Distinguishing data with and without hydrocarbon in scaled tank experiments using spline interpolation and normalized mean square error

This work applies spline Interpolation and Normalized Mean Square Error (NMSE) techniques to process data acquired from scaled tank experiments that replicates seabed logging (SBL) technique. SBL uses controlled source electromagnetic (CSEM) mechanism in its operation. It works by using resistivity contrast as hydrocarbon (HC) reservoirs are known to have high resistivity value of 30 - 500Ωm in contrast to sea water of 0.5 - 2Ωm and sediments of 1 - 2Ωm. Acquiring data in real SBL environment is very costly therefore scaled down tank of scale factor 2000 was built to replicate the environment. The experiment started by collecting signal from the tank with and without presence of HC at various parameters variations. In this work parameters such frequency and transmitted EM amplitude were varied to investigate whether presence of HC had contributed to the magnitude of the received EM waves. All the acquired data were processed using cubic spline interpolation and NMSE were calculated. It was found that when HC was present in the tank NMSE calculated were higher than when no HC in the tank for all the parameter variations. This indicates that combination of spline interpolation and NMSE are able to distinguish data with and without HC information for SBL environment using scaled down environment. © 2014 AIP Publishing LLC.