), both experimental research fields but different from physio geographic points of view. Thirty topsoils (0-10 cm) samples were collected for each study area using conditioned Latin Hypercube Sampling strategy. Environmental covariates were extracted from a digital elevation model (DEM) and satellite images based on the representation of soil forming factors. We validated the results by randomly splitting the dataset into a train (two-third) and test (one-third) and calculated the root mean square error and R2. Our results showed that RF provided the most accurate spatial prediction with R2 of about 80% for each soil property in both study areas. This study highlighted the importance of terrain attributes (including plan and profile curvature, elevation and valley depth) and NDVI derived from satellite images in presenting a spatial distribution of selected soil properties in two different areas. We conclude that comparing these methods can help to determine the most accurate maps under diverse geographical conditions and heterogeneities at different scales, which can be used in precision soil quality management.

Soil salinization is one of the main threats to soils worldwide, which has serious impacts on soil functions. Our objective was to map and assess salt-affectedness on arable land (0.85 km²) in Hungary, with high spatial resolution, using a combination of ensemble machine learning and multivariate geostatistics on three salt-affected soil indicators (i.e., alkalinity, electrical conductivity, and sodium adsorption ratio (n = 85 soil samples)). Ensemble modelling with five base learners (i.e., random forest, extreme gradient boosting, support vector machine, neural network, and generalized linear model) was carried out and the results showed that ensemble modelling outperformed the base learners for alkalinity and sodium adsorption ratio with R² values of 0.43 and 0.96, respectively, while only the random forest prediction was acceptable for electrical conductivity. Multivariate geostatistics was conducted on the stochastic residuals derived from machine learning modelling, as we could reasonably assume that there is spatial interdependence between the selected salt-affected soil indicators. We used 10-fold cross-validation to check the performance of the spatial predictions and uncertainty quantifications, which provided acceptable results for each selected salt-affected soil indicator (for pH value, electrical conductivity, and sodium adsorption ratio, the root mean square error values were 0.11, 0.86, and 0.22, respectively). Our results showed that the methodology applied in this study is efficient in mapping and assessing salt-affectedness on arable lands with high spatial resolution. A probability map for sodium adsorption ratio represents sodic soils exceeding a threshold value of 13, where they are more likely to have soil structure deterioration and water infiltration problems. This map can help the land user to select the appropriate agrotechnical operation for improving soil quality and yield.

Drought has emerged as a major challenge to global food and water security, and is particularly pronounced for Algeria, which frequently grapples with water shortages. This paper sought to monitor and assess the temporal and spatial distribution of drought severity across northern Algeria (excluding the Sahara) during the growing season from 2011 to 2022, while exploring the relationship between the normalized difference vegetation index (NDVI) anomaly and climate variables (rainfall and temperature). Temporal NDVI data from the Terra moderate resolution imaging spectroradiometer (MODIS) satellite covering the period 2000–2022 and climate data from the European Reanalysis 5th Generation (ERA5) datasets collected during the period 1990–2022 were used. The results showed that a considerable portion of northern Algeria has suffered from droughts of varying degrees of severity during the study period. The years 2022, 2021, 2016, and 2018 were the hardest hit, with 76%, 71%, 66%, and 60% of the area, respectively, experiencing drought conditions. While the relationship between the NDVI anomaly and the climatic factors showed variability across the different years, the steady decrease in vegetation health indicated by the NDVI anomaly corroborates the observed increase in drought intensity during the study period. We conclude that the MODIS-NDVI product offers a cost-efficient approach to monitor drought in data-scarce regions like Algeria, presenting a viable alternative to conventional climate-based drought indices, while serving as an initial step towards formulating drought mitigation plans.

Soil health is critical for sustaining ecosystem functions and addressing environmental challenges. Effective soil health management requires reliable methods for assessing soil properties. Soil spectroscopy may allow resource‐effective assessment of soil properties, but more knowledge is needed to transfer knowledge from laboratory‐grade spectrometers to in‐field data acquisition. This study explores the predictive potential of selected spectral subsets from the full visible and near‐infrared (VIS–NIR) range, using various machine learning algorithms (MLAs), as a theoretical exercise to support the design of practical soil sensing tools. Specifically, we evaluated whether narrower spectral ranges can provide predictions comparable to those achieved with the full VIS–NIR spectrum. The ranges are chosen to emulate the spectral coverage and resolution of commercially available sensors which are candidates for widespread and resource‐effective data collection. We used the VIS–NIR spectral data (400–2500 nm) alongside laboratory analyses of several soil properties to be predicted from the pan‐European LUCAS dataset. We employed four different MLAs for estimating soil properties: support vector regression (SVR), cubist, random forest (RF), and multi‐layer perceptron (MLP), which were benchmarked against ordinary least squares regression. Our results showed that spectral subset ranges of 1000–2500 nm and 1350–2500 nm (emulating Trinamix and NeoSpectra sensors, respectively) yielded prediction accuracies similar to the full spectrum. Spectral subsets limited to the visible and early NIR range (350–1000 nm) were less effective. The most informative spectral features were found in wavelengths above approximately 1750 nm. Among MLAs, MLP consistently delivered the best performance, particularly when estimating organic carbon, nitrogen, pH and clay, which were predicted with greater accuracy compared to potassium (K), phosphorus (P) and coarse fragments (CF) which cannot yet be robustly predicted from spectral data alone. This study provides preliminary insight into the spectral regions most relevant for soil property prediction. These findings may inform future development and optimisation of real‐world soil sensors. Validation with actual sensor data, both on dried and in situ samples, remains an important next step.