Abstract: A Different Seismic Interpretation Approach to Define Stratigraphic Sequences and Predict Facies Distributions in Continental Strata: an Example from the Middle Magdalena Basin, Colombia
Olaya, Ivan D. - Ecopetrol
Conventional methods of seismic stratigraphic analysis, which depend on angular discordance and terminations of seismic reflections, are not very dependable or accurate in continental strata due to the dominance of nearly horizontal and parallel and/or discontinuous and incoherent seismic rejections.
We have developed a philosophically different, robust approach to analysis of seismic reflections from continental strata. This approach relies on prediction of seismic reflection and wave-form attributes from stratigraphic analysis of well-log data; comparison of the predicted with the observed reflection attributes; adjustment, if necessary; and prediction of well-log attributes from observed seismic reflection attributes at other locations.
This method acquires robustness for the following reasons. (1) Our stratigraphic analysis of well logs identifies chronostratigraphic units which are correlated across facies tracts. Since seismic reflections are generated from chronostratigraphic units, observations from the two data sets are comparable. (2) Stratigraphic cycles and their facies components (which are laterally variable within cycles) are identified at high vertical resolution in well logs. This high-resolution information is transferred to the seismic domain in the form of a prediction, using the relation that the magnitude of the impedance contrast multiplied by the continuity of the stratigraphic unit under consideration generates reflections. Well-log predictions of continuity, amplitude and wave-form character of the seismic response are compared with the observed, and the observed seismic character acquires geologic meaning. (3) Since the two data types generate the same types of stratigraphic information, but at different vertical and lateral resolution, the approach of hypothesis formulation and hypothesis testing using alternating data types is a powerful means to evaluate the accuracy of stratigraphic predictions.
AAPG Search and Discovery Article #90933©1998 ABGP/AAPG International Conference and Exhibition, Rio de Janeiro, Brazil