ABSTRACT: Reconstructing Flow Strength in Curved Alluvial Channels

David F. Dominic

Previous attempts to deduce flow parameters of rivers from their deposited sediments have focused on either the channel-scale suite of lithofacies or on the character of grain-size distributions. No attempt has yet been made to evaluate the correspondence between these two methods. The present study provides such a comparison, utilizing data from Upper Pennsylvanian and Lower Permian sandstones of northern West Virginia.

Detailed sedimentological logs and measurements of sandstone body geometry were compared with predicted sedimentation patterns generated by a mathematical model of flow in curved alluvial channels. Matching model output to outcrop measurements verified input parameters as those prevailing during deposition. Estimates of bed shear stress were then derived from the inferred position of samples below the reconstructed flow field for the channel bend.

Measured cumulative distributions of grain size were matched to those generated by a simulation model that considers the rate of sediment transport of individual size fractions as bed load and their removal from the bed as suspended load. Correspondence between measured and simulated distribution is good, allowing a second estimation of bed shear stress for each sample.

Estimates of bed shear stress derived from these two methods are well correlated. However, those derived from the channel-bend model consistently exceed those from the grain-size model by a factor of about 5. These results, together with those of previous studies, support the conclusion that estimates of bed shear stress from grain size are too low rather than channel-bend estimates are too high. This underestimation for the grain-size model may be partly explained if grains are less easily suspended than previously assumed.

AAPG Search and Discovery Article #91003©1990 AAPG Annual Convention, San Francisco, California, June 3-6, 1990