Full 3D Compositional Study of the Franklin Field, North Sea
François LORANT1, Françoise BEHAR1, and Jean-Marie LAIGLE2
1 Institut Français du Pétrole, 1-4 avenue de Bois Préau, 92852 Rueil-Malmaison, France
2 Beicip-Franlab, 1-4 avenue de Bois Préau, 92852 Rueil-Malmaison, France
The Franklin Field is located within block 29/5b in the Central Graben Area of the North Sea. This field, under more than 5000 m of sediments, contains high-pressure (P > 100 MPa) and high-temperature (T > 180°C) gas condensate. The reservoir lies at the top of a titled block, within the Upper Jurassic Fulmar sandstone. The petroleum system in this area is composed by three source rocks. The Pentland Fm (Middle Jurassic), located below the Fulmar Sands, contains a coaly organic matter (original TOC = 5%). Over the Franklin reservoir, the Heather Fm and the Kimmeridge Clay Fm (Upper Jurassic) both correspond to marine source rocks with original TOC of 2% and 7% respectively.A 2D study of the Elgin Field, that is adjacent to the Franklin Field, was previously conducted to account for fluid composition in the Fulmar reservoir (Vandenbrouck et al., Org. Geochem. 1998, 30, 1105-1125). This work, that assumed a simplified thermal history, succeeded to reproduce hydrocarbons composition tendencies, especially the preservation of heavy saturates and the increase of the saturated-to-aromatics ratio at high temperature. However, simulations did not provide a quantitative prediction of the reservoir fluid composition. Especially they failed to account for the high proportion of methane (~25 wt %) and the low aromatics content (~3 wt %). The authors suggested that the model might be improved by better taking into account late methane generation, retention of aromatic compounds in the Pentland Fm, and hydrocarbon segregation during migration.In order to address these questions and improve the HP-HT fluid prediction in this region, we achieved a 3D study of the Franklin Field using a more realistic thermal model and a new compositional model for primary and secondary cracking (Behar and Lorant, this conference). This work was conducted in two steps:
Firstly, a thermal model was calibrated on present-day reservoir temperatures, through a sensitivity analysis of both surface temperatures, thermal thinning profile, and temperatures at the bottom of the lithosphere. For that purpose, we developed a specific experimental design approach where each time and/or spatial distribution was treated as a single parameter. Hence, a reduced number of runs were required to successively complete the sensitivity analysis (3 'meta-parameters') and to find the best thermal history scenario for this 3D block.
Secondly, once the thermal model calibrated, 3D runs using a full compositional Darcy migration were performed. The model encompassed 28 layers with 100x100 meshes per layer. Each source rock generated 8 mobile HC fractions, including 3 fractions for the formation of methane from the marine and coal source rocks, and from oil secondary cracking: this allowed to estimate the contribution of each source of gas within the reservoir.This model perfectly reproduced the fluid composition in Franklin, including the amount of methane (we found 27 wt%) and the low concentration in heavy aromatics. The gas source tracking indicated that 2/3 of methane originated from the Pentland Fm, and about 20% from oil secondary cracking. Even though secondary migration might affect the composition of the cumulated fluid, we found that the key-parameters in this issue were the retention of NSOs, the aromatics secondary cracking and the late gas potential in source rocks.
AAPG Search and Discover Article #90066©2007 AAPG Hedberg Conference, The Hague, The Netherlands