HomeMy WebLinkAboutDuck Infrastructure Vulnerability Assessment-2020
ΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗ
Coastal Hazards Infrastructure Vulnerability Assessment
Duck, North Carolina
ΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗΗ
CĻĬƩǒğƩǤЋЉЋЉ
Program for the Study of Developed Shorelines
Western Carolina University
Cullowhee, NC 28723
Program for the Study of Developed Shorelines
Western Carolina University
This document was prepared by the Town of Duck under grant award # NA15NOS4190066 to the
Department of Environmental Quality, Division of Coastal Management from the Office for Coastal
Management, National Oceanic and Atmospheric Administration. The statements, findings, conclusions,
and recommendations are those of the authors and do not necessarily reflect the views of DEQ, OCM or
NOAA. The authors, at the Program for the Study of Developed Shorelines at Western Carolina
University, include: Blair Tormey (btormey@wcu.edu), Katie Peek (kmcdowell@wcu.edu), Holli
Thompson (hthompson@wcu.edu), Robert Young (ryoung@wcu.edu), Hanyu (Helen) Wang
(hanyu.wang@duke.edu).
Cover Photo: Town of Duck commercial center on Currituck Sound (Photo credit: Program for the Study
of Developed Shorelines at Western Carolina University).
This Page: Cypress tree in Currituck Sound (Photo credit: Program for the Study of Developed Shorelines
at Western Carolina University).
2 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Table of Contents
Executive Summary......................................................................................................................................4
Vulnerability Assessment Products & Deliverables.........................................................................5
Digital Data Sources.........................................................................................................................5
Introduction & Project Description...............................................................................................................6
Methodology................................................................................................................................................7
Step 1: Exposure Analysis & Mapping..............................................................................................7
Step 2: Sensitivity Analysis...............................................................................................................8
Step 3: Vulnerability Calculation......................................................................................................9
Results Summary & Discussion.....................................................................................................................9
Exposure Results..............................................................................................................................9
Sensitivity Results..........................................................................................................................13
Vulnerability Results......................................................................................................................13
Vulnerability Assessment Caveats.................................................................................................14
Conclusions & Potential Adaptation ..........................................................................................................18
Appendix: Summary Data Tables................................................................................................................20
3 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Executive Summary
Coastal
Hazards Infrastructure Vulnerability Assessment for the Town of Duck, North Carolina.Vulnerability is
generally defined as the extent to which a resource is susceptible to harm from hazards or climate
change impacts. For infrastructure (assets), vulnerability is most often calculated as a combination of
exposure and sensitivity. Exposure refers to the extent or degree to which climate change or a natural
hazard is likely to affect an asset, and sensitivity refers to how it will fare when exposed to a
hazard/impact. This assessment evaluated individual buildings and roads, allowing managers to compare
the vulnerability of individual assets to develop more detailed adaptation plans and strategies. The
coastal hazards evaluated in this study include flooding, storm surge, sea-level rise, and erosion.
This assessment focused on roads and a subset of commercial, professional, retail, and publicly-owned
buildings near the town center. Most of the buildings evaluated are concentrated along Duck Road (NC
HWY 12) near Currituck Sound. In total, 65 buildings and 308 road segments (almost 32 miles) were
included in the vulnerability assessment. Specific scores are reported for each individual asset in the
supplied Excel datasheets, and summary data tables can be found in the Appendix. Final exposure and
vulnerability results are also provided as GIS maps and layers.
The three highest vulnerability buildings evaluated are Beach Realty, Sunset Grill & Raw Bar, and Kitty
Hawk Surf Company. An additional 20 buildings have high-moderate vulnerability. The parcels
containing high or high-moderate vulnerability buildings have an estimated value over $24 million, while
the associated buildings have an estimated value over $10.5 million. Almost one-third of the evaluated
buildings have moderate vulnerability. An additional 21 buildings have low-moderate vulnerability, while
the lowest vulnerability building evaluated is Ocean Atlantic Rentals.
The highest vulnerability road segment is a short portion of Duck Road, just north of the Waterfront
Shops. An additional 6 road segments (totaling over 1 mile) have high-moderate vulnerability, including
several portions of Duck Road. In total, 2.60 miles of road have moderate vulnerability, 13.65 miles of
road have low-moderate vulnerability, and 7.31 miles have low vulnerability. Nearly 7 miles of road have
minimal vulnerability to coastal hazards (due to lack of exposure).
Compared to many barrier island communities, Duck has relatively low overall vulnerability. This is due
to its unique sheltered coastal setting and significant interior elevation. Although Duck has these
beneficial factors, it still has significant exposure to coastal hazards: primarily coastal erosion on the
oceanfront and flooding on the soundside. Win
the coast to erosion, flooding, and waves over multiple days.
Due to low elevations and soundside shoreline retreat (from loss of wetlands and marsh), much of the
infrastructure in the town commercial center is exposed to flooding, erosion, and sea-level rise. Because
Duck Road (NC HWY 12) is the only road with a continuous north-south connection, it is the most critical
transportation corridor in Duck. The high vulnerability along several segments of this critical road has
significant impacts for the entire town, as well as communities to the north. Recommendations for
adaptation along the soundside shoreline include the development of a long-term transportation plan
for Duck Road, and strategies that restore or slow the loss of marshes and wetlands (e.g., living
shorelines).
The vulnerability of existing structures in Duck can be reduced through two primary adaptation
measures: elevation and/or relocation. It is also important to consider reducing the vulnerability of any
associated utilities, including coordinating with utility providers on the development and installation of
more resilient infrastructure. While these adaptation actions may not always be practical, they are
4 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
important to consider, as they are the most efficient way to decrease the vulnerability and increase the
resiliency of existing infrastructure. These adaptation options should certainly be considered following
storms when funds may become available for resilience actions. It is also vital to increase the resilience
of any future development in Duck. The safest bet is to place new infrastructure in areas that have
minimal exposure to hazards. When that is not possible, adopting higher standards for building
elevation and construction can lead to more sustainable infrastructure over the long term.
Vulnerability Assessment Products & Deliverables
1. Excel datasheets: All results are provided in tables, including scoring of individual buildings and
roads. The exposure, sensitivity, and vulnerability scores are reported alongside any additional
asset attributes, as well as intermediate scores in the analysis.
2. Geographic Information Systems (GIS) Maps and Layers: WCU will provide all GIS data, including
the exposure layers, exposure results, and final vulnerability results as a separate file. Digital
data sources can be found in the next section of this document.
3. Vulnerability Results Summary Document: This summary document (herein) explains the
deliverables, results, and methodology. It briefly summarizes the vulnerability assessment
results in the aforementioned datasheets and maps, as well as the methodology utilized. This
document does not fully describe all results from the analysis; see the Appendix and Excel
datasheets for detailed results.
Digital Data Sources
1. FEMA Flood Zones: Preliminary FEMA flood maps were obtained from
Products website. According to FEMA, the VE zones are areas subject to inundation by the 1-
percent-annual-chance flood event, with additional hazards due to storm-induced velocity wave
action, and the AE zones are areas subject to inundation by the 1-percent-annual-chance flood
event (determined by detailed methods). The shaded X zone (referred to as 500-year)
represents areas between the limits of the base flood and the 0.2-percent-annual-chance (or
500-year) flood.
2. NOAA Sea-Level Rise Inundation Scenarios: Sea-level rise inundation scenarios were obtained
sea-level rise viewer.
3. NOAA Storm Surge (SLOSH) Model: Storm surge data were National
Hurricane Center.
4. Erosion/Coastal Proximity: Simple coastal proximity buffers of 25, 50, and 100 feet were applied
to the soundside shoreline at Duck. Shorelines were digitized by WCU along the marsh/water
interface.
5 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Introduction & Project Description
of Developed Shorelines has completed a Coastal
Hazards Infrastructure Vulnerability Assessment for the Town of Duck, North Carolina. Unlike many
studies, this assessment evaluated individual buildings and roads, allowing managers to compare the
vulnerability of individual assets to develop more detailed adaptation plans and strategies.
In the past, vulnerability has been defined in many ways, but can be generally described as the extent to
which a species, habitat, or resource is susceptible to harm from hazards or climate change impacts. For
infrastructure, vulnerability is calculated as a combination of two components: exposure and sensitivity.
Vulnerability = Exposure + Sensitivity
Exposure refers to the extent or degree to which
climate change or a natural hazard is likely to
affect a resource.
Sensitivity refers to how a resource will fare
when exposed to a hazard/impact.
The ability of a resource to adjust or cope with climate
change or natural hazards (referred to as adaptive
capacity) is often included as an additional component
of vulnerability, particularly in natural resource and
socioeconomic studies. However, this concept is difficult
to apply in the built environment (i.e., buildings, roads,
etc.). For example, buildings cannot inherently adapt to
climate change or other hazards, while natural resources
often can (a salt marsh can adapt to sea-level rise by
migrating upland, whereas a building cannot). Thus, the
ability of an asset to adapt to changes is not included in
the vulnerability score. This does not mean that
understanding the adaptation potential of an asset is not
important. Identifying the range of effective adaptations
for key vulnerable infrastructure is the logical next step
in the hazard/climate change planning process. Effective
adaptations will reduce exposure and/or sensitivity,
which is the key to reducing vulnerability.
Roads and non-residential buildings were the focus of
the infrastructure vulnerability assessment of Duck.
Because this assessment calculates vulnerability at the
asset level, it was not practical to evaluate all buildings
in the town. Therefore, a subset of commercial,
professional, retail, and publicly-owned buildings near
the town center were chosen for evaluation (after
discussions with Town officials). Most of the buildings
Figure 1. Buildings (white dots) and roads (lines)
evaluated are concentrated along Duck Road (NC HWY
included in the vulnerability assessment of Duck.
12) near Currituck Sound. Roads were evaluated using
Duck Road (NC HWY 12) is highlighted in yellow.
pre-determined segments obtained from Dare County. In
total, 65 buildings and 308 road segments (totaling
6 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
almost 32 miles) were included in the vulnerability assessment (Figure 1). Road segments were further
categorized into the following groups: 1) primary roads, 2) secondary roads, and 3) tertiary roads. Duck
Road (NC HWY 12) is the only road considered primary, as it provides the main north-south access route
through the town. All secondary roads branch from Duck Road, and tertiary roads connect to the
secondary roads.
Methodology
The Coastal Hazards Infrastructure Vulnerability Assessment for Duck has three primary steps: 1)
Exposure Analysis and Mapping, 2) Sensitivity Analysis, and 3) Vulnerability Calculation. A detailed
description of the scoring information for buildings and roads can also be found in the Excel results
sheets that accompany this report.
Step 1: Exposure Analysis & Mapping
The first step in this assessment was to analyze the exposure of the evaluated infrastructure to coastal
hazards. Five primary factors or hazards were determined for asset exposure; these factors indicate how
exposed an asset is to coastal hazards. The general exposure factors/hazards include flooding potential,
storm surge, sea-level rise inundation, coastal proximity/erosion, and historic flooding/damage (for
buildings only). Table 1 summarizes these exposure factors/hazards, as well as utilized data sources.
Table 1. Exposure factors/hazards evaluated and data sources.
Exposure Factors/Hazards Data Sources & Description
Flooding Potential Preliminary FEMA Flood Zones (VE or AE); 1% annual flood chance ± velocity/waves
Storm Surge NOAA SLOSH model; Category 1-5 for buildings, category 3 for roads; LiDAR DEM
Sea-Level Rise Inundation NOAA SLR modeling; 1-6 ft for buildings, 3 ft for roads
Coastal Proximity/Erosion Shoreline proximity buffers; 25 ft, 50 ft, & 100 ft buffers
Historic Flooding/Damage Post-storm damage reports
*SLOSH - Sea, Lake, and Overland Surges from Hurricanes
The exposure analysis utilized data imported into Geographical Information Systems (GIS) format, as
exposure is directly dependent on location and mapped hazard data (whether the area near the
infrastructure experiences the hazard). Digital hazard data were gathered for each of the exposure
factor/hazards, such as the online georeferenced FEMA flood map layers. The only dataset that does not
come from a widely available, well established source is historic flooding/damage, which was derived
from post-storm damage reports. Each exposure data layer thus represents an exposure factor or hazard
zone. Assets located within a particular zone are assigned a higher score than assets located outside of
the hazard zone. Scores for each exposure factor are then summed and grouped together (binned) to
get a total exposure score. Final exposure scores fall into one of six ranking categories: high, high-
moderate, moderate, low-moderate, low, and minimal (asset does not lie within any mapped zone).
Although the exposure factors/hazards are similar for buildings and roads, there are some differences in
the specific scoring methodology. This is due to the fact that buildings were analyzed as point features,
while roads were analyzed as segmented linear features (road segments were obtained from Dare
County). Buildings were compared to multiple categories of storm surge (category 1-5) and sea-level rise
scenarios (1-6 feet). However, due the complexity of scoring road segments (varying lengths, linear
roads intersecting polygon hazard data, etc.) only one storm surge category (category 3) and one sea-
level rise scenario (3 feet) were evaluated.
7 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Step 2: Sensitivity Analysis
The second step in this assessment was to analyze the sensitivity of the evaluated infrastructure to
coastal hazards. Similar to exposure, a set of factors was determined for asset sensitivity (Table 2); these
factors indicate how sensitive an asset is to coastal hazards. Unlike exposure, sensitivity is evaluated
independent of geographic location (only exposure is location-dependent). Sensitivity refers to how that
asset would fare when exposed to the hazard, which is a function of the inherent properties or
characteristics of the asset. For buildings, sensitivity factors include: protective engineering, building
age, first floor elevation (FFE) compared to base flood elevation (BFE), FFE compared to storm surge
inundation levels (height above ground), and design/construction. For roads in Duck, only the protective
engineering factor was applicable for sensitivity.
Table 2. Sensitivity factors evaluated and data sources.
Sensitivity Factor Data Sources & Description
Protective Engineering
Field/aerial imagery analysis; Town officials
buildings & roads
Building Age
Dare county parcel data; Town officials
buildings only
FFE vs BFE
FFE data estimated using field/imagery analysis & Town officials; BFE data from FEMA
buildings only
FFE vs Surge FFE data estimated using field/imagery analysis & Town officials; Surge inundation levels
buildings only from NOAA
Design & Construction
Field & imagery analysis
buildings only
The protective engineering factor represents whether an asset is protected by hard structures (e.g.,
seawalls, bulkheads) or landscape modifications (e.g., significant drainage alteration, major restored
landscape). This factor assumes that infrastructure protected with engineering is less likely to be
damaged by coastal hazards. For this analysis, this sensitivity factor was only considered if the asset was
in a coastal proximity/erosion buffer zone (exposed).
The remaining four sensitivity factors were only applied to buildings (Table 2). The building age factor
(scored in 15 year increments) assumes that older buildings are more likely to sustain damage when
exposed to coastal hazards. First floor elevation was utilized for two sensitivity factors: 1) comparison to
and 2) comparison to
Category 5 was chosen because all buildings evaluated are exposed to this surge level, allowing for a
uniform comparison ķĻƭƦźƷĻ ğ ƌƚǞ ƦƩƚĬğĬźƌźƷǤ ŅƚƩ Ʒŷźƭ ķĻŭƩĻĻ ƚŅ ƭǒƩŭĻ ĻǣƦƚƭǒƩĻ źƓ 5ǒĭƉ͵ The final
sensitivity factor
construction and/or finishes.
Because digital sensitivity data are not generally available, the primary data for much of the sensitivity
analysis was obtained from field/imagery analysis as well as discussions with town officials (Table 2). A
higher score was given for an unfavorable sensitivity factor result (e.g., an older building was scored
higher than a newer building). For buildings, the sensitivity scores for each factor were summed to
obtain a total raw score, then grouped together (binned) into six categories: high, high-moderate,
moderate, low-moderate, low, and minimal. Since only the protective engineering factor was applicable,
no specific sensitivity score was calculated for roads. Instead, if a road segment was in a coastal
proximity zone and had protective engineering, its raw exposure score was reduced.
8 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Step 3: Vulnerability Calculation
To obtain a vulnerability score, the exposure and sensitivity scores are summed, and then grouped
together (binned) into six vulnerability ranking categories. The ranking categories are as follows: high,
high-moderate, moderate, low-moderate, low, and minimal vulnerability. Specific scoring ranges for
vulnerability can be found within the Excel results sheets.
Results Summary & Discussion
A total of 65 buildings and 308 road segments (totaling almost 32 miles) were included in the
vulnerability assessment of Duck. Specific scores are reported for each individual asset in the supplied
Excel datasheets and summarized in the Appendix; final exposure and vulnerability results are also
provided as GIS maps and layers.
Exposure Results
A notable result of the exposure analysis is that over 50% of the buildings evaluated have a moderate
exposure or higher. This overall high exposure is partially due to the buildings selected for evaluation, as
the town commercial center is situated along the Currituck Sound waterfront (Figure 1), which exposes
these assets to more coastal hazards. Ten of the evaluated buildings have the highest exposure, most of
which are part of the Waterfront Shops commercial center (Figures 2 & 3).
Over 45% of the evaluated buildings have a low-moderate or low exposure to coastal hazards (Table 3).
Most of these buildings are located on the east side of Duck Road, and therefore, are outside the FEMA
flood, coastal proximity, and sea-level rise hazard zones. None of the buildings evaluated have minimal
exposure, as each of these assets are in at least the category 5 storm surge hazard zone.
Table 3. Exposure results for evaluated buildings and road segments at Duck.
Exposure Rank # buildings % all buildings # road segments length (miles) % total road length
2 0.14 0.4%
High 10 15.4%
6 1.01 3.2%
High-Moderate 18 27.7%
36 3.47 11.0%
Moderate 7 10.8%
138 12.73 40.3%
Low-Moderate 5 7.7%
30 7.66 24.2%
Low 25 38.5%
96 6.61 20.9%
Minimal 0 0.0%
Over 4.60 miles of road in Duck have a moderate exposure or higher to coastal hazards (Table 3).
However, only 1.15 miles have high or high-moderate exposure, most of which are segments of Duck
Road. Over 20 miles of road have a low-moderate or low exposure to coastal hazards, because many of
these road segments are located on the higher elevation portions of the community near the center of
the island (Figure 4). Over 6.6 miles of road have minimal exposure using this methodology, which
means these road segments did not fall within ğƓǤ of the mapped exposure hazard zones (flooding,
storm surge, sea-level rise, or erosion/coastal proximity). Exposure is directly dependent on location;
thus, if an asset is located beyond the influence of a particular coastal hazard, its exposure is diminished.
9 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Figure 2. Exposure results for select buildings and roads in Duck.
10 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Figure 3. Exposure results for select buildings and roads in Duck near the town commercial center.
11 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Figure 4. Digital elevation map of Duck with roads (white lines) and select buildings (white dots).
12 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Sensitivity Results
The sensitivity results for buildings show that over one-quarter have either high or high-moderate
sensitivity to coastal hazards (Table 4). Four of the evaluated buildings have the highest sensitivity rank,
Bait & Tackle. All of these buildings have a high sensitivity to storm surge inundation, largely due to
lower first floor elevations. Almost 40% of buildings have a moderate sensitivity, while approximately
one-third have a low-moderate or low sensitivity. The lowest sensitivity buildings are the Village Table &
, and Aqua. For roads in Duck, only the
protective engineering factor was applicable for sensitivity, and therefore is not scored separately.
Table 4. Sensitivity results for evaluated buildings in Duck.
Sensitivity Rank # buildings % all buildings
High 4 6.2%
High-Moderate 13 20.0%
Moderate 25 38.5%
Low-Moderate 20 30.8%
Low 3 4.6%
Minimal 0 0.0%
Vulnerability Results
The three highest vulnerability buildings evaluated in Duck are Beach Realty, Sunset Grill & Raw Bar, and
Kitty Hawk Surf Company. An additional 20 buildings (31%) have high-moderate vulnerability (Table 5,
Figures 5-7). The parcels containing high or high-moderate buildings have an estimated value over $24
million, while the associated buildings have an estimated value over $10.5 million. Over 30% of the
evaluated buildings have moderate vulnerability. An additional 21 buildings (32%) have low-moderate
vulnerability, while the lowest vulnerability building evaluated in Duck is Ocean Atlantic Rentals (Table 5,
Figures 5-7).
Table 5. Vulnerability results for evaluated buildings and road segments in Duck.
Vulnerability Rank # buildings % all buildings # road segments length (miles) % total road length
1 0.06 0.2%
High 3 4.6%
6 1.04 3.3%
High-Moderate 20 30.8%
34 2.60 8.2%
Moderate 20 30.8%
141 13.65 43.2%
Low-Moderate 21 32.3%
29 7.31 23.1%
Low 1 1.5%
97 6.96 22.0%
Minimal 0 0.0%
The highest vulnerability road segment in Duck is the portion of Duck Road (NC HWY 12) between Sea
Colony Drive and Dune Road, north of the Waterfront Shops (Figure 6). This ~340-foot segment of road
is within 25 feet of the shoreline, and is in a low-lying area that is easily flooded. Minor protective
engineering (e.g., rip-rap) has been installed along a portion of this road segment. An additional 6 road
segments have high-moderate vulnerability. These are primarily portions of Duck Road, including: 1) 2
contiguous segments north of Station Bay Drive, 2) the segment between Oyster Catcher Lane and
Ocean Pines Drive, 3) the segment between Olde Duck Road and Sea Colony Drive, and 4) the segment
between Cook Drive and Marlin Drive (Figures 5 & 7). The high-moderate vulnerability road segments
total just over 1 mile in length. In total, 2.60 miles of road have moderate vulnerability, 13.65 miles of
road have low-moderate vulnerability, and 7.31 miles have low vulnerability (Table 5). Nearly 7 miles of
road in Duck have minimal vulnerability to coastal hazards (due to lack of exposure).
13 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Road segments were also categorized into primary, secondary, and tertiary roads (Table 6). Just over
one mile of Duck Road (the only primary road) has high or high-moderate vulnerability to coastal
hazards, particularly where portions of this main highway cross low-lying areas and/or are in close
proximity to Currituck Sound. Most the secondary and tertiary roads have low-moderate or lower
vulnerability, as most branch off from Duck Road and generally run east-west across the higher
elevation portions of island (Figure 4).
Table 6. Vulnerability results for primary, secondary, and tertiary roads in Duck.
Vulnerability Rank primary* roads (miles) secondary roads (miles) tertiary roads (miles)
High 0.06 0.00 0.00
High-Moderate 1.02 0.02 0.00
Moderate 0.40 1.72 0.49
Low-Moderate 2.74 8.78 2.13
Low 0.88 4.92 1.51
0.92 4.29 1.74
Minimal
*Duck Road (NC HWY 12)
Vulnerability Assessment Caveats
1. This assessment analyzes the combined vulnerability of Duck to coastal hazards (e.g. flooding,
storm surge, sea-level rise, and erosion). Therefore, a section of the town that has maximum
exposure to one or more of these factors will inherently have a higher overall exposure, and
thus, vulnerability. At the same time, some of the assets were given a lower vulnerability rank
for the analyzed coastal hazards. This does not mean that these assets will not be affected by
one, or more, of these hazards in the future, but instead, that the asset is not within the
mapped hazard layers utilized (Table 1). It is important to note that on a barrier island, assets
could be destroyed by a hurricane within any given year.
2. Approximately one mile of road in Duck has high or high-moderate vulnerability. However, these
segments are primarily along Duck Road (NC HWY 12), which is the only road with a continuous
north-south connection in the town. In fact, very few secondary roads (which run perpendicular
to Duck Road) connect with each other, making detours unlikely.
3. For this study, all roads and a select subset of buildings were analyzed. If more buildings (e.g.,
residential) were included, the statistics would likely change substantially, as many are located
in higher elevations portions of the island (see Figure 4). In addition, other commercial/public
assets could be examined in the future, including parking lots, boardwalks, and recreational
spaces.
4. Typically, hurricane risk planning focuses on a direct, shore-perpendicular, landfall from a major
hurricane on the oceanfront. While this type of storm is a legitimate concern (particularly for
oceanfront homeowners), a primary concern for the town commercial center is a hurricane that
would produce sustained winds and surge from Currituck Sound. These are typically slow-
moving storms that track more oblique to the coast, focusing the strongest winds onto the
soundside shoreline.
5. As higher quality data become available for the components of vulnerability (exposure and
sensitivity), the final rankings for these assets can be updated (and may change).
14 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Figure 5. Coastal hazard vulnerability results for roads and select buildings in north Duck. Select assets are labeled.
15 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Figure 6. Coastal hazard vulnerability results for roads and select buildings in central Duck. Select assets are
labeled.
16 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Figure 7. Coastal hazard vulnerability results for roads and select buildings in south Duck, near the town
commercial center. Select assets are labeled.
17 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Conclusions & Potential Adaptation
Compared to many barrier island communities, the Town of Duck has relatively low overall vulnerability.
This is due to its unique coastal setting and significant interior elevation. The northern Outer Banks of
North Carolina (which includes Duck) are northeast facing, and are sheltered from tropical storms by
Cape Hatteras and various shoals to the south. This makes a direct hit from a tropical storm or hurricane
much less likely. Rather, northward tracking storms tend to push water from the Currituck Sound
towards the estuarine shoreline of Duck. Currituck Sound is much smaller than the neighboring
Albemarle and Pamlico sounds, which reduces the amount of fetch (open water distance) available for
producing large waves and surge. Duck also has significant elevation across the interior of the island,
with relic dune heights up to 50 feet, and largely continuous oceanfront dunes up to 25 feet high (above
sea level). Although Duck has these beneficial factors, there is still significant exposure to coastal
hazards, primarily through coastal erosion on the oceanfront and flooding on the soundside. Winter
impact this part of the coast, with erosion, flooding, and waves over
multiple days.
The town commercial center of Duck is located on the soundside of the barrier island and, therefore, is
sheltered from oceanfront processes and hazards. However, due to low elevations and shoreline retreat
(from loss of wetlands and marsh), much of the infrastructure on the soundside is exposed to flooding,
erosion, and sea-level rise. This is particularly true where infrastructure is close to the shoreline. High
exposure to these hazards is concentrated in several areas (e.g., portions of Duck Road near the water,
the Waterfront Shops, and roads in the far north where the island is narrow). Because Duck Road (NC
HWY 12) is the only road with a continuous north-south connection, it is the most critical transportation
corridor in Duck. The high vulnerability along several segments of this critical road has significant
impacts for the entire town, as well as communities to the north.
This study focused solely on the public and commercial infrastructure of the town, therefore, the
preliminary adaptation recommendations relate only to the soundside (Currituck) shoreline. To address
the transportation concerns, it is recommended that the town develop a long-term transportation plan
that considers relocation and/or elevation of the highest vulnerability segments of Duck Road. To
address the vulnerability of infrastructure along the soundside shoreline, it is also recommended that
the town consider strategies that restore or slow the loss of marshes and wetlands. These strategies
should focus initially on the establishment of living shorelines. There are several locations where
artificial access channels were created through the marsh, which may no longer be necessary (e.g.,
north of Kitty Hawk Kites and south of Kitty Hawk Surf Co.). Strategies could be implemented that close,
restore, and/or preserve these sections of marsh, reducing the vulnerability of nearby infrastructure.
The vulnerability of existing structures in Duck can be reduced through two primary adaptation
measures: elevation (raising the level of the first finished floor) and/or relocation (moving the structure).
likelihood of damage during a storm event. Relocating a building to a less exposed location (e.g., higher
ground or further from the water) reduces its exposure, by lowering the likelihood of being impacted by
coastal hazards. It is also important to consider reducing the vulnerability of any associated utilities,
including coordinating with utility providers on the development and installation of more resilient
infrastructure. While these adaptation actions may not always be practical, they are important to
consider, as they are the most efficient way to decrease the vulnerability and increase the resiliency of
existing infrastructure. These adaptation options should certainly be considered following storms when
funds may become available for resilience actions.
18 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
In addition to considering adaptation measures on existing infrastructure, it is also vital to increase the
resilience of any future development in Duck. The safest bet is to place new infrastructure in areas that
have minimal exposure to hazards. When that is not possible, adopting higher standards for building
elevation and construction can lead to more sustainable infrastructure over the long term. Many
communities have recently implemented stricter local standards for building elevation, such as
to 3 feet or higher. Along much of the soundside portion of Duck, the BFE in the proposed FEMA flood
zone (the AE or 100-year flood) is only 4 feet (above the NAVD88 datum, which is approximately mean
sea-level), and most of the evaluated commercial buildings are already elevated above this value.
Ho
shoreline could be much higher, up 7 to 8 feet above the ground. Therefore, it is recommended that
future building elevation guidelines consider these additional data sources. Increasing the requirements
for elevating buildings could significantly lower the risk of flooding and damage from future storms. We
consider the proposed FEMA BFE to be a minimum value and recommend that some freeboard become
standard practice for future construction.
19 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Appendices
Table A1. List of commercial, professional, retail, and publicly-owned buildings evaluated in Duck. Buildings are
listed generally from north to south.
# Building Name Exposure Sensitivity Vulnerability
1 Beach Realty high-moderate high high
2 North Duck Watersports high-moderate low-moderate moderate
3 Sun Realty low-moderate low-moderate low-moderate
4 Village Table/Nor' Banks Sailing & Watersports low-moderate low low-moderate
5 Watersports Outbuildings moderate low-moderate moderate
6 Duck Fire Station low moderate low-moderate
7 USACE Field Research Facility low low-moderate low-moderate
8 Sunset Grill & Raw Bar high high-moderate high
9 Barrier Island Shoppes low high moderate
10 Resort Realty high moderate high-moderate
11 Tommy's Natural Foods Market & Wine Shop low-moderate high high-moderate
12 Blue Point high moderate high-moderate
13 Barr-EE Station high moderate high-moderate
14 Donna Designs high moderate high-moderate
15 Sunset Ice Cream high moderate high-moderate
16 Designs Amity high moderate high-moderate
17 Outer Banks Olive Oil Company high moderate high-moderate
18 Coastal Cantina high moderate high-moderate
19 Duck's Cottage high-moderate high-moderate high-moderate
20 Kayak Corolla Outerbank Adventures high-moderate moderate high-moderate
21 Zen & Zip, Yoga high-moderate moderate high-moderate
22 Islands high-moderate low-moderate moderate
23 Allie June high-moderate low-moderate moderate
24 Stan White Realty high-moderate moderate high-moderate
25 Duck Deli low moderate low-moderate
26 Water Tower high-moderate high-moderate high-moderate
27 Kitty Hawk Kites high-moderate high-moderate high-moderate
28 Brindley Beach Realty low low-moderate low-moderate
29 Life is Good moderate moderate moderate
30 Wee Winks Market and Deli low-moderate moderate moderate
31 Kitty Hawk Surf Co high high-moderate high
20 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Table A1. Continued
# Building Name Exposure Sensitivity Vulnerability
32 ABC Store low-moderate moderate moderate
33 Dockside N' Duck high-moderate high-moderate high-moderate
34 Donuts on Stick high-moderate high-moderate high-moderate
35 Duck United Methodist Church high-moderate low-moderate moderate
36 Nags Head Hammocks high-moderate low-moderate moderate
37 BP Gas Station/Coastal Cravings low high-moderate moderate
38 Duck Village Outfitters low moderate low-moderate
39 PNC Bank low low-moderate low-moderate
40 Kellogg Supply Company low low-moderate low-moderate
41 Town Hall Garage low low-moderate low-moderate
42 Red Sky Café/Carolina Designs Realty low low-moderate low-moderate
43 Town Hall/Keller Meeting Hall moderate moderate moderate
44 Town Hall Pavilion low low-moderate low-moderate
45 Town Hall Picnic Shelter moderate moderate moderate
46 Roadside Bar & Grill low high-moderate moderate
47 Loblolly Pines Shopping Center/US Post Office low high-moderate moderate
48 Town Hall Amphitheater low moderate low-moderate
49 Ocean Atlantic Rentals low low low
50 Cotton Gin low moderate low-moderate
51 Osprey Landing Shopping Center high-moderate low-moderate moderate
52 Twiddy & Company Realty low low-moderate low-moderate
53 Super Wings moderate moderate moderate
54 Scarborough Faire Shopping Village low high-moderate moderate
55 NC Coast Restaurant high-moderate low-moderate moderate
56 Bob's Bait & Tackle moderate high high-moderate
57 Seagreen Gallery & Plum Crazy high-moderate high-moderate high-moderate
58 Farmers Daughters high-moderate moderate high-moderate
59 Scarborough Lane Shoppes low low-moderate low-moderate
60 Costin Creations low low-moderate low-moderate
61 Outer Banks Surf low low-moderate low-moderate
62 Urban Cottage low high-moderate moderate
63 Finch Construction low moderate low-moderate
64 Twiddy Realty Maintenance low moderate low-moderate
65 Aqua moderate low low-moderate
21 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Table A2. Vulnerability of roads in Duck, in feet.
Road Name High High-Mod Mod Low-Mod Low Minimal
ABRON CT 163
ACORN OAK AVE 652 734
ALGONKIAN CT 93
AMY LN 306
ARROWHEAD CT 84
AZALEA CT 187
BALDPATE DR 315
BARRIER ISLAND STATION 964 1,822 2,003
BAUM TRL 93 667
BAYBERRY DR 1,963
BEACHCOMBER CT 445
BETSY CT 260
BIAS DR 469 946
BIAS LN 1,207
BLUE HERON LN 648
BRANDON CT 152
BUFFELL HEAD RD 1,696 867
BUNTING LN 99 666
BUNTING WAY 377
CANVAS BACK DR 1,424
CARROL DR 555 1,089
CEDAR DR 291
CHARLES JENKINS LN 1,161 1,845
CHEROKEE CT 179
CHIP CT 519
CHOCTAW CT 200
CHRISTOPHER DR 341 336 1,675
CLAY ST 493
COFIELD CT 161
COOK DR 2,674
CYPRESS DR 352
DIANNE ST 290 587 544
DOCKS CT 211
DUCK HUNT CLUB LN 893
DUCK LANDING LN 2,423
DUCK RD (NC HWY 12) 338 5,380 2,089 14,462 4,623 4,874
DUCK RIDGE VILLAGE CT 673
DUNE RD 2,627
DUNES CREST 245
ELM DR 178
FAWN CT 168
FLIGHT DR 1,447
FORESAIL CT 58
FOUR SEASONS LN 688 1,440 559 1,173
FRAZIER CT 91
GANNET CV 235
GANNET LN 454 211
GEORGETOWN SANDS RD 1,678
22 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Table A2. Continued
Road Name High High-Mod Mod Low-Mod Low Minimal
GIFFORD CIR 516
HALYARD CT 456
HATCH COVER CT 345
HILLSIDE CT 155
JASMINE CT 137
JAY CREST RD 1,857
LALA CT 161
LONE WAY 147
MAGNOLIA CT 221
MAINSAIL CT 62
MALLARD CT 441
MALLARD DR 1,584
MANTOAC CT 214
MAPLE DR 237
MARLIN CT 278
MARLIN DR 2,604
MARTIN LN 911 178
N BAUM TRL 759
NASH RD 373
NOR BANKS DR 793 736
OCEAN BAY BLVD 683 858
OCEAN CREST WAY 383
OCEAN FRONT DR 745
OCEAN PINES DR 1,517
OCEAN WAY 616
OCEAN WAY CT 495
OLD SQUAW DR 1,385
OLDE DUCK RD 2,356
OSPREY RIDGE RD 1,044
OYSTER CATCHER LN 519 851
PAMELA CT 231
PELICAN WAY 593
PINTAIL DR 1,523
PLOVER DR 2,721
POTESKEET DR 660 1,942
PUFFER CT 31
QUAIL WAY 738
QUARTERDECK DR 393
RAKIOCK CT 85
RENE CT 121
ROCKFISH LN 301
ROYAL TERN LN 460
RUDDY DUCK LN 718
SAILFISH CT 30
SANDCASTLE CT 148
SANDPIPER CV 216
SANDY RIDGE RD 1,673
SCARBOROUGH LN 2,701
23 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
Table A2. Continued
Road Name High High-Mod Mod Low-Mod Low Minimal
SCHOONER RIDGE DR 4,348 365
SEA COLONY DR 2,075
SEA EIDER CT 197
SEA HAWK CT 234
SEA HAWK DR 1,468 1,447
SEA HAWK DR E 191
SEA TERN DR 1,502
SEABREEZE DR 255 1,697
SETTLERS LN 895
SHEARWATER WAY 418
SHELDRAKE CT 124
SHIPS WATCH DR 835 1,140
SHIPS WHEEL CT 537
SKIMMER WAY 287 1,898
SNIPE CT 58
SNOW GEESE DR 481 3,199
SOUND SEA AVE 1,435
SPECKLE TROUT DR 1,018
SPINDRIFT CT 152
SPINDRIFT LN 633
SPINNAKER CT 156 49
SPRIGTAIL DR 1,383
SPYGLASS RD 763 900
STATION BAY DR 1,059
SUNFISH CT 61
SUNFLOWER CT 218
TERESA CT 267
THRUSH CT 151
TIDES DR 1,002
TOPSAIL CT 66
TRINITIE DR 1,309
TUCKAHOE DR 1,218 1,839
TURNBUCKLE CT 280
UPPOWOC CT 88
VICTORIA CT 271
VIREO WAY 661
VIVIAN CT 192
WAMPUM DR 2,757
WAXWING CT 398
WAXWING LN 891
WHISTLING SWAN DR 368
WIDGEON DR 1,509
WILLOW DR 309
WINAUK CT 189
WINDSURFER CT 143
WIROANS CT 143
WOOD DUCK DR 1,500
YOLANDA TERR 211
24 | Duck Infrastructure Vulnerability Assessment
Program for the Study of Developed Shorelines
Western Carolina University
25 | Duck Infrastructure Vulnerability Assessment