--> Integrated Analysis of the Bakken Petroleum System, U.S. Williston Basin, by Jack Flannery and Jeff Kraus #10105 (2006).

Datapages, Inc.Print this page

PSIntegrated Analysis of the Bakken Petroleum System, U.S. Williston Basin

By

Jack Flannery1 and Jeff Kraus2

 

Search and Discovery Article #10105 (2006)

Posted May 23, 2006

 

*Poster presentation, at AAPG Annual Convention, Houston, Texas, April 10-12, 2006 (with adaptation for HTML version)

 

Click to view posters in PDF format.

Poster 1 (3.6 mb)      Poster (4.3 mb)      Poster (3.4 mb)

 

1Tethys Geoscience, Denver Colorado

2Formerly Tethys Geoscience, Denver Colorado, currently ExxonMobil, Houston Texas ([email protected])

 

Abstract 

As much as 300 billion barrels of oil have been generated from Upper Devonian-Lower Mississippian Bakken shales in the U.S. Williston Basin. Recent industry activity has been focused on the middle Bakken siltstone trend in Richland County, Montana. Operators there are enjoying impressive success rates from wells that test 500 barrels of oil per day, on average. Horizontal drilling, completion, and fracturing technology are generally credited with opening up the historically disappointing play. Companies are now extending the play in to other parts of the Basin. Future success will rely largely upon developing a thorough understanding of the play as it is currently being exploited and, especially, upon using that understanding to identify key geologic controls of Bakken prospectivity that can be capitalized on elsewhere. 

Regional structure and isopach maps, along with geochemical, thermal, and rock properties data, are used to construct a three-dimensional thermal and fluid flow model of the basin. The model provides unique insight into the evolution of the Bakken petroleum system and allows us to predict reservoir quality, source maturation, and volumes of oil expelled and currently trapped within the middle Bakken. Integration and spatial analysis of modeled results, regional maps, and measured data shed light upon the fundamental geologic variables and relationships that control Bakken prospectivity. Key factors include maximum reservoir temperature, stratigraphic architecture, and small-scale porosity development. We interpret potential for additional middle Bakken exploration downdip from the current siltstone play where the middle Bakken thickens and becomes sandier.

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uAbstract

uGeologic framework

uAcknowledgments

uPetroleum system

  uThermal calibration

  uGeochemical calibration

  uModeling results

uPlay elements

  uAnalysis

  uPlay fairway

uConclusions

 

 

 

 

 

 

 

 

 

 

Geologic Framework

Figure 1-1. The U.S. portion of the Williston Basin contains more than 30,000 wells drilled since the early 1900’s that produce from more than 20 intervals.

Figure 1-2. Major structural elements and crustal boundaries can be interpreted from reduced-to-pole (RTP) magnetic data.

Figure 1-3. Stratigraphic column.

Figure 1-4. Thinning across the Nesson Anticline documents its inception during the latest Cretaceous. Maximum depth of burial and temperature were also reached during the latest Cretaceous.

Figure 1-5. Laramide uplift shed a thick package of Lower Cretaceous clastics into the western part of the basin. Deposition was localized by reactivated basement faults in the northwest part of the basin.

Figure 1-6. Triassic sedimentation returned to the basin depocenter. The thick in the southern portion of the area is related to deposition in the Powder River Basin to the southwest.

Figure 1-7. Uplift of the ancestral Rocky Mountains deposited a thick sequence of post-Bakken Mississippian strata along the western margin and in the basin’s depocenter.

Figure 1-8. The Silurian depocenter was developed in its present-day location. A carbonate platform developed along the basin’s eastern edge, while sedimentation was minimal to the west.

Figure 1-9. The Cambrian Deadwood Sandstone was deposited unconformably above Precambrian basement. Basement faults localized Cambrian sedimentation patterns.

 

Click to view the stratigraphic evolution of the Williston Basin (Figures 1-9 to 1-4).

Figure 1-10. Stratigraphy and lithofacies of Middle Bakken.

Figure 1-11. Log features of Bakken Formation in A.H.E. L. #12-31 Grassy Butte H-3.

Figure 1-12. Stratal geometry of the Bakken Formation.

Figure 1-13. Lower Bakken shales pinch-out along coincident with the Richland County MT trend. 

Figure 1-14. Middle Bakken dolomites thicken subtly in the current area of activity. The sandy unit is well developed and thickens to the north.

Figure 1-15. Upper Bakken shales thicken dramatically toward the basin center. 

Click to view Bakken section in sequence (lower to middle to upper).

Figure 1-16. Annual production from the Bakken showing early exploration in the basal Sanish sand, Bakken shale exploration in the early 1990’s, and the recent drilling in the middle Bakken in Richland County, Montana.

Return to top.

 

The Bakken Formation straddles the Devonian-Mississippian boundary and is one of more than 20 oil and gas producing formations in the Williston Basin (Figures 1-1 and 1-2). The Williston Basin chronostratigraphic column (Figure 1-3) highlights stratigraphic units used to make the regional 22 depth structure and 19 isopach maps (e.g., Figures 1-4, 1-5, 1-6, 1-7, 1-8, and 1-9) that form the foundation of our regional analysis, as well as the primary input in to the three-dimensional basin model.

The Bakken Formation (Figures 1-10, 1-11,1-12, 1-13, 1-14, and 1-15) is informally subdivided into a middle dolomitic or silty member, which is sandwiched between upper and lower organic-rich, black shales. The black shales have generated approximately 300 BBO. Drilling in the active trend in Richland County, Montana is focused on the uppermost lithologic unit in the middle Bakken, which is predominantly dolomite in Montana and becomes more siliciclastic in North Dakota. 

The middle Bakken also contains a thick sand-rich unit that is present in North Dakota and Canada, but not in Montana. This sandstone unit presents an additional, largely untested target in the middle Bakken. It also provides the migration pathway for Bakken-sourced oils in to Saskatchewan. 

The Bakken Formation pinches out along the southwestern margin of the Basin. In the Richland County trend, the middle Bakken directly overlies low-permeability Three Forks carbonates and is overlain by the upper shale, which serves as the top seal. This stratigraphic trap is highly effective and is a major control over the geographic extent which is termed the Siltstone Pinch-Out Play. 

Figure 1-16 shows, on an annual basis, production from the Bakken. The fundamental controls of middle Bakken prospectivity are:

1) Trapping Mechanism

  • The pinch-out forms a highly effective stratigraphic trap along the southwestern basin margin.

  • The pinch-out also enhances oil accumulation, with oil saturation >90% measured in the reservoir.

  • Structural or combination structural/stratigraphic traps will be necessary for the middle Bakken to be prospective elsewhere.

2) Maximum Temperature

  • Maximum temperature was reached in the Early Tertiary and has decreased since, “freezing” the petroleum system. Maximum temperature greater than 100C is important for:

    • Oil generation and expulsion

    • Natural fracture development

    • Optimal reservoir quality development.

 

Acknowledgments 

Data Sources:                           Technology:

IHS Energy                              ESRI ArcGIS

NDGS                                     IES PetroMod3D

USGS                                      Spotfire DecisionSite

Humble Geochem. Serv.

 

Bakken Petroleum System 

Thermal Calibration 

   

Geochemical Calibration 

 

Modeling Results 

 

The basin model was calibrated with more than 12,000 corrected bottom hole temperature (BHT) measurements provided by the North Dakota Geologic Survey (NDGS) and the North America heat flow database, from Southern Methodist University (Figures 2-1 and 2-2). A relatively good correlation between regional heat flow and geothermal gradient provided additional support and was used as input to the 3D thermal modeling. A local thermal anomaly exists in the southwestern portion of North Dakota. Previous workers comment on a thermal anomaly along the Nesson Anticline, but this anomaly is not apparent in the complete database (Figures 2-1, 2-2, and 2-3). 

Thermal calibration was also substantiated byTmax data (Figure 2-4) from the upper and lower shale units. We had no vitrinite reflectance (only modeled reflectance [Figure 2-5]) or other paleothermometers to further calibrate the model. 

The middle Bakken reached its maximum temperature in the Early Tertiary and has since cooled 20-30C. 

Both the upper and lower shale units are very organic-rich across much of the basin (Figure 2-6). Measured hydrogen index decreases as maturity increases towards the basin center (Figure 2-7).  

Bakken-sourced oils (Figures 2-8 and 2-9) are generally found where the Bakken is mature, except in the Poplar Dome area where faulting has provided cross-formational migration pathways and in the northern part of the U.S. Williston Basin, where Bakken oil is migrating northward.  

The greatest volume of oil generated from the Bakken was in the northwest of the basin center (Figure 2-10), where both the upper and lower Bakken shale members are thickest and mature. The map of the relative volume of oil generated has been regridded and is unit-less (Figure 2-11).

 

Analysis - Building the Play Elements 

Spotfire DecisionSite Analysis 

 

Critical Element Analysis 

Figure 3-1 demonstrates that oil test flow rate is depth dependent. Most successful Bakken wells have been drilled between 9,300 feet and 11,500 feet. The upper (shallow) limit is governed by thermal maturity and correlates with the depth at which significant oil expulsion begins. 

Figure 3-2 shows a gradual increase in gas flow rate, which abruptly declines below 11,500 feet. This may signify a decline in reservoir quality with depth. However, more drilling is required to sufficiently determine whether a depth control over reservoir quality exists.  

Figure 3-3 suggests a gradual increase in Gas:Oil Ratio (GOR) with depth, but the trend is poorly defined. Because GOR does not correlate well with depth or maturity, and the Bakken is a high-quality oil-prone source rock, possibly reservoir quality (RQ) or completion practices are the fundamental controls. 

Figure 3-4 shows that tests with low oil flow rates tend to have higher gas flow rates. Because the Bakken is not overmature for oil generation anywhere in the basin, this relationship reinforces the idea that RQ and well completion practices govern gas production and GOR. Vertical Bakken tests have higher GOR than do horizontal. Recent Bakken completions in Richland Co., MT have lower GOR. 

Figure 3-5 illustrates the depth-dependant relationship between oil expulsion and depth. Thickness and maturity control oil expulsion in the Bakken. 

Figure 3-6 shows that local charge does not appear to govern oil test rates. However, in areas where the Bakken is immature, oil tests are low.

 

Play Fairway Interpretation 

 

The first oil saturation map (Figure 3-7) is the result from PetroMod’s default compaction model. The second oil saturation map (Figure 3-8) is derived from the calibrated porosity-depth curve, and illustrates the result of 300 BBO with too little pore space to occupy. The actual distribution of oil saturation probably lies somewhere between these two results. We attempted to predict the occurrence of fracturing caused by overpressure during peak oil generation (>120C). We also modified default compaction parameters of the middle Bakken siltstones to match published core porosity data (Figure 3-9). 

Depth of burial, estimated reservoir quality (Figure 3-10) and modeled oil saturation (along with lower Bakken data [Figure 3-11) are used to construct a map of middle Bakken play areas. The spots on the map are middle Bakken wells completed since January 1, 2005, when we completed our interpretation of the play fairways (Figure 3-12).

 

Conclusions 

1) Maximum temperature, which was reached in the early Tertiary, is the governing factor over:

  • Oil generation

  • Fracture development

  • Primary reservoir quality.

2) The effectiveness of the stratigraphic trap along the southwestern basin margin is primarily responsible for:

  • >90% oil saturation

  • High drilling success rates.

3) These conditions also exist:

  • Downdip, toward the basin center

  • Along strike to the southeast.

4) Additional middle Bakken reservoirs exist in North Dakota; this relationship may expand the play if:

  • Structural traps can be identified

  • Fracture porosity is developed.

5) Bakken shales may prove productive with modern drilling and completion.

Return to top.