High quality 3D seismic data over a large UAE field combined with high-end geophysical techniques resulted in spectacular porosity detail from seismic data over this carbonate field. By using special post stack filtering techniques to improve data continuity and discontinuity, two optimal data sets were generated. One data volume was optimized for structure and stratigraphic interpretation and the other for event mapping and quantitative seismic attribute analysis. A multi-attribute calibration method was used to estimate the porosity from the seismic data. Attributes used include seismic and geologic or interpretative attributes. The result was a porosity prediction that was 20% improved over the traditional single attribute approach, as measured on hidden well data. The predicted porosity volume provided high quality detail of reservoir heterogeneity and was very useful in understanding flood front advance in the platform interior and flow patterns in the clinoforms. In addition, the seismic porosity volume provides a means to place future wells to tap by-passed oil and to optimize the location of planned injector and producer zones.
Seismic data is often used as a means to estimate rock properties between well controls. A typical method is to calibrate seismic attribute information with rock properties, such as porosity, from well log information. The seismic attribute data can then be transformed using the calibration function into an estimate of rock properties. In our study a multi-attribute neural network calibration is used to produce a 3D volume of rock property predictions from the seismic data (Hampson 2001).This seismic-scale rock property volume is then used to place new wells in favorable locations or in building geologic models for reserve estimation and/or reservoir simulation.
The study field is located in Abu Dhabi within the United Arab Emirates. The field is positioned over a low relief doubly plunging anticline that is oriented N-NE. The reservoir interval is comprised of a low relief, prograding carbonate platform. The main producing interval is in the Lower Cretaceous. The complex stratal architecture is driven largely by the longer-term (second-order) accommodation history. The field is characterized by significantly different stratal geometries that are clearly imaged by the enhanced 3D seismic. The northern part of the field is characterized by strong basinward progradation that occurs in the late highstand and falling-stage systems tracts. Clinoforms exhibit spatial variability in geometry, composition and texture that significantly impact reservoir geometry and quality. The southern part of the field is characterized by aggradational platform facies that show variations in stacking patterns and composition keyed to position in the sequence-stratigraphic hierarchy. Most notably, strong aggradation during the highstand sequence set, coupled with proliferation of rudist "reef" builders, resulted in development of a complex system of moderate relief, higher-energy rudistid shoals (moderate to high reservoir quality). These shoals are separated by low relief, restricted lagoonal "ponds" that are filled largely with low-energy, mud-prone facies (low reservoir quality).