Abstract: 4D Heat, Pressure and Fluid Flow Simulation of Sedimentary Basins.

ERENDI, ALEX, and LAWRENCE M. CATHLES III, GBRN Group, Cornell University, Ithaca, NY.

Introduction

Three dimensional modeling of the heat and fluid flow processes that occur in sedimentary basins are becoming increasingly important in oil exploration. Over the last several years we have developed a simple method to capture the geology of a three dimensional volume so that thermal and fluid flow processes in that volume may be simulated. We have also developed tools to visualize the gridded input geology and the model results using Data Explorer, a visualization package developed by IBM.

Methods

The 3D data cube is built from parallel interpreted 2D sections through a 3D survey. Individual cross-sections of data are discretized along strata and pseudowells and connected to form a 3D grid (Fig. 1). The grid consists of quadrilateral elements, each grid section contains the same number of wells and horizons, but the wells can fully or in part fall on top of other wells and need not be vertical. This allows description of complicated fault systems. Three dimensional grids are generated for each time step from the time of basin formation by backstripping and making simple assumptions regarding salt diapirism.

Every grid element is assigned a specific lithology. Numerical algorithms combine lithology and fluid pressure to determine the porosity, thermal conductivity, and permeability of each element. A time- and space-variable basal heatflux and water depth-dependent surface temperature determine temperature in the growing basin. Fluid overpressures are generated by sediment loading and hydrocarbon maturation subject to user defined kinetics.

Basin simulation begins with the solution of the pressure equation. Pressure gradients and Darcy's law determine fluid fluxes, which are transferred to the heat equation to calculate advective heat transfer. Hydrocarbon and other reactions are then advanced one timestep, which may require several iterations of the finite element solver. After completing the first time step, variables are copied into the next grid which has newly deposited and further compacted and faulted strata, and the calculations are performed again for the next time step. Within the solution iteration, additional functions (hooks) are called. Some of these hooks affect the convergence of the solution. For instance, if the pressure below a seal exceeds the lithostatic pressure, the seal is ruptured. This is simulated by changing the seal permeability as a function of the pressure below the seal and iterating so that the pressure is maintained at some predefined fraction of lithostatic. Other hooks, such as the maturation of hydrocarbons, do not strongly affect the number of iterations the solver must take to advance one timestep.

Example

The South Eugene Island Block 330 area is used to demonstrate these numerical procedures. We distribute kerogen as indicated geologically, we place a seal— believed to be capillary— in the basin to produce overpressured compartments in positions observed. The basal heatflux is calibrated to match well log data. We then run 3D models over 30 million years using maturation kinetics and determine where hydrocarbons are produced and where they migrate. The results are visualized as streamlines showing the most likely path of fluid particles.

AAPG Search and Discovery Article #90937©1998 AAPG Annual Convention and Exhibition, Salt Lake City, Utah