Modeling groundwater total dissolved solid from derived electromagnetic data using multiple linear regression analysis: a case study of groundwater contamination

• Published in: Modeling earth systems and environment 2020-09, Vol.6 (3), p.1863-1875
• Main Authors: Aduojo, Ameloko Anthony, Adebowole, Ayolabi Elijah, Uchegbulam, Okezie
• Format: Article
• Language: English
• Subjects: Boreholes; Chemistry and Earth Sciences; Computer programs; Computer Science; Conductivity; Contamination; Data; Dipoles; Dug wells; Earth and Environmental Science; Earth Sciences; Earth System Sciences; Ecosystems; Environment; Geophysical surveys; Geophysics; Groundwater; Groundwater pollution; Groundwater studies; Integrity; Math. Appl. in Environmental Science; Mathematical Applications in the Physical Sciences; Original Article; Parameters; Physicochemical processes; Physicochemical properties; Physics; Regression analysis; Regression models; Software; Statistics for Engineering; Terrain; Total dissolved solids; Waste materials; Water analysis; Water resources; Water sampling
• ISSN: 2363-6203; 2363-6211
• DOI: 10.1007/s40808-020-00796-9

The high concentration of total dissolved solids (TDS) and other physicochemical parameters in groundwater around dumpsites have been used to implicate contamination from decomposed waste materials. A simple multiple linear regression (MLR) TDS model that integrates the TDS data derived from boreholes and hand-dug wells to the geophysical parameters obtained from the frequency-domain electromagnetic (EM) data was developed in this research. This is with a view to efficiently monitor groundwater resources and exploration around the Olusosun dumpsite and its communities. With the aid of the MLR equation, the observed TDS concentration of water samples collected from boreholes and hand-dug wells, and the corresponding estimated ground conductivity data in the vertical dipole mode (VD 40) and horizontal dipole modes (HD 40 and HD 20), obtained from geophysical surveys were regressed in Microsoft Excel software to generate a MLR TDS model. The integrity of the derived TDS model was appraised to examine the possibility of deploying it to investigate the TDS content of groundwater around the study area. The EM data and the resistivity models obtained around the study area confirmed contamination going on around the dumpsite. The developed TDS model can be put to use with high confidence, for groundwater TDS prediction around the study area where there are only terrain conductivity data but with no boreholes parameters. Also, terrain conductivity data alone can be applied to the model to predict the concentration of TDS in groundwater where there are no boreholes and hand-dug wells, therefore reducing the cost and time of determining and monitoring both parameters independently. With the aid of the ArcGIS software, the TDS model was used to generate TDS estimate map for the area. The knowledge of the TDS variability in such a map could give a clue about the integrity of the underground water around the site.