Applying Multivariate Statistical Analysis to Marine Electromagnetic Sounding Data

Summary In 2012-2015, on the Norwegian shelf, a marine electromagnetic sounding surveys were carried out using the differential-normalized electro prospecting method (DNME) with length of several thousand kilometers. The effectiveness of the method is confirmed in reference fields such as Troll. More than 20 wells were drilled, a significant part of forecasts was confirmed. However, new deposits were later discovered at the survey sites, which anomalies had not previously distinguished according to the inversion data. The fact of lack of anomalous effects made us look at the results of the shooting of those years in a new way. For one of the survey lines crossing the reference field, the following results are presented: a multivariate statistical analysis applied to the reprocessed data; one-dimensional inversion of transients simultaneously with the apparent resistivity curve with a more detailed step; inversion within the 2.5D model. The research carried out shows that at the first stage, it is advisable to highlight the area of most significant interest based on the result of multivariate statistical analysis. The results of the multivariate statistical analysis do not contradict the results of 2D modeling.