complete mild-slope equation (except for very small domains). Typically, coastal wave

propagation problems involve the modeling of very large domains. For example, consider

the case of 12 second waves in water of 15 m depth. The wavelength L is about 136 m;

an 8 km by 8 km domain is about 3600L2 in size. The difficulties associated with solving

such large problems spawned the development of several simplified models (e.g. the

"parabolic approximation" models (Dalrymple et al. 1984; Kirby, 1986), RCPWAVE

model (Ebersole, 1985), EVP model (Panchang et al 1988), etc.).

However, these

simplified models compromised the physics of the mild-slope equation: they model only

one- or two-way propagation with weak lateral scattering.

Such models are hence

applicable only to rectangular water domains for a very limited range of wave directions

and frequencies. Most realistic coastal domains with arbitrary wave scattering cannot be

modeled with these simplified models.

This manual describes a wave model called CGWAVE developed at the University

of Maine under a contract for the U.S. Army Corps of Engineers, Waterways Experiment

Station. CGWAVE is a general purpose, state-of-the-art wave prediction model. It is

applicable to estimation of wave fields in harbors, open coastal regions, coastal inlets,

around islands, and around fixed or floating structures. While CGWAVE simulates the

combined effects of wave refraction-diffraction included in the basic mild-slope equation,

it also includes the effects of wave dissipation by friction, breaking, nonlinear amplitude

dispersion, and harbor entrance losses. CGWAVE is a finite-element model that is

interfaced to the SMS model (Jones & Richards, 1992) for graphics and efficient

implementation (pre-processing and post-processing). The classical super-element method

as well as a new parabolic approximation method developed recently (Xu, Panchang and

Demirbilek 1996), are used to treat the open boundary condition. An iterative procedure

(**c**onjugate **g**radient method) introduced by Panchang et al (1991) and modifications

suggested by Li (1994) are used to solve the discretized equations, thus enabling the

modeler to deal with large domain problems. This manual provides a brief review of the

basic theory in Sections 2 and 3, an overview of how this theory is implemented in

4

Integrated Publishing, Inc. |