Method for determining formation properties from seismic attributes
AssigneeExxon Production Research Company
InventorsCraig S. Calvert, Vijay Khare, Kenneth E. Dahlberg, Leslie A. Wahrmund
Life of this patent
Estimating the term…
Abstract
A method of geophysical prospecting using seismic trace attribute analysis for formation characterization. More particularly, a method for selection of seismic traces which most accurately represent the subsurface formation in the area of a seismic survey is disclosed. In one embodiment, inflection-point interval analysis is used to account for effects of local surfaces.
Claims17
A method for predicting subsurface formation properties of a designated subsurface formation, which method comprises the steps of:
- obtaining a seismic survey of said designated subsurface formation, said seismic survey comprising a set of seismic data traces;
- selecting a calibration well;
- generating a synthetic seismogram representative of the subsurface formations proximate to said calibration well;
- selecting a study interval and identifying said study interval on both said synthetic seismogram and said seismic data traces;
- selecting one or more seismic attributes for calibration and extracting said calibration attributes from said study interval on both said synthetic seismogram and at least a portion of said seismic data traces;
- determining subsurface formation properties proximate to said calibration well;
- selecting one or more of said seismic data traces, which traces have calibration attributes for said study interval which approximate those of said synthetic seismogram;
- using calibration attributes extracted from said selected seismic data traces and said subsurface formation properties proximate to said calibration well to construct a calibration model for said designated subsurface formation;
- and
- using said calibration model to predict subsurface formation properties for said designated subsurface formation.
A method for predicting subsurface formation properties for a specified subsurface formation, which method comprises the steps of:
- obtaining a seismic survey of said specified subsurface formation, said seismic survey comprising at least two seismic data traces proximate to a well;
- selecting a study interval within said seismic survey, said study interval comprising at least three seismic intervals;
- deriving a synthetic seismogram representative of the portion of said subsurface formation within said study interval and proximate to said well;
- selecting at least one calibration attribute and extracting said attribute from the portions of said synthetic seismogram and said at least two seismic data traces falling within said study interval;
- comparing said calibration attributes and selecting each seismic data trace having calibration attributes which approximate those of said synthetic seismogram;
- using said selected seismic data traces to develop a scaling function for each said calibration attribute;
- applying said scaling function to the portions of said seismic data traces falling within said study interval to produce rescaled calibration attributes for said portions of said seismic data traces;
- selecting seismic data traces having rescaled attributes which approximate said extracted attributes from said synthetic seismogram;
- using said selected seismic data traces to generate a calibration model;
- and
- using said calibration model to predict subsurface formation properties for said subsurface formation.
In a method of seismic attribute analysis of a set of seismic data traces, a method for dividing a designated study interval into a plurality of thinner seismic intervals, said method comprising the steps of:
- locating inflection points on the portions of said seismic data traces falling within said designated study interval;
- for each of said seismic data traces, designating the seismic intervals between consecutive inflection points as inflection point intervals;
- and
- using said inflection point intervals in performing said seismic attribute analysis.
Dependencies are inferred from the claim text.