ABSTRACT: An Attempt to Define Sandy Reservoirs in a Proximal Platform Environment Using Impedance Inversion and Sequence Stratigraphy

Pierre Allix, Jean-Pierre Blanche, Herve Coutrix, Van Thang Nguyen

In the everyday work of the explorationist, the location and the determination of the extent of reservoir sands in heterolithic sequences are among the major problems for intensive hydrocarbon exploration. In response to the need for a predictive and precise knowledge of lithology, a new methodology has been developed. The methodology uses the new concepts of sequential stratigraphy and the most recent geophysical techniques. Initially, a sedimentological analysis, which is reinforced by the biostratigraphical information, is conducted on all the well logs and compared with the sampled data from cores, sidewall cores, etc. This sedimentological analysis has two principal objectives: (1) interpreting the well data in chronostratigraphic sequences (recognizing stratigraphic sequences with maximum flood surfaces, erosional surfaces, and determination of trends), which provides the data for the elaboration of the conceptual sedimentological model and (2) classifying the lithofacies in acoustic impedance. After the physical approach is done on the logs, the geophysical analysis of the problem is started. One important phase, stratigraphic deconvolution, is concerned with the improvement of the seismic data. This process optimizes the well calibration and improves the seismic resolution.

The interpretation of the improved seismic data is done jointly by the geophysicist and the sedimentologist, who must

-- follow the major discontinuities observed in the wells (with iteration on the initial data),

-- analyze the geometric relationships between sequence (onlap and toplap), and

-- analyze the architectural information (offlap and downlap).

This interpretation enables the construction of a geometric stratigraphic model (organizing and ordering the intrasequence and intersequence sedimentary relationships). This two-dimensional geologic model, constrained by the a priori information on impedance and seismic travel time is then submitted to a double one-dimensional iterative inversion resulting in: the creation of a synthetic seismogram, an iterative comparison with the initial section and automatic reduction of errors, an optimized model of acoustic impedance and travel time after n iterations, and the calculation of a residual section (the difference between the initial data and the synthetic data of the optimized model).

The quality of the result is analyzed on the residual section. The optimized impedance section obtained constitutes a document directly interpretable in terms of lithology and reservoir (reference wells).

AAPG Search and Discovery Article #91003©1990 AAPG Annual Convention, San Francisco, California, June 3-6, 1990