STRATEGIE DE PROSPECTION HYDROGEOLOGIQUE PAR PREVISION DE L’ÉPAISSEUR DE SOCLE A FORER

Abstract

This study aims to test the predictive power of a statistical model (model “ESOF”) to determine the thickness of base to drill in a context of water drilling in crystalline and cristallophyllian bedrock. This model was developed using the approach of multiple linear regression based on the depth of the first significant arrival of water and the thickness alteration in the geological context of the Ivorian base. The assessment model includes an estimate and analysis of the performance criteria and robustness in calibration and validation phases. The results obtained showed that the correlation coefficient (R) is between 0.582 and 0.731, the bias (B) varies from 10,265 to 12.466, and the mean square error (RMSE) is between 12.664 and 14.213. The results for the robustness showed that the model is robust overall (8.2% -14.9%). The "ESOF" model is an efficient, robust and parcimonious model. It can thus be used in the strategy of hydrogeological prospection in crystalline and cristallophyllian bedrock in general and particularly in Ivory Coast.