Petroleum Systems Modeling on the Move: Scientific Concepts, Numerical Models and Exploration Practices

Abstract

The evolution of Basin and Petroleum Systems Modeling (BPSM) has certainly changed the exploration world since it appeared in the 1970s and 80s. Massive innovations were necessary to work out, refine, simplify and implement models for the most relevant thermal, mechanical and fluid flow processes acting from pore to basin scale, and to derive the values of the key parameters from geophysical measurements on field, well and laboratory scale until commercially applicable solutions became available to support exploration decisions. We want to describe typical workflows from four different periods with application examples to illustrate the enormous changes during this development. 1980‐1990: The Concepts: Basin Modeling first focused on geochemistry, and the development teams evolved as spin‐offs mostly with close connections to geochemical labs. Models and data for source rock generation and expulsion linked with burial histories was the main focus in many studies. Simulating thermal and compaction histories was performed at one‐dimensional (1D) points or as multi‐1D simulations in 2D and 3D data models, and linked to source and reservoir maps to enable estimates of generated and expelled volumes. This new method developed rapidly in exploration applications, investments increased rapidly and many development teams were founded in the research centers of oil companies and in academic institutes. These original seed points made it possible for team members to build startup companies and almost all current BPSM geoscience companies and groups are successors of these early developments. 1990‐2002: The Science: This was the main period for the key numerical implementations to describe the entire chain of Petroleum Systems Modeling processes from generation to migration and accumulation. Especially the three main fluid simulator types for Flowpath, Invasion Percolation and Schlumberger‐Private Darcy Flow and combinations between them were very much discussed. Multi‐component (PVT) fluid models were added. In addition, geological models for the structural evolution (erosion, salt, tectonics) were added and refined and additional formulations for in‐reservoir processes such as biodegradation, or models for H2S were developed. BPSM expertise was based on the integration of various scientific disciplines including geochemistry, physics and geology. 2002‐2010: The Products: The intensive research and development efforts resulted in two clear commercial trends for software products with an emphasis either on usability and rapid applications or on the technically advanced modeling. Solutions were found for easy to use and understand models that set a new standard in the geoscience world and enabled the user community to emerge from their expert niche. On the other hand, high end solvers were developed for advanced 3D process modeling. Numerical methods, local grid refinement methods and computer platforms increased model resolution by an order of magnitude every 2‐3 years resulting in better accuracy, bigger models, smaller cells, more fluid components and phases. The introduction of probabilistic concepts has been of special importance during the entire development. 2010‐2018: The Data: There was an increasing focus on the ability to consume data such as well logs and seismic from growing exploration data bases. Many products were integrated with and linked to more general geoscience products and platforms. Integrated workflows with other geoscientific modeling techniques were developed, such as the modeling of lithospheric processes, structural formation and tectonics (structural restoration), clastic sedimentation and the formation of carbonates (stratigraphic forward modeling) rock diagenesis (reservoir assessment) and geomechanical modeling (seal assessment). More complex structural models could be analyzed in 3D, and interesting new workflows with seismic modeling were developed. The application to unconventional Shale plays was an incentive to refine source rock models (kinetics, components, phases, adsorption) and to understand prospect‐scale geomechanical effects and integrate multiple‐ Schlumberger‐Private well data. Biogenic gas and gas hydrate exploration also resulted in new modeling techniques. The integration of petroleum systems modeling with resource assessments is a specific new development that is enforcing the integration of deterministic and probabilistic approaches. Attempts are made to directly use volume and properties from high definition (HD) Petroleum Systems Modeling in the play and prospect assessment analysis. The input for such HD models are seismic derived facies cubes and log derived fluid component refinements, also taken into account asphaltenes and resins on the heavy petroleum side and inert gases on the low end. The ability to work with prospect scale models is a general trend and is expected to improve applications of in‐reservoir process modeling such as diffusion vs segregation. We will show typical examples from the different periods to illustrate this story, and mention some of the current changes in the digital world regarding the handling of big data in the cloud and the usage of machine learning methods to open opportunities for better data integration, semi‐automatic model building and calibration, and the development of basin catalogs.

AAPG Datapages/Search and Discovery Article #90349 © 2019 AAPG Hedberg Conference, The Evolution of Petroleum Systems Analysis: Changing of the Guard from Late Mature Experts to Peak Generating Staff, Houston, Texas, March 4-6, 2019