A new technique is developed for modeling 3D permeability distributions. The technique integrates all available data into a fluid flow simulation model. The integrated modeling process honors the essential aspects of the established reservoir descriptions as well as the geological facies model and engineering data. The added value of data integration of the fluid flow simulation is illustrated by the improved accuracy of the resulting well performance predictions and the decrease in time requirements for reservoir modeling history matching.

The technique utilizes diverse data at different scales to condition reservoir models of facies, porosity, and permeability. Such data includes 3D seismic, well logs, core measurements, geologic facies distribution, flow meter logs, and pressure buildup tests. The model building process explicitly accounts for the difference in scale of the various measurements. The model calculates the porosity, facies, and permeability in the inter well volume using geostatistical techniques that are constrained by seismic impedance derived from the 3D seismic data. The use of engineering data in the permeability modeling constrains the results and decreases the history matching time requirements.

A case study demonstrates the modeling technique. A reservoir model is developed for the Unayzah Formation in the Hawtah Field of Saudi Arabia. The Unayzah is a highly stratified clastic reservoir in a mixed fluvial and eolian depositional environment. Data integration provided more realistic reservoir model for this complex geologic setting than the conventional approach. Specifically, the integrated approach provide a reservoir model that captured the complex and highly stratified nature of the lithological units. Fluid flow simulation was carried out for both the new integrated reservoir model and the conventional reservoir model. Results show tremendous savings in history matching time and more accurate results for use in reservoir management production strategies when applying the new technique.


At the present time there is an increasing demand for detailed geological numerical models which incorporate all available data into reservoir characterization studies for the purpose of fluid flow simulation. Conventional modeling techniques, which lack the ability to quantitatively integrate data, tend to produce homogenous results of reservoir properties in the inter-well regions. These models, when fed into reservoir simulations for performance predictions, may generate biased and unreliable results. This necessitates the development of an method that integrates all available data, despite differences in scale, improving the predictive power of the models and making it possible to obtain quicker production history matching from the reservoir simulation.

One of the primary reasons for using geostatistics in the reservoir modeling process is data integration. That is, it allows the incorporation of diverse data of varying scale. This can include very descriptive data, such as conceptual geologic interpretations, or measurements such as 3D seismic time traces, their derivatives, and the resulting interpretations. Geostatistical tools can use data such as 3D seismic to directly or indirectly contribute to the modeling of the inter-well regions. This may provide significant risk reduction in reservoir development and management.

This paper presents a geostatistical methodology that has been adopted for integrating geophysical, geological, and engineering data in reservoir modeling. The Hawtah Field, located in the central part of Saudi Arabia (Fig. 1), has been chosen to demonstrate the approach. Hawtah is a recently developed field with a wealth of modern geological, petrophysical, geophysical, and production data. Incorporating all of this information into the reservoir model exceeded the capability of conventional numerical models.

This content is only available via PDF.
You can access this article if you purchase or spend a download.