EST simulation results - vertical erosion versus frequency-of-occurrence

each event. To define an erosion magnitude consistent with the tropical storm

analysis, the total summation of erosion magnitudes per year was selected as the

parameter of interest. For example, if three storm events were simulated during

the first year, the sum of the three vertical erosions would be used to define the

parameter for which frequency-of-occurrence relationships would be computed.

The computational process is described in the following section.

EST simulation results - vertical erosion versus frequency-of-

occurrence

To most effectively use the results from the EST simulations for cap erosion

layer thickness design, frequency of vertical erosion curves and tables should be

generated from the data. For the Mud Dump site example, vertical erosion

versus frequency-of-occurrence relationships were generated for each of the

100 simulations described above for each of the six depth/mound height

configurations for both tropical and extratropical storms.

The frequency curves for each simulation are generated by (a) rank-ordering

the computed erosion magnitudes, (b) generating a cumulative distribution

function (cdf, P(x) versus magnitude), and (c) interpolating an erosion magnitude

for an n-year event from the cdf for a probability of occurrence P(x) of the form

resulting in an erosion versus frequency curve for each simulation.

Tropical storms. In the analysis of the 100 frequency relationships, an

average vertical erosion magnitude is computed relative to each return period.

From the EST simulations of tropical storms, an example plot of the 100 recur-

rence relationships and mean value (indicated by O) for the 8-ft mound located

at an 83-ft depth is shown in Figure G7. Note that the spread of data points

about the mean demonstrates a reasonable degree of variability, as would be

expected of a stochastic process.

Finally, the standard deviation of the 100 events relative to the mean is com-

puted as a measure of variability. Output for design purposes contains only the

mean frequency-of-occurrence relationship with a +/- one standard deviation

band. An example of this output is shown in Figure G8 for the 8-ft mound at the

83-ft depth shown in Figure G7. Table G8 summarizes the frequency-of-

occurrence of vertical erosion from tropical storms for all six mound configura-

tions in the form of a mean value and +/- standard deviation error that can be

added to or subtracted from the mean value.

Extratropical storms. A set of analyses identical to those made for tropical

storms should be made for the extratropical storms. From the Mud Dump site

analysis, an example plot of the 100 recurrence relationships and mean value

(indicated by O) for the 8-ft mound located at an 83-ft depth is shown in Fig-

ure G9. As for the tropical storms, the spread of data points about the mean

demonstrates a reasonable degree of variability, as would be expected of any

stochastic process. An example of the mean frequency-of-occurrence relation-

ship with a +/- one standard deviation band is shown in Figure G10 for the 8-ft

G23

Appendix G Procedures for Conducting Frequency-of-Erosion Studies