A multidisciplinary approach to modelling secondary migration: a Central North Sea example

Abstract

Abstract Numerical simulations assist in understanding secondary migration, and can help estimate which traps in a basin may have received a hydrocarbon charge. This is illustrated by a suite of numerical models used to track the secondary migration of hydrocarbons into an Upper Jurassic Fulmar discovery in the Central North Sea. The models include areal flow calculations using Exxon’s proprietary reservoir simulator, and cross-section calculations using TEMISPACK, a commercial basin modelling package. While the application of conventional reservoir simulation techniques to geologic time-scale problems is itself a novel approach, the most important development has been in the integration of fluid flow modelling with techniques for dating the timing of hydrocarbon charge. These techniques have been used to constrain and validate the model results. In addition, sensitivity studies permit evaluation of the required permeability for effective lateral migration. The areal calculations provide information about migration timing and the fill history of several traps. These fields are charged from mature Kimmeridge source rocks, mapped within the local drainage area. Several model runs were used to evaluate the sensitivity of migration timing to carrier bed permeability and capillary pressure characteristics. Calculations indicated that for reservoir quality carrier beds (permeability ≥ 1 mD) secondary migration can be geologically instantaneous, and that even low permeability silts can serve as effective migration pathways. A cross-sectional model was used to integrate cross-stratal and strata-parallel migration. The model included hydrocarbon expulsion from Kimmeridge source rocks into the Fulmar sandstone and lateral migration up the flank of the structure. Drilling results indicate that leakage into younger Tertiary strata has occurred. The calculation sugests a portion of the top seal breached by 10–20 Ma. Evidence of partial seal loss includes a seismic chimney interpreted above the predicted breached seal, and an appraisal well high on the Jurassic structure which encountered hydrocarbon-stained reservoir rocks with an apparent palaeohydrocarbon contact. Furthermore, hydrocarbons in the overlying Tertiary reservoirs can be geochemically tied to the remaining Jurassic accumulation. Finally, analysis of authigenic cements, particularly the K-Ar ages of fibrous illite, suggests that hydrocarbons occupied the trap until approximately 10 Ma. Analysis of Tertiary hydrocarbon fluid inclusions further indicates that hydrocarbons entered these Tertiary reservoirs approximately 10 Ma. These latter two techniques now enable the modeller to apply temporal constraints to hydrocarbon charge simulation. Such constraints allow ‘history matching’ analogous to field scale reservoir simulations.