Ground water quality evaluation using hydrogeochemical characterization and novel machine learning in the Chikun Local Government Area of Kaduna State, Nigeria.

Researchers

Journal

Modalities

Models

Abstract

The investigation collected 50 random water samples from wells and bore holes in the five wards. In the meantime, the Water Quality Index (WQI) in this region was assessed using a novel machine learning model. In this sphere of science, the Emotional Artificial Neural Network (EANN) was used as an innovative technique. The training dataset comprised 80% of the available data, while the remaining 20% was used to assess the performance of the network. The laboratory analysis revealed that the levels of magnesium (0.581 mg/L), mercury (0.0143 mg/L), iron (0.82 mg/L), lead (0.69 mg/L), calcium (2.03 mg/L), and total dissolved solid (105 mg/L) in the water sample were quite high and exceeded the maximum permissible limits established by the National Standard Water Quality (NSWQ) and Water Quality Association (WQA). Except for magnesium, mercury, iron, and lead, all physicochemical parameters are below the utmost permissible limit. Results showed that hydrogeological effects and anthropogenic activities, such as waste management and land use, impact groundwater pollution in the Chikun Local Government Area of Kaduna State up to 60 m deep. The results of the EANN showed that R2 index and normalized root mean square error (RMSENormalized) values for the training and test stages are 0.89 and 0.18, and 0.83 and 0.23, respectively.