Uncovering Drivers of Atmospheric River Flood Damage Using Interpretable Machine Learning
Abstract
The intensity of an atmospheric river (AR) is only one of the factors influencing the damage it will cause. We use random forest models fit to hazard, exposure, and vulnerability data at different spatial and temporal scales in California to predict the probability that a given AR event will cause flood damage, as measured by National Flood Insurance Program (NFIP) claims. We first demonstrate the usefulness of data-driven models and interpretable machine learning to identify and describe drivers of AR flood damage. Hazard features, particularly measures of AR intensity such as total precipitation, increase the probability of damage with increasing values up to a threshold point, after which the probability of damage saturates. Although hazard is generally the most important risk dimension across all models, exposure and vulnerability contribute up to a third of the explanatory power. Exposure and variability features generally increase the probability of damage with increasing values, apart from a few instances which can be explained by physical intuition, but tend to affect the probability of damage less for the largest AR events. Comparisons between random forest models at different spatial and temporal scales showed general agreement. We then examine limitations inherent in publicly available exposure, vulnerability, and loss data, focusing on the difference in temporal resolution between variables from different risk dimensions and discrepancies between NFIP claims and total flood losses, and describe how those limitations may affect the model results. Overall, the application of interpretable machine learning to understand the contributions of exposure and vulnerability to AR-driven flood risk has identified potential community risk drivers and strategies for resilience, but the results must be considered in the context of the data that produced them.
Get full access to this article
View all available purchase options and get full access to this article.
Data Availability Statement
All data used in this study are publicly available, and all code created to generate results is available in a Github repository (Bowers 2023). In particular, the data sets described in Table 1 are available as downloadable CSV files, the reproduce_figures.html markdown file re-creates all figures and numerical results from this paper, and the figure4.html markdown file re-creates Fig. 4 for models at all spatial and temporal resolutions.
Acknowledgments
This material is based upon work supported by both the Stanford Graduate Fellowship and the National Science Foundation (NSF) Graduate Research Fellowship under Grant No. 1000265549. Any opinions, findings, and conclusions expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. We additionally thank Jenny Suckale and two anonymous reviewers for their helpful feedback that improved the quality of this work.
Author contributions: Corinne Bowers: Conceptualization, Data curation, Methodology, Formal analysis, Validation, Visualization, Writing–original draft, Writing–review and editing. Katherine A. Serafin: Conceptualization, Supervision, Writing–review and editing. Jack W. Baker: Conceptualization, Supervision, Writing–review and editing, Funding acquisition.
References
ACS (American Community Survey). 2023a. “DP04: Selected housing characteristics, 2009-2021.” Accessed November 21, 2023. https://data.census.gov/.
ACS (American Community Survey). 2023b. “DP05: ACS demographic and housing estimates, 2009-2021.” Accessed November 21, 2023. https://data.census.gov/.
Alipour, A., A. Ahmadalipour, P. Abbaszadeh, and H. Moradkhani. 2020. “Leveraging machine learning for predicting flash flood damage in the Southeast US.” Environ. Res. Lett. 15 (2): 024011. https://doi.org/10.1088/1748-9326/ab6edd.
Apley, D. W., and J. Zhu. 2020. “Visualizing the effects of predictor variables in black box supervised learning models.” J. R. Stat. Soc. B 82 (4): 1059–1086. https://doi.org/10.1111/rssb.12377.
Atreya, A., S. Ferreira, and E. Michel-Kerjan. 2015. “What drives households to buy flood insurance? New evidence from Georgia.” Ecol. Econ. 117 (Sep): 153–161. https://doi.org/10.1016/j.ecolecon.2015.06.024.
August, L., K. Bangia, L. Plummer, S. Prasad, K. Ranjbar, A. Slocombe, and W. Wieland. 2021. CalEnviroScreen 4.0. Sacramento, CA: California Office of Environmental Health Hazard Assessment.
Bakkensen, L. A., C. Fox-Lent, L. K. Read, and I. Linkov. 2017. “Validating resilience and vulnerability indices in the context of natural disasters.” Risk Anal. 37 (5): 982–1004. https://doi.org/10.1111/risa.12677.
Bergstrand, K., B. Mayer, B. Brumback, and Y. Zhang. 2015. “Assessing the relationship between social vulnerability and community resilience to hazards.” Social Indic. Res. 122 (2): 391–409. https://doi.org/10.1007/s11205-014-0698-3.
Blum, A. G., P. J. Ferraro, S. A. Archfield, and K. R. Ryberg. 2020. “Causal effect of impervious cover on annual flood magnitude for the United States.” Geophys. Res. Lett. 47 (5). https://doi.org/10.1029/2019GL086480.
Bowers, C. 2023. “Supplemental code release: Uncovering effects of exposure and vulnerability on atmospheric river flood damage using interpretable machine learning.” Accessed November 21, 2023. https://github.com/corinnebowers/damagedrivers.
Bradt, J. T., C. Kousky, and O. E. Wing. 2021. “Voluntary purchases and adverse selection in the market for flood insurance.” J. Environ. Econ. Manage. 110 (Oct): 102515. https://doi.org/10.1016/j.jeem.2021.102515.
Brody, S. D., H. Kim, and J. Gunn. 2013. “Examining the impacts of development patterns on flooding on the gulf of Mexico coast.” Urban Stud. 50 (4): 789–806. https://doi.org/10.1177/0042098012448551.
Brunner, M. I., A. E. Sikorska, and J. Seibert. 2018. “Bivariate analysis of floods in climate impact assessments.” Sci. Total Environ. 616-617 (Mar): 1392–1403. https://doi.org/10.1016/j.scitotenv.2017.10.176.
Cao, Q., A. Gershunov, T. Shulgina, F. M. Ralph, N. Sun, and D. P. Lettenmaier. 2020. “Floods due to atmospheric rivers along the U.S. West Coast: The role of antecedent soil moisture in a warming climate.” J. Hydrometeorol. 21 (8): 1827–1845. https://doi.org/10.1175/JHM-D-19-0242.1.
CDC (Centers for Disease Control). 2022. “CDC/ATSDR social vulnerability index, 2000-2020.” Accessed November 21, 2023. https://www.atsdr.cdc.gov/placeandhealth/svi/data.
Chawla, N. V., K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer. 2002. “SMOTE: Synthetic minority over-sampling technique.” J. Artif. Intell. Res. 16 (Jun): 321–357. https://doi.org/10.1613/jair.953.
Corringham, T. W., and D. R. Cayan. 2019. “The effect of El Niño on flood damages in the western United States.” Weather Clim. Soc. 11 (3): 489–504. https://doi.org/10.1175/WCAS-D-18-0071.1.
Corringham, T. W., F. M. Ralph, A. Gershunov, D. R. Cayan, and C. A. Talbot. 2019. “Atmospheric rivers drive flood damages in the western United States.” Sci. Adv. 5 (12): eaax4631. https://doi.org/10.1126/sciadv.aax4631.
Cutter, S. L. 2016. “The landscape of disaster resilience indicators in the USA.” Nat. Hazards 80 (2): 741–758. https://doi.org/10.1007/s11069-015-1993-2.
Czajkowski, J., G. Villarini, M. Montgomery, E. Michel-Kerjan, and R. Goska. 2017. “Assessing current and future freshwater flood risk from north Atlantic tropical cyclones via insurance claims.” Sci. Rep. 7 (1): 41609. https://doi.org/10.1038/srep41609.
Darlington, J. C., and N. Yiannakoulias. 2022. “Experimental evidence for coverage preferences in flood insurance.” Int. J. Disaster Risk Sci. 13 (2): 178–189. https://doi.org/10.1007/s13753-022-00397-3.
Debbage, N. 2019. “Multiscalar spatial analysis of urban flood risk and environmental justice in the Charlanta Megaregion, USA.” Anthropocene 28 (Dec): 100226. https://doi.org/10.1016/j.ancene.2019.100226.
DeFlorio, M. J., D. W. Pierce, D. R. Cayan, and A. J. Miller. 2013. “Western U.S. extreme precipitation events and their relation to ENSO and PDO in CCSM4.” J. Clim. 26 (12): 4231–4243. https://doi.org/10.1175/JCLI-D-12-00257.1.
Dewitz, J., and USGS. 2021. “National land cover database (NLCD) 2019 products (ver. 2.0, June 2021).” Accessed November 21, 2023. https://doi.org/10.5066/P9KZCM54.
Erlingis, J. M., M. Rodell, C. D. Peters-Lidard, B. Li, S. V. Kumar, J. S. Famiglietti, S. L. Granger, J. V. Hurley, P. Liu, and D. M. Mocko. 2021. “A high-resolution land data assimilation system optimized for the western United States [dataset].” J. Am. Water Resour. Assoc. 57 (5): 692–710. https://doi.org/10.1111/1752-1688.12910.
Estabrooks, A., T. Jo, and N. Japkowicz. 2004. “A multiple resampling method for learning from imbalanced data sets.” Comput. Intell. 20 (1): 18–36. https://doi.org/10.1111/j.0824-7935.2004.t01-1-00228.x.
FEMA. 2006. Hazus flood model technical manual. Washington, DC: Dept. of Homeland Security.
FEMA. 2020. “National flood hazard layer (NFHL).” Accessed November 21, 2023. https://www.fema.gov/flood-maps/tools-resources/flood-map-products/national-flood-hazard-layer.
FEMA. 2023a. “CRS participating communities.” Accessed November 21, 2023. https://www.fema.gov/floodplain-management/community-rating-system.
FEMA. 2023b. “OpenFEMA data sets.” Accessed November 21, 2023. https://www.fema.gov/about/openfema/data-sets.
Flanagan, B. E., E. W. Gregory, E. J. Hallisey, J. L. Heitgerd, and B. Lewis. 2011. “A social vulnerability index for disaster management.” J. Homeland Secur. Emerg. Manage. 8 (1). https://doi.org/10.2202/1547-7355.1792.
Gelaro, R., et al. 2017. “The modern-era retrospective analysis for research and applications, version 2 (MERRA-2) [dataset].” J. Clim. 30 (14): 5419–5454. https://doi.org/10.1175/JCLI-D-16-0758.1.
Gourevitch, J. D., and N. Pinter. 2023. “Federal incentives for community-level climate adaptation: An evaluation of FEMA’s community rating system.” Environ. Res. Lett. 18 (3): 034037. https://doi.org/10.1088/1748-9326/acbaae.
Heiman, E. R. 2022. “Protecting renters from flood loss.” Univ. Pennsylvania Law Rev. 170 (3): 783–809.
Highfield, W. E., and S. D. Brody. 2017. “Determining the effects of the FEMA Community Rating System program on flood losses in the United States.” Int. J. Disaster Risk Reduct. 21 (Mar): 396–404. https://doi.org/10.1016/j.ijdrr.2017.01.013.
Highfield, W. E., S. D. Brody, and C. Shepard. 2018. “The effects of estuarine wetlands on flood losses associated with storm surge.” Ocean Coastal Manage. 157 (May): 50–55. https://doi.org/10.1016/j.ocecoaman.2018.02.017.
Highfield, W. E., W. G. Peacock, and S. Van Zandt. 2014. “Mitigation planning: Why hazard exposure, structural vulnerability, and social vulnerability matter.” J. Plann. Educ. Res. 34 (3): 287–300. https://doi.org/10.1177/0739456X14531828.
Ismail-Zadeh, A. T., S. L. Cutter, K. Takeuchi, and D. Paton. 2017. “Forging a paradigm shift in disaster science.” Nat. Hazards 86 (Mar): 969–988. https://doi.org/10.1007/s11069-016-2726-x.
James, G., D. Witten, T. Hastie, and R. Tibshirani. 2013. An introduction to statistical learning. Springer texts in statistics. New York: Springer.
James, L. A., and M. B. Singer. 2008. “Development of the lower Sacramento valley flood-control system: Historical perspective.” Nat. Hazard. Rev. 9 (3): 125–135. https://doi.org/10.1061/(ASCE)1527-6988(2008)9:3(125).
Jonkman, S. N. 2005. “Global perspectives on loss of human life caused by floods.” Nat. Hazards 34 (2): 151–175. https://doi.org/10.1007/s11069-004-8891-3.
Knighton, J., B. Buchanan, C. Guzman, R. Elliott, E. White, and B. Rahm. 2020. “Predicting flood insurance claims with hydrologic and socioeconomic demographics via machine learning: Exploring the roles of topography, minority populations, and political dissimilarity.” J. Environ. Manage. 272 (Oct): 111051. https://doi.org/10.1016/j.jenvman.2020.111051.
Komolafe, A., S. Herath, and R. Avtar. 2018. “Sensitivity of flood damage estimation to spatial resolution.” J. Flood Risk Manage. 11 (Jan): S370–S381. https://doi.org/10.1111/jfr3.12224.
Konrad, C. P., and M. D. Dettinger. 2017. “Flood runoff in relation to water vapor transport by atmospheric rivers over the western United States, 1949–2015.” Geophys. Res. Lett. 44 (22): 11456–11462. https://doi.org/10.1002/2017GL075399.
Kotsiantis, S., D. Kanellopoulos, and P. Pintelas. 2006. “Handling imbalanced datasets: A review.” GESTS Int. Trans. Comput. Sci. Eng. 30 (1): 25–36.
Kousky, C. 2011. “Understanding the demand for flood insurance.” Nat. Hazard. Rev. 12 (2): 96–110. https://doi.org/10.1061/(ASCE)NH.1527-6996.0000025.
Lamjiri, M. A., M. D. Dettinger, F. M. Ralph, and B. Guan. 2017. “Hourly storm characteristics along the U.S. West Coast: Role of atmospheric rivers in extreme precipitation.” Geophys. Res. Lett. 44 (13): 7020–7028. https://doi.org/10.1002/2017GL074193.
Lundberg, S. M., and S.-I. Lee. 2017. “A unified approach to interpreting model predictions.” In Proc., 31st Conf. on Neural Information Processing Systems (NIPS 2017). La Jolla, CA: Neural Information Processing Systems Foundation.
Merz, B., H. Kreibich, and U. Lall. 2013. “Multi-variate flood damage assessment: A tree-based data-mining approach.” Nat. Hazards Earth Syst. Sci. 13 (1): 53–64. https://doi.org/10.5194/nhess-13-53-2013.
Merz, B., H. Kreibich, R. Schwarze, and A. H. Thieken. 2010. “Assessment of economic flood damage.” Nat. Hazards Earth Syst. Sci. 10 (8): 1697–1724. https://doi.org/10.5194/nhess-10-1697-2010.
Mobley, W., A. Sebastian, R. Blessing, W. E. Highfield, L. Stearns, and S. D. Brody. 2021. “Quantification of continuous flood hazard using random forest classification and flood insurance claims at large spatial scales: A pilot study in southeast Texas.” Nat. Hazards Earth Syst. Sci. 21 (2): 807–822. https://doi.org/10.5194/nhess-21-807-2021.
Molnar, C. 2023. “Interpretable machine learning: A guide for making black box models explainable.” Accessed November 21, 2023. https://christophm.github.io/interpretable-ml-book/.
NOAA (National Oceanic and Atmospheric Administration). n.d.-a. “Multivariate ENSO Index Version 2 (MEI.v2).” Accessed November 21, 2023. https://psl.noaa.gov/enso/mei/.
NOAA (National Oceanic and Atmospheric Administration). n.d.-b. “Pacific Decadal Oscillation (PDO).” Accessed November 21, 2023. https://www.ncei.noaa.gov/access/monitoring/pdo/.
NOAA NCEI (National Oceanic and Atmospheric Administration and National Centers for Environmental Information). 2023. “U.S. billion-dollar weather and climate disasters.” Accessed November 21, 2023. https://www.ncei.noaa.gov/access/billions/.
PEP (Population Estimates Program). 2023. “Population and Housing Unit Estimates Tables.” Accessed November 21, 2023. https://www.census.gov/programs-surveys/popest/data/tables.html.
Pollack, A. B., I. Sue Wing, and C. Nolte. 2022. “Aggregation bias and its drivers in large-scale flood loss estimation: A Massachusetts case study.” J. Flood Risk Manage. 15 (4): 1–16. https://doi.org/10.1111/jfr3.12851.
Ralph, F. M., J. J. Rutz, J. M. Cordeira, M. D. Dettinger, M. Anderson, D. Reynolds, L. J. Schick, and C. Smallcomb. 2019. “A scale to characterize the strength and impacts of atmospheric rivers.” Bull. Am. Meteorol. Soc. 100 (2): 269–289. https://doi.org/10.1175/BAMS-D-18-0023.1.
Rözer, V., H. Kreibich, K. Schröter, M. Müller, N. Sairam, J. Doss-Gollin, U. Lall, and B. Merz. 2019. “Probabilistic models significantly reduce uncertainty in Hurricane Harvey pluvial flood loss estimates.” Earth’s Future 7 (4): 384–394. https://doi.org/10.1029/2018EF001074.
Rufat, S., E. Tate, C. G. Burton, and A. S. Maroof. 2015. “Social vulnerability to floods: Review of case studies and implications for measurement.” Int. J. Disaster Risk Reduct. 14 (Part 4): 470–486. https://doi.org/10.1016/j.ijdrr.2015.09.013.
Rutz, J. J., W. J. Steenburgh, and F. M. Ralph. 2014. “Climatological characteristics of atmospheric rivers and their inland penetration over the western united states.” Mon. Weather Rev. 142 (2): 905–921. https://doi.org/10.1175/MWR-D-13-00168.1.
Sadiq, A. A., J. Tyler, and D. Noonan. 2020. “Participation and non-participation in FEMA’s community rating system (CRS) program: Insights from CRS coordinators and floodplain managers.” Int. J. Disaster Risk Reduct. 48 (Sep): 101574. https://doi.org/10.1016/j.ijdrr.2020.101574.
Sadler, J., J. Goodall, M. Morsy, and K. Spencer. 2018. “Modeling urban coastal flood severity from crowd-sourced flood reports using Poisson regression and random forest.” J. Hydrol. 559 (Apr): 43–55. https://doi.org/10.1016/j.jhydrol.2018.01.044.
Saito, T., and M. Rehmsmeier. 2015. “The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets.” PLoS One 10 (3): e0118432. https://doi.org/10.1371/journal.pone.0118432.
Sanders, B. F., J. E. Schubert, D. T. Kahl, K. J. Mach, D. Brady, A. AghaKouchak, F. Forman, R. A. Matthew, N. Ulibarri, and S. J. Davis. 2022. “Large and inequitable flood risks in Los Angeles, California.” Nat. Sustainability 6 (1): 47–57. https://doi.org/10.1038/s41893-022-00977-7.
Smith, A. B., and R. W. Katz. 2013. “US billion-dollar weather and climate disasters: Data sources, trends, accuracy and biases.” Nat. Hazards 67 (2): 387–410. https://doi.org/10.1007/s11069-013-0566-5.
Solomatine, D. P., and A. Ostfeld. 2008. “Data-driven modelling: Some past experiences and new approaches.” J. Hydroinf. 10 (1): 3–22. https://doi.org/10.2166/hydro.2008.015.
Stein, L., M. P. Clark, W. J. Knoben, F. Pianosi, and R. A. Woods. 2021. “How do climate and catchment attributes influence flood generating processes? A large-sample study for 671 catchments across the contiguous USA.” Water Resour. Res. 57 (4): 1–21. https://doi.org/10.1029/2020WR028300.
Strobl, C., A. L. Boulesteix, A. Zeileis, and T. Hothorn. 2007. “Bias in random forest variable importance measures: Illustrations, sources and a solution.” BMC Bioinf. 8 (25): 1–21. https://doi.org/10.1186/1471-2105-8-25.
Szczyrba, L., Y. Zhang, D. Pamukcu, D. I. Eroglu, and R. Weiss. 2021. “Quantifying the role of vulnerability in hurricane damage via a machine learning case study.” Nat. Hazard. Rev. 22 (3): 1–12. https://doi.org/10.1061/(ASCE)NH.1527-6996.0000460.
Tate, E., M. A. Rahman, C. T. Emrich, and C. C. Sampson. 2021. “Flood exposure and social vulnerability in the United States.” Nat. Hazards 106 (1): 435–457. https://doi.org/10.1007/s11069-020-04470-2.
Thomson, H., H. B. Zeff, R. Kleiman, A. Sebastian, and G. W. Characklis. 2023. “Systemic financial risk arising from residential flood losses.” Earth’s Future 11 (4): 86. https://doi.org/10.1029/2022EF003206.
USGS. 2023. “National hydrography dataset.” Accessed November 21, 2023. https://www.usgs.gov/national-hydrography/access-national-hydrography-products.
Wagenaar, D., J. de Jong, and L. M. Bouwer. 2017. “Multi-variable flood damage modelling with limited data using supervised learning approaches.” Nat. Hazards Earth Syst. Sci. 17 (9): 1683–1696. https://doi.org/10.5194/nhess-17-1683-2017.
Willis, H., A. Narayanan, J. Fischbach, E. Molina-Perez, C. Stelzner, K. Loa, and L. Kendrick. 2016. Current and future exposure of infrastructure in the United States to natural hazards. Santa Monica, CA: RAND.
Woldemeskel, F., and A. Sharma. 2016. “Should flood regimes change in a warming climate? The role of antecedent moisture conditions.” Geophys. Res. Lett. 43 (14): 7556–7563. https://doi.org/10.1002/2016GL069448.
Zahran, S., S. Weiler, S. D. Brody, M. K. Lindell, and W. E. Highfield. 2009. “Modeling national flood insurance policy holding at the county scale in Florida, 1999–2005.” Ecol. Econ. 68 (10): 2627–2636. https://doi.org/10.1016/j.ecolecon.2009.04.021.
Information & Authors
Information
Published In
Natural Hazards Review
Volume 25 • Issue 3 • August 2024
Copyright
© 2024 American Society of Civil Engineers.
History
Received: Aug 14, 2023
Accepted: Jan 16, 2024
Published online: May 2, 2024
Published in print: Aug 1, 2024
Discussion open until: Oct 2, 2024
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.