Frequency of erosion for the combined impacts of tropical and extratropical storms (Cont.)

important that erosion per storm and the cumulative effects be computed as

realistically as practical.

A simple method to compute cumulative erosion is to compute an annual

average erosion then multiply that value by the number of years of interest. This

can be done by examining the full set of training storms modeled in the erosion

frequency analysis, then summing the maximum erosion from each storm and

dividing by the number of storms to compute the average maximum erosion per

storm. The average annual erosion could then be computed as the average

maximum erosion per storm times the average number of storms per year (e.g.,

2.375 for the Mud Dump site). This method would likely produce extremely

conservative estimates of annual erosion because successive storms would not

necessarily produce erosion in the same location. Also as the mound erodes, the

elevation decreases, which decreases the erosion rate during future storms. This

method also includes the erosion from severe, infrequent storms which would

perhaps cause some significant cap erosion such that the cap would have to be

repaired.

A correction for the gross annual erosion estimates computed by the above

method could be calculated by computing the total mound erosion resulting from

a series of low to moderate intensity storms (those with erosion frequencies of

less than 5-10 years) applied consecutively (using LTFATE) to a specific mound

configuration. The mound geometry from the first storm would be the initial

geometry for the second storm and so on. The maximum total erosion at any

location on the mound after a series of storms that could normally be experi-

enced in a year (say two to four for the Mud Dump) applied consecutively could

then be compared with the maximum total cap erosion of each storm summed

individually. The correction factor would then be the ratio of the consecutive

total maximum erosion divided by the individual total maximum erosion.

Average annual erosion would then be the number of storms per year times the

maximum average erosion per storm times the correction factor. Cumulative

erosion would then be the corrected average annual erosion times the number of

years of interest.

A more sophisticated estimate of cumulative annual erosion values would be

to use LTFATE to model erosion for a particular capped mound configuration

for a period of 10 to 20 years from which the training storms were selected. The

storm-induced capped mound geometry from the initial storm would be, as

above, the input geometry for the following storm, with the resulting capped

mound geometry from each preceding storm becoming the input geometry for

the subsequent storm.

At the end of each year, the maximum erosion, average erosion thickness, and

area of erosion (as defined in Figure G11) would be computed. Because of the

multiple years of data, running averages of each of the quantities could be com-

puted along with basic statistics such as the average, maximum, and standard

deviation. With these values a considerably more realistic estimate of annual

and cumulative annual erosion is more likely. Additional research on the

application of this suggested approach to actual projects is planned to determine

G28

Appendix G Procedures for Conducting Frequency-of-Erosion Studies