Résumé: This study examines the spatial and temporal variations of meteorological drought over northern Algeria based on the Standardized Precipitation Index (SPI) and Principal Component Analysis (PCA) with Varimax Rotation. It also evaluates the relationship between different atmospheric circulation indices and drought variability at seasonal and annual time scales. The atmospheric circulation indices considered in this assessment are the Southern Oscillation Index (SOI), the North Atlantic Oscillation (NAO) Index, the Westerly Index (WI), the two Mediterranean Oscillation Indices (MOI1) (MOI2), the Western Mediterranean Oscillation Index (WeMOI), the North Sea Caspian Pattern (NCP) Index, the Trans Polar Index (TPI), and the Eastern Mediterranean Pattern (EMP) Index. The obtained results indicate that drought varies widely over the study area, as four Rotated Principal Components (RPCs) were retained for the Varimax Rotation, which divided the study area into four drought sub-regions: the central and eastern coastal regions represented by RPC1, the western regions represented by RPC2, the eastern regions represented by RPC3, and the west-central southern regions represented by RPC4. Drought variability on seasonal basis was successfully associated with different atmospheric circulation indices, mainly with EMP at the eastern part of the study area, which gathers the sub-regions represented by RPC1 and RPC3, and with SOI, NAO, WI, MOI1, and MOI2 in the western part, which gathers the sub-regions represented by RPC2 and RPC4. Drought inter-annual variability was better explained by EMP in RPC1 sub-region, by SOI in RPC2 sub-region, by NCP in RPC3 sub-region, and by MOI1 in RPC4 sub-region. Another important outcome revealed by this study is the prevalence of a general increasing trend in drought conditions over the study area, with more statistically significant trends in the western regions. Overall, this study has resulted in a better understanding of the mechanisms responsible for drought variability over northern Algeria, constituting valuable material for enhancing drought forecasting and water resource planning.

Résumé: The current study was conducted to classify the quality of groundwater in wetlands. This research was carried out to evaluate the groundwater quality of wells in the surrounding area according to the Irrigation Water Quality Index (IWQI) to verify its suitability for agricultural purposes. The IWQI was developed in 2010 by Meireles; it was calculated based on the water quality parameters: EC, Na+1, Mg+2, Ca+2, Cl–1, and of 37 monitoring wells. The results of the IWQI show that the water quality declined significantly in July, with the majority of wells 76% within the severe restriction category (SR), which means that it’s suitable only for irrigation of crops with high salt tolerance, and 14% of wells falling within the High Restriction category (HR). The lake’s discharge decreases during the dry season, and most of the water quality classified as a severe restriction (SR), this indicates that the groundwater is influenced by the saline waters of Lake Fetzara, which are already loaded with chemical elements due to evaporation. As a result, the study of groundwater suitability for irrigation revealed that these waters are generally acceptable for irrigation of salt-tolerant crops on well-drained soils but require a prior control of salinity evolution.

Résumé: Developing an accurate map to control groundwater pollution is becoming increasingly important to fulfill requirements of the Algerian water directives. Evaluating groundwater vulnerability to nitrate contamination in the eastern Mitidja aquifer has become crucial for water resource control and preservation. In this research, some of the commonly used ensemble methods namely Categorical Boosting, Adaptive Boosting and Random Forest were employed to model the spatial groundwater vulnerability to nitrate concentration based on the maximum acceptable concentration in drinking water according to the Algerian directives (50 mg/L). The study were carried out on the evaluation of different possible influencing parameters including depth to groundwater, permeability, rainfall, vadose zone, distance to river, drainage density, land use land cover, Normalized Difference Vegetation Index, slope, topographic wetness index and elevation. The proposed approach aims to understand and control the vulnerability and pollution process of the groundwater aquifers with the smallest available dataset to assess vulnerability predictive mapping. The best results based on the receiver operating characteristic and area under the curve (ROC-AUC) showed that the Categorical boosting indicated the highest accuracy of 94% then the adaptive boosting algorithm with 92%, and the Random forest algorithm with 89%. It revealed that climatic and hydrogeological factors are the key factors determining groundwater vulnerability to NO3 concentration in the eastern Mitidja. It is concluded that the risk of groundwater pollution covers a large area of the groundwater resources in the city of Algiers, Blida, and Boumerdas in northern Algeria.

Résumé: Erosion can have a negative impact on the agricultural sustainability and grazing lands in the Mediterranean area, especially in northern Algeria. It is useful to map the spatial occurrence of erosion and identify susceptible erodible areas on large scale. The main objective of this research was to compare the performance of four machine learning techniques: Categorical boosting, Adaptive boosting, Convolutional Neural Network, and stacking ensemble models to predict the occurrence of erosion in the Macta basin, northwestern Algeria. Several climatologic, morphologic, hydrological, and geological factors based on multi-sources data were elaborated in GIS environment to determine the erosion factors in the studied area. The conditioning factors encompassing rainfall erosivity, slope, aspect, elevation, LULC, topographic wetness index, distance from river, distance from roads, clay mineral ratio, lithology, and geology were derived via the integration of topographic attributes and remote sensing data including Landsat 8 and Sentinel 2 within a GIS framework. The inventory map of soil erosion was created by integrating data from the global positioning system to locate erosion sites, conducting extensive field surveys, and analyzing satellite images obtained from Google Earth through visual interpretation. The dataset was divided randomly into two sets with 60% for training and calibrating and 40% for testing the models. Statistical metrics including sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (ROC) were used to assess the validity of the proposed models. The results revealed that machine learning and deep learning, as well stacking ensemble techniques, showed outstanding performance with accuracy over 98% with sensitivity 0.98 and specificity 0.98. Policy makers and local authorities can utilize the predicted erosion susceptibility maps to promote sustainable use of water and soil conservation and safeguard agricultural activities against potential damage.