Comparison between the Different Approaches of Secondary and Tertiary Hydrocarbon Migration Modelling in 3D Basin Simulators

Sylvie Pegaz-Fiornet, Bernard Carpentier, Anthony Michel, and Sylvie Wolf
IFP 1&4 avenue de Bois-Préau 92852 Rueil-Malmaison France

Hydrocarbon migration at basin scale and geological time-scales is modelled with different techniques. This paper aims to summarize, to compare and to illustrate the different approaches commonly used to simulate secondary hydrocarbon migration and dysmigration in 3D basin modelling. The models are distinguishable by many aspects: the physical modelling, the numerical methods of resolution and the obtained results.

Darcy flow modelling is continuous in time and space. Hydrocarbon migration is driven by buoyancy, fluid pressure field and capillary pressure. This model can take into account a multicomponent and multiphase description. It can be coupled with a pressure-compaction model. Even though the computing time of this model is long, due to the use of iterative algorithms with convergence criteria, it has been improved using new schemes and a parallel approach. Darcy model is well suitable for transient flow in low permeable areas; it is known to give a good description of hydrocarbon leakage through mud-rock sequences. However, it is not always enough efficient in permeable layers for very rapid secondary migration.

Invasion percolation modelling is decoupled from pressure-compaction computation. Hydrocarbons move instantaneously under the effects of buoyancy and capillary pressure. The migration temporal scenario is driven by the fluid expulsion speed from the source-rock. The secondary migration process is quite well simulated with this model but invasion percolation can sometimes give an unexpected description of hydrocarbon leakage out of traps. Moreover, in some cases, it can bypass small sand lenses.

Like invasion percolation, ray tracing method has a fast computing speed. This model is linked to traps knowledge: potential hydrocarbon accumulation areas have to be identified at the beginning of the simulation contrary to the other models. Ray tracing assumes that hydrocarbons migrate instantaneously, under the only influence of buoyancy, from the source-rock to the previously defined traps. It can give a compositional and multiphase description of hydrocarbons. Seal leakage can also be modelled.

The limits and the simplifying hypothesis of each method will be explained and clarified with synthetic examples. Then, we will focus on the comparison of the obtained results with a special attention to the differences in the traps filling history. This will be illustrated by examples, based on synthetic and real case studies, in various petroleum systems.

The coupling of these different models would be an interesting approach to develop in order to use the adequate method for each geological process that we want to simulate.

AAPG Search and Discovery Article #90091©2009 AAPG Hedberg Research Conference, May 3-7, 2009 - Napa, California, U.S.A.