Building Performance Assessment:
Hurricane Fran in North Carolina
Observations, Recommendations, and Technical Guidance
Federal Emergency Management Agency
Mitigation Directorate
Washington, DC and Region IV, Atlanta, Georgia

FEMA 290/March 1997


The Building Performance Assessment Team Process

In response to hurricanes, floods, earthquakes, and other disasters, the Federal 
Emergency Management Agency (FEMA) often deploys Building Performance Assessment 
Teams (BPATs) to conduct field investigations at disaster sites. The members of 
a BPAT include representatives of public sector and private sector entities who 
are experts in specific technical fields such as structural and civil 
engineering, building design and construction, and building code development and 
enforcement. BPATs inspect disaster-induced damages incurred by residential and 
commercial buildings and other manmade structures; evaluate local design 
practices, construction methods and materials, building codes, and building 
inspection and code enforcement processes; and make recommendations regarding 
design, construction, and code issues. With the goal of reducing the damage 
caused by future disasters, the BPAT process is an important part of FEMA’s 
hazard mitigation activities.

Cover Photograph:

Hurricane Fran at landfall, September 5, 1996, 7:45 p.m., e.d.t. National 
Oceanic and Atmospheric Administration GOES-8 Color Enhanced IR photograph.





PREFACE

The Federal Emergency Management Agency (FEMA) Mitigation Directorate 
administers the floodplain management provisions of the National Flood Insurance 
Program (NFIP). The Federal Insurance Administration (FIA), also part of FEMA, 
administers the insurance provisions of the NFIP. Together, the Mitigation 
Directorate and FIA have been involved in assessing the performance of buildings 
affected by flooding. To date, FEMA has prepared over 25 building performance 
assessment, damage assessment, and flood hazard mitigation reports. A list of 
these reports is provided in Appendix A of this report. Over ten thousand copies 
of  FEMA<0x2019>s report on Hurricane Andrew have been distributed, and the 
report has been cited by the national media and used by State and local 
governments and model building code organizations as the basis for changes to 
building codes and standards. The findings and recommendations of these reports 
have been used by all levels of government to enhance the performance of 
buildings subject to natural hazards.






TABLE OF CONTENTS


PREFACE

EXECUTIVE SUMMARY 1

1 INTRODUCTION 1-1
 1.1 Purpose 1-1
 1.2 Storm Conditions 1-1
 1.3 Building Siting and Construction Regulations 1-3
 1.3.1 Coastal Area Management Act 1-3
 1.3.2 North Carolina State Building Code 1-3
 1.3.3 National Flood Insurance Program 1-5

2 SITE OBSERVATIONS 2-1
 2.1 Assessment Team Approach 2-1
 2.2 Erosion and Scour 2-3
 2.3 Building Foundation Systems 2-6
 2.3.1 Piling Embedment for Structural Support 2-6
 2.3.2 Piling Embedment for Decks, Porches, and Roof Overhangs 2-10
 2.3.3 Debris Impact on Vertical Foundation Members 2-13
 2.3.4 Grade of Lumber Used for Timber Pilings and Cross-Bracing 2-13
 2.3.5 Elevation of Buildings 2-17
 2.3.6 Cross-Bracing Below Elevated Buildings 2-18
 2.3.7 Solid Perimeter Masonry Foundation Walls Supported on a Continuous 
Footing 2-18
 2.3.8 Manufactured (Mobile) Home and Permanently Installed RV Foundations 2-18
 2.4 Breakaway Walls Below Elevated Buildings 2-21
 2.4.1 Placement of Exterior Sheathing Over Pilings 2-21
 2.4.2 Improper Attachment of Breakaway Wall Panels to Foundation Members 2-22
 2.4.3 Placement of Breakaway Wall Placed Seaward of Cross-Bracing 2-22
 2.5 Below-Building Concrete Slabs 2-24
 2.5.1 Slab Thickness 2-24
 2.5.2 Slab Joints 2-24
 2.5.3 Wire Mesh 2-25
 2.5.4 Connecting the Slab to the Vertical Foundation Members 2-26
 2.5.5 Casting Concrete Grade Beams and Slabs-On-Grade Monolithically 2-27
 2.5.6 Concrete Collars 2-27
 2.6 On-Site Utility Systems 2-28
 2.6.1 Air Conditioning / Heat Pump Compressor Platforms 2-28
 2.6.2 Placement of Utilities on, Through, or Adjacent to Breakaway Wall Panels 
2-29
 2.6.3 Placement of Utilities Adjacent to Vertical Support Members 2-32
 2.6.4 Septic Tanks 2-32
 2.7 Dry Floodproofing in Coastal A Zones 2-33
 2.8 Wind Damage 2-35
 2.9 Corrosion of Structural Metal Components 2-35
 2.10 Concerns Regarding the Effective FIRMs for North Carolina Coastal 
Communities 2-35
 2.11 Building Performance Successes 2-37

3 RECOMMENDATIONS 3-1
 3.1 Building Foundation Systems 3-1
 3.1.1 Piling Embedment for Structural Support 3-1
 3.1.2 Piling Foundations for Decks, Porches, and Roof Overhangs 3-2
 3.1.3 Grade of Lumber Used for Timber Pilings and Cross-Bracing 3-3
 3.1.4 Proper Elevation of Coastal Buildings 3-3
 3.1.5 Cross-Bracing Below Elevated Buildings 3-4
 3.1.6 Solid Perimeter Masonry Foundation Walls Supported on a Continuous 
Footing 3-4
 3.1.7 Manufactured (Mobile) Home and Permanently Installed RV Foundations 3-4
 3.2 Breakaway Walls Below Elevated Buildings 3-6
 3.2.1 Placement of Exterior Sheathing Over Pilings 3-6
 3.2.2 Improper Attachment of Breakaway Wall Panels to Foundation Members 3-8
 3.2.3 Placement of Breakaway Wall Panels Seaward of Cross-Bracing 3-8
 3.3 Below-Building Concrete Slabs	3-8
 3.3.1 Slab Thickness 3-8
 3.3.2 Slab Joints 3-8
 3.3.3 Wire Mesh 3-8
 3.3.4 Connecting the Slab to the Vertical Foundation Members 3-8
 3.3.5 Casting Concrete Grade Beams and Slabs-On-Grade Monolithically 3-10
 3.3.6 Concrete Collars 3-10
 3.4 On-Site Utility Systems 3-10
 3.4.1 Air Conditioner / Heat Pump Compressor Platforms 3-10
 3.4.2 Placement of Utilities on, Through, or Adjacent to Wall Panels 3-12
 3.4.3 Placement of Utilities Adjacent to Vertical Support Members 3-12
 3.4.4 Septic Tanks 3-12
 3.5 Dry Floodproofing 3-12
 3.6 Wind Damage Protection 3-14
 3.7 Protection of Metal Structural Components from Corrosion 3-15 

APPENDIXES

Appendix A 
Building Performance Assessment, Damage Assessment, and Hazard Mitigation 
Reports Prepared by the Federal Emergency Management Agency

Appendix B  
Members of the Building Performance Assessment Team for Hurricane Fran

Appendix C 
Executive Summary from Piling Embedment Study Performed by 	Woodward-Clyde 
Federal Services on Topsail Island, North Carolina

LIST OF FIGURES

1-1 High water mark elevations, in feet above Mean Sea Level, surveyed after 
Hurricane Fran. 1-2

1-2 Minimum oceanfront setback requirements under the North Carolina Coastal 
Area Management Act. 1-4

2-1 Areas of BPAT aerial survey and ground observations. 2-2

2-2 Erosion resulted in significant loss of supporting sand, averaging 4 to 6 
feet, under oceanfront buildings. 2-3

2-3 Determination of localized scour from changes in sand color, texture, and 
bedding. 2-4

2-4 Loss of the frontal dune and the resulting erosion and scour left many 
coastal houses unable to resist wind and flood loads acting simultaneously. 2-5

2-5 Overwash of barrier islands generated high-velocity flows that caused 
extensive scour adjacent to large objects. 2-5

2-6 The disruption of velocity flows by large, non-breakaway objects generated 
extensive scour that undermined vertical foundation members and slabs-on-grade. 
2-6

2-7 Over 100 oceanfront houses were washed off their foundations or completely 
destroyed. 2-7

2-8 Many oceanfront houses built prior to current (1986) North Carolina State 
Building Code requirements were found to be leaning or destroyed. 2-7

2-9 Typical collapse mechanism of post-FIRM building based on pre-1985 embedment 
requirements in North Carolina coastal areas. 2-8

2-10 One of several buildings observed to be leaning landward that were 
suspected of having been constructed to current North Carolina State Building 
Code requirements.2-9

2-11 The BPAT observed several hundred decks and porches that collapsed as a 
result of insufficient foundation support. 2-10

2-12 Example of building constructed to current North Carolina State Building 
Code requirements with insufficient embedment of pilings/columns under two-story 
deck. 2-11

2-13 Embedment of deck supports into frontal dune was often shallow. 2-11

2-14 Storm-generated debris impacted nearby structure. 2-12

2-15 As shown by this post-Fran photograph taken at Figure Eight Island North 
Carolina, a current architectural trend is the construction of multistory decks 
supporting roof overhangs. 2-13

2-16 Typical failure of deck/roof overhang and air conditioning/ heat pump 
compressor platform for landward home. 2-14

2-17 Precast concrete ring section of septic tank became waterborne debris, 
impacting building foundation members. 2-15

2-18 Impact of debris from a damaged deck appeared to have broken cross-bracing. 
2-15

2-19 Example of extreme impact — two houses floated and pushed into another 
house. 2-16

2-20 Example of broken piling. 2-16

2-21 Survival of this properly elevated North Carolina State Park public rest 
room demonstrates the State’s commitment to proper construction in coastal 
areas. 2-17

2-22 Collapse of footing and foundation wall under elevated wood-frame building. 
2-19

2-23 Catastrophic failure of landward building constructed on masonry wall and 
slab-on-grade foundation 2-19

2-24 Permanently installed RV overturned as a result of anchor pullout. 2-20

2-25 Localized scour beneath pre-storm footprint of manufactured home. 2-20

2-26 Exterior sheathing of breakaway wall spanned piling. 2-22

2-27 Breakaway wall panel failed to function as designed because continuous 
sheathing was installed across pilings. 2-23

2-28 Use of anchor bolts through the sill plate of a breakaway wall is improper. 
2-23

2-29 Improper installation of breakaway walls (on the seaward side of cross-
bracing) resulted in failure of cross-bracing. 2-24

2-30 Example of unnecessarily thick slab. 2-25

2-31 Concrete slab-on-grade flipped up, probably by wave action, came to a rest 
against two foundation members, generating large, unanticipated loads on the 
foundation. 2-26

2-32 Use of steel dowels to tie slabs to vertical foundation members prevented 
the proper breakup of slab. 2-27

2-33 Air conditioning / heat pump compressor platform leaning because erosion 
and scour caused loss of one supporting column. 2-28

2-34 Although the air conditioning/heat pump compressor platform on this 
oceanfront house was adequately elevated with cantilever bracing, the compressor 
was almost pushed off the platform by wind because of the lack of the necessary 
attachment. 2-29

2-35 Air conditioning / heat pump compressor platform leaning scour caused loss 
of two supporting posts. 2-30

2-36 Example of utility component (electric panel box) installed on a breakaway 
wall panel. 2-30

2-37 Utility components (dryer vent, air conditioning compressor feed line, main 
electrical service line out of meter box, and electric wiring) penetrating 
breakaway wall panel. 2-31

2-38 These utility components (wiring, electric panel box, ductwork and sewer 
line) were installed adjacent to breakaway wall panel and were damaged when the 
walls broke away.	2-31

2-39 Oceanfront erosion and scour unearthed and damaged septic tanks and 
systems. 2-32

2-40 Engineered flood shield installed over opening to large dry floodproofed 
commercial building. 2-33

2-41 Buildings that are floodproofed require extensive engineering detailing. 2-
34

2-42 Dry floodproofing often requires extensive human intervention. 2-34

2-43 Corrosion of galvanized floor truss plates was observed in many buildings. 
2-36

2-44 Corrosion of galvanized floor truss plates was observed in buildings. 2-36

2-45 Corrosion of hurricane straps and steel plates was also observed in many 
locations. 2-37

2-46 While this house experienced 6 or more feet of vertical erosion and scour, 
as well as the loss of breakaway wall panels, the foundation and superstructure 
performed as designed. 2-38

2-47 As shown by this post-Fran photograph taken at Emerald Isle, North 
Carolina, proper elevation and setback from the oceanfront in conjunction with 
substantial protection afforded by dunes resulted in little or no	flood damage 
to the oceanfront row of buildings. 2-39

2-48 This large, engineered oceanfront structure performed as designed. 2-39

3-1 Deepening foundation to account for scour. 3-5

3-2 Recommended practice for breakaway wall sheathing. 3-6

3-3 Recommended breakaway wall attachment. 3-7

3-4 Recommended contraction joint layout for frangible slab-on-grade below 
elevated building. 3-9

3-5 Cantilevered mechanical platform. 3-11

3-6 Mechanical platform supported by pilings. 3-11

3-7 Proper location of utilities. 3-13

3-8 Dry floodproofing in coastal A zones. 3-14






EXECUTIVE SUMMARY

On September 5, 1996, Hurricane Fran made landfall near Cape Fear, North 
Carolina (see Figure 1-1 in Section 1), and generated considerable rainfall, 
moderately high winds, and storm surge and waves along the coast. The National 
Oceanic and Atmospheric Administration estimated that Hurricane Fran generated 
1-minute sustained winds of 115 miles per hour. Storm surge elevations 
approached or exceeded National Flood Insurance Program (NFIP) base flood 
elevations from Kure Beach North, Carolina, to North Topsail Beach, North 
Carolina, along approximately 50 miles of coastline. The recorded maximum high 
water, assumed to include wave effects, was 15.4 feet above mean sea level 
(m.s.l.) at Kure Beach. Although the storm generated high winds along the coast 
and well inland, severe damage to buildings was concentrated in those areas also 
impacted by the flood surge and waves. This report focuses on the damages along 
the North Carolina coast that resulted from flood surge, wave action, erosion, 
and scour.

On September 12, 1996, the Mitigation Directorate of the Federal Emergency 
Management Agency (FEMA) deployed a Building Performance Assessment Team (BPAT) 
to coastal North Carolina to assess damage caused by Hurricane Fran. The team 
was composed of FEMA Headquarters and regional engineers, a State 
representative, a consulting structural engineer, a consulting specialist in 
coastal construction and shoreline erosion, a consulting coastal engineer, the 
Chief Underwriter of the NFIP, and an engineer from the Insurance Institute for 
Property Loss Reduction. (See Appendix B for a list of team members.) Some 
members of the BPAT also represented the American Society of Civil Engineers 
(ASCE) Committee on Flood-Resistant Design and Construction.

The mission of the BPAT was to assess the performance of buildings on the 
barrier islands most directly affected by Hurricane Fran and to make 
recommendations for improving building performance in future events. Better 
performance of building systems can be expected when the causes of observed 
failures are determined and repair and reconstruction are undertaken in 
accordance with recognized standards of design and construction. The immediate 
goal of the BPAT process is to provide guidance to State and local governments 
for post-hurricane reconstruction. In addition, the BPAT<0x2019>s findings can 
enhance future coastal design and construction.

The BPAT made its assessments by conducting site investigations to observe the 
condition of buildings in selected areas affected by the storm. The scope of the 
BPAT process did not include recording the numbers of buildings damaged by the 
hurricane, determining the frequency of specific types of damage, or collecting 
other data that could serve as the basis of statistical analyses. Collectively, 
the team invested over 600 hours of effort conducting site investigations, 
inspecting damages, and preparing documentation. Documentation of observations 
made during ground-level and aerial surveys included field notes and  
photographs. 

The BPAT assessed the performance of primary structural systems of buildings, 
i.e., systems that support the building against lateral and vertical loads 
experienced during a hurricane; building extensions, such as decks, porches, and 
roof overhangs; nonstructural building components such as breakaway walls and 
below-building concrete slabs; and on-site building support utilities such as 
electrical, water, and sewage services. The team focused its efforts on primary 
structural systems. It is extremely important to note, however, that damage to 
other portions of buildings often contributed to the damage incurred by the 
primary structural systems.

The building types observed were primarily one- and two-family, one- to three-
story, wood-frame structures elevated on wood pilings. Other types of 
construction observed included one- and two-family wood-frame slab-on-grade 
houses, manufactured homes and permanently installed recreational vehicles (RVs) 
on dry-stack masonry foundations, and a small number of wood-frame structures 
elevated on solid-perimeter masonry walls. In general, wood-frame structures 
elevated on piling foundations outperformed structures on all other types of 
foundations (e.g., masonry pier, continuous masonry wall [crawl space], slab-on-
grade) in resisting flood effects, including velocity flow from storm surge, 
wave action, debris impact, erosion, and scour. The team also observed two 
commercial structures: a hotel in which dry floodproofing measures helped 
protect the building from flood damage and a large oceanfront engineered 
concrete structure that performed well.

Coastal areas from Cape Fear to Cape Lookout experienced significant erosion and 
scour. <cLeading:-1.000000>Erosion caused by Hurricane Fran was exacerbated by 
the previous dune erosion caused by Hurricane Bertha, which made landfall in the 
same area only 2 months earlier. In many locations, especially from Topsail 
Beach to North Topsail Beach, localized frontal dunes were eroded  and the beach 
profile was lowered 2 to 3 feet. Erosion of up to 4 to 6 vertical feet beneath 
oceanfront homes was measured in many locations. In addition, localized scour 
measured at vertical foundation members generally reached one to 1 to 1.5 times 
the diameter or width of the member. Measurements of combined erosion and scour 
commonly totaled 5 to 7 vertical feet at oceanfront homes in the area from 
Topsail Beach to North Topsail Beach. This erosion and scour, added to the 
average long-term erosion rate of 1 to 2 feet a year, left many oceanfront homes 
unable to withstand the loads experienced.

The combined effects of erosion and scour resulted in the collapse of well over 
100 oceanfront homes with shallow piling foundation systems in the area from 
Topsail Beach to North Topsail Beach. Several similar oceanfront homes were lost 
in the Kure Beach-Carolina Beach area. The loss of supporting sand left many 
short pilings either completely exposed or embedded less than 2 feet. In either 
case, some pilings gave way. As a result, the remaining foundation pilings were 
overloaded and the elevated building collapsed. In those rare instances where 
oceanfront homes were constructed on slabs-on-grade, the loss of supporting sand 
coupled with the impact of velocity flow and breaking waves on the walls of the 
structures caused the structures to collapse.

The team observed very little damage in some areas, even oceanfront areas where 
velocity flows, wave action, and severe erosion occurred. The successful 
performance of buildings in these areas demonstrates the value of compliance 
with NFIP requirements regarding the elevation of buildings in coastal flood 
hazard areas and current State of North Carolina requirements regarding setback 
and piling embedment depth for oceanfront structures. The observations of the 
team and the findings of a separate study of piling embedment depth conducted 
for FEMA on Topsail Island (see Appendix C) suggest that the more stringent 
embedment depth requirements incorporated into the North Carolina State Building 
Code in 1986 helped reduce damage. The use of flood-resistant construction 
materials and techniques, such as in engineered concrete buildings was also 
effective. In addition, breakaway walls, although generally not installed 
properly, usually broke away as intended under the impact of flood forces and 
helped prevent structural damage.

Although the BPAT noted that breakaway walls generally performed as intended, 
three design and construction errors were observed that are worth noting:

- Breakaway wall panels were often installed immediately adjacent to and seaward 
of cross-bracing. When the panels broke away, they were pushed against the 
cross-bracing by flood waters. The resulting force on the vertical surface 
generated loads far in excess of the design strength of the cross-bracing. As a 
result, cross-bracing was broken or torn away.

- Utilities were installed on, through, or adjacent to breakaway wall panels. As 
a result, the panels were often prevented from breaking away cleanly under flood 
loads, and when they did break away, the utilities were damaged. Much of the 
utility damage observed was a direct result of improper installation.

- Sheathing was installed on the exterior of breakaway wall panels, continuously 
over the outside face of vertical foundation members. Sheathing installed in 
this way inhibits the ability of the breakaway wall panels to break away 
cleanly.

Most slabs-on-grade below elevated buildings broke apart under the hydrodynamic 
and impact loads imposed by flood waters and therefore did not transfer those 
loads to the foundation system. Also, the resulting slab fragments were usually 
small enough that when they became waterborne debris, they did not damage 
foundation system components. However, the BPAT observed some design and 
construction errors worth noting:

- In some instances, slabs were attached to vertical foundation members with 
steel dowels placed in the piling and cast into the slab. This practice resulted 
in the transfer of unanticipated loads to the foundation system and may have 
caused the failure of some foundations systems.

- The slabs observed generally did not have a sufficient number of contraction 
joints to promote the slab<0x2019>s breaking into small pieces. In one instance, 
a large section of a slab was observed to have been lifted by flood forces and 
to have come to rest against vertical foundation members. Although evidence of a 
cause and effect relationship was not directly observed, slabs that reacted in 
this way may have led to the failure of some buildings as well.

- The use of wire mesh cast into slabs further complicated matters by holding 
pieces of the slabs together after the slabs had fractured.

- Concrete collars were occasionally placed around pilings during the 
construction of slabs. Although the collars were intended to provide stability, 
they increased wave loads and scour by presenting larger obstructions to flow. 
Also, once the underlying sand was removed by erosion, the collars increased the 
dead weight of the pilings to which they were attached.

Utilities that were not installed in a manner that afforded the greatest extent 
of flood protection possible were damaged. Although portions of most utility 
services must extend below the flood level, many simple techniques are available 
to minimize or eliminate damages. Observed damage to water, sewage, electrical, 
telephone, and cable TV services could have been avoided. Septic tanks were 
routinely left exposed by storm-induced erosion and scour, and their connections 
to buildings were severed. The tanks were then filled with flood water and 
debris.

On oceanfront homes, many porches, decks, and roof overhangs supported on 
vertical foundation members collapsed or became structurally unsound. Similar 
failures occurred in the porches, decks, and roof overhangs attached to some 
inland homes. These failures were observed in both new and old structures. The 
vast majority of the vertical foundation members were found to have been 
embedded only 4 to 5 feet below existing grade without any regard for erosion or 
scour. In a few situations, undersized vertical support members, usually 
measuring 4 inches by 4 inches or 6 inches by 6 inches, were shattered, probably 
by the impact of waterborne debris.

Many manufactured homes and permanently installed RVs were installed on dry-
stack masonry block foundations with metal tiedown straps attached to ground 
anchors. This method of installation performed very poorly. The failure of these 
foundations resulted in the loss of approximately 50 percent of the manufactured 
homes and RVs observed in the Surf City and North Topsail Beach areas. The 
causes of failure observed by the BPAT were undermining of the dry-stack block 
by scour resulting from relatively shallow velocity flow, failure of the tiedown 
straps due to corrosion from salt spray, and pullout of the ground anchors. 
Pullout of ground anchors occurred when the pullout resistance of the soil was 
exceeded because of improper anchor selection and/or saturation of the 
restraining sandy soil when the site flooded.

The BPAT developed recommendations for reducing future hurricane damage. The 
recommendations address areas of concern such as building materials (including 
corrosion protection for metal structural components, e.g., hurricane clips, 
straps, and fasteners), design practices, construction techniques, and quality 
of construction. The recommendations presented in this report are applicable in 
other communities that experience similar coastal flooding.

This report presents the BPAT<0x2019>s observations of the successes and 
failures of buildings that experienced the flood effects of Hurricane Fran, 
comments on building failure modes, and provides recommendations intended to 
enhance the performance of buildings in future hurricanes.






1 INTRODUCTION

1.1 Purpose

The purpose of this report is to present the observations of the Federal 
Emergency Management Agency’s (FEMA’s) Building Performance Assessment Team 
(BPAT) regarding the successes and failures of buildings that experienced the 
wind and flood effects of Hurricane Fran in North Carolina, to comment on the 
failure modes of damaged buildings, and to provide recommendations for 
improvements intended to enhance the performance of coastal buildings in future 
hurricanes.

1.2 Storm Conditions

On September 5, 1996, Hurricane Fran made landfall in the vicinity of the Cape 
Fear, North Carolina (see Figure 1-1). According to the National Hurricane 
Center, Fran was ranked as a Category 3 (major) hurricane on the Saffir-Simpson 
Scale. Hurricane Fran was the most intense hurricane to make landfall along the 
U.S. mainland during 1996. Although Fran’s destructive storm surge, waves, and 
winds impacted the immediate coastal areas east and north of Cape Fear, heavy 
rainfall and high winds occurred well inland and resulted in riverine flooding 
and wind damage to residential and commercial buildings, manufactured homes, 
trees and crops, and power distribution systems in North Carolina, Virginia, 
West Virginia, and Maryland. Much of the wind-related damage was not caused 
directly by the wind but by wind-downed trees. In areas where soils were 
saturated by the heavy rainfall, many trees were unable to resist the high winds 
and caused extensive damage when they fell.

[Begin figure]
Figure 1-1 High water mark elevations, in feet above Mean Sea Level, surveyed 
after Hurricane Fran.
[End figure]

The National Hurricane Center and the National Weather Service estimated that 
Hurricane Fran’s maximum 1-minute sustained wind speed was 115 miles per hour 
(mph). It appears that Hurricane Fran may have reached design wind speeds (110 
mph, fastest mile for 50-year return frequency) in a small area along the 
immediate oceanfront near Figure Eight Island. However, most coastal buildings 
in the study area appear to have received less than design wind speeds. A peak 
gust of 95 mph was recorded 940 feet from the ocean in Kure Beach. Although the 
storm generated high winds along the coast and well inland, severe damage to 
buildings was concentrated in those areas also affected by the storm surge and 
waves. 

Independent of the BPAT process, FEMA’s Mitigation Directorate and the Federal 
Insurance Administration conducted a high water mark survey in the wake of 
Hurricane Fran from just west and south of Cape Fear to just west of Cape 
Lookout. The goal of the survey was to determine and map approximate high water 
mark elevations that indicate the stillwater storm surge elevation and the 
combined effect of storm surge and waves in areas significantly affected by 
Hurricane Fran. The resulting historical record will prove useful to FEMA in the 
revision of Flood Insurance Studies and to the insurance industry in the 
settlement of claims regarding flood and wind damage. Selected elevation 
measurements made during this survey are shown in Figure 1-1. Storm surge 
elevations approached or exceeded National Flood Insurance Program (NFIP) Base 
Flood Elevations (BFEs) from Kure Beach to North Topsail Beach, along 
approximately 50 miles of coastline. As shown in the figure, a maximum storm 
surge elevation of 11.9 feet above mean sea level (m.s.l.) — as measured inside 
a structure — was recorded at Figure Eight Island, North Carolina. The maximum 
recorded high water mark of 15.4 feet m.s.l. (assumed to reflect storm surge 
plus wave height) was at the southern end of Kure Beach, North Carolina.

The severity of erosion of oceanfront sand dunes is closely related to the storm 
surge elevation at the shoreline. It is reasonable to assume that dune erosion 
due solely to Fran was a 100-year event. However, 2 months prior to Fran, 
Hurricane Bertha made landfall in the same area. Wind speeds and water levels 
were significantly less than those associated with Fran and significantly below 
design conditions. Storm-induced dune erosion is at least partly temporary, but 
there had been insufficient time for much recovery following Bertha’s estimated 
storm surge of 6 to 9 feet m.s.l. The cumulative effect of back-to-back 
hurricanes appears to have caused dune erosion distances in excess of what would 
be expected to occur in a single 100-year storm surge.

This report focuses on the damages along the North Carolina coast that resulted 
from storm-induced flood surge, wave action, erosion, and scour.

1.3 Building Siting and Construction Regulations

Building construction regulations on the North Carolina coast have been 
established by the North Carolina Coastal Area Management Act (CAMA), the North 
Carolina State Building Code, and the NFIP. CAMA identifies ocean hazard areas, 
establishes oceanfront setback lines for new construction, and protects sand 
dunes. The State Building Code regulates most structural requirements. NFIP 
Flood Insurance Rate Maps (FIRMs) identify flood hazard areas and provide BFEs. 
BFEs are used to establish minimum floor elevations for buildings in 100-year 
flood hazard areas and other prescriptive and descriptive requirements of the 
NFIP. State requirements regarding most other construction criteria are more 
stringent than those of the NFIP.

1.3.1 North Carolina Coastal Area Management Act

In 1979 CAMA identified ocean hazard areas along the North Carolina coastline. 
All new buildings were required to be set back from the seaward line of stable 
dune vegetation at least 30 times the long-term erosion rate determined by the 
North Carolina Division of Coastal Management (see Figure 1-2 A). A minimum 
erosion rate of 2 feet per year was adopted. Additional setbacks were required 
on the largest primary or frontal sand dunes. On previously subdivided lots too 
small to meet the setback requirement, exemptions were allowed for single-family 
houses as close as 60 feet from the vegetation line. In 1985 the minimum setback 
distance for commercial buildings larger than 5,000 square feet was increased to 
60 times the long-term erosion rate (see Figure 1-2 B), with additional 
exemptions where the rate is greater than 3.5 feet per year.

[Begin figure]
Figure 1-2 Minimum oceanfront setback requirements under the North Carolina 
Coastal Area Management Act.
[End figure]

1.3.2 North Carolina State Building Code

The North Carolina State Building Code is based on the Standard Building Code 
with significant revisions adopted by the North Carolina Building Code Council. 
A separate Residential Building Code provides more prescriptive criteria for 
one- and two-family dwellings and is now based on the Council of American 
Building Officials (CABO) Code with substantial amendments by the Council. Most 
of the buildings observed near the coast had been constructed under the 
Residential Code, which was first adopted in the mid-1960’s and has undergone 
several major revisions. After seven major hurricanes affected the North 
Carolina coast in the 1950’s, the Council adopted specific hurricane-resistant 
criteria for small residential buildings on the barrier islands. These were 
initially an optional appendix but soon became mandatory, with enforcement by 
local building officials. With a push from the State Building Code, building 
designs quickly shifted away from low floor elevations with shallow foundations 
to piling foundations elevated for underhouse parking. Since the 1960’s, most 
houses on the barrier islands have been constructed on pilings.

The State Building Code initially required pilings to be embedded 8 feet below 
grade. Later, it became apparent that this piling penetration requirement was 
inadequate for erosion-prone oceanfront buildings. The Code was revised on 
January 1, 1986, to require piling foundations in all coastal high hazard areas 
(NFIP V zones) and ocean hazard areas (identified by CAMA). Buildings 
constructed closer to the seaward edge of the vegetation line than 60 times 
CAMA’s long-term erosion rate are now required to have pilings extending to -5.0 
feet m.s.l. or 16 feet below grade, whichever is less. At the same time, 
requirements for cross-bracing between pilings were added to improve wind 
resistance, making buildings with longer pilings readily distinguishable from 
older buildings on unbraced pilings with shallower embedment.

Wind-resistant construction techniques emphasizing improved connections from 
roof to foundation were in standard practice before 1970. Major increases in the 
wind criteria in the Residential Building Code have already been adopted and are 
scheduled to take effect sometime in 1997. The new criteria will, for the first 
time, apply up to 100 miles inland from the coast, rather than only on the 
barrier islands.

1.3.3 National Flood Insurance Program

All communities on North Carolina barrier islands participate in the NFIP. The 
NFIP was created by an act of Congress in 1968 to make flood insurance available 
to property owners in communities that agree to enact and administer floodplain 
management regulations that meet program requirements. The regulations require 
that new and substantially improved buildings in floodprone areas be built in 
such a manner as to reduce flood hazards and loss of life and property resulting 
from floods. In coastal areas, this means that buildings must be adequately 
elevated and protected from the effects of high-velocity flood flow. In V zones, 
buildings must be elevated on piling foundations and the lowest horizontal 
structural member of the lowest floor must be at or above the BFE. In addition, 
the area below the building must be free of obstructions or enclosed by non-
supporting breakaway walls intended to collapse under wind an water loads 
without causing damage to the foundation or the elevated portion of the 
building. In coastal A zones, which are less likely to be affected by high-
velocity flow, the lowest floor of the building must be at or above the BFE and 
the areas below the BFE can be enclosed with non-breakaway walls.

In the mid-1970’s, FEMA issued a FIRM for each of the barrier island communities 
in North Carolina. When the communities began implementing their required 
floodplain management regulations in the late 1970’s, the minimum lowest floor 
elevation requirements based on the BFEs shown on the FIRMs superseded the 
previous State Building Code requirement that the lowest floor be 2 feet above 
the highest known historical water mark. The resulting common use of piling 
foundations with underhouse parking generally placed the elevated floors well 
above minimum elevations required by the NFIP. However, finished underhouse 
enclosures constructed with non-load-bearing walls were common in older 
buildings and, in some communities, in new buildings.

Concerns about the accuracy of the information shown on FIRMs for areas near the 
ocean had been previously raised in North Carolina communities affected by 
Hurricane Fran. According to the FIRMs, many oceanfront lots are within B zones 
and C zones, outside the 100-year flood hazard area. In general, minimal 
elevation requirements at the building sites on these lots did not include 
consideration of waves above the stillwater flood elevation. The accuracy of the 
FIRMs and the steps being taken by FEMA in response to this issue are discussed 
in Section 2.10 of this report.

An important provision that communities participating in the NFIP must include 
in their floodplain management regulations is the requirement that substantially 
damaged buildings, if restored, meet the same requirements imposed for new 
buildings. The NFIP defines substantial damage as “damage of any origin 
sustained by a structure whereby the cost of restoring the structure to its 
before damage condition would equal or exceed 50 percent of the market value of 
the structure before the damage occurred.” The BPAT observed several hundred 
buildings in the area between Kure Beach and North Topsail Beach that may have 
been substantially damaged. The vast majority of these were oceanfront 
residential buildings removed from their foundations by flood forces.






2 SITE OBSERVATIONS

2.1 Assessment Team Approach

On September 12, 1996, the FEMA Mitigation Directorate deployed a BPAT to 
coastal North Carolina to assess damage caused by Hurricane Fran. The team was 
composed of FEMA Headquarters and regional office engineers, a State 
representative, a consulting structural engineer, a consulting specialist in 
coastal construction and shoreline erosion, a consulting coastal engineer, the 
Chief Underwriter of the NFIP, and an engineer from the Insurance Institute for 
Property Loss Reduction. (See Appendix B for a list of team members.) Some 
members of the BPAT also represented the American Society of Civil Engineers 
(ASCE) Committee on Flood-Resistant Design and Construction.

The mission of the BPAT was to assess the performance of buildings on the 
barrier islands most directly affected by Hurricane Fran and to make 
recommendations for improving building performance in future events. Better 
performance of building systems can be expected when the causes of observed 
failures are determined and repair and reconstruction are undertaken in 
accordance with recognized standards of design and construction. The immediate 
goal of the BPAT process is to provide guidance to State and local governments 
for post-hurricane reconstruction. In addition, the BPAT’s findings can enhance 
future coastal design and construction.

The BPAT made its assessments by conducting site investigations to observe the 
condition of buildings in selected areas affected by the storm. The scope of the 
BPAT process did not include recording the numbers of buildings damaged by the 
hurricane, determining the frequency of specific types of damage, or collecting 
other data that could serve as the basis of statistical analyses. Collectively, 
the team did invest over 600 hours of effort conducting site investigations, 
inspecting damages, and preparing documentation. Documentation of observations 
made during ground-level and aerial surveys included field notes and 
photographs. 

On Friday, September 13, 1996, the BPAT conducted an aerial survey along the 
North Carolina coast from Wrightsville Beach (in the south) to Emerald Isle (in 
the north). Ensuing ground observations were made in the area extending from 
Kure Beach (in the south) to North Topsail Beach (in the north). Figure 2-1 
shows the areas where the aerial surveys and ground observations were made. 
Other communities in the studied area include Carolina Beach, Wrightsville 
Beach, Topsail Beach, and Surf City. Documentation of observations made during 
the ground and aerial surveys included field notes and photographs. 

[Begin figure]
Figure 2-1 is a map showing areas of BPAT aerial survey and ground observations.
[End figure]

The BPAT assessed the performance of primary structural systems of buildings, 
i.e., systems that support the building against lateral and vertical loads 
experienced during a hurricane; building extensions, such as decks, porches, and 
roof overhangs; nonstructural building components such as breakaway walls and 
below-building concrete slabs; and on-site building support utilities such as 
electrical, water, and sewage services. The team focused its efforts on primary 
structural systems. It is extremely important to note, however, that damage to 
other portions of buildings often contributed to the damage incurred by the 
primary structural systems.

The building types observed were primarily one- and two-family, one- to three- 
story, wood-frame structures elevated on wood pilings. Other types of 
construction observed included one- and two- family wood-frame, slab-on-grade 
houses, manufactured homes and permanently installed recreation vehicles (RVs) 
on dry-stack masonry foundations, and a small number of wood-frame structures 
elevated on solid perimeter masonry walls. In general, wood-frame structures 
elevated on piling foundations outperformed all other types of foundations 
(e.g., masonry pier, solid perimeter masonry wall [crawl space], slab-on-grade) 
in resisting flood effects, including velocity flow, storm surge, breaking 
waves, debris impact, erosion, and scour. The team also observed two commercial 
structures: a hotel in which dry floodproofing measures helped protect the 
structure from flood damage and a large oceanfront engineered concrete building 
that performed well.

2.2 Erosion and Scour

Oceanfront Residential Buildings

Coastal areas from Cape Fear to Cape Lookout experienced significant erosion and 
scour. In many locations, especially from Topsail Beach to North Topsail Beach, 
localized frontal dunes were eroded and the beach profile was lowered 2 to 3 
feet. Erosion beneath oceanfront homes averaged 4 to 6 vertical feet (see Figure 
2-2). In addition, erosion and localized scour at vertical foundation members 
was observed to have occurred.

[Begin figure]
Figure 2-2 is a photo showing how erosion resulted in significant loss of 
supporting sand, averaging 4 to 6 feet, under oceanfront buildings.
[End figure]

A cursory study of localized scour was performed during the site investigation. 
Sand surrounding pilings was excavated to identify the maximum localized scour 
that occurred. From changes in sand color, texture, and bedding, the team 
determined that, in general, localized scour occurred to a depth of 
approximately 1 to 1.5 times the diameter or width of the piling (see Figure 2-
3). The depth of scour around 8-inch-diameter round pilings and 8-inch x 8-inch 
square pilings supporting oceanfront structures was measured to be approximately 
10 to 11 inches. Erosion and scour were commonly observed to total 5 to 7 
vertical feet at oceanfront homes in the area from Topsail Beach to North 
Topsail Beach. This erosion and scour, added to the long-term erosion rate of an 
average 1 to 2 feet a year, left many homes unable to withstand the loads 
imposed by flood and wind forces acting simultaneously (see Figure 2-4).

[Begin figures]
Figure 2-3 is a photo showing the determination of localized scour from changes 
in sand color, texture, and bedding.

Figure 2-4 is a photo showing how loss of the frontal dune and the resulting 
erosion and scour left many coastal houses unable to resist wind and flood loads 
acting simultaneously.
[End figures]

Landward Residential Buildings

No evidence of general erosion was observed in the areas around landward 
structures, but evidence of localized scour around pilings and other 
obstructions was plentiful (see Figures 2-5 and 2-6). In general, scour did not 
result in the failure of the piling foundations of landward structures. However, 
scour around the vertical members supporting air conditioner platforms and 
building extensions such as decks, porches, and roof overhangs occasionally 
decreased the ability of the vertical members to withstand flood forces and led 
to their collapse. 

[Begin figures]
Figure 2-5 is a photo showing how overwash of barrier islands generated high-
velocity flows that caused extensive scour adjacent to large objects.

Figure 2-6 is a photo showing how the disruption of velocity flows by large, 
non-breakaway objects generated extensive scour that undermined vertical 
foundation members and slabs-on-grade. 
[End figures]

2.3 Building Foundation Systems

In assessing the performance of structure foundation systems, the BPAT addressed 
a variety of issues related to the performance of oceanfront and landward 
structures: piling and column embedment for structures and their extensions 
(e.g., utility platforms, decks, porches, and roof overhangs), the grade of 
lumber used for vertical foundation members, elevation of structures in relation 
to the flood depth, cross-bracing of vertical support members, and solid 
perimeter foundation walls on continuous footings. The BPAT also assessed the 
performance of foundations under manufactured homes and permanently installed 
RVs.

2.3.1 Piling Embedment for Structural Support

Lack of sufficient embedment of vertical structural foundation members may well 
have contributed to the collapse of over 100 oceanfront residential buildings 
(see Figure 2-7). Of those that did not collapse, many were found to be leaning 
(see Figure 2-8). The majority of these structures met the pre-1986 requirement 
for an 8-foot embedment of pilings and columns (measured from existing grade). 
Many front-row houses were placed near or on the landward slope of the frontal 
dune, where the ground elevations were often 8 to 9 feet m.s.l. As a result, the 
bottoms of the pilings or columns were at approximately 0 feet m.s.l. (see 
Figure 2-9)

[Begin figures]
Figure 2-7 is a photo showing that over 100 oceanfront houses were washed off 
their foundations or completely destroyed. 

Figure 2-8 is a photo showing how many oceanfront houses built prior to current 
(1986) North Carolina State Building Code requirements were found to be leaning 
or destroyed. 

Figure 2-9 is an illustration showing typical collapse mechanism of post-FIRM 
building based on pre-1985 embedment requirements in North Carolina coastal 
areas.
[End figures]

As noted in Section 1.3.1, the North Carolina State Building Code was revised in 
1986 to require that vertical foundation members in erosion-prone areas be 
embedded 16 feet below existing grade or to -5 feet m.s.l., whichever is 
shallower. The 1986 requirement was generally successful in protecting 
structures in areas of low ground elevation, where pilings had to be embedded to 
-5 feet m.s.l. This is significant because most of the buildings undermined by 
erosion were in areas where the ground elevations were low. For structures on 
higher dunes (i.e., where ground elevations exceed 11 feet m.s.l.) the piling 
embedment requirement changes to only 16 feet below grade. This embedment depth 
is not sufficient to allow the pilings to survive a similar storm or continuing 
long-term erosion of moderate to high dunes.

Although post-1986 oceanfront structures generally performed better than 
oceanfront structures built prior to 1986, several foundations supporting 
oceanfront structures that were observed to be leaning were suspected of being 
post-1986 (see Figure 2-10). The remaining embedment depth of the foundation 
members beneath these structures was not determined by the BPAT; however, for 
example, with a pre-storm grade of 11 feet m.s.l., erosion of approximately 6 
vertical feet to an elevation of approximately 5 feet m.s.l., and localized 
scour of an additional 1 vertical foot, the vertical foundation members should 
still have been embedded approximately 9 feet below grade during the height of 
the storm. This depth should have been sufficient to prevent leaning in many 
cases. One possible explanation is that the pilings under these leaning 
structures did not meet the current embedment depth requirement.

[Begin figure]
Figure 2-10 is a photo showing one of several buildings observed to be leaning 
landward that were suspected of having been constructed to current North 
Carolina State Building Code requirements.
[End figure]

To follow up on this issue and investigate the effects of the current North 
Carolina State Building Code requirements on the performance of foundation 
pilings, FEMA contracted with Woodward-Clyde Federal Services (W-C) to determine 
piling embedment depths for oceanfront buildings on Topsail Island, North 
Carolina, where Hurricanes Bertha and Fran damaged a number of structures. Using 
aerial photographs, W-C identified 205 post-1986 oceanfront buildings. Of the 
identified buildings, 92 percent had not sustained any significant foundation 
damage. The remainder had pilings that were damaged or leaning. W-C conducted 
tests to determine the embedment depths of selected pilings under 11 of the 
identified buildings, including 7 leaning buildings, and found that over 80 
percent of the tested pilings did not meet the 1986 embedment requirement. The 
testing procedure and the findings are presented in a separate report prepared 
by W-C. The Executive Summary from the W-C report is contained in Appendix C of 
this report. Recommendations based on W-C’s findings are presented in Section 
3.1.1.

2.3.2 Piling Embedment for Decks, Porches, and Roof Overhangs

Lack of sufficient embedment of vertical foundation members for decks, porches, 
and roof overhangs attached to oceanfront and landward residential buildings 
resulted in the collapse of several hundred of these building extensions (see 
Figure 2-11).

[Begin figure]
Figure 2-11 is a photo showing that the BPAT observed several hundred decks and 
porches that collapsed as a result of insufficient foundation support.
[End figure]

Oceanfront Residential Buildings

Vertical foundation members supporting unroofed decks did not have to meet the 
pre-1986 State Building requirement for 8-foot piling embedment, nor do they 
have to meet the post-1986 requirement for 16-foot embedment. Vertical 
foundation members for covered porches and roof overhangs are supposed to meet 
the criterion applied to the foundation members for the main structure. The BPAT 
found that vertical foundation members for decks, porches, and roof overhangs 
were often embedded to a depth of only 2 to 6 feet below existing grade (see 
Figure 2-12).

[Begin figure]
Figure 2-12 is a photo of an example of building constructed to current North 
Carolina State Building Code requirements with insufficient embedment of 
pilings/columns under two-story deck. 
[End figure]

Decks, porches, and roof overhangs were often built on the seaward side of 
oceanfront structures and were therefore often embedded into the frontal dune 
(see Figure 2-9). With embedments of only 2 to 6 feet into the dune, the bottoms 
of the pilings or columns were often at elevations of 4 to 8 feet m.s.l. The 
remaining embedment depth of those deck, porch, and roof overhang supports that 
survived the hurricane appears to be as little as 1 to 2 feet in many cases. 
Since these supports are usually seaward of main structures, they are subject to 
amounts of storm surge, velocity flow, wave action, vertical erosion, and 
localized scour at least as great as those that affect the main structure (see 
Figure 2-13). 

[Begin figure]
Figure 2-13 is a photo showing that the embedment of deck supports into frontal 
dune was often shallow. After erosion of the dune, the bottom of the support for 
this deck was several feet above grade.
[End figure]

In the areas where decks, porches, and roof overhangs were observed, erosion was 
approximately 7 vertical feet, to an elevation of approximately 4 feet m.s.l. 
Localized scour of an additional vertical foot would result in total loss of 
embedment to an elevation of 3 feet m.s.l. during the peak of the storm (see 
Figure 2-9). When vertical foundation members lost their ability to support the 
structure above, the deck, porch, or roof overhang often collapsed, damaging the 
structure to which it was attached and becoming waterborne debris that was then 
carried into the main structure or nearby structures (see Figure 2-14). This 
damage may have contributed to the collapse of some buildings.

[Begin figure]
Figure 2-14 is a photo showing how storm-generated debris impacted nearby 
structure.
[End figure]

For decks, porches, and roof overhangs to have survived, their supporting 
vertical members would have to have had a post-storm embedment of approximately 
8 feet below grade. The findings of the team regarding decks, porches, and roof 
overhangs are particularly important because it appears that the construction of 
multilevel decks and porches supporting roof overhangs is becoming increasingly 
popular in oceanfront architecture (see Figure 2-15). Usually, these building 
extensions are larger and more complex than required solely for building access.
Landward Residential Buildings

[Begin figure]
Figure 2-15 A shown by this post-Fran photograph taken at Figure Eight Island 
North Carolina, a current architectural trend is the construction of multistory 
decks supporting roof overhangs.
[End figure]

Decks, porches, and roof overhangs supported by vertical foundation members were 
observed to have been installed on many landward homes on barrier islands. In 
general, these building extensions were not protected from localized scour 
caused by velocity flow. The loss of supporting soil due to scour often left 
vertical foundation members of decks, porches, and roof overhangs unable to 
resist the velocity flow, wave action, and debris impact forces that occurred in 
coastal areas (see Figure 2-16). Vertical foundation members were found to not 
be embedded to the same depth as the main building supports. It was reported 
that the North Carolina State Building Code requires vertical supports for the 
main structure outside of a V Zone to be embedded 8 feet below existing grade, 
but that no such requirement was enforced for building extensions such as decks 
and, in some instances, porches and roof overhangs.

[Begin figure]
Figure 2-16 is an illustration of a typical failure of deck/roof overhang and 
air conditioning/heat pump compressor platform for landward home.
[End figure]

2.3.3 Debris Impact on Vertical Foundation Members

Debris observed by the BPAT included 8-inch x 8-inch pilings up to 20 feet long 
(see Figure 2-14), round 6-inch diameter posts, septic tank sections (see Figure 
2-17), materials from collapsed adjacent houses, the remains of collapsed decks 
(from the house impacted and from adjacent and other nearby oceanfront houses — 
see Figure 2-18), and portions of collapsed fishing piers. An extreme example of 
debris impact is shown in Figure 2-19. Although debris impact generally was not 
suspected of causing significant failure of vertical foundation members, it did 
damage foundation cross-bracing, as discussed in Section 2.3.6.

[Begin figures]
Figure 2-17 is a photo showing how a precast concrete ring section of septic 
tank became waterborne debris, impacting building foundation members.

Figure 2-18 is a photo showing how the impact of debris from a damaged deck 
appeared to have broken cross-bracing. 

Figure 2-19 is a photo showing an example of extreme impact — two houses floated 
and pushed into another house.
[End figures]

2.3.4 Grade of Lumber Used for Timber Pilings and Cross-Bracing

To resist coastal flood forces, timber pilings depend largely on their 
dimensions and depth of embedment, but another important factor is the grade of 
lumber used. Lower grades of lumber may have knots, cracks, or other 
imperfections that contribute to failure when the piling is acted on by water 
and debris impact forces. For example, Figure 2-20 shows a failed timber piling 
whose strength was compromised by closely spaced knots. Failures of this type 
were not widely observed by the BPAT but, as indicated by Figure 2-20, are a 
potential problem that can lead to structure failure and even collapse.

[Begin figure]
Figure 2-20 is a photo showing an example of broken piling. The piling broke at 
the location of several knots, where cross-bracing was attached (note remaining 
bolt and piece of bracing).
[End figure]

2.3.5 Elevation of Buildings

NFIP regulations require that structures in Coastal High Hazard Areas (V zones) 
be elevated so that the lowest horizontal structural member of the lowest floor 
is at or above the BFE shown on the FIRM in effect at the time of construction. 
In the areas visited by the BPAT, structures in V zones appeared to have been 
built in compliance with this requirement. For structures in A zones, the NFIP 
regulations require that the lowest floor be elevated to or above the BFE; no 
requirements are imposed for structures in B, C, and X zones. Although elevating 
on open foundations with lowest horizontal structural members at or above the 
BFE is not required outside of V zones, this practice was widely observed in A, 
B, C, and X zones on the barrier islands within the study area (see Figure 2-
21). 

[Begin figure]
Figure 2-21 is a photo showing how the survival of this properly elevated North 
Carolina State Park public rest room demonstrates the State’s commitment to 
proper construction in coastal areas.
[End figure]

Homes in A, B, C, and X zones were often elevated 8 to 9 feet on embedded piling 
foundations to allow below-building parking and storage. This practice 
undoubtedly resulted in less damage than would have occurred if the lowest 
floors of these structures had been elevated only to the BFE in A zones and not 
elevated at all in B, C, or X zones. However, the areas below many of these 
elevated buildings had been enclosed with nonstructural wall panels and were 
being used for living space rather than solely for parking, storage, and 
building access. When acted on by velocity flows, the wall panels often 
collapsed. As a result, the affected buildings incurred extensive nonstructural 
damage.

2.3.6 Cross-Bracing Below Elevated Buildings

The BPAT found widespread damage to 2x cross-bracing, especially below 
oceanfront homes, including braces split along the grain, braces shattered 
across the grain, and pull-through of brace attachment bolts. (The term “2x” 
refers to lumber with nominal dimensions of 2 inches x 8 inches, 2 inches x 10 
inches, etc.) Wave and debris impact appeared to have generated the greatest 
amount of damage. As noted in Section 2.3.3, the debris observed by the team 
included 8-inch x 8-inch pilings and 6-inch diameter posts, septic tank 
sections, and materials from collapsed houses, decks, and fishing piers. These 
types of objects can result in point-loading impacts that generate loads well 
beyond the material strengths of 2x cross-bracing. Although damage was most 
prevalent in areas where extensive debris was observed, no definitive cause and 
effect relationship could be established

Debris was also observed lying against or draped over cross-bracing. When 
exposed to the hydrodynamic loads imposed by flood waters, debris draped over or 
lying against cross bracing increases the drag coefficient and the area of the 
obstruction, thereby increasing the lateral loads transferred to the foundation. 
Although cross-bracing was frequently damaged, this damage did not appear to 
result in damage to the elevated building as long as the pilings were embedded 
deep enough to resist erosion.

2.3.7 Solid Perimeter Masonry Foundation Walls Supported on a Continuous Footing

Solid perimeter masonry foundation walls supported by a continuous footing are 
not a prevalent form of construction on the barrier islands within the study 
area. Where this type of construction was found in areas of high-velocity flow, 
poor building performance was generally observed. High-velocity flood flows 
moving around the perimeter foundation walls generated localized scour that 
propagated to a depth greater than that of the bottom of the continuous footing 
supporting the perimeter foundation wall. Once the soil underlying the footing 
was lost, the footing and foundation wall collapsed, leaving the floor diaphragm 
unsupported (see Figure 2-22). This scenario occurred not just in oceanfront 
areas, but also in areas set back more than 600 feet from the ocean shoreline 
(see Figure 2-23). Even in areas of relatively shallow flooding (1 to 2 feet 
deep) and where deposition of beach sand had occurred, scour and collapse of 
solid perimeter foundation walls was observed.

[Begin figures]
Figure 2-22 is a photo showing the collapse of footing and foundation wall under 
elevated wood-frame building. Collapse resulted because obstruction of flow by 
building caused scour to extend below the bottom of the footing. Note propane 
gas line extending through foundation wall.

Figure 2-23 is a photo showing the catastrophic failure of landward building 
constructed on masonry wall and slab-on-grade foundation. Failure resulted 
because obstruction of flow by building caused extensive scour. Note compressor 
collapsed into scour hole.
[End figures]

2.3.8 Manufactured (Mobile) Home and Permanently Installed RV Foundations

Many manufactured homes and RVs were significantly damaged by Hurricane Fran. 
The vast majority of manufactured homes and RVs were anchored on top of dry-
stack masonry block piers and anchored with metal straps and helical anchors (2 
feet long with 3-inch helical plates) embedded into the sand (see Figure 2-24). 
While most were exposed to relatively shallow flood depths (1 to 3 feet), many 
were moved 50 feet or more laterally and flipped over by wind forces acting 
alone or in conjunction with flood forces (see Figures 2-24 and 2-25).

[Begin figures]
Figure 2-24 is a photo showing how the permanently installed RV overturned as a 
result of anchor pullout. The anchor was 2 feet long.

Figure 2-25 is a photo showing the localized scour beneath pre-storm footprint 
of manufactured home. Note the collapsed dry-stack block foundation and the 
termination of root mat, which otherwise would have helped stabilize the 
adjacent ground.
[End figures]

The team observed depressions from 1 to 2 feet deep left by localized scour 
within the original footprint of the structure (see Figure 2-25). The scour may 
have been caused by numerous factors, including a discontinuity between the 
stabilizing root mat provided by grass surrounding the site and the 
corresponding loss of unprotected sand beneath the home, the creation of a large 
obstruction by the solid skirt surrounding the foundation system, and localized 
scour around the dry-stack masonry piers supporting the structure.

Even units tied down with straps and helical anchors were displaced from their 
foundations because of pier undermining and subsequent collapse, strap failure, 
or anchor pullout. Strap failure may have occurred when the tensile strength of 
the strap was exceeded. Anchor pullout occurred when the resisting force of the 
surrounding soil was exceeded. Both strap failure and anchor pullout occurred in 
several scenarios, which include the following:

- Collapse of the supporting dry-stack masonry foundation due to localized 
scour. When the foundations gave way, the unit fell onto the ground, exposing 
the seaward face to the full force of the velocity flow and debris impact.

- Failure of the strap due to corrosion. Several corroded straps were observed 
to have failed when they were exposed to minimal tensile loading. The coastal 
environment, where salt and moisture are present, can accelerate the rate of 
corrosion. Straps that are exposed to salt spray and that are not periodically 
cleansed by rainfall can lose much of their design tensile strength in a little 
as 3 to 5 years.

- Pullout of the anchor due to soil saturation. All anchors observed had been 
embedded in sand. During flooding conditions, sand can quickly become saturated 
and thereby loose it capacity to resist pullout of the helical anchor plates. 
Because anchors that had pulled out were observed to have small-diameter helical 
plates and shallow embedments, it is assumed that soil saturation played at 
least a contributing role in anchor pullout.

In addition, the use of anchors of the wrong size and the installation of 
anchors not in accordance with manufacturers’ recommendations may have 
contributed to the observed failures.

2.4 Breakaway Walls Below Elevated Buildings

Many of the areas below BFE beneath elevated structures observed by the BPAT had 
been enclosed with wall panels intended to break away under the impact of 
hydrodynamic flood forces. Under the NFIP, this practice is permitted. When 
properly installed, these wall panels break away under the impact of 
hydrodynamic flood forces and therefore do not transfer loads to the foundation 
of the structure and the structure frame. Although the BPAT observed that 
breakaway wall panels generally performed as intended, some problems are worth 
noting. The placement of exterior sheathing of breakaway panels continuously 
over adjacent vertical foundation members, the improper attachment of breakaway 
panels to foundation members, and the improper position of the panels in 
relation to foundation cross-bracing were often found to affect their 
performance. These issues are discussed in the following sections.

2.4.1 Placement of Exterior Sheathing Over Pilings

On some structures, exterior sheathing consisting of oriented strand board (OSB) 
had been installed over breakaway wall panels in such a way that it traversed 
adjacent panels and the faces of intervening vertical foundation members. 
Sheathing installed in this fashion is not in conformance with breakaway wall 
designs recommended by the NFIP. It interferes with the function of the 
breakaway panels because it must fail before the panels can break away (see 
Figures 2-26 and 2-27). The OSB installed across breakaway panels and foundation 
members did not appear to have caused structural damage; however, when acted on 
by flood forces, it can potentially place unnecessary and unanticipated lateral 
loads on vertical foundation members.

[Begin figures]
Figure 2-26 is a photo showing the exterior sheathing of breakaway wall spanned 
piling.

Figure 2-27 is a photo showing how the breakaway wall panel failed to function 
as designed because continuous sheathing was installed across pilings. No 
structural damage was observed; however, note damage to utility components 
installed on breakaway wall panel.
[End figures]

2.4.2 Improper Attachment of Breakaway Wall Panels to Foundation Members

In general, the BPAT observed that breakaway wall panels had been attached to 
structure foundation members with an excessive number of fasteners (nails). The 
BPAT did not observe any instances of structural failure or structural damage 
that appeared to have resulted from this practice. However, when an excessive 
number of fasteners are used between the structural members and the perimeters 
of the breakaway wall panels, the loads necessary to make the panels break away 
increase significantly, far beyond the flood load expected to cause the panel to 
break away.  Another example of improper attachment is shown in Figure 2-28. 
Placing anchor bolts through the sill plate of the breakaway wall panel prevents 
it from breaking away until the forces on it have increased significantly beyond 
those under which the wall is intended to break away.

[Begin figure]
Figure 2-28 is a photo showing how the use of anchor bolts through the sill 
plate of a breakaway wall is improper. Even though this bolt does not have a nut 
and washer, it prevented the wall from breaking away laterally.
[End figure]

2.4.3 Placement of Breakaway Wall Panels Seaward of Cross-Bracing

On a few structures, breakaway wall panels were observed to have been installed 
directly seaward of cross-bracing (see Figure 2-29). When the panels broke away 
under the loads imposed by flood waters, they moved back and came to rest 
vertically against the cross-bracing. As a result, the vertical surface exposed 
to velocity flow, breaking waves, and debris impact increased tremendously and 
so did the corresponding loading on the cross-bracing. For cross-bracing 
installed across a typical 8-foot span between pilings, the resulting loading 
far exceeds the bending moment capacity of 2x or 3x wood braces in the narrow 
dimension. As a result, the cross-bracing often failed.

[Begin figure]
Figure 2-29 is a photo showing how the improper installation of breakaway walls 
(on the seaward side of cross-bracing) resulted in failure of cross-bracing.
[End figure]

2.5 Below-Building Concrete Slabs

With some minor exceptions, below-building concrete slabs generally performed as 
intended. Because some of the exceptions resulted in building damage, they are 
worth noting.

2.5.1 Slab Thickness 

Slabs thicker than 4 inches were observed to have caused two problems: 

-·The thicker the slab, the greater the force needed to break the slab and 
therefore the greater the load transferred to the building foundation system 
until the slab breaks free of the foundation (see Figure 2-30).

[Begin figure]
Figure 2-30 is a photo showing an example of unnecessarily thick slab.
[End figure]

-·The thicker the slab the more it weighs per square foot of surface area. When 
a thicker slab breaks apart, the sections weigh more than those of the same size 
from a thinner slab and they create greater impact loads when they are thrown up 
against the building foundation by velocity flow and wave action.

2.5.2 Slab Joints 

Three general types of joints are used in concrete slabs under elevated 
buildings: tooled and sawcut contraction (crack control), expansion, and 
isolation:

- Contraction joints are cut into the surface of the slab after the slab is 
poured and floated level. The joints become vertical planes of weakness that are 
intended to control cracking. These planes of weakness can serve a dual purpose 
by creating a frangible slab, since they are also the planes along which the 
slab is expected to break during a coastal erosion and scour event such as a 
hurricane or Nor’easter.

- Expansion joints are set in place before the slab is poured. They separate 
independent slab sections and are filled with a compressible material that 
allows the sections to expand and contract in response to changes in 
temperature. They also help create a frangible slab.

- Isolation joints are used to separate the slab from structural members, such 
as vertical foundation members, and other slab penetrations. These joints are 
similar to expansion joints in that they are filled with a compressible material 
and are set before the slab is poured. When used in frangible slabs, they help 
ensure that the slabs will break away cleanly from the vertical foundation 
members and other slab penetrations such as sewer riser pipes.

The problems associated with slab joints involved the number of contraction 
joints and the effect of reinforcing wire mesh. In the slabs beneath many 
structures, the number of contraction joints was observed to be insufficient to 
make the slabs frangible. When the slabs broke up, the pieces were too large and 
generated unnecessary impact loads on the foundation system. Occasionally, the 
lack of an adequate number of contraction joints prevented the slab from 
breaking up. Figure 2-31 shows a large, unbroken section of a below-building 
slab-on-grade that was flipped up, probably by wave action, and came to rest 
against two vertical foundation members. The flipped slab created an obstruction 
that increased the flood loads on the foundation. In fact, the vertical 
foundation members behind the slab in Figure 2-31 were found to be leaning 
landward.

[Begin figure]
Figure 2-31 is a photo showing how a concrete slab-on-grade flipped up, probably 
by wave action, came to a rest against two foundation members, generating large, 
unanticipated loads on the foundation. Note slab is supported on concrete 
collars around piles and contains wire mesh.
[End figure]

2.5.3 Wire Mesh

The BPAT observed that wire mesh was used in most slabs. The mesh is laid out 
before the concrete is poured. It usually extends across contraction joints but 
usually does not extend across expansion joints. The BPAT observed that where 
wire mesh was present, it usually had been installed improperly (i.e., at the 
bottom rather than in the middle of the slab). Even so, the team observed that 
when reinforced slabs broke apart, broken sections were held together by the 
wire and came to rest against, or became wrapped around, vertical foundation 
members. As a result, large, unanticipated loads were transferred to the 
foundation system.

2.5.4 Connecting the Slab to the Vertical Foundation Members

Some engineers and architects may have specified, or contractors chose to use, 
dowels to connect concrete slabs-on-grade to vertical foundation members. The 
dowels, intended to help prevent the differential settlement of the slabs, were 
inserted into or through the vertical foundation members before the slab was 
poured. They caused serious problems when the slabs broke apart under flood 
loads. The dowels made it more difficult for the slab to break into small pieces 
and separate cleanly from the vertical foundation members. Even when the slabs 
broke into small pieces, the dowels acted like pins in a hinged connection, 
keeping the slab connected to the vertical foundation members (see Figure 2-32). 
As a result, unnecessary and unanticipated flood loads were transferred to the 
vertical foundation members.

[Begin figure]
Figure 2-32 is a photo showing the use of steel dowels to tie slabs to vertical 
foundation members prevented the proper breakup of slab. A portion of the slab 
was left attached to the piling, resulting in large, unanticipated loads on 
foundation members, Note leaning vertical member, on right.
[End figure]

Although the BPAT was unable to define a cause-and-effect relationship, several 
buildings with this slab-to-foundation pin detail were found to be leaning. The 
team members believe that the inability of the slab to break free of the 
vertical foundation members was at least partially responsible the failure of 
the vertical foundation members to remain plumb.

2.5.5 Casting Concrete Grade Beams and Slabs-On-Grade Monolithically 
Some concrete grade beams and slabs-on-grade were poured monolithically in a 
continuous concrete pour. As a result, large loads were transferred to the 
foundation system when velocity flow, breaking waves, and debris forces were 
applied to the sections of slabs attached to the grade beam. Although the BPAT 
was unable to define a cause-and-effect relationship, several buildings with 
this monolithic grade beam and slab detail were found to be leaning. The team 
members believe that the inability of the slab to break free of the vertical 
foundation members and grade beam was at least partially responsible for the 
failure of the vertical foundation members to remain plumb.

2.5.6 Concrete Collars 

The BPAT observed that concrete collars were often poured around foundation 
pilings in conjunction with the construction of below-building concrete slabs 
(see Figures 2-14 and 2-31). Although intended to provide stability, these 
collars presented a large obstruction to flow, thereby increasing flood loads 
on, and scour around, the pilings to which they were attached. The increased 
scour resulted in a loss of sand supporting the foundation (see Figure 2-31). As 
shown in Figure 2-14, collars did not prevent piling failure.

2.6 On-Site Utility Systems

Building performance issues concerning on-site utility systems usually involved 
air conditioning / heat pump compressors and their supporting platforms, the 
placement of utility system components in relation to breakaway wall panels and 
vertical foundation members, and septic tanks.

2.6.1 Air Conditioning / Heat Pump Compressor Platforms

The majority of air conditioning and heat pump compressor platforms supported by 
posts collapsed when acted on by flood forces. The posts were generally observed 
to be embedded only 1 to 2 feet. Once platforms were dislodged, they often 
collapsed, leaving compressors submerged in flood water (see Figure 2-23. Many 
compressors were observed to have become waterborne debris. Occasionally, when a 
platform survived, the compressor was insufficiently elevated and was inundated 
with salt water and sand. Once inundated, the compressor is no longer 
salvageable and must be replaced.

Oceanfront Residential Buildings

When post-supported platforms adjacent to oceanfront houses collapsed, the cause 
was almost always erosion and scour combined (see Figure 2-33). Erosion in the 
areas where the platforms were set was generally 2 to four 4 feet in depth, 
often exceeding the embedment depth of the support posts. Cantilevered 
platforms, which do not depend on vertical support members, escaped the scour- 
and erosion-induced damage incurred by post-supported platforms. Compressors 
installed on adequately elevated cantilevered platforms were not subject to 
flood-related damages, including inundation, but were still subject to wind 
damage when they were not adequately anchored (see Figure 2-34).

[Begin figures]
Figure 2-33 is a photo showing an air conditioning/heat pump compressor platform 
leaning because erosion and scour caused loss of one supporting column.

Figure 2-34 is a photo showing that although the air conditioning/heat pump 
compressor platform on this oceanfront house was adequately elevated with 
cantilever bracing, the compressor was almost pushed off the platform by wind 
because of the lack of the necessary attachment.
[End figures]

Landward Residential Buildings

Many of the post-supported compressor platforms adjacent to homes landward of 
the oceanfront row collapsed as a result of localized scour due to velocity 
flows (see Figures 2-23 and 2-35). Localized scour of approximately 1 foot in 
depth was generally observed. Where posts were embedded only 1 to two 2 feet, 
localized scour allowed the velocity flow and debris impact forces to dislodge 
the platform. Platforms with properly embedded posts performed much better. 
Often, the only damage sustained by landward residential buildings was the loss 
of the compressor unit due to water or wind damage.

[Begin figure]
Figure 2-35 is a photo showing an air conditioning/heat pump compressor platform 
leaning because scour caused loss of two supporting posts.
[End figure]

2.6.2 Placement of Utilities on, Through, or Adjacent to Breakaway Wall Panels

In the vast majority of structures with breakaway wall panels observed by the 
BPAT, utilities were improperly placed on, through, or adjacent to breakaway 
wall panels.

The BPAT observed electric meter boxes, telephone service boxes, cable TV boxes, 
sewer service lines, and domestic water service feeds all mounted on breakaway 
wall panels (see Figures 2-27 and 2-36). Utilities placed through breakaway wall 
panels included telephone and cable TV lines, the electric feed from the back of 
the meter box to the electric panel box, and water service feeds (see Figure 2-
37). Under the effects of flood forces, these utilities either were torn out or 
prevented breakaway wall panels from breaking away cleanly. Another deficiency 
observed was the placement of utilities adjacent to or near breakaway wall 
panels (see Figure 2-38). These utilities were damaged when flood forces caused 
the panels to break away.

[Begin figures]
Figure 2-36 is a photo showing an example of utility component (electric panel 
box) installed on a breakaway wall panel. 

Figure 2-37 is a photo showing utility components (dryer vent, air conditioning 
compressor feed line, main electrical service line out of meter box, and 
electric wiring) penetrating breakaway wall panel. 

Figure 2-38 is a photo showing how these utility components (wiring, electric 
panel box, ductwork and sewer line) were installed adjacent to breakaway wall 
panel and were damaged when the walls broke away. Note broken sewer line riser 
pipe on seaward side of piling.
[End figures]

2.6.3 Placement of Utilities Adjacent to Vertical Support Members

Utilities on the vast majority of structures were found to be in locations 
exposed to velocity flow and debris impact, e.g., mounted to vertical foundation 
members on sides other than the landward side (see Figure 2-38). Utilities 
installed on the landward side of vertical foundation members generally survived 
since the foundation member shielded them from velocity flow and debris impact.

2.6.4 Septic Tanks

Septic tanks installed near oceanfront homes were often left exposed by storm-
induced erosion and scour (see Figure 2-39). Occasionally, thin-walled septic 
tanks made of precast concrete rings were observed to have become waterborne 
debris. Concrete ring sections were found dislodged and under elevated 
structures (see Figure 2-17). When a tank was exposed, the sewer line from the 
home was usually severed. On many exposed tanks there were openings where the 
access lid was missing and where the connection to the sewer line from the house 
was exposed when the pipe broke away. The openings allowed sewage to leak out 
and flood water and debris to enter the tank. Homes that were otherwise not 
significantly damaged had been posted “Unoccupiable” by the local building 
official because of the lack of an operating sanitary disposal system.

[Begin figure]
Figure 2-39 is a photo showing how oceanfront erosion and scour unearthed and 
damaged septic tanks and systems. Note precast ring section and precast tank.
[End figure]

The State of North Carolina has established regulations concerning the 
installation of septic tanks and leach fields in areas subject to coastal flood 
hazards. When a new building is constructed, the tank and leach field must be 
installed on its landward side. When repairs to an existing septic system 
located on the seaward side of a building become necessary for any reason, the 
tank and leach field must be moved to the landward side of the building.

2.7 Dry Floodproofing in Coastal A Zones

One unusual building in Wrightsville Beach is worth noting. The building is a 
slab-on-grade hotel that was renovated several years ago. During the renovation 
of the conference rooms, the owner reconstructed the exterior walls to make them 
watertight and installed tracks in the door openings for the placement of 
removable flood shields (see Figure 2-40). The flood shields were approximately 
3 feet high and were manufactured by a firm that specializes in producing flood 
shields. Each shield was equipped with a pneumatic gasket that could be inflated 
to seal the gap between the track and the edge of the shield (see Figure 2-41). 
Construction of a solid masonry wall inside the exterior walls of the conference 
rooms completed the floodproofing.

[Begin figures]
Figure 2-40 is a photo showing an engineered flood shield installed over opening 
to large dry floodproofed commercial building.

Figure 2-41 is a photo showing how buildings that are floodproofed require 
extensive engineering detailing. Note that this flood shield is sealed with a 
pneumatic gasket.
[End figures]

The BPAT interviewed the hotel manager and engineer during the site 
investigation. Both were directly involved in the renovation of the hotel. Both 
said that the floodproofing was worth the effort but that they wished the 
shields had been 1 foot higher. While the shields prevented the storm surge from 
entering the protected area, some water splashed over the tops of the shields. 
The water that passed over the top flooded the conference rooms to a depth of 4 
inches. It damaged the carpeting and mildewed the wall paper. However, in the 
remaining portions of the hotel, which were not floodproofed, the depths of 
flood waters reached 18 inches. In these areas, the water damaged the sheetrock 
and left over 2 inches of sand on the floor. Both the manager and the engineer 
stated that keeping sand out of the conference area was, in itself, 
justification for the expense of the floodproofing.

Although flood shields and other elaborate floodproofing measures can be quite 
effective, they often require extensive human intervention to function properly 
(see Figure 2-42). It should be noted that very few floodproofed buildings in 
coastal A zones are known to be subject to wave action. Floodproofing in areas 
known to be subject to wave action presents special challenges that must be 
addressed in the design, installation, and operation of the components of the 
floodproofing system.

[Begin figure]
Figure 2-42 is a photo showing now dry floodproofing often requires extensive 
human intervention. Note the detailed instructions affixed to this flood shield
[End figure]

2.8 Wind Damage

Although the BPAT focused on flood damage, the team also observed wind damage to 
many buildings. Porch roofs and large overhangs often failed because of poor 
connections, particularly base and roof connections of support columns. Porch 
and overhang failures often caused severe damage to otherwise well-connected 
main roofs. An additional contributor to observed wind damage was the failure of 
corroded metal connectors.

2.9 Corrosion of Structural Metal Components

The BPAT continued to see a trend toward the increased use of partially exposed 
metal structural components, such as hurricane straps and clips, stamped metal 
plates on floor diaphragm trusses, and  manufactured home and RV tiedowns (see 
Section 2.3.8), in coastal structures. With this trend, comes a trend toward an 
increase in the observed corrosion of these components. Components that are 
partially exposed, i.e., those that are exposed to direct contact with ambient 
exterior air but not cleansing rainfall, continued to show the highest rate of 
corrosion.

Examples of such components are shown in Figures 2-43, 2-44, and 2-45. The 
presence of rust indicates an obvious loss of galvanization on metal components 
observed in both oceanfront and landward structures in the study area and may 
indicate that the connectors are nearing the end of their useful life.

[Begin figures]
Figure 2-43 is a photo showing that the corrosion of galvanized floor truss 
plates was observed in many buildings. 

Figure 2-44 is a photo showing that the corrosion of galvanized floor truss 
plates was observed in buildings. Note that painting does little to slow the 
process of corrosion in coastal environments.

Figure 2-45 is a photo showing that the corrosion of hurricane straps and steel 
plates was also observed in many locations.
[End figures]

As noted in Section 2.8, the team observed wind damage to some structures that 
was due in part to the failure of corroded metal connectors.

2.10 Concerns Regarding the Effective FIRMs for North Carolina Coastal 
Communities

Throughout the damaged oceanfront area, the effective FIRMs for the affected 
communities do not account for the effects of dune erosion, wave setup, or wave 
runup. Prior to Hurricanes Fran and Bertha, the V-zones were located on the 
ocean beach, well seaward of building locations. Oceanfront dunes were 
identified as B and C zones, outside the influence of 100- and 500-year 
flooding. Fran caused severe erosion in the oceanfront row of buildings and 
allowed waves greater than 3 feet high to extend several rows of buildings 
farther landward. Smaller waves swept the entire barrier island in many 
locations. 

In addition to the false sense of safety that results when erosion is neglected, 
the FIRM deficiencies allowed finished underhouse enclosures in oceanfront B and 
C zones to be constructed on slab foundations not supported by the piling 
foundation. In areas like Kure Beach, erosion and wave damage caused significant 
damage to the finished enclosures. Without the application of FEMA models for 
dune erosion, wave setup, and wave runup, the wave model used in the preparation 
of the existing FIRMs underestimates the wave heights above the stillwater 
elevations. This results in BFEs that are lower than those needed to avoid wave 
damage to coastal construction.

[Editor’s note: FEMA Region IV in Atlanta has issued advisory flood hazard maps 
for several communities severely impacted by Hurricane Fran and has begun the 
preparation of revised FIRMs. The communities have adopted the advisory maps and 
will use them until FEMA issues the revised FIRMs.]

2.11 Building Performance Successes

The shift in construction from low floor elevation with shallow footings to 
elevated piling foundations with underhouse parking, as described in Section 
1.3.1, significantly reduced the flood damage to buildings landward of any 
erosion. Where not subject to erosion, second row and more landward buildings on 
pilings consistently survived wave heights of at least 3 feet and overwash 
deposition of up to 3 feet of sand under the building. The use of 8-inch x 8-
inch square pilings embedded 8 feet below grade was successful in protecting 
landward three-story buildings elevated up to 10 feet above grade. Outside of 
areas impacted by erosion, the BPAT did not observe a single piling failure. 
Since most landward building sites in the beach communities are not subject to 
erosion, the piling standards initiated by the by the State Building Code in the 
1960’s were extremely successful. However, underhouse non-load-bearing 
enclosures below the elevated floor were regularly flooded and when used as 
finished living space were often severely damaged.

The shift in the State Building Code to require longer pilings for erosion-prone 
buildings along the ocean was generally successful. As noted by W-C (see 
Appendix C), of the 205 post-1986 oceanfront structures on Topsail Island, over 
90 percent sustained no significant foundation damage. Only a few were seriously 
damaged or destroyed. In comparison, adjacent oceanfront houses on shallow 
pilings were often destroyed when the foundation was undermined by erosion. The 
current requirement is for piling embedments to -5.0 feet m.s.l. or 16 feet 
below grade, whichever is less. As noted in Section 2.3.1, the natural grade on 
most of the eroded lots was relatively low, allowing the -5.0-foot m.s.l. 
requirement to control the design. In those conditions, the piling standard was 
usually adequate. However, on higher dunes where the requirement for 16 feet 
below grade applies, the required depth is too shallow to keep buildings stable 
after erosion of the dune and beach profile.

It is also important to note that the BPAT observed few situations in which the 
performance of breakaway walls below an elevated building may have resulted in 
structural damage to the building. Successful building performance during 
Hurricane Fran also demonstrates the value of compliance with elevation and 
setback requirements, the use of flood-resistant construction materials and 
techniques, such as in engineered concrete buildings, and compliance with other 
coastal design and construction requirements (see Figures 2-46, 2-47, and 2-48).

[Begin figures]
Figure 2-46 is a photo showing that while this house experienced 6 or more feet 
of vertical erosion and scour, as well as the loss of breakaway wall panels, the 
foundation and superstructure performed as designed.

Figure 2-47 As shown by this post-Fran photograph taken at Emerald Isle, North 
Carolina, proper elevation and setback from the oceanfront in conjunction with 
substantial protection afforded by dunes resulted in little or no	flood damage 
to the oceanfront row of buildings.

Figure 2-48 is a photo showing how this large, engineered oceanfront structure 
performed as designed.
[End figures]

Beach nourishment with construction of a hurricane protection dune substantially 
reduced damage in Wrightsville Beach and Carolina Beach. In these areas, the 
manmade dune eroded but prevented erosion failures and reduced wave damage to 
structures. Such dunes are considered expendable but require periodic 
maintenance and replacement after the worst storms.






3 RECOMMENDATIONS

3.1 Building Foundation Systems

3.1.1 Piling Embedment for Structural Support

It is critical that coastal foundations be designed to survive the anticipated 
amount of erosion and scour. Erosion and scour combine to impact coastal piling 
foundations in three distinct ways. First, in the absence of cross-bracing, the 
loss of soil adjacent to a thin vertical foundation member results in a longer 
unsupported length. The increase in unsupported length allows for greater 
deflection of the vertical member. Second, the loss of soil adjacent to pilings 
leaves less soil to counteract lateral loads applied to the pilings by the 
structure above, the velocity flow of the storm surge, wave action, and debris 
impact. Third, pilings, which rely on friction between the piling and the 
adjacent soil to transfer loads into the ground, lose some of the resisting 
friction when the adjacent soil is eroded and scoured. The loss of friction 
reduces the ability of the piling to resist uplift loads from wind.

For the concerns discussed above to be adequately addressed, designs of coastal 
foundation systems must account for the conditions expected to occur during the 
base flood (100-year flood) and long-term erosion for the life of the building. 
The following documents offer guidance to designers of coastal foundations:

- The ASCE standard ASCE 7-95, Minimum Design Loads Building and Other 
Structures. The 1995 version of this standard includes, for the first time, 
criteria for determining flood loads (Chapter 5) and for combining flood and 
other loads to determine combined load factors (Chapter 2) for buildings that 
experience simultaneous wind and flood loads. This standard is available from 
ASCE. Portions, if not the whole standard, may well be incorporated by the model 
building code organizations into future versions of model building codes. This 
standard also meets, or exceeds, the minimum requirements of the NFIP for 
determining loads.

- The ASCE standard Flood Resistant Design and Construction Practices. This 
standard is currently in development and should be completed in 1997. It will be 
available from ASCE. This new standard will provide descriptive as well as 
prescriptive requirements regarding the design and construction of buildings 
that are to be located in floodprone areas. The draft of this standard presents 
the recommendations of some of the Nation’s leading experts in coastal 
construction and meets or exceeds the minimum requirements of the NFIP.

- FEMA’s Coastal Construction Manual (FEMA 55). This document provides further 
guidance on coastal foundation systems. It recommends that pilings be embedded 
to a depth of -10 feet m.s.l. 

- FEMA’s Technical Bulletin No. 5, Free of Obstruction Requirements for 
Buildings Located in Coastal High Hazard Areas. This document provides 
information on NFIP-compliant design and construction practices that can prevent 
damage to coastal buildings caused by below-building obstructions.

The BPAT recommends that in the absence of State or local requirements based on 
detailed engineering studies or the historical performance of coastal buildings 
subjected to base flood conditions, pilings for structures in areas subject to 
erosion and scour be embedded to -10 feet m.s.l. In areas with terminating 
strata that inhibit erosion and scour, and areas with rocky shorelines, 
foundation systems with shallower depths may be justified.

The findings of the W-C study suggest that in coastal areas of North Carolina 
where the ground elevations are at or below 11 feet m.s.l. the current 
requirements of the State Building Code regarding piling embedment depth may 
have been adequate to prevent foundation failure during Hurricane Fran. However, 
given the limitations of that study (as discussed in Appendix C), and the 
anticipated revisions to BFEs for North Carolina coastal communities as 
discussed in Section 2.10 of this report, it may be appropriate for the state to 
consider the need for piling embedment depth requirements more stringent than 
those in its current Code. These issues are discussed below.

All of the structures tested by W-C were located where the ground elevation was 
11 feet m.s.l. or less. To meet the applicable code requirement, the pilings for 
these structures had to be embedded to a minimum of -5 feet m.s.l. In areas 
where the grade is greater than 11 feet m.s.l., pilings will meet the State Code 
if they are embedded 16 feet below average original grade. As a result, the 
bottoms of the pilings in these areas will be above -5 feet m.s.l. This 
embedment depth may not be sufficient for structures built on or directly behind 
frontal dunes, where extensive erosion and scour can cause the loss of the 
entire dune or remove so much of it that piling support becomes inadequate. 
There are areas within the state where this situation can occur. For structures 
in these areas, the state may want to consider revising the Code to require that 
piling embedment depth be -5 feet m.s.l. or 16 feet below grade, whichever is 
greater.

As noted in Section 2.10, FEMA is preparing revised FIRMs for several North 
Carolina communities affected by Hurricane Fran. Because the revised FIRMs are 
expected to include BFEs higher than those shown on the current FIRMs, new 
structures and substantially damaged structures that are reconstructed will have 
to be elevated higher to meet minimum NFIP requirements. As a result, these 
structures may be subject to different wind and flood loads than structures 
built previously, when lower elevation requirements applied, and increased 
piling embedment may be necessary.

In addition, the findings of the W-C study suggest that the State of North 
Carolina needs to emphasize the importance of inspection and code enforcement to 
ensure that all structure pilings meet current and any future Code requirements.

3.1.2 Piling Foundations for Decks, Porches, and Roof Overhangs

The design criteria for vertical foundation members for building extensions such 
as porches, decks, and roof overhangs must be equal to those for the foundation 
system of the main structure. 

Oceanfront Residential Buildings

The vertical foundation members for decks, porches, and roof overhangs must be 
designed and constructed to maintain their ability to support the structure 
above. These building extensions are often on the ocean side of oceanfront 
structures, where they are exposed to amounts of storm surge, velocity flow, 
wave action, vertical erosion, and localized scour at least as great as those 
that affect the main structure. The foundation requirements for these building 
extensions should never be less stringent than those for the building itself. 
Because of the damage caused to the main structure when building extensions 
collapse and the debris they generate once they collapse, they should not be 
considered sacrificial. The only exceptions would be stairways and narrow 
walkways required for building access.

In areas subject to erosion and scour, embedment of vertical foundation members 
for building extensions should be based on a depth related m.s.l., not a depth 
below existing grade. The BPAT recommends that in these areas, the requirements 
applied to the main building support system also be applied to the piling 
foundations for decks, porches, and roof overhangs. This recommendation is based 
on the observed performance of coastal buildings in many hurricanes, including 
Hurricane Fran. 

Landward Residential Buildings

In landward areas also, vertical foundation supports for building extensions 
such as decks, porches, and roof overhangs should meet the same requirements 
applied to the main building support system. In areas not subject to erosion but 
subject to localized scour, embedment of these foundation members may be based 
on a depth related to existing grade. From the observed performance of landward 
coastal buildings and building extensions, the continued use of an embedment 
depth of 8 feet is recommended for all vertical foundation members.

3.1.3 Grade of Lumber Used for Timber Pilings and Cross-Bracing

The requirements of the North Carolina State Building Code regarding the quality 
of wood materials used in the construction of residential building foundations, 
including wood piles and dimensional lumber, are set forth in Section R-309 of 
the Code. Section R-309 specifically addresses protection against decay. It 
states that all pressure-treated wood foundation members must bear the quality 
mark of a quality control inspection agency for pressure-treated wood that has 
been approved by the State Building Code Council. It does not contain any 
explicit requirements regarding grades of lumber used for vertical foundation 
members or cross bracing. Contractors and inspectors must therefore depend on 
their judgment in determining acceptable grades of lumber for these 
applications. To help avoid situations in which foundation strength is 
compromised by the use of lower-quality wood materials (as depicted in Figure 2-
20), the State should consider amending the Code to include prescriptive 
language regarding the grade of wood materials used for vertical members and 
cross bracing in structure foundations.

3.1.4 Proper Elevation of Coastal Buildings

Buildings constructed in Coastal High Hazard Areas (V zones — as shown on NFIP 
FIRMs) must be elevated so that the lowest horizontal structural member of the 
lowest floor is at or above the BFE and the area below the building is free of 
obstructions. These requirements are intended to allow velocity flows and waves 
to pass freely under the building. When a building in a V zone meets these 
requirements, its lowest floor is usually 12 inches or more above BFE, because 
of the thickness of the floor diaphragm and the supporting beams. FIRMs for 
coastal communities usually show A zones landward of V zones. In A zones, a 
building’s lowest floor must be elevated to or above the BFE, and areas below 
the BFE may contain obstructions. In B, C, and X zones, no elevation 
requirements apply.

The practice of elevating buildings on open foundations and ensuring that the 
lowest horizontal structure member is above the BFE was widely used in A, B, C, 
and X zones on the barrier islands impacted by Hurricane Fran. Homes in these 
zones were often elevated 8 to 9 feet on embedded piling foundations to allow 
below-building parking and storage. This practice undoubtedly resulted in much 
lower damages than would have occurred if the lowest floors of these homes had 
been elevated to the BFE in A zones and not elevated in  B, C, and X zones. The 
practice of requiring V zone construction standards in coastal A zones exceeds 
NFIP minimum requirements. Coastal A zones subjected to velocity flow and wave 
action are not distinguished from other A zones on FIRMs. Communities on barrier 
islands may want to consider adopting stricter standards in areas of known high 
coastal hazard, whether or not they are currently identified as V zones.
A building elevated so that its lowest floor is above BFE can qualify for an 
extremely favorable flood insurance rate. Rates for buildings in both Zone A and 
Zone V are lowered incrementally for each 1-foot increase in the height of the 
lowest floor above the BFE, up to a maximum of 4 feet. For buildings at this 
height, the rate reduction is 33 percent in Zone A and 60 percent in Zone V. 
Rate reductions are justified because a building that exceeds the minimum 
elevation requirements generally has a low risk of being significantly damaged 
by flooding. Therefore, in the design process for a new building, elevating 
above the BFE is well worth considering. Local insurance agents can provide 
information concerning the flood insurance rates associated with elevating above 
the BFE.

3.1.5 Cross-Bracing Below Elevated Buildings

Whenever possible, piling foundations should be designed to withstand 
simultaneous wind and flood loads without the use of cross-bracing. Alternatives 
to cross-bracing that may provide the necessary stability include incorporating 
a structural, unroofed deck into the building design to increase the building 
footprint; using pilings that are larger, longer, or both; and reducing piling 
spacing. When cross-bracing is necessary, its use should be minimized to the 
extent possible, especially where it would be perpendicular to velocity flow, 
wave action, and debris impact. Whenever cross-bracing is used, regardless of 
its orientation, it must be designed to withstand the anticipated wind and flood 
loads.

3.1.6 Solid Perimeter Masonry Foundation Walls Supported on a Continuous Footing

The use of solid perimeter foundation walls in coastal flood hazard areas should 
be scrutinized carefully. Since these foundations create large obstructions to 
velocity flow and are usually backfilled with the native sandy material, they 
are extremely susceptible to extensive localized scour. This condition was 
observed to have occurred even in overwash areas where accretion of sand 
occurred.  Scour occurred where the velocity flow was disrupted at the seaward 
face of the obstruction and where the flow reconverged at the landward face of 
the obstruction. 

Only where an engineering analysis of potential scour has been completed by a 
professional engineer should coastal communities consider allowing solid 
perimeter wall foundation systems in landward areas subject to high-velocity 
flow. Engineering solutions to this problem could include backfilling the 
foundation excavation with soil that is resistant to scour and installing the 
footing at a depth that is below the expected depth of scour (see Figure 3-1). 
The preferred solution is to construct piling foundations in landward areas 
subject to high-velocity flow. Of course, solid perimeter walls should never be 
used in oceanfront areas and are not permitted in V Zones.

[Begin figure]
Figure 3-1 is an illustration showing deepening foundation to account for scour.
[End figure]

3.1.7 Manufactured (Mobile) Home and Permanently Installed RV Foundations

Manufactured (mobile) homes and permanently installed RVs are usually supported 
on dry-stack masonry foundations. When manufactured homes and RVs are installed, 
steps should be taken to protect them from the damage caused by foundation 
collapse due to scour, anchor strap failure due to corrosion, and anchor strap 
pullout due to the use of the wrong size or type of anchor.

Protecting the foundation from localized scour requires either controlling scour 
or providing a foundation that extends to a depth that is below the expected 
depth of scour. For example, controlling scour may involve excavating the area 
under the footprint of the home and replacing the excavated soil with a non-
scourable soil or installing a geotextile fabric beneath the home. If a fabric 
is installed, it must be keyed-in around the edges so that the scour will not 
undercut the fabric. Providing a deep foundation can be done at minimal expense. 
A power-driven auger can be used to drill holes to a depth that exceeds the 
depth of the scour. The augured holes should be at least 1 foot in diameter and 
should exceed the expected scour depth by at least 1 foot. The deep foundation 
can consist of a wood post or cast-in-place concrete footing.

Metal straps used to tie down manufactured homes and RVs can become loose over 
time. In coastal areas, metal straps are also subject to higher rates of 
corrosion. Loose or corroded straps exposed to the loads imposed by high winds 
and flood waters are prone to failure, which can lead to damage to the 
foundation and home. It is critical that straps be checked periodically for both 
proper tension and corrosion. Checks for tension should be made frequently 
throughout the hurricane season and are especially important after high-wind 
events. All loose straps should be tightened according to the manufacturer’s 
specifications. Straps that show visible signs of corrosion should be replaced. 
In coastal areas, metal straps have been shown to exhibit signs of corrosion 
after 3 to 5 years of exposure.

Pullout of anchors to soil saturation can be minimized through the use of the 
proper size and type of anchor. Anchors used in sandy soils prone to saturation 
by flood waters must either be long enough that the helical plates extend to a 
depth below the saturated soil, into soil that can resist anchor pullout, or be 
designed to work properly in saturated soil. In coastal areas subject to high-
velocity flow that have loose to medium-dense sands and other granular soils, 
anchors should be  at least 4 feet long and 3/4 inch in diameter and should have 
helical plates at least 6 inches in diameter. Major anchor manufacturers provide 
guidelines for selecting and installing the appropriate anchor according to the 
size of the home, the soil type at the installation site, and other conditions 
that can affect anchor performance. The manufacturer’s specifications and 
recommendations should always be followed in the selection and installation of 
anchors.

To minimize the impact of tidal surge on manufactured homes and their 
foundations, it is recommended that in all Special Flood Hazard Areas subject to 
coastal flooding, including A zones, manufactured homes be elevated such that 
the bottom of the lowest horizontal structural member of the lowest floor is at 
or above the BFE. This means that the bottoms of the chassis I-beams would be at 
or above the BFE.

3.2 Breakaway Walls Below Elevated Buildings

When an area below an elevated building is enclosed with breakaway wall panels, 
special care should be taken to ensure that the placement and attachment of the 
panels does not interfere with the ability of the panels to break away when 
acted on by flood forces. 

3.2.1 Placement of Exterior Sheathing Over Pilings

Exterior sheathing attached to breakaway wall panels must not extend over the 
vertical foundation members. This also applies to wire mesh used in exterior 
stucco systems. There should be a free and clear joint between the panels and 
vertical foundation members so that unnecessary lateral loads are not 
transferred to the foundation (see Figures 3-2 and 3-3). FEMA’s Coastal 
Construction Manual provides guidance on how to fabricate wood-frame, metal-
frame, and masonry breakaway panels. 

[Begin figures]
Figure 3-2 is an illustration showing recommended practice for breakaway wall 
sheathing.

Figure 3-3 is an illustration showing recommended breakaway wall attachment.
[End figures]


3.2.2 Improper Attachment of Breakaway Wall Panels to Foundation Members

The attachment of breakaway wall panels to surrounding surfaces should be done 
in such a way that the panels will safely resist wind loads but break away under 
coastal flood loads (see Figure 3-3). Local building codes provide information 
on applicable wind loads that building components must resist. FEMA’s Coastal 
Construction Manual provides guidance on how to properly attach breakaway wall 
panels.

3.2.3 Placement of Breakaway Wall Panels Seaward of Cross-Bracing

As a matter of practice, breakaway walls should never be installed immediately 
seaward of cross-bracing. This practice exposes the cross-bracing to lateral 
loads that far exceed those that 2x bracing is able to resist. In most 
instances, an alternative is to install latticework or move cross-bracing away 
from breakaway wall panels, into an interior area.

3.3 Below-Building Concrete Slabs

When a slab-on-grade is constructed below an elevated building, it should be 
designed and constructed in such a way that it will not damage the building 
foundation when acted on by flood forces. Issues requiring special consideration 
include the thickness of the slab, slab joints, and construction practices that 
are not appropriate for coastal flood hazard areas subject to erosion and scour.

3.3.1 Slab Thickness 

Slabs below elevated buildings in areas subject to erosion and scour should be 
no thicker then 4 inches. Thicker slabs present two problems: they are harder to 
break into small pieces and each piece weighs more per unit of surface area than 
a same-sized piece of a thinner slab.

3.3.2 Slab Joints 

Of the three types of joints described in Section 2.5.2, contraction joints are 
the most important for ensuring the frangibility of below-building slabs. As 
shown in Figure 3-4, contraction joints should be cut into the surface of the 
slab from piling to piling in both directions across the entire slab. Expansion 
and isolation joints should be installed as appropriate in accordance with 
standard practice or as required by State and local codes.

[Begin figure]
Figure 3-4 is an illustration of recommended contraction joint layout for 
frangible slab-on-grade below elevated building.
[End figure]

3.3.3 Wire Mesh 

Wire mesh retards the ability of the slab to break apart and therefore should 
not be used.

3.3.4 Connecting the Slab to the Vertical Foundation Members

Slabs should never be connected to vertical foundations members when the slab is 
underlain by granular soil in areas subject to erosion and scour. This practice 
unnecessarily threatens the stability of the foundation system of elevated 
buildings.

3.3.5 Casting Concrete Grade Beams and Slabs-On-Grade Monolithically 

Grade beams and slabs-on-grade should never be cast monolithically in areas 
subject to erosion and scour. In these areas, grade beams must be designed to be 
self-supporting (to account for the loss of supporting soil from erosion and 
scour) and to withstand velocity flow and debris impact as well as stiffen the 
foundation system.  All slabs-on-grade must designed to act separately from 
grade beams.

3.3.6 Concrete Collars

In areas subject to erosion and scour, concrete collars should not be placed 
around foundation pilings.

3.4 On-Site Utility Systems

On-site utilities need to be installed with much greater attention to the 
effects of flooding. In some cases, such as placement of electrical meters, 
installation will need to be coordinated with local public utility companies. 
Installation of other items, such as septic systems, may fall under the purview 
of local or State health departments.

3.4.1 Air Conditioner / Heat Pump Compressor Platforms

Platforms that support air conditioner / heat pump compressors must be designed 
to withstand the forces associated with the base flood. In a coastal floodplain, 
the best way to avoid damage to these platforms is to employ the method used for 
the protection of buildings — elevation. Therefore the bottom of the lowest 
horizontal structural member of the platform should be elevated to or above the 
BFE. Ideally, platforms should be cantilevered from an elevated floor diaphragm 
(see Figure 3-5). An alternative is to support the platform partially or 
completely on pilings (see Figure 3-6).

[Begin figures]
Figure 3-5 is an illustration of a cantilevered mechanical platform

Figure 3-6 is an illustration of a mechanical platform supported by pilings.
[End figures]

Platforms designed and constructed with vertical foundation members must be 
protected from localized scour and, in oceanfront areas, protected from erosion 
so that the foundation members can resist the velocity flow, wave action, and 
debris impact found in coastal areas. When a vertical foundation member looses 
it ability to support the platform, the platform collapses, becoming waterborne 
debris that is then carried into the structure or nearby structure. Because of 
the cost of the compressor (often $2,000 and up), the potential loss of 
habitability when the compressor is rendered inoperable, and the debris that 
platforms generate once they collapse, these platforms cannot be considered 
sacrificial. 

Vertical foundation members for compressor platforms in landward areas should 
meet the same requirements as the main building support system. In areas subject 
to scour, embedment of these foundation members should be based on a depth 
related to existing grade. From the observed performance of oceanfront 
buildings, an embedment depth of -10 feet m.s.l. is recommended for all vertical 
foundation members for oceanfront buildings.  From the observed performance of 
landward coastal buildings, an embedment depth of 8 feet below existing grade is 
recommended for platform vertical foundation members. These recommendations are 
considered prudent in the absence of specific State and local requirements.

3.4.2 Placement of Utilities on, Through, or Adjacent to Breakaway Wall Panels

Utility services should never be placed in a such a manner that they can be 
damaged when a breakaway wall panel breaks away or in such a way that they will 
interfere with the function of the breakaway panels. As noted in Section 2.6, 
the BPAT observed numerous types of utility system components mounted on or 
adjacent to breakaway wall panels. This practice is unacceptable and almost 
always results in the utility components being severely damaged during a 
hurricane. The BPAT also observed utility lines penetrating breakaway wall 
panels. This practice is also unacceptable. When it becomes necessary to extend 
these lines through a wall panel, a utility blockout, as shown in Figure 3-3, 
should be built into the wall.

3.4.3 Placement of Utilities Adjacent to Vertical Support Members

Utilities installed on the landward side of vertical foundation members are 
shielded by the foundation members against damage from velocity flow and debris 
impact. Service connections such as electrical meters, telephone junction boxes, 
and cable junction boxes that must be exposed to flooding should be placed on 
the landward side of the most landward vertical foundation member (see Figure 3-
7). Vertical utilities such as sewer and water risers should also be placed on 
the landward side of vertical foundation members.

[Begin figure}
Figure 3-7 is an illustration of the proper location of utilities.
[End figure]

3.4.4 Septic Tanks

Septic tanks should be installed as far landward as practical and permitted by 
the authority having jurisdiction. Before septic tanks are installed, local and 
State Health Departments having regulatory control should be consulted 
concerning whether such tanks are permissible and how and where they should be 
installed.

3.5 Dry Floodproofing

The NFIP regulations do not permit the use of dry floodproofing in Coastal High 
Hazard Areas (V zones), which are subject to deep flooding, high-velocity flow, 
debris impact, and wave heights. The regulations do allow nonresidential 
buildings in A zones, including coastal A zones, to be dry floodproofed through 
the construction of walls that are substantially impervious to the passage of 
flood waters. Dry floodproofing in coastal A zones, however, presents special 
challenges. Although the potential for high-velocity flow, debris impact, and 
wave action in coastal A zones is significantly less than in V zones, it still 
must be assessed and accounted for in the dry floodproofing design.

Studies performed by the U.S. Army Corps of Engineers (COE) National Flood 
Proofing Committee, found in the COE publication Floodproofing Tests, 1988, 
indicate that dry floodproofing can be used in areas that experience flooding to 
a depth of 3 feet. The COE did not study coastal effects, including wave action. 
Wave action, especially waves breaking on the vertical dry floodproofing 
components, will add significant loads to the floodproofing system. 

Therefore, it is recommended that dry floodproofing be used in coastal 
floodplains only where the stillwater depth is no more than 2 feet above grade 
during the base flood. Where the stillwater depth is 2 feet, the wave crest 
elevation will be approximately 1 foot above the stillwater elevation, and wave 
breaking and overtopping will reach even higher. To protect a building from wave 
overtopping where the stillwater depth is 2 feet, 4 feet of dry floodproofing is 
recommended (see Figure 3-8). In no case should the increase in floodproofing 
height be used to compensate for higher stillwater depths without a detailed 
engineering analysis.

[Begin figure}
Figure 3-8 is an illustration of dry floodproofing in coastal A zones.
[End figure]

Dry floodproofing is normally designed for velocity flows that do not exceed 10 
feet per second. Velocities up to 10 feet per second can be converted to an 
approximate equivalent hydrostatic head (see FEMA 259: Engineering Principles 
and Practices for Retrofitting Flood Prone Residential Buildings). In the case 
of structure subjected to a 2-foot stillwater flood depth, a velocity of 10 feet 
per second can be converted to an approximate increase of 1 foot of flood depth. 
This provides further justification for restricting floodproofing to coastal 
structures that are subject to no more than 2 feet of stillwater flooding. For 
additional information on proper floodproofing design and construction criteria, 
see FEMA 259, Engineering Principles and Practices for Retrofitting Flood-Prone 
Residential Structures, and FEMA’s Technical Bulletin No. 3, Non-Residential 
Floodproofing — Requirements and Certification.

3.6 Wind Damage Protection

The observed wind damage to porch roofs and large overhangs can be avoided in 
the future through the use of easily implemented and inexpensive retrofits, such 
as the addition of corrosion-resistant metal connectors at critical locations. 
Revising the State Building Code to include construction details for top-of-
column and bottom-of-column connections would help ensure that future 
construction is better able to resist high winds.

3.7 Protection of Metal Structural Components from Corrosion

Maintaining the design strength of all structural components is critical. Any 
loss of strength can lead to structural failure during subsequent hurricanes. 
FEMA recently issued NFIP Technical Bulletin No. 8, Corrosion Protection for 
Metal Connectors in Coastal Areas, which provides guidance concerning the 
selection, installation, and maintenance of metal connectors such as truss 
plates and hurricane straps. 






APPENDIX A

Building Performance Assessment, Damage Assessment, and Hazard Mitigation 
Reports Prepared by the Federal Emergency Management Agency

The Oklahoma City Bombing: Improving Building Performance Through Multi-Hazard 
Mitigation, in conjunction with the American Society of Civil Engineers, August 
1996

Hurricane Opal in Florida, A Building Performance Assessment, August 1996

FEMA DR-TX-1041, Flooding in Southeast Texas from the Storm of October 15 - 21, 
1994, August 1995

FEMA DR-GA-1033, Flooding in Georgia from Tropical Storm Alberto, January 1995

Preliminary Field Assessment — Hurricane Emily on the North Carolina Outer 
Banks, January 1994

Building Performance: Hurricane Iniki in Hawaii — Observations, Recommendations, 
and Technical Guidance, January 1993

Building Performance: Hurricane Andrew in Florida — Observations, 
Recommendations, and Technical Guidance, December 1992

Building Performance Assessment Team Report: Noreaster, Delaware and Maryland, 
January 1992

Flood Damage Assessment Report: Hurricane Bob, August 1991

Damage Assessment of Flooded Buildings 1985 - 1990, June 1991

Flood Damage Assessment Report: Tropical Storm Allison, June 1990

Flood Damage Assessment Report: Noreaster of April 1990, June 1990

Flood Damage Assessment Report: Riverine Flooding in Central Kentucky, February 
1990

Flood Damage Assessment Report: Hurricane Hugo, October 1989

Flood Damage Assessment Report: Texas, June 1989

Flood Damage Assessment Report: Noreaster, Mid-Atlantic Coast, March 1989

Flood Damage Assessment Report: Riverine Flooding in Maine, June 1988

Flood Damage Assessment Report: Noreaster, Mid-Atlantic Coast, April 1988

Flood Damage Assessment Report: Riverine Flooding in Central Michigan, May 1987

Flood Damage Assessment Report: Riverine Flooding in Allegheny County, 
Pennsylvania, January 1987

Flood Damage Assessment Report: Riverine Flooding in Clive, Iowa, September 1986

Flood Damage Assessment Report: Hurricane Gloria, February 1986

Improving Resistance of Buildings to Wind Damage:  Hurricane Elena, September 
1985

Hazard Mitigation Team:  Hurricane Diana, 1984

Proposed Changes to Building Codes in Response to Hurricane Alicia, August 1983

Hazard Mitigation Report: Noreaster, Outer Banks, North Carolina, October 1982

Hazard Mitigation Report: Hurricane Frederic, September 1979






APPENCIX B

Members of the Building Performance Assessment Team for Hurricane Fran

Clifford Oliver
Senior Engineer / Project Officer and Team Leader
Mitigation Directorate
Federal Emergency Management Agency
Washington, DC

Donald Beaton
Chief, Regulations & Underwriting Branch
Policy Analysis & Emergency Management Coordination Division
Federal Insurance Administration
Federal Emergency Management Agency
Washington, DC

Paul Tertell
Senior Engineer
Mitigation Directorate
Federal Emergency Management Agency
Washington, DC

Mark Vieira, P.E.
Civil Engineer
Hazard Identification and Risk Assessment Branch
Federal Emergency Management Agency, Region IV
Atlanta, Georgia

Charles Bornman, P.E.
Senior Structural Engineer
Greenhorne & O’Mara, Inc.
Greenbelt, Maryland

Christopher Jones, P.E.
Senior Coastal Engineer
Earth Tech
Charlottesville, Virginia

Spencer Rogers
Coastal Construction and Shoreline Erosion Specialist
North Carolina Sea Grant
Kure Beach, North Carolina

Greg Chiu
Insurance Institute for Property Loss Reduction
Boston, Massachusetts

Dan Deegan
Senior Coastal Engineer
Woodward-Clyde Federal Services
Atlanta, Georgia






APPENDIX C
Executive summary from piling embedment study performed by Woodward-Clyde 
services on Topsail Island, North Carolina

Executive Summary

Purpose

As a result of Hurricane Bertha (July 12, 1996) and Hurricane Fran (September 5, 
1996), a large number of structures were damaged on Topsail Island, North 
Carolina.  Topsail Island includes the communities of Topsail Beach, Surf City, 
and North Topsail Beach. An initial review of the structures indicated that 
shallow embedment depth of their foundation pilings could have been the primary 
cause of structural failure due to the storms. FEMA contracted with Woodward-
Clyde Federal Services (Woodward-Clyde) to perform testing and evaluation of the 
piling lengths and embedment depths. The purpose of the testing was to determine 
whether the pilings meet the embedment depth requirements of the current North 
Carolina State Building Code. The current version of the Code was  implemented 
on January 1, 1986.

Project Execution

To identify oceanfront structures built after the implementation of the current 
Code (January 1, 1986), Woodward-Clyde obtained aerial photographs for Topsail 
Island from the North Carolina Department of Transportation. The photographs 
represent the periods immediately after January 1, 1986, prior to Hurricane 
Bertha, after Hurricane Bertha, and after Hurricane Fran. From these 
photographs, a total of 205 post-1985 oceanfront structures were identified. 
Duplex or multiplex units (two- to four-family structures) were considered 
single structures for purposes of this study.

Of the 205 structures, a total of 16 (7.8 percent) were identified as having 
leaning pilings (11 structures) or were identified as total losses (5 
structures). Structures identified as total losses were either completely washed 
away by the storm or were so severely damaged that the structures were totally 
destroyed. Many of the 205 structures identified received other damage such as 
roof, wall, deck, and concrete damage caused by both flooding and high winds. 
However, the focus of this study was on damage to piling foundations that 
supported elevated residential one- to four-family structures.

Field inspections of the 205 oceanfront structures were conducted to identify 
piling damage and general building parameters. A total of 20 damaged and 
undamaged structures were initially identified for piling testing. However, 
after homeowner approvals were requested and received, it was determined that 
only 11 structures would be tested. These 11 structures include 7 structures 
with leaning pilings and 4 structures with no leaning pilings.

Testing Procedures

Using a nondestructive test methodology to determine total piling length, 
Woodward-Clyde tested 5 pilings at each structure, a sampling of approximately 
25 percent. For the test, accelerometers were mounted directly to each piling. 
The piling then was struck on its side with a hammer. The hammer blow created 
dispersive stress waves that traveled the length of the piling. Data recorded by 
the accelerometers were then digitally processed and analyzed. Analysis of the 
data yielded a computation for the total length of the piling. Ground level and 
top-of-piling elevations were surveyed and used to determine piling embedment 
depth.

Typical accuracy using this type of testing is ± 10 percent. In three piling 
test cases, the return signals recorded were not accurate enough to allow for a 
piling length determination. Therefore, for 3 of the eleven 11 structures, only 
4 instead of 5 pilings were tested, yielding a total of 52 tested pilings.

Piling Test Results

The current North Carolina State Building Code requires that piling tips be at -
5 feet m.s.l. or 16 feet below grade, whichever is shallower. The findings of 
the tests are based on the evaluation of the piling embedment depths in relation 
to the -5 feet m.s.l. criterion, since this, rather than the 16 feet below grade 
criterion, is the controlling factor for piling embedment in the test area. This 
is because pre-storm grade elevations for most oceanfront houses on Topsail 
Beach were less than 11 feet m.s.l. 

Of the 11 structures tested, 4 were one-story above the pilings and 7 were two 
or more stories above the pilings. The following table summarizes the findings.

No.: 1
Stories above top of piling: 1
Pilings leaning: yes
Cross bracing: yes
Total pilings per structure: 21
Piling not meeting code: 4 of 5
Average differential to meet code footnote 1: 0.9

No.: 2
Stories above top of piling: 1
Pilings leaning: Yes
Cross bracing: Yes
Total pilings per structure: 21
Piling not meeting code: 4 of 5
Average differential to meet code footnote 1: 2.0

No.: 3 
Stories above top of piling: 2
Pilings leaning: Yes
Cross bracing: No
Total pilings per structure: 20
Piling not meeting code: 4 of 4
Average differential to meet code footnote 1: 4.7

No.: 4
Stories above top of piling: 2
Pilings leaning: yes
Cross bracing: no
Total pilings per structure: 20
Piling not meeting code: 4 of 4
Average differential to meet code footnote 1: 5.6

No.: 5
Stories above top of piling: 2
Pilings leaning: Yes
Cross bracing: No
Total pilings per structure: 15
Piling not meeting code: 5 of 5
Average differential to meet code footnote 1: 3.3

No.: 6
Stories above top of piling: 2
Pilings leaning: Yes
Cross bracing: Yes
Total pilings per structure: 15
Piling not meeting code: 4 of 4
Average differential to meet code footnote 1: 4.2

No.: 7
Stories above top of piling: 2
Pilings leaning: Yes
Cross bracing: Yes
Total pilings per structure: 12
Piling not meeting code: 5 of 5
Average differential to meet code footnote 1: 6.1

No.: 8
Stories above top of piling: 3
Pilings leaning: No
Cross bracing: No
Total pilings per structure: 30
Piling not meeting code: 2 of 5
Average differential to meet code footnote 1: 3.4

No.: 9
Stories above top of piling: 2
Pilings leaning: No
Cross bracing: No
Total pilings per structure: 15
Piling not meeting code: 1 of 5
Average differential to meet code footnote 1: 0.7

No.: 10
Stories above top of piling: 1
Pilings leaning: No
Cross bracing: Yes
Total pilings per structure: 50
Piling not meeting code: 5 of 5
Average differential to meet code footnote 1: 2.6

No.: 11
Stories above top of piling: 1
Pilings leaning: No
Cross bracing: No
Total pilings per structure: 25
Piling not meeting code: 5 of 5
Average differential to meet code footnote 1: 4.7

Footnote 1 1Average differential is the average distance from the tip of piling 
to -5 feet m.s.l. 
Note: Information on the effects of erosion and scour is provided in the 
sections of the report preceding this appendix.

One-Story Structures (Total of Four)

- Two had leaning pilings.

- Three had some bracing, although none is required by the current Code.  The 
structure that had no cross-bracing was not damaged.

- Ninety percent of the pilings tested for these four one-story structures (both 
with and without leaning pilings) did not meet the current Code requirement for 
piling embedment depth.

Two- and Three-Story Structures (Total of Seven)

- Five had leaning pilings.

- Two of the five structures with leaning pilings had cross-bracing at the time 
of the storm, but none of the structures had cross-bracing in accordance with 
the current Code. The two structures that did not have leaning pilings had no 
cross-bracing. It should be noted that the current Code allows for alternative 
bracing systems if they are designed and sealed by a Licensed Professional 
Engineer or Architect.

- Including the pilings within the 10-percent accuracy range, 78 percent of the 
pilings tested for the seven two-story structures (both with and without leaning 
pilings) did not meet the current Code requirement for piling embedment depth.

The following table provides a breakdown of the number of pilings by the amount 
of additional embedment depth necessary for the piling to meet the Code 
requirement. Of the 52 pilings tested (including those pilings within the 10-
percent accuracy range), over 80 percent did not meet the Code requirement.

Additional Depth (feet): 0
Number of Pilings Required to Meet Code: 9

Additional Depth (feet): 0-1
Number of Pilings Required to Meet Code: 3

Additional Depth (feet): 1-2
Number of Pilings Required to Meet Code: 6

Additional Depth (feet): 2-3
Number of Pilings Required to Meet Code: 8

Additional Depth (feet): 3-4
Number of Pilings Required to Meet Code: 4

Additional Depth (feet): > 4
Number of Pilings Required to Meet Code: 22

It was observed that all on the identified post-1985 structures that had leaning 
pilings, the pilings leaned inland in a westerly direction (the direction of the 
storm surge).

Conclusions

From the field observations and test results, Woodward-Clyde concludes the 
following:

- Approximately 92 percent of oceanfront structures built after the 
implementation of January 1986 Code changes did not sustain significant piling 
damage.

- All tested post-1985 structures that had leaning pilings did not meet the 
requirements of the current Code. Of all the pilings tested, both leaning and 
not leaning, over 80 percent did not meet the Code requirement.

- It appears that the Code requirement for piling embedment depth may be more 
effective in preventing piling damage than the requirement for cross-bracing. 
This includes those pilings within the 10-percent accuracy range.

From the test results and the field observations, it appears that a structure 
should sustain minimal piling damage if it is constructed according to the 
current Code requirements. However, several factors exist that prevent a 
complete evaluation of the piling requirements in the Code:

- The study involved testing only 11 of 205 post-1985 oceanfront structures.

- The findings for the piles not meeting Code are limited to structures whose 
piling embedment depth is controlled by the -5 foot m.s.l. criterion.

- The majority of the structures tested did not meet the embedment depth 
requirement of the Code, including those that did not have leaning pilings.

Recommendations

For the reasons cited above, the relative effectiveness of the two embedment 
criteria, “tip penetration of at least 5.0 below mean sea level or 16 feet below 
average original grade which ever is least,” cannot be made. Therefore, 
Woodward-Clyde can not recommend a change to the piling embedment depth 
requirement of the North Carolina State Building Code at this time. Woodward-
Clyde does, however, recommend that better construction and inspection practices 
be implemented to ensure proper installation of the pilings so that they at 
least meet the current Code requirements.