Predicting the restoration pattern from hurricane-induced power outages from facebook data

Abstract

Extreme events such as tropical storm, tornado, hurricane cause significant disruptions to infrastructure systems including power, water, transportation, telecommunication services. Faster restoration from power outages is critical since power outages substantially impact various sectors including education, financial transactions, healthcare, and leisure. Thus, it is important to study outage restoration patterns. To develop data-driven models and test its performance on unseen hurricanes, high-resolution data from multiple hurricanes are required. However, such high-resolution power outage data from utility companies are proprietary and not easily accessible to all. Thus, the aim of this study is to demonstrate the use of macroscopic location data available from Facebook for analyzing power outage during hurricanes. First, it shows the association between population activity in Facebook and hurricane-induced power outage using the data for Hurricane Ida at a ZIP Code level. Second, it develops a data-driven model to predict power outage restoration pattern at a ZIP Code level utilizing Facebook data for Hurricanes Ida and Ian. We found that Facebook data can explain 59 % of variance in by power outages at daily level and it can explain 65 % of variance in restoration times from power outages at a ZIP code level. The data-driven model can reliably predict the restoration pattern from power outages (R 2=0.816). This study can aid researchers to choose alternative data for power outage analysis and help emergency managers and utility companies gain data-driven insights enhancing their decision-making for an impending hurricane.

Keywords

Power outage / Restoration / Hurricane / Facebook data

References

[1] U.S. Hurricane Strikes by Decade. National hurricane center and central pacific hurricane center . https://www.nhc.noaa.gov/pastdec.shtml. Accessed Jul. 29, 2024.

[2] Grenier RR, Sousounis P, Schneyer J, Raizman D. Quantifying the impact from climate change on U.S. hurricane risk. 2020. https://www.air-worldwide.com/siteassets/Publications/White-Papers/documents/air_climatechange_us_hurricane_whitepaper.pdf.

[3] Cangialosi JP, Latto AS, Berg R. National hurricane center cyclone report: hurricane irma. 2017. https://www.nhc.noaa.gov/data/tcr/AL112017_Irma.pdf.

[4] Weinner J, Postal L. Central Florida lights could Be out for days, weeks. Orlando Sentinel 2017. https://www.orlandosentinel.com/2017/09/11/central-florida-lights-could-be-out-for-days-weeks/.

[5] Beryl repair crews face threats in Houston after a week without power: NPR. https://www.npr.org/2024/07/17/nx-s1-5043279/beryl-houston-repair-crews-face-threats-still-without-power. Accessed Jul. 19, 2024.

[6] Koks E, Pant R, Thacker S, Hall JW. Understanding business disruption and economic losses due to electricity failures and flooding. Int J Disast Risk Sci 2019; 10(4): 421-38. https://doi.org/10.1007/s13753-019-00236-y.

[7] Kuntke F, Linsner S, Steinbrink E, Franken J, Reuter C. Resilience in agriculture: communication and energy infrastructure dependencies of German farmers. Int J Disast Risk Sci 2022; 13(2): 214-29. https://doi.org/10.1007/s13753-022-00404-7.

[8] NERC. Hurricane irma event analysis report. 2018. https://www.nerc.com/pa/rrm/ea/Hurricane_Irma_EAR_DL/September%202017%20Hurricane%20Irma%20Event%20Analysis%20Report.pdf.

[9] Jongman B, Wagemaker J, Revilla Romero B, Coughlan De Perez E. Early flood detection for rapid humanitarian response: harnessing near real-time satellite and twitter signals. ISPRS Int J Geoinf 2015; 4(4): 2246-66. https://doi.org/10.3390/ijgi4042246.

[10] Hsu C-W, Mostafavi A. Untangling the relationship between power outage and population activity recovery in disasters. Resil Cities Struct 2024; 3(3): 53-64. https://doi.org/10.1016/j.rcns.2024.06.003.

[11] Fan C, Mostafavi A, Gupta A, Zhang C. A system analytics framework for detecting infrastructure-related topics in disasters using social sensing. Lect Notes in Comput Sci (Including Subseries Lect Notes Artif Intell Lect Notes Bioinf) 2018: 74-91. https://doi.org/10.1007/978-3-319-91638-5_4. No. 10864 LNCS.

[12] Kryvasheyeu Y, Chen H, Moro E, Van Hentenryck P, Cebrian M. Performance of Social network sensors during Hurricane Sandy. PLoS ONE 2015; 10(2). https://doi.org/10.1371/journal.pone.0117288.

[13] Huang Q, Xiao Y. Geographic situational awareness: mining tweets for disaster preparedness, emergency response, impact, and recovery. ISPRS Int J Geoinf 2015; 4(3): 1549-68. https://doi.org/10.3390/ijgi4031549.

[14] Roy KC, Hasan S, Mozumder P. A multilabel classification approach to identify hurricane-induced infrastructure disruptions using social Media data. Comput-Aided Civil Infrastruct Eng 2020; 35(12): 1387-402. https://doi.org/10.1111/mice.12573.

[15] Liu H, Davidson RA, Apanasovich TV. Statistical forecasting of electric power restoration times in hurricanes and ice storms. IEEE Trans Power Syst 2007; 22(4): 2270-9. https://doi.org/10.1109/TPWRS.2007.907587.

[16] Homeland Security. Power outage incident annex to the response and recovery federal interagency operational plans. 2017. https://www.dhs.gov/xlibrary/assets/foia/mgmt_directive_110421_safeguarding_sensitive_but_unclassified_information.pdf.

[17] Bonanno, M. Hurricanes and Utilities: strengthening power grid resilience against Hurricanes. Constellation . https://constellationclearsight.com/blog/power-grid-resilience-against-hurricanes/. Accessed Jul. 24, 2024.

[18] Jamal TB, Hasan S. A generalized accelerated failure time model to predict restoration time from power outages. Int J Disast Risk Sci 2023; 14(6): 995-1010. https://doi.org/10.1007/s13753-023-00529-3.

[19] Mitsova D, Esnard AM, Sapat A, Lai BS. Socioeconomic vulnerability and electric power restoration timelines in Florida: the case of Hurricane Irma. Natural Hazards 2018; 94(2): 689-709. https://doi.org/10.1007/s11069-018-3413-x.

[20] Nateghi R, Guikema SD, Quiring SM. Forecasting hurricane-induced power outage durations. Nat Hazards 2014; 74(3): 1795-811. https://doi.org/10.1007/s11069-014-1270-9.

[21] Nateghi R, Guikema SD, Quiring SM. Comparison and validation of statistical methods for predicting power outage durations in the event of hurricanes. Risk Analysis 2011; 31(12): 1897-906. https://doi.org/10.1111/j.1539-6924.2011.01618.x.

[22] Dargin JS, Mostafavi A. Human-centric infrastructure resilience: uncovering well-being risk disparity due to infrastructure disruptions in disasters. PLoS ONE 2020; 15(6). https://doi.org/10.1371/journal.pone.0234381.

[23] Azad S, Ghandehari M. A study on the Association of Socioeconomic and Physical Cofactors contributing to power restoration after Hurricane Maria. IEEE Access 2021; 9: 98654-64. https://doi.org/10.1109/ACCESS.2021.3093547.

[24] Román MO, Stokes EC, Shrestha R, Wang Z, Schultz L, Sepúlveda Carlo EA, Sun Q, Bell J, Molthan A, Kalb V, Ji C, Seto KC, McClain SN, Enenkel M. Satellite-based assessment of electricity restoration efforts in Puerto Rico after Hurricane Maria. PLoS ONE 2019; 14(6). https://doi.org/10.1371/journal.pone.0218883.

[25] McRoberts DB, Quiring SM, Guikema SD. Improving hurricane power outage prediction models through the inclusion of local environmental factors. Risk Anal 2018; 38(12): 2722-37. https://doi.org/10.1111/risa.12728.

[26] Guikema SD, Nateghi R, Quiring SM, Staid A, Reilly AC, Gao M. Predicting hurricane power outages to support storm response planning. IEEE Access 2014; 2: 1364-73. https://doi.org/10.1109/ACCESS.2014.2365716.

[27] Quiring SM, Zhu L, Guikema SD. Importance of soil and elevation characteristics for modeling hurricane-induced power outages. Nat Hazards 2011; 58(1): 365-90. https://doi.org/10.1007/s11069-010-9672-9.

[28] Han SR, Guikema SD, Quiring SM. Improving the predictive accuracy of hurricane power outage forecasts using generalized additive models. Risk Anal 2009; 29(10): 1443-53. https://doi.org/10.1111/j.1539-6924.2009.01280.x.

[29] Liu CF, Mostafavi A. Revealing hazard-exposure heterophily as a latent characteristic of community resilience in social-spatial networks. Sci Rep 2023; 13(1). https://doi.org/10.1038/s41598-023-31702-9.

[30] Arora P, Ceferino L. A quasi-binomial regression model for hurricane-induced power outages during early warning. ASCE-ASME J Risk Uncert Eng Syst, Part A: Civil Eng 2024; 10(2). https://doi.org/10.1061/ajrua6.rueng-1215.

[31] Ulak MB, Kocatepe A, Konila Sriram LM, Ozguven EE, Arghandeh R. Assessment of the Hurricane-induced power outages from a demographic, socioeconomic, and transportation perspective. Nat Hazards 2018; 92(3): 1489-508. https://doi.org/10.1007/s11069-018-3260-9.

[32] Wanik DW, Anagnostou EN, Astitha M, Hartman BM, Lackmann GM, Yang J, Cerrai D, He J, Frediani MEB. A case study on power outage impacts from future Hurricane Sandy scenarios. J Appl Meteorol Climatol 2018; 57(1): 51-79. https://doi.org/10.1175/JAMC-D-16-0408.1.

[33] Mukherjee S, Nateghi R, Hastak M. A multi-hazard approach to assess severe weather-induced major power outage risks in the U.S. Reliab Eng Syst Safety 2018; 175: 283-305. https://doi.org/10.1016/j.ress.2018.03.015.

[34] Watson PL, Spaulding A, Koukoula M, Anagnostou E. Improved quantitative prediction of power outages caused by extreme weather events. Weather Clim Extremes 2022; 37. https://doi.org/10.1016/j.wace.2022.100487.

[35] Kabir E, Guikema SD, Quiring SM. Power outage prediction using data streams: an adaptive ensemble learning approach with a feature- and performance-based weighting mechanism. Risk Analysis 2024; 44(3): 686-704. https://doi.org/10.1111/risa.14211.

[36] Alpay BA, Wanik D, Watson P, Cerrai D, Liang G, Anagnostou E. Dynamic modeling of power outages caused by thunderstorms. Forecasting 2020; 2(2): 151-62. https://doi.org/10.3390/forecast2020008.

[37] Liu H, Davidson RA, Asce AM, David VRosowsky, Asce M, Stedinger JR. Negative binomial regression of electric power outages in hurricanes. J Infrastruct Syst 2005; 11(4): 258-67. https://doi.org/10.1061/ASCE1076-0342200511:4258.

[38] Alemazkoor N, Rachunok B, Chavas DR, Staid A, Louhghalam A, Nateghi R, Tootkaboni M. Hurricane-induced power outage risk under climate change is primarily driven by the uncertainty in projections of future Hurricane frequency. Sci Rep 2020; 10(1). https://doi.org/10.1038/s41598-020-72207-z.

[39] Mitsova D, Li Y, Einsteder R, Briggs TRoberts, Sapat A, Esnard A-M. Using Nighttime light data to explore the extent of power outages in the Florida Panhandle after 2018 Hurricane Michael. Remote Sens (Basel) 2024; 16(14): 2588. https://doi.org/10.3390/rs16142588.

[40] Data for good. Meta . https://dataforgood.facebook.com/dfg/about. Accessed Jul. 24, 2024.

[41] Maas P, Iyer S, Gros A, Park W, Mcgorman L, Nayak C, Dow PA. Facebook disaster maps: aggregate insights for crisis response & recovery. In: Proceedings of the 16th ISCRAM Conference; 2019. p. 836-47. https://support.google.com/crisismaps/.

[42] Cutter SL, Ash KD, Emrich CT. Urban-Rural differences in disaster resilience. Ann Am Assoc Geogr 2016; 106(6): 1236-52.

[43] Fraser T, Page-Tan C, Aldrich DP. Social capital’s impact on COVID-19 outcomes at local levels. Sci Rep 2022; 12(1). https://doi.org/10.1038/s41598-022-10275-z.

[44] Greene WH. Econometric analysis. Prentice Hall; 2003.