Multiscale methods for CSEM data interpretation

Abstract

During my work, I focused my attention on “marine Controlled Source Electro-Magnetic sounding (mCSEM), which is a EM getting information about the resistivity distribution beneath the sea-floor. mCSEM uses a low frequency EM signal generated by a transmitter antenna towed by a ship and received by receivers deployed at the sea-floor. This is very useful for oil companies because it can be used to detect, locate and monitoring oil or gas. Further applications of marine CSEM are the exploration for gas hydrates as a methane resource, and possibly pre-drilling surveys to mitigate hazard represented by hydrates and shallow gas. During my work, I have developed two methods for a fast interpretation of mCSEM data. The first method, called “Singular Function Normalization” method (SFN) is a fast and computationally low cost to get information about the areal resistivity distribution. The is based on the study of the “Magnitude Versus Offset” signals (MVO), which are the values of the amplitude of the electric field measured by a receiver versus the distance source-receiver (offset) represented in a semi-logarithmic scale. Our aim is at emphasizing the presence of anomalous resistive buried bodies, by approximating the MVO signal obtained at each receiver by a singular function, such as the Lipschitz-Holder singularity function and estimating, for each receiver, its exponent. This parameter is expected to vary on the set of the MVO curves acquired during the survey, so to reflect the presence of the anomalous body. The second proposed is the Depth from Extreme Points (DEXP) applied to mCSEM data. The DEXP method, developed by Fedi in 2007, is used in potential field to get a fast imaging of the source distribution. In particular, we get information about the depth to the source and the structural index that is a source-dependent parameter corresponding to the fall-off rate of the field with distance for many, but not all, ideal sources. In my work I have shown that it is possible to apply this also to low-frequency electromagnetic fields, under specific assumptions. I have applied the to the electric field scattered by buried resistive sources. The DEXP is based on the evaluation of the static field at altitudes higher than the measurement altitude thanks to a routine procedure called upward-continuation. I shown that upward-continuation may be well established also for very low-frequency EM fields, under the condition that the distance from the source is kept less than the skin-depth. So, similarly to potential fields, we can get in a fast way, and without any a-priori information, the position and the structural index of the anomalous resistive bodies buried beneath the sea-floor. In a similar way, I have demonstrated that is possible to apply to mCSEM data also a geometric called multi-ridge method, developed by Fedi et al.(2009) for potential field data, again under the condition that the distance from the source is less than the skin-depth. This is very fast and gives information about the depth and horizontal position of the sources, while it does not provide a direct estimation of the structural index. As the DEXP method, the multi-ridge is based on the upward-continuation of the electromagnetic field scattered by the buried resistive sources. The DEXP and multi-ridges methods were tested successfully on synthetic data and on real data-set provided by eni. The results were compared with the results obtained using 3D anisotropic inversion, showing a good agreement with them.