Introduction

According to the Intergovernmental Panel on Climate Change (IPCC) special report on extreme events, the risk associated with disaster can be defined as the likelihood of events that severely alter the normal functioning of a community or society. Specifically, these hazardous, physical events threaten vulnerable peoples’ social conditions1. Risk assessments across the literature prevalently follow the IPCC framework, which quantifies risk as a function of hazard, exposure, and vulnerability2,3,4. However, in a recent study, response was included as a fourth component of risk5. Response refers to the ability to react to a situation and is often excluded as a risk driver. One of the novelties of our study lies in adding a response component to the flood risk framework due to its potential to subdue the adversity of the event.

In the IPCC risk framework, hazard is a function of scale, including the extent and probability of occurrence of a flood event at a given location. Studies have come up with different ways to quantify the hazard associated with flooding, for example, using the Federal Emergency Management Agency’s (FEMA) maps2, flood inundation maps3, climatic projections to account for risk associated with climate change4, and peak discharge for different flood return periods (1000, 100, and 50 years)6. The IPCC defines exposure as the presence of people, livelihood, environmental services and resources, infrastructure, and social and cultural aspects the flooding could adversely affect1. Different factors can account for exposure, such as population density, housing units, impervious surfaces, and elevation and density of infrastructures4. Vulnerability is the propensity or predisposition to be adversely affected by flooding1. To assess vulnerability, studies have considered social dimensions such as age, education, ethnicity, gender, and socioeconomic status2, 4, 7. We classified vulnerability into social, ecological, economic, and health dimensions, considering multiple variables under each section. Our study is unique in that it adds different dimensions to vulnerability and gives insights to quantify response while assessing the risks associated with flooding across Nebraska on a county scale.

Study area, methodology, and datasets information

Study area

Nebraska is a state located in the Midwestern region of the United States. It consists of 93 counties and is surrounded by six states. The topography of Nebraska can be divided into two major land regions – the Dissected Till Plains and the Great Plains (Fig. 1). Dissected Till Plains are located in the eastern part of the state, which consists of rolling hills and fertile agricultural land. This region includes major Nebraska cities – Omaha (Douglas County), Lincoln (Lancaster County), Bellevue (Sarpy County), and Papillion (Sarpy County) (based on existing population). The Great Plains region of Nebraska consists of rolling terrain suitable for agricultural operations. This region is mostly dedicated to corn and wheat fields. The Great Plains of Sand Hills are located in the north-central part of the state, mainly covered with grassy dunes.

Figure 1
figure 1

Topographical map of Nebraska. This map was generated using ArcGIS software – version 10.7.1.

Full size image

Nebraska’s climate and topography play a crucial role when flooding occurs in the state. The Midwestern United States is susceptible to weather patterns, including heavy rainfall, snowstorms, and tornadoes. The region's topography primarily consists of rolling terrain, which makes it even more prone to flooding, as there are no significant natural barriers to counter peak flows. The region also has many major river systems, including the Missouri River, Platte River, and Niobrara River. These rivers and their tributaries drain a large portion of the state and tend to quickly rise and overflow during heavy rain or snowmelt, leading to widespread flooding.

Methodology

The risk assessment of this study aims to understand and quantify the potential dangers posed by flooding in a comprehensive manner, which can help in making more informed decisions related to effective flood management. The multiplication of the components of risk, i.e., hazard, exposure, and vulnerability is a common approach in risk assessment and has been widely reported in the literature3, 4, 7. The multiplication reflects the idea that high exposure to a hazard, combined with high vulnerability, leads to a greater overall risk. Hazard is related to the potential occurrence of a natural or human-induced physical event that can result in loss of life, injury, or other health impacts1. Exposure refers to elements such as population, infrastructures, and natural or artificial resources in an area where hazard events may occur. Vulnerability in literature is defined as the propensity to be adversely affected by a disaster event1.

In this study, we added one more component, i.e. response, which refers to actions and measures taken to reduce risk5. The extent of hazard’s exposure can amplify the risk, especially if that system is highly vulnerable. However, response, in the form of proper actions and measures against flooding, can potentially reduce the risk. In this study, we multiply hazard, exposure, and vulnerability, and divide that outcome by response (as it subdues the risk) to account for the combined effect of drivers on the risk level (Eq. 1).

$$\mathrm{Risk}=\frac{\mathrm{Hazard }\times \mathrm{ Exposure }\times \mathrm{ Vulnerability}}{\mathrm{Response}}.$$
(1)

Equation (1) considers risk as a function of hazard, exposure, vulnerability, and response, and reflects the understanding that all components contribute to the overall level of flood risk. We considered different variables under each driver of risk, which involved different units of measurement or scales. To allow for meaningful comparison and combination, the variables used to quantify the drivers of risk were scaled from 0 to 1 using (Eq. 2)4:

$$X=\left(x-\mathrm{min}(x)\right)/(\mathrm{max}\left(x\right)-\mathrm{min}\left(x\right)).$$
(2)

By standardizing the variables, we eliminated the potential bias introduced by the magnitude of the values. We assumed equal importance for each variable within a component. Therefore, the values of each risk driver were combined and divided by the number of drivers used4, resulting in an index ranging from 0 to 1 (Eq. 3):

$$D(\mathrm{0,1})={\sum }{\mathrm{X}}_{D,i} (\mathrm{0,1})/\mathrm{n},$$
(3)

where D is the driver's index, X.D., i is the ith scaled variable, and n is the number of variables. Averaging the variables under each component simplified the calculation and interpretation of flood risk.

Datasets information

The properties at risk of flood hazard were used as a proxy to quantify the hazard. We obtained this information from the First Street Foundation8 for 2020. The First Street Foundation uses a probabilistic flood model, which considers the hazard associated with rainfall, riverine flooding, and coastal surge flooding. The model identifies property boundaries and then uses elevation data to determine the likelihood of water reaching the premises. The model also considers adding infrastructure protection, including dunes, wetlands, seawalls, and pumps. To quantify hazard, we reviewed properties in Nebraska counties and collected flood factor scores ranging from 1 to 10, where 1 refers to the lowest chance of a property flooding and 10 refers to the highest. We only considered properties with scores of 8, 9, and 10, as they have severe consequences on properties and the associated population. The exposure variable referring to population density was obtained from the U.S. Census Bureau9 for 2020. It should be noted we also used county properties as a component of the exposure variable, but observations and information on properties were only included during the hazard and hazard index Sects. (“Hazard” and “Hazard index”) to avoid repetition in the exposure Sects. (“Exposure” and “Exposure index”).

We collected demographic and social data from the Census and DataUSA10 platform for the year 2020, which were used to quantify vulnerability (Table 1). To quantify response, we collected information from different state organizations such as the Nebraska Department of Natural Resources11 (NeDNR), State of Nebraska Flood Hazard Mitigation Plan12 (SNFHMP), Nebraska Department of Health and Human Services13 (NDDHS), U.S. Army Corps of Engineers14 (USACE), and Homeless Shelters Directory15, an open-source website, which includes information about different emergency shelters located across the state at the county level for 2022 as we did not have information for 2020 (Table 2).

Table 1 Variables considered for quantifying vulnerability.
Full size table
Table 2 Variables considered for quantifying response.
Full size table

Flood risk framework

The risk associated with flooding signifies the possibility of adverse effects on humans and surroundings after the occurrence of flooding events. This risk is derived from the interaction of social and environmental processes, the combination of flood hazards, and the vulnerabilities of exposed elements1. Most studies across the literature used the IPCC’s risk framework to estimate risk, i.e. risk as the function of hazard, exposure, and vulnerability. We considered response as the fourth driver of risk (Fig. 2). The response has the ability to reduce the risk of flooding and is required to be included in the existing flood risk frameworks to carry out an effective assessment.

Figure 2
figure 2

Flood risk framework showing interaction among variables under each driver. These drivers interact with each other resulting in risk.

Full size image

Hazard

Hazard is considered the potential occurrence of a natural or human-induced physical event that can result in loss of life, injury, or other health impacts1. The quantification of flood hazards is carried out at a local and global scale. On a basin scale, one way to quantify hazard is by estimating peak discharge for different flood return periods through statistical modeling and direct observation6. On a global or country scale, studies have used climate and flood model outputs16. In this study, we considered the number of properties at risk of flooding obtained from flood models to quantify the hazard. The First Street Foundation flood model is a probabilistic flood model considering hazards associated with rainfall, riverine flooding, and coastal surge flooding. The model identifies property boundaries and uses elevation data to determine the likelihood of water reaching properties. It considers community protection and land features such as dunes, wetlands, seawalls, and pumps while quantifying flooding hazards. To assess flood hazards, we considered a flood factor score of 8, 9, and 10, based on the probability of flooding in the area.

Exposure

Exposure refers to elements such as population, infrastructures, and natural or artificial resources in an area where hazard events may occur. The concentration of the population increases the exposure to extreme events4. In the case of flooding, many studies have used population to calculate the risk of exposure4, 17, 18. Further, as the number of flooding events is projected to rise, the number of housing units at risk associated with flooding across the United States is likely to triple by the 2050s19. Consequently, a larger risk for housing infrastructure will increase maintenance costs, influence public health, and profoundly disrupt struggling families. In this study, housing exposure was also considered, though the information was only included under the hazard component as all properties and their associated risks have already been recorded there.

Vulnerability

Vulnerability in literature is defined as the propensity to be adversely affected by a disaster event1. Vulnerability indicators are often single variables, though they indicate multidimensional factors such as historical, cultural, social, and economic processes that affect the community’s ability to cope with hazards and respond to them20. In this study, we quantified population vulnerability with reference to four themes, i.e. social, ecological, economic, and health (Fig. 3).

Figure 3
figure 3

We classified vulnerability under four themes. All variables under each theme were considered for quantification of vulnerability at the county scale.

Full size image

Social vulnerability

Social vulnerability refers to socio-demographic factors like age, gender, and ethnicity that affect the population’s resilience against flooding. A socially vulnerable population is more likely to be adversely affected during flooding and can take longer to recover21. This type of vulnerability plays a crucial role in flooding evacuation processes. For instance, it is more difficult for older people to move to a safe place in the event of a hazard. In some cases, health complications can increase during movement, worsening the situation2. The infant population can be considered vulnerable as they require more attention, especially in post-flooding scenarios when hospitals and daycare facilities are affected3. Gender can indicate vulnerability due to a lack of resources and differential exposure2. Literature has found that women have higher risk perceptions, demonstrate higher preparedness planning, and are more likely to respond to warnings than men; however, in some cases, they are more likely than men to be single parents or primary caregivers to families. Literature has also reported that in many cases, females have relatively lower income, resources, and autonomy than males, which makes them more vulnerable7, 22, 23. Race and ethnicity factors are also essential considerations when assessing vulnerability. Ethnic inequality associated with language and cultural barriers, as in the case of immigrants, may hamper flood preparation and evacuation3.

Ecological vulnerability

Ecological vulnerability can be defined as changes in climatic and environmental conditions that trigger or facilitate other adverse impacts besides flooding. It is difficult to quantify as ecological systems are complex and interconnected, with multiple species, habitats, and environmental processes interacting with each other. Literature has considered different proxies, including proximity of the ecosystem to toxic release inventory and super fund sites, type of existing slope, land use and soil, and availability of green space in dense regions24, 25.

Air pollution is one potential quantifying factor. During flooding, it is seen that when a hazardous chemical comes in contact with air, it creates a toxic environment26. Fine particulate matter causing air pollution is often attributed to adverse health outcomes27. Further, food environment factors, such as accessibility to resources, quality of food available, and price and availability of a product were also considered when assessing ecological vulnerability. Characteristics such as existing biodiversity, access to fresh water and land resources, and overall ecosystem health in a region influence various aspects of food production, distribution, and waste management, which directly impact the food environment index. Ecologically conscious practices can contribute to a healthier, more sustainable, and resilient food environment.

Apart from the drivers of ecological vulnerability mentioned above, literature has also included the effect of habitat disruption, which leads to the loss of food sources for various species, breeding grounds, and nesting sites. These changes disrupt ecological balances, thus impacting biodiversity and causing a rise in health-related issues for human, plant, and animal species24, 28, 29. Flooding also alters fish movement, which affects fishing operations30, and damaging the forest ecosystem leads to ecological and economical deficits31. Therefore, farming, fishing, and forestry are among the primary sectors susceptible to flooding.

Economic vulnerability

Economic vulnerability can be defined as the degree to which individuals with low economic status are susceptible to or unable to cope with the adverse effect of flooding. Economic vulnerability is often considered a function of wealth and income32. Wealth is an essential factor to consider while assessing people's vulnerability to flooding. Those with low income are more susceptible due to a lack of resources, poor housing, and the inability to recover quickly2. Higher wealth increases the possibility of preparing for disaster and leads to a quicker recovery after the event3. Literature has also reported that a low economic status results in the prevalence of ill health and societal issues, such as violence, lack of trust, and poor educational facilities, which can be considered a source of economic vulnerability33.

Several studies have shown that flooding leads to job loss, financial crises, and adverse health outcomes34,35,56, while Rufat et al. utilized the weights developed by prior studies57. It would be valuable for future studies to weigh each variable according to their importance in estimating flood risk.

We assessed flood risk associated with properties and population, which can be calculated for other dimensions of floods (geographic, social, and infrastructural). It is difficult to quantify flood hazards at a county or state level at a high resolution due to the computation challenges associated with detailed spatial–temporal modeling. In such cases, studies can work toward develo** high-resolution flood maps, which can be used for quantification of flood hazards.

We classified county flood vulnerability into four major types: social, ecological, economic, and health. This allowed us to explore different aspects which affect humankind. There is potential for future studies to expand this list of factors resulting in flood-associated vulnerability. Further, in the context of health vulnerability, more publicly accessible data can be beneficial. Information related to different diseases caused by floods can be helpful in establishing essential risk frameworks.

Quantification of flood-risk response was one of the novelties of this work. We classified response into four classes based on the information obtained from the Department of Natural Resources in Nebraska, the Nebraska Department of Health and Human Services, state flood hazard management plans, and the U.S. Army Corps of Engineers website. It will be essential to develop a system to collect and quantify response information associated with flooding.

Conclusion

This study gave insights into the flood risk linked to properties and associated population at a county scale for the state of Nebraska. We quantified flood risk by modifying the existing IPCC's risk framework (with hazard, exposure, and vulnerability) to add a response component. We obtained information about properties at risk of flooding, which constituted our hazard component. Exposure was quantified by including population information in the framework. We split the vulnerability factors into four sub-classes to explore how vulnerability can affect flood risk and preparedness. We quantified response to include different mitigation strategies counties are using to counter flooding. The overall flood risk was mostly concentrated in the eastern part of the state, particularly in Sarpy, Dakota, Adams, Wayne, Cass, Pierce, Platte, and Colfax counties. The methodology implemented in this study is not limited to the quantification of flood risk associated with properties. It can be applied to other dimensions of floods and hazardous events like droughts, forest fires, and cold waves. However, the drivers may need to be modified for different cases and hazardous events. A county-scale approach like this will make policymakers aware of the existing risks and vulnerabilities. This can help make tailored, effective, and locally relevant policies at the county level.