Improved Reservoir Characterization in Low-Permeability Reservoirs With Geostatistical Models
- D.N. Meehan (Stanford U.) | S.K. Verma (Stanford U.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- August 1995
- Document Type
- Journal Paper
- 157 - 162
- 1995. Society of Petroleum Engineers
- 0 in the last 30 days
- 534 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
In fill drilling has been commercially successful in many low permeability,heterogeneous gas reservoirs. Reservoir discontinuities have often beensuspected as a factor in poor gas recoveries on wide spacing. Large verticaland lateral variations in permeability make it difficult to account for partialdrainage at in fill locations. How many wells must be drilled to recover thegas? What are the effects of heterogeneities on optimal well spacing andfracture length?
In this paper, a case history illustrates the power of incorporating highresolution, fine grid geostatistical models in simulating reservoir behavior.Previous reservoir simulation studies provided acceptable matches of flow ratesand pressures by fairly arbitrary reductions in the log derived net pay for theentire reservoir or away from the well. However, these models failed to matchextended pressure buildups. The buildups indicate significantly highergas-in-place in the reservoir than indicated by simulation matches based onsimpler reservoir descriptions. The geo-statistical model presented hereresulted in a good, multiple well history match and matched the long termbuildup.
The techniques for generating the reservoir description are summarized alongwith the reservoir simulation results. Predictions of in fill drilling successwith this model are better than for prior models. Predictions of incrementalgas recoveries from in fill drilling from this model are consistent withobserved results. Reservoir heterogeneities (specifically the lateralcontinuity of permeability) appear to be the most important factors in thisreservoir controlling inadequate drainage of the uppermost intervals. Theselateral heterogeneities appear to be diagenetic permeability alterationsresulting in partial compartmentalization of the many individual sands.
Optimal well spacing in very low permeability reservoirs has been addressedby numerous authors. Wells with permeabilities in the Cotton Valley range(< 0.01 md) generally indicate extremely long "optimal" fracturelengths (often in excess of 1000 ft fracture half-lengths) It is doubtful thatsuch fractures can be created without vastly larger jobs than predicted byconventional hydraulic fracture models. This means that the cost functions forconventional fracture optimization models may be inappropriate in thickintervals with few stress barriers. Inadequate barriers to fracture heightgrowth and reservoir heterogeneities indicate the need for closer spacing andmore moderate fracture lengths.
Continued in fill drilling accomplishes two things, viz. increased access topoorly drained portions of the reservoir with better stimulations andacceleration of recoveries from the most continuous portions of the reservoir.Current well costs can justify incremental recoveries at the current spacinglevels; however, significant gas will remain unrecovered. The importance oflowering well costs is described.
|File Size||3 MB||Number of Pages||6|