EM 1110-2-1100 (Part II)
30 Apr 02
Sandy Hook, NJ, has a return period of 10 years. A northeaster with a 6-ft surge has a return period of
approximately 2 years (Gravens, Scheffner, and Hubertz 1989).
(c) Extratropical events cannot be parameterized in the same manner as hurricanes. Therefore, the
development of stage-frequency relationships based on the Joint Probability Method (to be discussed in the
following section) is not a viable approach for assigning frequency-of-occurrence relationships to
extratropical events. As an alternative approach, frequency indexing is now established through application
of a new statistical procedure called the empirical simulation technique. This approach will be discussed in
the following section.
(d) Because a large number of extratropical storm events have impacted coastal areas of the United
States, the 20- to 30-year database of hindcast storms of the Wave Information Study (WIS) is adequate for
representing the historical population of storms along the coasts of the United States. Wind fields produced
by the storms of this database are input to numerical hydrodynamic long-wave models to produce storm
surges as a function of historically hindcast wind fields contained in the WIS database. Because of the
frequency of events and large area of impact, this database is considered adequate for determining stage-
frequency relationships for design use.
(3) Surge interaction with tidal elevations.
(a) A final consideration on the specification of tropical and extratropical storm surge elevations relates
to the time of occurrence. The timing of storm events with respect to the phase of the astronomical tide
cannot be overemphasized. When a storm surge coincides with a spring high tide, the resulting total surge
can be many times more devastating than the surge alone. For example, a moderate event at low tide can
become the storm surge of record at high tide. An example situation is shown for Hurricane Gloria, which
moved along the Delaware and New Jersey coasts, making landfall on Long Island, NY, at 1100 EST on
27 September 1985. The total storm track is shown on Figure II-5-24, and a more detailed plot of the track
in the vicinity of Long Island is shown on Figure II-5-25 (Jarvinen and Gebert 1986).
(b) The significance of timing between the surge peak and the tide phase can be seen in Figure II-5-26.
The observed storm surge at Sandy Hook, NJ, had a peak value of 7.5 ft and occurred near low tide (1.0 ft).
As a result, the total measured water level was 8.5 ft, corresponding to just 3.5 ft above normal high tide. If
the storm surge had occurred 4.5 hr earlier to coincide with high tide (5.0 ft), the peak surge value would have
been 12.5 ft. The differences in levels of damage resulting from an 8.5-ft versus a 12.5-ft surge level can be
considerable, and the only difference between the two scenarios is a 4.5-hr difference in time of occurrence.
As shown in this example, the phasing of the storm and tide impacts both the design of the structure and the
Storm event frequency-of-occurrence relationships.
(a) The majority of coastal structures are designed to provide some specific level of protection to the
beach and its surrounding population and supporting structures. This level of protection is generally based
on the frequency of occurrence of a storm surge of some specified maximum elevation selected by assessing
the risks of structural failure or consequences of overtopping versus the economics of the cost of the design
project. Therefore, one important aspect of coastal design criteria is the development of stage-frequency or
frequency-of-occurrence relationships for the design area.
Water Levels and Long Waves