New hybrid method for vulnerability assessment in floodplain areas exposed to dam break

Natural and man‐made factors can cause dam failure, which constantly threatens downstream assets. Flood risk assessment is typically developed according to the hydraulic and geographic characteristics of floodplain regions. However, previous approaches have not considered asset values and the impacts of flooding on them, including threat probability and consequences. This report presents a new integrated hybrid model based on security vulnerability assessment and hydraulic analysis of floods and uses the source‐pathway‐receptor‐consequences (SPRC) approach to comprehensively examine all factors. A questionnaire survey was used for performing threat analysis and attractiveness analysis and determining the asset values and consequences. A geographical information system was applied to superimpose layers based on the SPRC approach in five different scenarios based on piping discharge and downstream flow in the case of a dam break. These scenarios were developed considering the basin water level. The results illustrate that residential areas have the highest vulnerability, and farmlands exhibit the lowest exposure to flooding caused by dam failure. Based on the attraction to terrorism, criticality due to consequences, and flooding characteristics, high‐risk zones were identified for implementing mitigation or protection countermeasures.


INTRODUCTION
Dams are defined as artificial obstructions to natural water flows, which help accumulate water for agricultural purposes, energy generation, potable water supply, flood mitigation, and recreational reasons. 1 Furthermore, dams are designated as one of the key assets in the Critical Infrastructures and Key Resources by the Department of Homeland Security in the United States.There was a massive initiative from the federal government to coordinate a set of resilience and protection plans, which aims to clarify the features that affect resilience. 2 According to this mission, dams and their environment should be designed to be safer, more secure, and more resilient by preventing, deterring, neutralizing, or mitigating the effects of deliberate efforts by terrorists to destroy, incapacitate, or exploit these elements. 3There are also other programs being developed to help protect critical infrastructure, such as the Swiss Federal program for critical infrastructure protection, intended to analyze threats, identify vulnerabilities, and assess risks.
Policymakers must consider all relevant hazards and threats to assess the risk and provide a protection plan for dams and their downstream areas.The risks or vulnerabilities related to dams often refer to natural hazards such as earthquakes, floods, or technological failures.Furthermore, the wide range of hazards and threats can be multi-faceted, either the result of technological or human error, related to improper operation or a lack of maintenance.Most of the previous research has focused on different natural hazards and threats.However, the emergence of terrorist groups, such as Al-Qaeda and ISIS, revived the idea of protection from human-related threats.The scope of human-related threats ranges from unintentional errors to targeted malicious attacks, either physical (e.g., explosive devices) or digital, which require sophisticated models that are capable of describing complex ideology related to terror and destruction. 4Thus, risk assessment and vulnerability assessment should expand to all properties related to the dams, including farms, transportation routes, energy lines, pipes, business areas, industries, cities or villages, and even biosphere assets.6][7][8] Researchers have divided different methods concerning loss estimation into three categories: macro-level by characterizing the event, meso-level by describing the location, and micro-level by investigating the individuals. 6In the developed equation, the most critical property is people, and any risk number is usually calculated based on the number of people at risk.Two reducing factors, such as the sheltering fraction 9 and evacuation fraction (F E ), are expressed in Equation (1).The mortality fraction is the other factor (F d ) used in risk assessment equations (Equation 2), where N is the number of people at risk, and N PAR is the number of people at risk.
F E : evacuation fraction; F d : mortality fraction; F S : sheltering fraction; N PAR : number of people at risk.The other approach for estimating flood mortality risk is the scalable approach, developed by the US army corps of engineers, which is based on three factors: flood severity, exposed population, and shelter type. 10Penning-Rowsell et al. introduced a risk assessment framework to estimate injury and loss of life.This framework contains four stages for risk assessment: hazard identification, hazard characterization, exposure assessment, and risk characterization.However, the definition of hazards is solely based on natural hazards. 11t is imperative to identify parameters related to the flood risks that result from breaking dams, especially by malicious attacks.In the dam failure risk assessment model, only the dominant failure mode of the dam is usually identified.However, different parameters are required for evaluating and assessing the vulnerabilities of nearby assets to a dam break.2][13][14][15][16][17][18] Sun et al. computed the vulnerability based on consequences in three categories, such as the loss of life, economic loss, and social and environmental impact, although the speed and depth of flow downstream were also considered in dam-break flood simulations. 12,19The hazard type is also connected to damage classes for assessing vulnerability in other methods, like in the GenMR framework for large dams.In contrast, some hazards are more performance based, considering hydropower or outlet malfunction, rather than natural hazards. 13Other researchers assigned the value of property exposure to flooding as an essential element for vulnerability assessment. 14One of the comprehensive descriptions for the vulnerability was presented as a dependency, mainly on the degree of exposure, the system's capability to withstand exposure, the magnitude of the phenomenon, and the social factor. 15In the HURAM model, depth, speed, and duration are three essential factors used to estimate vulnerability to dam-break floods. 18Figure 1 illustrates the different vulnerability assessment parameters used in previous methods.
Catastrophic losses in terms of human casualties, property destruction, economic damages, and loss of public confidence can result from dam failure, especially downstream and in the floodplain areas.Traditionally, when man-made disasters are caused by sabotage or terrorist actions, the risk is expressed as a function of the likelihood of an initiating event, the possibility of system failure, and the consequences associated with the loss.However, other approaches cover threat analysis, asset analysis, attractiveness, and vulnerability assessment to achieve the risk number for each property is security vulnerability assessment (SVA).SVA is one of the comprehensive approaches currently used in different sectors and infrastructure.One of the advantages of SVA is the flexibility of techniques that risk assessors can use in each part of SVA. 20,21ams are considered superstructures, and they are critical forms of infrastructure in most countries, especially considering that dams are often built in crowded areas, making them high-potential targets for terrorist action. 22or instance, the Mosul Dam, the largest dam in Iraq, became one of the tools of ISIS for threatening downstream cities during Mosul occupancy in 2016.However, dams are essential for supplying water resources and potable water, agricultural activities, producing electricity, and controlling floodwaters.Thus, dams are inherently responsible for bolstering the downstream areas and increasing the populations in these areas.As a result, cities, industries, and farms tend to grow while environmental resources, such as trees and forests, tend to vanish.These factors highlight the importance of basing vulnerability assessment and risk analysis on downstream properties and assets.Several studies have been conducted in areas under the effects of a flood caused by dam destruction.Messervey 23 has created a framework and discovered modeling techniques for dam destruction.Awal studied dam destruction caused by landslides and flyovers, increasing water height, 24 and other studies described alternative signals and decreasing floodwater risk. 25ost of the research in this field tried have used hydraulic software or considered crisis management in response to natural threats such as earthquakes and floodwater.However, there is no research on the accurate vulnerability assessment of dam destruction assets based on hydraulic modeling in collaboration with water flow factors and destruction scenarios, particularly man-made threats.This has become more important from the perspective of homeland security agencies and private sectors, who own infrastructure or at least one of the stockholders is invested in those infrastructure.Vulnerability assessment is a part of risk assessment, which shows how vulnerable the dam or any infrastructure is to the identified threat.Accordingly, infrastructure vulnerability can be categorized into three categories, which show the enhancement of three generations of assessment.In the first generation, the inherent properties of an asset become the leading factor for the assessment of vulnerability or risk assessment of the asset.In the next generation, it is necessary to consider the attractiveness of the asset and exposure to the hazard, which has been used in the Khalid Sheikh Mohammed and Security Risk Factors Table methods, and mentioned in previous reports by Kröger, Zio. 4,20Finally, the third-generation uses a triangulated approach based on asset property type, attractiveness and dependencies of the asset, and interdependencies following the network of properties.Model-based risk assessment is one of the integrated third-generation models. 26

F I G U R E 2
The new hybrid approach for vulnerability assessment.
The SVA approach also considers attractiveness of properties, or their ability to attract attention.However, a more scientific method is required, especially regarding the hydraulic characteristics of floods and the interactions between floods and downstream properties in the floodplain area.
Therefore, based on the latest vulnerability assessment generation, a new hybrid approach must be developed that considers the vulnerability of the dam's downstream area.The present study aims to modify an integrated hybrid model and analyze the vulnerability of the dam's downstream assets in the case of man-made threats, to provide risk analysis and management for the protection and mitigation of threats in high-risk areas.Figure 2 shows the theoretical framework of this approach.
First, it is necessary to illustrate the main innovations and differences between the current approach with others.Previous risk assessment approaches are mainly based on one scope.For instance, it is common to use a floodplain risk map based on hydrological information without considering the sensitivity of properties to the water and flood.Standard methods typically provide risk zone maps based on geographical altitude and flood range in the riverbank.Otherwise, the source of the flood is neglected, especially for man-made threats.Thus, using the integrated source-pathway-receptor-consequences (SPRC) and SVA method helps us integrate the hydrological risk evaluation with social and geographical impacts and added evaluation of hazards and man-made threats for risk assessment.It may provide policymakers and decision-makers with the necessary information to determine the appropriate applications for downstream areas.Specifically, this research attempts to answer the following questions: What is the best approach for combining hydrologic and hydraulic criteria in a floodplain with social and geographical information on floodplain properties according to flood characteristics and asset sensitivity?How can hazards and threats be interpreted as a source of flooding using risk assessment?What is the optimized, integrated approach for risk and vulnerability assessment in floodplain areas?

METHODOLOGY
As mentioned previously, it is necessary to generate a new approach that combines security risk modeling, consisting of the inherent characteristics of assets and their attractiveness, with hydraulic and hydrological characteristics of the flood and floodplain areas in contact with downstream assets.Thus, a parallel approach is needed to provide appropriate risk analysis steps, in addition to flood mapping and horological analysis.As illustrated in Figure 3, the task of risk analysis is conducted by performing a modified SVA method.Furthermore, a hydraulic model is developed using HEC-RAS to extract the vulnerable factors of flood risk based on the height and velocity of water flow.Then, the SPRC method is applied as a connection between these two parallel calculations to finalize the risk map based on a geographic information system (GIS).Figure 3 illustrates the models and tools used in this research.

F I G U R E 3
The process of research.

SVA model
Dams and their facilities are designated as critical infrastructure and assets.Thus, the risk assessment model should consider the attractiveness of facilities and the probability of a threat occurring, and it should simultaneously consider the vulnerability of each type of threat.The SVA method is the primary risk assessment model in this study.Ranking in the SVA method includes a 5-level Likert scale that is classified with a query according to experts' suggestions.Level one of the assets is considered the minimum value, and level five represents the maximum. 21Figure 4 depicts the 5-level SVA methodology.
It is essential to improve and adjust the SVA in water infrastructure, and the innovative approaches in this study are mostly related to this issue.Employing the SPRC approach to evaluate vulnerability in the SVA method is the first improvement.Integrating hydraulic assessment of flood flow downstream with GIS and applying it in SVA by developing a GIS-based risk map is another improvement to the traditional SVA approach.These modifications help make this method more suitable for water infrastructure and floodplain areas.

SPRC model
It is crucial to identify a transparent platform for connecting security risk assessment to the engineering approach for vulnerability assessment.Flood risk experts proposed the conceptual SPRC model in 2001.It shows a simple causal chain ranging from the primary sources of danger (sources) through the links and lines of interactions (pathways) that make impacts on elements at risk (receptors), and finally, it assesses the effects of disasters on receptors (consequences). 27The SPRC method is a conceptual method used to evaluate vulnerability, and the risk of assets against floodwater can be estimated and measured.Figures 5 and 6 show the process of this methodology.As illustrated in Figure 5, the interactions between assets and sources of danger play a vital role, which are defined as pathways.Figure 6 illustrates the general approach of the presented risk assessment method according to the SPRC approach.This figure identifies the SPRC relations based on the interactions between sources and receptors, which are the properties downstream of the dam, which starts with the first source of flooding.There are three types of relations among the properties.The first relation is the impact of the source on the receptor (direct impact-black lines), the second is the impact of the source of the secondary effects (red line), and the third is the probable indirect impacts from receptor to receptor (green and purple lines), which can be neglected in this research.This approach helps to apply the principles of SPRC to choose the primary source, secondary sources, and impacted properties.

Flood mapping model
One of the essential criteria for risk assessment is vulnerability, which was calculated in the SVA using the qualitative approach.This study aims to combine the vulnerability of assets with the source of danger, which is a flood caused by dam destruction.Thus, it is necessary to assume different scenarios for the dam to break due to terrorist attacks and to obtain outputs.The selection of the scenarios depends on the situation and dam properties.For this research, the type of dam is selected to be an earth dam.Different failure modes are possible.Accordingly, to cover most of the aspects of piping failure and flooding, five scenarios are assigned for this research.The first scenario concerns the piping of 36.6 million m 3 of water from the lake (40% of its capacity).In the second scenario, releasing 55% of the volume of the dam's

F I G U R E 6
The integrated approach for vulnerability assessment based on source-pathway-receptor-consequence (SPRC) approach (vulnerability of receptor K).
reservoir, ∼50 million m 3 of water.The third scenario involves releasing 70% of the lake capacity (64 million m 3 ).The fourth scenario involves refilling 85% of the lake capacity (77.7 million m 3 ).Finally, the fifth scenario considers spilling the total lake capacity.The outputs of the breaking dam simulate the flood load on downstream areas to obtain the flood mapping, including the cross-section and velocity of the surge in the downstream floodplain area.Using HEC-RAS, an open-source software compatible with GIS, enables merging of the data from security sources with hydraulic and flood data.This software also allows the design of culverts, bridges, side weirs, and spillways to provide countermeasures.HEC-RAS software also incorporates boundary conditions (BC) to model the flow of the flood.These include upstream and downstream BCs, maintaining a specific flow rate and water surface balance, water surface ramp (angle), and critical depth.Figure 7 demonstrates the simulation processes of flood mapping in HEC-RAS software and the transition of outputs to GIS.

Vulnerability modeling in GIS
The typical SVA model determined the vulnerability assessment based on questionnaire results.However, the SPRC model suggested measuring vulnerability caused by the flood using the velocity, water level, and value (sensitivity) of an asset (receptor sensitivity).In Table 1, the factors used in asset vulnerability assessment are classified into five classes (Likert scales).First, the information layers are prepared numerically according to the elements.Then, layers (height, velocity, and assets value) have been multiplied and integrated with the GIS by employing Spatial Analyst tools and the Raster Calculator sub-menu to obtain a final asset vulnerability map (Figure 8).This research applied the integrated risk model to conceptual earth dams.Table 2 illustrates the characteristics of the earth dam, which has a clay core.
Next, it is necessary to define the downstream assets that can be affected by the flood and dam destruction.For the reliability of the questionnaire, the Cronbach alpha value was assessed, which was equal to 0.83, which demonstrates that the reliability of the questionnaire was in the appropriate range.
The vulnerability was produced first by averaging the value of properties, flood velocity, and height of water (flood level), as shown in Equation ( 3), to develop a risk analysis map for each zone.
Where velocity and height are flooding characteristics that come from HEC-RAS, and V value is the asset value introduced in Table 4.After calculating the vulnerability of the flood zone, Equation (4) was employed to develop the risk map.This equation is used in the GIS to produce a risk map by superimposing the maps and averaging criteria for each zone.
where A and T are attractiveness and threat probability, respectively, C is the consequence for each asset as a result of the flood impact, and V is the vulnerability of each zone that was calculated in Equation (3).

Rationality and validity of the presented model
There are two important issues regarding the reliability and validity of this method.First, the result of vulnerability is achieved from the multiplication of the three indicators, such as water flow rate, height, and the vulnerability of downstream assets, with completely different dimensions.Doubt may also arise concerning the BCs used for calculation.
To address these issues, the validity of this method is compared with similar methods, and then, the rationality of the boundaries is investigated.

Validation and comparing to similar methods
First, it is necessary to clarify the target of this method.The target of this method is to assess downstream land and properties.Thus, it shows the conflict between technical and societal criteria for any vulnerability analysis.When the study deals with the vulnerability of land, it is not just considering the hydraulic effects of flooding, for instance, the drag force or flotation probability or debris impact.It also focuses on the inherent reflection of the assets and properties inside each area while facing flood water, from wetting to hydrodynamic forces.Therefore, linguistic variables should be employed to create nondimensional variables for more accurate combination and improved intuition, for precisely evaluating vulnerability.Moreover, the basis of risk assessment is the multiplication and product of three factors: threat, vulnerability, and consequence.As it was mentioned in Figure 1, the product of several variables, such as speed and depth with time, or social factors, is common in multiple methods, such as HURAM and the approach proposed by Tsakiris.As a result, his type of multiplication is often considered valid.

Rationality of the method
Concerning the rationality, it is sufficient to indicate the method of implication as averaging the values of height, velocity of water (impact of drag), and value of the asset.The main question that should be answered is: "What is the meaning of vulnerability here?"The vulnerability is the level of impact of water on the asset, based on its importance according to the probable casualty, damage, and security.Accordingly, the boundary of vulnerability is the same as the boundary for the maximum and minimum of all three variables based on the range of applied linguistic variables.Thus, overly exaggerating or overly weakening the overall impact is not expected.

SVA assessment based on questionnaire survey
As mentioned previously, it is necessary to assess the main parts of SVA based on the questionnaire.Accordingly, asset vulnerability for dam destruction scenarios on the downstream side is obtained and analyzed.As illustrated in Table 5, downstream dwellers have the maximum value and environmental properties have the minimum value in the asset value-determination process.The preliminary step is prioritizing threats.Based on the questionnaire results, the terrorist attack and government-supported threats are considered the highest threat probability that may occur in this dam.Government-supported threats (wars or struggles between two countries) and dissatisfied employees are rated second and third, respectively.Table 6 lists the rankings of the threats.
According to Table 7, the questionnaire results reveal that adversary forces are equally attracted to infrastructure and residential areas in the next step of SVA.Finally, biological assets (farmlands and environmental assets) are the lowest priority.The results of the attractiveness analysis show that, for these assets, such as reservoir TA B L E 5 Asset priority in terms of value.

TA B L E 6
Threat prioritization in terms of occurrence.

Threat prioritization based on (threat sources and records, potential measures, enemy's ability, and motive)
Internal and international terrorists 4 Adversary countries 3 Dissatisfied employees and employers 2 TA B L E 7 Asset prioritization in terms of target attractiveness.

Assets prioritization based on (destructions, disorders, and consequences)
Infrastructure 3

TA B L E 8
Property prioritization based on casualties, financial loss, and ecological effects.

Residential areas 5
Infrastructures 3 Farmlands 2 Environmental 2 dams, the essential and most vulnerable properties are the ones for people who live near the dam or use the potable water from this asset.The other dependent infrastructure, such as roads and residential areas near the dam, also achieve the highest attractiveness rating among other properties because of the higher probability of casualties.
Next, the consequence of any successful adversary attack for each asset is evaluated.This evaluation is based on two parameters, namely casualties and estimated damages.Table 8 shows the and ranking consequences for each of the properties.The results of the questionnaires show that residential areas are ranked first, and infrastructure ranked second.Then, farmlands and environmental properties followed, achieving similar status according to their consequence level.
Finally, the primary step of the SVA method is accomplished.In the next step, the model determines the vulnerability based on flood modeling and different types of dam-break scenarios.

3.2
Flood modeling results in different dam failure scenarios

Flood mapping, height, and speed of the flood
The maximum flood hazard zone The maximum flood hazard zone is 1309.82m, which occurs in Scenario 5 at a distance of 9026 m from the dam and 127 min after the dam fracture, presented in Figure 9.This figure illustrates the maximum flooded area in the floodplain area.

The maximum height of the flood
The maximum height of the flood is 11.248 m, which occurs in Scenario 5 at a distance of 2448 m from the dam and 115 min after fracture occurs in the dam body, presented in Figure 10.This figure illustrates the highest elevation of flooding from the downstream river's centerline.

The maximum speed of the flood
The maximum speed of the flood is 11.248 m/s, which occurs in Scenario 5 at a distance of 2448 m from the dam and 115 min after fracture occurs in the dam body.Figure 11 helps to calculate the speed of flood in this section.The figure illustrates the highest speed of flood with the highest drag force for properties in a floodplain area.

FLOOD OCCURRENCE ANALYSIS IN DIFFERENT SCENARIOS
The outputs of HEC-RAS software provided the estimated time of the flood's arrival in residential areas and population centers.Table 9 demonstrates the arrival time of the flood to population centers.

RESULTS OF ASSET VULNERABILITY MODELING IN GIS
Based on the classifications in Table 1 and application of each area, as well as the results from HEC-RAS for height, velocity, and value of the asset for each site in each scenario, the vulnerability is estimated by multiplying and integrating them into raster maps where a pixel's value shows the asset's exposure, so a higher value in each pixel indicates higher vulnerability.

F I G U R E 13
Vulnerability level of each asset in Scenario 2.

F I G U R E 14
Vulnerability level of each asset in Scenario 3.
at the end of the Section 2, the average of three criteria (water flow height, and the value), was calculated based on a range of applied linguistic variables and assigned to the zone as a vulnerability contour from red to green.It is clear that two red zones exist in the south and west of the map, revealing that residential areas and flooding may cause casualties.However, the development of the orange and yellow regions of the map mostly depends on the geographical characteristics and flood hydropower, especially in the last three scenarios.Since most of the zones are farmland, they can be evacuated during flooding, and infrastructure zones are located too far from flooding impact areas, even in the worst scenario.It is also clear that the vulnerable zones are limited to the river shoreline for the lower flooding scenarios, such as scenarios one to three.

F I G U R E 16
Vulnerability level of each asset in Scenario 5.

ASSET RISK MODELING RESULTS IN GIS
The SVA methodology was used to develop the risk map by multiplying the likelihood of vulnerability, consequences, and threat-attractiveness from Tables 5-8.The outputs of the integrated risk model were obtained in the form of raster maps, where the pixel value shows the risk for each asset and area.The pixels with higher values exhibited higher risk in each scenario.Figures 17-21 illustrate the risk of each downstream property for the five scenarios, as mentioned earlier.
In the risk maps, there are five high-risk zones with residential application.At the east end of the downstream region, there is also an infrastructure zone with moderate risk, which shows the impact of infrastructure and their serviceability on the area's risk level.Four roads and bridges also exhibited intermediate risk factors, resulting from the consequence of their probable destruction in a flood.For this case study, the risk number did not pass the high level for scenarios one to four.However, in the fifth scenario, the residential areas and infrastructure, such as roads, bridges, and pipelines (at the end of the map), exhibited high and very high-risk levels.On the other hand, farmlands and environmental assets showed less risk, which provides policymakers with improved spatial design selections.Finally, the fifth scenario should be a design scenario for the spatial design of the downstream area.The high-risk areas are identified in each scenario, and the risk-management approaches should be applied to them first, before other areas.
There are three essential contributions of this new integrated approach.First, analysts can ignore the impact of flooding in open land without particular assets when optimizing the solutions for the whole downstream area.Second, consequences play an important role in addition to vulnerability.Third, threat assessment, vulnerability analysis, and asset characterization provide spatial risk assessment, which is unique and innovative among other flood risk assessments.As discussed in the results, the risk level assessed using this hybrid method can be more precise than previous methods.Moreover, the results and their comparison can reveal the intensity of risk by changing the scenario in each   zone.The asset value and consequence intensity tend to have different levels of risk in different zones.This indicates that policymakers should integrate both approaches before implementing countermeasures for high-risk floodprone zones.

CONCLUSION
The occurrence of human threats on dams, such as military intervention or terrorist attacks, cannot be wholly restrained, but risk management can reduce losses and damages resulting from these threats.Therefore, the preliminary and most important step in reducing the risk of flooding caused by dam breaks is assessing the risk downstream.The solutions to reduce the risk of downstream properties, such as early warning or portable flood protection solutions, can be assigned for high-risk areas.This study developed a new hybrid method for vulnerability assessment and risk analysis by combining security risk approaches with the hydro-impacts of floods caused by dam breaks.Thus, SVA was selected as the primary vulnerability and risk assessment method, which is approved by the Environmental Protection Agency and American Petroleum Institute.Then, for vulnerability assessment, HEC-RAS is used to assess the velocity and height of flooding and produces flood maps for different flood scenarios.Finally, using GIS, all risk elements are incorporated in separate layers, and by integrating them, the risk map is obtained for each scenario.The most important results of the case study are expressed as follows: 1.According to the result of the questionnaire, residential areas were considered the most valuable asset using the SVA method.2. Terrorist action was identified as the highest priority among other possible threats to the earth dam, according to the questionnaire using the SVA method.3. Population centers were considered the highest priority in terms of consequences based on SVA and SPRC methods.4. The maximum height and speed of the flood are 11/248 m and 17/63 m/s, according to scenario five, acquired by HEC-RAS software.5.The minimum arrival time of the flood to a population center is 10 min, obtained for city 1 in scenario five, and the maximum arrival time of the flood to a population center is 205 min, obtained for city 7 in Scenario 1. 6. Risk modeling illustrated that farmlands had the lowest risk and population centers had the most significant risk of flooding caused by a dam break in the most dangerous scenario (Scenario 5).
As a summary of the theoretical contribution, in the first step, the method evaluates the asset and analyzes the threat based on the SVA approach.The probability of threat occurrence and threat level help to develop destruction and flood scenarios.The second step is performed to establish the quantitative analysis for flood mapping and zoning.Software, such as HEC-RAS (open-source software), is used to calculate values for different riverside land zones and integrate the speed, height, and asset values to develop vulnerability levels for each zone.Finally, the third step correlates the consequences based on the land application (infrastructure, environment, farmland, residential) with vulnerability and probability of threat for each scenario to develop a risk number for different zones.
Furthermore, the advantages of the model proposed in this paper compared with other risk assessment models can be divided into two categories.The first is related the technical approach.In previously developed models for water infrastructure, such as the CARVER or FMEA models, the total calculation is based on semi-quantitative data gathered from questionnaire surveys or interviews.However, the current method integrated the semi-quantitative data with technical data collected from hydro analysis based on velocity and height.Second, the output is based on GIS maps, which help the policymakers determine their strategies based on the application of each region.Another innovative approach is the SRPC approach, which improves the asset assessment and risk management.Although this new integrated model for vulnerability assessment and risk analysis is advised for downstream hydraulic facilities and infrastructure, such as dams, the detailed geographic information associated with applying the lands should be available for precise calculation.Using HEC-GEO RAS to directly add flood mapping in GIS maps is advisable.Future research can apply game theory, rather than a questionnaire, to evaluate the probability of threats and model destruction in the dam, potentially providing more realistic scenarios.

F I G U R E 1
Different vulnerability assessment parameters used in previous research.

F I G U R E 4
Security vulnerability assessment (SVA) model.

F I G U R E 5
Source-pathway-receptor-consequence (SPRC) model.

F I G U R E 7 Note:
Floodwater simulation process in HEC-RAS software.TA B L E 1 Classification indices used to determine the vulnerability.Colors should follow the rates of measure column.

Figures 12 -
16 show asset vulnerabilities based on each asset's height, velocity, and value in different scenarios according to the five classes.Generally, in the output of HEC-RAS, the height and velocity of the flood can be observed for each section of the floodplain area, as mentioned in Figures9-11.However, to develop the vulnerability of downstream zones, it was necessary to combine the speed, height, and value of the properties in each zone as one integrated GIS-based map.As mentioned F I G U R E 9 Cross-section of the maximum flood zone area F I G U R E 10 Cross-section of the maximum height of the flood.

FF I G U R E 12
I G U R E 11 Cross-section of the flooded area with maximum speed.TA B L E 9 Arrival time of the flood to population centers in different scenarios.Arrival time of flood to city centers (minutes) Vulnerability level of each asset in Scenario 1.

F I G U R E 15
Vulnerability level of each asset in Scenario 4.

F
I G U R E 17 Risk level of each asset in Scenario 1.

F I G U R E 18
Risk level of each asset in Scenario 2. F I G U R E 19 Risk level of each asset in Scenario 3.

F I G U R E 20
Risk level of each asset in Scenario 4.F I G U R E 21Risk level of each asset in Scenario 5.

Table 3
Downstream assets.Goals of the questionnaires.
shows tangible assets determined by experts.TA B L E 3