ON THE INTEGRATION OF GEOLOGICAL, GEOCHEMICAL, AND GEOPHYSICAL DATA

The integration of geological, geochemical and geophysical data to generate physical and chemical models of geological systems (e.g. aquifers, geothermal systems, oil reservoirs) is one of the most challenging activities in geological research. Those models are needed for a better location of drilling targets. In this work, three methods are proposed to integrate the geophysical and geochemical data and to compare them with the geological features of the system: 1) Standardization of each variable and sum of the standardized values at each point of the grid to produce an integrated map of all the studied variables, 2) Geographic weighted regression and/or artificial neural networks that considers the geographic dependence of the variables when multivariate regression is applied to obtain a dependent variable (e.g. CO2, temperature) with respect to other geophysical and chemical independent variables, and 3) Cluster analysis that considers the existence of clusters of high or low values of the different variables (univariate and bivariate) in the geographic area allowing a clear definition of the areas where the different variables coincide as indicators of good properties for drilling purposes. The results of all these methods should be compared with the geological features of the area, such as faults, contacts, rock types. These methods are applied to the San Vicente Geothermal Field in El Salvador.