water

Article
Development of Resilience Index Based on Flooding
Damage in Urban Areas
Eui Hoon Lee 1 and Joong Hoon Kim 2, *
1 Research Center for Disaster Prevention Science and Technology, Korea University, Seoul 02841, Korea;
[email protected]
2 School of Civil, Environmental and Architectural Engineering, Korea University, Seoul 02841, Korea
* Correspondence: [email protected]; Tel.: +82-02-3290-3316

Received: 10 April 2017; Accepted: 9 June 2017; Published: 13 June 2017

Abstract: Flooding volume in urban areas is not linearly proportional to flooding damage because,
in some areas, no flooding damage occurs until the flooding depth reaches a certain point, whereas
flooding damage occurs in other areas whenever flooding occurs. Flooding damage is different
from flooding volume because each subarea has different components. A resilience index for urban
drainage systems was developed based on flooding damage. In this study, the resilience index
based on flooding damage in urban areas was applied to the Sintaein basin in Jeongup, Korea.
The target watershed was divided into five subareas according to the status of land use in each
subarea. The damage functions between flooding volume and flooding damage were calculated by
multi-dimensional flood damage analysis. The extent of flooding damage per minute was determined
from the results of flooding volume per minute using damage functions. The values of the resilience
index based on flooding damages were distributed from 0.797292 to 0.933741. The resilience index
based on flooding damage suggested in this study can reflect changes in urban areas and can be used
for the evaluation of flood control plans such as the installation, replacement, and rehabilitation of
drainage facilities.

Keywords: flooding damage; resilience index; multi-dimensional flood damage analysis

1. Introduction
Heavy rainfall resulting from climate change can lead to inundation in urban areas; therefore,
a proper evaluation of urban drainage systems is required. The proper evaluation of urban drainage
systems is an important factor in the design and operation of urban drainage facilities. Many
researchers have studied resilience in urban cities and drainage systems. In the 1990s, the concept
of sustainability was introduced in urban cities and research on developing sustainable urban
development models was conducted [1]. The role of urban parks in a sustainable city and the meanings,
models, and metaphors of resilient cities for various factors such as sporting, relax, being with children,
meeting others, escaping from city, walking the dog, being in nature, meditating and getting inspiration
were suggested [2,3]. A sustainable city model with global sustainability with global capital, city
capacity and city condition was determined [4]. Risk and resilience were incorporated to enhance the
sustainability of urban water systems using the concept of ecosystem resilience including potable water
use, gross pollutants, cost and social acceptance [5]. The roles of risk, resilience, and environmentally
sustainable cities were studied in order to frame national security and energy policies [6]. The concept
of resilience was introduced based on these studies.
In the 2010s, a new concept of sustainability and resilience in urban systems was proposed.
Sustainability and resilience with non-equilibrium, adaptive planning and design considering bus
routes and urban drainage swales were applied in the new urban world [7]. Collaborative research
in urban areas to quantify the cost-effectiveness of resilience and integrative flood management

Water 2017, 9, 428; doi:10.3390/w9060428 www.mdpi.com/journal/water

Water 2017, 9, 428 2 of 15

was announced [8]. A resilience-based design and a management paradigm for unpredictable
and manmade catastrophic disasters was suggested as well as the concept of sustainable, robust,
and resilient water distribution systems [9,10]. A conceptual framework for designing, planning,
and managing resilient cities was constructed, considering climate change and environmental
risk [11,12]. The integration of risk and resilience in engineering systems was adapted to manage
natural and man-made catastrophes [13]. A conceptual framework with various factors such as climate
change, rapid urbanization, and population growth was created for mitigation, adaptation, and coping
strategies [14]. Resilience-based failure mode effects and a criticality analysis for regional water supply
system were proposed [15]. Recently, various resilience studies on urban systems including various
factors emerged.
Recently, resilience in urban drainage systems has been studied though research that focused on
the resilience in urban cities and water distribution systems. A global analysis approach considering
link failure, flood volume and flood duration was used to investigate structural resilience in urban
drainage systems and resilient urban drainage as options for optimized area management [16,17].
A social-ecological index for measuring flood resilience and multifunctional urban flood resilience
enhancement strategies were suggested [18,19]. A global analysis approach considering flood volume
was used for the evaluation of functional resilience in urban drainage systems [20]. Cooperative
operation of centralized and decentralized reservoirs was proposed for flood reduction and resilience
about flooding volume in urban drainage systems [21]. Resilience studies applied to urban cities and
drainage systems were introduced.
Resilience of urban area to flooding has also been studied. In urban drainage systems, the details
of the flooding and the measures were conducted to mitigate such floods in the future [22]. Reflexivity,
knowledge and adaptation were used for describing the potential urban form of a flood-resilient
urban area [23]. A methodology of urban flooding resilience was aimed to be organized into a
software tool the choice of vulnerability indicators and the integration of the point of view of various
stakeholders [24]. Learning and action alliances for the integration of flood risk management into
urban planning was suggested [25]. Urban flooding resilience was introduced as a key focus of
flooding management [26].
The resilience of earlier research focused on the flood volume in urban drainage systems. Failure
in urban drainage systems was defined as flooding or malfunctioning of urban drainage facilities [21].
A regional classification was required because flooding depth is not linked to flooding damage in
some areas, but it is linked to flooding damage in other areas. In this study, multi-dimensional flood
damage analysis (MD-FDA) was applied to divide the target area into several subareas according to
the status of land use and to obtain regional damage functions between flooding volume and damage.
The results of minutely flooding volume at each node by rainfall runoff simulation were obtained and
were converted to minutely flooding damage. The resilience index for flooding damage was suggested
and applied to the target area.

2. Methodologies
This study consists of six parts. The first part is application of MD-FDA for obtaining flooding
damages at each subarea. The second part is obtaining damage functions based on relations between
flooding volume and damage at each subarea. The third part is generation of synthetic rainfall data
for rainfall runoff simulation. The fourth part is obtaining the results of minutely flooding volume
by rainfall runoff simulations. The fifth part is conversion of flooding volume to flooding damage by
damage functions. The last part is application of the results of flooding damage to resilience index.
The flowchart for this study is shown in Figure 1.
Water 2017, 9, 428 3 of 15
Water 2017, 9, 428 3 of 14
Water 2017, 9, 428 3 of 14

Figure 1. Resilience evaluation flowchart for this study.

Figure 1. Resilience evaluation flowchart for this study.
Figure 1. Resilience evaluation flowchart for this study.
2.1. Multi-Dimensional Flood Damage Analysis
2.1. Multi-Dimensional Flood Damage Analysis
2.1. Multi-Dimensional Flood Damage Analysis
The MD-FDA was developed for economic analysis in flood control projects [27]. It is based on
The MD-FDA was developed for economic analysis in flood control projects [27]. It is based on the
the flooding depthwas
The MD-FDA determined
developed byfor
administrative districts.
economic analysis The control
in flood components of[27].
projects the ItMD-FDA
is based are
on
flooding depth determined by administrative districts. The components of the MD-FDA are damage
damage
the to humans,
flooding building damage,
depth determined damage to districts.
by administrative contents Theof buildings,
componentsfarmland
of the damage,
MD-FDAcrop are
to humans, building damage, damage to contents of buildings, farmland damage, crop damage,
damage, to
damage inventory
humans,damage,
buildingand public damage
damage, facility damage. The of
to contents schematic of estimating
buildings, flood damage
farmland damage, crop
inventory damage, and public facility damage. The schematic of estimating flood damage by MD-FDA
by MD-FDA
damage, is shown
inventory in Figure
damage, and2.public facility damage. The schematic of estimating flood damage
is shown in Figure 2.
by MD-FDA is shown in Figure 2.

Figure 2. Schematic of estimating flood damages by multi-dimensional flood damage analysis (MD-FDA).
Figure 2. Schematic of estimating flood damages by multi-dimensional flood damage analysis (MD-FDA).
Figure
Damage2. typesSchematic
in MD-FDA of estimating floodfrom
are different damages by multi-dimensional
each other. floodthedamage
In residential areas, damage of
analysis (MD-FDA).
buildings
Damage andtypesthe indamage
MD-FDA of are
contents
differentof buildings
from each areother.classified as structure
In residential areas, the anddamage
contents,
of
respectively. In agricultural areas, the buried and lost areas of rice paddy
buildings and the damage of contents of buildings are classified as structure and contents, fields and the damage of
Damage
respectively. types
crops are categorized in MD-FDA
as farmland
In agricultural are
areas, anddifferent from each
crops, respectively.
the buried other.
and lost areasThe In residential
industrial
of rice paddyarea areas,
covers
fields the
and the damage
tangible
damage of
assets
of
buildings
crops are and
and inventories. theInfrastructure
categorized damage of contents
as farmland andofcrops,
represents buildings are classified
flood-related The as
damage
respectively. ofstructure andcovers
roads, bridges,
industrial area contents, respectively.
railroads,
tangibleandassets
river
In
andagricultural
structures. The areas,
inventories. person the buried
represents
Infrastructure and
death
representslostandareas of ricedamage
refugees.
flood-related paddy
The fields
MD-FDA
of and the damage
requires
roads, bridges, flooding ofdepth
railroads, crops are
data,
and river
categorized
frequency of
structures. as
The farmland
rainfall,
personand and crops,
Geographic
represents respectively.
deathInformation
and refugees. The
Systemindustrial
The (GIS)
MD-FDA area covers
land requires tangible
use information.
floodingThe assets
depthvalueand
of
data,
inventories.
frequency of Infrastructure
flooding damage is obtained
rainfall, represents
and Geographic flood-related
by calculating
Information damage
the property
System costof based
(GIS) roads, bridges,
on
land use land railroads,
use and The
information. theand river
relevant
value of
portion ofdamage
flooding flooding is depth andby
obtained area. The ratiothe
calculating of damage
propertytocost
flooding
baseddepth in residences
on land use and the is shown in
relevant
Table 1 [28].
portion of flooding depth and area. The ratio of damage to flooding depth in residences is shown in
Table 1 [28].
Water 2017, 9, 428 4 of 15

structures. The person represents death and refugees. The MD-FDA requires flooding depth data,
frequency of rainfall, and Geographic Information System (GIS) land use information. The value
of flooding damage is obtained by calculating the property cost based on land use and the relevant
portion of flooding depth and area. The ratio of damage to flooding depth in residences is shown in
Table 1 [28].

Table 1. The ratio of damage by flooding depth in residences.

Depth (m)
0~0.5 0.5~1.5 1.5~2.5 Over 2.5 Comments
Types
The ratio of damage in
15 40 83 100 -
detached houses (%)
The ratio of damage in n is the number of
15/n 40/n 83/n 100/n
apartments and buildings (%) floors

The ratio of damage by flooding depth in buildings is the same as it is in apartments. The ratio
of damage by flooding depth in buildings in which the contents, tangible assets, and inventories are
damaged is shown in Table 2 [28].

Table 2. The ratio of damage by flooding depth in contents of buildings.

Depth (m)
0~0.5 0.5~1.0 1.0~2.0 2.0~3.0 Over 3.0
Types
The ratio of damage in
14.5 32.6 50.8 92.8 100
contents of buildings (%)
The ratio of damage in
25 50 80 100 100
tangible assets (%)
The ratio of damage in
15 30 60 100 100
inventories (%)

The values in Tables 1 and 2 represent the ratios of damage to flooding depth for each factor.
The factors are houses, apartments, contents of buildings, tangible assets, and inventories. The data in
Tables 1 and 2 are somewhat outdated and can be replaced if new data is uploaded. The property value
of each factor is required to obtain the results of flooding damage. The ratio of damage by flooding
depth in farmland, tangible assets, inventories and crops of agricultural area and industrial area is
shown in Table 3 [28].

Table 3. The ratio of damage by flooding depth in crops.

Depth (m)
0~0.5 0.5~1.0 1.0~2.0 2.0~3.0 Over 3.0
Types
Farmland (%) 0 0 100 100 100
Tangible assets (%) 23.2 45.3 78.9 96.6 100
Inventories (%) 12.8 26.7 58.6 89.7 100
Depth (m) Under 1.0 Over 1.0
Duration (day) Less than 1 1~2 3~4 5~6 Over 7 All duration
Crops Rice paddy 14 27 47 77 95 100
(%)
Field 35 51 67 81 95 100

In agricultural area, the property value of farmlands consists of rice paddy and dry field. The price
per area of rice paddy and dry field is based on the data of Korea Appraisal Board [29], and each
Water 2017, 9, 428 5 of 15

buried area and dry field are priced differently. The property value of farmlands in agricultural area
can be calculated by multiplying areas of rice paddy and dry field by assessment about price of the
buried and lost areas. The property value of crops in agricultural area consists of types of crops, areas
of rice paddy and dray field and consumer price index. The cost by types of crops is based on the
production cost of agricultural products [29]. Areas of rice paddy and dry field are same as farmland.
The consumer price index is also based on the data of Statistics Korea [29]. Table 4 shows the estimation
of property values [27].

Table 4. Estimation of property values.

Classification Requirements Calculation

(1) Number of houses and building
Property value of
(2) Price per area (won/m2 )
Structure (house) structures = (1) × (2) ×
(3) Number of floors
(3) × (4)
(4) Construction industry deflator
Residential area
(1) Assessment about the price of contents
Property value of contents
Contents (household) (2) Number of households
= (1) × (2) × (3)
(3) Consumer price index
(1) Areas of rice paddy and dry field Property value of contents
Farmlands (area)
(2) Assessment about price of the buried and lost areas = (1) × (2)
Agricultural area (1) Types of crops
Property value of crops =
Crops (area) (2) Areas of rice paddy and dry field
(1) × (2) × (3)
(3) Consumer price index

(1) Assessment about price for industrial type Property value of tangible
Tangible assets (won)
Industrial area (2) Number of works assets and inventories =
Inventories (won) (3) Consumer price index (1) × (2) × (3)

In industrial areas, the property value of tangible assets consists of assessment about price for
industrial type, number of works and consumer price index. Assessment about price for industrial
type is based on the data of Statistics Korea [29]. Consumer price index is also based on the data of
Statistics Korea [29]. The property value of industrial area is calculated by assessment about price for
industrial type, number of works and consumer price index.
Flooding volume and flooding depth obtained by Storm Water Management Model (SWMM)
simulation and the results of property value calculated in Table 3 were used for the calculation of
flooding damage. Flooding damage was calculated using Equation (1) [28]:

Fd = Pv × R a × Rd , (1)

where Fd is the flooding damage, Pv is the property value, Ra represents the ratio of flooding area, and
Rd is the ratio of damage to flooding depth.

2.2. Generation of Synthetic Rainfall Data

In this study, synthetic rainfall data was required for rainfall runoff simulations. The design
of drainage facilities in urban drainage systems is based on the Huff distribution [30]. The Huff
distribution includes four quartiles and each quartile has a different peak time. The damage functions
of the Huff distribution for Jeongup, Korea are shown in Equations (2)–(5), respectively [31]:

y = 30.417x6 − 95.785x5 + 113.87x4 − 62.09x3 + 13.469x2 + 1.1268x − 0.0019, (2)

y = −30.678x6 + 93.204x5 − 101.35x4 + 43.908x3 − 4.5765x2 + 0.4869x + 0.0014, (3)

y = 36.029x6 − 98.986x5 + 95.279x4 − 38.622x3 + 7.4086x2 − 0.1088x + 0.0002, (4)

y = −29.739x6 + 81.235x5 − 81.962x4 + 37.734x3 − 7.1157x2 + 0.8466x − 0.0001, (5)

Water 2017, 9, 428 6 of 15

where y refers to the cumulative rainfall ratio and x represents the cumulative time ratio. In Korea,
including Jeongup, the third quartile of the Huff distribution was suggested as a proper rainfall
distribution [32]. The process of generating synthetic rainfall data consists of three steps. The first step
is the distribution of cumulative rainfall amount according to the rainfall duration. The second step
is the acquisition of rainfall distribution according to the rainfall duration by using the results of the
first step. The final step is to use the amount of rainfall amount and duration to generate synthetic
rainfall data using the Huff distribution. The process of generating synthetic rainfall data using the
Huff distribution is shown in Figure 3.
Water 2017, 9, 428 6 of 14

Water 2017, 9, 428 6 of 14

Figure 3. Generation of synthetic rainfall data by Huff distribution.

Figure 3. Generation of synthetic rainfall data by Huff distribution.

2.3. Development of ResilienceFigure Index Based on

3. Generation Floodingrainfall
of synthetic Damage data by Huff distribution.
2.3. Development of Resilience Index Based on Flooding Damage
In previous
2.3. Development studies, various
of Resilience resilience
Index indices Damage
Based on Flooding for urban drainage systems were suggested to
In previous
determine systemstudies,
resiliencevarious resilience
in urban indices
draining for urban
areas. Almostdrainage
all of thesystems wereindices
resilience suggested to
were
determine In system
previous studies, in
resilience various
urban resilience
draining indices
areas. for urban
Almost all drainage
of the systemsindices
resilience were suggested
were to
calculated
calculated using flooding volume obtained by rainfall runoff simulations. As mentioned before, the
usingdetermine
flooding system resilience inby
urban draining
runoffareas. Almost allAs of the resilience indices were
resilience indexvolume obtained
is required rainfall
to evaluate appropriate simulations.
flooding damage mentioned
in urbanbefore, the resilience
drainage systems
calculated
index is flooding using
required to flooding
evaluate volume obtained
appropriate by
flooding rainfall
damage runoff simulations.
in urban As mentioned
drainage systems before,
because the
flooding
because
resilience index
volume
is
is different
required to
from
evaluate
flooding
appropriate
damage.
flooding
The comparison
damage in
of
urban
the resilience
drainage
index
systems
volume
based is different
on theflooding from
floodingvolume flooding
volume damage.
and flooding The comparison of the resilience
4. of the resilience index the
index based on
because is different from damage is shownThe
flooding damage. in Figure
comparison
flooding volume and flooding damage is shown in Figure 4.
based on the flooding volume and flooding damage is shown in Figure 4.

FigureFigure
Figure 4.Comparison
Comparison
4.4. Comparison of of
of resilience
resilience index
index
resilience based
based
index on on
on
based flooding
flooding volume
volume
flooding and
andflooding
volume flooding
and damage.
damage.
flooding (a) (a)
damage.
Conceptual
Conceptual illustration
illustration of resilience
of resilience based
based on
onon flooding
flooding volume; (b) Conceptual illustration of
(a) Conceptual illustration of resilience based floodingvolume;
volume;(b)
(b)Conceptual
Conceptualillustration
illustration of
of
resilience
resilience based on flooding volume.
resiliencebased
based on
on flooding
flooding volume.
volume.
Failure in Figure 4b occurred if flooding also occurred in residential and commercial areas.
Failure in Figure 4b occurred if flooding also occurred in residential and commercial areas.
Figure 4 shows the difference between resilience based on the flooding volume and flooding damage.
Figure 4 shows the difference between resilience based on the flooding volume and flooding damage.
In this study, the concept in Figure 4b was applied to evaluate the system resilience in the target area.
In this study,
Utility the concept
performance in in Figure
urban 4b wassystems
drainage applied to evaluate
should be ablethe system
to show theresilience in thedrainage
status of urban target area.
Utility performance
systems in urban
for each time. drainage
In this systems
study, the status should
of urbanbe able tosystems
drainage show theforstatus offrequencies
various urban drainage
of
systems for each time. In this study, the status of urban drainage systems for various
rainfall compared to the extreme rainfall with 100-year frequency. Utility performance for frequencies
each time of
Water 2017, 9, 428 7 of 15

Failure in Figure 4b occurred if flooding also occurred in residential and commercial areas.
Figure 4 shows the difference between resilience based on the flooding volume and flooding damage.
In this study, the concept in Figure 4b was applied to evaluate the system resilience in the target area.
Utility performance in urban drainage systems should be able to show the status of urban drainage
systems for each time. In this study, the status of urban drainage systems for various frequencies of
rainfall compared to the extreme rainfall with 100-year frequency. Utility performance for each time
can be calculated by Equation (6):
D
u ( T )i = 1 − i , (6)
De
where u(T)i represents the utility performance function at time i, Di represents the flooding damage
(cost) based on selected rainfall data at time i, and De is the flooding damage (cost) based on extreme
rainfall data at time i. There are two requirements. The first requirement is that the rainfall durations
Di and De should be the same. The second requirement is that the frequency of De should be higher
than that of Di . A high value of utility performance function means that the urban drainage system in
the target watershed has the ability to resist system failure including flooding damage. The resilience
based on flooding damage is calculated by the utility performance functions at each time. The resilience
index showing the degree of resilience according to the flooding damage when rainfall events of each
frequency occur by comparing flooding damage when extreme rainfall occurs. The resilience based on
the flooding damage in urban drainage systems is calculated by Equation (7):
Z T
1
Rd = u( T )dT, (7)
T 0

where Rd represents the resilience in urban drainage systems and T is the total time rainfall runoff
simulation. In this study, 15 rainfall events generated by the third quartile of the Huff distribution were
used for calculation of resilience in the target urban drainage system. These 15 rainfall occurrences
had durations of 30, 60, and 90 min, and frequencies of 10, 30, 50, 70, and 100 years. The rainfall data
with the 100-year frequency was selected from among the five frequencies as extreme rainfall data.
Other rainfall data with other frequencies was used for selected rainfall data.

3. Applications and Results

3.1. Study Area

Jeongup is an urbanized city of Jeollabuk-do province in Korea. Many apartments and buildings
are located in Sintaein basin, Jeongup. The total drainage area of Sintaein basin is 67.9 ha. This area
was inundated six times in ten years, from 2002 to 2011. In 2011, there was inundation by extreme
rainfall over 400 mm. The sewer network in Sintaein basin was constructed to simulate rainfall runoff.
The sewer network in the drainage area, including 175 subcatchments and 175 conduits, is shown in
Figure 5.
There are five subareas: A1 (17.6 ha), A2 (21.5 ha), A3 (0.7 ha), A4 (15.1 ha), and A5 (13.0 ha). A1 is
upstream in the drainage area and is a residential area with schools and apartments. A2 is midstream
in the drainage area and includes both residential and commercial areas. A3 consists of buildings and
empty land. A4 is a commercial area including stations and churches. A4 is susceptible to inundation
because it receives the discharge from A1, A2, and A3. A5 is downstream in the drainage area and
includes public facilities such as offices, post offices, and schools.
 Jeongup is an urbanized city of Jeollabuk-do province in Korea. Many apartments and buildings
are located in Sintaein basin, Jeongup. The total drainage area of Sintaein basin is 67.9 ha. This area
was inundated six times in ten years, from 2002 to 2011. In 2011, there was inundation by extreme
rainfall over 400 mm. The sewer network in Sintaein basin was constructed to simulate rainfall runoff.
The sewer
Water 2017, 9, network
428 in the drainage area, including 175 subcatchments and 175 conduits, is shown
8 of in
15
Figure 5.

Figure 5. Sewer network in target area (from Google Earth Imagery ©, Google Inc.).
Figure 5. Sewer network in target area (from Google Earth Imagery ©, Google Inc.).

There are five subareas: A1 (17.6 ha), A2 (21.5 ha), A3 (0.7 ha), A4 (15.1 ha), and A5 (13.0 ha). A1
3.2. Conversion from Flooding Volume to Flooding Damage
is upstream in the drainage area and is a residential area with schools and apartments. A2 is
midstream in the drainage
The calculation areatoand
of damage includes
human both residential
is excluded and commercial
from this study because it isareas. A3 to
difficult consists of
calculate
buildings and empty
and it is uncertain. land.
The A4 is avalue,
property commercial area
the ratio of including
flooding areastations andratio
and the churches. A4 is susceptible
of damage to flooding
to inundation
depth because
are required it receives
to calculate the discharge
flooding damage. The fromfirstA1, A2,
step andcalculation
is the A3. A5 isofdownstream
the propertyin the
value
drainage
about each area andIn
factor. includes publicvalue
the property facilities such as offices,
of structures post offices,
in residential areas,and schools.
Table 5 shows the number of
building types by subareas.

Table 5. The number of residence types by subareas.

Subareas
A1 A2 A3 A4 A5
Types
Detached houses 89 36 44 41 49
Apartments 5 0 0 0 0
Buildings 17 16 7 9 22

The price per area of residence types in Table 6 is quoted from the data of Korea Appraisal
Board [33]. Construction industry deflator is based on the data of Korea Development Institute [34].
The property value of structure in residential areas is calculated by multiplying the number of houses
and building, price per area, number of floors and construction industry deflator in each subarea.

Table 6. The price per area of residence types.

Types Detached Houses Apartments Buildings

Price per area (won/m2 ) 1,385,000 1,690,000 853,000

In the property value of contents in residential areas, assessment about the price of contents is
calculated as the household appraisal value (12,182,399 won/household) for each household [33].
There is no agricultural area in the target watershed and the property value of the agricultural area is
not calculated. For calculating the property value of industrial areas, the number of works in the target
watershed is shown in Table 7.
Water 2017, 9, 428 9 of 15

Table 7. The number of works in target watershed.

Subareas A1 A2 A3 A4 A5
Manufacturing industry 0 0 0 0 0
Electronic enterprise Gas and water utility 0 3 1 4 1
Waste water and environmental restoration 0 0 0 1 0
Construction industry 0 0 0 2 0
Wholesale business and retail trade 13 18 4 24 38
Transportation industry 0 0 1 0 0
Accommodation and restaurant business 4 38 7 18 25
Broadcasting and information industry 0 2 0 2 0
Financial and insurance business 0 4 0 2 0
Real estate industry 0 1 0 2 0
Science and technology services 0 11 2 22 5
Business facilities management and
0 0 0 0 1
business support services
Public administration 0 1 0 2 0
Educational services industry 0 3 0 1 3
Health care and social welfare 0 10 2 8 3
Art and leisure service 0 8 2 6 3
Association and private service 1 3 0 1 1
Water 2017, 9, 428 9 of 14

Total property value of all subareas including residential and industrial area is shown in Table 8.
ThereTotal
is noproperty
property value
value of all subareas values
of agricultural including residential
in all and industrial
subareas because there isarea is shown inarea
no agricultural Table
in
8. There is
all subareas. no property value of agricultural values in all subareas because there is no agricultural
area in all subareas.
Table 8. The results of property value of subareas.
Table 8. The results of property value of subareas.
Subareas A1 A2 A3 A4 A5
Subareas A1 A2 A3 A4 A5
Property
Property value of
value of structure
structure(won)
(won) 8,908,320,000
8,908,320,000 2,505,465,000
2,505,465,000 1,855,900,000
1,855,900,000 3,711,800,000
3,711,800,000 2,041,490,000
2,041,490,000
Property value of contents (won) 755,308,738 328,924,773 243,647,980 487,295,960 268,012,778
Property value of contents (won) 755,308,738 328,924,773 243,647,980 487,295,960 268,012,778
Property value of tangible assets (won) 19,293,116 710,845,218 225,385,429 937,125,269 276,831,416
Property
Propertyvalue
valueofoftangible assets
inventories (won)
(won) 19,293,116
7,200,137 710,845,218
18,552,083 225,385,429
4,331,518 937,125,269
26,725,653 276,831,416
23,541,351
Property
Total value of inventories
property value (won)(won) 7,200,137
9,690,121,991 18,552,083
3,563,787,074 4,331,518
2,329,264,927 26,725,653
5,162,946,882 23,541,351
2,609,875,545
Total property
Area (m value
2) (won) 9,690,121,991
184,100 3,563,787,074
114,190 2,329,264,927
7,430 5,162,946,882
125,310 2,609,875,545
71,060
Property valueArea
per(m 2)
area (won/m2 ) 184,100
52,635 114,190
31,209 7,430
313,495 125,310
14,201 71,060
36,728
Property value per area (won/m2) 52,635 31,209 313,495 14,201 36,728

Synthetic rainfall data generated by the third quartile of Huff distribution is used as input data
Synthetic rainfall data generated by the third quartile of Huff distribution is used as input data
to obtain the flooding area through SWMM simulations. Flooding areas when the total amount of
to obtain the flooding area through SWMM simulations. Flooding areas when the total amount of
synthetic rainfall is 80 mm and 150 mm among various results are shown in Figure 6.
synthetic rainfall is 80 mm and 150 mm among various results are shown in Figure 6.

Figure 6. Flooding areas.

In a previous study, damage functions between flooding damage and flooding volume were
calculated [35]. Damage functions between flooding volume and flooding damage were obtained by
MD-FDA [35]. The type of damage functions was the sifting power function. In this study, the same
damage functions were used for conversion from flooding volume to flooding damage. Damage
functions in each subarea are shown in Equations (8)–(12). These are the damage functions for A1,
Water 2017, 9, 428 10 of 15

In a previous study, damage functions between flooding damage and flooding volume were
calculated [35]. Damage functions between flooding volume and flooding damage were obtained
by MD-FDA [35]. The type of damage functions was the sifting power function. In this study, the
same damage functions were used for conversion from flooding volume to flooding damage. Damage
functions in each subarea are shown in Equations (8)–(12). These are the damage functions for A1, A2,
A3, A4, and A5, respectively:

D = 1.107 × 106 × (V − 63.9)0.46012 , (8)

D = 5.927 × 105 × (V )0.44332 , (9)

D = 1.331 × 107 × (V − 291.5)0.25969 , (10)

D = 4.074 × 105 × (V )0.5774 , (11)

D = 3.549 × 104 × (V )0.9496 , (12)

where D is the flooding damage and V is the flooding volume in urban drainage systems. Rainfall
runoff simulations for obtaining the flooding volume were conducted using SWMM [36]. The results
of minutely flooding volume were converted to minutely flooding damage by damage functions.
The conversion process from flooding volume to flooding damage is shown in Figure 7.
Water 2017, 9, 428 10 of 14

Figure 7. Conversion process from flooding volume to flooding damage.

Figure 7. Conversion process from flooding volume to flooding damage.
3.3. Resilience Based on Flooding Damage
3.3. Resilience Based on Flooding Damage
In this study, 100-year frequency rainfall was selected as the extreme rainfall for the application
ofIn this
the study, 100-year
resilience frequency
index. The rainfall was
result of resilience selected
in the targetas theisextreme
area rainfall
shown in for the application of
Table 9.
the resilience index. The result of resilience in the target area is shown in Table 9.
Table 9. The result of resilience in the target area.

Frequency (Years)
10 30 50 70
Duration (Minutes)
30 0.933741 0.895701 0.876585 0.860621
60 0.932841 0.868809 0.839358 0.826333
Water 2017, 9, 428 11 of 15

Table 9. The result of resilience in the target area.

Frequency (Years)
10 30 50 70
Duration (Minutes)
30 0.933741 0.895701 0.876585 0.860621
60 0.932841 0.868809 0.839358 0.826333
90 0.924555 0.851202 0.813135 0.797292

As frequency increases, the resilience in the urban drainage system decreases. As duration
increases, the resilience in the urban drainage system decreases. This means that this drainage system
has low resilience when rainfall occurs with a high frequency and for a long duration. The values of
resilience and
Water 2017, 9, 428 time are shown in Figures 8–10. 11 of 14
Water 2017, 9, 428 11 of 14

1
1
10 year
10 year
30 year
0.8 30 year
0.8 50 year
50 year
70 year
70 year
0.6
0.6

0.4
0.4

0.2
0.2

0
00 20 40 60 80 100 120
0 20 40 60 80 100 120
Time (min)
Time (min)

Figure 8.
Figure 8. Variance of
of resilience at
at 30-min duration.
duration.
Figure 8. Variance
Variance of resilience
resilience at 30-min
30-min duration.

In Figure
In Figure 8,8, the
the value
value of of resilience
resilience isis lowest
lowest at at about
about 20 20 min
min in in all
all frequencies
frequencies including
including 10, 10, 30,
30,
50, In Figure
and 70 8, the
years. Aftervalue
40 of resilience
min, the is lowest
resilience at about
returned to 20 in
one min allin all frequencies
cases. As frequency including
increases, 10, the
30,
50, and 70 years. After 40 min, the resilience returned to one in all cases. As frequency increases, the
50, and 70
resilience in years.
in the After
the urban 40 min,
urban drainage the
drainage system resilience returned
system decreases. to
decreases. Flooding one in
Flooding occurs all cases.
occurs when As frequency
when rainfall
rainfall occurs increases,
occurs with the
with aa 10-
10-
resilience
resilience in
year frequency the urban
frequency because drainage
because the the designsystem decreases.
design frequency
frequency of Flooding
of the
the sewer occurs
sewer network when
network in rainfall
in the occurs
the target
target area with
area is a
is 10 10-year
10 years.
years.
year
frequency
Failure becausedamage)
(flooding the design frequency
duration for of
allthe sewer network
frequencies is in theThe
similar. target area is
reason is 10 years.
that the Failure
time of
Failure (flooding damage) duration for all frequencies is similar. The reason is that the time of
(flooding damage)
concentration in duration
the target for allisfrequencies
area 20 min and isthe
similar.
amount Theof reason is that
rainfall in thetarget
the time ofareaconcentration
is drained
concentration in the target area is 20 min and the amount of rainfall in the target area is drained
in the target
rapidly. Thearea is 20
result minresilience
of the
the and the amount
goes downof rainfall
down to 0.4 in
0.4 in the target
in 10-year
10-year area is drained
frequency and totorapidly.
below 0.2 The
0.2 inresult
30-year of
rapidly. The result of resilience goes to frequency and below in 30-year
the resilience
frequency. The goes down
The result
result of to 0.4
of resiliencein 10-year
resilience goes frequency
goes down
down to to 0.1 and
0.1 in to
in 50-yearbelow 0.2
50-year frequency in
frequency and 30-year
and to frequency.
to below
below 0.1 The
0.1 in result
in 70-year
70-year
frequency.
of resilience
frequency. This goes down
This result to
result means 0.1
means thatin 50-year
that the
the areafrequency
area is and
is susceptible
susceptible whento below
when rainfall0.1 in 70-year
rainfall occurs
occurs for frequency.
for aa short This
short duration
duration result
and
frequency. and
means
in large that the
amounts. area is susceptible when rainfall occurs for a short duration and in large amounts.
in large amounts.

Figure 9.
Figure 9. Variance
Variance of
of resilience
resilience at
at 60-min
60-min duration.
duration.
Figure 9. Variance of resilience at 60-min duration.
In Figure
In Figure 9,
9, the
the value
value ofof resilience
resilience isis lowest
lowest atat about
about 4040 min
min in
in all
all frequencies
frequencies including
including 10,
10, 30,
30,
50, and
50, and 70
70 years.
years. After
After 60
60 min,
min, the
the resilience
resilience returned
returned toto one
one inin all
all cases.
cases. Although
Although the
the shape
shape ofof the
the
graph for the 60-min duration is similar to that for 30 min, it spreads significantly to the
graph for the 60-min duration is similar to that for 30 min, it spreads significantly to the left and rightleft and right
in the
in the graph
graph for
for the
the 60-min
60-min duration.
duration. Generally,
Generally, the the lowest
lowest value
value in
in the
the 60-min
60-min duration
duration isis higher
higher
than that in the 30-min duration, though the resilience in the urban drainage
than that in the 30-min duration, though the resilience in the urban drainage system decreases as system decreases as
frequency increases. The result of resilience goes down to 0.6 in a 10-year frequency and to 0.3 in aa
frequency increases. The result of resilience goes down to 0.6 in a 10-year frequency and to 0.3 in
Water 2017, 9, 428 12 of 15

In Figure 9, the value of resilience is lowest at about 40 min in all frequencies including 10, 30,
50, and 70 years. After 60 min, the resilience returned to one in all cases. Although the shape of the
graph for the 60-min duration is similar to that for 30 min, it spreads significantly to the left and right
in the graph for the 60-min duration. Generally, the lowest value in the 60-min duration is higher
than that in the 30-min duration, though the resilience in the urban drainage system decreases as
frequency increases. The result of resilience goes down to 0.6 in a 10-year frequency and to 0.3 in a
30-year frequency. The results of resilience in a 50-year frequency and a 70-year frequency go down to
0.2. This result means that the area is better able to withstand rainfall of 60-min duration than rainfall
of 30-min duration. The resilience in this area, with rainfall occurring at a frequency of over 10-years,
goes down dramatically. The results of resilience with rainfall occurring at 50- and 70-year frequencies
are similar, which means that the increase in the amount of rainfall from a 50-year frequency to a
70-year frequency will not result in additional flooding damage.
Water 2017, 9, 428 12 of 14

1
10 year
30 year
0.8 50 year
70 year
Resilience

0.6

0.4

0.2

0
0 20 40 60 80 100 120
Time (min)

Figure 10. Variance of resilience at 90-min duration.

Figure 10. Variance of resilience at 90-min duration.

In Figure 10, the value of resilience is lowest at about 60 min in all frequencies including 10, 30,
50, andIn Figure 10, After
70 years. the value of resilience
80 min, is lowest
the resilience at about
returned to 60
one minin in
allall frequencies
cases. Althoughincluding
the shape10, of
30,the
50,
and 70 years. After 80 min, the resilience returned to one in all cases. Although
graph for the 90-min duration is similar to that for 30 min and 60 min, it spreads most widely to thethe shape of the graph
for the
left and90-min
right induration
the graph is for
similar to that duration.
the 90-min for 30 minGenerally,
and 60 min, theitlowest
spreads mostinwidely
value to theduration
the 90-min left and
right
is in the
higher graph
than thatfor
inthe
the90-min
30-minduration.
durationGenerally,
and similar theto
lowest
that invalue
the in the 90-min
60-min duration
duration. is higher
The result of
than that in the 30-min duration and similar to that in the 60-min duration.
resilience goes down to 0.6 in 10-year frequency and to 0.4 in 30-year frequency. The result The result of resilience goes
of
down to 0.6 in 10-year frequency and to 0.4 in 30-year frequency. The result
resilience goes down to 0.3 in 50-year frequency and to 0.2 in 70-year frequency. This result meansof resilience goes down
to 0.3the
that in area
50-year frequency
is better able to and to 0.2 in rainfall
withstand 70-year of
frequency. This result
90-min duration thanmeans thatofthe
rainfall 30-area
andis60-min
better
able to withstand rainfall of 90-min duration than rainfall of 30- and 60-min duration.
duration. The resilience in this area, with rainfall occurring at a frequency of over 10 years, goes below The resilience in
thisbecause
0.5 area, with rainfall
of the occurring
design frequency at aof
frequency
the sewerofnetwork
over 10 years, goes below
in the target area. 0.5 because of the design
frequency of the sewer network in the target area.
4. Conclusions
4. Conclusions
In urban drainage systems, flooding damage can pose a critical threat to human life. In the
In urban drainage systems, flooding damage can pose a critical threat to human life. In the process
process of making flooding control plans, it will be useful to have a resilience index based on flooding
of making flooding control plans, it will be useful to have a resilience index based on flooding damage.
damage. Flooding volume in subareas is different from flooding damage in subareas because some
Flooding volume in subareas is different from flooding damage in subareas because some subareas are
subareas are immediately damaged by a certain amount of flooding while other subareas are not. A
immediately damaged by a certain amount of flooding while other subareas are not. A new resilience
new resilience index based on flooding damage has been suggested.
index based on flooding damage has been suggested.
Multi-dimensional flood damage analysis has been selected to obtain the flooding damage in
Multi-dimensional flood damage analysis has been selected to obtain the flooding damage in each
each subarea. Sintaein basin in Jeongup, Korea, has been selected as a target area. Sintaein basin is
subarea. Sintaein basin in Jeongup, Korea, has been selected as a target area. Sintaein basin is divided
divided into five subareas, namely, A1, A2, A3, A4, and A5, according to land use. In subareas of the
into five subareas, namely, A1, A2, A3, A4, and A5, according to land use. In subareas of the target
target area suggested in the previous study, damage functions were used for the conversion from
area suggested in the previous study, damage functions were used for the conversion from flooding
flooding volume to flooding damage [32]. To obtain the results of flooding volume, the SWMM was
volume to flooding damage [32]. To obtain the results of flooding volume, the SWMM was used in
used in rainfall runoff simulations.
rainfall runoff simulations.
Synthetic data for rainfall runoff simulations was distributed using the Huff distribution [30].
The Huff distribution is generally used for the design of drainage facilities in Korea. The third quartile
of the Huff distribution was selected because it was appropriate for application in Korea [32]. Fifteen
rainfall events had durations of 30, 60, and 90 min and frequencies of 10, 30, 50, 70, and 100 years,
generated by the third quartile of the Huff distribution, which were used for the calculation of
resilience in the target area.
Water 2017, 9, 428 13 of 15

Synthetic data for rainfall runoff simulations was distributed using the Huff distribution [30].
The Huff distribution is generally used for the design of drainage facilities in Korea. The third quartile
of the Huff distribution was selected because it was appropriate for application in Korea [32]. Fifteen
rainfall events had durations of 30, 60, and 90 min and frequencies of 10, 30, 50, 70, and 100 years,
generated by the third quartile of the Huff distribution, which were used for the calculation of resilience
in the target area.
The results of flooding volume were converted into the results of flooding damage by damage
functions after obtaining the results of flooding volume in each subarea by simulating SWMM.
The flooding damage in each subarea was used to calculate the utility performance per minute.
The results of the utility performance in each subarea were combined with the results of the utility
performance in the target area. The resilience in the target area was calculated by the results of the
utility performance. The rainfall events with 100-year frequency were selected because the calculation
of resilience requires the determination of extreme rainfall. The resilience decrease as the duration
decreased because rainfall events with short duration burden urban drainage systems.
In 30-min duration, the resilience go down at about 10-min and rise up at about 40-min. The
result of the resilience in 30-min goes down to 0.4 in 10-year frequency and to below 0.2 in 30-year
frequency. The results of the resilience go down to 0.1 in 50-year frequency and to below 0.1 in 70-year
frequency. In 60-min duration, the resilience goes down at about 20-min and rises up at 50-min. The
results of the resilience in 60-min go down to 0.6 in 10-year frequency and to 0.3 in 30-year frequency.
The results of resilience in 50-year frequency and 70-year frequency go down to 0.2. In 90-min duration,
the resilience goes down at 30 min and rises up at 80-min. The results of resilience go down to 0.6 in
10-year frequency and to 0.4 in 30-year frequency. The results of resilience go down to 0.3 in 50-year
frequency and to 0.2 in 70-year frequency.
In this study, the contents of uncertainty and vulnerability in socioeconomic and physical aspects
were not included in estimating flooding damage. Future studies would yield more detailed results if
there were uncertainty and vulnerability in socioeconomic and physical aspects. This study excluded
the damage to humans and did not consider the total damage of urban areas because there are resident
and floating people in urban areas. Age and gender in the damage to humans should be considered.
The resilience index based on flooding damage can be used for the evaluation of flood control plans
such the installation, replacement, and rehabilitation of drainage facilities. Resilience index can be
used as a fundamental indicator for the regional evaluation considering the flooding damage. Because
various structural measures such as installation of new pump stations and detention reservoirs and
replacement and rehabilitation of sewer conduits need to be evaluated, the resilience index based on
flooding damage will support decision making in flood control plans. In future research, the resilience
index based on flooding damage will be simplified by reports and recorded data.

Acknowledgments: This research was supported by a grant (13AWMP-B066744-01) from the Advanced Water
Management Research Program funded by the Ministry of Land, Infrastructure, and Transport of the Korean
government and The National Research Foundation (NRF) of Korea, funded by the Korean government (MSIP)
(No. 2016R1A2A1A05005306).
Author Contributions: Eui Hoon Lee carried out the survey of previous studies and wrote the draft of the
manuscript. Eui Hoon Lee revised the draft until the final version of the manuscript and simulated the inlet/outlet
operation. Eui Hoon Lee and Joong Hoon Kim conceived the original idea of the proposed method.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Haughton, G. Developing sustainable urban development models. Cities 1997, 14, 189–195. [CrossRef]
2. Chiesura, A. The role of urban parks for the sustainable city. Landsc. Urban Plan. 2004, 68, 129–138. [CrossRef]
3. Pickett, S.T.; Cadenasso, M.L.; Grove, J.M. Resilient cities: Meaning, models, and metaphor for integrating
the ecological, socio-economic, and planning realms. Landsc. Urban Plan. 2004, 69, 369–384. [CrossRef]
4. Egger, S. Determining a sustainable city model. Environ. Model. Softw. 2006, 21, 1235–1246. [CrossRef]
Water 2017, 9, 428 14 of 15

5. Blackmore, J.M.; Plant, R.A. Risk and resilience to enhance sustainability with application to urban water
systems. J. Water Resour. Plan. Manag. 2008, 134, 224–233. [CrossRef]
6. Coaffee, J. Risk, resilience, and environmentally sustainable cities. Energy Policy 2008, 36, 4633–4638.
[CrossRef]
7. Ahern, J. From fail-safe to safe-to-fail: Sustainability and resilience in the new urban world.
Landsc. Urban Plan. 2011, 100, 341–343. [CrossRef]
8. Djordjević, S.; Butler, D.; Gourbesville, P.; Mark, O.; Pasche, E. New policies to deal with climate change and
other drivers impacting on resilience to flooding in urban areas: The CORFU approach. Environ. Sci. Policy
2011, 14, 864–873. [CrossRef]
9. Park, J.; Seager, T.P.; Rao, P.S.C. Lessons in risk-versus resilience-based design and management.
Integr. Environ. Assess. Manag. 2011, 7, 396–399. [CrossRef] [PubMed]
10. Lansey, K. Sustainable, robust, resilient, water distribution systems. In Proceedings of the 14th Water
Distribution Systems Analysis Conference (WDSA 2012), Adelaide, South Australia, 24–27 September 2012;
Engineers Australia: Adelaide, Australia, 2012; p. 1.
11. Desouza, K.C.; Flanery, T.H. Designing, planning, and managing resilient cities: A conceptual framework.
Cities 2013, 35, 89–99. [CrossRef]
12. Jabareen, Y. Planning the resilient city: Concepts and strategies for coping with climate change and
environmental risk. Cities 2013, 31, 220–229.
13. Park, J.; Seager, T.P.; Rao, P.S.C.; Convertino, M.; Linkov, I. Integrating risk and resilience approaches to
catastrophe management in engineering systems. Risk Anal. 2013, 33, 356–367. [PubMed]
14. Butler, D.; Farmani, R.; Fu, G.; Ward, S.; Diao, K.; Astaraie-Imani, M. A new approach to urban water
management: Safe and SuRe. Procedia Eng. 2014, 89, 347–354.
15. Hwang, H.; Lansey, K.; Quintanar, D.R. Resilience-based failure mode effects and criticality analysis for
regional water supply system. J. Hydroinform. 2015, 17, 193–210.
16. Mugume, S.N.; Gomez, D.E.; Fu, G.; Farmani, R.; Butler, D. A global analysis approach for investigating
structural resilience in urban drainage systems. Water Res. 2015, 81, 15–26. [PubMed]
17. Siekmann, T.; Siekmann, M. Resilient urban drainage—Options of an optimized area-management.
Urban Water J. 2015, 12, 44–51.
18. Kotzee, I.; Reyers, B. Piloting a social-ecological index for measuring flood resilience: A composite index
approach. Ecol. Indic. 2016, 60, 45–53.
19. Mugume, S.N.; Melville-Shreeve, P.; Gomez, D.; Butler, D. Multifunctional urban flood resilience
enhancement strategies. Proc. Inst. Civ. Eng. Water Manag. 2017, 170, 115–127.
20. Mugume, S.N.; Butler, D. Evaluation of functional resilience in urban drainage and flood management
systems using a global analysis approach. Urban Water J. 2016, 1–10. [CrossRef]
21. Lee, E.H.; Lee, Y.S.; Joo, J.G.; Jung, D.; Kim, J.H. Flood Reduction in Urban Drainage Systems: Cooperative
Operation of Centralized and Decentralized Reservoirs. Water 2016, 8, 469.
22. Gupta, K. Urban flood resilience planning and management and lessons for the future: A case study of
Mumbai, India. Urban Water J. 2007, 4, 183–194. [CrossRef]
23. White, I. The absorbent city: Urban form and flood risk management. Proc. Inst. Civ. Eng.-Urban Des. Plan.
2008, 161, 151–161. [CrossRef]
24. Barroca, B.; Bernardara, P.; Mouchel, J.M.; Hubert, G. Indicators for identification of urban flooding
vulnerability. Nat. Hazards Earth Syst. Sci. 2006, 6, 553–561. [CrossRef]
25. Van Herk, S.; Zevenbergen, C.; Ashley, R.; Rijke, J. Learning and Action Alliances for the integration of flood
risk management into urban planning: A new framework from empirical evidence from The Netherlands.
Environ. Sci. Policy 2011, 14, 543–554. [CrossRef]
26. Lamond, J.E.; Proverbs, D.G. Resilience to flooding: Lessons from international comparison. Proc. Inst. Civ.
Eng.-Urban Des. Plan. 2009, 162, 63–70. [CrossRef]
27. Choi, S.A.; Yi, C.S.; Shim, M.P.; Kim, H.S. Multi-Dimensional Flood Damage Analysis (I): Principle and
Procedure. J. Korea Water Resour. Assoc. 2006, 50, 1–9.
28. Study on Economical Analysis Method of Flood Control Project; Ministry of Construction and Transportation:
Seoul, Korea, 2004.
29. Statistics Korea. Available online: kostat.go.kr (accessed on 25 February 2017).
30. Huff, F.A. Time distribution of rainfall in heavy storms. Water Resour. Res. 1967, 3, 1007–1019. [CrossRef]
Water 2017, 9, 428 15 of 15

31. Korea Precipitation Frequency Data Server. Available online: www.k-idf.re.kr (accessed on 27 February 2017).
32. Yoon, Y.N.; Jung, J.H.; Ryu, J.H. Introduction of design flood estimation. J. Korea Water Resour. Assoc. 2013,
46, 55–68.
33. Korea Appraisal Board. Available online: www.kab.co.kr (accessed on 22 February 2017).
34. Korea Development Institute. Available online: www.kdi.re.kr (accessed on 3 March 2017).
35. Choi, H.; Lee, E.H.; Joo, J.G.; Kim, J.H. Determining optimal locations for rainwater storage sites with the
goal of reducing urban inundation damage costs. KSCE J. Civ. Eng. 2016, 1–13. [CrossRef]
36. United States Environmental Protection Agency (USEPA). Storm Water Management Model User’s Manual
Version 5.0; EPA: Washington, DC, USA, 2010.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).