International Journal of Disaster Risk Reduction 74 (2022) 102955

Contents lists available at ScienceDirect

International Journal of Disaster Risk Reduction

journal homepage: www.elsevier.com/locate/ijdrr

Comprehensive flood vulnerability analysis in urban communities:

Iowa case study
Yazeed Alabbad a, b, *, Ibrahim Demir a, c
a
Department of Civil and Environmental Engineering, University of Iowa, Iowa City, IA, USA
b
Department of Civil Engineering, King Saud University, Riyadh, Saudi Arabia
c
IIHR - Hydroscience and Engineering, University of Iowa, Iowa City, IA, USA

A R T I C L E I N F O A B S T R A C T

Keywords: Floods are one of the most prevalent types of natural disasters affecting communities worldwide.
Flood damage It is expected to persist with increasing magnitude and frequency as a result of climate change,
Impact assessment resulting in both direct and indirect negative consequences. The flood impact assessment is
Flood vulnerability considered as a key component of flood risk management strategies, such as benefit-cost analysis
Floodplain management for mitigation planning. This research aims to provide a comprehensive analysis of socio-
Urban flooding
economic impacts of the 100 and 500-year riverine flood scenarios for major communities in
the State of Iowa, United States of America. The analysis includes impact on essential facilities,
businesses, vehicles, loss of life, amount of debris transported downstream, and displaced pop­
ulation, using Hazards-United States (HAZUS) and the Hydrologic Engineering Center’s Flood
Impact Analysis (HEC-FIA) functions along with GIS-based analysis. Also, we have investigated
the floodplain vulnerability using the combination of flood depth and velocity. Due to the dif­
ferences in the topography and the spatial distribution of buildings and infrastructure, and the
absence of sufficient flood prevention measures within the study area, our findings illustrate that
some communities will suffer significant damage and losses during flooding. This research could
aid communities in making more effective decisions regarding damage reduction, prioritizing
mitigation actions, and preparing for future flood events.

1. Introduction
Increased population, rapid urbanization, and climate change are all contributing to aggravate flood risk in communities around
the world [1,2]. Flooding can cause adverse effects on people, properties, agriculture, and infrastructure and cause significant eco­
nomic damage and loss of life [3,4]. From 1980 to 2021, the annual flood disasters in the U.S. caused nearly $3.8 billion in damage and
killed 15 people [5]. Also, productivity, labor hours, and GDP may all be affected by flooding [6]. The state of Iowa has been severely
impacted by flood events over the past two decades. In 2019, flood disasters struck many parts of Iowa, causing an estimated $1.6
billion in damage [7].
Flood hazard changes over time for communities with severe consequences [8]. By 2100, the average projected 100-year floodplain
will increase 45%, with annual flood-related damage of 750 million [9]. State and local authorities perform measures (e.g., dams,
levees) to control floodwater. However, the continued development in flood-prone areas and change in weather patterns due to climate
change may make existing flood controls unable to withstand future flood events, resulting in increased flood damage [10]. Additional

* Corresponding author. Department of Civil and Environmental Engineering, University of Iowa, Iowa City, IA, USA.
E-mail address: [email protected] (Y. Alabbad).

https://doi.org/10.1016/j.ijdrr.2022.102955
Received 28 January 2022; Received in revised form 31 March 2022; Accepted 4 April 2022
Available online 9 April 2022
2212-4209/© 2022 Elsevier Ltd. All rights reserved.
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

plans and strategies can be developed, such as riverine planning to better cope with, respond to, and recover from flood events in the
long and short-term in order to minimize flood risk and increase the resilience of communities [11].
Effective flood risk assessment leads to better understanding of flood risk and actions for flood resilience [12]. Flood risk can be
defined as a combination of the probability of a flood event with the consequences associated with it [13–15]. In other words, when a
flood hits a community (Hazard), it can affect people and infrastructure (Exposure) and generate devastating economic, environ­
mental, and social losses (Vulnerability). The flood risk assessment and management process include identifying the risk, evaluating
the significance of the risk, and then taking actions to control and manage the risk [16]. The key question is how much risk there is, and
how it will be mitigated. Identification and monitoring of flood events [17] and risk and loss assessment have become necessary to
prioritize mitigation projects, allocate resources, and analyze the benefit-cost of mitigation measures. Different stakeholders, including
state and local governments, city planners, emergency response providers, and the public, can use the assessment to make successful
flood risk mitigation decisions [18,19]. Web systems have been actively used in risk communication while providing extensive ca­
pabilities for terrain analysis [20] and geospatial forecasting [21].
Flood hazard models coupled with the built environment have been studied extensively from different perspectives. Risk models
can be used in flood risk management for risk communication [22] and evacuation plans [23]. Researchers have utilized flood hazard
models to explore flood damage on buildings [24–26], road network topology [27,28], and agriculture [29]. In addition, Alabbad et al.
[30] developed a data analytics framework for efficient flood mitigation practices at the property level.
The potential flood-related damage, including physical, economic, and social, can be estimated using HAZUS model software
developed by FEMA [31]. Assessment of direct and indirect losses provides a comprehensive picture of flood impacts and helps focus
on effective projects to minimize the consequences. Previous studies have utilized HAZUS functions focusing on estimating the direct
impacts of flood damage, including structure and content [32,33]. Indirect losses such as business losses may be worse than the direct
impacts. The absence of reliable methodologies and data may be an obstacle for estimating the indirect losses [34]. Researchers have
performed a comprehensive flood damage analysis, including indirect losses (e.g., business interruption) using HAZUS, and proven its
capability in generating reliable results [35–37]. FEMA accepts the estimated losses generated by HAZUS to be used for community
flood hazard planning [38]. This planning is required for communities to obtain post-flood disaster funding from FEMA. Also, it could
help develop strategies for flood mitigation, such as buyout and analyzing the cost-benefit of the measures [39,40].
Flooding can have an economic and functional impact on critical facilities (e.g., hospitals, fire departments). The necessity of these
facilities during flooding becomes more critical than in normal circumstances due to its capability to provide essential services. For
example, the fire department has a vital role in rescuing people from dangerous areas. However, these facilities might be already
exposed to flooding, reducing their functionality. Qiang [41] spatially analyzed critical infrastructure exposed to the 100-year flood
extent in the U.S. with a limited focus on varying flood events and the functionality assessment.
Flood Impact Analysis (HEC-FIA) software developed by the Hydrologic Engineering Center can perform flood-based analysis on
the potential for life loss [42]. HEC-FIA is a geospatially based model for a single flood event that takes into account flood depths,
detailed structure inventory with population, and evacuation information. It is considered as advanced software for life loss estimation
[43]. Researchers have explored the capability of HEC-FIA to estimate the expected life loss resulting from dam failure [44], storm
surge [45], and levee breach case studies in the U.S [46]. Evacuation effectiveness is crucial for fatality number estimation [47].
During the 2016 flood in Freeport, IA, the sound of the basement wall collapsing was the only flood notification for some people [48].
Based on the HEC-FIA findings, decision-makers can select appropriate measures [49] to reduce the potential loss of life, including
ensuring early warning of major floods utilizing various technologies and approaches, evacuating planning, and shelter in place
options.
The flood’s depth, velocity, and duration are essential elements in determining flood losses [50,31]. Due to the complexity and
uncertainty of producing detailed flood characteristics for each flood event, flood depth is considered reasonable to be employed to
estimate flood-related damage [51]. However, other flood characteristics should be included to extract a wide range of advanced flood
hazard assessments. The combination of flood depth and velocity can go beyond the traditional estimation of flood damage for
floodplain vulnerability classifications for better flood management [52]. Researchers have studied flood depth-velocity (DxV)
functions to create flood-hazard classification mapping [53,54] and assess the stability of buildings [55], vehicles [56], and people
[57] during flooding. Smith et al. [58] have created hazard classifications using the product of DxV for overall floodplain assessment as
recommended by USACE [59] and Australian Institute for Disaster Resilience [52]. Following the computation of flood consequences,
flood risk management strategies such as resource allocation can be evaluated and prioritized.
This research requires significant resources and skills in data integration, spatial analysis, and computation [60], which may be a
challenge for most communities with limited resources. Due to complex nature of flood hazard and the significant adverse impacts
associated with flooding, using tools such as HAZUS and HEC-FIA contributes to mitigation decisions in communities. They are
considered the most comprehensive tools for flood risk assessment [61,62]. Many Iowan communities lack flood hazard planning,
which may lead communities not to be eligible for post-flood disaster funds from FEMA [38]. Based on major scientific databases (i.e.,
Google Scholar), Iowa has produced detailed flood hazard impact analysis only for a limited number of communities (i.e., [35]).
Due to the lack of a comprehensive flood loss estimation analysis within Iowa communities, this study focuses on assessment of the
flood-related losses for several Iowa communities during the 100 and 500-yr riverine flood events. We have computed damage to
critical facilities (i.e., police station), vehicles, and other building-related losses (i.e., income & wage losses, relocation expenses) along
with the amount of debris and displaced population. Also, there is a need to investigate areas prone to life losses during flooding in
order to prepare well for future floods. To date, no previous analysis has examined the loss of life resulting from flood scenarios in Iowa.
Our analysis investigates the potential life losses for major Iowa communities caused by the 100 and 500-yr riverine flood scenarios
and the economic impacts associated with it. It can help decision-makers identify vulnerable population and areas, take mitigation

2
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

actions such as moving the most vulnerable people out of the hazard areas and minimize the economic consequences. In addition,
another focus of this research is to assess the vulnerability of people, vehicles, and buildings to flooding using not only flood depth but
also flood velocity. The study categorizes the floodplain areas into different classes (i.e., safe, unsafe for vehicles) based on the depth-
velocity interaction for several Iowa communities to be used for flood risk management, including deciding where to develop new
residential areas and prioritizing flood management efforts (e.g., for evacuation routes to examine vehicle stability).
These assessments could increase knowledge regarding expected consequences resulting from flood events, provide a compre­
hensive insight into the vulnerability of Iowa communities to flooding and help make potential improvements within the floodplain,
either structure or non-structural, to minimize flood-related losses and as a guide to improving the response to flooding and speed up
the recovery.
The rest of the paper is devoted to the flood loss and damage analysis methodology (section 2), including the case study, data
preparation, and damage functions. Research results along with discussion are presented in section 3. The conclusion and further
analysis follow these sections.

2. Methodology
2.1. Case study
The study has been conducted on selected communities in the State of Iowa, United States of America. Cedar, Des Moines, and Iowa
Rivers are running inside the State. Iowa communities distribute in 99 counties with a total population of 3.1 million. The past two
decades have seen a history of recurring flood events in Iowa, possessing serious damage to people, agriculture, and infrastructure at
various levels. In 2008, 85 counties in Iowa were affected by a significant flood event and were designated as federal disaster areas,
resulting in $10 billion in damage cost and more than 40,000 people affected [63]. Our work provides a comprehensive look at the
consequences of the 100 (1% chance) and 500-year (0.2% chance) riverine flood events on eight major Iowa communities named Des
Moines, Cedar Rapids, Iowa City, Waterloo, Cedar Falls, Charles City, Adel, and Independence (Fig. 1). River stage and discharge
associated with the two analyzed flood events for each community are presented in Table 1. The reason behind selecting those
communities is due to the data availability (i.e., flood models) along with considering different topographical, residential surveys, and
river stages (Table 1).

2.2. Data preparation

Table 2 summarizes the datasets incorporated in this research to analyze the socio-economic impacts resulting from flood events.
The datasets are collected from different sources, including but not limited to federal and state agencies (e.g., FEMA, USACE) and
organizations. The interaction between flood models and building and vehicle inventory has led to the estimation of damage and losses
to buildings, vehicles, and lives, as well as floodplain classification analysis. Building and vehicle depth-damage functions from HAZUS
are utilized to extract the percentage of damage.
Our analysis also requires building-related information (e.g., area) for business interruption functions. To estimate vehicle damage,
it is required to determine the used and new vehicle costs and the type and number of vehicles within the impact area. Bridge damage

Fig. 1. The geo-spatial representation for the analyzed cities.

3
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

Table 1
Population, area, and river stages for the study area.

City Population Area (mi2) 100-yr flood 500-yr flood

3
Stage (ft) Discharge (ft /s) Stage (ft) Discharge (ft3/s)

Des Moines 214,133 88.18 31.5 72,000 34 106,000

Cedar Rapids 137,710 71.62 23.5 94,100 28.5 122,000
Iowa City 74,828 25.6 29 31,010 32.5 45,260
Waterloo 67,314 61.39 26 105,000 30.7 137,000
Cedar Falls 40,713 28.93 99.5 94,786 103 122,323
Charles City 7,396 6.22 24.75 32,800 27.25 41,550
Adel 6,153 4.87 21.5 35,000 23 45,000
Independence 6,064 6.01 24.2 35,538.4 29 50,795

Table 2
Datasets with description and source information used in this research.

Data Description

Flood model The model consists of 3.28-foot resolution flood depth and velocity raster for the inundated areas using a coupled 1D/2D
hydrodynamic MIKE FLOOD model software. The model utilizes LiDAR data to delineate the accurate centerline of streams,
drainage areas, basin boundaries, and cross-section geometry and develop digital elevation models (DEMs). Peak annual
streamflow records are obtained from the USGS and the USACE, and the land-use dataset is extracted from The United States
Department of Agricultural (USDA). Flood maps are created for different return period flows (i.e., 2-, 5-, 10-, 25-, 50-, 100-,
500-yr) and hosted on ifis.iowafloodcenter.org.
Source: Iowa Flood Center (IFC) [64,65]
Building inventory & essential It includes building-spatial distribution and building-related information (e.g., demographic, occupancy type, area)
facilities Source: HAZUS [31]; 2010 U S. Census Bureau [66]; HIFLD Open datasets [67]; County Tax Office
Essential facility depth-damage It defines the percentage of damage based on a flood depth and an occupancy type
functions Source: HAZUS [31]
Business interruption functions Equations include the cost of losses ($) (e.g., income) per square foot for different occupancy types, reconstruction times, %
recapture, and % owned occupied.
Source: HAZUS [31]
Vehicle inventory The inventory provides vehicle-spatial distribution, vehicle values, and counts
Source: [68]; 2000 Census building areas and 2008 Dun & Bradstreet data [69]
Vehicle depth-damage function It is the percentage of damage based on the flood depth and vehicle type
Source: HAZUS [31]
Highway bridges It is a database that contains a spatial representation of bridge points with bridge-related information
Source: National Bridge Inventory [70]
Road segments It is a spatial representation of road networks
Source: OpenStreetMap [71]
Life loss functions It illustrates the fatality zones for each occupancy type and the average fatality rate.
Source: HEC-FIA [42]
Floodplain classification Curves represent thresholds for the vulnerability of people, vehicles, and buildings in the floodplain. Source: Smith et al. [58]

during flooding is analyzed using information about highway bridges. The potential life loss during flooding is estimated according to
flood depth, occupancy types, and the number of people occupying properties. Flood-prone areas are analyzed to examine the stability
of people, vehicles, and buildings using the vulnerability curves.
Data integration and analysis have been performed in the Geographic Information System (ArcGIS 10.5.1) to produce the economic

Fig. 2. The overall methodology for the comprehensive flood loss and damage analysis.

4
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

and human impacts during flooding. Fig. 2 illustrates the research methodology for flood impact assessment, which includes three
phases: HAZUS, HEC-FIA, and floodplain vulnerability. Sections 2.3, 2.4, and 2.5 provide detailed explanations of the approach used in
each phase.

2.3. Community flood impact analysis

HAZUS utilizes Geographic Information System (GIS) to analyze data and generate potential flood-related losses [31]. HAZUS can
be used to estimate a wide range of damage, including damage to people, buildings, vehicles, and infrastructure. It can operate at three
levels based on the available data, user expertise, and spatial scale analysis. Level 1 analysis produces results using default HAZUS data
and models. Level 2 analysis requires a combination of default HAZUS data with more detailed user-supplied data (e.g., high-resolution
flood model and updated inventory data). The user can incorporate data resulting from more advanced studies and models, such as
modifying the depth-damage function, into the Level 3 analysis. In this research, Level 2 analysis was performed to estimate flood
hazard damage. HAZUS depth-damage functions are employed to estimate damages associated with different flood return periods. We
incorporated flood simulation models from Iowa Flood Center (IFC) while building, vehicle, and other infrastructure inventory data
extracted directly from the HAZUS 4.2 inventory database, using the 2010 Census data and primarily RSMeans 2018 economic values.
Detailed HAZUS inventory data can be found in Ref. [72].
HAZUS can run the analysis at different spatial scales using a national inventory database. The census block level is the smallest
spatial scale for data aggregation. The user can assess losses resulting from a single flood event or a series of flood occurrences,
allowing for annualized damage estimates. HAZUS distributes buildings evenly within each census block, which may affect the ac­
curacy of the results. For example, HAZUS may count damage for unflooded buildings located inside partially flooded census blocks.
Damage estimates in HAZUS are area-weighted to account for the block-to-block variation in flood depth. Our analysis is performed on
urban regions where buildings tend to be uniformly distributed per census block [34]. User-defined data from tax assessor offices can
improve the spatial representation of structures and gain more accurate flood loss estimation [73]. Direct damage to structure and
content within the study area has been studied extensively by our team [74]. Our work focuses on other flood-related damage and
losses, including business interruption, displaced population and shelter needs, vehicle damage, essential facility and infrastructure
damage, and transported debris.

2.3.1. Business interruption

Business interruption includes income, relocation, rental, and wage losses. Equations and parameters associated with each loss are
presented as follows:
Income losses: This estimation relies on the restoration time, income recapture factor, income per day, and area of occupancy
(Equation (1)). The losses cover each building subjected to flooding.
Loss ​ of ​ Incomei ($) = (1 − IRFi ) ​ * ​ Ai,j ​ * ​ IDi *LOFi,j ​ (1)

where all i belongs to a specific occupancy and all j belongs to a specific flood depth, IRFi is the income recapture factor, Ai,j is the
occupancy area, IDi is the income per day (per square foot), and LOFi,j is the loss of function time (days).
Relocation losses: Occupants need to relocate once a building reaches a damage threshold of 10%. The damage costs include
shifting and transferring (disruption costs) as well as renting a temporary place. Some occupancies (e.g., theaters) will not be
considered under this estimation. Relocation expenses can be estimated through the area of occupancy, disruption, and rental costs,
recovery time, and percent owner-occupied (Equation (2)).
[ ( )]
Relocation ​ Expensesi ($) ​ = ​ Ai,j * (1 − %OOi ) * (DCi ) + %OOi * DCi + RENTi * RTi,j (2)

where all i belongs to a specific occupancy and all j belongs to a specific flood depth, Ai,j is the occupancy area, OOi is the percent owner
occupied, DCi is the disruption cost, RENTi is the rental cost, and RTi,j is the recovery time (days).
Rental losses: It can be estimated through the area of occupancy, rental costs, recovery time, and percent owner-occupied
(Equation (3)). The losses are calculated for buildings with damage of more than 10%.
Rental ​ Lossesi ($) = (1 − %OOi )*Ai,j *RENTi *RTi,j (3)

where all i belongs to a specific occupancy and all j belongs to a specific flood depth, OOi is the percent owner occupied, Ai,j is the
occupancy area, RENTi is the rental cost, and RTi,j is the recovery time (days).
Wage losses: This estimation relies on the restoration time, wage recapture factor, wage per day, and area of occupancy (Equation
(4)). The losses cover each building exposed to flooding.
Wage ​ Lossesi ($) = (1 − WRFi )*Ai,j *WDi *LOFi,j (4)

where all i belongs to a specific occupancy and all j belongs to a specific flood depth, WRFi is the wage recapture factor, Ai,j is the
occupancy area, WDi is the wage per day (per square foot), and LOFi,j is the loss of function time (days).

2.3.2. Displaced population and shelter needs

The displaced population is determined once the flood depth equals or is more than 0.5 ft. Estimating the number of people who
need public shelters due to flooding depends on displaced people with factors related to age information and household income

5
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

acquired from the US Census database.

2.3.3. Vehicle damage

The HAZUS procedure to estimate flood damage of vehicles requires determining vehicle inventory within a study area, day and
nighttime vehicle allocation, vehicle ages and values, and a percent of loss damage associated with the flood depth. Based on the
occupancy type (i.e., residential, commercial) and its square footage, the number of parked vehicle spaces is estimated. A utilization
percentage of parked vehicles is used to distribute vehicles with regard to the time of day in order to estimate the actual number of
vehicles at risk of flooding.
In HAZUS, vehicles are classified into three categories: cars, light trucks, and heavy trucks. This classification is proposed using data
from the [68]; the US Department of Transportation’s comprehensive Truck Size and Weight Study (TSWS), and the 1995 National
Personal Transportation Survey (NPTS). The age distribution of vehicles and the ratio of trucks to cars were derived from NADA data.
Based on data obtained from a variety of sources, including NADA, the average prices of the new vehicles used by HAZUS are $22,
618.47 (cars), $20,969.2 (light truck), and $76,087.67 (heavy trucks). The average value of a used vehicle is estimated to equal 50% of
the value of a new vehicle. The total value of vehicles in an area is calculated by multiplying the total number of vehicles by the
percentage of cars/light trucks/heavy trucks, the percentage of new/used vehicles, and the average prices of vehicles that fall into both
categories.
To estimate the damage to vehicles, the percentage of damage for each vehicle type with regard to the flood level is determined
using depth damage functions (Fig. 3), which is developed by HAZUS with support from experts in this field. HAZUS uses a default
vehicle height above the ground (car: 1.5 ft, light truck: 2.7 ft, large truck: 5 ft), resulting in an estimated damage of less than 20%. The
damage percentage increases as the flood level reaches to significant components (i.e., dashboard) of vehicles. Our analysis is applied
at a census block level and based on the assumption that all vehicles are present within the census block at the time of the flood. The
area-weighted depth damage approach is used to account for flood depth variation along the census block. The vehicle damage cost of a
census block will be the percentage of the census block affected at a given flood depth multiplied by the percentage of damage at that
depth multiplied by the vehicle values (Equation (5)).
Vehicle ​ damage ​ ($) = (%) ​ Di,j *(%) ​ CBj *VV($) (Eq. 5)

where all i belongs to a specific occupancy and all j belongs to a specific flood depth, Di,j is the damage percentage, CBj the percentage
of impacted census block, and VV is the vehicle value.

2.3.4. Essential Facilities and Infrastructure Damage

Critical facilities (e.g., police station) and utilities (e.g., treatment plants) are represented as a point location and classified into
different occupancy types. As an assumption, some facilities will have a basement and others not. The model can estimate their po­
tential direct impacts of flooding, including functionality assessment using the depth level at their point-representation. The analysis
assigns default replacement and content values based on the occupancy type. The flood depth will allow users to estimate the damage
percentage from depth-damage curves for each occupancy type (Table 3, Fig. 4) and, then the amount of damage is determined
(Equation (6)). The damage functions provided in this section are only for essential facilities and utilities impacted withing the study
area. Additional damage functions (e.g., hospitals) can be found in HAZUS software package.
Loss($) = (%)D*IV ​ ($) (6)

where D is the damage percentage at a given flood depth and IV is the inventory value.
The national bridge inventory database can be accessed to extract and represent the geolocation of the bridges. Highway bridge
damage is based on the bridge cost, scour potential rating, and the flood scenario. The scour rate for a bridge range from 0 to 9. The
lowest rate means that the bridge is closed due to scouring, and nine indicates that the bridge is open. HAZUS focuses on bridge scour
rating from 1 to 3 for damage estimation. Above this rate, the bridge would not be likely damaged. The probability of bridge failure
depends on the number of spans for a bridge. For example, during the 100-yr flood, bridges rated as three will be 1% and 0.25% failure

Fig. 3. Vehicle depth-damage functions [31].

6
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

Table 3
The percentage of damage for wastewater treatment plant [31].

Flood depth (ft) 0 1 2 3 4 5 6 7 8 9 10

Damage (%) 0 5 8 10 17 24 30 30 30 30 40

Fig. 4. Depth-damage functions for essential facilities [31].

probability for single and continuous bridges, respectively. The damage cost is determined by multiplying the probability of failure by
the bridge cost, and functionality can be estimated (1- the likelihood of failure).

2.3.5. Transported debris

The amount of flood debris transported downstream will be a result of building-related damage. The debris could include finishes
(e.g., carpeting), structural components, and foundation materials. The analysis relies on occupancy area, flood depth, and foundation
types to estimate the debris weight (ton) following a flood event (Table 4). The removal cost is assumed to be $18.50 per cubic yard
[50].

2.4. Life loss assessment during flood events

Expected life losses associated with people’s movement and behavior during a flood event can be determined using HEC-FIA
software [42] which utilizes the simplified LIFESim methodology and runs the analysis using GIS. The necessary geospatial data to
carry out the analysis are flood depth and extent, flood arrival time, detailed building inventory, and evacuation information. HEC-FIA
can access the HAZUS database to import building inventory data into the study area, including population distribution. HEC-FIA uses
an algorithm to redistribute people based on day and nighttime.
The amount of time that the population has to evacuate from the hazard area is the core component in estimating life loss. Thus,
determination of the evacuation time needs detailed information about the evacuation routing, warning delay time, warning diffusion
time, and protective action initiation time. HEC-FIA assumes that people will move out of the hazard areas horizontally (shortest path).
Also, once the flood depth reaches 2 ft at a building (the end of the opportunity to evacuate a structure), the model will implement
vertical evacuation.
Three fatality zones (safe, compromised, chance) will be assigned for the impacted buildings depending on the maximum water
depth and occupancy type (Fig. 5). The fatality zones and their details described below:
Safe: The structure is usually dry or exposed to little or shallow floods unlikely to knock people off their feet. The average fatality

Table 4
An example of the amount of debris resulting from flooding [31].

Occupancy Type Flood depth (ft) Debris weight (ton/1000sq.ft)

Finishes Structure Foundation

Footing Slab on Grade

Single Family Dwelling (With no basement) 0–4 4.1 – – –

4–8 6.8 – – –
>8 6.8 6.5 12 25

7
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

rate for this zone is 0.0002.

Compromised: The structure has been severely damaged by the flood, exposing people to dangerous floodwater. Its average fatality
rate is 0.12.
Chance: Flood victims are often washed downstream or stuck underwater, and their survival is primarily dependent on chance. The
average fatality rate is 0.9145.
Flooding may happen unexpectedly or with little notice [75]. Even though there is a warning system (e.g., siren), flooding can
strike communities at night where people fall asleep, ignoring the warning. Therefore, it is crucial to ensure that people will not be in
the greatest danger of staying home during flooding. Our analysis will consider that people will remain in their homes during the 100
and 500-yr flood events, which means that impacted people have to evacuate vertically. The potential life losses are estimated by
multiplying the number of people caught in buildings by the average fatality rate. We consider the largest number of life losses from the
day and night, and then the analysis will show the benefit of avoiding life losses in 2021-dollar amounts ($7.9 million/life) [14].

2.5. Floodplain vulnerability analysis

Flood vulnerability classification is based on hazard curves proposed by Ref. [58] and recommended by Ref. [52,59]. It is built into
six vulnerability zones by employing the combination of flood depth and velocity (DxV). It can inform spatial mapping of hazardous
areas to vehicles, buildings, and people within the floodplain. As a result, mitigation activities can be performed, such as prioritizing
regions for evacuation.
The combined hazard vulnerability curves can be used for baseline hazard classification (Fig. 6). It sets threshold criteria to
determine vulnerable people, vehicles, and buildings. Additional specific hazard classifications can be more appropriate than the
baseline classification assessment [58]. For example, in evacuation scenarios, specific people and vehicle stability curves can be used to
assess the evacuation routes. These vulnerability curves are based on an international literature search. Experiments utilize the
relationship between a person (height X mass) and flow regime (DxV) to provide guidelines for people’s safety and stability in floods. In
this research, flood behavior impacts on the 100 and 500-yr floodplain will be analyzed using flood hazard vulnerability curves.
Integration, geoprocessing, and visualization of layers have been performed using GIS. After generating the floodplain vulnerability
maps, two-dimensional interaction has been performed between the floodplain vulnerability map and the built environment (build­
ings, road segments) to identify unsafe locations within the studied communities.

3. Results and discussion

3.1. Community flood impact analysis
This section presents and discusses the economic and human impacts of flood-related damage and losses (i.e., businesses, vehicles,
displaced population and shelter needs, essential facilities and infrastructure, debris removal) for the selected communities.
Business Interruption: As can be seen in Fig. 7, the business interruption losses during the 100 and 500-year floods are estimated for
the selected communities. It is obvious that wage losses are the most affected, while losses from rental income seem relatively the least
affected during the two flood scenarios. The figure shows that Cedar Rapids and Waterloo losses increases even though the population
and areas go down.
Des Moines, the largest city in Iowa in terms of population and area, was found to be resilient to flooding compared to Cedar Rapids
and Waterloo, which their losses approximately two times and more than what Des Moines will experience during flooding. The
estimated total business interruption losses for the analyzed cities during the 100 and 500-yr flood events are $836 M and $2.5 B,
respectively. We found that the variation in business interruption losses is attributed to the increase of buildings exposed to floods.
Therefore, limiting new development in floodplains, enforcing new policies for the existing buildings (e.g., raising a structure above

Fig. 5. An example of fatality zones for a one-story building [42].

8
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

Fig. 6. Floodplain vulnerability threshold adapted from Ref. [58].

Fig. 7. Business interruption losses (Millions of dollars) sorted by population.

9
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

the flood level), and relocating the impacted buildings to safe locations all can minimize the business interruption losses.
Displaced Population and Shelter Needs: Floods can last for days or weeks, so people have to move from hazardous areas to safe
places. Not all affected people can pay for temporary relocation rooms, so the analysis considers income and age for estimating persons
seeking shelters. Communities show differential impacts in terms of the number of displaced populations during flooding (Table 5).
During the 100-yr flood, 2516 people of Cedar Rapids will have to seek temporary shelter. Also, the 500-yr flood will affect
approximately 22% of Waterloo’s population significantly. That can be attributed to the number of buildings impacted by flooding.
Vehicle Damage: Vehicles can experience damage and cause significant losses based on the flood depth and the type and distri­
bution of vehicles. The distribution of day and night vehicles is based on the occupancy type of the buildings. Fig. 8 shows the total
vehicle damage for the studied communities during the 100 and 500-yr flood scenarios. The analysis indicates that vehicle losses
during the day appear to be more severe than night losses. That could be a result of shutting down businesses that lead to empty vehicle
parking lots at night. Also, we found a significant link between the number of damaged buildings and the vehicle losses. However, the
population and area of communities appear not strongly connected to increased vehicle losses, as can be seen for Des Moines. Also,
some communities show a big shift in losses between the two flood events, as estimated for Cedar Rapids and Waterloo. Early flood
warning helps move vehicles out of hazard areas and avoid losses.
Essential Facilities and Infrastructure Damage: The damage costs associated with the impacted essential facilities and infrastructure
during flooding are presented in Table 6. The dollar losses include structure and content damage. Under the 100-yr flood, we noticed
only Cedar Rapids could experience damage to one of its police stations and bridges with a total estimated loss of $2.4 M. During the
500-yr flood event, most of the studied communities can experience damage to some essential facilities and bridges. The total damage
estimation will be approximately $77 M. Waterloo will have severe impacts during the 500-yr since fire departments (count:3), police
stations (count:2), and schools (count:3) are not safe from flooding. Allocation of amenities in Iowa City appears to be less vulnerable
to flooding. No critical amenity was damaged in Iowa City during the 100 and 500-yr flood events. Also, all impacted amenities will be
non-functional except the fire and police stations in Cedar Falls. Relocation of critical amenities or using flood mitigation measures
help reduce damage and speed up the response to emergency cases during flooding.
Debris Removal Cost: Debris amounts resulting from building damage during the 100-yr and 500-yr flood events are illustrated in
Table 7. Cedar Falls generates the highest amount of debris under the 100-yr flood, while Waterloo shows significant amounts of debris
during the 500-yr flood. Although Des Moines has the largest population and area among the studied areas, the debris estimation
appears the lowest compared to the other communities during the two scenarios. The total cost of removing the debris during the two
flood scenarios is $0.7 M and $3.6 M, respectively.

3.2. Life loss assessment during flood events

We investigated the expected life loss during the 100-yr and 500-yr flood scenarios. Flood depth is the only flood characteristic
involved in the mortality estimation analysis. We assume that all people decided to remain in their homes during flooding. So,
impacted people will have to seek safety through vertical evacuation. This assumption is considered because of the absence of flood
arrival time data over the study area, which prevents us from analyzing people evacuating horizontally. Also, since flooding can
happen without or with little warning (i.e., flood), our assumption is important to explore vulnerable areas to life losses and take
action. Based on the occupancy type (i.e., residential, commercial), population, and the maximum flood depth, the analysis will assign
lethality zones (Safe, Compromised, Chance) for each building exposed to more than 2 ft flood depth, the default depth for non-
evacuation out of hazard area. Then, we used the average fatality rate (Chance: 91%, Compromised: 12%, Safe: 0.02%) in the liter­
ature for people caught in the affected buildings. Five communities are selected to carry out the life loss analysis.
It is noted that the studied communities are differentially vulnerable to life losses during the two flood events (Table 8). Although
Cedar Rapids is the highest population among the studied communities, its buildings have little life loss impacts compared to other
studied cities. That might be attributed to the existence of efficient flood controls (e.g., levee) that make the floodplain less vulnerable
or the spatial planning of buildings away from the waterways. On the other hand, Charles City appears to experience significant life
losses (50 lives) during the 100-yr flood scenario, while Waterloo is the most vulnerable to life losses among the studied communities
under the 500-yr flood with an estimate of 245 lives. Also, when compared to the 100-year scenario, the number of lives lost in the
analyzed cities increased dramatically during the 500-year flood, reaching twice or more losses. That is because of the increase of flood
depth and extent as well as the number of impacted buildings.
In addition, Table 8 shows the economic losses resulting from the potential fatalities over the study area, which can reach up to
$395 M and $1.9 B under the 100 and 500-yr flood events, respectively. These economic values can be used as benefits of avoided
mortality in the context of implementing benefit-cost analysis for flood mitigation projects.

Table 5
Number of displaced population and persons seeking shelter during flooding.

City 100-yr flood 500-yr flood

Displaced population People seeking shelter Displaced population People seeking shelter

Des Moines 479 13 1475 67

Cedar Rapids 2516 115 6290 296
Iowa City 1224 99 2220 205
Waterloo 174 2 15,306 1104
Cedar Falls 668 38 1115 75

10
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

Fig. 8. Day and night vehicle damage (in $Million) sorted by population.

Table 6
Damage to essential facilities and infrastructure ($1000s).

Facilities Des Moines Cedar Rapids Iowa City Waterloo Cedar Falls

100 yr 500 yr # 100 yr 500 yr # 100 yr 500 yr 100 yr 500 yr # 100 yr 500 yr #

Fire Department – – – – – – – – – $4187 3 – $217 1

Police Station – – – $2367 $4419 1 – – – $4092 2 – $46 1
School – $7750 1 – $1845 1 – – – $2486 3 – – –
Wastewater treatment plant – – – – – – – – – – – – $51,682 1
Highway bridge – $50 1 $41 $81 1 – – – – – – $8 1

Table 7
The amount of debris and removing cost sorted by population.

Cities Debris (ton) Removing cost ($)

100-yr 500-yr 100-yr 500-yr

Des Moines 881 5833 $32,597 $215,821

Cedar Rapids 6752 21,933 $249,824 $811,521
Iowa City 1541 4889 $57,017 $180,893
Waterloo 2164 46,455 $80,068 $1,718,835
Cedar Falls 7624 19,374 $282,088 $716,838

Table 8
The fatality rate and economic losses ($) for the analyzed communities.

Cities No. of life losses Fatalities to 2021 dollars ($7.9 M/life [13];

100-yr 500-yr 100-yr 500-yr

Cedar Rapids 14 30 $111 M $237 M

Waterloo 20 245 $158 M $1936 M
Cedar Falls 12 75 $95 M $593 M
Charles City 50 68 $395 M $529 M
Independence 4 18 $32 M $134 M

The life loss estimates are spatially represented per census block during the 500-yr flood event (Fig. 9). Some blocks are expected to
lose up to 40 people’s lives, as in Cedar Falls and Charles City. These maps can illustrate the most vulnerable census block in terms of
life loss. Therefore, mitigation strategies can take place, including adjusting existing impacted structures to be in a safe lethality zone
(e.g., add more stories), allocating shelters, and prioritizing the evacuation areas.

11
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

Fig. 9. The potential life losses per impacted census block during the 500-yr flood scenario for Cedar Rapids (a), Cedar Falls (b) Charles City (c), Independence (d), and
Waterloo (e).

12
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

3.3. Floodplain vulnerability

Due to data availability, two cities are chosen for studying the comprehensive flood behavior impact. Fig. 10 shows the interaction
between flood depth and velocity during the 100-yr and 500-yr flood events for Iowa City and Adel communities. These maps can

Fig. 10. Floodplain vulnerability classification during the 100-yr flood (left) and 500-yr flood (right) for Adel (a, b) and Iowa City (c, d); SV: small vehicles, C:
children, E: elderly, V: all vehicles, P: all people, SB: structural buildings or buildings without special requirements, B: all buildings).

13
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

inform decision-makers of a wide flood risk planning and management [76], including the decision on road avoidance or entering,
pedestrian safety, amenity and shelter locations, mitigation allocation, where to develop new lands, and critical evacuation locations.
Also, it would be beneficial to transform the analysis into open-access interactive maps [77], so the public can be aware of risk easily
and prepare well.
We examined the building locations acquired from County Tax Office and Microsoft building footprint dataset, and road segments
from Open Street Maps (OSM) located inside the floodplain vulnerability maps. We found many buildings and roads categorized as
unsafe for people, vehicles, and buildings (Table 9). During the 100 and 500-yr flood scenarios, the largest vulnerable zone is within
the Iowa City floodplain that is unsafe for children, the elderly, and vehicles, while the unsafe areas for people, vehicles, and buildings
(without special requirements) are the largest for Adel city. There are 41 building locations categorized as unsafe for small vehicles in
Iowa City under the 100-yr flood, and the number goes up to 2243 buildings during the 500-yr flood. Also, up to 120 Iowa road
segments intersect with unsafe zones. Most of the vulnerable Adel buildings and roads fall in the zone classified as unsafe for children,
the elderly, and vehicles. Floodplain vulnerability classification could enhance floodplain management by using techniques (e.g.,
wetland, floodwall) to reduce the flood depth and velocity.

3.4. Limitations
Conducting flood impact analysis requires obtaining detailed datasets regarding buildings and infrastructure with advanced flood
models. This research attempts to provide comprehensive flood impact analysis using freely available data. However, data quality and
analysis assumptions may affect the results but still can provide a useful estimate for communities. One of the challenges in this
research is the absence of accurate spatial building representation with detailed information. Even though County Tax Office data can
provide accurate spatial representation; however, it lacks critical information (e.g., population information, foundation type) that can
introduce some barriers to completing the analysis.
The absence of a detailed high-resolution flood behavior model (i.e., depth, velocity, duration, and arrival time) may restrict flood
impact analysis. It requires an extensive process that involves high computational and data requirements that may be a challenge in
most cases for communities with limited resources. HAZUS relies on flood depth as the only flood characteristic driver to estimate the
damage and social impacts. This can increase uncertainties in the modeled outcomes. For example, flood duration can play a significant
role in estimating flooded vehicle damage [78]. Also, the analysis assumed that all vehicles were subjected to flooding without
considering the possibility of moving the vehicles out of the hazard areas before flooding. The procedure to carry out some HAZUS
analysis (e.g., income losses) is based on the occupancy characteristics. It might be beneficial to improve the methods by embedding
additional parameters such as the number of people. Also, flood impact assessment (e.g., health impacts, evacuation planning) from
systems like HAZUS can be limited and uncertain at some scale levels.
In the context of life loss estimation, our analysis relied on only flood depth and assumed all people occupied their houses during
flooding. The results are still important to ensure people’s safety staying at their homes during flooding and determine the most
vulnerable areas for life losses. To improve the analysis, evacuation scenarios can be examined, which require detailed information of
flood components, warning systems, and demographic [42]. Besides, it is important to explore life loss analysis caused by flood control
system failure (e.g., a levee breach) where the time is limited for evacuation. Also, HEC-LifeSim software [59] provides advanced
capability to estimate the loss of life during flooding, whether on roads or in structures.
When it comes to floodplain vulnerability classification, analyzing the flood depth-velocity interaction should not be limited to
final levels of flood return periods (i.e., 100-yr, 500-yr) but include all stages of the flood hydrograph to understand the worst flood
behavior that may occur before the peak of flood levels [58]. Also, it is recommended in some cases (e.g., planning for building a
shelter) to consider a specific flood hazard classification (e.g., building stability curve) available [58] instead of the combined criteria
presented in this research.

4. Conclusion
With the increase in the frequency and magnitude of flooding due to human activities and climate change, flood vulnerability
analysis is a crucial stage to enhance the flood risk management cycle (i.e., planning) and, as a result, strengthen community resilience.
This study carried out comprehensive flood damage and loss estimation analysis for several Iowa communities during the 100 and 500-
yr flood scenarios. It has been found that flooding can generate costly economic impacts on essential facilities, infrastructure, busi­
nesses, and vehicles. That may be attributed to the higher number of buildings located in the floodplain. Besides, we noticed that some

Table 9
The area and the number of buildings and roads within floodplain vulnerability regions (Fig. 6).

Iowa City Adel City

Area (mi2) Buildings Road segments Area (mi2) Buildings Road segments

Floodplain vulnerability 100 yr 500 yr 100 yr 500 yr 100 yr 500 yr 100 yr 500 yr 100 yr 500 yr 100 yr 500 yr
H1 0.74 0.75 2090 1591 129 157 0.18 0.15 38 78 31 28
H2 0.45 0.56 41 2243 44 122 0.12 0.12 16 34 23 25
H3 0.77 1.28 120 581 19 100 0.37 0.32 13 24 17 26
H4 0.06 0.32 0 5 1 18 0.59 0.47 10 17 9 9
H5 0.49 0.43 0 5 1 1 0.83 1.19 5 12 5 8
H6 0.06 0.29 0 0 0 0 0.31 0.34 0 0 3 3

14
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

essential facilities (e.g., schools) lie within the floodplain and become non-functional during flooding, which may raise questions about
location-allocation without accounting for flood scenarios. Also, our findings illustrate significant human impacts, such as loss of life,
as seen in Charles City and Waterloo. During large-scale flooding, some of the Iowa City and Adel flood-prone locations were spatially
classified as unsafe for people, vehicles, and buildings. Providing detailed information about vulnerable population and areas due to
flooding in real-time and accessible platforms could open paths for improvements to minimize the consequences of flooding. Results
can be used for existing policy improvement, land use, river basin, and emergency planning, and tell us which and where the mitigation
efforts would be more effective.
Future research can transform this study into a web-based geospatial analytical system [79] or integrate it into virtual reality
environments [80] that promote easily accessible flood impact information for decision-makers, including the public. Mitigation
strategies are often evaluated by only the structure and content values of the buildings for the benefit-cost analysis. Our detailed
economic and human impacts (e.g., income and life losses) can increase the benefits of implementing such measures at the community
and property level. Advanced flood loss analysis components (i.e., arrival time) can be utilized to investigate the evacuation of people
dynamically from hazard areas in order to assess people’s lives caught in buildings or streets.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

References
[1] W. Zhang, G. Villarini, G.A. Vecchi, J.A. Smith, Urbanization exacerbated the rainfall and flooding caused by hurricane Harvey in Houston, Nature 563 (7731)
(2018) 384–388, https://doi.org/10.1038/s41586-018-0676-z.
[2] C. Wasko, S. Westra, R. Nathan, H.G. Orr, G. Villarini, R. Villalobos Herrera, H.J. Fowler, Incorporating climate change in flood estimation guidance,
Philosophical Transact. Royal Soc. 379 (2021) 20190548, https://doi.org/10.1098/rsta.2019.0548, 2195.
[3] F. Dottori, W. Szewczyk, J.C. Ciscar, F. Zhao, L. Alfieri, Y. Hirabayashi, L. Feyen, Increased human and economic losses from river flooding with anthropogenic
warming, Nat. Clim. Change 8 (9) (2018) 781–786, https://doi.org/10.1038/s41558-018-0257-z.
[4] E. Yildirim, I. Demir, Agricultural Flood Vulnerability Assessment and Risk Quantification in Iowa, Science of The Total Environment, 2022, p. 154165, https://
doi.org/10.1016/j.scitotenv.2022.154165.
[5] NOAA National Centers for Environmental Information (NCEI) U.S, Billion-dollar weather and climate disasters. https://www.ncdc.noaa.gov/billions/, 2021.
[6] C. Brown, R. Meeks, Y. Ghile, K. Hunu, Is water security necessary? An empirical analysis of the effects of climate hazards on national-level economic growth,
Phil. Trans. Math. Phys. Eng. Sci. 371 (2013) 20120416, https://doi.org/10.1098/rsta.2012.0416, 2002.
[7] Iowa.gov., Office of the governor of Iowa, Reterieved from, https://governor.iowa.gov/2019/03/governor-reynolds-launches-211-flood-hotline-and-one-stop-
2019-iowa-floods-website-to-assist, 2019.
[8] J.M. Sadler, N. Haselden, K. Mellon, A. Hackel, V. Son, J. Mayfield, J.L. Goodall, Impact of sea-level rise on roadway flooding in the Hampton Roads region,
Virginia, J. Infrastruct. Syst. 23 (4) (2017), 05017006, https://doi.org/10.1061/(ASCE)IS.1943-555X.0000397.
[9] US EPA, Green infrastructure, Retrieved from Manage Flood Risk, https://www.epa.gov/green-infrastructure/manage-flood-risk, 2020.
[10] NAI, How-to guide for infrastructure, Retrieved from, https://asfpmlibrary.s3uswest2.amazonaws.com/FSC/NAI/ASFPM_NAI_Infrastructure_2016.pdf, 2014.
[11] United Nations., Disaster Risk and Resilience, 2012. Retrieved from, https://www.un.org/en/development/desa/policy/untaskteam_undf/thinkpieces/3_
disaster_risk_resilience.pdf.
[12] M.J. Hammond, A.S. Chen, S. Djordjević, D. Butler, O. Mark, Urban flood impact assessment: a state-of-the-art review, Urban Water J. 12 (1) (2015) 14–29,
https://doi.org/10.1080/1573062X.2013.857421.
[13] FEMA, Benefit-cost sustainment and enhancements, Retrieved from, https://www.caloes.ca.gov/RecoverySite/Documents/Benefit%20Cost%20Sustainment.
pdf, 2016.
[14] FEMA, Resilience report, Retrieved from, https://dnr.wi.gov/topic/FloodPlains/documents/UpperFox_ResilienceReport.pdf, 2016.
[15] Construction Industry Research and Information Association (CIRIA), International Levee Handbook, 2013 (London, UK).
[16] O.D. Cardona, M.K. Van Aalst, J. Birkmann, M. Fordham, G. Mc Gregor, P. Rosa, F. Thomalla, Determinants of risk: exposure and vulnerability, in: Managing the
Risks of Extreme Events and Disasters to Advance Climate Change Adaptation: Special Report of the Intergovernmental Panel on Climate Change, Cambridge
University Press, 2012, pp. 65–108, https://doi.org/10.1017/CBO9781139177245.005.
[17] I. Haltas, E. Yildirim, F. Oztas, I. Demir, A comprehensive flood event specification and inventory: 1930–2020 Turkey case study, Int. J. Disaster Risk Reduc. 56
(2021) 102086, https://doi.org/10.1016/j.ijdrr.2021.102086.
[18] H. Xu, M. Windsor, M. Muste, I. Demir, A web-based decision support system for collaborative mitigation of multiple water-related hazards using serious
gaming, J. Environ. Manag. 255 (2020) 109887, https://doi.org/10.1016/j.jenvman.2019.109887.
[19] G. Ewing, I. Demir, An ethical decision-making framework with serious gaming: a smart water case study on flooding, J. Hydroinf. 23 (3) (2021) 466–482,
https://doi.org/10.2166/hydro.2021.097.
[20] M. Sit, Y. Sermet, I. Demir, Optimized watershed delineation library for server-side and client-side web applications, Open Geospat. Data Softw. Standard. 4 (1)
(2019) 1–10, https://doi.org/10.1186/s40965-019-0068-9.
[21] H. Xu, M. Muste, I. Demir, Web-based geospatial platform for the analysis and forecasting of sedimentation at culverts, J. Hydroinf. 21 (6) (2019) 1064–1081,
https://doi.org/10.2166/hydro.2019.068.
[22] L.J. Weber, M. Muste, A.A. Bradley, A.A. Amado, I. Demir, C.W. Drake, W.F. Krajewski, T.J. Loeser, M.S. Politano, B.R. Shea, N.W. Thomas, The Iowa
Watersheds Project: Iowa’s prototype for engaging communities and professionals in watershed hazard mitigation, Int. J. River Basin Manag. 16 (3) (2018)
315–328, https://doi.org/10.1080/15715124.2017.1387127.
[23] V. Radosavljevic, G. Belojevic, N. Pavlovic, Tool for decision-making regarding general evacuation during a rapid river flood, Publ. Heath 146 (2017) 134–139,
https://doi.org/10.1016/j.puhe.2017.01.025.
[24] E. Yildirim, I. Demir, An integrated flood risk assessment and mitigation framework: a case study for middle cedar river basin, Iowa, US, Int. J. Disaster Risk
Reduc. 56 (2021) 102113, https://doi.org/10.1016/j.ijdrr.2021.102113.
[25] J.C. Aerts, N. Lin, W. Botzen, K. Emanuel, H. de Moel, Low-probability flood risk modeling for New York City, Risk Anal. 33 (5) (2013) 772–788, https://doi.
org/10.1111/risa.12008.
[26] S.D. Brody, S. Zahran, W.E. Highfield, H. Grover, A. Vedlitz, Identifying the impact of the built environment on flood damage in Texas, Disasters 32 (1) (2008)
1–18, https://doi.org/10.1111/j.1467-7717.2007.01024.x.
[27] Y. Alabbad, J. Mount, A.M. Campbell, I. Demir, Assessment of Transportation System Disruption and Accessibility to Critical Amenities during Flooding: Iowa
Case Study, Science of The Total Environment, 2021, p. 148476, https://doi.org/10.1016/j.scitotenv.2021.148476.
[28] K.C. Van Ginkel, F. Dottori, L. Alfieri, L. Feyen, E.E. Koks, Flood risk assessment of the European road network, Nat. Hazards Earth Syst. Sci. 21 (3) (2021)
1011–1027, https://doi.org/10.5194/nhess-21-1011-2021.

15
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

[29] F.O. Tapia-Silva, S. Itzerott, S. Foerster, B. Kuhlmann, H. Kreibich, Estimation of flood losses to agricultural crops using remote sensing, Phys. Chem. Earth, Parts
A/B/C 36 (7–8) (2011) 253–265, https://doi.org/10.1016/j.pce.2011.03.005.
[30] Y. Alabbad, E. Yildirim, I. Demir, Flood Mitigation Data Analytics and Decision Support Framework: Iowa Middle Cedar Watershed Case Study, Science of The
Total Environment, 2022, p. 152768, https://doi.org/10.1016/j.scitotenv.2021.152768.
[31] FEMA, Multi-hazard loss estimation methodology, flood model, Retrieved from, https://www.fema.gov/sites/default/files/2020-09/fema_hazus_flood-model_
technical-manual_2.1.pdf, 2020.
[32] C.A. Cummings, P.E. Todhunter, B.C. Rundquist, Using the Hazus-MH flood model to evaluate community relocation as a flood mitigation response to terminal
lake flooding: the case of Minnewaukan, North Dakota, USA, Appl. Geogr. 32 (2) (2012) 889–895, https://doi.org/10.1016/j.apgeog.2011.08.016.
[33] E. Yildirim, I. Demir, Flood Risk Assessment and Quantification at the Community and Property Level in the State of Iowa, 2021, https://doi.org/10.31223/
X5TC92. EarthArxiv, 2803.
[34] E. Tate, C. Munoz, J. Suchan, Uncertainty and sensitivity analysis of the HAZUS-MH flood model, Nat. Hazards Rev. (2015), https://doi.org/10.1061/(ASCE)
NH.1527-6996.0000167.
[35] Story County Iowa, Multi-jurisdictional hazard mitigation plan, Retrieved from, https://cityofcollins.municipalimpact.com/documents/200/Story_County__IA_
Hazard_Mitigation_Plan_Update_2018_Final_-_9.22_MB.pdf, 2018.
[36] C. Kousky, S.M. Olmstead, M.A. Walls, M. Macauley, Strategically placing green infrastructure: cost-effective land conservation in the floodplain, Environ. Sci.
Technol. (2013), https://doi.org/10.1021/es303938c.
[37] H. McGrath, E. Stefanakis, M. Nastev, Sensitivity analysis of flood damage estimates: a case study in Fredericton, New Brunswick, Int. J. Disaster Risk Reduc.
(2015) 379–387, https://doi.org/10.1016/j.ijdrr.2015.09.003.
[38] NOAA, Economic Assessment of Green Infrastructure Strategies for Climate Change Adaptation: Pilot Studies in The Great Lakes Region. Retrieved from, 2014.
https://coast.noaa.gov/data/digitalcoast/pdf/climate-change-adaptation-pilot.pdf.
[39] K.O. Atoba, S.D. Brody, W.E. Highfield, W.J. Merrell, Estimating residential property loss reduction from a proposed coastal barrier system in the Houston-
Galveston region, Nat. Hazards Rev. 19 (3) (2018), 05018006, https://doi.org/10.1061/(ASCE)NH.1527-6996.0000297.
[40] K.S. Nelson, J. Camp, Quantifying the benefits of home buyouts for mitigating flood damages, Anthropocene (2020) 100246, https://doi.org/10.1016/j.
ancene.2020.100246.
[41] Y. Qiang, Flood exposure of critical infrastructures in the United States, Int. J. Disaster Risk Reduc. 39 (2019), https://doi.org/10.1016/j.ijdrr.2019.101240,
101240.
[42] USACE, HEC-FIA, flood impact analysis user’s manual, Retrieved from, https://www.hec.usace.army.mil/software/hec-fia/default.aspx, 2018.
[43] World Bank, Understanding Risk: Review of Open Source and Open Access Software Packages Available to Quantify Risk from Natural Hazards, World Bank
Group, Washington, United States, 2014. Available online from: http://documents.worldbank.org/curated/en/109501468327344629.
[44] F. Mokhtari, S. Soltani, S.A. Mousavi, Assessment of flood damage on humans, infrastructure, and agriculture in the Ghamsar Watershed using HEC-FIA
software, Nat. Hazards Rev. 18 (3) (2017), 04017006, https://doi.org/10.1061/(ASCE)NH.1527-6996.0000248.
[45] J.T. Brackins, A.J. Kalyanapu, Using ADCIRC and HEC-FIA modeling to predict storm surge impact on coastal infrastructure, in: World Environment and Water
Resources Congress, 2016. https://ascelibrary.org/doi/pdf/10.1061/9780784479841#page=213.
[46] S.N. Jonkman, B. Maaskant, B. Kolen, M. Zethof, W. Lehman, Loss of Life, Evacuation and Emergency Management: Comparison and Application to Case Studies
in the USA, DELFT UNIV OF TECHNOLOGY (NETHERLANDS), 2013. https://apps.dtic.mil/sti/citations/ADA577448.
[47] S.B. Jonkman, B.B. Maaskant, B.B. Kolen, J.J. Needham, Loss of life estimation–Review, developments and challenges, in: In E3S Web of Conferences, 2016,
p. 06004, https://doi.org/10.1051/e3sconf/20160706004.
[48] R. Brummel, J. Jensen, H. Pyzdrowski, On the path to community flood resilience for the Upper Iowa Watershed: documenting 2016 flood experiences in
Freeport, Iowa, Freeport Flood Report (2019). https://upperiowariver.org/wp-content/uploads/2019/08/Report___Community_Flood_Resilience_in_Freeport.
pdf.
[49] A. Teague, Y. Sermet, I. Demir, M. Muste, A collaborative serious game for water resources planning and hazard mitigation, Int. J. Disaster Risk Reduc. 53
(2021) 101977, https://doi.org/10.1016/j.ijdrr.2020.101977.
[50] FEMA, Debris removal, Retrieved from, https://www.fema.gov/appeal/debris removal-1, 2020.
[51] USACE, Economic Guidance Memorandum (EGM) 04-01: Generic Depth-Damage Relationships for Residential Structures with Basements, 2003 (Washington,
DC).
[52] Australian Institute for Disaster Resilience (AIDR), Flood hazard. Australian disaster resilience handbook collection, Guideline (2017), 7-3.
[53] V.A. Rangari, N.V. Umamahesh, A.K. Patel, Flood-hazard risk classification and mapping for urban catchment under different climate change scenarios: a case
study of Hyderabad city, Urban Clim. 36 (2021) 100793, https://doi.org/10.1016/j.uclim.2021.100793.
[54] L.T.H. Binh, N.V. Umamahesh, E.V. Rathnam, High-resolution flood hazard mapping based on nonstationary frequency analysis: case study of Ho Chi Minh City,
Vietnam, Hydrol. Sci. J. 64 (3) (2019) 318–335, https://doi.org/10.1080/02626667.2019.1581363.
[55] M. Jakob, D. Stein, M. Ulmi, Vulnerability of buildings to debris flow impact, Hazards 60 (2) (2012) 241–261, https://doi.org/10.1007/s11069-011-0007-2.
[56] J. Xia, R.A. Falconer, Y. Wang, X. Xiao, New criterion for the stability of a human body in floodwaters, J. Hydraul. Res. 52 (1) (2014) 93–104, https://doi.org/
10.1080/00221686.2013.875073.
[57] C. Arrighi, H. Oumeraci, F. Castelli, Hydrodynamics of pedestrians’ instability in floodwaters, Hydrol. Earth Syst. Sci. 21 (2017) 515–531, https://doi.org/
10.5194/hess-21-515-2017.
[58] G.P. Smith, E.K. Davey, R.J. Cox, Flood Hazard, Water Research Laboratory, University of New South Wales, Sydney, 2014. Technical report 2014/07.
[59] USACE., HEC-LifeSim, Retrieved from https://www.hec.usace.army.mil/error.html?aspxerrorpath=/software/hec-lifesim/download.aspx, , 2020.
[60] R. Agliamzanov, M. Sit, I. Demir, Hydrology@ Home: a distributed volunteer computing framework for hydrological research and applications, J. Hydroinf. 22
(2) (2020) 235–248, https://doi.org/10.2166/hydro.2019.170.
[61] J.C. Banks, J.V. Camp, M.D. Abkowitz, Adaptation planning for floods: a review of available tools, Nat. Hazards 70 (2) (2014) 1327–1337, https://doi.org/
10.1007/s11069-013-0876-7.
[62] J.L. Gutenson, A.A. Oubeidillah, P. Hicks, L. Durham, A.N.S. Ernest, L. Zhu, X. Zhang, Using HAZUS-MH and HEC-RAS to evaluate real world flooding events in
the Upper Alabama River watershed, in: Proc., World Environmental and Water Resources Congress 2015, ASCE, Reston, VA, 2015, pp. 1607–1627, https://doi.
org/10.1061/9780784479162.157.
[63] National Weather Service, Central Iowa floods of 2008, Rederived from, https://www.weather.gov/media/dmx/SigEvents/2008_Central_Iowa_Floods.pdf,
2009.
[64] D. Gilles, N. Young, H. Schroeder, J. Piotrowski, Y.J. Chang, Inundation mapping initiatives of the Iowa Flood Center: statewide coverage and detailed urban
flooding analysis, Water 4 (1) (2012) 85–106.
[65] Iowa Flood Center., Iowa flood inundation maps, Retrieved from, https://iowafloodcenter.org, 2020.
[66] United States Census Bureau, Retrieved from, https://www.census.gov, 2010.
[67] Homeland Infrastructure Foundation-Level Data (HIFLD), Open Data, 2019. Retrieved from, https://hifld-geoplatform.opendata.arcgis.com/.
[68] National Automobile Dealers Association (NADA). Retrieved from https://www.nada.org/.
[69] Dun & Bradstreet, Business Population Report Aggregated by Standard Industrial Classification (SIC) and Census Block, 2006. May 2006.
[70] National Bridge Inventory, Bridges and structures, Retrieved from, https://www.fhwa.dot.gov/bridge/nbi.cfm, 2018.
[71] Contributors, OpenStreetMap Wiki, 2020. Retrieved from, https://wiki.openstreetmap.org/w/index.php?title=Contributors& oldid=2162996.
[72] FEMA, Hazus inventory technical manual, Retrieved from, https://www.fema.gov/sites/default/files/documents/fema_hazus-inventory-technical-manual-4.2.
3.pdf, 2021.
[73] E. Ghimire, S. Sharma, Flood damage assessment in HAZUS using various resolution of data and One-Dimensional and Two-Dimensional HEC-RAS depth grids,
Nat. Hazards Rev. (2021) 04020054, https://doi.org/10.1061/(ASCE)NH.1527-6996.0000430.

16
Y. Alabbad and I. Demir International Journal of Disaster Risk Reduction 74 (2022) 102955

[74] E. Yildirim, I. Demir, An integrated web framework for HAZUS-MH flood loss estimation analysis, Nat. Hazards 99 (1) (2019) 275–286, https://doi.org/
10.1007/s11069-019-03738-6.
[75] V. Matthews, J. Longman, H.L. Berry, M. Passey, J. Bennett-Levy, G.G. Morgan, R.S. Bailie, Differential mental health impact six months after extensive river
flooding in rural Australia: a cross-sectional analysis through an equity lens, Front. Public Health 7 (2019) 367, https://doi.org/10.3389/fpubh.2019.00367.
[76] A. Carson, M. Windsor, H. Hill, T. Haigh, N. Wall, J. Smith, R. Olsen, D. Bathke, I. Demir, M. Muste, Serious gaming for participatory planning of multi-hazard
mitigation, Int. J. River Basin Manag. 16 (3) (2018) 379–391, https://doi.org/10.1080/15715124.2018.1481079.
[77] A. Hu, I. Demir, Real-time flood mapping on client-side web systems using HAND model, Hydrology 8 (2) (2021) 65, https://doi.org/10.3390/
hydrology8020065.
[78] E. Martínez-Gomariz, M. Gómez, B. Russo, P. Sánchez, J.A. Montes, Methodology for the damage assessment of vehicles exposed to flooding in urban areas,
J. Flood Risk Manag. 12 (3) (2019) e12475, https://doi.org/10.1111/jfr3.12475.
[79] H. Xu, I. Demir, C. Koylu, M. Muste, A web-based geovisual analytics platform for identifying potential contributors to culvert sedimentation, Sci. Total Environ.
692 (2019) 806–817, https://doi.org/10.1016/j.scitotenv.2019.07.157.
[80] Y. Sermet, I. Demir, Virtual and augmented reality applications for environmental science education and training, in: New Perspectives on Virtual and
Augmented Reality, Routledge, 2020, pp. 261–275.

17