Reactive Transport Modeling and Reservoir Characterization: Recent Advances in Geoscience and Engineering Applications

Yitian Xiao* and Gareth Jones
ExxonMobil, USA
*[email protected]

One of the key challenges in reservoir characterization is the accurate prediction of reservoir quality distribution and alteration associated with various diagenetic reactions at both geological and production timescales. Reactive transport modeling is an emerging technology that can be used to simulate ground-water flow coupled with chemical reactions to facilitate predictions of the dynamic reservoir behavior of multi-phase fluid-rock interactions. We have applied reactive transport modeling to a number of case studies that significantly advanced our reservoir characterization capabilities for both geoscience and engineering applications. For example, we have developed the first fully-coupled 3-D reactive transport models for dolomitization in both early reflux and late fault-controlled hydrothermal systems. The modeling results revealed complex spatial and temporal variations of limestone, dolomite, and anhydrite distribution and the significant implications on carbonate reservoir quality. Other successful applications included modeling early diagenesis associated with a fresh-water lens in an isolated carbonate platform; geothermal convection and burial diagenesis in a salt-buried isolated platform; and fault-induced hydrothermal flow and illitization and permeability reduction in a sandstone reservoir. For engineering applications, we applied reactive transport modeling to study formation damage associated with water and steam injections in sandstone reservoirs; acid stimulations and worm-hole development in carbonate reservoirs; and the long-term fate and risks of injecting CO₂ and H₂S in saline formations. Modeled positive and negative feedbacks induced by artificial diagenesis provided guidance to optimize injection strategies. Reactive transport modeling, when sufficiently integrated with traditional methods and calibrated with field data, has the potential to generate physically viable predictive concepts and reduce the uncertainty in predicting the spatial distribution of reservoir properties that impact both geoscience and engineering business decisions.

AAPG Search and Discovery Article #90077©2008 GEO 2008 Middle East Conference and Exhibition, Manama, Bahrain