No abstract

The Agbami project team was challenged to generate a field development plan and to assess that plan over a wide range of uncertainty. Complete stakeholder alignment around the technical aspects of the development plan was a critical issue. Field data, laboratory data, and analog data were incorporated into a range of reservoir simulation models. The field data included appraisal well logs, cores, 3D seismic, fluid samples, pressure data, and drill stem tests. Three hundred reservoir models were constructed and ranked using streamline simulation for both static and dynamic attributes to assess the range of subsurface uncertainty. A project team with significant but diverse deepwater development experience was formed and co-located. A formal phased decision making process was implemented. During each phase of the process key parameters of uncertainty were identified and ranked in terms of project impact. Field development options were evaluated in distinct phases over the full range of uncertainty. Experimental Design (ED) methodology was used throughout the evaluation to obtain the maximum information with the minimum computational effort. The paper shows how the results from this process facilitated the identification of the key uncertainties and provided direct input into economic models for decision analysis. In Phase Two of the evaluation the pressure maintenance scheme selected for the 17 million year (MY) reservoir was crestal gas re-injection with peripheral water injection. Crestal gas re-injection was selected for the 14MY and 16MY reservoirs. The facility capacity requirements were also selected during this phase at 250,000 stock tank bbl of oil per day (STB/D) and 450,000 thousand standard cubic ft per day (Mscf/D). Peer reviews and assists were held and alignment was obtained along with action items to be included in the subsequent phase. In Phase Three of the evaluation the well count parameter was investigated and an optimum number of 38 wells was selected. Production profiles were generated and presented in terms of P10, P50, and P90. Introduction This paper will focus on the application of Experimental Design (ED) during Phase Three at Agbami. With drilling depths of up to 10,000 ft below mudline in 4,800 ft of water, well costs at Agbami are at the high end of typical deepwater costs. Therefore, the number of wells including their type and location is an important optimization parameter in the field development plan. The probabilistic forecast of all produced and injected fluids is also a key deliverable of the field development plan. Agbami project stakeholders are aligned on all technical aspects of the field development plan and the project is rapidly proceeding toward sanction and awarding of major contracts.

No abstract

No abstract

Geologic reservoir characterization of Hatter's Pond Field, Mobile County, Alabama was conducted using a combination of lithologic and petrophysical data in order to determine appropriate reservoir layering and flow units for engineering simulation and reservoir management. 15,000 feet of core provided the foundation for the geologic study. Layer models were constructed via computer, and numerical grids were supplied directly to the simulator. Three dimensional computer modelling enhanced understanding of reservoir quality distribution. Lithologic analyses indicate that productive Norphlet Formation lithologies are aeolian dune sandstones capped by a reworked transgressive marine sand sequence. Smackover Formation reservoir units consist of dolomitized transgressive marine carbonates shoaling upward into pellet/oolite bars and sealed by the intertidal and supratidal carbonates and evaporates of the overlying Buckner Formation. Significant variations in porosity and permeability occur both vertically and laterally within the producing formations. Geologic controls on reservoir quality in the Smackover Formation at Hatter's Pond are strongly diagenetic. Texturally destructive dolomitization has obliterated depositional fabrics, and created pore systems controlled by dolomite crystal size/shape and the presence of early fabric-selective leaching in grainstones overrun by exposed facies. Norphlet Formation reservoir character is related to the original depositional fabric, which has been modified by pressure solution and quartz cementation. Without available core, the model would have been much less accurate, because logs alone do not reflect the diagenetic changes that control the reservoir. By combining core, log, and other petrophysical data, the layering and flow units provided to the simulator produced acceptable history matches in as few as 6 simulation runs. Introduction Hatter's Pond Field, located in southwestern Alabama (Figure 1) was discovered in 1974. As of 1/1/92, the field had produced 46 million barrels of condensate, 104 BCF of dry gas, and 11 million barrels of natural gas liquids from two Upper Jurassic units, the Norphlet Formation (sandstone) and the overlying Smackover Formation (carbonate). The combined pay section of the Smackover and Norphlet Formations averages 200 to 300 feet thick at subsea elevations ranging from approximately 18,000' (5,480 m) to 18,300' (5,574 m). The field is a 1 × 5-mile (1.6 × 8-km) north-south trending anticline bounded on the east by the western graben fault of the Mobile Graben (Figure 2). P. 521^