--> Geostatistical Integration of Multiscale Data to Construct a Hunton Group Geocellular Model: Upscaling Logs and Downscaling Seismic Impedance Volumes

AAPG ACE 2018

Datapages, Inc.Print this page

Geostatistical Integration of Multiscale Data to Construct a Hunton Group Geocellular Model: Upscaling Logs and Downscaling Seismic Impedance Volumes

Abstract

The Hunton Group on the carbonate ramp often exhibits highly variable porosity and lithology resulting in heterogenous production. The generation of an accurate geocellular model requires upscaling of sparse porosity logs followed by their integration with denser seismic impedance volumes. The integration of multi-scalar data consisting of core, logs, and seismic data to construct a 3D geocellular model remains an ongoing challenge. Seismic data provide laterally dense but vertically low resolution, “soft” estimates of lithology and rock properties. In contrast, log data provide corresponding “hard” estimates that are laterally sparse but exhibit high resolution. To bridge this gap in scales, we develop a workflow to upscale well log and core data, and downscale seismic impedance estimates, resulting in an integrated gridded reservoir model with reduced uncertainty.

Our workflow begins with principal component analysis (PCA) of electric logs followed by self-organizing map (SOM) to construct electrofacies logs. We then corroborate the geologic interpretation of the electrofacies predictions using thin sections and borehole images. Porosity logs are correlated with core porosity measurements. Lithology and porosity logs are then upscaled to the size at which the vertical heterogeneity of log properties can be preserved by comparing the logs before and after upscaling. We then construct a vertical variogram from the upscaled well logs and a horizontal variogram from the downscaled acoustic impedance volume.

To populate the 3D volume we (1 establish seismic impedance attribute which correlate to the electrofacies and porosity to design the horizontal variograms for the 3D lithology and porosity models, (2 construct the relationship between the acoustic impedance and lithology and porosity logs at well locations, (3 perform a 3D seismic inversion of acoustic impedance volume (4 downscale (laterally interpolate) the inverted prestack acoustic impedance volume computed at the seismic bin size resolution to geocellular model grid size, (5 create horizontal variogram maps from the downscaled seismic acoustic impedance, and (6 obtain the horizontal variogram parameters and substitute them into lithology horizontal variogram.

The seismic data helps to design the geocellular horizontal variograms of the lithology and porosity 3D models. We illustrate the value of this workflow through application to a Hunton Group reservoir over the Cherokee Platform, Oklahoma, USA.