Storm tide frequency analysis for the coast of North Carolina, north of Cape Lookout

[From the U.S. Government Printing Office, www.gpo.gov]

c25~~r

Coastal Zone .a- OFco
Informnation , f
Center 4

NOAA Technical Memorandum NWS HYDRO-27 * -

';rES�
NOV 24 4:

STORM TIDE FREQUENCY ANALYSIS FOR THE COAST OF
NORTH CAROLINA, NORTH OF CAPE LOOKOUT

Francis P. Ho
Robert J. Tracey

COASTAL Z0ONK
, Office of Hydrology COASTAL ZONE
Silver Spring, Md. INFORMATION CENTER
November 1975

* QC
851
.U6
H6
no.27

�

n oaao NATIONAL OCEANIC AND / National Weather
llnoaa ATMOSPHERIC ADMINISTRATION Service

,o6 ,v

NOAA TECIINICAL MEMORANDA

National Weather Service, Office of Hydrology Series

The Office of Hydrology (IHYDRO) of the National Weather Service (NIWS) develops procedures for making
river and water supply forecasts, analyzes hydrometeorological data for planning and design criteria for
other agencies, and conducts pertinent research and development.

NOAA Technical Memoranda in the NWS IHYDRO series facilitate prompt distribution of scientific and
technical material by staff members, cooperators, and contractors. Information presented in this series
may be preliminary in nature and may be published formally elsewhere at a later date. Publication 1 is
in the former series, Weather Bureau Technical Notes (TN); publications 2 to 11 are in the former ser-
ies, ESSA Technical Memoranda, Weather Bureau Technical Memoranda lTDrF). Beginning with 12, publica-
tions are now part of the series, NOAA Technical Memoranda, NIWS.

Publications listed below are available from the National Technical Information Service, U.S. Depart-
ment of Commerce, Sills Bldg., 5285 Port Royal Road, Springfield, Va. 22151. Price: $3.00 paper copy;
$1.45 microfiche. Order by accession number shown in parentheses at end of each entry.

Weather Bureau Technical Notes

TN 44 HYDRO 1 Infrared Radiation from Air to Underlying Surface. Vance A. Myers, May 1966. (PB-170-
664)

ESSA Technical Memoranda

WBTM HYDRO 2 Annotated Bibliography of ESSA Publications of Hydrological Interest. J. L. EH. Paulhus,
February 1967. (Superseded by 11BT! IHYDRO 8)

IBTFl HYDRO 3 The Role of Persistence, Instability, and Moisture in the Intense Rainstorms in Eastern
Colorado, June 14-17, 1965. F. K. Schwarz, February 1967. (PB-174-609)

1"'BT IHYDRO 4 Elements of River Forecasting. Marshall M. Richards and Joseph A. Strahl, October 1967.
(Superseded by WB?1 IHYDRO 9)

I'BTM HYDRO S Mleteorological Estimation of Extreme Precipitation for Spillway Design Floods. Vance A.
Myers, October 1967. (PB-177-687)

W1BT Y-IYDRO 6 Annotated Bibliography of ESSA Publications of Ilydrometeorological Interest. J. L. II.
Paulbus, November 1967. (Superseded by 1VBTh IIYDRO 8)

IIBTM HYDRO 7 Meteorology of Major Storms in Western Colorado and Eastern Utah. Robert L. Wleaver,
January 1968. (PB-177-491)

WBTM HYDRO 8 Annotated Bibliography of ESSA Publications of lHydrometeorological Interest. J. L. E!.
Paulhus, August 1968. (PB-179-855)

WTBTM HYDRO 9 Elements of River Forecasting (Revised). Marshall 1M. Richards and Joseph A. Strahl,
Harch 1969. (PB-185-969)

11IBTM HYDRO 10 Flood larning Benefit Evaluation - Susquehanna River Basin (Urban Residences). Harold
J. Day, March 1970. (PB-190-984)

WBT1 IHYDRO 11 Joint Probability Method of Tide Frequency Analysis Applied to Atlantic City and Long
Beach Island, N.J. Vance A. Mlyers, April 1970. (PB-192-745)

NOAA Technical Memoranda

Nl1S HYDRO 12 Direct Search Optimization in Mathematical Modeling and a Watershed Model Application.
John C. Monro, April 1971. (COMl-71-00616)

NI!S HYDRO 13 Time Distribution of Precipitation in 4- to 10-Day Storms--Ohio River Basin. John F.
Miller and Ralph 1H. Frederick, May 1972. (COr-72-11139)

NWS IIYDRO 14 National Weather Service River Forecast System Forecast Procedures. December, 1972.
COM-73-10517)

(Continued on inside back cover)

NOAA Technical Memorandum NWS HYDRO-27

STORM TIDE FREQUENCY ANALYSIS FOR THE COAST OF

NORTH CAROLINA, NORTH OF CAPE LOOKOUT

Francis P. Ho
Robert J. Tracey

Office of Hydrology ('
Silver Spring, Md. Jh' _ " e
November 1975 ijFf G-, -TE

U.S. DEPARTMENT OF COMMERCE NOAA
COASTAL SERVICES CENTER
2234 SOUTH HOBSON AVENUE
.HARLESTON, SC 29405-2413

An_ . ..r
elf _ Property of CSC Library

T-
~,o pTMOp,

UNTE STATES
UNITED STATES /NATIONAL OCEANIC AND/ National Weather
DEPARTMENT OF COMMERCE ATMOSPHERIC ADMINISTRATION Service
Rogers C. B. Morton, Secretary Robert M. White Administrator George P Cressman Drector _
George P Cress~m anDr e t ~

CONTENTS

Abstract ................................ 1

1. Introduction .......................... 1

1.1 Objective and scope ...................... 1
1.2 Authorization ......................... 2
1.3 Study method ......................... 2

2. Summary of historical storms .................. 2

2.1 Hurricane tracks ............... 3
2.2 Historical notes on hurricanes ..3
2.3 Winter coastal storms ..................... 8

3. Climatology of hurricane characteristics ............ 9

3.1 The general method - South of Cape Hatteras ... 9
3.1.1 Probability distribution of hurricane intensity .......10
3.1.2 Probability distribution ot radius of maximum winds .....10
3.1.3 Probability distributions of speed and direction of forward
motion . .........................10
3.1.4 Frequency of hurricane tracks . ...............10
3.2 Special considerations North of Cape Hatteras .........11
3.2.1 Conditional probability question ..............11
3.2.2 Parameters for landfalling hurricanes from northeast
quadrant .........................11
3.2.3 Parameters for landfalling hurricanes from southeast
quadrant .........................12
3.2.4 Landfalling track frequency .................12
3.2.5 Hurricanes passing inland ..................12
3.2.6 Exiting hurricanes .....................14
3.2.7 Alongshore hurricanes ....................14

4. Hurricane surge .........................14

4.1 Surge model ..........................14
4.2 Shoaling factor ........................15

5. Winter coastal storm tides ..................15

5.1 Evaluation of northeaster tide levels in study area ......15
5.2 Analysis of tide gage records .................15

6. Tide frequency analysis by joint probability method .......16

6.1 The joint probability method ..................16
6.2 Astronomical tides .......23
6.3 Tide frequencies at selected points..23
6.4 Adjustment along coast .....................24

iii

CONTENTS

6.5 Reference datum. ......................24
6.6 Comparison of frequency curve. ...............25

7. Relation of this report to disaster planning. ........25
8. Coordination and comparison with other reports. .......26
References .. ...........................27

TABLES

1. Hurricane and tropical storm parameters - N.C. - Va. border .. . 19

2. Hurricane and tropical storm parameters - Wright Monument, N.C. 20

3. Hurricane and tropical storm parameters - Salvo, N.C. .....21

4. Hurricane and tropical storm parameters - Ocracoke Inlet, N.C. .22

5. Hurricane and tropical storm parameters - Cape Lookout, N.C .. . 22

6. Comparison of tide frequencies. ................26

LIST OF FIGURES

1. Location map. .........................30

2. Tracks of major hurricanes from the southeast quadrant affecting
the study area, 1871-1974. ...................31

3. Same as figure 2 but for hurricanes from the southwest
quadrant. ...........................32

4. Track of tropical storms and hurricanes showing motion from
northeast. ...........................33

5. Probability distributions of central pressure of hurricanes and
tropical storms adopted for Wright Monument, N.C., (a) landf ailing
storms from the southeast quadrant and alongshore storms, (b) land-
falling storms from the northeast quadrant, (c) exiting storms, and
(d) all landfalling storms, reproduced from Climatology
Report .. ............................34

6. Probability distributions of speed of storm motion adapted for
Wright Monument, N.C., for (a) landfalling storms from the south-
east quadrant and exiting storms, (b) landfalling storms from the
northeast quadrant (c) inland storms, and (d) alongshore storms. 35

List of Figures -- Continued

7. Frequency of landfalling hurricanes and tropical storms. Circled
dots are count for 50-n.mi. segments from Climatology Report. 36

8. Accumulative frequency of alongshore and inland hurricanes and
tropical storms for Wright Monument, N.C., 1871-1974 ...... 37

9. Schematic diagram illustrating distances L and LL for hurricanes
inland ............................. 38

10. Shoaling factor, North Carolina coast north of Cape Lookout.
Adapted from Barrientos and Chen (1975) ............ 39

11. Tide frequencies at Hampton Roads, Va., from winter coastal
storms. Plotted points are seasonal (October-May) maxima .... 40

12. Probability distribution of astronomical high and low tides
for hurricane season, Avon, N.C., 1955-1973 .......... 41

13. Tide frequencies at Wright Monument, N.C., for several classes of
storms: (a) landfalling, (b) alongshore, (c) inland, and
(d) exiting hurricanes and tropical storms; (e) winter storms;
(f) all storms ......................... 42

14. Tide frequencies at Ocracoke Inlet for several classes of storms:
(a) landfalling, and (b) alongshore hurricanes and tropical storms;
(c) winter storms; (d) all storms ................ 43

15. Total tide frequencies at selected points on the open coast at
Cape Lookout, and Salvo, N.C., and at the N.C.-Va. border. .. . 44

16. Figure 16.--Coastal tide frequencies. North Carolina, north of
Cape Lookout .......................... 45

17. Comparison of tide frequency curves at Ocracoke Inlet, (-) this
report, (---) Corps of Engineers report (1963) ......... 46

STORM TIDE FREQUENCY ANALYSIS FOR THE COAST OF NORTH CAROLINA,
NORTH OF CAPE LOOKOUT

Francis P. H-o and Robert J. Tracey
Water Management Information Division
National Weather Service, NOAA, Silver Spring, Md.

A report on work for the Federal Insurance Administration, Department of
Housing and Urban Development, by the National Oceanic and Atmospheric
Administration, Department of Commerce.

ABSTRACT. Storm tide height frequency distributions are
developed on the coast of North Carolina, north of Cape
Lookout, for the National Flood Insurance Program. Storm
tides are computed from a full set of climatologically
representative hurricanes using the National Weather
Service numerical-dynamic storm surge model. Winter
storm effects are estimated from tide gage records.
Storm tide levels from all storms are shown in coastal
profile between annual frequencies of 0.10 and .002.
This report is intended for use in estimating actuarial
risk to buildings from coastal floods and in land use
management.

1. INTRODUCTION

1.1 Objective and Scope

The Federal Insurance Administration (FIA), Department of Housing and Urban
Development (HUD), requested the National Oceanic and Atmospheric Administra-
tion (NOAA) to study flood levels from storm tides an the open coast of North
Carolina, north of Cape Lookout (fig. 1). This includes the Atlantic Ocean
coast of Carteret, Hyde, Dare, and Currituck Counties, Outer Banks of North
CarolinA~. The assignment is limited to determining storm tide frequencies
along the Atlantic Ocean coast on a common regional basis. Modifications of
the storm tide levels within Pamlico Sound and in other bays are not included.
These modifications have to be assessed by separate investigations, using the
present study as part of the relevant information.

The tide frequencies are of still water levels that would be measured in a
stilling well or tide gage house designed to exclude wave action. The
destructive effects of waves on the beach front must be taken into account
separately.

Storm tides in the study area are caused both by hurricanes -- which are
storms of tropical origin occurring in the summer and fall months -- and by
a winter type of coastal storm, commonly called a "northeaster." Both types
of storms are included in the study.

1.2 Authorization

The National Flood Insurance Act of 1968, Title KIII, Public Law 90-448,
enacted August 1, 1968, authorizes and directs the Secretary of Housing and
Urban Development to establish and carry out a National Flood Insurance Pro-
gram. The Secretary is authorized to secure the assistance of other Federal
Departments on a reimbursement basis in assessing flood frequencies. Authori-I
zation for this particular study is Project Order No. 2, dated November 13,
1974, under Agreement No. IAA-H-1975 between the Federal Insurance Administra-
tion and NOAA.

1.3 Study Method

The technique used in this tide frequency analysis for the oceanic coast of
North Carolina, north of Cape Lookout, is basically the same as that applied
earlier to the North Carolina coast south of Cape Lookout (Ho and Tracey
1975) and in other studies (e.g., Ho 1974). The procedure is explained in
detail in a separate report (Myers 1975).

First, the behavior of hurricanes along the coast both striking from the sea
and passing on the inland side is assessed from past records. Factors analyz-
ed included depression of the atmospheric pressure at the storm center below
the surrounding value, forward speed and direction of motion of the storm,
and distance from the storm center to the band of maximum winds. All these
factors relate to a storm's potential to produce high tides.

The second step in the tide frequency analysis is to calculate the coastal
tide levels that each of a number of hypothetical but representative hurri-
canes from various combinations of the hurricane parameters would produce.
For this a dynamic calculation method is used that has been demonstrated to
reproduce observed storm tides of past hurricanes within acceptable toler-
ances.

Finally, the computed storm surges are combined with the astronomical tide
variation by using a joint probability method to obtain a frequency distribu-
tion of several classes of storms. The effect of winter coastal storms are
also taken into account. These are discussed in section 5. Summing all
the frequencies yields the total tide at several points on the coast. Fre-
quency profiles along the coast are then constructed by interpolation, taking
into account water-depth variations ("shoaling factor," defined in par. 4.2)
and trend along the coast in climatological parameters.

These three steps are amplified in sections 3, 4, and 6 of the report,
respectively.

2. SUMMARY OF HISTORICAL STORMS

This section summarizes the major hurricanes that have affected the study
area since 1800 and two recent severe winter-type coastal storms. Lesser
storms are omitted.

2.1 Hurricane Tracks

Selected tracks of major hurricanes affecting the study area since 1871 are
shown in figures 2 and 3. These are segregated into hurricanes approaching
from the southeast and southwest quadrants, respectively. Tracks of hurri-
canes bypassing the coast in recent years are also included.

A few hurricanes strike the Atlantic coast north of Cape Hatteras from the
northeast. This type is discussed in par. 2.3. The tracks of tropical storms
and hurricanes that moved from the northeast in the general region of the
study area, are shown in figure 4. The information on hurricane tracks is
taken from the charts of North Atlantic tropical cyclones compiled by Cry
(1965). For 1964 through 1974, similar tracks are published in the Monthly
Weather Review.

2.2 Historical Notes on Hurricanes

Brief notes on the history of hurricanes and damages caused by them are
abstracted from published papers. Wind speeds are included to indicate the
intensity of storms. Wind speeds from Weather Bureau stations from storms be-
fore 1920's have been adjusted by the instrumental corrections to anemometers
developed at that time (Harrison 1963). Since that time, official wind re-
ports include the corrections. Prior to 1940, the highest wind given for a
storm was usually the "maximum velocity,?? an average wind speed for a five-
minute period. In recent years, the highest sustained wind is an average over
a one-minute period. For a complete chronology of tropical cyclones since
1586, the reader is referred to the publication on "North Carolina Hurricanes"
(Hardy and Carney 1962).

September 4, 1815

A major hurricane cut across extreme eastern North Carolina in early Sep-
tember 1815. This hurricane moved inland on the morning of September 4, pass-
ing close to New Bern, N. C., on the Neuse River and recurved northeastward.
At Beaufort, N. C., the tide flowed 4 ft higher than ever known. Every one
of the more than 20 vessels at Ocracoke Inlet, along the Outer Banks of North
Carolina,was driven ashore by the shifting gale. In New Bern, the tide was
one foot higher than in any storm since 1795 and reached an elevation of near-
ly 12 ft above common high-water mark (Ludlum 1963).

June 3-4, 1825

This early season tropical cyclone swept up the Atlantic coast with reports
of major damage all the way from Florida to New York City. Very high winds,
which lasted 30 hrs., were reported by the post surgeon of Fort Johnston at
Cape Fear, N.C. Along the Outer Banks of North Carolina the hurricane lashed
at shipping settlements. Near Ocracoke Inlet, 25 vessels were driven ashore.
A press dispatch from Adam's Creek, N.C., near Cape Lookout reported very
heavy losses with crops destroyed and cattle drowned as the storm tide rose
14 ft above low water. At New Bern, the tide rose 6 ft and considerable
damages near-the water front were reported (Ludlum 1963).

August 25, 1827

This hurricane was traced to its origin in the Windward islands on August
17. It struck the coast between Cape Fear and Cape Hatteras on August 25.
T-udlum (1963) gives some descriptive accounts of the storm tides along this
stretch of the coast: "The towns of New Bern and Washington, both heads of
navig2ation for tidal rivers emptying into Pamlico Sound, suffered severely
from high tides, and such high waters, too, are always caused by wind with an
easterly component. At Washington the tide was 12 to 15 ft above ordinary
tides and houses on Water Street found the river 5 to 6 ft deep in their first
floor during the height of the storm tide. At New Bern all communication for
a while was by canoe. Near Cape Hatteras two New York-to-North Carolina pack-
ets were driven ashore and smashed to pieces by the tremendous breakers. The
new Cape Hatteras Lightship off Ocracoke Inlet broke loose and piled up on the
south side of Ocracoke Island.??

July 12-15, 1842

A very destructive hurricane swept the entire North Carolina coast, appar-
ently most severe in the Ocracoke - Portsmouth, N.C., area. This storm was
reported to have been the most violent experienced at Ocracoke Bar for 80
years (another great hurricane in 1761 changed much of the coast-line of the
Outer Banks and cut through the New Inlet near Wilmington, N.C.). The damage
along the Outer Banks was immense. The entire village on Portsmouth Island
near Ocracoke Inlet, with the exception of one building, was wrecked. A store
at the settlement was blown down and floated away at the height of the storm.
Fourteen vessels went aground on the ocean beach near Ocracoke Inlet and f our-
teen more on the inside beaches. Two unknown vessels were dashed to pieces
in the breakers on Diamond Shoals. A number of dead horses and cattle were
seen drifted down the Sound after the storm was over (Ludlum 1963).

September 7-8, 1846

This hurricane apparently approaching slowly from the south opened up two
new inlets of major commercial importance across the Outer Banks. To the
south of Cape Hatteras, a new Hatteras Inlet between Ocracoke and Hatteras
Islands provided a new entrance into Pamlico Sound. To the north, Oregon In-
let (named after the first ship to pass through) split Bodie Island below Nags
Head for a more direct route to Albemarle Sound ports. The hurricane caught
20 ships at Ocracoke Inlet and drove all but two of them ashore or out to sea.
The small community of Hatteras just south of the Cape had all but six houses
flattened by the storm. At Nags Head the tide rose about 9 ft higher than
common tide (Ludlum. 1963).

October 23, 1878

This hurricane moved northward across Cuba, skirted the east coast of Flor-
ida and moved inland between Wilmington and Morehead City, N.C., on October
23rd. It struck the Outer Banks with full hurricane force, with maximum winds
of 77 mph recorded at Cape Lookout and 63 mph at Portsmouth. The steamer City
of Houston was lost on Frying Pan Shoals; a great many ships were damaged or
lost in the storm all along the Atlantic coast (Hardy and Carney 1962).

August 18, 1879
A severe hurricane moved inland near Wilmington on the 18th and back out to
sea near Norfolk with highest winds reported at Cape Lookout. The anemometer
cups at Cape Lookout were blown away and the wind was afterward estimated to
have reached 127 mph. Anemometers were also destroyed at Hatteras, Fort
Macon, Kitty Hawk, Portsmouth, N.C., and Cape Henry, Va., with speeds estima-
ted at 100 mph or more. A ship report indicated waves forty feet from trough
to crest. This storm was most destructive in the Morehead City - Beaufort,
N.C., area where two hotels were destroyed and 1,090 ft of railroad track torn
up. All the wharves were washed away and the chimneys of most houses were
blown away. On the Outer Banks, the storm caused great damage at Diamond
City, which was near Cape Lookout (Hardy and Carney 1962).
August 17-18, 1899
One of the most severe hurricanes on record for the Hatteras area moved
slowly northward across the Outer Banks during August 17-18. By early morning
of the 17th, the wind was blowing from the northeast at 54 mph at Hatteras;
it had reached 71 mph at 1:00 p.m. with extreme velocities of 90 to 105 mph.
The anemometer then blew away; stronger winds probably occurred. Hatteras
reported a barometer reading of 968.9 mb (28.61 in) at 8:00 p.m. of the same
day (U. S. Weather Bureau 1899). The Weather Bureau observer at Hatteras
reported that "the entire island" was covered with water to a depth of 4 to
10 ft. There were not more than four houses in which the tide did not rise
to a depth of 1 to 4 ft. All fishing piers and equipment were destroyed; all
bridges were swept away; a great proportion of the homes on the island were
damaged. There was much destruction at Diamond City, N.C. 'Flooding of much
of the coastal areas and strong winds and heavy rains inland as far as Raleigh
did great damage to crops (Hardy and Carney 1962).
July 31, 1908
This storm had its inception as a tropical storm off the east coast of Flor-
ida. It then moved to the east-northeast, did a complete "loop" and became a
hurricane as it moved northeastward off the coasts of Georgia and the Caro-
linas. It moved inland near Cape Lookout on July 31 then across Pamlico Sound,
continuing its northeastward movement. Highest reported wind was 46 mph at
Hatteras, but the storm piled up considerable water an the North Carolina
coast south of Hatteras. This combined with torrential downpours (10.73 inches
in 72 hours at New Bern and 9 inches at Kinston) caused much flooding in the
eastern counties. Damage was "immense," but no injuries or fatalities were
recorded. At New Bern, this was "the worst storm in history" (Hardy and
Carney 1962; Corps of Engineers, Wilmington, N.C. 1961; and Sugg, et al.,
1971).
September 16, 1933
This hurricane formed east of the Leeward Islands, moved northwest and then
northward, increasing in intensity and striking the coast a little west of
Hatteras about 8 a.m. on the 16th. Maximum wind speed at Hatteras was esti-
* ~mated at 76 mph because a portion of the anemometer was blown away. Winds
were estimated up to 125 mph in New Bern and Beaufort. Minimum barometric
pressure at Hatteras was 957 mb. Damage was heavy from a short distance south
of New Bern to the Virginia line. It was reported that hardly a building was
left standing in several coastal towns. High winds, waves, and piling up of

water in Pamlico and Albemarle Sounds caused 21 deaths. Wind and water did
great damage at New Bern where water was reported to reach "a height of 3 to
4 ft" (Hardy and Carney 1962). An estimated high tide of 7.0 ft MSL occurred
at Ocracoke, N.C. (Corps of Engineers, Wilmington, N.C. 1963).
September 18, 1936
This hurricane was one of the most severe hurricanes of record at Hatteras.
Maximum winds of 80 mph from the northwest were reported at Hatteras and 90
mph at Manteo, N.C. As the hurricane approached Hatteras it began recurving
northward and the storm center passed close to the station on the coast. The
highway from Currituck, N.C., to Norfolk, Va., was washed out. About 35 ft
of beach was cut away at Nags Head, N.C. Tides were very high at Manteo and
Hatteras (Hardy and Carney 1962). A high tide of 6.0 ft MSL was reported at
Hatteras and an estimated 6.3 ft MSL at Ocracoke, N.C. (Corps of Engineers,
Wilmington, N.C. 1963).
September 14, 1944
The "Great Atlantic Hurricane" of September 1944 caused destruction to 900
miles of the Atlantic coast from Hatteras northward. The center of the hurri-
cane passed a short distance to the east of Hatteras with estimated maximum
winds of 110 mph. At Hatteras a lowest barometric pressure of 947.2 mb (27.97
in) was recorded, the lowest pressure reading on record at the station. Cape
Henry, Va.,reported maximum winds of 134 mph with gusts estimated at 150 mph
(Sumner 1944). There was heavy damage in Elizabeth City, N.C., and the Nags
Head area. The storm was very severe and caused considerable property damage
on Ocracoke Island. Local residents reported the highest tide on record at
Ocracoke Village, 7.5 ft MSL (Corps of Engineers, Wilmington, N.C. 1963).
A highest tide of 7.0 ft above mean low water (6.0 ft MSL) was reported by
the U. S. Weather Bureau at Hatteras, N.C. (Sumner 1944).
August 12, 1955 - Connie
Hurricane Connie entered the North Carolina coast close to Cape Lookout
about 8:30 a.m. on August 12. The prolonged pounding of high waves against
the coast caused tremendous beach erosion estimated to have been worse than
that caused by Hazel in 1954. Tides of about 4.0 ft MSL were reported at
Ocracoke and Hatteras, N.C. (Corps of Engineers, Wilmington, N.C. 1963), while
water in the sounds and near the mouths of rivers were 5 to 7 ft MSL (Harris
1963). At Fort Macon, N.C., winds of 75 mph with peak gusts of 100 mph and
lowest pressure of 962 mb were reported. The storm also brought torrential
rains with the maximum, ranging around 12 inches within 48 hr falling near
Morehead City. Total damage throughout the state was estimated at $50
million (Hardy and Carney 1962).
September 19, 1955 - Tone
Hurricane Tone, moving from the south, crossed the North Carolina coast near
Salter Path, about 10 mi west of Morehead City, about 5 a.m. on September 19.
It then slowly curved to the northeast, passing out to sea near the Virginia
stateline early on September 20. When Tone entered North Carolina, highest
winds were a little over 100 mph in gusts. The highest recorded wind speed
was 75 mph gusting to 107 mph at Cherry Point. Minimum barometric pressure
over North Carolina was 960 mb. Heavy rains accompanied Tone. At the same
time, prolonged easterly winds drove tide water onto the beaches and into the

sounds and their estuaries to a height of 3 to 10 ft above normal. The result
was inundation of the greatest area of eastern North Carolina ever known to
have been flooded. At New Bern, the depth of water was the greatest of
record, being about 10.5 ft above mean low water, with 40 city blocks flooded.
Several hundred homes were washed away and thousands were flooded by water
with depths ranging up to 4 ft (Hardy and Carney 1962). A high tide of 7.2
ft MSL was reported at Atlantic Beach, N.C. (Harris 1963). High tides of 5.7
anid 3.8 ft MSL at Ocracoke and Hatteras, N.C., respectively, were reported by
the Corps of Engineers (Corps of Engineers, Wilmington, N.C. 1963).

September 27, 1958 - Helene

Hurricane Helene was one of the most powerful storms of recent history and,
fortunately for North Carolina, the storm center moved up the coast staying
well out at sea, on September 26-27. Even so, the highest winds of record
were recorded at Wilmington, with peak gusts of 135 mph and fastest mile 85
mph. The lowest reported central pressure was 932 mb at a point south-south-
east of Cape Fear early on the morning of the 27th (aircraft reconnaissance).
There was some beach erosion due to seas and tides but this was minimized by
the passage of the storm at time of low astronomical tide. The highest tides
on ocean beaches were estimated at 3 to 5 ft above normal. Tides were higher
on the southern edge of Pamlico Sound, where a sudden rise following the wind
shift as the center passed brought the tides to 7 or 8 ft above normal (Hardy
and Carney 1962). At Ocracoke, N.C., local residents stated it was the most
severe storm since 1944. Water covered most of the island and swept into
about 25 homes in the village. The northern end of the island was breached
at six different locations during the storm. Over 2.5 mi of state highway
steel-mat pavement was washed out, and approximately 12 mi of National Park
Service sand fence located near the ocean shore was destroyed. High-water mark~
of 5.5 ft MSL was reported by local residents in the village of Ocracoke. A
high tide of 5.1 ft MSL was recorded at Hatteras, N.C. (Corps of Engineers,
Wilmington, N.C. 1963).

September 11, 1960 - Donna

Hurricane Donna passed inland over the North Carolina coast between
Wilmington and Morehead City on September 11. The center of the storm passed
a few miles east of Wrightsville Beach, although Wilmington and Wrightsville
Beach were in the "eye" for about an hour. Lowest barometric pressure at
Wilmington was 962 mb. Tides of 6 to 8 ft above normal combined with high
winds caused severe damage at many points. Maximum winds were of hurricane
force with Wilmington reporting a peak gust of 97 mph. The storm center
moved northward along a path slightly east of a line from Wilmington to
Norfolk during the night of the 11th. Coastal communities suffered heavy
structural damage from Wilmington to Nags Head, with considerable beach
erosion. Wind gusts were in excess of 100 mph and tides were 4 to 8 ft above
normal (Hardy and Carney 1962). High tides of 10.6 ft MSL were reported at
Atlantic Beach, N.C., and 4 to 6 ft MSL in the sounds and near the mouths
of rivers. Tides of 3 to 4 ft MSL were reported on the Outer Banks near
Hatteras (Harris 1963).

September 30 - October 1, 1971 - Ginger

Ginger will be noted chiefly for its longevity and circuitous track. This
storm was tracked for 31 days during 20 of which it was a hurricane. On
September 27, Ginger moved northwestward and set a steady course toward the
North Carolina coast. Its center crossed the coast near Morehead City on the
evening of September 30 with maximum sustained winds of 75 mph and minimum
central pressure of 993 mb. Total damage caused by the storm in North Carolina
is estimated at $10 million. Tides on Pamlico Sound were 4-7 ft above normal.
At Washington, Aurora, New Bern, and Cherry Point, N.C., tides were 6 ft above
normal. On the ocean beaches at Hatteras,, tides were 2-3 ft above normal
(Simpson and Hope 1972).

2.3 Winter Coastal Storms

The study area is exposed to winter coastal storms, especially from Cape
Hatteras northward. Strong winds from the northeast quadrant are experienced
over long reaches of coast, hence the familiar name "northeaster." These
winds are part of a counter-clockwise cyclonic atmospheric circulation about
a center of atmospheric pressure, at sea. The area or proximity of warm gulf
stream water off the North Carolina Banks is a favored region for the develop-
ment and intensification of such storms. They are most common and most
severe during winter and spring months. Many of these storms move rapidly to
the north and northeast but under favorable conditions of the general atmos-
pheric circulation they can stall and intensify with little forward motion
for a couple of days.

These storms raise the tide (still water level that would be observed in a
tide gage house if one existed) but not as high as in the more severe hurri-
canes. The aspect of a northeaster most damaging to the exposed beaches of
the Outer Banks is the persistence of pounding waves, developed by wind blow-
ing toward the coast over long stretches of ocean. The extent of these dam-
ages may be illustrated by citing reports on recent coastal storms of March
7, 1962 and February 17, 1973.

March 7, 1962

The second worst northeaster of the present generation was on March 7, 1962,
locally known as the "Ash Wednesday storm." The effects were similar to the
1973 storm described below, though there was less destruction to buildings
(in the study area) because the several communities were less built up at the
time. There was considerable damage to private property along the shore in
the Nags Head, Kill Devil Hills area (Corps of Engineers 1966). The storm
opened a new inlet just north of Buxton, N.C. This was filled and closed at
considerable expense by dredges several months later (Corps of Engineers
1963). On some parts of the Delmarva Peninsula, to the north of the study
area, this was the most damaging of all coastal storms, including hurricanes,
of record to date.

February 17, 1973

The most damaging winter coastal storm of the present generation was on
February 17, 1973. The following newspaper quote succintly describes the
storm:

"North Carolina's land mass is smaller by several hundred acres this week
as a result of a severe coastal storm which gnawed away at beaches from
Corolla to Cape Fear.

"During the two days following a freak storm that frosted the state's
eastern lowlands with up to 15 inches of snow, savage 56-knot winds and power-
ful 40-foot waves hauled tons of sand from the shore.

"The wind-whipped sea toppled large buildings in resort communities on the
Outer Banks, nearly bisected at least two offshore islands, ripped up high-
ways, filled roadbeds with sand and carved away large chunks of sandy beach.

"The worst of the storm's fury was directed at Buxton, Kitty Hawk and Nags
Head, but severe erosion and some property damage occurred at the more
southerly resort communities of Wrightsville Beach, Carolina Beach and Top-
sail Beach." (Raleigh News and Observer February 19, 1973).

A typical description of local damage, from the same source, is "At Kitty
Hawk four beach cottages were washed away, nine were toppled by the waves
but left partially standing, 12 others received structural damage from the
pounding surf and dozens of others are left standing so near the encroaching
sea that another storm would undermine them."

3. CLIMATOLOGY OF HURRICANE CHARACTERISTICS

This section describes important characteristics of hurricane parameters
that are needed for calculating tide levels on the coast. Basic parameters
of hurricanes affecting the U. S. coast, including central pressure, radius
of maximum winds,, speed of forward motion, and direction of forward motion,
all-factors affecting storm-tide producing capability, were published in 1959
(Graham and Nunn 1959). This compendium of hurricane characteritics has been
updated through 1973 and adapted to the needs of the Flood Insurance Program,
including specification of probability distributions of the individual para-
meters. These data are published in a separate report (Ho, Schwerdt, and
Goodyear 1975) hereafter referred to as the Climatology Report, and are the
primary data source on hurricanes for the present study. These data are used
directly for the portion of the study area south of Cape Hatteras, with
certain refinements in hurricane track count described later. North of Cape
Hatteras the methods of the Climatology Report were extended by certain addi-
tional analyses, described in par. 3.2.

3.1 The General Method - South of Cape Hatteras

For the purpose of determing parameter probabilities, in the Climatology
Report hurricanes and tropical storms are grouped into three classes, those

which landfall on the coast, those which bypass alongshore with the center
remaining at sea, and those which exit the coast (after an earlier entry of
the coast elsewhere). In this study, south of Cape Hatteras, the first two
classes are used. Exiting storms are not significant there.

3.1.1 Probability Distribution of Hurricane Intensity

Storm surge magnitude varies approximately with the strength of the wind
that is putting stress on the water surface, other factors being constant.
An index of this wind stress in hurricanes is the intensity of the storm as
measured by the depression of the storm' s central pressure below representa-
tive peripheral pressure (D). The assessment of probability distributions of
this parameter for landfalling and alongshore storms is described in the
Climatology Report.

3.1.2 Probability Distribution of Radius of Maximum Winds

in all hurricanes, proceeding from the storm center outward, winds increase
from low values at the center of the eye to their most intense velocity just
beyond the edge of the eye, then decrease gradually. The average distance
from the storm center to the circle of maximum wind speed is called the radius
of maximum winds (R) and is adopted as a convenient single number to be used
as an index of the size or lateral extent of the hurricane, a factor which
affects the surge profile along the coast. R values are taken from the Clima-
tology Report and applied to both classes of storms.

3.1.3 Probability Distributions of Speed and Direction of Forward Motion

The speed (T) and direction (9) of forward motion of hurricanes also affect
surge height. The height of the surge an the coast increases with increasing
storm speed up to a forward speed higher than that of any hurricane in the
study area. Thus, the occasional fast-moving storms, especially if they are
large and moving directly toward the coast, pose the greatest hazard. Prob-
ability distributions of forward speed for the landfalling and alongshore
storms are given in the Climatology Report previously cited. The probability
distribution of direction of forward motion for landfalling hurricanes and
tropical storms is also discussed in the Climatology Report.

3.1.4 Frequency of Hurricane Tracks

The overall frequency of hurricane occurrences is basic to calculating the
resulting tide frequencies. The frequency with which hurricane and tropical
storms have entered or exited the coast and have moved approximately parallel
to the coast at sea ("alongshore") based on 103 years of data smoothed along
the coast is given in the Climatology Report. These counts are used in this
study, except that further analysis was made to the original data to take in-
to account in greater detail the influences of seaward extensions of the land
of the turning of the coastal orientation near Cape Hatteras. Additional
counts of alongshore storms immediately opposite Cape Hatteras, Ocracoke and
Cape Lookout were made from Cry's (1965) annual track charts to secure more
detail in the vicinity of these features. The frequency distribution of land-
falling storms in the study area is discussed in par. 3.2.4.

3.2 Special Considerations North of Cape Hatteras

Landf ailing hurricanes and tropical storms in the study area north of Cape
Hatteras are further classified to deal with the joint probability questions
referred to in Chapter 6 of the Climatology Report, and to take into account
the effects on the coast of storms passing inland to the coast. The latter
are not covered in detail in the Climatology Report.

3.2.1 Conditional Probability Question

It will be explained in section 6 that assessing the probability of a
certain combination of hurricane parameters involves multiplying together
the probabilities of each of the parameters, on a scale of 0 to 1.0, provided
the distributions of the several parameters are independent in the statistical
sense (Chapter 6 of the Climatology Report). If the distributions for any
two parameters are not essentially independent, then this must be recognized
by either using conditional probabilities or by segregating the possible hur-
ricane events into subsamples. An example of the subsample segregation in the
Climatology Report is the division of hurricanes into t"landfalling"t and
"alongshore." Separate forward speed distributions are then determined for
each subsample.

North of Cape Hatteras we must deal with the correlation between hurricane
intensity and the direction of storm motion relative to the coast. This is
alluded to in paragraph 6.3 of the Climatology Report as an example of con-
ditional probability questions that may be expected in various regions,
but specific data are not developed. Hurricanes landfalling from the south-
east quadrant cover the full range of intensities from very severe to weak.
From time to time in this region a hurricane meanders and strikes the coast
from the northeast quadrant. These storms are either of a weaker category
initially or, if originally intense, have been weakened by transit over cold
water and other effects. The speed distribution for these storms is also
different from those moving from the southeast quadrant. These storms all
move at less than 15 kt. A separation for all parameters, including track
frequency, was made between landfalling storms from the SE and NE quadrants
north of Cape Hatteras. This is described in the next several paragraphs.

3.2.2 Parameters for Landfalling Hurricanes from Northeast Quadrant

To provide the hurricane parameter data needed for this part of the study,
a special analysis was made of hurricanes and tropical storms landfalling
from the northeast quadrant. For this the original data from the Climatology
Report were used plus that for other hurricanes and tropical storms farther
at sea to expand the sample. Hurricanes and tropical storms moving from a
northeasterly direction within an area west of 70'W and north of 30'N were
examined. The speeds of forward motion were measured from Cr's storm tracks,
and a probability distribution formed. A sample of eight central pressure
values from the most recent storms were used to derive a probability curve
for this parameter. There is insufficient information to determine a separate
frequency distribution of radius of maximum winds for this class of storms,

but there is no reason to think that it would differ much from the distribu-
tion for all hurricanes at this latitude. The R distribution as given in the
Climatology Report is therefore adopted.

3.2.3 Parameters for Landfalling Hurricanes from Southeast Quadrant

To obtain the probability distribution of central pressure for the storms
landf ailing from the southeast quadrant, the probabilities for northeast quad-
rant hurricanes and tropical storms were subtracted from the overall proba-
bility for all landfalling hurricanes in the Climatology Report, weighted by
their percentage of occurrence. The probability distribution thus obtained
by subtraction was also checked against a direct sample of storm data. The
resultant distribution for the SE storms differs only slightly from that of
all landf alling storms. This is illustrated in figure 5 (curves a and d).
Speed of forward motion probabilities were evaluated in a similar manner.
Figure 6 shows the resulting comparative probability distributions of forward
speed. The probability distribution of radius of maximum winds from the Cli-
matology Report is adopted for the southeast quadrant hurricanes.

3.2.4 Landfalling Track Frequency

The frequency of landfalling storms from the northeast and southeast quad-
rants are not given in the Climatology Report separately. The segregation of
landfalling hurricanes and tropical storms into quadrant of approach calls for
a discontinuity at Cape Hatteras in the overall landf ailing frequency. The
frequency of storms landfalling from the sector 910 - 1600 is approximately
the same immediately north and south of the Cape. But landfalls from the
other possible directions -- 1610 - 2400 south of the Cape and from the north-
east quadrant north of the Cape--are not of equal frequency. Thus there is a
difference in the total landfall frequency. The overall storm landfalling
frequency, a factor in storm tide frequency computation, smoothed along the
coast by weighted moving averages in the Climatology Report, is adjusted to
define this discontinuity. A track count of storms from the northeast quad-
rant and the 910 - 1600 sector crossing overlapping two-degree latitude and
longitude squares was examined separately. The sum of these frequencies was
checked against the frequencies of all landfalling storms. Figure 7 depicts
the resulting frequencies with which hurricanes and tropical storms entered
the coast from different sectors both north and south of the Cape. The plot-
ted points (circled dots) show the frequencies of direct track counts at 50-
n.mi. intervals as given in the Climatology Report. SE-quadrant parameters
(par. 3.2.3) were applied to the 910 - 1600 sector storms.

3.2.5 Hurricanes Passing Inland

Because of the North Carolina coastal configurations and the propensity, of
hurricanes to recurve and move northward in this region, many hurricanes land-
fall on the coast south of Cape Hatteras and then move on a course approxi-
mately parallel to the coast north of the Cape. The tracks of some of these
are shown in figures 2 and 3. Hurricanes and tropical storms which bypassed
within 100 n.mi. inland of the coast were assessed by separate counts of
storm tracks crossing 350N and the N.C.-Va. border on the same 103 years of
track charts. The cumulative track counts along each line were plotted and a
smooth curve fitted by eye to the data on each of these frequency plots.

The accumulated frequencies for storms bypassing inland for Wright Monument,
N.C., obtained by interpolation is shown in figure 8 (together with along-
shore frequency at sea). A similar analysis was made for Salvo, N.C.

The central pressure for inland storms was estimated by applying an average
empirically-derived filling rate after the storm center landfalls.

Malkin (1959) studied the filling rates of 11 hurricanes moving inland in
the southern and eastern United States, with the Florida Peninsula excluded,
and constructed an empirical average curve of central pressure vs. time over
land (his figure 1). An exponential decay of the form

D. Dce (1)
1 c
where D. is the pressure deficit (surrounding pressure minus central pressure)
of the storm inland, D the central pressure deficit at the coast, t time in-
c
land, and a the fractional change in D in one time unit, fits these data rath-
er well. Expanding in a series and dropping higher order terms

D. = D (et = D (1 + a)t. (2)

Substitute LL/T = t, where LL is distance inland and T average speed,

D. = D (l + a)LL/T (3)
1 c
A value of a = -.04 per hour provides a good fit of eq. (3) to Malkin's curve
and was adopted.

Representative D.'s for storms passing inland of the coastal stations north
:1
of Cape Hatteras were obtained by making the following approximations and
substituting in (3). (a) Lay out tracks paralleling the coast; for LL use
distance from coastal entry point to latitude of station as shown in figure 9.
This is done separately for various distances, L, inland fromi the coast.
(b) For T use the median speed for landfalling storms at the latitude of the
station, from the Climatology Report. (c) For D. use pressure depression
values from the Climatology Report for landfalling storms at the latitude of
the station.

Figure 5 shows the three derived probability distribution curves of central
pressure of hurricanes and tropical storms for Wright Monument, N.C. Curve d
from the Climatology Report is shown for comparison. (Combining a and b,
weighted by frequency, yields d.) It will be noted that the diagram shows
only the fraction of all storms with intensities below certain levels and
makes no reference to frequency in terms of events per year.

The speed of forward motion for inland storms was measured from storm tracks
crossing the N.C.-Va. border within 100 n.mi. of the coast, and a probability
distribution of this parameter was formed. For R there are insufficient data
for a direct analysis. R may expand faster for storms over land than at sea
but this has not been verified. In any event, for this study the contribution
of inland storms to coastal tide frequencies is small and does not demand

attempts to refine R probabilities. The distribution of R in the Climatology
Report is used for inland storms.

3.2.6 Exiting Hurricanes

The frequency with which hurricanes and tropical storms exited the coast is
given in the Climatology Report. Other parameters are adapted from those for
the landfalling and inland categories. Approximating the parameters and
grouping into fewer class intervals suffices for these computations as exiting
storms produce lower storm tides, and, as it turns out, make an almost negli-
gible contribution to the overall storm tide frequencies (e.g., curve d, fig-
ure 13).

"I.2.7 Alongshore Hurricanes

Alongshore hurricane parameters are given in the Climatology Report. In-
tense hurricanes belong in this group. The probability distribution of cen-
tral pressure used in the computations is the same as that of the SE land-
falling storms. This class of hurricanes has a higher percentage of fast-
moving storms than those of landfalling and exiting categories.

4. HURRICANE SURGE

4.1 Surge Model

The National Weather Service has developed a two-dimensional numerical dyna-
mic surge model for calculating the water levels induced by hurricanes on the
continental shelf (Jelesnianski 1967, 1972, and 1974). The objective of this
work was to develop a tool to forecast open coast surges when hurricanes were
approaching. The model has become the backbone of NOAA's tide-frequency
studies for the flood insurance program. The development of the model is de-
scribed by Jelesnianski in the 1967 paper and operational applicatons (desig-
nated as SPLASH I and II) in the others. Both limitations and verification
of the model are described in the references. Replication of surge profiles
produced by past hurricanes agree well with observed storm tides and high-
water marks adjusted to exclude astronomical tide. The model computes the
surge, the difference between the local storm-induced level and the normal
water levels for the area. Thus, the computed storm. surge must be added to
the predicted astronomical tide.

Inputs to a computer surge calculation are hurricane central pressure defi-
cit, radius of maximum winds, storm direction of motion and forward speed, and
ocean depth at a series of grid points. The hurricane climatology just de-
scribed is oriented toward providing these parameters. The computer program
generates the needed moving sea-level pressure field and moving wind field
from the basic parameters by predetermined relations which are part of the
model.

The SPLASH I version is limited to storms moving forward at constant velo-
city and intensity toward a specified landfall point while SPLASH II has
been expanded to accommodate storms with generalized motions of not too great

complexity. Also, storm strength and size are allowed to vary in a contin-
uous monotonic manner with time. In the present study, SPLASH II was used
to compute surges generated by alongshore and inland hurricanes. Since SPLASH
I and II give the same result for landfalling hurricanes with a constant
velocity and intensity, landfalling hurricane surges were computed by the
SPLASH I program as it is slightly simpler to use.

4.2 Shoaling Factor

The capacity of a hurricane of given characteristics to produce a coastal
surge depends on the profile of water depth. The shallower the coastal water
the higher the surge. This variation along the East coast is depicted in
a report by Barrientos and Chen (1975, figure 3b) as the ratio of the surge
that would be produced locally at each coastal point by a standardized hurri-
cane to the surge from the same hurricane moving over a continental shelf
of average or standard slope. This ratio is called the shoaling factor,
and is generated by computing surges by the model that has been described at
the various coastal points and over the "standard basin" and taking ratios
of the peak surges. The North Carolina portion of this diagram is reproduced
in figure 10. The shoaling factor is implicit in calculations of hurricane
surge by the model at selected coastal points, since the sloping depths of
the continental shelf are introduced by input data to the calculation. The
shoaling factor is specified at 4-mi intervals in the SPLASH program (rela-
tive to the value at the center of the "basin") and is a primary guide to
interpolating between coastal computation points. The shoaling factor curve
of figure 9 reveals that a minimum factor is reached near Ocracoke Inlet and
comparatively higher factors appear to the north of Salvo, N.C. The top curve
(for R=26 n.mi.) is applied to hurricanes with R>19.5 n.mi. in this study and
the lower curve to smaller hurricanes.

5. WINTER COASTAL STORM TIDES

5.1 Evaluation of Northeaster Tide Levels in Study Area

Some of the 10-yr return period magnitude (10% probability per year) storm
tides in this study area are caused by northeasters, as will be shown later
in the report. At the 100-yr return period magnitude (1% chance of being
exceeded in any year) the activating storms are hurricanes and the additional
contribution by northeasters is negligible by comparison. For purposes of
this study, the northeaster contribution to storm-tide frequencies may be ev-
aluated with sufficient precision by direct analysis of long period tide
gage records and interpolation for frequencies at selected points along the
coast.

5.2 Analysis of Tide Gage Records

There are no permanent long period tide gage records within the study area
itself. We go to the nearest National Ocean Survey Reference Stations to the
north and south, with more than 30 yr of record. These are at Hampton Roads,
Va., (45 yr) and Wilmington, N.C. (37 yr). To focus on northeasters and ex-
clude hurricanes, at each station the maximum tide was abstracted for each
October through May season, with an additional check of weather maps to ex-

elude any hurricanes which might occur in early October. These data series
were treated as "annual series" and frequency curves fitted. The plot for
Hampton Roads is shown in figure 11. A frequency curve was fitted by eye us-
ing two curves from standard statistical distributions as guides. The 100-yr
northeaster tide level from the eye-titted curve is 7.0 ft MSL. At Wilming-
ton, the annual maxima conform to the two standard statistical distributions
and either yields a 100-yr northeaster tide level of 5.0 ft MSL. The maximum
ot record at Wilmington is 4.7 ft MSL in 1972. All of the data were adjusted
for sea-level trends to 1970 conditions by using factors from Hicks and Crosby
(1974).

Estimated northeaster tide frequency curves for selected points within the
study area were constructed by assuming that Wilmington values are lower than
on the open coast, then interpolating approximately linearly along the coast.
Reported highest storm tides are an additional guide. These include 7.6 ft
MSL at Norfolk, Va., 6.5 ft at Cape Hatteras, 5.6 ft at Wrightsville Beach,
N.C., (Cooperman and Rosendel 1962), and 8 ft MSL at Nags Head (Corps of
Engineers 1966). A subjective evaluation was required of relative effects on
the eastward-facing coast north ot Cape Hatteras, and the southeastward-facing
coast south of the Cape, from sustained northeast or eastnortheast winds.
Several investigations (e.g., Pore 1964) indicate that northeaster tide levels
are at least as well correlated with the alongshore component of the wind as
with the onshore component. This justifies an approximately linear interpola-
tion in spite of the change in direction of the coast. The estimated Wright
Monument and Ocracoke Inlet curves are shown in tigures 13 and 14, respec-
tively.

6. TIDE FREQUENCY ANALYSIS BY JOINT PROBABILITY METHOD

6.1 The Joint Probability Method

The first step in the joint probability method is to define the probability
distributions of the several hurricane parameters. This has been described
in Section 3. Each distribution is divided into class intervals, and the mid-
point values are read out for each class interval. The parameters derived in
this way and used in the subsequent computations are listed in tables 1 to 5
for the coast at the N.C.-Va., border, Wright Monument, Salvo, Ocracoke Inlet,
and Cape Lookout, N.C., respectively. As indicated in Section 3, many of
these parameter values are taken directly from the Climatology Report; others
were derived in the course of this study. Each combination of D, R, T, and 9
represents a climatologically possible landfalling hurricane or tropical
storm. For example, the parameters for landfalling hurricanes and tropical
storms in table 4 (8D's, 2 or 3 R's 6 T's, and 3 9's) define 396 different
storms which in the aggregate represent the climatological possibilities in
the vicinity. The probability (fraction of all hurricanes) of each of these
is obtained by multiplying the respective parameter probabilities in the
table. The sum of the probabilities ot the 396 hurricanes, of course, equals
1.0. The parameters in the table are considered independent in the statisti-
cal sense except that the three R's are not the same for all pressure depres-
sion categories, smaller values being used with the more intense pressure
depressions in line with the discussion in the Climatology Report and as shown
in the table.

As the second step, calculations are made with the SPLASH computer program
of the coastal surge profile for each landfalling hurricane. Many of the
surge profiles are obtained by adjustment of other profiles rather than by
* ~complete surge calculations. Each storm is allowed to strike the coast not
only at the most critical point but at 8-mi intervals.(being a multiple of the
SPLASH grid) on both sides of a location under study, and the storm surge pro-
* ~files shifted along the coast accordingly. The calculations involved in this
are explained in one of the reports cited (Myers 1975).

As the third step, all profiles in all shifted p ositions are randomly added
to the astronomical tide. The time phasing of surge and astronomical tide is
handled in a probabilistic manner as described in the report previously cited
(Myers 1975). The range of astronomical tide oscillation is represented by
four separate classes. The requisite analysis of a 19-yr cycle of the astro-
nomical tide to obtain these ranges is further discussed in the next section.
Since each surge profile has a prescribed frequency, as have the astronomical
tides,, all the profiles may be combined into a single tide frequency curve for
a fixed coastal point.

As a fourth step, storm tides are similarly computed for the alongshore
storms from the data listed in tables I to 5, and for the exiting and inland
storms for locations north of Cape Hatteras (tables 1-3). To the south of
Cape Hatteras, the occurrence of exiting storms is rather infrequent and
storms bypassing inland are included in the landfalling category. For the
fifth step, winter coastal storm-tide frequencies are also evaluated from tide
gage records (sec.5). Finally, summing all the possibilities yields the total
tide frequency.

Legend tor Tables I to 5

= Orientation of coast, measured clockwise from north (deg).
C
P = Central pressure (mb).
0
D = Central pressure deficit (mb).

P. = Proportion of total storms with indicated D value.

R = Distance from center of storm to principal belt of maximum winds
(n.mi.)

P = Proportion of storms with indicated R value.
r
T = Forward speed of storm (kt).

Pt = Proportion of storms with indicated T value.
Pt
F = Frequency of storm tracks crossing coast (storm tracks per
n.mi. of coast per year).

0L = Direction of entry or exit, measured clockwise from the coast (deg).
0L

PQ = Proportion of storms with indicated 9L value.

L = Distance of storm track from coast inland or seaward (n.mi.).

F = Frequency of storm tracks crossing line normal to coast (storm tracks
per n.mi. per year).

LL = Effective distance of storm over land (n.mi.).

Table l.--Hurricane and tropical storm parameters
N.C;-Va. border. 0 =346�
C

P D P R P T P F L p
0 1i r L P

934.8 78.4 .02 * * 7.5 .2 .00033 140 1.0
942.9 70.3 .03 * * 11.3 .2
�9 949.2 64.0 .05 23.8 .33 15.2 .2
.- 955.8 57.4 .10 36.9 .33 20.0 .2
4-4 964.7 48.5 .20 42.8 .33 25.2 .1
975.2 38.0 .20 31.0 .1
;j 985.0 28.2 .20
rw 996.7 16.5 .20

962.0 51.2 .02 23.8 .33 6.1 .2 .00021 93 1.0
0 964.8 48.4 .03 36.9 .33 6.8 .2
968.8 44.4 .05 42.8 .33 7.5 .2
975.2 38.0 .10 8.7 .2
,x 984.3 28.9 .20 10.4 .1
c 991.7 21.56 .20 12.9 .1
996.0 17.2 .20
z 999.6 13.6 .20

959.7 53.5 .1 23.8 .33 9.4 .3 .00240 233 .5
967.9 45.3 .1 36.9 .33 15.2 .4 266 .5
975.4 37.8 .2 42.8 .33 25.2 .3
~ 985.1 28.1 .2
x 992.8 20.4 .2
1000.2 13.0 .2

L Fa LL R Pr T Pt Remarks

5 .0008 * * 12.5 .2 P0, D, and Pi are
w 15 .0008 23.8 .33 17.2 .2 the same as those
~o 25 .0009 36.9 .33 19.9 .2 for SE landfalling
35 .0011 42.8 .33 24.4 .2 storms.
50 .0012 29.9 .1
-1 70 .0012 35.6 .1

5- .0015 50 23.8 .33 11.2 .3 D's are adjusted
1 15 .0016 60 36.9 .33 17.7 .4 from landfalling
0 25 .0017 65 42.8 .33 25.1 .3 storms by eq (3)
35 .0017 70 using LL and T
'- 50 .0013 90 values.

SE landfalling and alongshore storms with PO<945 mb:
R = 23.8 and 36.9 n.mi. are each assigned a probability of 0.5.

Table 2.--Hurricane and tropical storm parameters
Wright Monument, N.C. 0 =3370

Po D Pi R Pr T Pt F L
933.9 79.3 .02 * 7.2 .2 .00037 145 1.0
= 941.2 72.0 .03 * * 10.8 .2
947.6 65.6 .05 23.3 .33 14.7 .2
955.2 58.0 .10 36.5 .33 19.3 .2
-e 964.3 48.9 .20 42.7 .33 24.3 .1
974.9 38.3 .20 30.2 .1
985.3 27.9 .20
c; 996.6 16.6 .20

962.0 51.2 .02 23.3 .33 6.1 .2 .00037 98 1.0
964.8 48.4 .03 36.5 .33 6.8 .2
-1 968.8 44.4 .05 42.7 .33 7.5 .2
c 975.2 38.0 .10 8.7 .2
11: 984.3 28.9 .20 10.4 .1
c 991.7 21.5 .20 12.9 .1
996.0 17.2 .20
z 999.6 13.6 .20

954.4 58.8 .1 23.3 .33 9.0 .3 .00273 238 .5
963.9 49.3 .1 36.5 .33 14.7 .4 281 .5
0 972.1 41.1 .2 42.7 .33 24.3 .3
~H 982.7 30.5 .2
x 991.5 21.7 .2
999.1 14.1 .2

L Fa LL R Pr T Pt Remarks

5 .0011 * * 12.0 .2 Po, D, and Pi are the
i- 15 .0012 23.3 .33 16.3 .2 same as those for SE
25 .0012 36.5 .33 19.2 .2 landfalling storms.
oo 35 .0013 42.7 .33 23.3 .2
o 50 .0013 28.7 .1
< 70 .0014 34.4 .1

5 .0014 20 23.3 .33 11.2 .3 D's are adjusted from
15 .0014 30 36.5 .33 17.7 .4 landfalling storms by
= 25 .0014 35 42.7 .33 25.1 .3 eq (3) using LL and
35 .0014 40 T values.
50 .0013 60

SE landfalling and alongshore storms with PO<945 mb:

R = 23.3 and 36.5 n.mi. are each assigned a probability of 0.5.

Table 3.--Hurricane and tropical storm parameters
Salvo, N.C. 9c=3600

PO D P.i T Pt F PL Pe

932.7 80.5 .02 * * 6.8 .2 .00051 122 1.0
939.7 73.5 .03 * * 10.1 .2
-i 946.4 66.8 .05 22.8 .33 13.8 .2
954.6 58.6 .10 36.0 .33 18.4 .2
964.1 49.1 .20 42.5 .33 23.2 .1
974.7 38.5 .20 29.3 .1
985.4 27.8 .20
mci 996.6 16.6 .20

962.0 51.2 .02 22.8 .33 6.1 .2 .00051 75 1.0
,r 964.8 48.4 .03 36.0 .33 6.8 .2
-4 968.8 44.4 .05 42.5 .33 7.5 .2
c 975.2 38.0 .10 8.7 .2
: 984.3 28.9 .20 10.4 .1
991.7 21.5 .20 12.9 .1
996.0 17.2 .20
Z 999.6 13.6 .20

947.2 66.0 .1 22.8 .33 8.4 .3 .00075 215 .5
958.1 55.1 .1 36.0 .33 13.8 .4 248 .5
967.6 45.6 .2 42.5 .33 23.2 .3
. 979.3 33.9 .2
989.3 23.9 .2
w 997.6 15.6 .2

L Fa LL R Pr T Pt Remarks

5 .0018 * * 11.4 .2 Po D, and Pi are the
� 15 .0019 22.8 .33 15.5 .2 same as those for
o 25 .0020 36.0 .33 18.2 .2 SE landfalling storms
35 .0021 42.5 .33 22.3 .2
r 50 .0022 27.7 .1
0
� 70 .0022 33.2 .1

<20 # 22.8 .33 11.2 .3 D's are adjusted frcm
-d 25 .0012 5 36.0 .33 17.7 .4 landfalling storms by
35 .0012 10 42.5 .33 25.1 .3 eq (3) using LL and
50 .0012 20 T values.

* SE landfalling and alongshore storms with PO<945 mb:
R = 22.8 and 36.0 n.mi. are each assigned a probability of 0.5.
# included with landfalling storms

Table 4.--Hurricane and tropical storm parameters
Ocracoke, N.C. oc=55�

P D Pi R Pr T Pt F OL P
930.0 83.2 .02 * * 6.5 .2 .00159 93 .33
936.7 76.5 .03 * * 9.4 .2 124 .33
,H 944.2 69.0 .05 21.0 .33 12.7 .2 147 .33
953.2 60.0 .10 34.5 .33 16.8 .2
963.8 49.4 .20 42.0 .33 21.5 .1
976.2 37.0 .20 27.3 .1
987.6 25.6 .20
996.3 16.9 .20

L Fa LL R Pr T Pt Remarks
t
5 .0030 * * 11.1 .2 PO, D, and Pi are
15 .0045 21.0 .33 14.7 .2 the same as those for
o 25 .0056 34.5 .33 17.3 .2 landfalling storms.
35 .0059 42.0 .33 21.3 .2
c 50 .0055 26.5 .1
< 70 .0051 31.8 .1

* Landfalling and alongshore storms with Po<940 mb:
R = 21.0 and 34.5 n.mi. are each assigned a probability of 0.5.

Table 5.--Hurricane and tropical storm parameters
Cape Lookout, N.C. ec=90�

PO D Pi R Pr T Pt F eL Pe

928.5 84.7 .02 * * 5.3 .1 .00155 044 .33
, 935.6 77.6 .03 * * 8.0 .2 086 .33
943.3 69.9 .05 20.4 .33 10.7 .2 112 .33
= 952.9 60.3 .10 33.6 .33 14.8 .2
4 963.5 49.7 .20 41.9 .33 19.1 .2
c 976.1 37.1 .20 25.8 .1
987.5 25.7 .20
996.2 17.0 .20

L Fa LL R Pr T Pt Remarks

4 .0038 * * 9.0 .1 Po, D, and Pi are te
13 .0059 20.4 .33 12.3 .2 same as those for
o 22 .0065 33.6 .33 15.0 .2 landfalling storms.
c 30 .0066 41.9 .33 17.9 .2
6 43 .0062 23.1 .2
0
61 .0058 29.4 .1

* Landfalling and alongshore storms with P <940 mb:
R
R = 20.4 and 33.6 n.mi. are each assigned a probability of 0.5.

6.2 Astronomical Tides

Most of the combinations of forces producing the astronomical tides are ex-
perienced during a 19-yr cycle. There is also a seasonal variation in the
mean water level with a maximum in September-October. The months July,
August, September, and the first half of October are taken to represent the
hurricane season. Astronomical high and low tides at Morehead City and Avon,
N.C., and Virginia Beach, Va., for these representative months were recomputed
for a 19-yr period by running the standard tide computation program written by
Pore and Cummings (1967). The accumulated frequencies of the high and low
tides were calculated separately by months, then weighted in proportion to
hurricane occurrences in each month (July 13%, August 27%, September 39%,
October 21%). The weighted mean distributions are shown for Avon, N.C. in
figure 12. The resulting probability distributions for Morehead City are
almost the same as those for Avon. The range of high and low tides at
Virginia Beach, Va., is slightly greater than that of Avon, N.C. Interpolat-
ing from the probability curves of the latter two stations leads to distribu-
tions used in tide frequency analyses for locations along the northern por-
tion of the study area, while the probability distributions for Avon are
adopted for locations to the south. As in previous studies, each distribution
is divided into four class interval values of equal range. The representative
astronomical tide marigrams needed to combine with each hurricane surge mani-
gram were then approximated as cosine waves with a period of 12.42 hrs oscil-
lating between corresponding high tide and low tidk class interval values.
This assumes that the highest high tides occur with the lowest low tides, etc.

6.3 Tide Frequencies at Selected Points

The tide frequency curves for the five classes of storms, landfalling,
alongshore, inland, and exiting hurricanes and tropical storms, and winter
coastal storms, are plotted together on figure 13 and combined to give an all-
storm frequency curve for Wright Monument. Figure 14 is a similar plot for
Ocracoke Inlet, N.C., for the classes of storms applicable there. The all-
storm frequency curves for Cape Lookout, Salvo, N.C., and at the N.C.-Va.
border are shown in figure 15. The tide levels are stated as heights above
local mean sea level, adjusted to the 1941-59 epoch. Datum levels and dif-
ferences between datums in use are covered in par. 6.5. The "~open coast" tide
frequencies apply to ocean beaches at or near the locations indicated. It
should be emphasized that these frequency values are of still-water levels on
the open coast that would be measured in a tide gage house or other enclosure,
excluding wave action. The destructive effects of waves on the beach front
must be taken into account separately. In insurance rating this is taken
into account by the ocean front '"velocity zone."

Local effects can modify the elevation of the storm tide. Local features
diminishing "oe coast" elevations in the landward direction include narrow
passes and inlets and obstructions to inundation such as dunes and swamp veg-
etation. Converging shores of bays and strong winds over long fetches of
shallow water (wind "1setup") have the opposite effect. The net results of
these effects can result in either higher or lower storm tide levels of a
given mean return period at estuarine, bay, and inland locations compared to
the open coast. These differences have to be determined by localized studies.

The hurricane surge data presented here are only a portion of that needed for
such localized studies.

6.4 Adjustment Along the Coast

The estimated tide frequencies for locations other than those selected for
computation along this stretch of the coast were obtained by interpolation.
The interpolation was based on consideration of the frequency of storms, the
variation in the shoaling factor (fig. 10), and the trend in the hurricane
climatology parameters along the coast. Figure 16 shows the variation of the
total tide heights along the coast for 10-, 50-, 100-, and 500-yr return
periods, scaled from these diagrams and interpolated as indicated.

6.5 Reference Datum

The National Ocean Survey, NOAA, and its component, the National Geodetic
Survey, have developed the following standards for elevation control. Both
standards are defined here for convenience in consistent application of the
results of this study. The difference between the two is not large.

The National Geodetic Vertical Datum (NGVD) of 1929. This is a level surface,
perpendicular everywhere to the earth's gravity field. Its position is de-
fined by precise leveling between geodetic benchmarks throughout the United
States. Elevation contours on current topographic maps are generally related
to this datum. The NGVD is approximately, but not exactly, at the mean level
of the sea on the coasts. It cannot coincide exactly because of the fact that
the sea is not a geopotential surface, and sea level has risen since 1929.

Local Mean Sea Level (local MSL). The arithmetic mean hourly sea-level
heights over a specific 19-yr series of observations. A nineteen-year period
is required to complete certain lunar cycles (Shureman 1975). The reference
19-yr period, or tidal epoch, is changed about every 25 years. Storm tide
levels in this report are referred to local MSL, 1941-59 epoch. Two reasons
for using local MSL in coastal work are: First, this datum is defined in
terms of actual sea conditions, and therefore meets legal requirements for es-
tablishing "mean low water" and the like which are determined from tide obser-
vations. Second, it is much more economical in many coastal communities to es-
tablish a benchmark by observation of the height of the sea over a period of
time by a gage installed for the purpose, than by leveling from the geodetic
net. The National Ocean Survey has an active program of establishing tidal
benchmarks in this way. Short-period or recent records from these gages are
adjusted to the 1941-59 epoch based on comparison with simultaneous tide ob-
servations at long-record primary stations.

Differences between NVGD and local M'SL have been established at primary tide
gages by leveling, and at certain-s-ubordinate tide stations. These dif-
ferences are interpolated between points at which they have been established.
Differences determined in this way in the study area are within + 0.1 ft.

The methods of developing storm-tide frequencies in this study inherently
yield heights above local MSL. The heights of the frequency graphs are in
terms of this datum referred to the 1941-59 epoch. Adjustment to NVGD is
essentially zero for the stretch of coast under study, as indicated.

Additional information on tidal and geodetic datums can be obtained from the
National Ocean Survey, Rockville, Md., 20852. The tidal datum program is de-
scribed in the NOAA publication "Variability of Tidal Datum and Accuracy in
Determining Datums from Short Series of Observations" (Swanson 1974).

Local MSL relative to land is increasing slowly on the east coast, at the
rate of about a foot a century. Data on trends in sea level relative to land
are given in the above cited publication and by Hicks and Crosby (1974). This
trend is thought to be due mostly to slow subsidence of the land, but may also
reflect change in volume of water in the sea from melting of ice. No adjust-
ment has been made for this secular trend in this study.

6.6 Comparison of Frequency Curve

Figure 17 shows a tide frequency curve (dashed line) for Ocracoke, N.C., and
tide levels of 10 selected hurricanes of recent years (plotted data points) at
Ocracoke Village on the shores of Pamlico Sound, close to Ocracoke Inlet, re-
produced from a report prepared by the Corps of Engineers, Wilmington, N.C.,
District (1963). The tide frequency curve for Ocracoke Inlet of the present
study (copied from fig. 14 curve d) is also shown in the same diagram (solid
line). The tide levels at the 100-yr and 500-yr return period on both curves
reveal a difference of about 0.2 ft. This comparison is intended to verify
the correspondence of the computed frequency curve in figure 14 to the local
historical record.

7. RELATION OF THIS REPORT TO DISASTER PLANNING

The most recent disastrous display of hurricane forces on the U. S. coast
was by Camille, which struck the Bay St. Louis - Pass Christian - Gulfport -
Biloxi, Miss., areas in 1969. According to high-water marks, tne storm tide
reached a level of 24.6 ft above mean sea level (Corps of Engineers 1969). The
central pressure at landfall was about 908 mb (Ho et al. 1975). This is the
most intense hurricane so far to strike the United States mainland during the
period of record keeping. Other disasters could also be recounted, including
Hazel at the N.C.-S.C. border in 1954 and the Ga.-S.C. hurricane of 1893. All
of the eastern seaboard south of Cape Hatteras is exposed to these. For this
part of the area of this study, the National Weather Service recommends a re-
peat of Camille, the worst hurricane to strike the mainland, as a disaster
planning objective without regard to the statistical frequency of such a storm
at an individual point. Such a storm on a critical path for the south end of
the study area of this report would reproduce a maximum of about 13 to 15 ft
MEL.

North of Cape Hatteras a hurricane slightly less intense than Camille (be-
cause of mid-latitude modifications) making a direct strike from the southeast
* ~is a very real, if rare, possibility. Disaster and evacuation planning
should take into account the possibility of storm tide somewhat above the 500-
* ~yr level of figure 16.

The central purpose of this report is to develop actuarial frequencies for
insurance rating and related uses; therefore, all frequencies, including the
coastal profiles of figure 16, are stated in terms of probabilities or mean

recurrence intervals at points. The likelihood of Camille or any other given
intensity of storm somewhere within the study area in any given year is much
greater than the point recurrence interval for the same storm, a difference
that needs to be taken into account in regional planning against disasters.
Regional disaster planning should be based on studies for that particular
purpose.

8. COORDINATION AND COMPARISON WITH OTHER REPORTS

This report has been coordinated with the Corps of Engineers, Department of
the Army, which has made and is making various Flood Insurance studies in the
area.

Comparison with frequency levels in earlier studies for Outer Banks communi-
ties on the ocean side of the island, made during the first part of the Flood
Insurance Program, are given in table 6.

The tide frequencies on the open coast at Cape Lookout, N.C., are the same
as in a report for North Carolina, south of Cape Lookout, prepared by NOAA
for the Federal Insurance Administration (Ho and Tracey 1975).

Table 6.--Comparison of tide frequencies

Location and agency studies Tide level for return period
10-yr 100-yr 500-yr
(ft MSL)

Nags Head, N.C.
FIS study*, September 1972 6.4 8.8 10.0

Present study 5.6 9.0 11.5

Kill Devil Hills, N.C.
FIS study*, October 1972 6.4 8.8 10.0

Present study 5.6 9.0 11.6

*Flood Insurance Study prepared for FIA by Corps of Engineers, U. S. Army,
Wilmington, N.C., District.

References

Barrientos, Celso S., and Chen, Jye, 1975: "Storm Surge Shoaling Corrections
Along the East Coast of the United States," Techniques Development Labora-
tory, National Weather Service, NOAA, U. S. Dept. of Commerce, Silver
Spring, Md., Report to Dept. of Housing and Urban Development (to be
published).

Cooperman, Arthur I., and Rosendal, Hans E., 1962: "Great Atlantic Coast
Storm, 1962," Mariners Weather Log, Vol. 6, No. 3, Weather Bureau, U. S.
Dept. of Commerce, Weather Bureau, Washington, D. C., pp. 79-85.

Corps of Engineers, 1963: U. S. Army Engineer District, Wilmington, N.C.,
"Combined Hurricane and Beach Erosion Control Report - Ocracoke Island,
N.C.," Wilmington, N.C., 39 pp.

Corps of Engineers,1966: "Outer Banks Between Virginia State Line and
Hatteras Inlet, North Carolina," House Document No. 476, 89th Congress, 2nd
Session, U. S. Congress.

Corps of Engineers, 1969: U. S. Army Engineer District, Mobile, Ala., "After
Action Report -- Hurricane Camille/17-18 August 1969," Mobile, Ala., 109 pp.

Cry, George W. 1965: "Tropical Cyclones of the North Atlantic Ocean," U. S.
Weather Bureau Technical Paper No. 55, Weather Bureau, U. S. Dept. of
Commerce, Washington, D. C., 148 pp.

Graham, Howard E., and Nunn, Dwight E., 1959: "Meteorological Considerations
Pertinent to Standard Project Hurricane, Atlantic and Gulf Coasts of the
United States," National Hurricane Research Project Report No. 33, U. S.
Weather Bureau and Corps of Engineers, Washington, D. C., 76 pp.

Hardy, Albert V., and Carney, Charles B., 1962: "North Carolina Hurricanes,"
Weather Bureau, U. S. Dept. of Commerce, Raleigh, N.C., 26 pp.

Harris, D. Lee, 1963: "Characteristics of the Hurricane Storm Surge,"
Technical Paper No. 48, Weather Bureau, U. S. Dept. of Commerce, Washington,
D. C., 139 pp.

Harrison, Louis P., 1963: "History of Weather Bureau Wind Measurements,"
Key to Meteorological Records Documentation No. 3.151, Weather Bureau, U. S.
Dept. of Commerce, Washington, D. C., 68 pp.

Hicks, Steacy D., and Crosby, James E., 1974: "Trends and Variability of
Yearly Mean Sea Level 1893-1972," NOAA Technical Memorandum, NOS 13,
National Ocean Survey, NOAA, U. S. Dept. of Commerce, Rockville, Md., 14 pp.

Ho, Francis P., 1974: "Storm Tide Frequency Analysis for the Coast of
Georgia," NOAA Technical Memorandum, NWS Hydro-19, National Weather Service,
NOAA, U. S. Dept. of Commerce, Silver Spring, Md., 28 pp.

Ho, Francis P., Schwerdt, Richard W., and Goodyear, Hugo V., 1975: "Some
Climatological Characteristics of Hurricanes and Tropical Storms, Gulf and
East Coasts of the United States," NOAA Technical Report NWS-15, National
Weather Service, NOAA, U. S. Dept. of Commerce, Silver Spring, Md., 87 pp.

Ho, Francis P., and Tracey, Robert J., 1975: "Storm Tide Frequency Analysis
for the Coast of North Carolina, South of Cape Lookout," NOAA Technical
Memorandum, NWS Hydro-21, National Weather Service, NOAA, U. S. Dept. of
Commerce, Silver Spring, Md., 44 pp.

Jelesnianski, Chester P., 1967: "Numerical Computations of Storm Surges with
Bottom Stress," Monthly Weather Review, Vol. 95, pp 740-756.

Jelesnianski, Chester P., 1972: "SPLASH - (Special Program to List Amplitudes
of Surges from Hurricanes) I. Landfall Storms," NOAA Technical Memorandum,
NWS TDL-46, National Weather Service, NOAA, U. S. Dept. of Commerce, Silver
Spring, Md., 52 pp.

Jelesnianski, Chester P., 1974: "SPLASH (Special Program to List Amplitudes
of Surges from Hurricanes), Part Two: General Track and Variant Storm
Conditions," NOAA Technical Memorandum,NWS TDL-52, National Weather Service,
NOAA, U. S. Dept. of Commerce, Silver Spring, Md., 55 pp.

Ludlum, David M., 1963: Early American Hurricanes 1492-1870, American
Meteorological Society, Boston, Mass., 198 pp.

Malkin, William, 1959: "Yilling and Intensity Changes in Hurricanes Over
Land," National Hurricane Research Project Report No. 34, Weather Bureau,
U. S. Dept. of Commerce, Washington, D. C., 18 pp.

Myers, Vance A., 1970: "Joint Probability Method of Tide Frequency Analysis
Applied to Atlantic City and Long Beach Island, N.J.," ESSA Technical
Memorandum WBTM Hydro 11, Weather Bureau, ESSA, U. S. Dept. of Commerce,
Silver Spring, Md., 109 pp.

Myers, Vance A., 1975: "Storm Tide Frequencies on the South Carolina Coast,"
NOAA Technical Report, NWS-16, National Weather Service, NOAA, U. S. Dept.
of Commerce, Silver Spring, Md., 133 pp.

Pore, N. A., 1964: "The Relation of Wind and Pressure to Extratropical Storm
Surges at Atlantic City," Journal of Applied Meteorology, Vol. 3, No. 2,
pp. 155-163.

Pore, N. A. and Cummings, R. A., 1967: "A Fortran Program for the Calcula-
tion of Hourly Values of Astronomical Tide and Time and Height of High and
Low Water," ESSA Technical Memorandum WBTM TDL-6, Weather Bureau, ESSA,
U. S. Dept. of Commerce, Silver Spring, Md., 17 pp.

Rhodes, C. E., 1955: "North Atlantic Hurricanes and Tropical Disturbances -
1954" Climatological Data for the United States, National Summary.Annual
1954," Vol. 5, No. 13, Weather Bureau, U. S. Department of Commerce,
Asheville, N.C., pp. 72-88.

Shureman, Paul, 1949: "Tide and Current Glossary," U. S. Coast and Geodetic
Survey, Special Publication No. 228, revised by Steacy D. Hicks, 1975,
National Ocean Survey, NOAA, Rockville, Md.

Simpson, Robert H., and Hope, John, 1972: "Atlantic Hurricane Season of
1971," Monthly Weather Review, Vol. 100, No. 4, pp. 25b-267.

Sumner, H. C., 1944: "North Atlantic Hurricanes and Tropical Disturbances of
1944," Monthly Weather Review, Vol. 72, Washington, D. C., pp. 221-223.

Swanson, Robert L., 1974: "Variability of Tidal Datums and Accuracy in
Determining Datums from Short Series of Observations," NOAA Technical
Report, NOS 64, National Ocean Survey, NOAA, U. S. Dept. of Commerce,
Rockville, Md., 41 pp.

U. S. Weather Bureau, 1899: Monthly Weather Review and Annual Summary,
Vol. 27, Washington, D. C., p. 344.

Richmond

73W
77W ;, Cape Charles 75W 37N
37N + Hampton Roads + 37N

Virginia Beach
VA
N.C.

Elizabeth City

Wri ght Monument

-~ Washington . . . Salvo

w Bern 'p ,,oa Cap e Hatteras <
w.1 73W
7N 'W Portsmouth Ocracoke Inlet 75W 35N 5N

'= <f\Cape Lookout

Wilmington

Cape Fear

73W
77W 75W 33N
+ 33N 33N

Figure 1.--Location map.

X.G/t~~ / ,~ +~193

SEPT.

/T /\ SEPT. SEPT.

OCT. AUG. SEPT. AUG.
/ 1955 \9 61

AUG. OCT. AUG. SEPT. AUG. SEPT. AUG OCT. AUG. PT.
1899 1964 1953 1881 1955 1933 1958 1971 1893 1913

Figure 2.--Tracks of major hurricanes from the southeast quadrant affecting the
study area, 1871-1974.

1879

1960

OCT. / AUG.
SEPT. 1951 1949P
19PT AUG. AUG. SEPT. 1938
1958 1887 1955 1944

Figure 3.--Same as figure 2 but for hurricanes from the southwest quadrant.

SEPT.
1889
SEPT.
,8~~~~~ ~~~~s~~~~1967

~ :N

+i~ ~~~~ ~ ~~~~~~~~SEPT.
1972

JUL. 1934

OCT.
1913

NOV.
1935

OCT. JUL.
1897 1901

Figure 4.--Track of tropical storms and hurricanes showing motion
from northeast.

100 l I

920 940 9
so

U
z
w~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -

0
UA.
o 60-
z
w
U
LU

40 -

20
0-

20 -

920 940 960 980 1000

CENTRAL PRESSURE (mb)

Figure 5.--Probability distribution of central pressure of hurricanes and
tropical storms adopted for Wright Monument, N.C., (a) landfalling
storms from the southeast quadrant and alongshore storms, (b) land-
falling storms from the northeast quadrant, (c) exiting storms, and
(d) all landfalling storms, reproduced from Climatology Report.

100

80 -

0
z
w
in
060- b

IL
I.-

a.
0

0
II

20 -

10 20 30 40
SPEED OF STORM MOTION (kt)

Figure 6.--Probability distributions of speed of storm motion adapted for
Wright Monument, N.C., for (a) landfalling storms from the
southeast quadrant and exiting storms, (b) landfalling
storms from the northeast quadrant,(c) inland storms, and
(d) alongshore storms.

-WI

o w I I I -
o~ >

Os 2.0-
0Q O rO 0

-I IDSAC FR MNC--
0 cc
o Io

10 150 -0 240 3

CD
o I

-j -11- 9 I
o I -

> = I 9 1609 - 1I00

o I

100 150 200 250 300

Figure 7.--Frequency of landfalling hurricanes and tropical storms. Circled dots are count for 50-n.mi.
segments from Climatology Report.

� 10--

5 --

80 60 40 20 20 40 60 80
DISTANCE INLAND (n. mi.) / DISTANCE FROM COAST (n. mi.)

- 15

Figure 8.-Accumulative frequency of alongshore and inland hurricanes
and tropical storms for Wright Monument, N.C., 1871-1974.

+~~~ +
Va.-.C

L k~~~~~~ ~Wright Monument

+~~~

Figure 9.--Schematic diagram illustrating distances L and LL for
hurricanes inland.

N I Il Z

i 0 0 WI~~~YZ l WI
LU z~~~~~~~~~~~~~~~~~~r

i ~ O Irzr

*9 I
LT~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~L

LL~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ L
- ~ ~ ~ 0 0 I0- I I _

I~~~

R 26 n. mi.

Ft 1 n. M..

DISTAN;CE FROM N.C.-S.C. BORDER (n. mi.)
.6 - I I I
0 100 150 200 260 300

Figure IO.--Shoaling factor, North Carolina coast north of Cape
Lookout.Adapted from Barrientos and Chen (1975). W

_- HAMPTON ROADS

a 5-

It! ?

3-
Gumbel
__ _ __Log Pearson
-- Eye Fit

2 , 1 I I I I I , 11
2 5 10 25 50 100 500
RETURN PERIOD (yr)

Figure 11.-Tide frequencies at Hampton Roads, Va., from winter coastal
storms. Plotted points are seasonal (October-May) maxima,

4 ,~~ ~~ ~~ I ' I

~0-
LU

-2 -

-4,
I~~~

0 .2 .4 .6 .8
ACCUMULATED PROBABILITY

Figure 12.--Probability distribution of astronomical high and low tides
for hurricane season, Avon, N.C., 1955-1973.

16 I. .

WVRIGHT MONUMIEN I -

1~~~ 2

t~~~~~~~~~ . . .. .
-J~ ~ ~ ~ ~ ~ ~~~~
. ...1T. f 9.. ;..~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
lo~~~~~~~~v:
~~~F=
LO-LA7:: ~ I-' t - ~
w U-

T'~~~~~~~~~~~~~~~-j T---- -
I- -Q ;I>.. I -..~~~~~~~~~-1t
8-6-
i Y Y "4<~~~~~~~~~~~~~~7

4. - . -

RETURN PEID(r
Fiur 1.- Tde frqecs at9 Wrgh Mo uet N.. oIeea lse fsom:()lnflig

Fi ~ ~ b)aogshre, 13.- Tinad, frquncie at) exiinght oumnfrsvrilclasses adtopia storms; (a) windteilng
storms; (f) all storms.

OCRACOK E INLET~'ij1

1 i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~tm
w~~~~~~~~~~~~~~~~~~~~~~~I '1

12-- ~ ~ 4-_K j+wr.!t

8 ~~~~~..X..J, L~~~~~~~~~~~~~~ If~~~

6 . tri~~Lt ~Tc1~b.LU

[K~~~~ I ~~~~ v:V1 T1~~~~~~~~~~IiV.T~~~~~~i~~7~~~ P I-I~~73
Figue 1.-Tde fequncis a Ocrcok Inet or sverl casss ofstoms:(a)land aiing an
(b lnsoehricnsadtoia som;()wne strms d l ts

16 Ili''.

~~~K14

'---�4~~~~~~~~~~~~~~-

1 ~ ~ ~ t 2

3: ~ ~ ~ - .. I r

4I.,

10 10. ,5O -,0 5

and~~ at the NC-V. bor de r.

15 -vV
I Ii LI Id2I Ie
0 ccI
0 C.U > w
LU& U z
500-yr 0

10 1~~~~o0. 1 ~-0.-7-
I~5 I~~

r I% m100-yr
9.1
8.8 8.8
50-yr
11.0~~~~~.
I- I '-S 1.

z I 7.8 I 77 8.0
8 7.7 ~~~~~~~~~~~~~~~~~7.5-
LU

0 - I ~ 10-yr
I 6.0 I 5.8 i 5 I -
5 ;' 1 55 5.4

DISTANCE FROM N.C.-S.C. BORDER (n.mi.)
- I IIIIII
- I~

100 150 200 250 300

Figure 16.-Coastal tide frequencies. North Carolina, north of Cape Lookout. Ln

8 **

6- ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I 193 .-..' F,

010 1938

M~~~I A~I; K ~ 1958.4ET .

LU~~~~~~~~-. .1381 . .'

2 __ I i ~.i1ILL 1~ j~T- I! ~fI
- j . ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ I~~~~~~~~~~~~~~~~~
4~~~~~~~~~~~~~~__ II~I~~,'

0 I~~~~~~~~~~~~~~~~~~~-
5 50 100 500~~~~~~~~~~~~~~~~~~~~~7
REUR ERO (r

Figure 7.--Comarisonof tidefrequecy curvs at Oracoke nlet, -) thisreport (---) orps o
Enier2epr 16)

*~~~~~~~~~~~~~~~~~~~~~~~~~~~1 - .

(Continued from inside front cover)

NWS HYDRO 15 Time Distribution of Precipitation in 4- to 10-Day Storms--Arkansas-Canadian River Ba-
sins. Ralph H. Frederick, June 1973. (COM-73-11169)

NWS HYDRO 16 A Dynamic Model of Stage-Discharge Relations Affected by Changing Discharge. D. L.
Fread, December 1973. (COM-74-10818)

NWS HYDRO 17 National Weather Service River Forecast System--Snow Accumulation and Ablation Model.
Eric A. Anderson, November 1973. (COM-74-10728)

NW1S HYDRO 18 Numerical Properties of Implicit Four-Point Finite Difference Equations of Unsteady
Flow. D. L. Fread, March 1974.

NUS IHYDRO 19 Storm Tide Frequency Analysis for the Coast of Georgia. Francis P. Hlo, September 1974.
(COM-74-11746/AS)

NWS HYDRO 20 Storm Tide Frequency for the Gulf Coast of Florida From Cape San Blas to St. Petersburg
Beach. Francis P. Ho and Robert J. Tracey, April 1975. (COM-75-10901/AS)

NWS HYDRO 21 Storm Tide Frequency Analysis for the Coast of North Carolina, South of Cape Lookout.
Francis P. Ho and Robert J. Tracey, May 1975. (COM-75-11000/AS)

NNWS HYDRO 22 Annotated Bibliography of NOAA Publications of Hydrometeorological Interest. John F.
Miller, May 1975.

NI'S HYDRO 23 Storm Tide Frequency Analysis for the Coast of Puerto Rico. Francis P. Hlo, May 1975.
(COM-11001/AS)

NWS HYDRO 24 The Flood of April 1974 in Southern Mississippi and Southeastern Louisiana. Edwin !-I.
Chin, August 1975.

NWS HYDRO 25 The Use of a Multizone Hydrologic Model With Distributed Rainfall and Distributed Par-
ameters in the National Weather Service River Forecast System. David J. Morris, August
1975.

NWS HYDRO 26 Moisture Source for Three Extreme Local Rainfalls in the Southern Intermountain Region.
E. Marshall Hansen, October 1975.

A;C$6~~b ER~"C

C~-, 7-~c~?
m NOAA-S/T 76-1847

H1 1'I l Illllill
hiS i1170 TE!EII 666A00I 813
3 6668 00002 8813