Processing - CSEM/MT - Technology
Processing
Figure 2 Left - Processed CSEM data: magnitude of the electric field at the fundamental (blue) and harmonics as a function of source receiver offset, and phase of the electric field (bottom). Such data forms the basis of the interpretation.
Right - MT data from three stations in the North Atlantic. Apparent resistivity (top) and phase (bottom) calculated from the impedence tensor trotated to principle axis are shown, for the TE and TM modes.
CSEM data processing takes the raw electric and magnetic field time series recorded by the receivers, and uses an optimized FT process to extract the amplitude and phase of the signal corresponding to the transmitter at the fundamental transmission frequencies, and harmonics. Typically data up to the 23rd harmonic of the fundamental can be used, giving a dataset with rich frequency content. Stack window lengths are optimized dynamically to ensure good signal to noise ratio. The data are then merged with source and receiver navigation information to give the magnitude and phase versus offset data that forms the basis of the interpretation.
MT data are processed using a robust remote reference technique to provide impedence tensors at each station, from which the apparent resistivity and phase that form the basis of the data interpretation can be calculated.
Data Interpretation
Data interpretation proceeds in stages gradually adding complexity to the structure. At all points during the interpretation, well log and seismic information is used to guide the analysis to ensure that the final result is consistent with all the available data.
Initial 1D analysis using a forward modelling and inversion approach allows the class of structures to which the data are sensitive to be assessed, highlights lateral changes in the measured resistivity, and provides a starting point for higher dimensional modelling. Any inconsistencies between the data and the pre-survey model can also be investigated and addressed. Common offset normalized anomalies are also plotted at this stage. Although direct interpretation of these can be misleading, they provide a helpful guide to large scale lateral variations in the resistivity structure across the survey area.
Although 1D analysis can be extremely useful as a first step, it cannot account for higher dimensional effects in the data. To this 2D/3D forward modelling, hypothesis testing and inversion must be used. These approaches take into account electrical anisotropy in the structure, and the effects of rapidly varying water depth. Unconstrained inversions are a useful first step, but they result in low resolution images of the resistivity. These are improved considerably by the inclusion of structural information from seismic, and if appropriate resistivity information from well logs, to constrain the inversion process.
The result is a resistivity section or volume which can be integrated with seismic, gravity or other sources of information using our WISE workflows to provide a geologically consistent earth model.