19 June 2001




K. T. Holland
Naval Research Laboratory
Stennis Space Center, MS 39529-5004
C. L. Vincent
US Army Corps of Engineers Waterways Experiment Station
Vicksburg, MS
R. A. Holman
Oregon State University
Corvallis, OR 97331


The stability of nearshore sand bars was investigated using a long-term, video-derived, dataset. The cross-shore position of the inner sand bar relative to the shoreline at Duck, N.C., was observed to be at approximately the same mean location over multi-year time intervals, but varied dramatically over much shorter time scales on the order of a few days. The simple, lowest order model that the Duck inner bar is stable at a location of 87 m will result in a root-mean-square error of over 25 m. In contrast, nonlinear forecasting methods, which use past behavior as the model basis, showed greater skill in predicting bar change, having a prediction horizon for significant forecasts on the order of five days. The observations and the data in general, suggest that other deterministic models of bar behavior may be limited to forecasts of only a few days because of a strong sensitivity to initial conditions.

Sponsored by the Office of Naval Research.

Published in the Proceedings of Coastal Sediments 99, Volume 3, June 21, 1999.
Naval Research Laboratory Contribution Number NRL/PP/7442 98-0014.
Conference Proceedings