CARBONATE RESERVOIR APPROACH
Mentoring - Consulting
The first step prior of facies modeling in integrated study is the facies identification, calibration and propagation. During this step, the geologist is studying cores available in a set of strategic « key » wells and provides core descriptions with lithologic identification, mineral associations, sedimentary features and depositional environment characterization.
Once lithologic logs are delivered, the second step consists in the lithofacies/electrofacies calibration with diagraphies, to characterize each cored facies with conventional logs, prior a propagation to uncored wells. Finally, a petrophysical and dynamic calibration of facies with core measurements allows providing dynamic behavior of each facies
When facies are propagated to a maximum of wells, the phase of data analysis and interpretation starts. A high resolution sequence stratigraphy pattern is interpreted to refine geological model in uniform depositional sequences, and better handle facies variations and associations. Then, for each sequence determined, a vertical stationarity analysis allows understanding the vertical evolution of our reservoir, and a lateral stationarity analysis allows understanding the lateral evolution of our reservoir, and better predict the facies variation in both sampled and non-sampled areas
These two phases of facies identification and data analysis will provide the keys of comprehension of geological evolution of the reservoir, and guide the geologist in the conceptual model elaboration. During this critical phase , the geologist build a conceptual sedimentological block, supported by data, and providing depositional environments, facies associations rules, heterogeneities size, propagation and occurence.
This conceptual model will be the base of the geological model calibration and implementation and will help the geologist to choose the appropriate method of modeling, to select the geostatistical tools adapted to encountered geobodies and select the good interpolator able to predict the system evolution with the highest confidence. A final quality control procedure « back to data » will control that model prediction is fitting both data and concepts, and eventually highlight uncertainties area where data quality/quantity is not sufficient.
All right reserved by GEOXILIA
Copyright © GEOXILIA