Support Vector Machine: A Case Study in the Kert Aquifer for Predicting the Water Quality Index in Mediterranean Zone, Drouich Province, Oriental Region, Morocco

Hicham Gueddari, Mustapha Akodad, Mourad Baghour, Abdelmajid Moumen, Ali Skalli, Yassine El Yousfi, Hanane Ait Hmeid, Mohamed Chahban, Ghizlane Azizi, Mohamed Chaibi, Ouassila Riouchi, Mostapha Maach, Ahmed Ismail and Muhammad Zahid

Support Vector Machine: A Case Study in the Kert Aquifer for Predicting the Water Quality Index in Mediterranean Zone, Drouich Province, Oriental Region, Morocco

2022

Hicham Gueddari, Mustapha Akodad, Mourad Baghour, Abdelmajid Moumen, Ali Skalli, Yassine El Yousfi, Hanane Ait Hmeid, Mohamed Chahban, Ghizlane Azizi, Mohamed Chaibi, Ouassila Riouchi, Mostapha Maach, Ahmed Ismail and Muhammad Zahid

The expansion of urbanization and the amplification of anthropic activities in the Rif region require the establishment of wells. However, the irrational exploitation of water and natural conditions have generated the rise of the water table and the increase in pollution. Thus, the assessment of water quality has emerged as a significant concern. This study’s goal is to assess the adequacy of groundwater quality in two aquifers in the vicinity of the Mediterranean Zone - Drouich Province and Oriental Region, Morocco, for drinking water needs by taking 62 water samples of the Kert aquifer for 2019. The Water Quality Index (WQI) classifies water quality: as excellent, good, poor, very poor, etc. That is essential for conveying information about water quality to people and decision-makers in the affected area. The WQI in the Kert aquifer varies from 62.3 to 392.3. The calculation of the water quality index (WQI) of the Kert aquifer view is based that 45.16% of groundwater samples are of poor quality, making them acceptable for drinking. The study’s analysis is established with a geographic information system (GIS) setting. The index map provides decision-makers with a complete and interpretable picture for better water resource planning and management. SVM models are shown to account for 87.71% of the varying water quality score. Different statistical and intelligence models may make the index more predictable. These forecasts assist us in better managing the aquifer’s water quality.

Mostrar más [+]