Datapages, Inc.Print this page

AAPG GEO 2010 Middle East
Geoscience Conference & Exhibition
Innovative Geoscience Solutions – Meeting Hydrocarbon Demand in Changing Times
March 7-10, 2010 – Manama, Bahrain

Improved Extraction and Quality Control of Pre-Stack Seismic Attributes

Roy Burnstad1; Timothy Keho1

(1) EXPEC ARC, Saudi Aramco, Dhahran, Saudi Arabia.

We present a procedure for improved extraction and quality control of pre-stack seismic attributes from wide azimuth 3D land surveys. Both the pre-processing and extraction stages of the procedure rely on a target oriented, multi-term decomposition algorithm similar to that used in surface consistent processes; such as, statics, amplitude balancing and deconvolution. Quality control plots indicate that the decomposition procedure produces pre-stack attributes, which correlate better with porosity models derived from well logs. Decomposition allows consideration of other effects, such as anisotropy prior to Poisson reflectivity estimation from amplitude versus offset measurements. Our method utilizes normalization at two stages: first, during processing, trace normalization factors are computed using a large time window defined by the target horizon, and second, normalization is applied to the extracted attributes. During the first stage, a large background time window is selected for normalization on a swath-by-swath basis in a source, receiver, offset and azimuth consistent manner. Both the offset and azimuth terms in this step do not vary significantly with spatial position. The second stage consists of four components: (1) swath-by-swath amplitude mode before noise removal, (2) swath-by-swath deconvolution mode, (3) swath-by-swath amplitude mode post-deconvolution, and (4) survey wide amplitude mode prior to AVO measurements. The second stage occurs after AVO analysis. Here we normalize target Poisson reflectivity measurements by an estimate taken across a large background time window. This stage is performed in a sub-surface consistent manner with offset and azimuth terms allowed to vary spatially across the survey area. Application to a 3D survey over a carbonate oil field in Saudi Arabia showed correlation of pre-stack attributes with the oil water contact. Synthetic models indicate that the correlation is due to porosity variation, not fluid type.