Avulsion
Clusters in Alluvial Basins: Statistical Tools for Quantifying Sand-Body
Distributions and Implications for Reservoir Modeling and Interpretation
Hajek, Elizabeth1, Paul Heller1,
Snehalata Huzurbazar2, Benjamin Sheets3, Chris Paola4
(1) University of Wyoming, Laramie, WY (2) University of Wyoming, Laramie, (3)
ExxonMobil Upstream Research Company, Houston, TX (4) University of Minnesota,
Minneapolis, MN
Current models for predicting basin-scale
fluvial stratigraphy stem largely from theory and typically focus on the
spatial and temporal distribution of coarse-grained channel-belt deposits in
fine-grained overbank material. Such stratigraphy is dominantly controlled by
long-term channel avulsion behavior within a basin. Avulsion models used to predict
channel-belt distributions typically assume channels move, either randomly or
deterministically, to low spots throughout the model space. Recent observations
of ancient deposits and physical experiments indicate that there is a
previously-unrecognized pattern of channel-belt organization, where clusters of
closely-spaced channel-belt deposits are separated from each other by extensive
intervals of overbank deposits. Using a combination of field, remote sensing,
3D seismic, well-log, and experimental data, we quantitatively characterize the
distribution of channel deposits in the Ferris Formation (Upper
Cretaceous/Lower Paleogene; Hanna Basin, Wyoming) and experimental stratigraphy
generated at the St. Anthony Falls Research Lab (University of Minnesota). We
use Ripley's K-Function to show that both basins exhibit
statistically-significant channel-belt clustering over certain spatial scales.
This parameter describes degree of channel-body clustering at a range of
distances and might ultimately be used as a model input to characterize spatial
lithologic heterogeneity in alluvial reservoirs and basins. Furthermore, our
observations suggest that avulsion clusters form reservoir-scale features that
may not appear channel-like in shape but rather may form irregular
sand-dominated zones. Hence, seismic cubes processed to enhance lithologic
differences may offer advantages over volumes processed to maximize vertical
resolution in clustered alluvial intervals.
AAPG Search and Discover Article #90063©2007 AAPG Annual Convention, Long Beach, California