GUREVICH, ALEXANDER E., University of Southern California, Los Angeles, CA

ABSTRACT: Simulation of Basin Fluid Dynamics: A Way to Higher Efficiency

Powerful computers made basin fluid dynamics simulation possible. But it is difficult to get reliable quantitative estimations because of inadequacy of models of many processes and media properties. Just a few examples. Reconstruction of actual history of subsidence and uplift lacks information for periods of sedimentation interruption. Actual compaction is mostly creep and is a 3-D process hydrodynamically, differing radically from models based on laboratory tests and approximation of porosity/depth plots. Distribution of fractures, crucial for vertical permeability, depends on stress zones migration and mineral substance deposition, that are impossible to reconstruct now. Even deformation of rocks in laboratory tests cannot be directly applied to geologic environment, because cylind ical samples expand horizontally, which is impossible in beds. Utilization of reversed problem solutions, being mostly just data fitting, does not improve physical model but only adjust mathematical formula to get better approximation for a particular case.

Much more promising and efficient way to use basin simulation is to analyze relative roles of different acting factors and select separate geologic characteristics and develop structures of their combinations for statistical correlations. This approach, combining deterministic cause-and-effect basis with statistical way of getting numerical results, provides maximum efficiency, more reliable results, and controllable application.

AAPG Search and Discovery Article #90987©1993 AAPG Annual Convention, New Orleans, Louisiana, April 25-28, 1993.