Modelling approaches are becoming an efficient tool in the forecasting of the salinization spread impact on the local and global scales. In the Lower-Cheliff plain, updating the information on soil salinity expansion is more than required as it continues to damage the agricultural environment in there. Through this study, we adopted an artificial methodology that consists of a combination of buffer analysis using Sentinel-2 data and machine learning modelling to assess their aptitude in the prediction and mapping of soil salinity. The adopted random forest (RF) algorithm included the reflectance information from of the bands from Blue ( ${\rho }_{\mathrm{B}\mathrm{l}\mathrm{u}\mathrm{e}}$), Green ( ${\rho }_{\mathrm{G}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}$), Red ( ${\rho }_{\mathrm{R}\mathrm{e}\mathrm{d}}$), Near infrared (NIR), Vegetation red-edge (VRE) and Shortwave Infrared (SWIR), optimized with the geospatial buffering based on the 91 soil random samples collected during the summer of 2019 and measured for the Electrical Conductivity (EC) in the laboratory. The outputs from the geospatial buffering refined the goodness of the correlation between field data and the variables set from bands reflectance and selected salinity indices. The obtained coefficient of determination (R ²) with Multiple Linear Regression (MLR) and Partial Least Square regression (PLS) models proved an improvement in the multivariable prediction of soil salinity with the optimized Sentinel-2 data (R ² = 0.61 and 0.68 + RMSE of 3.36 and 2.87 dS m⁻¹ respectively), compared to the results from the 2015 multiple regression model, Given the high values of R ² = 0.77 and 0.95 and the low values of RMSE = 2.29 and 1.18 dS m⁻¹ respectively; the RF regressor was very efficient in predicting EC either with all variables or with selected variables from the variable importance measure based on the mean decrease accuracy (MDA). The random forest classifier performed top of the range classification with an overall Accuracy (OA) about 99%, estimating that over 75% of the study area surface is suffering from the salinization extent.

In semi-arid and arid areas, soil salinity has adverse effects both on the environment and agricultural production. The main causes of this salinization come from natural or anthropogenic processes, which is certainly an environmental problem that affects more than 20% of the world's land. This study was made in order to map the spatial distribution of soil salinity of the irrigated perimeter of El Ghrous in southeastern Algeria. These maps were performed based on data collected from 190 soil samples from 0 to 15 cm deep. We used ordinary kriging (OK) to analyze the spatial variability of soil salinity, while indicator kriging (IK) was used to analyze salinity versus threshold values. The salinity map predicted by the electrical conductivity (EC) values using the ordinary kriging (OK) method showed the different classes of salinity according to Durand's classification with moderately saline 3rd order dominance, while the unsalted soil (EC < 0.6 dS m-1) represents a very low percentage (1.5%). The indicator kriging (IK) was carried out by four thresholds which correspond to the salinity class limits: EC > 0.6, EC > 1, EC > 2, EC > 3, and EC > 4 dS m-1, for developing probability maps to determine risk areas. This study has shown the spatial trend of soil salinity by geolocation of different classes, and to carry out risk maps using geostatistical techniques.

The assessment of groundwater irrigation using robust tools is essential for the sustenance of the agro-environment in arid and semi-arid regions. This study presents a reliable method consisting of a combination of multivariate analysis and geostatistical modeling to assess groundwater irrigation resources in the Western Middle Cheliff (Algeria). For this goal, mean data from 87 wells collected during April to July 2017 were used. The hierarchical cluster analysis (HCA) using the Q-mode approach revealed three distinct water types, with mineralization increasing from cluster 1 to cluster 3. The Principal Component Analysis (PCA) utilizing the Varimax method approach allowed the extraction of three main components: the first and second (PC1, PC2), revealing that the geogenic process, have influenced the hydrogeochemical composition of groundwater. The pollution induced by agriculture activities has been related to PC3. Based on the combination of multivariate analysis and geostatistical modeling, the distribution maps were created by interpolating the factor distribution values acquired in the study region using the ordinary kriging (OK) interpolation method. The findings revealed that both natural processes and man-made activities have a substantial impact on the quality of groundwater irrigation. Cluster mapping, another often used combining approach, has shown its effectiveness in assisting groundwater resource management.

Chestnut cultivation is a type of traditional centennial exploitation in southern areas of Extremadura. Currently, chestnut is in danger of extinction due to the effects of climate change, some diseases ( e.g. Cryphonectria parasitica ) and land mismanagement. The high temporal variability of rainfall leads to soil erosion in chestnut cultivation. New forms of management are proposed to try, among other things, to reduce soil losses. This study quantifies soil losses in areas under tree canopies and open areas considering two different age plantations; 1990s and 1960s. To achieve the proposed goal 18 erosion plots were installed as paired plots under tree canopies and open areas in both plantations. The total amount of rainfall per event, tree cover, bare soil, runoff coefficient and sediment concentration were also measured in every plot. The results showed that the highest percentage of bare soil (> 45%) coincides with the period of greatest tree cover (> 75%). The open areas and the youngest plantation showed soil losses higher than the areas under the tree and the oldest plantation. In addition, soil losses increase as higher percentages of bare soil are recorded. We conclude that the size of the trees and a better soil stability in older plantations help reduce soil losses.

Extremadura is the region that stores the greatest amount of fresh water in Spain. Such water is mainly used for power generation, irrigation in agriculture, biodiversity conservation, tourism, recreation, and human and livestock consumption. Nevertheless, crucial information on the total number of water bodies and their geometrical characteristics and spatial distribution patterns are still missing. Thus, our main goal was to characterize the Extremenian water bodies geometrically and spatially through different statistical techniques such as kernel density, Moran’s index, the Getis-Ord Gi*, and principal component analysis (PCA). Firstly, all existing hydrological information was gathered, and using aerial aircraft imagery and satellite images, each water body (WB) was then carefully collected, checked, and corrected. We have inventoried 100,614 WBs (mean density: 2.45 WB km²), irregularly distributed on the territory. WBs with an area < 0.01 km² (100 ha) represent 64.5% of the total. A multivariate statistical study was conducted, showing that livestock, aridity of the climate, and topography are the main factors controlling the density of water bodies (WBs) in this area. It can be concluded that monitoring of small bodies is crucial to understand their spatial distribution, since they are spread over areas in which extensive farming and commercial crops such as tobacco strongly influence the way of living of many families.