Frequency of erosion studies (Cont.)

storms but possess sufficient variability to fill in the gaps in the historical

data.

To perform the EST, historical storms impacting a site are broken down

into the parameters that impact the engineering aspect of interest: storm

track, maximum winds, radius to maximum, pressure deficit, etc. These

variables are termed input vectors. The storm response of interest, in this

case vertical erosion of the capped mound, is also calculated for each his-

torical storm using an appropriate model (in this case LTFATE is used).

The response of interest is referred to as a response vector. During EST

simulations, N-repetitions (say 100 or more) of T-year responses (say 100

to 200 years) of the response vector of interest (vertical erosion for cap-

ping projects) are produced providing mean value frequency relationships

with accompanying confidence limits such that probability of occurrence

can be defined with error band estimates. In other words, the mean verti-

cal erosion for a range of return intervals with confidence limits (based on

the number of standard deviations) are produced by the EST procedure.

Application of the EST to a capping project involves a series of sequen-

tial steps to calculate the cap erosion thickness. A description of these

specific steps are provided in Appendix G, using the Mud Dump study

mentioned above as an example. The remainder of this section summarizes

the required steps and concludes with specific recommendations on how

to translate frequency of erosion values into a cap erosion layer thickness.

To define the required cap erosion layer thickness as a function of

depth at a specific site, the following procedure was developed. It con-

sists of a site-specific quantitative analysis approach. First, an appropri-

ate set of storms, both tropical and extratropical for east coast sites, and

tropical for Gulf coast sites, have to be selected. Next, the hydrodynamic

inputs (the time series of storm surge levels and tide elevations, their re-

sulting currents, and wave heights and periods) for the selected storms

have to be developed for input to an erosion model such as the LTFATE

model. These inputs are often developed using a 3-D ocean circulation

model such as ADCIRC (Luettich, Westerink, and Scheffner 1992) or

CH3D (Scheffner et al. 1994).

After the water level, current, and wave data for specific storms are

available and in the proper format, LTFATE can be run to calculate the

thickness of the layer eroded by each storm for a range of capped mound

configurations (elevations and cap materials). These data are then input

into the EST program, which makes 100 or more simulations of mound

erosion over a long time period (100-200 years). The results can then be

analyzed with standard statistical techniques to produce frequency of ero-

sion estimates for the various mound configurations tested. Finally, the

frequency of erosion estimates, including expected annual erosion and the

longer return period erosion estimates, are converted into a design erosion

layer thickness.

The following paragraphs discuss the results of such a study and how

these can be used to compute erosion layer thickness.

Chapter 8 Long-Term Cap Stability