Sea-Bed Logging (SBL) is an application of the marine Controlled-Source Electro-Magnetic (CSEM) method to detect the presence of hydrocarbon layer beneath the sea bed. This method depends on the large resistivity contrast between the hydrocarbon reservoir, and the surrounding layers of different resistivity. Thus, the ability to detect the presence of different resistive layers is important in processing CSEM data. In this paper, discriminant analysis is applied to simulated data aimed at classifying them into two groups based on the resistivity contrast. Discriminant function analysis is a statistical technique used to predict membership in two or more mutually exclusive groups. Two types of data, with (1000) and without (1) hydrocarbon was used to develop the discriminant model. Wilks' lambda is used to test the significance of the discriminant function as a whole, while measure of F-value for a variable is used to indicate its statistical significance in the discrimination between groups and extent to which that variable makes a unique contribution to the prediction of group membership. The results obtained imply that discriminant analysis has potential in detecting the contrast. © 2011 IEEE.