The prediction and study of air pollution is a complex process due to the presence of controlling factors, different land use, and different sources for the elaboration of pollution. In this study, we applied the machine learning technique (Random Forest) with time series of particulate matter pollution records to predict and develop a particulate matter pollution susceptibility map. The applied method is to strict measures and to better manage particulate matter pollution in Ras Garib city, Egypt as a case study. Air pollution data for the period between 2018 and 2021 is collected using five air quality stations. Some of these stations are located near highly urbanized locations and could be dense with the current rates of development in the future. The random forest was applied to verify and visualize the relationships between the particulate matter and different independent variables. Spectral bands of Landsat OLI 8 imaginary and land cover/land use indices were used to prepare independent variables. Analysis of the results reveals that the proper air quality distribution monitoring stations would provide a deep insight into the pollution distribution over the study site. Distance from the roads and the land surface temperature has a significant effect on the distribution of air quality distribution. The obtained probability and classification maps were assessed using the area under the receiver operating characteristic curve. The outcome prediction maps are reasonable and will be helpful for future air quality monitoring and improvements. Furthermore, the applied method of pollutant concentration prediction is able to improve decision-making and provide appropriate solutions.

The demands upon the arid area for water supply pose threats to both the quantity and quality of social and economic activities. Thus, a widely used machine learning model, namely the support vector machines (SVM) integrated with water quality indices (WQI), was used to assess the groundwater quality. The predictive ability of the SVM model was assessed using a field dataset for groundwater from Abu-Sweir and Abu-Hammad, Ismalia, Egypt. Multiple water quality parameters were chosen as independent variables to build the model. The results revealed that the permissible and unsuitable class values range from 36 to 27%, 45 to 36%, and 68 to 15% for the WQI approach, SVM method and SVM-WQI model respectively. Besides, the SVM-WQI model shows a low percentage of the area for excellent class compared to the SVM model and WQI. The SVM model trained with all predictors with a mean square error (MSE) of 0.002 and 0.41; the models that had higher accuracy reached 0.88. Moreover, the study highlighted that SVM-WQI can be successfully implemented for the assessment of groundwater quality (0.90 accuracy). The resulting groundwater model in the study sites indicates that the groundwater is influenced by rock-water interaction and the effect of leaching and dissolution. Overall, the integrated ML model and WQI give an understanding of water quality assessment, which may be helpful in the future development of such areas.

The increasing water demand in Egypt causes massive stress on groundwater resources. The high variability in the groundwater depth, aquifer properties, terrain characteristics, and shortage of rainfall make it necessary to identify the groundwater potentiality in semi-arid regions. This study used the possibilities of multi-criteria decision approaches (MCDA), geographical information system (GIS), and groundwater field data to delineate potential groundwater zones in the Tushka area, west of Lake Nasser, South Egypt. Furthermore, groundwater potentiality identification can help decision-makers better plan and manage the water resources in this promising area. Eight controlling factors were utilized to achieve the objective of the present work using multi-criteria decision analysis (MCDA) approaches, namely the analytical hierarchy process (AHP) and frequency ratio (FR) models. The controlling parameters were integrated with the geographic information system (GIS) to develop the zones of groundwater potentialities. The results revealed that high and moderate-potential zones cover approximately 61% and 52% of the total area in the AHP and FR models, respectively. A total of 44 groundwater production wells along with the well yield were collected and used to validate the models. The results were evaluated using the receiver operating characteristics (ROC) curve. The best-performing prediction rates achieved by AHP and FR were 83% and 81%, respectively. Finally, the obtained results indicated that the AHP model achieved better performance than the FR model.