--> Abstract: Integrated 3-Dimensional Reservoir Modelling to Develop a Thin Oil Column with Horizontal Wells in the Mahogany Field, Offshore Trinidad, by Y. K. Bally, M. Staines, J. Finneran, J. Ali-Nandalal, and H. Gamero; #90923 (1999)

Datapages, Inc.Print this page

BALLY, YATINDRANATH K., MIKE STAINES, JOE FINNERAN, JOANN ALI-NANDALAL, BP Amoco Trinidad, and HELENA GAMERO, Schlumberger GeoquestCaracas, Venezuela

Abstract: Integrated 3-Dimensional Reservoir Modelling to Develop a Thin Oil Column with Horizontal Wells in the Mahogany Field, Offshore Trinidad

The Mahogany Field (Fig. 1) in Trinidad and Tobago was discovered in 1968 and 1974 with the EM-1 and EM-2 exploratory wells. In 1994, exploration resumed in the field with the drilling of 4 exploratory wells, EM-3, EM-4, EM-5, and EM-5XST, resulting in the discovery and delineation of significant gas, oil and condensate accumulations. Field development commenced in December 1997, with gas supply dedicated to the Atlantic LNG plant currently being built in Trinidad.

The Mahogany Field is a faulted anticlinal structure with Pleistocene age stacked sand and shale sequences. There are 15 reservoirs in this field, with over 2TCF in gas reserves. The 21 sand is a major reservoir with a 63 ft to 75 ft oil leg, overlain by a large gas cap (430 ft at crestal position). Oil reserves at the end of 1998 were estimated at 27MMSTBO (150MMSTBO in place).

The challenge to develop these oil reserves with horizontal wells is complicated by lateral stratigraphic variations, and structural complexity. As a result, a 3D reservoir model is being utilised to plan horizontal wells so that there is optimum placement within targeted flow units. The model incorporates the use of 3D seismic data, cores, Formation Micro Imager stratigraphic interpretation, DGI's Earthvision 3D structural software package, and an in-house multiphase flow reservoir simulator. Approximately 10 wells are planned to develop these oil reserves.

Discussion

The Mahogany Field is a NW-SE trending anticlinal structure (Fig. 1) cut by major down-to-the-east (synthetic) and minor down-to-the-west (antithetic) normal faults. Bed dips average five to ten degrees. The 21 sand is a shelfal sheet (shoreface) sand deposited in the Columbus Basin. Deposition was as a result of the major northeasterly progradation of the Orinoco delta in Pleistocene time. The 21 sand (Fig. 2) consists of generally stacked coarsening upward sands interbedded with shales and silts reflecting deposition in lower to upper shoreface. North to south stratigraphic variations together with structural complexity, add to the challenges of the horizontal program.

A 3D seismic survey was acquired in 1993 specifically for field development. Time structure maps were created and then depth converted for input into the 3D structural model using Earthvision (Fig. 2). The 20 and 21 sands are the largest gas bearing reservoirs in the field, and thus exhibit an anomalous high amplitude character. This allows for confident interpretations in time, and places the higher risk in the depth conversion process. With no flank velocity control, amplitude and coherency extractions are being used to aid in determining the areal extent of the reservoirs. Structural depth maps correlate well with amplitude and coherency outline maps. The seismic data at the 21 sand level has been dramatically attenuated as a result of the shallow gas zones, and faulting patterns on the crest of the field. The dominant frequency is 20-25 hertz and does not allow for resolution of the various stratigraphic boundaries within the thick, 21 sand reservoir.

These rocks are world class reservoir rocks. Core porosities and permeabilities average 30% and 1000md respectively. The 21 sand core data was integrated with the Formation Microscanner Image (FMI) data to form an integrated depositional model. From core analysis, flow units were defined based on permeability to porosity ratios, storage capacity and flow. Utilising flow units, speed zones were identified as high potential targets for the horizontal wells. The flow units along with their petrophysical properties (porosity, permeability, net to gross ratio, connate water saturation), were input into the in-house multiphase flow simulator for full integration of the petrophysical, structural and engineering data.

Well location planning to target the best flow units is challenging because of variations in stratigraphy, a five degree bed dip, placement to avoid early water and gas coning, and limits in the TVDSS window so there are no "peaks and valleys" in the lateral section. Integrated analysis using structure maps, 3D seismic interpretations, and 3D visualisation, allow for viewing and planning the wells so that optimum locations are identified. 3D visualisation of the structure and property maps (effective porosity and volume of shale) key the interpreter into "sweet" spots for horizontal targets.

Integrated Modelling to Develop a Thin Oil Leg

The results of the first well are indeed very encouraging. The well encountered 90% net sand in the 2000 ft lateral section and is producing at 2500BOPD. Techniques like barefoot openhole completions in these unconsolidated sands are being planned for future wells. The 3D model will be refined as wells are drilled so that the challenges of optimally exploiting the field are met.

AAPG Search and Discovery Article #90923@1999 International Conference and Exhibition, Birmingham, England