Induced polarization effect in land and marine CSEM data: applications for hydrocarbon exploration

The recent improvements in data acquisition and data interpretation techniques, and the encouraging results obtained by Russian and Chinese geophysicists using the induced polarization (IP) method have stimulated a renewed interest in its use in hydrocarbon (HC) exploration. In this thesis we show the results of IP measurements conducted on shale HC source rocks samples, and apply the information gained to model typical geoelectrical formations of onshore and offshore HC deposits. With these results we can consider the feasibility of detecting a measurable IP response associated with the deposit. We also develop and test an inversion technique for quantitative interpretation of the IP data. Over recent years a large amount of effort has been applied to understanding and improving the inversion theory for recovering a true model in a controlled source electromagnetic (CSEM) setting. Most recent 3D inversion schemes are inverted strictly for real resistivity values even though there is sufficient evidence that the resistivity of rocks is described by a complex function. This property of rocks is manifested by the IP phenomenon and its principles have been used extensively in the search for mineral deposits for many years. Historically, this method of investigating the complex resistivity has been restricted to land based IP surveys. In this thesis we show that a similar technique can be used for offshore HC exploration. The Cole-Cole conductivity model is tested with complex resistivity data and detailed mineralogy of hydrocarbon bearing shale rock samples. Inversion routines are developed and tested using synthetic data to recover the three empirical variables from recorded complex resistivity data. These empirical variables are the decay coefficient (C), chargeability (m), and the time constant (T). For three shale samples, detailed geologic analysis using QEMSCAN was conducted to help better understand the inversion results. Complex resistivity values are calculated from recorded EM data from 0.0156 Hz to 9216 Hz. Using the observed mineralogical data we were able to fit the recorded complex resistivity data using the Cole-Cole model for two of the three samples and a spherical inclusion, three-phase Generalized Effective Medium Theory of IP (GEMTIP) model for the last sample. The results showed both the effects of electrode and membrane polarization at different frequencies. The gained knowledge from the QEMSCAN results helped in understand how varying grain size and pyrite concentration controls the IP effect. Forward geophysical modeling of hydrocarbon systems is accomplished using geologic inputs from the shale sample analysis to deposit-scale HC models. The IBCEM3DIP code which is based on the Integral equation method, developed by the Consortium for Electromagnetic Modeling and Inversion (CEMI), is used for forward modeling. Using the developed HC models, the effect of deposit-scale changes in Cole-Cole parameters is tested on synthetic IP data. The results clearly show that while the membrane polarization is at the lower end of IP detectability, the electrode polarization (from pyrite) is at the very high end. Thus it should be easily detectable, even in an environment with a higher level of noise. We have also proved that one can invert for both the real and imaginary parts of the EM signal in a marine setting with sufficient accuracy to delineate between the two bodies. With advances in the understanding of the IP effect through IP modeling, forward modeling and inversion, detection and discrimination capability will be improved for hydrocarbon source and possible reservoir targets, leading to a greater efficiency in petroleum exploration.