1. IntroductionA natural disaster is an actual event that causes detrimental effects while a natural hazard is the threat of an event that could cause a detrimental effect [1]. Natural disasters are created by shifts in the Earth’s general stability—whether it is movement of plates in the Earth’s crust to form an earthquake, excess rain that cannot fully infiltrate into the ground, or extremely dry areas catching fire from the heat. These often create secondary events, such as landslides or mudslides, as a result of a flooding event. While these events are not able to be restrained, it is possible to lessen the impacts and prepare as best as possible [2]. Natural disasters negatively affect people’s lives as they can be fatal, economically devastating, and environmentally depleting. This loss of life, damage to important infrastructure, and loss of resources all creates life-changing impacts that are physically, socially, economically, and environmentally damaging. Physical impacts can include damage or contamination to property, built infrastructure, and land. This results in injury, death, and loss of people, structures, animals, and crops [3]. Social impacts can be physical and/or mental health effects or destruction of household structures [3]. Economic losses are interconnected with physical impacts as well, and can be represented by costs associated with repair, replacement, and recovery [3]. Negative environmental impacts are also caused by natural disasters; for example, droughts alter water availability which causes biodiversity crises [4].Vulnerability connects natural disaster events and the level of their risk by describing the degree that the afflicted places or people may be negatively impacted [5]. There are innumerous classification systems and methods of categorizing natural hazards and natural disasters for different areas of the world and from different sources. The most significant natural hazards and natural disasters of which to investigate vulnerability using lists and indexes by the Center for Disease Control and Protection [6], United States Geological Survey [7], Center for Disease Philanthropy [8], and Federal Emergency Management Agency [1] include, but are not limited to: avalanche, drought, earthquake, extreme temperature, flood, hail, heat wave, hurricane, ice storm, landslide, lightning, strong wind, tornado, tsunami, wildfire, winter weather, and volcanic activity.Resilience represents the response to and the ability to recoup losses and recover stability after a natural disaster [5]. The Environmental Protection Agency (EPA) stated that focus on preparedness and recovery aligned with smart growth methods can help with a community’s response to natural disasters [9]. Resilience, therefore, does not only represent the reaction post-natural disaster, but is largely affected by the awareness and preparedness of a community to their vulnerability to the natural disaster in the first place. The Department for International Development (DFID) stated that overall resilience includes adaptation of livelihoods and infrastructure, anticipation of vulnerability in climate and extreme scenarios, absorption of the effects and response for recovery, and response when the actual events occur [10]. Resilience begins with awareness and protective measures for infrastructure and concludes with disaster response.Infrastructure is an important part towards the functioning of society, thus improving and maintaining infrastructure in a way that is resilient is important. A process of planning and assessing the vulnerability, designing reasonable resilience actions, implementing these actions in the area, and consistently reviewing and adapting is best advised. Some examples of proactive changes as resilience efforts are green roofs to combat extreme heat in cities or wetlands to help with coastal flooding along shorelines [11].This review focused on the vulnerability and resilience related to natural disaster events, specifically involving infrastructure that is important to the function of society during and after a natural disaster. For investigating most relevant studies, three stages of the review process were conducted, as seen in Figure 1. The first two stages were to tailor and find the most pertinent studies. Stage one revealed that flooding was the most pertinent natural disaster to investigate based on studies related to types of natural hazard and natural disaster vulnerability. Stage two determined transportation as the most critical infrastructure type in relation to flood resilience. Stage three determined keywords based on the examination of abstracts and titles of relevant studies, and then the final keywords were used to select studies most related to transportation infrastructure resilience to flood events, as directed by stages one and two. The final studies selected were reviewed. These stages are further explained and delineated in the section of Materials and Methods.

The following questions were addressed through this review considering the results of the searches of recent research:

(1)

Which natural disaster is most pertinent for vulnerability study?

(2)

Which aspect of infrastructure should be included in flood resilience study?

(3)

What is the current stage of research related to transportation infrastructure resilience to flood events?

2. Materials and MethodsThis review utilized Google Scholar and Scopus to search for scholarly articles and papers published from 1900 to 2021. Google Scholar searches scholarly literature from articles, theses, and books from multiple publishers, societies, and repositories. It was chosen as a widely used starting ground for scientific research [12]. Scopus is a database of peer-reviewed literature that is collected from journals, books, and conferences regarding science, technology, social sciences, arts, and humanities. It was chosen as it represents a main data source for over three thousand academic and corporate institutions [13]. The results found from these searches were very widespread from a variety of major journals, databases, and websites including: SpringerLink, ASCE, MDPI, Sage Journals, ScienceDirect, and Wiley Online Library. Result totals mentioned below are equivalent to the sum of both database search’s results together. An advanced search was used by one independent reviewer with the criteria of: (1) custom range in the beginning of the review from 1900 to 2021 for Google Scholar and 1961 to 2021 for Scopus since Scopus does not provide data from 1900 to 1960, (2) exclusion of citations and patents results in Google Scholar, and (3) search keywords in the title of the article in both Scopus and Google Scholar. Citations and patents were excluded as these represented sources without publication access and patents were not the format represented in studies for this review. The search criteria within Scopus were limited to article title and within Google Scholar to title only to exclude results of which the topic was not the primary focus. A variety of publications were accepted including articles, journal papers, reports, and theses until the third stage in which only peer-reviewed journal publications in English were considered. As aforementioned, this review contained three stages. Each stage’s key features can be seen in Figure 2 and each is explained in greater detail below.As mentioned previously, this review initiated with a search to find which natural hazard or natural disaster was most studied regarding vulnerability. Stage one conducted a search with seventeen natural hazards and natural disasters as mentioned above, and the word ‘vulnerability,’ since vulnerability refers to a possible level of destruction due to a natural disaster. Table 1 presents the number of studies found with each type of natural hazard or natural disaster; a total number of 6541 results were found from all natural disaster vulnerability studies. As seen in Table 1, the amount of studies related to natural hazard and natural disaster vulnerability was nearly zero from 1900 to 1980, but it began to increase from 1981 to 1990. This can be likely attributed to two factors: the increase of occurrence of several natural disasters and efforts to prepare and respond to natural disasters, such as the development of corporations that initiated extensive amounts of studies [14]. Since the 1980’s, large corporations including the Centre for Research on the Epidemiology of Disasters (CRED) and the US Agency for International Development (USAID) initiated efforts to investigate natural disasters [14]. These two factors could be linked with climate change, as the early 1980’s felt increased temperature and the late 1980s experienced drought and wildfire, and the Intergovernmental Panel on Climate Change was formed in 1989 [15].As seen in Table 1, ‘flood vulnerability’ was the most prominent with 2223 results, which confirmed this as the most decisive direction to conduct the rest of the review. The next highest was ‘drought vulnerability’ with 1383 results, and all others had lower result totals. Since studies regarding the vulnerability of floods represented the natural disaster with the highest amount of studies from a total of seventeen natural hazard and natural disaster vulnerability searches, flood was chosen as the natural disaster to further investigate. Figure 3 presents a similar trend as all natural hazards and natural disasters observed; flood vulnerability studies also increased rapidly after the 1980′s. Therefore, the authors further focused on database from 1981 to 2021 to conduct the remainder of this review.With stage two, this review persisted to find which infrastructure was most studied with flood resilience. Resilience is one of the key aspects to consider with floods since it accommodates proper management of floodwater during flooding events which lessen risk to people and infrastructure [16]. Increasing resilience is crucial to ensure the well-being of communities that are affected by flood events, and infrastructure is a component that affects resiliency of the communities. To influence resilience of communities, infrastructure handles, withstands, and restores operability to floods and thus requires alterations, changes, and proper development to handle these events. Since climate change has increased the intensity and frequencies of floods, infrastructure resilience is a high priority. This study considered critical infrastructure including the chemical, commercial facilities, communications, critical manufacturing, dams, defense industrial base, emergency services, energy, financial services, food and agriculture, government facilities, healthcare and public health, information technology, nuclear, transportation, and water and wastewater systems sectors [17].

Stage two used the keyword phrase ‘flood resilience infrastructure.’ Results from the search keyword phrase ‘flood resilience infrastructure’ totaled to 79 results. 55 results were considered since 24 results were repeated between the two databases. Each study was screened, and these 55 studies were categorized by the primary types of critical infrastructure which were involved in the study: transportation, wastewater treatment, water supply, energy, green infrastructure, health care, housing, communications, and emergency services. Transportation was focused on in 57% of these studies, wastewater treatment in 42%, energy in 34%, water supply in 32%, green infrastructure in 23%, health care in 21%, communications in 21%, housing in 19%, and emergency services in 8%. Many articles featured more than one type of infrastructure, so total percentages are not one hundred. Since transportation was the most prevalent infrastructure type, this was considered in relation to floods and resilience studies for the rest of the review.

In stage three, this study searched literature related to transportation infrastructure resilience to flood events. Based on titles and abstracts, final keywords (i.e., ‘transportation’, ‘road(s)’, and ‘transit’ with ‘flood’ and ‘flooding’) were determined. Authors included ‘flood’ and ‘flooding’ in keywords since these terminologies have slightly different definitions, and either is commonly used in studies of transportation infrastructure resilience to flood events. Flood is the natural disaster itself while flooding is the act of the natural disaster occurring. Furthermore, an option used by the authors within Google Scholar to search relevant studies was including the exact keywords in the title of the article. By using keyword combinations with ‘flood’ and ‘flooding’, the authors included all relevant studies. The searches yielded total 700 studies: 475 studies with ‘flood’ and 236 studies with ‘flooding.’ ‘Road’ and ‘roads’ were used for the same reason with Google Scholar.

This review then checked these 700 studies and excluded 566 studies. The accepted studies for this third stage were: (1) written in English and (2) peer-reviewed published journal publications with available access. Conference proceedings, books, reports, or academic papers (i.e., thesis or dissertation) were not included. Irrelevant studies (e.g., habitat modification due to road-killed snakes caused by summer flooding) were also excluded. Therefore, a total of 133 studies were further investigated.

Based on reviewing abstracts of these 133 studies, this study first determined six main research categories as they relate to transportation infrastructure resilience to flood events. These studies were categorized as aligned with the Infrastructure Resilience Planning Framework (IRPF) established by the Cybersecurity and Infrastructure Security Agency (CISA), as seen in Figure 4. The IRPF consisted of 5 total steps: (1) Lay the Foundation; (2) Critical Infrastructure Identification; (3) Risk Assessment; (4) Develop Actions; and (5) Implement and Evaluate. This framework supported the Federal Management Agency (FEMA) National Mitigation Investment Strategy and the U.S. Government Accountability Office (GAO) Disaster Resilience Framework. Therefore, this framework is applicable to any of the sixteen critical infrastructure types, including transportation infrastructure [18].

This framework is a flexible guidance to help lay the groundwork for success, prioritize critical infrastructure, understand risk, identify opportunities to improve resilience, and influence decision-making related to resilience for planning and investment decisions. Since this framework expressed this flexibility with its use, the first two steps were covered by first two stages of this review as transportation infrastructure was determined as the main area for stage three.

Research category A: analysis of flood risk in relation to transportation infrastructure. Recognition of flood risk is imperative to help future planning and investment decisions related to resiliency of transportation infrastructure [19].Research category B: flood prediction and real-time flood forecasting. According to Fan, C., et al. (2020), accurate flood forecasting would increase transportation resiliency that allows emergency managers, public officials, and other decision-makers to have more accurate and real-time flood prediction data [20].Research category C: investigation of the physical impacts of flooding on transportation infrastructure components. The World Economic Forum (2015) noted that proper assessment, understanding, and explanation of the existing risks of flooding is beneficial to heighten resilience to floods. For a proper response method to be established for floods, the problem itself must first be identified [21].Research category D: analysis of the vulnerability of transportation systems and elements related to flooding. As stated by Colon, C., et al. (2020), transport systems hold high vulnerabilities and are important before and after flooding events. By evaluating vulnerability of components of the transport network, prioritization of resilience efforts can be made to benefit economics and general function [22].Research category E: mitigation strategies or preparatory systems developed for transportation infrastructure for before and after flood events. As Gersonius, B., et al. (2016) noted, resilience strategies utilize prevention and preparedness measures to reduce effects and risks of flooding [23]. Improving effectiveness of design standards for more resilient transportation infrastructure, disaster recovery plans, and consideration of better planning measures for redundancy and flexibility of transportation infrastructure is critical to improve [19,24,25].

Research category F: the study of all other areas that could relate to transportation infrastructure resilience to flood events but fall outside the six research categories determined.

As discussed above, six research categories were aligned with Steps 3 and 4 of the IRPF. Categories A and B worked for identifying threats and hazards. Category C applied to assess consequences and infrastructure system risks. Category D represented assess vulnerability. Category E worked for refining goals and objectives, identifying and selecting resilience solutions, and developing implementation strategies. Category F applied to assess consequences, identify resilience solutions, select resilience solutions, and develop implementation strategies. There is a research gap for assessing existing resources and capabilities, implementing through existing planning mechanisms, monitoring and evaluating effectiveness, and updating plans. This is discussed in greater detail in the Discussion section. This final stage of the review investigated 133 studies, which consist of 17 studies in Category A, 11 studies in Category B, 29 studies in Category C, 25 studies in Category D, 20 studies in Category E, and 31 studies in Category F.

3. ResultsAs aforementioned in the Materials and Methods section, a final 133 studies were investigated to review the studies conducted to increase transportation infrastructure resilience to flood events. Table 2, Table 3, Table 4, Table 5, Table 6 and Table 7 present these 133 studies including the title, year of publication, authors, country of study area conducted, and the journal published within for each category. All studies are listed in a publication year order. In case a study did not apply to a specific area, the country of study area was presented as N/A.Table 2 represents the 17 studies within research category A, regarding flood risk correlated to transportation infrastructure [26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42].

Table 2. 17 Studies of category A.

Table 2. 17 Studies of category A.

Study Number:Study Title:Year:Authors:Country of Study Area:Journal:1Flood analysis and hydraulic competence of drainage structures along Addis Ababa light rail transit [26]2021Kiwanuka, M., et al.EthiopiaJournal of Environmental Science and Sustainable Development2Flooding and its relationship with land cover change, population growth, and road density [27]2021Rahman, M., et al.BangladeshGeoscience Frontiers3Flood risk assessment using the CV-TOPSIS method for the Belt and Road Initiative: an empirical study of Southeast Asia [28]2020Yan, A., et al.AsiaEcosystem Health and
Sustainability4Assessing flood probability for transportation infrastructure based on catchment characteristics, sediment connectivity and remotely sensed soil moisture [29]2019Kalantari, Z., et al.SwedenScience of The Total Environment5A Method for Urban Flood Risk Assessment and Zoning Considering Road Environments and Terrain [30]2019Chen, N., et al.ChinaSustainability6Changes concerning commute traffic distribution on a road network following the occurrence of a natural disaster—The example of a flood in the Mazovian Voivodeship (Eastern Poland) [31]2019Borowska-Stefańska, M., et al.PolandTransportation Research Part D: Transport and Environment7Analysis of Flood Vulnerability and Transit Availability with a Changing Climate in Harris County, Texas [32]2019Pulcinella, J. A., et al.USATransportation Research Record: Journal of the Transportation Research Board8Flood risk analysis for flood control and sediment transportation in sandy regions: A case study in the Loess Plateau, China [33]2018Guo, A., et al.ChinaJournal of Hydrology9A Location Intelligence System for the Assessment of Pluvial Flooding Risk and the Identification of Storm Water Pollutant Sources from Roads in
Suburbanised Areas [34]2018Szewrański, S., et al.PolandWater10The Increased Risk of Flooding in Hampton Roads: On the Roles of Sea Level Rise, Storm Surges, Hurricanes, and the Gulf Stream [35]2018Ezer, T.USAMarine Technology Society Journal11Flood probability quantification for road infrastructure: Data-driven spatial-statistical
approach and case study applications [36]2017Kalantari, Z., et al.SwedenScience of The Total Environment12Climate change in asset management of infrastructure: A riskbased methodology applied to disruption of
traffic on road networks due to the flooding of tunnels [37]2016Huibregtse, E., et al.N/AEuropean Journal of Transport and Infrastructure Research13Modeling flash floods in southern France for road management purposes [38]2016Vincendon, B., et al.FranceJournal of Hydrology14A method for mapping flood hazard along roads [39]2014Kalantari, Z., et al.SwedenJournal of Environmental Management15Flash flood risk estimation along the St. Katherine road, southern Sinai, Egypt using GIS based morphometry and satellite imagery [40]2011Youssef, A. M., et al.EgyptEnvironmental Earth Sciences16Development of a screening method to assess flood risk on Danish national roads and highway systems [41]2011Nielson, N. H., et al.DenmarkWater Science & Technology17The Environmental Impact of Flooding on Transportation Land Use in Benin City, Nigeria [42]2010Adebayo, W. O. and Jegede, O. A.NigeriaAfrican Research ReviewWithin category A, which is the flood risk analysis studies, hydrological and/or hydrodynamic modeling were often utilized to analyze flood depths. Geospatial tools were then used to display these depths which translated to flood risks. Sanyal, J., et al. (2014) used a hydrological model (HEC-HMS) to determine how land use and land cover change affected a sub-catchment and influenced the flood risk [43,44]. Kiwanuka, M., et al. (2021) conducted hydrological analysis using HEC-HMS along several roadways in Addis Ababa City, Ethiopia. Geospatial tools then helped to display the physical aspects of elevation data [26,44]. Szewrański, S., et al. (2018) developed a location intelligence system, extended from the Pluvial Risk Flood Assessment Tool. It included spatial and temporal pluvial flood analysis, elevation, and hydrologic analyses. This was used to find runoff depths and distribution of flood risks in Wrocław, Poland [34]. Nielson, N. H., et al. (2011) investigated flood risk in Jutland, Denmark with the 1-D hydrodynamic model, Mike Urban [41,45]. Geospatial methods illustrated elevation-based depressions of land surfaces that experienced flooding [41]. Youssef, A. M., et al. (2011) investigated qualitative flash flood risk analysis by incorporating remote imagery and physical data in geospatial systems in Sinai, Egypt. Morphometric analysis of the individual sub-basins was evaluated to determine the hazard from flash floods [40]. Through many of these studies, drainage systems (i.e., culverts, drains) were influential characteristics in affecting flood risk [26,34,40,41].Furthermore, there are several other efforts to investigate the flood risk. For example, Yan, A., et al. (2020) investigated historical flood risks in 11 countries within Southeast Asia, using the CV (coefficient of variation) and TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) methods. The CV method was utilized to find weights of the indicators for the flood risk assessment, and the TOPSIS method assessed the flood risk by utilizing a decision matrix [28]. Chen, N., et al. (2019) used a road risk zoning model that determined submerged depths, assessed urban flood risk with a neural network algorithm, and created flood risk maps. Spatial distribution of this flood risk varied greatly among the cities in the Chang-Zhu-Tan Urban Agglomeration (CZTUA), China [30]. Kalantari, Z., et al. (2017) utilized spatial analysis with ArcHydro to obtain the physical characteristics of the watershed and used statistical methodology (i.e., regression models) to determine and display flood probability in Västra Götaland and Värmland counties of Sweden [36,46]. Sanyal, J. and Lu, X. (2004) reviewed applications of remote imagery and spatial analysis for flood management and highlighted the importance of accurate analysis of flood depths for flood hazard mapping. This application was recommended to understand impacts of monsoons which are strong winds prevalent in south and southeastern Asia that can bring rains [47]. Islam, A., and Barman, S. D. (2020) considered morphometric characteristics (e.g., basin areas, stream number and length) to measure the floods of the Mayurakshi River, India [48]. Islam, A. and Ghosh, S. (2021) created a community-based risk assessment for riverine floods in the Rarh Plains, India that utilized the analytical hierarchy process (AHP). Flood depth was used as the determiner for flood hazard and demographic, social, infrastructure, and economic characteristics were considered [49].Table 3 represents the 11 studies related to flood prediction and real-time flood forecasting which is Research category B [20,50,51,52,53,54,55,56,57,58,59].

Table 3. 11 Studies of category B.

Table 3. 11 Studies of category B.

Study Number:Study Title:Year:Authors:Country of Study Area:Journal:1Flash flood susceptibility prediction mapping for a road network using hybrid machine learning models [50]2021Ha, H., et al.VietnamNatural Hazards2Estimating Flood Inundation Depth along the Arterial Road Based on the Rainfall Intensity [51]2021Suharyanto, A.IndonesiaCivil and Environmental Engineering3A network percolation-based contagion model of flood propagation and recession in urban road networks [20]2020Fan, C., et al.USAScientific Reports4Validating an Operational Flood Forecast Model Using Citizen Science in Hampton Roads, VA, USA [52]2019Loftis, J. D., et al.USAJournal of Marine Science and Engineering5Modeling the Impacts of Sea Level Rise on Storm Surge Inundation in Flood-Prone Urban Areas of Hampton Roads, Virginia [53]2018Castrucci, L. and Tahvildari, N.USAMarine Technology Society Journal6A Case Study for the Application of an Operational Two-Dimensional Real-Time Flooding Forecasting System and Smart Water Level Gauges on Roads in Tainan City, Taiwan [54]2018Chang, C., et al.TaiwanWater7Impact of Sea-Level Rise on Roadway Flooding in the Hampton Roads Region, Virginia [55]2017Sadler, J. M., et al.USAJournal of Infrastructure Systems8Estimation of Real-Time Flood Risk on Roads Based on Rainfall Calculated by the Revised Method of Missing Rainfall [56]2014Kim, E., et al.KoreaSustainability9Spatially distributed flood forecasting in flash flood prone areas: Application to road network
supervision in Southern France [57]2013Naulin, J., et al.FranceJournal of Hydrology10Use of radar rainfall estimates and forecasts to prevent flash flood in real time by using a road inundation warning system [58]2012Versini, P.FranceJournal of Hydrology11Vulnerability of Hampton Roads, Virginia to Storm-Surge Flooding and Sea-Level Rise [59]2006Kleinosky, L. R., et al.USANatural HazardsStudies within this category considered historical and current flood threats and/or future scenarios to better predict the flood events. Since having sufficient rainfall and water data would increase the accuracy of the prediction models, there are some related discussions and investigations. Kim, E., et al. (2014) estimated real-time flood risks by investigating historical rainfall and the probability of precipitation in Busan, Korea [56]. Chang, C., et al. (2018) found highly accurate flood forecasts by utilizing a two-dimensional real-time forecasting model with improved water gauges that includes recording and transmission of data. It helps track road inundation in real-time in Tainan City, Taiwan [54]. Naulin, J., et al. (2013) utilized spatial and temporal rainfall estimate data where water gauges were not present in the Gard Region, France and utilized this data with the hydro-meteorological forecasting approach [57]. Loftis, J. D., et al. (2019) validated accuracy for the street-level flood forecasting tool for Virginia, USA by addition of atmospheric wind and pressure data, tidal harmonic predictions, and ocean currents to their hydrodynamic model (SCHISM) and with a citizen science GPS data collection made in Hampton Roads located in Virginia to map the inundated areas as well as validate and improve predictive models for future flooding [52].Table 4 represents the 29 studies within Research category C, examination of the physical impacts of flood events on transportation infrastructure [60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88].

Table 4. 29 Studies of category C.

Table 4. 29 Studies of category C.

Study Number:Study Title:Year:Authors:Country of Study Area:Journal:1Quantifying Road-Network Robustness toward Flood-Resilient Transportation Systems [60]2021Tachaudomdach, S., et al.ThailandSustainability2Flood Impact Assessments on Transportation Networks: A Review of Methods and Associated Temporal and Spatial Scales [61]2021Rebally, A., et al.N/AFrontiers in Sustainable Cities3Flood risk assessment of the European road network [62]2021van Ginkel, K. C. H., et al.EuropeNatural Hazards and Earth System Sciences4A River Flood and Earthquake Risk Assessment of Railway Assets along the Belt and Road [63]2021Wang, Q., et al.AsiaInternational Journal of Disaster Risk Science5Flood impacts on urban transit and accessibility—A case study of Kinshasa [64]2021He, Y., et al.Democratic Republic of the CongoTransportation Research Part D: Transport and
Environment6Assessment of transportation system disruption and accessibility to critical amenities during flooding: Iowa case study [65]2021Alabbad, Y., et al.USAScience of The Total Environment7Towards Resilient Critical Infrastructures: Understanding the Impact of Coastal Flooding on the Fuel Transportation Network in the San Francisco Bay [66]2021He, Y., et al.USAInternational Journal of Geo-Information8Mere Nuisance or Growing Threat? The Physical and Economic Impact of High Tide Flooding on US Road Networks [67]2021Fant, C., et al.USAJournal of Infrastructure Systems9A systematic assessment of the effects of extreme flash floods on transportation infrastructure and circulation: The example of the 2017 Mandra flood [68]2020Diakakis, M., et al.GreeceInternational Journal of Disaster Risk Reduction10Probabilistic modeling of cascading failure risk in interdependent channel and road networks in urban flooding [69]2020Dong, S., et al.USASustainable Cities and Society11A physically based spatiotemporal method of analyzing flood impacts on urban road networks [70]2019Li, Y., et al.USANatural Hazards12Assessing the knock-on effects of flooding on road transportation [71]2019Pyatkova, K., et al.SpainJournal of Environmental
Management13Analysis of Transportation Disruptions from Recent Flooding and Volcanic Disasters in Hawaii [72]2019Kim, K., et al.USATransportation Research Record14The characteristics of road inundation during flooding events in Peninsular Malaysia [73]2019Ismail, M. S. N., et al.MalaysiaInternational Journal of GEOMATE15A topological characterization of flooding impacts on the Zurich road network [74]2019Casali, Y. and Heinimann, H. R.SwitzerlandPLoS ONE16Local floods induce large-scale abrupt failures of road networks [75]