Benchmarks for Numerical Coastal Models
 
 

Don L. Boyer

Environmental Fluid Dynamics Program

Department of Mechanical and Aerospace Engineering

Arizona State University, Tempe, AZ 85287-6106
 
 

Laboratory experiments considering the interaction of an oscillatory along-shelf current with a long shelf, shelf-break, continental-slope topography incised by a single isolated canyon are conducted; the fluid is linearly stratified and the effects of background rotation are considered. The laboratory physical system is a circular tank in which an annular shelf, shelf-break, continental slope topography is placed in its center, with the shelf and coastline being in the inner portion of the system and the deep water being away from the tank center. Modulating the turntable rotation rate about the mean, f/2, drives background oscillatory motions, where f is the Coriolis parameter. Both subinertial and superinertial frequencies are investigated. Limitations of the lab experiments relating to the method of forcing the background motions and to the requirement that the flows be hydrostatic are discussed.
 
 

In addition to better understanding the motion in the vicinity of submarine canyons, a primary objective of the laboratory study is to produce data sets of laboratory observations that can be used as benchmarks (or tests) of numerical models. To meet this objective it is necessary to conduct a careful and objective error analysis of the laboratory experiments so as to demonstrate clearly the level of uncertainty of the laboratory observations. This is accomplished in the present study by performing repeated experiments under seemingly the same system parameters and then doing a statistical analysis of the various observables such as the time-dependent and mean horizontal velocity, stream function and horizontal divergence fields at various levels and the time-dependent and mean isopycnal displacements at selected locations.
 
 

The laboratory observations are compared with a finite element coastal model developed by Dr. Haidvogel, Rutgers University, New Jersey, USA. The laboratory and numerical models show that the oscillatory forcing leads to a rectified (mean) flow driven along the shelf break in a direction with the shelf on the right facing downstream. A description of the physical mechanisms leading to the rectified flow is given and differences between the laboratory and numerical models are discussed.