W. Huang et al. / Ocean Engineering 30 (2003) 22752295
2294
effective long-term water level data for the study of coastal hydrodynamics and
shoreline change in the Long Island's South Shore. In addition, because water levels
at temporary field stations can now be predicted by the RNN--WL model, expensive
data monitoring instruments can be relocated to other new sites.
NOAA has a national water level observation network that covers all the coastal
regions in USA (http://co-ops.nos.noaa.gov/usmap.html). Within a scale ranging
from 60 to 591 km as given in this study, one can always find one or more NOAA
water level stations along the US coast. For many NOAA stations, long-term data
over a period of several decades have been processed and are available for online
download from the NOAA Web site. However, due to the long distance between a
local station and a NOAA station, the differences of phase and amplitude of water
levels are usually significant. This disparity often makes it difficult to apply conven-
tional regression method to transfer the valuable data at a NOAA monitoring station
to a local station in a specific study site. The regional neural network model (RNN--
WL) developed in this study will provide a practical tool for coastal engineers and
researchers to predict long-term historic water level data at a local station from a
remote NOAA station. The RNN--WL model has been programmed for application
to general coastal regions. Two sets of hourly water level data are needed in model
training and verification. Because field data collection is usually expensive, the
RNN--WL model provides a cost-effective alternative for coastal engineers to obtain
long-term data in the regional coastal study area.
Acknowledgements
This study was supported by the Coastal Inlet Research Program, US Army Corps
of Engineers (USACE). The views expressed in this paper are those of the authors
and do not necessarily reflect the views of the funding agency. The authors would
like to thank Thomas C. Wilson, Stony Brook University, for providing the data for
this study. Permission was granted by Headquarters, U.S. Army Corps of Engineers,
to publish this information.
References
Bibike, Y., Solomatine, D., Abbott, M., 1999. On the encapsulation of numerical-hydraulic models in
artificial neural network. Journal of Hydraulic Research 37 (2), 147161.
Deo, M.C., Naidu, C.Sridhar, 1999. Real time wave forecasting using neural networks. Journal of Ocean
Engineering 26, 191203.
Fletcher, D., Goss, E., 1993. Forecasting with neural networks: an application using bankruptcy data. Inf
Management 24, 159167.
Greenman, R., Roth, K., 1999. High-lift optimization design using neural networks on a multi-element
airfoil. Journal of Fluids Engineering, ASME 121, 434440.
Grosskopf, W.G., Aubrey, D.G., Mattie, M.G., Mathiesen, M., 1983. Field Intercomparison of Nearshore
Directional Wave Sensors, IEEE Oceanic Engineering Society. IEEE Journal of Oceanic Engineering
8 (4), 254271.
Previous Page
