Why That Great Discovery Wasn't So Great: The Limitations of 2D Seismic

Abstract

ABSTRACT Many small to medium-sized oil and gas discoveries turn out to be much smaller than anticipated at the time of the discovery. One reason for this is the continued use of conventional 2D seismic data for prospect development and pre-drilling economic evaluation, without ensuring that only dip lines are used for interpretation. The limitations of 2D data have been known for many years1234. A computer model study of a small gas-filled sand bar embedded in shale shows that interpretation of a " bright spot" from a grid of lines with a spacing as small as 500 m can give an area for the amplitude anomaly nearly six times the area of the sand body. This type of error can be avoided by using 3D seismic data for prospect development. THE PROBLEM Many small to medium-sized oil and gas prospects me successful in that they are classified as discoveries, and are perhaps even completed and produce hydrocarbons, but are abandoned before even a small fraction of the reserves estimated to be present (before or even after the discovery) have been produced. Sometimes, a 3D seismic survey is used after the discovery to appraise it for efficient development, and this shows that the size of the prospect has been grossly exaggerated in the previous interpretation based on 2D seismic data. In this case, development of the discovery is usually no longer economical. My personal experience is that this kind of problem is more common when the prospect is based solely or mainly on seismic reflection amplitude anomalies (" bright spots") than when there is a defined structural trap. The study reported here was prompted when I was told an operator who had looked at about twenty blocks for a Gulf of Mexico lease sale had decided not to bid on any of the blocks for which he had 3D seismic data. ANALYZING THE ERRORS- METHOD To investigate the possible size of errors in the interpretation of a 2D seismic survey, I built a simple computer model of a small gas-filled sand bar. The base of the sand bar was placed on a horizontal surface, to remove the effect of dip as far as possible. Using this model, I generated a grid of four synthetic seismic lines to be interpreted as if they were line shot across a real prospect. Two of the lines wereclose to " dip" lines, that is, lines which are close to normal to the strike of the sand bar, and two were close to strike lines. The synthetic data set showed the expected amplitude anomalies from the sand body. I interpreted the four lines as if they were real data, and generated a contour map of the amplitude Anomaly. From this map I picked a location for an exploration well. A well in this location on the model would find nearly 5 m of gas-filled sand.