Explore the vast range of titles available.

Accurate mapping of soil organic carbon (SOC) supports sustainable land management practices and carbon accounting initiatives for mitigating climate change impacts. This study presents a novel meta-learner framework that combines multiple machine learning algorithms and spectra processing algorithms to optimize SOC prediction using the PRISMA hyperspectral satellite imagery in the Doukkala plain of Morocco. The framework employs a two-layer structure of prediction models. The first layer consists of Random Forest (RF), Support Vector Regression (SVR), and Partial Least Squares Regression (PLSR). These base models were configured using data smoothing, transformation, and spectral feature selection techniques, based on a 70/30% data split. The second layer utilizes a ridge regression model as a meta-learner to integrate predictions from the base models. Results indicated that RF and SVR performance improved primarily with feature selection, while PLSR was most influenced by data smoothing. The meta-learner approach outperformed individual base models, achieving an average relative improvement of 48.8% over single models, with an R2 of 0.65, an RMSE of 0.194%, and an RPIQ of 2.247. This study contributes to the development of methodologies for predicting and mapping soil properties using PRISMA hyperspectral data.

Soil salinity severely constrains agricultural productivity and soil health, particularly in arid and semi-arid regions. Conventional salinity assessment methods are labor-intensive, time-consuming, and spatially limited. This study developed a data-scarce workflow integrating proximal sensing (EM38-MK2), very high-resolution multispectral imagery, and machine learning to map soil salinity at field scale in the semi-arid Sehb El Masjoune area, central Morocco. A total of 26 soil samples were analyzed for Electrical Conductivity (EC), and 500 Apparent Electrical Conductivity (ECa) measurements were collected and calibrated using the field samples. Spectral and topographic covariates derived from Unmanned Aerial Vehicle (UAV) and PlanetScope imagery supported model training using Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), Random Forest (RF), and a Stacked Ensemble Learning Model (ELM). Regression Kriging (RK) was applied to model residuals to improve spatial prediction. ELM achieved the highest accuracy (R2 = 0.87, RMSE ≈ 4.15), followed by RF, which effectively captured nonlinear spatial patterns. RK improved PLSR accuracy (by 11.1% for PlanetScope, 13.8% for UAV) but offered limited gains for RF, SVR, and ELM. SHAP analysis identified topographic covariates as the most influential predictors. Both UAV and PlanetScope delineated similar saline–sodic zones. The study demonstrates the following: (1) a scalable, data-efficient workflow for salinity mapping; (2) model and RK performance depend more on algorithmic design than sensor type; (3) interpretable ML and spatial modeling enhance understanding of salinity processes in semi-arid systems.

Depending on the band position on the electromagnetic spectrum, optical and electronic characteristics, sensors collect the reflected energy by the Earth’s surface and the atmosphere. Currently, the availability of the new generation of medium resolution, such as the Multi-Spectral Instrument (MSI) on board the Sentinel-2 satellite, offers new opportunities for long-term high-temporal frequency for Earth’s surfaces observation and monitoring. This paper focuses on the analysis and the comparison of the visible, the near-infrared (VNIR), and the shortwave infrared (SWIR) spectral bands of the MSI for soil salinity discrimination in an arid landscape. To achieve these, a field campaign was organized, and 160 soil samples were collected with various degrees of soil salinity, including non-saline soil samples. The bidirectional reflectance factor was measured above each soil sample in a goniometric laboratory using an ASD (Analytical Spectral Devices) spectroradiometer. In the laboratory work, pHs, electrical conductivity (EC-Lab), and the major soluble cations (Na+, K+, Ca2++, and Mg2+) and anions (CO32−, HCO3−, Cl−, and SO42−) were measured using extraction from a saturated soil paste, and the sodium adsorption ratio (SAR) was calculated using a standard procedure. These parameters, in addition to the field observations, were used to interpret and investigate the spectroradiometric measurements and their relevant transformations using the continuum removed reflectance spectrum (CRRS) and the first derivative (FD). Moreover, the acquired spectra over all the soil samples were resampled and convolved in the solar-reflective spectral bands using the Canadian Modified Herman transfer radiative code (CAM5S) and the relative spectral response profiles characterizing the Sentinel-MSI band filters. The statistical analyses conducted were based on the second-order polynomial regression (p < 0.05) between the measured EC-Lab and the reflectances in the MSI convolved spectral bands. The results obtained indicate the limitation of VNIR bands and the potential of SWIR domain for soil salinity classes’ discrimination. The CRRS and the FD analyses highlighted a serious spectral-signal confusion between the salt and the soil optical properties (i.e., color and brightness) in the VNIR bands. Likewise, the results stressed the independence of the SWIR domain vis-a-vis these soil artifacts and its capability to differentiate significantly among several soil salinity classes. Moreover, the statistical fit between each MSI individual spectral band and EC-Lab corroborates this trend, which revealed that only the SWIR bands were correlated significantly (R2 of 50% and 64%, for SWIR-1 and SWIR-2, respectively), while the R2 between the VNIR bands and EC-Lab remains less than 9%. According to the convergence of these four independent analysis methods, it is concluded that the Sentinel-MSI SWIR bands are excellent candidates for an integration in soil salinity modeling and monitoring at local, regional, and global scales.

Accurate and efficient crop maps are essential for decision-makers to improve agricultural monitoring and management, thereby ensuring food security. The integration of advanced artificial intelligence (AI) models with hyperspectral remote sensing data, which provide richer spectral information than multispectral imaging, has proven highly effective in the precise discrimination of crop types. This systematic review examines the evolution of hyperspectral platforms, from Unmanned Aerial Vehicle (UAV)-mounted sensors to space-borne satellites (e.g., EnMAP, PRISMA), and explores recent scientific advances in AI methodologies for crop mapping. A review protocol was applied to identify 47 studies from databases of peer-reviewed scientific publications, focusing on hyperspectral sensors, input features, and classification architectures. The analysis highlights the significant contributions of Deep Learning (DL) models, particularly Vision Transformers (ViTs) and hybrid architectures, in improving classification accuracy. However, the review also identifies critical gaps, including the under-utilization of hyperspectral space-borne imaging, the limited integration of multi-sensor data, and the need for advanced modeling approaches such as Graph Neural Networks (GNNs)-based methods and geospatial foundation models (GFMs) for large-scale crop type mapping. Furthermore, the findings highlight the importance of developing scalable, interpretable, and transparent models to maximize the potential of hyperspectral imaging (HSI), particularly in underrepresented regions such as Africa, where research remains limited. This review provides valuable insights to guide future researchers in adopting HSI and advanced AI models for reliable large-scale crop mapping, contributing to sustainable agriculture and global food security.

Africa’s rapidly growing population is driving unprecedented demands on agricultural production systems. However, agricultural yields in Africa are far below their potential. One of the challenges leading to low productivity is Africa‘s poor soil quality. Effective soil fertility management is an essential key factor for optimizing agricultural productivity while ensuring environmental sustainability. Key soil fertility properties—such as soil organic carbon (SOC), nutrient levels (i.e., nitrogen (N), phosphorus (P), potassium (K), moisture retention (MR) or moisture content (MC), and soil texture (clay, sand, and loam fractions)—are critical factors influencing crop yield. In this context, this study conducts an extensive literature review on the use of hyperspectral remote sensing technologies, with a particular focus on freely accessible hyperspectral remote sensing data (e.g., PRISMA, EnMAP), as well as an evaluation of advanced Artificial Intelligence (AI) models for analyzing and processing spectral data to map soil attributes. More specifically, the study examined progress in applying hyperspectral remote sensing technologies for monitoring and mapping soil properties in Africa over the last 15 years (2008–2024). Our results demonstrated that (i) only very few studies have explored high-resolution remote sensing sensors (i.e., hyperspectral satellite sensors) for soil property mapping in Africa; (ii) there is a considerable value in AI approaches for estimating and mapping soil attributes, with a strong recommendation to further explore the potential of deep learning techniques; (iii) despite advancements in AI-based methodologies and the availability of hyperspectral sensors, their combined application remains underexplored in the African context. To our knowledge, no studies have yet integrated these technologies for soil property mapping in Africa. This review also highlights the potential of adopting hyperspectral data (i.e., encompassing both imaging and spectroscopy) integrated with advanced AI models to enhance the accurate mapping of soil fertility properties in Africa, thereby constituting a base for addressing the question of yield gap.

Soil salinity is a major constraint to soil health and crop productivity, especially in arid and semi-arid regions. The most accurate measurement of soil salinity is considered to be the electrical conductivity of saturated soil extracts (ECe). Because this method is labor-intensive, it is unsuitable for routine analysis in large soil sampling campaigns. This study aimed to identify the best models to estimate soil salinity based on ECe in relation to a rapid electrical conductivity (EC) measurement in soil/water (referred to as S:W henceforward) extracts. We evaluated the relationship between ECe and the ECS:W extract ratios (1:1, 1:2, and 1:5) in salt-affected soils from the semi-arid Sehb El Masjoune region of Morocco. The soil salinity in this region is 0.5 to 235 dS/m, as determined by the ECe method. A total of 125 soil samples, from topsoil (0–15 cm) and subsoil (15–30 cm) with mainly fine to medium textures, were analyzed using linear, logarithmic, and second-order polynomial regression models. The models included all samples or grouped samples according to soil texture (fine, medium) or specific textural classes. The mean ECe values were 2.6, 3.1, and 7.9 times greater than the EC of 1:1, 1:2, and 1:5 S:W extracts, respectively. Polynomial regression models had the best predictive accuracy, R2 = 0.98, and the lowest root mean square error of 10.6 to 10.7 dS/m for the ECS:W extract ratios of 1:5 and 1:2. The polynomial models could represent the non-linear relationships between ECe and salinity indicators, especially in the 80–170 dS/m salinity range, where other models typically underestimate the salinity. These results confirm that advanced regression techniques are suitable for predicting soil salinity in a salt-affected semi-arid region. The site-specific models outperformed previously published models, because they consider the spatial variability and heterogeneity of the salinity in the study area explicitly. This confirms the importance of calibrating soil salinity models according to the local soil and environmental conditions. Consequently, we can undertake soil salinity assessments in hundreds of samples by using the simple, rapid ECS:W extraction method as a direct indicator of EC and extrapolate to ECe with a polynomial regression model. Our approach enables the widespread soil salinity assessments that are needed for land-use planning, irrigation management, and crop selection in salt-affected landscapes.

Spatial analyses of water-quality-monitoring networks in coastal waters are important because pollution sources vary temporally and spatially. This study was conducted to evaluate the spatial distribution of the water-quality-monitoring network of Kuwait Bay using both geostatistical and multivariate techniques. Three years of monthly data collected from six existing monitoring stations covering Kuwait Bay between 2009 and 2011 were employed in conjunction with data collected from 20 field sampling sites. Field sampling locations were selected based on a stratified random sampling scheme oriented by an existing classification map of Kuwait Bay. Two water quality datasets obtained from different networks were compared by cluster analysis applied to the Water Quality Index (WQI) and other water quality parameters, after which the Kriging method was used to generate distribution maps of water quality for spatial assessment. Cluster analysis showed that the current monitoring network does not represent water quality patterns in Kuwait Bay. Specifically, the distribution maps revealed that the existing monitoring network is inadequate for heavily polluted areas such as Sulaibikhat Bay and the northern portion of Kuwait Bay. Accordingly, the monitoring system in Kuwait Bay must be revised or redesigned. The geostatistical approach and cluster analysis employed in this study will be useful for evaluating future proposed modifications to the monitoring stations network in Kuwait Bay.

La glace de rivière est un phénomène récurrent et caractéristique de la majorité des rivières du Canada. Sa présence affecte le régime d'écoulement et peut influencer ainsi le niveau d'eau, la structure des berges et le transport des sédiments. Sans aucun doute, ces changements perturbent l'équilibre de l'écosystème en place. En plus, la glace de rivière est aussi réputée comme cause principale d'inondations hivernales dues aux embâcles, de perturbations à la navigation et à la production d'hydroélectricité. Dans cette perspective, une bonne caractérisation du couvert de glace de rivière est un atout considérable.Nous avons été les premiers à adopter l'approche dite, orientée objet, pour la caractérisation de la glace de rivière. Cette approche repose sur l'objet comme unité de base de classification, elle se distingue des méthodes de classification conventionnelles par sa prise en considération de la topologie et les relations sémantiques des objets. Une méthode de segmentation en objets des images RADARSAT-l a alors été développée. L'objectif était d'avoir des objets représentatifs des différents phénomènes de glace observés à diverses échelles spatiales, tout en préservant la représentativité des relations géométriques et sémantiques des objets.La zone d'étude est le tronçon de la rivière Saint-François reliant la ville de Windsor à celle de Drummondville (Qc). Ce site est connu par la présence de problèmes dus à la glace de rivière, une morphologie variable. De plus, une banque de données historiques sur la glace est disponible. Un total de 19 images RADARS A T -1 mode fin ont été acquises au cours de trois hivers et plusieurs campagnes d'observations de terrain ont été menées.Un système d'information géographique (SIG) a été monté pour fournir les couches d'information contextuelles des images RADARSAT. Largeur, sinuosité et profondeur du chenal sont des exemples de couches matricielles calculées et présentées par le SIG. Des couches thématiques telles que rives, grandes et petites îles et zones de rapides ont aussi été intégrées dans le SIG. Afin de comprendre les liens qui existent entre l'information contenue dans les images RADARSA T -1 et celle présente dans le SIG, nous avons réalisé une analyse qualitative et une quantitative. L'analyse qualitative a mis en évidence les liens entre (1) la théorie de la glace de rivière et du signal radar, (2) le contexte morphologique établi dans le SIG et (3) les observations terrain et historiques disponibles pour le site d'étude. Aussi, par rapport à l'utilisation unique des caractéristiques de l'image RADARSAT-l, l'analyse quantitative a confirmé l'apport des caractéristiques morphologiques de la rivière et ceux liées à la forme des l'objet dans l'amélioration de la classification. L'information extraite à partir de ces deux analyses a été exprimée sous forme de règles logiques de décision dans l'approche de classification. Ces règles sont synthétisées dans une structure appelée Structure Hiérarchique de la Classification orientée objet. Élément essentiel de la classification orientée objet, la structure hiérarchique que nous avons définie est flexible quant à son montage et facilement transférable d'une image à une autre.

The present study focuses on the validation and comparison of eight different physical models for soil salinity mapping in an arid landscape using two independent Landsat-Operational Land Imager (OLI) datasets: simulated and image data. The examined and compared models were previously developed for different semi-arid and arid geographic regions around the world, i.e., Latino-America, the Middle East, North and East Africa and Asia. These models integrate different spectral bands and unlike mathematical functions in their conceptualization. To achieve the objectives of the study, four main steps were completed. For simulated data, a field survey was organized, and 100 soil samples were collected with various degrees of salinity levels. The bidirectional reflectance factor was measured above each soil sample in a goniometric laboratory using an analytical spectral device (ASD) FieldSpec-4 Hi-Res spectroradiometer. These measurements were resampled and convolved in the solar-reflective bands of the Operational Land Imager (OLI) sensor using a radiative transfer code and the relative spectral response profiles characterizing the filters of the OLI sensor. Then, they were converted in terms of the considered models. Moreover, the OLI image acquired simultaneously with the field survey was radiometrically preprocessed, and the models were implemented to derive soil salinity maps. The laboratory analyses were performed to derive electrical conductivity (EC-Lab) from each soil sample for validation and comparison purposes. These steps were undertaken between predicted salinity (EC-Predicted) and the measured ground truth (EC-Lab) in the same way for simulated and image data using regression analysis (p ˂ 0.05), coefficient of determination (R2), and root mean square error (RMSE). Moreover, the derived maps were visually interpreted and validated by comparison with observations from the field visit, ancillary data (soil, geology, geomorphology and water table maps) and soil laboratory analyses. Regardless of data sources (simulated or image) or the validation mode, the results obtained show that the predictive models based on visible- and near-infrared (VNIR) bands and vegetation indices are inadequate for soil salinity prediction in an arid landscape due to serious signals confusion between the salt crust and soil optical properties in these spectral bands. The statistical tests revealed insignificant fits (R2 ≤ 0.41) with very high prediction errors (RMSE ≥ 0.65), while the model based on the second-order polynomial function and integrating the shortwave infrared (SWIR) bands provided the results of best fit, with the field observations (EC-Lab), yielding an R2 of 0.97 and a low overall RMSE of 0.13. These findings were corroborated by visual interpretation of derived maps and their validation by comparison with the ground truthing.

Cette recherche a pour objectif principal le suivi de l'évolution du couvert nival durant la période de fonte au sud du Québec en utilisant le capteur VEGETATION. En effet, ce capteur se caractérise par une bonne répétitivité temporelle, 24 heures, et un large champ de vue, 2500*2500 km. Il permet ainsi un bon suivi spatio-temporel du couvert nival durant la période de fonte. Cependant, la résolution spatiale du capteur VEGETATION n'est que d'1 km² . Pour des fins de modélisation hydrologique et pour des bassins versants de petite ou moyenne taille, il est utile de savoir combien il y a de neige, en terme de pourcentage de surface, dans un pixel VEGETATION.Dans cette étude l'estimation du couvert nival se base sur la combinaison d'une image VGT acquise durant la période de fonte, 11 avril 1999, avec deux autres images VGT, dites de références, acquises en période de couvert total de neige et en absence totale de neige, respectivement le 8 février et le 2 mai 1999. Ainsi, pour chacun des pixels VGT des indices spécifiques de neige (F et NDSI) sont obtenus. Parallèlement, pour chacun de ces pixels VGT, le pourcentage, en terme de surface, couvert de neige est extrait d'une image de haute résolution spatiale, capteur HRVIR de SPOT4, acquise aussi le 11 avril 1999. Les équations d'estimation du couvert nival relient les valeurs des indices de neige aux pourcentages de neige. Toutefois la répétitivité temporelle du capteur HRVIR étant de 26 jours, il est impossible de pouvoir compter sur ce capteur pour effectuer le suivi de la fonte de neige.En constatant que le 11 avril 1999 il ne restait pas beaucoup de neige dans la région d'étude, nous avons refait les mêmes étapes en utilisant une image VGT du 2 avril 1999 et une autre de haute résolution spatiale, capteur HRV (SPOT2), du même jour. Ainsi, nous avons cartographié la distribution du couvert nival au sud du Québec pour les deux dates citées et aussi pour le 27 mars et le 5 avril 1999.Un autre volet intéressant dans cette étude concerne l'estimation des réflectances des classes d'occupation du sol en utilisant les images VGT. En fait, la réflectance d'un pixel VGT est la combinaison de celles des éléments de surface qui le composent. En appliquant le modèle linéaire de composition des réflectances nous avons pu estimer à partir de l'image VGT du 11 avril 1999 les réflectances des classes d'occupation du sol issues de l'image de haute résolution spatiale.Finalement, les images du 11 et du 2 avril 1999 utilisées dans le cadre de cette étude ne contiennent pas assez de neige pour pouvoir définir des relations satisfaisantes. Toutefois, nous avons déterminé les pourcentages de neige sur chaque pixel VEGETATION avec une précision tout à fait acceptable, 80% des pixels de la sous-image VGT du 11 avril sont estimés à ± 10%. La méthodologie que nous avons adoptée tout au long de cette étude peut être appliquée dans un contexte opérationnel moyennant quelques petits ajustements.