Multivariate compound events drive historical floods and associated losses along the U.S. East and Gulf coasts

doi:10.21203/rs.3.rs-5040855/v1

Download PDF

Article

Multivariate compound events drive historical floods and associated losses along the U.S. East and Gulf coasts

https://doi.org/10.21203/rs.3.rs-5040855/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Compound flooding events are a threat to many coastal regions and can have widespread socio-economic implications. However, their frequency of occurrence, underlying flood drivers, and direct link to past socio-economic losses are largely unknown despite being key to supporting risk and adaptation assessments. Here, we present an impact-based analysis of compound flooding for 203 coastal counties along the U.S. Gulf and East coasts by combining data from multiple flood drivers and socio-economic loss information from 1980 to 2018. We find that ~ 80% of all flood events recorded in our study area were compound rather than univariate. In addition, we show that historical compound flooding events in most counties were driven by more than two flood drivers (hydrological, meteorological, and/or oceanographic) and distinct spatial clusters exist that exhibit variability in the underlying driver of compound flood events. Furthermore, we find that in more than 80% of the counties over 80% of recorded property and crop losses were linked to compound flooding. The median cost of compound events is more than 26 times that of univariate events in terms of property loss and 76 times in terms of crop loss. Our analysis overcomes some of the limitations of previous compound-event studies based on pre-defined flood drivers and offers new insights into the complex relationship between hazards and associated socio-economic impacts.

Earth and environmental sciences/Climate sciences

Earth and environmental sciences/Environmental sciences

Earth and environmental sciences/Hydrology

Earth and environmental sciences/Natural hazards

Scientific community and society/Geography

Scientific community and society/Water resources

Compound flooding, resulting from multiple flood drivers such as precipitation, river discharge, and storm surge, poses a significant risk to coastal communities, now and in the future¹. The interactions among flood drivers can lead to catastrophic socio-economic impacts that are often of longer duration and/or higher spatial extent² than events caused by individual flood drivers³. These impacts are expected to be exacerbated by climate change (including sea-level rise along the coast) and urban development. Thus, assessments of socio-economic losses due to compound flooding events are a pressing concern^4–6. These assessments require, however, comprehensive socio-economic loss records⁷, which are typically unreliable and/or largely unavailable⁸ along most coastal regions.

The United States (U.S.) is particularly vulnerable to compound flooding due to its exposure to different flood drivers combined with extensive coastal development and population density⁹. For instance, Hurricanes Harvey^10,11 and Sandy¹² evidenced how precipitation, storm surges, and ocean waves combined can have a drastic impact on coastal metropolitan areas. These events also exposed how vulnerable critical infrastructure is to compound flooding events ^9,13, emphasizing the urgent need for improved flood hazard assessment¹⁴ and risk mitigation and adaptation measures^15,16. The U.S. East and Gulf coasts have been found to be major regional hotspots of historical and future compound flooding potential where associated impacts are projected to increase due to sea level rise and increasing storm activity^1,9,13,17,18.

Despite the increasing risk of compound flooding, the underlying drivers of such events and their socio-economic impacts remain poorly understood^7,16, particularly when compared to univariate flooding events. Identifying the nature of a particular flood event (i.e., compound or univariate) and its historical damage is challenging as it requires integrating different model datasets and socio-economic loss data, which usually come with large uncertainties^9,19. To date, most assessments have focused on estimating the likelihood, intensity, and trends of compound flooding hazards (e.g., storm surges and precipitation) using statistical and process-based models^9,19–35. It remains unknown how many of the historically observed flood events were compound events and which flooding drivers were involved in generating them. Furthermore, no link between such events and historically recorded direct damage has been established, particularly based on comprehensive, high-quality socio-economic loss data^7,36,37.

We fill this gap by conducting an impact-based analysis using empirical loss data to quantify the socio-economic impacts of past compound flooding events along the U.S. East and Gulf coasts^6,9,37–39. Here, we establish links between historical flooding events and their socio-economic losses for 203 coastal counties by combining meteorological, hydrological, and oceanographic data and flood loss information derived from the Spatial Hazard Events and Loss Database for the U.S. (SHELDUS⁴⁰) between 1980 and 2018. We analyze the occurrence of compound flooding events, their underlying drivers and associated socio-economic impact for each coastal county. The results allow us to find regional hotspots of historical compound flooding, identify and contrast the drivers that caused them, and compare our results from this impact-based approach to those reported in previous studies that focused on compound flood potential using only flood driver information⁴¹.

Spatial distribution of compound flooding events and their drivers

The U.S. experiences a considerable number of flooding events, with a total of 5,506 recorded incidents in SHELDUS along the U.S. Gulf and East coasts between 1980 and 2018 (see Fig. S1). Fig. S2 shows the total number of compound events (see Methods for how we define and identify compound events). In Fig. 1, we show the ratio of compound events versus total number of flooding events (expressed in percent) and how this varies across counties. The percentage values vary from 50–100%. We find that over 70% of total recorded flood events were compound in 83% of the counties, and no county shows less than 50% of compound events. The highest percentage of compound flooding events (ranging from 90–100%) occurred in counties along the U.S. Gulf Coast, from Louisiana to Florida’s Panhandle, and on the U.S. East Coast, between North Carolina and Delaware, and in Maine (Fig. 1). These are areas where most recorded flood events were caused by multiple flood drivers. For example, in Atlantic County (New Jersey) and in Gloucester County (Virginia) 90% and 100% of historical flood events were compound in nature, respectively. Furthermore, we find that compound event occurrence varies considerably between states, with Mississippi (MS) experiencing the highest percentage of compound flood events (89%), when averaged across all the counties, and Georgia (GA) having the lowest percentage (59.5%) (Fig. 1).

In Fig. 2, we show how often the different flood drivers were involved in creating the identified compound flooding events. We find that precipitation is most often a contributing driver, i.e., precipitation exceeded its 95th percentile threshold during 91% of all compound events (see Methods) (Fig. 2a). This highlights the importance of extreme rainfall events in creating compound flooding, including those caused by extra-tropical storms as well as hurricanes^42,43. River discharge and soil moisture were also often involved in historical compound events for most counties (85% and 83% of all compound events, respectively) (Fig. 2b and 2c). The contribution of storm surge and ocean waves (described here as significant wave height) to compound flooding events varies considerably across space, with the highest percentage (ranging from 60–90%) in counties along the East coast between North Carolina and Maine (Fig. 2d and 2e).

Figure 3 shows more detailed results for the ten counties with the highest total number of recorded flooding events in SHELDUS. Doughnut charts show how often the different flood drivers contributed to compound events in these counties. For instance, in Harris County (Texas), precipitation exceeded its 95th percentile value during 119 out of 155 total compound events. This highlights the county's vulnerability to intense rainfall events, combined with other flood drivers, which, in this case, are often associated with tropical storms and hurricanes. River discharge and soil moisture also played important roles, contributing to 114 and 102 compound events, respectively, indicating the substantial influence of upstream hydrological conditions and saturated ground on the generation of compound flood events and associated impacts. Soil moisture contribution is highest for Harris County, TX, and Charleston County, SC, and markedly lower for the other counties shown. Those same two counties also have the lowest fraction of compound events where oceanographic drivers (storm surge and wave height) contributed compared to the other eight counties.

Clustering analysis of the impacts and drivers of compound flooding

Through clustering analysis (see Methods and Supplementary Fig. S3), we identify four key clusters where in each cluster, the counties exhibit, on average, similar percentages of occurrences in the different flood drivers during compound flooding events (Fig. 4). As shown above at the individual county level, precipitation, river discharge, and soil moisture (i.e., hydrological processes) have contributed to compound flooding more frequently than storm surges and waves (i.e., oceanographic processes) in many counties; those counties are part of cluster 1 and stretch along the entire Gulf and East coasts (Fig. 5). In clusters 2, 3 and 4, we find a higher occurrence of ocean waves and storm surge during compound flooding events compared to cluster 1, particularly in clusters 2 and 3 (> 16%). These clusters are comprised of coastal counties mostly between North Carolina and New Hampshire, including Chesapeake Bay, and individual counties along the U.S. Gulf coast (Fig. 5), consistent with Fig. 2. We note that while clusters 2 and 3 look similar in Fig. 4 in terms of the average values across counties, the variability can still be very different and lead to the separation of clusters. The dendrogram in Fig. S3 shows the hierarchical relationships between all counties and allows, for example, to further compare counties within our identified main clusters in terms of compound flooding drivers and how they relate to each other.

Socio-economic impact of compound flooding and their regional distribution

Flooding events can lead to property (infrastructure) and crop damage. Figure 6 shows the percentage of historical property (Fig. 6a) and crop losses (Fig. 6c), recorded in SHELDUS, associated with compound flooding events as identified in our analysis. Over 80% of historical property and crop damage has been caused by events when multiple drivers were extreme (rather than univariate) in 92% and 81% of counties, respectively. For example, Orleans Parish, Louisiana, experienced property losses of ~ US $24 billion, almost entirely attributable to compound flooding (~ 99%). Ocean County and Monmouth County in New Jersey have recorded over US$11 billion in flood damage with over 99% of this damage due to compound flooding. Those two examples from Louisiana and New Jersey also highlight the challenges when analyzing loss data that includes extreme outlier events. Hurricanes Katrina and Sandy were by far the costliest events in those counties, and both happened to be compound events where multiple flood drivers were extreme. Hence, it is not surprising that the majority of losses are linked to compound events. To account for this, we also derive the median losses from compound and non-compound events, and the ratio between the two, for all counties (Fig. 6b and 6d). A ratio larger than one indicates that compound events were more costly than non-compound events. For property losses, this is the case for 161 (out of 203) counties, and the average ratio across those counties is 26.68 (excluding the ones where no non-compound events were recorded, and a ratio cannot be calculated). This means that compound events (in terms of the median) were more than 26 times costlier than non-compound events; for crop losses that number increases to 76.02, further highlighting the damaging effects of compound flood events. Counties with ratios smaller than one indicate that non-compound events were more expensive. That is the case for 42 counties in terms of property loss and 28 counties for crop loss. Note that in some of those cases, the overall number of recorded flood loss events was small, and no compound events occurred, leading to a ratio of zero. Overall, these results show the widespread impact of compound flooding events on built infrastructure and agricultural assets along the U.S. Gulf and East coasts.

Our study analyses the drivers of compound flooding events and their link to recorded socio-economic impacts (between 1980–2018) for 203 U.S. coastal counties. We find that ~ 80% of all flood events recorded along the U.S. Gulf and East coasts were compound (i.e., caused by multiple flood drivers) rather than univariate. In addition, despite finding considerable variability in the occurrence of these events across counties, at least 50% of total flood events in all counties were compound. These findings are consistent with previous work showing a relatively high likelihood of joint occurrence for different flood processes in these regions (e.g., rainfall and storm surge)^9,22–24,28. Furthermore, historical compound flooding events in many counties were driven by more than two flood drivers, including multiple hydrological and oceanographic processes. This analysis underscores the critical roles of precipitation, river discharge, and soil moisture as dominant drivers of such events, while coastal processes such as storm surge and ocean waves contributed considerably to compound events certain regions and counties, but much less in others. The analysis, including empirical loss data, overcomes some of the limitations of previous work on compound event analysis where only certain pre-defined pairs of drivers were included and analyzed^{9,19,25,29,30,32,41,44–50}.

Through clustering analysis, we identify different regions and counties with similar occurrences of different flood drivers during compound events. In further analysis, we find that most historical property and crop damage (more than 80% of total flood losses) have been due to compound flooding (rather than univariate) for more than 80% of the counties. This highlights the importance of considering and integrating compound flood event analysis, with all possible drivers, into hazard and risk assessments to support current and future adaptation and risk mitigation measures. This is particularly important since the coastal mean sea level continues to rise, and storm climatology with associated extreme weather events are projected to increase and intensify due to climate change along the U.S. Gulf and East coasts^1,9,13. This could exacerbate future compound flooding events due to the intensification of one or more of their drivers.

The spatial distribution of coastal counties clusters along the U.S. East and Gulf Coasts shows important patterns, with several isolated counties exhibiting characteristics different from their surrounding regions. These clusters generally group counties with similar flood driver profiles, yet outliers can emerge due to different geographical features, historical storm impacts, or limited data samples. For instance, McIntosh County in Georgia emerges as an outlier of Cluster 3, despite being surrounded by Cluster 1 counties. This classification stems from only three recorded flood events, predominantly characterized by high contributions from waves and storm surges. Such cases underscore the importance of cautious interpretation when dealing with limited data points, as they may not fully represent the complete flood hazard profile of a county and can lead to apparent anomalies in spatial clustering analyses.

Despite the importance of our results, this study has some limitations. This includes, for instance, relying on multiple hindcast and reanalysis data sets (to overcome the lack of observational records in space and time), which may not adequately capture local extremes and their variability^51–53. We also rely on historical flood event information from a county-level database (SHELDUS), which does not provide the exact location of the recorded impact. It is also important to highlight that the socio-economic impacts of such events are not only defined by the extreme flood drivers assessed here but are also dependent on vulnerability and exposure characteristics such as infrastructure resilience and population density⁵⁴. Here, we define compound events based on the exceedance of the 95th percentile value by two or more drivers (see Methods), but other definitions are possible. To test the sensitivity of our results, we repeated the same analysis using the 99th percentile as a threshold (see Fig. S4 to S6). As expected, the number of compound events decreases but the overall conclusions drawn from the analysis are unchanged. Overall, our analysis indicates that research on the complex interactions between flood drivers and associated impacts needs to continue, including improved data for both the flood drivers and recorded losses as well as model development to explore unobserved events. The identification of different flood drivers during compound flooding events over different regions and counties, as shown here, suggests that a one-size-fits-all approach to compound flood risk assessment and forecasting may be inadequate.

Historical reanalysis/hindcast and socio-economic loss data

High-resolution data for five flood drivers (river discharge, precipitation, soil moisture, storm surge, and ocean waves) were obtained from multiple sources for 203 counties along the U.S. Gulf and East coasts. The data sets cover the period from 1980 to 2018 (Table 1). All data was re-gridded onto the precipitation spatial grid (~ 6 km) using the inverse distance weighted interpolation (Fig. 7). The data was temporally aligned to daily resolution. Historical direct property and crop losses ($USD) were extracted from the Spatial Hazard Events and Losses Database for the U.S. (SHELDUS) version 20 and cover 1980 to 2018. The SHELDUS data provides information on the hazard type (here, we only focus on flooding, which includes a total 5,506 recorded flood events) and the start and end dates of the recorded impacts.

Table 1

The table below summarizes the variables, their spatial and temporal resolutions, the time periods covered, and the sources of the hydrometeorological data used in our study:
Variable	Spatial resolution	Temporal resolution	Time period	Source
Flood driver data
River discharge	0.05° x 0.05°	Daily	1980–2018	GLOFAS⁵⁵
Precipitation	0.0625° x 0.0625	Daily	1980–2018	Non-split Livneh⁵⁶
Soil moisture	0.25° x 0.25°	Hourly	1980–2018	ERA5⁵⁷
Storm surge	2.5 km	Hourly	1980–2018	CODEC⁵⁸
Significant wave height	200 m	Hourly	1980–2018	WIS⁵⁹
Socio-economic loss data
Properly damage	County	Daily	1980–2018	SHELDUS⁴⁰
Crop damage	County	Daily	1980–2018	SHELDUS⁴⁰

Classification of compound flooding and link to socio-economic losses

To provide meaningful comparison across space, we consider the relative extremeness of each flood driver based on the local climatology. To this end, we calculate percentile values at each grid point from the time series data (Table 1), normalizing the data sets across space and time. This percentile-based approach enables us to consider not only the intensity of events but also the most extreme in their local context. For each grid point in a county, we calculate the average of the percentile values for all flood drivers and select the grid point with the maximum value, which is assumed to correspond to the approximate location of the flood impact; SHELDUS only provides information at the county level with no exact location of the event available. For oceanographic drivers (storm surge and waves), percentile values are determined using the closest grid point to the coastline if our estimate of the impact location is within 6 km from the shoreline (note that 6 km is the resolution to which all other flood drivers were re-gridded).

To classify historical flood events in SHELDUS as compound or non-compound, we identify instances where at least two flood drivers related to each flood event exceeded their 95th percentile values simultaneously, following previous work⁶⁰. To link the flood drivers to socio-economic losses, we associate the flood drivers with the loss data in SHELDUS using a window of ± 1 day. This allows us to account for short delays between the occurrence of extremes and their recorded impacts in SHELDUS. To avoid potential autocorrelation between soil moisture and precipitation and river discharge, percentile values for soil moisture come from one day prior to each flood event. This approach allows us to link each extreme event identified in the historical model data to its recorded socio-economic loss data. The frequency of each flood driver during compound events was calculated by summing the number of instances when they exceeded their 95th percentile values. To assess whether our conclusions are robust against the choice of using the 95th percentile threshold, we conducted the same calculations considering the 99th percentile (see Supplementary Fig. S4-S6) as a threshold to define a compound event. This analysis only focuses on the most extreme occurrences of each flood driver. This sensitivity analysis is useful to validate our findings by ensuring that the identified patterns and relative contributions of the flood drivers remain consistent even when considering only the most severe events.

Identification of coastal counties with similar compound flood drivers

Hierarchical clustering was performed using Ward's linkage method, a well-established approach that minimizes the total within-cluster variance⁶¹. This approach provides a robust framework for identifying and delineating counties with similar frequency of flood drivers into distinct clusters (Supplementary Fig. S3). We also determine the most frequent drivers in each cluster (Fig. 4). The data for the clustering included percentile values for the five flood drivers during compound events. This allows us to identify regional patterns and county-level similarities (Fig. 5). The initial cluster distances were computed using a multidimensional approach, where pairwise Euclidean distances ($\:{D}_{\text{i},\text{j}}$) were derived based on the percentile values of flood drivers for each county. This approach ensures that the clustering process identifies counties with high similarity in the frequency and magnitude of flood drivers during compound events.

$$\:{D}_{i,j}=\:\sqrt{\sum\:_{k=1}^{w}{\left({x}_{i,k}-\:{x}_{j,k}\right)}^{2}}$$

where, $\:{D}_{i,j}$ is the Euclidean distance between counties $\:i$ and $\:j$. This distance metric represents how similar or dissimilar two counties are in terms of the contributions from the flood drivers.

$\:{x}_{i,k}$ and $\:{x}_{j,k}$ are the percentile values of the $\:k$th flood driver (precipitation, river discharge, soil moisture, storm surge, and wave) for counties $\:i$ and $\:j$, respectively, during compound events.

$\:k$ is the index representing each of the flood drivers, where $\:k=1,\:2,\:3,\:4,\:5$. corresponding to each flood driver, respectively.

$\:w$ is the total number of flood drivers considered in the clustering.

6. DATA AVAILABILITY

All reanalysis and hindcast data used in this paper are publicly available (Table 1). The socio-economic impact data used in this study are not publicly available due to license restrictions.

7. CODE AVAILABILITY

The codes supporting this study are available from the corresponding author upon reasonable request.

ACKNOWLEDGEMENTS

This work was supported by the U.S. National Academies of Sciences, Engineering, and Medicine Gulf Research Program's Thriving Communities Grant (award number: 200010880), the Megalopolitan Coastal Transformation Hub (MACH) under the National Science Foundation award ICER-2103754 and National Science Foundation CAREER grant (award number: 2141461).

AUTHOR CONTRIBUTIONS

J.A. and T.W. conceptualized the study and developed the methodology. J.A. collected and curated the data, wrote the software code, performed the formal analysis and investigation, created the visualizations, and wrote the original draft. J.M., T.W., A.R.E., M.G. and C.T.E. participated in discussions and reviewed and edited the manuscript. T.W. acquired funding and supervised the project.

COMPETING INTERESTS

The authors declare no competing interests.

Ghanbari, M., Arabi, M., Kao, S. C., Obeysekera, J. & Sweet, W. Climate Change and Changes in Compound Coastal-Riverine Flooding Hazard Along the U.S. Coasts. Earths Future 9, e2021EF002055 (2021).
Sohrabi, M., Moftakhari, H. & Moradkhani, H. Efficient Tropical Cyclone Scenario Selection Based on Cumulative Likelihood of Potential Impacts. Earths Future 11, e2023EF003731 (2023).
Deb, M. et al. Interacting Effects of Watershed and Coastal Processes on the Evolution of Compound Flooding During Hurricane Irene. Earths Future 11, e2022EF002947 (2023).
Seneviratne, S. I., et al. Weather and Climate Extreme Events in a Changing Climate. Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (2022) doi:10.1017/9781009157896.013.
Singh, D. et al. Focus on compound events. in Fifth National Climate Assessment (eds. Crimmins, A. R. et al.) (U.S. Global Change Research Program, Washington, DC, USA, 2023). doi:10.7930/NCA5.2023.F1.
van den Hurk, B. J. J. M. et al. Consideration of compound drivers and impacts in the disaster risk reduction cycle. iScience 106030 (2023) doi:10.1016/j.isci.2023.106030.
Lee, R. et al. Reclassifying historical disasters: From single to multi-hazards. Science of The Total Environment 912, 169120 (2024).
Schlosser, C. A. et al. Assessing compounding risks across multiple systems and sectors: a socio-environmental systems risk-triage approach. Frontiers in Climate 5, 1100600 (2023).
Feng, D., Tan, Z., Xu, D. & Ruby Leung, L. Understanding the compound flood risk along the coast of the contiguous United States. Hydrol Earth Syst Sci 27, 3911–3934 (2023).
Valle-Levinson, A., Olabarrieta, M. & Heilman, L. Compound flooding in Houston-Galveston Bay during Hurricane Harvey. Science of The Total Environment 747, 141272 (2020).
Sebastian, A. et al. Hindcast of pluvial, fluvial, and coastal flood damage in Houston, Texas during Hurricane Harvey (2017) using SFINCS. Natural Hazards 109, 2343–2362 (2021).
Goulart, H. M. D. et al. Compound flood impacts from Hurricane Sandy on New York City in climate-driven storylines. Natural Hazards and Earth System Sciences 24, 29–45 (2024).
Sarhadi, A. et al. Climate Change Contributions to Increasing Compound Flooding Risk in New York City. Bull Am Meteorol Soc 105, E337–E356 (2024).
Aghakouchak, A. et al. Climate Extremes and Compound Hazards in a Warming World. Annu Rev Earth Planet Sci 48, 519–548 (2020).
Aerts, J. C. J. H. et al. Climate adaptation: Evaluating flood resilience strategies for coastal megacities. Science (1979) 344, 473–475 (2014).
Lewis, M., Moftakhari, H. & Passalacqua, P. Challenges for compound coastal flood risk management in a warming climate: a case study of the Gulf Coast of the United States. Frontiers in Water 6, 1405603 (2024).
Marsooli, R., Lin, N., Emanuel, K. & Feng, K. Climate change exacerbates hurricane flood hazards along US Atlantic and Gulf Coasts in spatially varying patterns. Nature Communications 2019 10:1 10, 1–9 (2019).
Kirezci, E. et al. Projections of global-scale extreme sea levels and resulting episodic coastal flooding over the 21st Century. Scientific Reports 2020 10:1 10, 1–12 (2020).
Xu, K., Wang, C. & Bin, L. Compound flood models in coastal areas: a review of methods and uncertainty analysis. Natural Hazards 2022 116:1 116, 469–496 (2022).
Diermanse, F. et al. Probabilistic compound flood hazard analysis for coastal risk assessment: A case study in Charleston, South Carolina. Shore & Beach 9–18 (2023) doi:10.34237/1009122.
Ming, X., Liang, Q., Dawson, R., Xia, X. & Hou, J. A quantitative multi-hazard risk assessment framework for compound flooding considering hazard inter-dependencies and interactions. J Hydrol (Amst) 607, (2022).
Moftakhari, H. R., Salvadori, G., AghaKouchak, A., Sanders, B. F. & Matthew, R. A. Compounding effects of sea level rise and fluvial flooding. Proceedings of the National Academy of Sciences 114, 9785–9790 (2017).
Naseri, K. & Hummel, M. A. A Bayesian copula-based nonstationary framework for compound flood risk assessment along US coastlines. J Hydrol (Amst) 610, 128005 (2022).
Wahl, T., Jain, S., Bender, J., Meyers, S. D. & Luther, M. E. Increasing risk of compound flooding from storm surge and rainfall for major US cities. Nat Clim Chang 5, 1093–1097 (2015).
Santos, V. M., Wahl, T., Jane, R., Misra, S. K. & White, K. D. Assessing compound flooding potential with multivariate statistical models in a complex estuarine system under data constraints. J Flood Risk Manag 14, e12749 (2021).
Nasr, A. A. et al. Temporal changes in dependence between compound coastal and inland flooding drivers around the contiguous United States coastline. Weather Clim Extrem 41, 100594 (2023).
Jane, R. et al. Assessing the Potential for Compound Storm Surge and Extreme River Discharge Events at the Catchment Scale with Statistical Models: Sensitivity Analysis and Recommendations for Best Practice. (2022) doi:10.1061/(ASCE)HE.1943.
Couasnon, A. et al. Measuring compound flood potential from river discharge and storm surge extremes at the global scale. Natural Hazards and Earth System Sciences 20, 489–504 (2020).
Green, J. et al. A Comprehensive Review of Coastal Compound Flooding Literature.
Mitu, M. F., Sofia, G., Shen, X. & Anagnostou, E. N. Assessing the compound flood risk in coastal areas: Framework formulation and demonstration. J Hydrol (Amst) 626, 130278 (2023).
Khanam, M. et al. Impact of compound flood event on coastal critical infrastructures considering current and future climate. Natural Hazards and Earth System Sciences 21, 587–605 (2021).
Moftakhari, H., Schubert, J. E., AghaKouchak, A., Matthew, R. A. & Sanders, B. F. Linking statistical and hydrodynamic modeling for compound flood hazard assessment in tidal channels and estuaries. Adv Water Resour 128, 28–38 (2019).
Hendry, A. et al. Assessing the characteristics and drivers of compound flooding events around the UK coast. Hydrol Earth Syst Sci 23, 3117–3139 (2019).
Peña, F. et al. Investigating compound flooding in a low elevation coastal karst environment using multivariate statistical and 2D hydrodynamic modeling. Weather Clim Extrem 39, 100534 (2023).
Sadegh, M. et al. Multihazard Scenarios for Analysis of Compound Extreme Events. Geophys Res Lett 45, 5470–5480 (2018).
Zscheischler, J. et al. Future climate risk from compound events. Nat Clim Chang 8, 469–477 (2018).
Zscheischler, J., van den Hurk, B., Ward, P. J. & Westra, S. Multivariate extremes and compound events. in Climate Extremes and Their Implications for Impact and Risk Assessment 59–76 (Elsevier, 2020). doi:10.1016/B978-0-12-814895-2.00004-5.
Kreibich, H. et al. How to improve attribution of changes in drought and flood impacts. Hydrological Sciences Journal 64, 1–18 (2019).
Jiang, S., Tarasova, L., Yu, G. & Zscheischler, J. Compounding effects in flood drivers challenge estimates of extreme river floods. Sci. Adv 10, 4005 (2024).
CEMHS. Spatial hazard events and losses database for the United States Version 20.0 [Online Database]. Center for Emergency Management and Homeland Security, Arizona State University, Phoenix, AZ (2022).
Nasr, A. A., Wahl, T., Rashid, M. M., Camus, P. & Haigh, I. D. Assessing the dependence structure between oceanographic, fluvial, and pluvial flooding drivers along the United States coastline. Hydrol Earth Syst Sci 25, 6203–6222 (2021).
Nederhoff, K. et al. Tropical or extratropical cyclones: what drives the compound flood hazard, impact, and risk for the United States Southeast Atlantic coast? Natural Hazards 120, 8779–8825 (2024).
Ye, F. et al. A cross-scale study for compound flooding processes during Hurricane Florence. Hazards Earth Syst. Sci 21, 1703–1719 (2021).
Liu, Z. et al. A Framework for Exploring Joint Effects of Conditional Factors on Compound Floods. Water Resour Res 54, 2681–2696 (2018).
Latif, S. & Simonovic, S. P. Compounding joint impact of rainfall, storm surge and river discharge on coastal flood risk: an approach based on 3D fully nested Archimedean copulas. Environmental Earth Sciences 2023 82:2 82, 1–32 (2023).
Jalili Pirani, F. & Najafi, M. R. Characterizing compound flooding potential and the corresponding driving mechanisms across coastal environments. Stochastic Environmental Research and Risk Assessment 37, 1943–1961 (2023).
Olbert, A. I. et al. Combined statistical and hydrodynamic modelling of compound flooding in coastal areas - Methodology and application. J Hydrol (Amst) 129383 (2023) doi:10.1016/J.JHYDROL.2023.129383.
Bevacqua, E., Vousdoukas, M. I., Shepherd, T. G. & Vrac, M. Brief communication: The role of using precipitation or river discharge data when assessing global coastal compound flooding. Natural Hazards and Earth System Sciences 20, 1765–1782 (2020).
Jane, R., Cadavid, L., Obeysekera, J. & Wahl, T. Multivariate statistical modelling of the drivers of compound flood events in south Florida. Natural Hazards and Earth System Sciences 20, 2681–2699 (2020).
Ward, P. J. et al. Dependence between high sea-level and high river discharge increases flood hazard in global deltas and estuaries. Environmental Research Letters 13, (2018).
Sun, Q. et al. A Review of Global Precipitation Data Sets: Data Sources, Estimation, and Intercomparisons. Reviews of Geophysics 56, 79–107 (2018).
Beck, H. E. et al. Daily evaluation of 26 precipitation datasets using Stage-IV gauge-radar data for the CONUS. Hydrol Earth Syst Sci 23, 207–224 (2019).
Lavers, D. A., Simmons, A., Vamborg, F. & Rodwell, M. J. An evaluation of ERA5 precipitation for climate monitoring. Quarterly Journal of the Royal Meteorological Society 148, 3152–3165 (2022).
Chang, H. et al. Assessment of urban flood vulnerability using the social-ecological-technological systems framework in six US cities. Sustain Cities Soc 68, 102786 (2021).
Grimaldi, S. et al. River discharge and related historical data from the Global Flood Awareness System. v4.0. European Commission, Joint Research Centre (JRC) (2022).
Pierce, D. W. et al. An extreme-preserving long-term gridded daily precipitation data set for the conterminous United States. J Hydrometeorol 22, 1883–1895 (2021).
Hersbach, H. et al. ERA5 hourly data on single levels from 1940 to present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS) (2023).
Muis, S. et al. A High-Resolution Global Dataset of Extreme Sea Levels, Tides, and Storm Surges, Including Future Projections. Front Mar Sci 7, 263 (2020).
U.S. Army Corps of Engineers. Wave Information Studies (WIS) [Data set]. U.S. Army Engineer Research and Development Center, Coastal and Hydraulics Laboratory (2023).
Ali, J. et al. The role of compound climate and weather extreme events in creating socio-economic impacts in South Florida. Weather Clim Extrem 42, 100625 (2023).
Everitt, B. S., Landau, S., Leese, M. & Stahl, D. Cluster Analysis. (Wiley Series in Probability and Statistics, 2011).

No competing interests reported.

Alietal2024supplementaryinformation.docx

Download PDF

Reviews received at journal
29 Oct, 2024
Reviewers agreed at journal
08 Oct, 2024
Reviewers invited by journal
16 Sep, 2024
Editor assigned by journal
12 Sep, 2024
Submission checks completed at journal
10 Sep, 2024
First submitted to journal
05 Sep, 2024

You are reading this latest preprint version

Multivariate compound events drive historical floods and associated losses along the U.S. East and Gulf coasts

Status:

Version 1

Abstract

Figures

1. INTRODUCTION

2. RESULTS

3. DISCUSSION AND CONCLUSION

4. MATERIALS AND METHODS

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1