[First Hit]

Datapages, Inc.Print this page

 Click to view article in PDF format.

  

GCSeismic Aids Heterogeneity Hunt*

Satinder Chopra1 and Yong Xu2

 

Search and Discovery Article #40489 (2010)

Posted February 19, 2010

 

*Adapted from the Geophysical Corner column, prepared by the authors, in AAPG Explorer, November, 2009. Editor of Geophysical Corner is Bob A. Hardage ([email protected]). Managing Editor of AAPG Explorer is Vern Stefanic; Larry Nation is Communications Director.

 

1Arcis Corp, Calgary, Canada  ([email protected])

2previously with Arcis Corp, Calgary, Canada

 

General Statement

 Canada’s Athabasca oil sands represent the biggest petroleum accumulation in the world and presently produce more than one million barrels of oil per day. These are the Lower Cretaceous bitumen sand reservoirs comprising the McMurray Previous HitFormationNext Hit that varies in thickness from 10 to 90 meters and occurs at depths of 0 to 400 meters. While the shallow oil sands are exploited by open-pit mining, the deeper reservoirs are produced through some type of insitu bitumen production like SAGD (Steam-Assisted Gravity Drainage).

 SAGD operations require extensive, bitumen-saturated, homogeneous formations for optimum production. The McMurray Previous HitFormationNext Hit, however, is heterogeneous in terms of reservoir continuity, mineralogy, sedimentary facies and Previous HitwaterNext Hit-saturation, and is too complex to be fully understood from the sparse available core database. Surface seismic data are one option for characterizing this reservoir heterogeneity, with a common approach being to use neural networks or statistical analysis at well locations to deduce relationships between seismic attributes and lithology. These relationships are then used to determine lateral lithology variations between wells. 

Figure Captions

uGeneral statement
uFigures
uMethodology
uConclusions



















uGeneral statement
uFigures
uMethodology
uConclusions



















uGeneral statement
uFigures
uMethodology
uConclusions



















uGeneral statement
uFigures
uMethodology
uConclusions




















 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

fig01

Figure 1. Cross-plots of gamma-ray versus (a) P-impedance, (b) Vp/Vs velocity ratios and (c) Previous HitdensityNext Hit. Red samples are from the McMurray Previous HitFormationNext Hit and blue samples are from theWabiskaw Previous HitFormationNext Hit. Note that bulk Previous HitdensityNext Hit and gamma-ray responses (c) increase in a quasi-linear manner.

fig02

Figure 2. Previous HitDensityNext Hit section derived from seismic data. Overlain on the section are Previous HitdensityNext Hit logs (black curves), gamma ray logs (purple) and impedance logs (blue). Wab is the Wabiskaw Previous HitFormationNext Hit; McM is the McMurray Previous HitFormationNext Hit.

 

Methodology

 We describe here a two-step approach to understand the heterogeneity of Athabasca oil-sand reservoirs. The first step involves a rock physics study to understand relationships between lithology and petrophysical parameters. From this effort, lithology-sensitive rock parameters are selected that can be detected seismically. The second step is to derive these lithology parameters from seismic data.

 The first step – rock physics analysis – is carried out for various rock physics parameters across the zone of interest. Parameters that exhibit the best sensitivity to lithology are selected. For example, on Figure 1 we show P-impedance, Vp/Vs velocity ratio and Previous HitdensityNext Hit cross-plotted against gamma ray using log samples from the study area. For these study wells, P-impedance shows a limited ability to distinguish lithology; Vp/Vs ratio shows a gentle variation with lithology; and Previous HitdensityNext Hit appears to be the best indicator of clay content.

 Now that the desired rock parameter – Previous HitdensityNext Hit – has been determined, step two is to do AVO analyses of prestack seismic data to estimate rock Previous HitdensityNext Hit along seismic profiles. Normally, Previous HitdensityNext Hit determination is done using a three-term AVO analysis that requires prestack seismic data with long offsets. In our study we improved this conventional approach by adopting innovative ideas like using a windowed approach instead of a sample-by-sample computation for deriving AVO attributes, reducing distortion due to NMO stretch and offset-dependent tuning, using error-based weights and accounting for the strong reflections from the McMurray Devonian interface.

 On Figure 2 we show a Previous HitdensityNext Hit section derived from seismic data. The lateral variation in seismic-based Previous HitdensityNext Hit shows the richest sand areas (in green color) within the mid-McMurray are around wells 5 and 6, with good shaly cap rocks in the upper-McMurray. These predictions are verified by gamma ray logs acquired in both wells. Recently drilled wells 3 and 7 served as blind well tests. Well 3 found mainly a shaly facies within the McMurray, and the seismic inversion for Previous HitdensityNext Hit agrees with these log calibration data. Well 7 was drilled at the edge of the richest sand zone, and its reservoir also matches the seismic-based Previous HitdensityNext Hit results. In addition, the sandy cap rock within the upper-McMurray in well 7 is predicted by the seismic inversion. 

Conclusions

 This study demonstrates the application of rock physics analysis and the determination of rock Previous HitdensityNext Hit from seismic data can be used to characterize heterogeneity with the McMurray Previous HitFormationNext Hit portion of the Athabasca oil sands. While other rock physics parameters can be used, Previous HitdensityNext Hit seems to be a good indicator of lithology at this study site. This same methodology can be applied to other areas where the objective is to determine heterogeneity within any Previous HitformationNext Hit of interest.

Return to top.

 

Copyright © AAPG. Serial rights given by author. For all other rights contact author directly.