Compounding Disasters in Gulf Coast Communities 2020-2021: Impacts, Findings, and Lessons Learned (2024)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Compounding Disasters in The Gulf of Mexico Region: 2020– 2021 This chapter begins with an overview of the disruptive weather-climate events of 2020– 2021 in the Gulf of Mexico (GOM) region. Although these events occurred with notable frequency during this period, extreme and disruptive weather-climate events are not an unusual occurrence in the region. What sets the events in this time frame apart is their co-occurrence with a public health emergency—the COVID-19 pandemic. In tandem with the pandemic, each of the weather-climate events in 2020–2021 had consequences for the public’s health in communities affected by these events. The chapter then turns to focus on the content of the regional information-gathering sessions in Harris and Galveston Counties, Texas; Cameron and Calcasieu Parishes, Louisiana; and Baldwin and Mobile Counties, Alabama. This section of the chapter includes a description of the format of the information-gathering sessions, summaries of the content of the sessions, and discussion of the common themes that arose in each session. SYNOPSIS OF DISRUPTIVE EVENTS Detailing the costs, frequency, and number of previous disasters in the GOM region provides important context; however, to fully grasp the toll these compounding disasters exact on affected individuals and communities, a closer, qualitative examination is required. Informed directly by the voices of residents and practitioners working and living in the GOM region, this chapter details how disaster impacts compounded, intersecting with a complex set of 104 PREPUBLICATION | UNCORRECTED PROOFS 

sociopolitical, economic, cultural, infrastructure, and public health vulnerabilities among highly exposed and vulnerable populations, challenging the capacity of communities to recover. The COVID-19 Pandemic The COVID-19 pandemic has been intrinsically and inherently a public health event, both in form—a communicable disease—and in terms of leadership, with public health agencies and professionals, from the global to the local level, leading the response. COVID-19 spread expansively in time and space throughout the GOM region, complicating the dynamics of all other disaster events throughout the region. A graphic from the Fifth National Climate Assessment (USGCRP, 2023) illustrates how the overarching pandemic overlapped temporally with and was overlaid spatially upon other U.S. disasters throughout 2020 and 2021 (see Figure 3-1). Table 3-1 shows a comparison of the impacts on public/population health from 2020–2021 hurricanes determined to be billion-dollar disasters, Winter Storm Uri, and COVID-19; Table 3-2 shows total deaths, deaths per 100,000 GOM residents attributable to COVID-19, and rating GOM states for COVID-19 mortality. 105 PREPUBLICATION | UNCORRECTED PROOFS 

FIGURE 3-1 Temporal compounding of disruptive events in 2020–2021. The Gulf of Mexico region is included as part of the East Coast. SOURCE: USGCRP, 2023. TABLE 3-1 Profile of Disaster Impacts on Public/Population Health for Some 2020–2021 Disasters Disaster Health Hurricanesa Winter Storm Uri COVID-19 Consequences GOM Economic Costs 1–75 billionb 200–300 billionc Trillionsd ($) Harms: Physical Health 106 PREPUBLICATION | UNCORRECTED PROOFS 

Deaths 173e 246f 167,000g Injuries Several thousand Several thousand Negligible emergency room emergency room visitsh visitsi Disease cases Negligible Hundreds (not 11 million confirmedj related to COVID- 19) Medically high- Increased risk Increased risk Life-threatening risk risk patients Matrix: Psychological Stressorsk Potentially Exposure to storm Exposure to storm COVID-19 illness traumatic events hazards hazards Hospitalization, ICU admission, Personal injury Personal injury Mechanical ventilation Perceived threat to Perceived threat to Life threat life life Exposures at work Loss Loss of life Loss of life Loss of life Damage/loss of Damage/loss of Isolation home, possessions home Loss of employment Life change Displacement from Displacement from Work/school disruptions home home Isolation Financial hardship Financial hardship Financial hardship Workforce stress Increased workload/hours Mental exhaustion Balancing work/home demands Health Care Access and Disruption of Services Damage to Moderate to severe Moderate to severe None facilities 107 PREPUBLICATION | UNCORRECTED PROOFS 

Damage to Moderate to severe Moderate to severe None power, infrastructure Redirected Facility-specific— Facility-specific— Extensive—entire units/facilities services related to impact on related to impact on shifted to COVID-19 facility facility Prolonged and periodic Time-limited Time-limited Disruption of Facility-specific— Facility-specific— Extensive, widespread services related to impact on related to impact on Prolonged and periodic facility facility Time-limited Time-limited Public Health Services and Workforce Public health role Active Active participatory Lead role participatory role role Negligible Negligible Risk of Tens of thousands of public harm/death for health professionals sickened and public health hundreds of frontline personnel personnel died in GOM region Disruption of Major but brief Major but brief Extreme and prolonged services Community level Statewide in Texas throughout GOM region Work stress for Significant Severe/relatively Unprecedented/prolonged personnel brief Increased workload/hours Mental exhaustion Balancing work/home demands a Espinel, Galea et al., 2019; Espinel, Kossin et al., 2019; Shultz, Kossin, Hertelendy et al., 2020; Shultz et al., 2022. b Xu et al., 2022. c TASCE, 2022. d CMS, n.d.; Hlávka and Rose, 2022. e Xu et al., 2022. f Texas Department of State Health Services (TDSHS), 2021. g CDC, 2023c h AHRQ, 2020. i Cindass et al., 2023; Orton et al., 2022. 108 PREPUBLICATION | UNCORRECTED PROOFS 

j JHU, 2024. k Shultz, Espinel, Galea et al., 2007; Shultz et al., 2013; Shultz et al., 2017. COVID-19 Mortality in Gulf of Mexico States COVID-19 was the third leading cause of death in the United States in 2020 and 2021. COVID-19 deaths dwarfed numbers of fatalities from all forms of disasters nationwide and in the GOM region. Four of the five GOM states ranked among the highest in COVID-19 morbidity rates in the nation in 2020, and three of the five ranked among the highest in 2021 (CDC, 2023c). The racial disparities associated with COVID-19 were significant. In Louisiana, for example, Black Americans accounted for 50 percent of known COVID-19-related deaths while representing only 32 percent of the state’s population (Hu et al., 2022). Links between disparate exposure to air pollution are predicted to be one of several factors contributing to the disproportionate deaths among Black Americans, racial/ethnic minorities, and other vulnerable groups (Terrell and James, 2022; Chakraborty, 2021). TABLE 3-2 Total Deaths, Deaths per 100,000 Residents Attributable to COVID-19, and GOM State Rankings for COVID-19 Mortality Rates 2020 2021 Deaths Rate/ State Rank Deaths Rate/ State Rank for COVID- for COVID- 100,000 100,000 19 Mortality 19 Mortality Residents Residents Rate Rate U.S.* 350,831 85.0 - 416,893 104.1 - Louisiana 6,533 118.0 6 6,329 116.9 16 109 PREPUBLICATION | UNCORRECTED PROOFS 

Mississippi 4,466 123.5 4 5,082 146.3 5 Texas 30,840 105.2 9 44516 151.4 3 Alabama 6,544 103.6 11 9,491 152.8 2 *U.S. data from Xu et al., 2022. SOURCE: CDC, 2023c; Xu et al., 2022. COVID-19 Complications and Impediments for Hurricane Response One of the most urgent problems posed by COVID-19 was how to safely evacuate and shelter populations in the path of hurricanes and tropical systems. Salas et al. (2020), Shultz, Kossin, Hertelendy et al. (2020), Shultz, Fugate et al. (2020), and Shultz, Kossin, Ali et al. (2020) have described the challenges and some clear incompatibilities that became evident in the early months of the pandemic. Specifically, Shultz, Kossin, Hertelendy et al. (2020, p. 494) characterize them as follows: Climate-driven hazards underscore the imperative for timely warning, evacuation, and sheltering of storm-threatened populations—proven life-saving protective measures that gather evacuees together inside durable, enclosed spaces when a hurricane approaches. Meanwhile, the rapid acquisition of scientific knowledge regarding how COVID-19 spreads has guided mass anti-contagion strategies, including lockdowns, sheltering at home, and physical distancing; life-saving strategies [that] separate and move people apart. Gathering people together (e.g., in an evacuation shelter) for hurricane protection conflicts logistically with moving people apart to minimize risk of transmission of airborne communicable diseases. It is impossible to implement both approaches optimally and concurrently (Salas et al., 2020; Shultz, Kossin, Hertelendy et al., 2020; Shultz, Fugate et al., 2020). 110 PREPUBLICATION | UNCORRECTED PROOFS 

GOM states were among the first to ease restrictions and reopen their economies (Miller, 2020) after the lockdowns began in March 2020. Five months later, by mid-August 2020, GOM states experienced a massive COVID-19 case surge, leading the nation in newly diagnosed cases, hospitalizations, and deaths. Simultaneously, the 2020 Atlantic hurricane season produced a record total 30 named storms, including 11 that made landfall in the United States (NOAA, 2021m). 1 In August 2021, the situation in the GOM region was repeated when, again, GOM states led the nation in a massive flare-up of COVID-19 cases. This time, however, it was the virulent Delta variant circulating—the most lethal variant during the course of the pandemic (Patrick et al., 2022). COVID-19 Vaccination Rates in Gulf of Mexico States Vaccines against COVID-19 were available by November 2021, but by then, Mississippi, Alabama, and Louisiana ranked first, second, and fourth, respectively, among states for the highest number of deaths per capita due to COVID-19. Florida was not far behind at 9th, and Texas was 19th (Statista, 2021). At that same time, just 46.6 percent of Mississippians, 48.4 percent of Louisianans, and 45.6 percent of Alabamans were fully vaccinated (CDC, 2023d). These 3 states were among the 10 states with the lowest vaccination rates; their hospitalization rates were nearly 4 times higher and death rates more than 5.5 times higher than those 10 states with the highest vaccination rates (Hanna et al., 2021). The low vaccination rates persisted through October 2022, when Louisiana, Mississippi, and Alabama, ranked 48th, 49th, and 50th, respectively, on the percentage of vaccinated 18- to 64-year-olds. COVID-19 and Economic and Employment Stability The COVID-19 pandemic significantly affected economic and employment stability nationwide (U.S. Census Bureau, 2024). Several years into the pandemic, incomes were lower 1 At the time of writing, the 2020 hurricane season continues to hold the title of the most active hurricane season on record. 111 PREPUBLICATION | UNCORRECTED PROOFS 

overall and by state compared with 2019, just prior to the pandemic (U.S. Census Bureau, 2021c). Beginning in 2021, the sharp rise in inflation outstripped the rise in income, producing the net effect of decreasing buying power and the value of household savings, which were already low for GOM states (U.S. Bureau of Labor Statistics, 2023d). The nation’s overall unemployment rate, which reached 14.8 percent during the pandemic, was extremely high (39.3 percent) for those working in the leisure and hospitality industries because of stay-at-home orders (Falk et al., 2020). Those who remained in the service, manufacturing, and transportation industries were required to report physically to work, putting them at higher risk of contracting COVID-19 because of increased exposure and lack of sufficient workplace safety measures. Throughout 2020–2021, the emergence and spread of COVID-19 transformed the public health risk of compounding disasters in the GOM region by amplifying health- compromising exposures and underlying vulnerabilities and modifying emergency procedures for weather-climate events. Trivedi (2023, p. 1) highlights that, “while compounding disasters have long existed, the duration and scope of the COVID-19 pandemic both drew additional attention to them and the challenges of addressing them” (citing Tozier de la Poterie et al., 2022). Hurricanes and Tropical Storms In 2020–2021, three Category 3 or higher storms—Laura, Zeta, and Ida—made landfall in the GOM region (all in Louisiana). Additionally, Hurricane Sally (Category 2) made landfall in Alabama, and Hurricane Delta (Category 2) made landfall in Louisiana, and Hurricanes Hanna and Nicholas (both Category 1) made landfall in Texas. The events of these 2 years were situated in the context of long-term climate change and short-term variability associated with a “double- dip” La Niña (NOAA, 2021b). The GOM region’s notable hurricanes and tropical storms are summarized in Table 3-3. 112 PREPUBLICATION | UNCORRECTED PROOFS 

TABLE 3-3 Summary of the Gulf of Mexico Region’s Hurricanes and Tropical Storms 2020– 2021 Hazard and Description Landfall Affected Monetary Direct Date Areas Damage in Fatalities the United States ($) Hurricane Hanna 7/23/20 S TX 1B+ 5 Hanna’s prehurricane impacts affected virtually the entire GOM region. Hanna’s primary hazards included excessive rainfall, rip currents, and winds. New Orleans experienced street flooding as the outer bands from the developing storm skirted the Gulf coast (NOAA, 2021c). Hurricane Laura 8/27/20 SW LA, 19B 47 Laura was a Category 4 storm that made landfall SE TX, in Cameron Parish, Louisiana. Winds in excess SW MS of 150 mph, rainfall, and storm surge inundation levels of 12 to 18 ft above ground were described as “unsurvivable” and “catastrophic.” Laura caused significant damage in coastal regions and inland cities, such as Lake Charles. Extensive housing and infrastructure damage included 568,000 power outages in Texas and Louisiana and an estimated 10,000 demolished homes in Louisiana. Shortly after the core of the hurricane passed Westlake, near Lake Charles, Laura triggered a fire at a chemical plant that discharged a cloud of potentially toxic smoke that stretched for miles. Laura was the strongest hurricane, in terms of maximum sustained wind speed at landfall, to affect Louisiana since 1856. 113 PREPUBLICATION | UNCORRECTED PROOFS 

The World Meteorological Organization retired the name Laura after this storm (NOAA, 2021d). Hurricane Sally 9/16/20 AL, W FL 7.3B 4 Sally was a Category 2 storm that made landfall and near Gulf Shores, Alabama. Like many recent Panhandle, hurricanes (Harvey in 2017, Florence in 2018, GA and Dorian in 2019), Sally was a slow-moving tropical cyclone, producing significant flooding. While the storm featured 100+ mph winds, the rainfall totals in excess of 2 ft is a signature fingerprint of this storm (NOAA, 2021e). Tropical Storm Beta 9/17/20 SE TX, 225M 1 Tropical Storm Beta caused widespread flooding LA, MS, in the Houston area, disrupting major AL transportation corridors and prompting numerous water rescues (NOAA, 2021f). Hurricane Delta 10/9/20 LA, MS 2.9B+ 2 Winds and rainfall from Delta compromised already weakened structures damaged by Hurricane Laura. Impacts (structural damage and power loss) from Delta were felt in Louisiana and Mississippi (NOAA, 2021g). Hurricane Zeta 10/28/20 SE LA, 4B 5 Hurricane Zeta attained “major hurricane” status Alabama, as it made landfall in Cocodrie, Louisiana. It is SW MS, N the latest landfalling major hurricane on record GA for the continental United States. Zeta was a Category 3 storm and produced storm surge inundation of 6 to 10 ft above ground level along the Mississippi coast (particularly in the back bays) (NOAA, 2021h). 114 PREPUBLICATION | UNCORRECTED PROOFS 

Hurricane Eta 11/9/20 FL, GA, 1.5B 7 The final hurricane of the 2020 season, SC, NC Hurricane Eta made three total landfalls, the latter two in Florida on November 9 and 12. Eta caused major flooding from storm surge and heavy rainfall, mainly in South Florida (NOAA, 2021i). Tropical Storm Claudette 6/18/21 SE LA, 375M 4 Claudette moved onshore south of Houma, MS, AL, Louisiana. Nearly a foot of rainfall caused GA, SC, damage in parts of Louisiana and Mississippi. NC Multiple tornadoes were spawned, including a long-track EF2* tornado in Alabama, where multiple fatalities were reported (NOAA, 2022d). Hurricane Ida 8/29/21 LA, NE 75B+ 96 Ida was a multihazard storm covering multiple MS, NW geographic regions, with initial impacts in the AL, E and GOM region. At Category 4, the hurricane made central and landfall close to Port Fourchon, Louisiana, with E TN, maximum sustained winds of at least 150 mph portions of (240 km/h) and a minimum central pressure of KY and 930 millibars. Ida was one of three hurricanes in VA, SW recorded history to make landfall in Louisiana VA with 150 mph winds, along with Hurricane Laura in 2020 and the Last Island hurricane of 1856. According to the National Oceanic and Atmospheric Administration, 100 percent of homes in Grand Isle, a barrier island about 10 miles from the port, received damage; 40 percent were completely destroyed. There was heavy damage to the energy infrastructure across 115 PREPUBLICATION | UNCORRECTED PROOFS 

southern Louisiana, causing widespread, long- duration power outages that affected millions of people. Parts of New Orleans were without power for more than a week during the hottest months of the year. This timing amplified heat- related risks among an already vulnerable population. The World Meteorological Organization retired the name Ida after this storm (NOAA, 2022a). Hurricane Nicholas 9/12/21 SE TX, S 1B 2 Nicholas made landfall in rural Matagorda LA County, Texas. Power outages (nearly 500,000 in Texas) and freshwater flooding were the primary outcomes in such places as Harris and Galveston counties. More than 100,000 Louisiana residents lost power during flood- producing rainfall, and Florida reported two fatalities due to rip currents (NOAA, 2021j). * The Enhanced Fujita Scale (EF Scale) became operational on February 1, 2007. It is used to assign a tornado a rating based on estimated wind speeds and related damage. When tornado-related damage is surveyed, it is compared with a list of damage indicators, or Dis, and degrees of damage, or DoD, to help estimate the range of wind speeds likely produced by the tornado. From that information, a rating (EF0 to EF5) is assigned (National Weather Service, n.d.). Other notable tropical systems impacting the GOM region in 2020–2021 include Hurricane Eta (in 2020) and the precursor storms associated with Tropical Storms Bertha (in 2020) and Fay (in 2020). Bertha ultimately affected the Carolinas, but flooding (more than a foot of rainfall) in South Florida caused by the evolving system resulted in structural damage in Miami and surrounding regions. A small tornado also affected the region. Tropical Storm Fred (in 2021) was a minor storm that made landfall in August on the western Florida panhandle, causing thousands of power outages. The storm ultimately caused $1.3 billion in damage and seven fatalities as it moved into the Northeast (NOAA, 2021k). The following month, another tropical storm (NOAA, 2021l) made landfall near St. Vincent Island, Florida. That storm 116 PREPUBLICATION | UNCORRECTED PROOFS 

spawned an EF0 tornado south of Tallahassee, Florida, and wind damage caused power outages and tree damage in much of North Florida. Figure 3-2 depicts the tracks of the hurricanes and tropical storms in the GOM region in 2020–2021. FIGURE 3-2 Tropical storm and hurricane tracks in the Gulf of Mexico region, 2020–2021. Source: NOAA, 2021. Tornadoes During 2020–2021, April 2020 was the most tornado-active month in the United States, seeing several multiday events (April 7–8, 12–13, 19, and 22–23; NOAA, 2020). Many of these outbreaks were in the GOM region (see Table 3-4). In 2021, tornado reports exceeded the 1991– 2010 U.S. annual average (NCEI, 2022) with many of these events occurring in the GOM region 117 PREPUBLICATION | UNCORRECTED PROOFS 

(see Table 3-4), including an EF2 tornado to touch down in Lake Charles, Louisiana, in late October 2021, damaging about a dozen homes and injuring several people (Misick, 2021). TABLE 3-4 Tornadic Events in Gulf of Mexico States in 2020–2021 Date(s) GOM State(s) Impacted January 10–11, 2020 Mississippi, Alabama, and Louisiana February 5–7, 2020 Mississippi and Alabama March 18–19, 2020 Texas March 30–31, 2020 Texas, Alabama, Mississippi, and Florida April 12–13, 2020 Texas, Louisiana, Mississippi, Alabama, and Florida April 19–20, 2020 Mississippi, Alabama, and Florida April 22–23, 2020 Texas, Louisiana, Mississippi, and Florida June 6–9, 2020 Florida November 24–25, 2020 Texas and Mississippi November 30, 2020 Florida December 23–24, 2020 Texas January 25–27, 2021 Alabama and Florida February 15, 2021 Florida March 13, 2021 Texas March 16–18, 2021 Mississippi and Alabama March 24–28, 2021 Texas and Alabama April 19–20, 2021 Louisiana, Mississippi, and Florida 118 PREPUBLICATION | UNCORRECTED PROOFS 

April 23–25, 2021 Texas and Alabama April 27–28, 2021 Texas May 2–4, 2021 Mississippi December 29–30, 2021 Alabama Note that there were multiple other wind and/or rain events that occurred in the GOM region in 2020–2021 that are not listed in Tables 3-3 and 3-4 but still caused loss and damages in the region. Extreme Temperatures Parts of the GOM region experienced extreme heat during 2020–2021. The impacts were amplified by episodic power losses associated with thunderstorms and hurricanes. The year 2020 was tied with 2016 for the warmest on record globally, and most of Florida endured record heat that year, even by its own rather warm standards (NOAA, 2021a). An early spring heat wave also broke temperature records as the result of a persistent high-pressure system over the GOM (Cappucci, 2020). Many people in the GOM region were exposed to these record-breaking heat events while enduring COVID-19 and episodic storms. During Hurricanes Delta, Zeta, Laura, and Ida, heat was the most frequent cause of deaths in Louisiana. Of the 65 deaths collectively attributed to the storms, 23 were due to extreme heat (Louisiana Department of Health, 2023). Extreme cold temperatures and winter weather overwhelmed parts of Texas, Louisiana, and Mississippi during a weeklong storm event—Winter Storm Uri—in February 2021. This event is often mischaracterized as a polar vortex storm. The polar vortex is not a storm, but a circulating wind pattern (see Figure 3-3) that is always present above the Arctic region. When it is strong, extremely cold air is “trapped,” but when it weakens (as it did in 2021), extremely cold Arctic air can spread south into the midlatitudes. This sequence of events enabled a strong Arctic cold front to plunge deep into the interior United States, bringing temperatures as much as 40°F (4.44°C) below normal throughout the region. 119 PREPUBLICATION | UNCORRECTED PROOFS 

FIGURE 3-3 The science of the polar vortex. SOURCE: NOAA, 2019. Within the past decade, significant advances have been made in the prediction of extreme weather-climate events (Sillmann et al., 2017). Yet, as is often seen with anomalous weather- climate events, forecasts for Uri provided a sufficient time window to initiate certain preparations. However, the urgency or awareness around preparations was lacking, demonstrating that better preparation is needed at the individual, community, and regional levels for winter weather extremes, even if they are currently rare. 120 PREPUBLICATION | UNCORRECTED PROOFS 

IMPACTS OF COMPOUNDING DISASTERS IN REGIONS OF TEXAS, LOUISIANA, AND ALABAMA To gain an understanding of how compounding disasters affect the lives and livelihoods of individuals who live and work in the GOM region, the Committee on Compounding Disasters in Gulf Coast Communities, 2020–2021, held a series of 13 in-person information-gathering sessions with local, state, and federal representatives from two adjacent counties/parishes in three GOM states: Alabama, Louisiana, and Texas. The primary data elicited from session participants in the information-gathering sessions formed the basis of the report’s summary, analysis, findings, and conclusions. Potential participants were identified through their professional roles within disaster-affected communities, and invitations were extended to a wide range of invitees. In total, 77 individuals participated in the committee’s in-person information-gathering sessions (see Appendix B for a list of panelists). Panelists who agreed to participate represented the subset of individuals who accepted the committee’s invitation and were present for an information- gathering session. The committee is aware that the experiences shared at these sessions are only a small glimpse into the realities that existed for those affected by these events in the GOM region during this 2-year period, a time when everyday stresses were exacerbated by the anxiety and uncertainty brought on by the COVID-19 pandemic and multiple, successive weather- climate events. Recovery Time Lines The information-gathering sessions all began with a participatory activity aimed at documenting the disasters and recovery time lines in each region where that panel took place. The disaster sequences collectively captured in each session in a particular region were aggregated into a single time line, shown in Figure 3-4. Darker gray-shaded regions represent COVID-19 surges based on available case-count data, while yellow-shaded boxes mark the 6- month Atlantic hurricane season, illustrating that both the 2020 and 2021 hurricane seasons were accompanied by surges in COVID-19 cases, as noted earlier in this chapter. These time lines were populated by asking participants to first identify the disruptive events that were most significant for their community. For each of these events, they then collectively defined the 121 PREPUBLICATION | UNCORRECTED PROOFS 

recovery period that followed, represented by the colored (at times overlapping) bars in Figure 3- 4. When characterizing the impacts and recovery time line for each of these events, respondents underscored that even when these events do not directly compound the losses from a preceding event, they continue to have indirect effects by interrupting and extending the ongoing recovery process (e.g., reconstruction and rebuilding of homes postdisaster will likely need to be paused in advance of an approaching storm). FIGURE 3-4 Aggregated recovery time lines for each state visited in the committee’s information- gathering sessions, identifying specific disruptive events (circles) and recovery periods (colored bars) 122 PREPUBLICATION | UNCORRECTED PROOFS 

relative to pandemic surges (dark gray bars) during an ongoing pandemic period (light gray strip) and intermittent hurricane seasons (light yellow strip). Recovery Indicators As part of disaster recovery, many organizations, residents, and state and federal agencies use indicators to measure response and recovery. When defining the recovery time lines in Figure 3-4, the committee asked session participants to identify specific indicators they view as signs that a community has “recovered” from a disaster. The committee found that there is no single indicator or consistent set of indicators that represent when or if a community and its members have fully recovered. The indicators cited most commonly by respondents were categorized based on six sectors that meet the critical needs of a community (see Table 3-5). Specific examples of indicators in each sector are detailed in the subsequent summaries of the information-gathering sessions. TABLE 3-5 Most Frequently Self-Reported Recovery Indicators in Information-Gathering Sessions Systems/Sectors Common Recovery Indicators Utilities (water, power) and services (transportation) restored,* debris removed* Public Services Businesses reopen,* tourism returns, entertainment resumes Local Economy Repairs complete,* population return,* building permits issued, tarps removed, Housing insurance claims paid Health Care Mental health indicators,* services restored, wait times reduced Education Schools reopened* Faith-based institutions reopened* Civil Society *Referenced by a majority of participants in a panel session. Following the definition of recovery indicators and time lines, discussions with participants focused on COVID-19’s impacts on disaster preparation, adaptations and/or changes in preparations for the 2020 and 2021 hurricane seasons, specific impacts of the identified 123 PREPUBLICATION | UNCORRECTED PROOFS 

disasters, “bright spots,” and lessons learned. The following sections summarize these discussions and identify key issues that surfaced across sessions. HARRIS AND GALVESTON COUNTIES, TEXAS On October 10 and 11, 2022, the committee held five 90-minute information-gathering sessions in Houston, Texas (Harris County) with representatives from community-based organizations (CBOs) and local, state, and federal government officials from Texas’s Galveston and Harris Counties. On October 10, the committee held sessions with (1) leaders representing nonprofit and nongovernmental organizations (NGOs) involved in disaster risk management, housing, and community development (Harris NGO Panel); (2) public officials representing floodplain and emergency management, homeland security and public safety, housing, engineering, and public health in Harris County (Harris Officials Panel); and (3) state and federal officials, including a regional Federal Emergency Management Agency (FEMA) coordinator and representatives from the Texas General Land Office and the Small Business Association (Texas Federal/State Officials Panel). On October 11, session participants included (1) Galveston County nonprofit and NGO leaders representing response and recovery, education, financial stability, community emergency response, the Chamber of Commerce, and food security (Galveston NGO Panel); and (2) Galveston County officials representing public health, housing, and emergency management (Galveston Officials Panel). In total, 20 people participated in the committee’s information-gathering sessions in Texas. 2020–2021 Disasters of Greatest Impact In addition to discussing challenges brought on by the global COVID-19 pandemic, panelists from the Galveston and Harris County information-gathering sessions focused on Winter Storm Uri 2 as their primary hazard event. This extreme weather-climate event prompted large-scale power grid failures resulting in the inability to heat homes, burst pipes and water, disruption of internet and cell phone service, severe supply chain delays, and far-reaching 2 For more information, see Donald (2021). 124 PREPUBLICATION | UNCORRECTED PROOFS 

economic hardship for residents and business owners unaccustomed to and thus unprepared for extended freezing temperatures (see Chapter 2 for more detail). Without power and heat, many businesses statewide, which were already hobbled by the pandemic, closed down for several weeks. Panelists noted that community members were experiencing anxiety, malaise, and disaster fatigue at this time. One Galveston NGO leader working in disaster recovery was discussing ongoing disaster recovery projects from as far back as Hurricane Ike, when she stopped and said: “One of the things that we noticed, and you all have probably seen some level of it, a lot of our population, and some of our leadership, are really having disaster fatigue. When we were getting threatened with Nicholas, a lot of people went, you know what? To heck with it. I don’t care. I’m exhausted. I’m exhausted from battling my way through my life in COVID. I’m exhausted from my pipes being broken and trying to get resources to get things fixed. I don’t care if they knock my house down. . . . It’s just been one trauma after another.” Box 3-1 provides more detail on community disaster fatigue. BOX 3-1 Community Disaster Fatigue During 2020–2021, the combination of high-profile, high-impact disruptive events (e.g., weather-climate events and the pandemic) and racial inequality and police brutality contributed to community disaster fatigue, especially among low-income communities and communities of color. Ingham and colleagues (2023, p. 2) define community disaster fatigue as a “collective situation in which the components of a complex system fail to adapt appropriately, leading to inactivity, public health issues, and dysfunctional institutions.” Community disaster fatigue can slow down or stymie much-needed changes in the health, economic, and societal arenas as the focus remains on the disaster recovery process (Ingham et al., 2022). Outcomes of community disaster fatigue appear to be associated, on a collective level, with elements reflecting the breakdown of community resilience (Ingham et al., 2023). With the recurrence of disasters, communities continue to struggle as disaster recovery becomes increasingly prolonged and unattainable. As a result, community resilience and adaptive capacity decline, compassion fatigue sets in, and key leaders and policymakers become complacent without delivering recognizable or meaningful change. Further, the struggles of daily survival create 125 PREPUBLICATION | UNCORRECTED PROOFS 

consistent stressors that impede effective healing in the midst of health, economic, and societal challenges. Recovery Indicators and Timing Hurricane Laura, which made landfall in a parish adjacent to the Texas border, did not cause structural damage in Harris or Galveston County, but triggered traumatic recollections of Harvey’s impacts. It also drew resources toward affected areas in southeast Texas and Louisiana and brought displaced Louisiana evacuees into the state. As a result, at the time of Winter Storm Uri, Texas was still sheltering Louisiana residents from Hurricane Laura, all within the restrictions necessary to manage the COVID-19 pandemic. A secondary hazard event, Hurricane Nicholas, a Category 1 storm that made landfall in mid-September 2021, also affected these areas. The Harris and Galveston NGO panelists stressed that recovery from Uri was taking longer than hurricane recovery typically would, with some people still unable to return to their homes. They informed the committee that recovery from prior hurricanes—Harvey (in 2017) and Ike (in 2008)—and floods—Tax Day (in 2016) and Memorial Day (in 2015)—was also still ongoing and was extended by Uri, particularly with respect to housing. Panelists listed recovery indicators that included repair of burst pipes and backed-up sewer lines, restoration of power, a decline in boil-water notices, the ability to ship donated blood, hospitals being “normal busy” instead of in “catastrophic mode,” and dialysis centers reopening with the ability to admit more patients. Harris County officials noted that disaster recovery is often seen at the community level and can be considered concluded when everyone is “whole” and back to work again. Meanwhile, panelists representing state and federal officials distinguished short-term recovery indicators such as restoration of utilities and resolution of transportation concerns, from long-term recovery indicators, such as finalizing of funding allocations and rebuilding or upgrading of infrastructure in accordance with new codes. One panelist mentioned the large number of homes typically listed in the market after a hurricane, defining recovery as the moment when these more-than-normal for-sale signs no longer appear. Other indicators that panelists identified included the restoration of access to adequate food and 126 PREPUBLICATION | UNCORRECTED PROOFS 

clean water, an increase in department staffing to predisaster levels, availability of housing for displaced populations, and shelters being closed. The timing of recovery from Uri varied by community and neighborhood. Many roads took up to a week to reopen. Regarding power restoration, participants reported that many neighborhoods waited 10–14 days to have electricity restored, though as noted previously, housing recovery had not yet concluded by the time the information-gathering sessions were held. Realized Impacts of 2020–2021 Disasters Winter Storm Uri was a weeklong event that included multiple days of dangerously cold temperatures that affected the GOM region, compounding the severity of power and water outages, disruption of renewable energy production, and transportation stoppages. According to the National Oceanic and Atmospheric Administration (NOAA), nearly 10 million people were without power for extended periods, and studies have confirmed that a disproportionate number of those affected were vulnerable populations (NOAA, 2021o). Depending on the source, estimates of the death toll range from 246 (TDSHS, 2021) to more than 700 (Aldhous et al., 2021), with the Houston Chronicle contending that there were more than 1,000 unexplained deaths within the duration of Uri (Despart, 2022). As one NGO panelist said, “Nobody was prepared for Uri.” The unprecedented scale of the Texas power grid failure made the storm response much more challenging, especially since power was necessary to maintain a safe level of warmth in residences. 3 The committee also heard about the extensive damage to infrastructure. Pipes froze, shifted, and broke, impeding water and sanitation services. The co- occurrence of Winter Storm Uri and the pandemic affected sectors at nearly every level. Hospitals were already overburdened before the power outages, which then further hindered their ability to operate effectively. One response organization resorted to using social media platforms to screen online posts in an effort to find people seeking help, explained one participant. Social 3 As discussed in Chapter 2, the death count for Winter Storm Uri varies depending on the data source. However, the most common cause of death is definitive: exposure to extreme temperatures. According to the Texas Department of State Health Services (2021), 65.4 percent of winter storm–related deaths were caused by extreme cold exposure-related injuries (hypothermia and frostbite). Hanchey et al. (2023) attribute only 47.1 percent of winter storm–related deaths to extreme cold exposure, but they still found that extreme cold exposure was the most common cause of death. 127 PREPUBLICATION | UNCORRECTED PROOFS 

media sites were also used to disseminate real-time information addressing community needs. Compared with Tropical Storm Beta and Hurricane Laura, “we did not see a lot of conversation . . . but when you come to the winter storm, it exploded. And it exploded for a little while, until peoples’ phones lost their charge. And then we saw a decline in what people were posting, because they were sitting shivering in their house with no electricity to charge their phones,” a participant explained. The compounding effects of Uri impaired risk communication in general. Most session panelists noted that their southern tier geographies were unfamiliar with how to prepare for such a deep freeze, and many did not believe that the event’s impacts would be so severe and sustained. Even with ample notice of the storm’s arrival, many GOM communities did not anticipate being immobilized for an extended period of time. One NGO representative from Harris County explained, “Almost everybody has their hurricane season stuff ready. Nobody thinks about the other disasters that could potentially happen during the year. Being prepared for those is just as important.” Box 3-2 provides more information as to cognitive bias challenges embedded in disaster understanding. BOX 3-2 Influence of Cognitive Biases on Risk Perception, Planning, and Communication When people confront the threat of an impending disruptive event or disaster, their decision- making can be impaired by the tendency to overemphasize recent events or personal experiences when estimating the impact of the future event (recency bias) and by the tendency to underestimate the likelihood of the occurrence or impact of the event (normalcy bias). Historical events often serve as benchmarks for how people perceive and plan for future hazards. This tendency can be dangerous. Recent examples of “benchmark” storms include Hurricanes Andrew (in 1992), Ivan (in 2004), Rita and Katrina (in 2005), and Gustav and Ike (in 2008) and the 2011 winter freeze. However, the problem with referring to storms as benchmark is that the bar is always shifting, and the potential for future events to surpass historical events is increasing. Hurricanes Harvey (in 2017) and Ida (in 2021) are examples of storms that render benchmark events obsolete. Accordingly, cognitive biases that inform how people interpret and respond to new threats can result in significant underestimation of risk, leading to inadequate planning and preparation, and ultimately to more severe impacts. The 128 PREPUBLICATION | UNCORRECTED PROOFS 

effects of cognitive bias are not limited to the decision-making of individuals; indeed, they may be reflected in regulations, codes, and policies. According to traditional risk assessment, planners are advised to identify and prioritize risks with the highest probability of occurring locally. This approach can lead to the neglect of atypical and rarer events and leave communities susceptible to an unanticipated, but not unimaginable, event such as Winter Storm Uri. Responses to atypical events such as extreme cold can be outside the purview of organizations accustomed to response and recovery for hurricanes, tornadoes, and flooding, further complicating relief efforts. Those responding have steep learning curves in anticipating and responding to the needs, damages, and other impacts of such an atypical hazard. Compounding events further complicate the traditional risk assessment methodology (Zscheischler et al., 2018). For example, the rainfall intensities experienced now and going forward are not like those of the rainstorms of past centuries or even decades. The literature shows that the top 1–2 percent of rainfall-intensity events have increased in recent decades (Easterling et al., 2017). In just the last 15 years, the percentage of the southern United States experiencing extreme 1-day rainfall events in any individual spring has increased significantly (USGCRP, 2023; Di Liberto, 2021). Relatedly, jurisdictional infrastructure (e.g., roads, bridges) was designed and engineered under an assumption of “stationarity”—that is, assuming that hazards such as rainfall would have future intensities similar to those of rainstorms of decades past. A significant portion of U.S. infrastructure was designed and built in the 20th century. Not only is it aging and often in a state of deferred maintenance, but it now exists in the changing climate of the 21st century. For centuries, European communities have placed physical markers to indicate the height of past floods (Glaser et al., 2010). Such public memorialization is one way to both document past events and provide an enduring reminder of past tragedies. Still, social memory can fade—within both professional circles and the public. Further, human recency and normalcy biases associated with disaster experience are vexing challenges. Programs dedicated to public and agency disaster risk communication require permanent funding, regular updating, community participation, and ongoing implementation. Deficiencies and deficits in risk communication are also linked to vulnerability, particularly as related to a lack of comprehensive information, which can lead to false perceptions (Birkmann and Fernando, 2008) and influence motivation and perceived ability or need to prepare for disruptive events. To address this challenge, the National Hurricane Center (NHC) has been working with social scientists to improve communication products, notably the tropical cyclone “cone of uncertainty” 129 PREPUBLICATION | UNCORRECTED PROOFS 

product. In early 2024, the NHC announced the upcoming release of an experimental tropical cyclone forecast cone graphic for the 2024 hurricane season (NOAA, 2024). It was necessary to get disaster information out to residents, especially that pertaining to preparation for freezing weather, a generally uncommon event in the region. One NGO leader noted, “The emergency broadcast system—it lacks preparedness in letting people know what’s happening. But if a child or something is abducted it flashes on everyone’s phone. But no one was prepared for Uri. Well, that’s because of the grid and everything. But we don’t get those notifications in a timely manner. So, I think that needs to be addressed.” Another state official expressed the need for greater collaboration with houses of worship and other local purveyors of information to improve pathways for trusted and accurate risk communication. Moreover, many people had relocated to Harris County prior to Winter Storm Uri, but many were unfamiliar with emergency protocols, shelter locations, and evacuation routes used in hurricane preparations that would have provided a level of familiarity with general emergency protocols. The effects of Uri caused a tremendous loss of life, explained participants. People did not have heat, fuel was not readily available, and residents resorted to barbecuing in their garages for warmth. Many carbon monoxide detectors did not function properly, and people used outdoor propane heaters to provide heat in their homes. “It was devastating,” summed up a panelist. “People have been traumatized by multiple successive disasters, and when another one strikes, it can feel like PTSD [post-traumatic stress disorder],” a Harris County NGO leader said. Relatedly, participants noted a dramatic rise in domestic violence and socioeconomic impacts, resulting from or exacerbating mental health issues. Panelists from both counties experienced high turnover rates in the health care industry, and new staff were not always familiar with disasters or disaster response. This was but a microcosm of broader trends, with Houston’s population declining for the first time after Uri. Businesses were also not returning at the typical rate, participants said. These ramifications had wide-ranging effects that, at the time of these panel sessions, were still ongoing. 130 PREPUBLICATION | UNCORRECTED PROOFS 

CAMERON AND CALCASIEU PARISHES, LOUISIANA The committee convened four information-gathering sessions in Lake Charles, Louisiana (Calcasieu Parish), on December 7 and 8, 2022, with representatives working and/or living in Louisiana’s Calcasieu and Cameron Parishes and working in the disaster space. The opening session included leaders of CBOs, NGOs, and nonprofit organizations representing general aid and community support services, senior services, faith-based communities, housing rebuild/construction, disability advocacy, and health care (physiological and behavioral) (Louisiana NGO Panel). In the second session, the committee heard from Cameron Parish government officials representing floodplain and emergency management, health care, and parish administration (Cameron Public Officials Panel). The third session consisted of Calcasieu Parish government officials from parish administration, public health, homeland security and emergency preparedness, housing, and behavioral health (Calcasieu Public Officials Panel). The final session included state and federal officials representing FEMA, public health, community development, homeland security and emergency preparedness, mental and behavioral health, and emergency services (Louisiana State/Federal Panel). In total, 30 people participated in the committee’s information-gathering sessions in Louisiana. 2020–2021 Disasters of Greatest Impact In addition to the COVID-19 pandemic, panelists agreed that Hurricane Laura, a Category 4 hurricane that made landfall near Cameron, Louisiana, on August 29, 2020, 15 years to the day after Hurricane Katrina in 2005, was the most significant weather-related event to impact their region in the 2020–2021 time frame. Panelists’ views differed as to the second most significant weather-related event because of the number of disruptive events occurring within a relatively short period. Hurricanes Delta and Zeta both hit in October 2020. Hurricane Delta, which mobilized debris and caused further damage to structures already damaged by Hurricane Laura, complicated recovery activities, as well as FEMA and insurance claims stemming from Hurricane Laura. Hurricane Zeta caused some flooding in Cameron Parish, and its impacts also 131 PREPUBLICATION | UNCORRECTED PROOFS 

siphoned resources to other areas, affecting Calcasieu and Cameron Parish residents still sheltering in New Orleans after Hurricane Laura. Winter Storm Uri in February 2021 brought extreme cold weather with freezing rain, causing pipes to burst and acutely affecting water systems for homes and public buildings, such as hospitals and dialysis centers, that were still damaged from Hurricane Laura. Some residents had yet to return to their homes because of Hurricane Laura’s impacts when Winter Storm Uri hit. Three months after Uri, a heavy May rainstorm brought about a foot of rain in less than 24 hours, flooding and reflooding areas in the region (Di Liberto, 2021) and further hindering ongoing recovery efforts. Panelists also made note of Hurricane Ida (in August 2021), clarifying that its direct impacts were not significant in this region, but that the storm drew resources and some attention away from Southwest Louisiana to the southeastern portion of the state; the added stress also compounded existing mental health challenges. The impacts and respective recovery time lines of each event make it difficult, if not impossible, to differentiate between or ascribe specific impacts to each disaster. Many panelists also drew comparisons between Hurricane Laura and September 2005’s Hurricane Rita. In a discussion about determining when an area can be considered “recovered,” an NGO leader asked the group about recovery from earlier storms: “Did we recover from Rita honestly and thoroughly? . . . There is so much still from Hurricane Rita that is not repaired.” Recovery Indicators and Timing Short-term recovery indicators, as reported by panelists, included the removal of blue tarps from roofs, schools reopening, rebuilding of the workforce (including having enough clinicians to meet the community’s mental health care needs), and the return of residents and the entertainment industry. Long-term recovery indicators included the return and reopening of small businesses and schools, the availability of affordable housing and new construction, access to counseling resources, family stabilization, children achieving educational objectives, local communities meeting their needs without outside support, and the number of substance abuse cases and overdose deaths returning to predisaster levels. Local officials added to the list of recovery indicators the substantial completion of construction projects, utility restoration, repairs 132 PREPUBLICATION | UNCORRECTED PROOFS 

to damaged homes, and resumption of essential services, as well as removal of road debris and blight remediation. They mentioned as well the return to predisaster levels of alcohol sales, home assessment values, school enrollment, domestic violence, and divorce rates. State and federal officials cited additional measures: the ability to be proactive rather than reactive, a majority of building damage repaired, reopening of small businesses, and resumption of social and community activities. These discussions portrayed recovery as a phased process: the first stage concerns achieving stability, said a Louisiana State official, while the second includes incorporating resilience into future planning and actively making improvements. Regarding recovery timing, all four Louisiana panels agreed that, for many residents of Calcasieu and Cameron Parishes, recovery from Hurricane Laura was still ongoing, particularly demolition, repairs, and restoration. Panelists noted that new construction was just starting at the time of the information-gathering session in late 2022. Panelists also agreed that COVID-19 and the series of climate and weather-related events following Hurricane Laura prolonged the overall recovery time line, prompting one panelist to quip, “The term compounding disasters—we’ve got to be the poster child for that.” Having sufficient federal support and cooperation was mentioned as a key factor in the recovery process. Realized Impacts of 2020–2021 Disasters Session participants noted that the COVID-19 pandemic, Hurricane Laura, and the hurricanes and storms that followed affected all sectors of the local economy in Calcasieu and Cameron Parishes. Since the pandemic began, businesses of all kinds—including restaurants, hotels, casinos, and the entertainment industry—had been struggling to stay open. Many small businesses had still not been able to, or had chosen not to, reopen at the time of the panel session. If businesses were operating, most had reduced hours because of understaffing. As one panelist pointed out: “It’s not good enough just to have a plan. You have a plan B, and a plan C, and a plan D.” Hurricane Laura’s effects included widespread damage to public sector–operated utilities, with restoration of electricity taking 5–6 weeks. Session participants recalled how the main broadband provider, responsible for one of the few essential services provided by a private 133 PREPUBLICATION | UNCORRECTED PROOFS 

company, “failed miserably.” One Calcasieu Parish official noted that perhaps the “biggest revelation for us was how deeply we had come to depend on broadband technology. . . . Overnight, businesses, schools, government, individuals [needed broadband] not just for entertainment but for basic services. . . . We realized we were very vulnerable.” Those without access for longer periods struggled to access or learn about disaster resources, and it severely limited students’ ability to attend virtual classes when their schools were damaged. The storm demolished radio towers, eliminating the ability to communicate hurricane response information. The Lake Charles radio tower still is not rebuilt, a panelist pointed out. Delta’s winds caused even more power outages, in part because of trees damaged in Laura, further exacerbating utility recovery. Calcasieu and Cameron Parish residents also struggled to access health care facilities after the storms. Hospitals were damaged and utility outages prevented adequate patient care. However, one of the complicating features of 2020–2021’s compounding disasters was that, in some cases, readily available funding was earmarked for COVID-19 care. Yet, following a hurricane, the most urgent need was to reestablish care in area hospitals that were damaged by the storm. One panelist laid out the scenario, saying, “Almost every dollar that came down to the local health department was related to COVID [and] those dollars were very specific in what they could be used for. And there were bigger threats to people’s health and well-being than COVID during this time period.” (See Chapter 4, Box 4-1, for more information on COVID-19 federal funding.) For example, the only hospital in Cameron Parish was flooded with 2–3 feet of water and remains inoperative as of November 2023 (Maschke, 2023). This situation interfered with all types of hospital-based care and was especially problematic for medically high-risk patients whose medical conditions could not be treated adequately. The education sector lost 6,000–7,000 students because of the 2020–2021 disasters, participants explained. As enrollment dropped, schools were allocated fewer resources or closed, and school budgets were reduced by about $50 million, said one panelist. Multiple local government buildings and staff homes were damaged or worse, destroyed, which further hindered response. Law enforcement personnel faced similar challenges and lacked functional facilities or vehicles that could navigate the flooding and blocked roads. Fifteen of the 18 fire stations in Calcasieu Parish were so damaged that they could no longer store their fire trucks safely. Even funeral homes were damaged to the extent that families 134 PREPUBLICATION | UNCORRECTED PROOFS 

struggled to find facilities to bury or cremate their loved ones. Participants reported that every church in Cameron Parish was damaged, and as of December 2022, only two had reopened. Areas flooded that had never flooded before, in part because of accumulating hurricane debris clogging waterways, explained one participant. At one point, before Delta’s landfall, residents were advised to “secure your debris,” which they had heaped streetside following the instructions after Hurricane Laura. The call, though futile for most, was intended to prevent debris from remobilizing, whether by wind or water, and causing further damage. This instruction, one panelist said, is “the synopsis of compounding disasters.” Delays and limited capacity in debris removal created obstructions to stormwater flow and contributed to flooding in areas that had never flooded before the May 2021 rainstorm. Panelists in Louisiana emphasized the strong sense of solidarity and willingness to help each other during the compounding disasters of 2020–2021. They agreed that the concerted response to the weather-climate disasters reinvigorated a sense of community closeness and care for one another that COVID-19 had taken away. One panelist noted that it was a matter of survival: “Heading into a disaster where isolation was the name of the game, we were not going to survive as a community in isolation.” The trust, rapport, collegiality, and genuine care expressed among individuals representing different governmental and nongovernmental actors was tangible in nearly all of the information-gathering discussions. “It takes a community,” one federal official noted. “We all just needed to love each other,” said an NGO panelist from Southwest Louisiana, “and that really, I think, brought us back as a community . . . because that is at the core of who we are.” BALDWIN AND MOBILE COUNTIES, ALABAMA The committee convened four information-gathering sessions in Mobile, Alabama, (Mobile County) on January 31 and February 1, 2023. The first session included representatives from a national nonprofit whose local office focuses on education, financial stability, and disaster response and recovery; the University of South Alabama; and Mobile-based CBOs with expertise in the areas of community and public health, resilient housing, emergency management, and insurance research (Alabama NGO Panel). The second session comprised city government officials from neighboring Baldwin County from the Offices of Environmental and Grants Management, Public Works, and Building (Baldwin Officials Panel). Mobile County and 135 PREPUBLICATION | UNCORRECTED PROOFS 

city officials who represented county and city environmental services, the Mobile County Health Department, the Mobile County Emergency Management Agency, the public school system, the city’s Neighborhood Development Department, and resiliency efforts based in the mayor’s office (Mobile Officials Panel). The fourth session included representatives from the Alabama Departments of Public Health, Insurance, and Geological Survey, and field officers from FEMA and the U.S. Department of Housing and Urban Development (HUD) (Alabama State/Federal Officials Panel). In total, 27 people participated in the committee’s information-gathering sessions in Alabama. 2020–2021 Disasters of Greatest Impact In addition to the COVID-19 pandemic, panelists agreed that the two main events with the greatest impact on their region during 2020–2021 were Hurricane Sally in September 2020 and Hurricane Zeta in October 2020. Hurricane Sally’s impacts were worst in Baldwin County, while Zeta affected primarily rural communities in Mobile County. Although Sally was a lower- intensity storm, its heavy rains and sustained winds (slow forward speed) led to widespread tree falls, damage to deferred-maintenance buildings, and power outages lasting 2–3 weeks. These effects were then compounded by Zeta’s arrival, interrupting and delaying recovery efforts. Many panelists noted that the compounding effect was predominantly on the organizations and agencies serving frontline communities. Several panelists mentioned that Hurricane Ida and Winter Storm Uri in February 2021 were additional weather-related events that further interrupted and delayed recovery from Hurricanes Sally and Zeta. These counties hosted people who fled storms from neighboring regions, and many of the evacuees were coping with such problems as alcoholism, mental health issues, and severe stress, which placed additional strain on the local Alabama communities and organizations. Panelists also mentioned the ways in which previous hurricanes—Ivan in 2004 and Katrina in 2005—influenced the response and recovery processes for Hurricanes Sally and Zeta. The 15-year break between the earlier damaging hurricanes and those in 2020–2021, panelists commented, caused complacency at nearly every level, including preparation, evacuation, and response. Box 3-3 provides an example of a scenario-building workshop held in preparation for compounding disasters. 136 PREPUBLICATION | UNCORRECTED PROOFS 

BOX 3-3 Scenario-Based Planning for Compounding Disasters During the Deepwater Horizon disaster in 2010, the U.S. Department of the Interior launched a Strategic Science Initiative to develop a method for mobilizing real-time science guidance for decision- makers in the midst of national emergencies. The group developed a scenario-building workshop procedure that envisioned the possibility of sequential events interrupting the recovery process and thrusting communities into a situation that would require a steeper and longer climb to recovery (Machlis and McNutt, 2010). Scenarios carried out by the group considered what would happen if a hurricane blew through the Gulf of Mexico during the response to the oil spill. This type of exercise is crucial to understanding and adequately planning for the co-occurrence of multiple disasters. In 2017, a nonprofit mental health organization in Mobile, Alabama, received a grant from the National Academies of Sciences, Engineering, and Medicine to conduct a tabletop exercise focused on a scenario in which pandemic influenza and a hurricane occurred concurrently. The exercise provided disaster-related trauma and mental health training for community members, mental health professionals, and social service providers, intended to support a train-the-trainer model that could sustain this network over time. According to a session panelist involved in the exercise, at the time, no one believed that such a scenario could occur. When the events of 2020–2021 began to unfold, however, the panelist explained that the organization was better prepared to understand the implications and address the community’s needs more effectively. Recovery Indicators and Timing Panelists listed short-term indicators such as debris removal (which typically is slower in more rural areas of the county), seeing fewer blue tarps on roofs, power restoration, rebuilding of homes or new construction, reengagement of social connections and networks, insurance claims being settled, schools back in session, and shelters closing. Other indicators mentioned included restoration of community functionality; hospitalization numbers returning to baseline; sediment back onshore and dunes replenished; and jobs, power, infrastructure, civic sectors, and population restored. Panelists also remarked that recovery from the events of 2020–2021 is ongoing and will take years. In addition to the public assistance projects underway or already completed, there are more in the queue, and they will take several years to complete. COVID-19 recovery indicators at the public health level included sustainable state staffing models, declining 137 PREPUBLICATION | UNCORRECTED PROOFS 

case counts, and reduced need for ventilators. Long-term health effects will be seen for years, commented one panelist, since so many people were forced to defer screenings and other medical procedures. Generally, this panelist noted, people are sicker now and have worse health outcomes. Realized Impacts of 2020–2021 Disasters The committee referred to the closely successive occurrence of Hurricanes Sally and Zeta, coincident with the pandemic, as a “joint pandemic event”—dubbing the events as “Zally.” Because Hurricanes Sally and Zeta made landfall within a short period, some panelists struggled to separate the events and their effects when trying to recall associated specifics. Panelists discussed the ways in which the joint pandemic event affected the social determinants of health in Mobile and Baldwin Counties. Housing was affected in several ways. Safe, sanitary, and secure housing stock was reduced by storm damage, particularly after Sally’s slow progression. One Mobile County official noted that Sally damaged 57 schools and Zeta, 34, explaining that “when you’ve got that kind of magnitude on the first round and then you’ve got to come back and determine what Sally damaged and Zeta damaged, and [deal with] insurance [and] claim adjusters,” it was infeasible to disentangle which storm did what damage and where. As was the case in Calcasieu and Cameron Parishes, pandemic protocols made it impossible for insurance adjusters and FEMA staff to conduct in-person assessments; instead, they had to use photos provided by property owners to document damages. It is worth noting that opinions from the FEMA representative and some panelists vastly differed as to whether this adaptation for documenting damages ultimately helped or hindered the claims process and affected payout amounts. In rural, bayou areas of Mobile County, many houses flooded, but residents were too overwhelmed by the pandemic-era requirements to apply for assistance online. Some landlords with homes in HUD’s Housing Choice Voucher program moved tenants out to make repairs, but when they realized they could double the rent as housing stock was reduced, they disenrolled from the program, further depleting affordable housing options and forcing families to move to other areas where they could afford housing. Residents with vouchers could relocate but were reluctant to do so because of commuting distances from schools and jobs. 138 PREPUBLICATION | UNCORRECTED PROOFS 

Coinciding with losses in the housing inventory, demand for food stamps surged, and the safety net of social services suffered. Utilities were severely compromised after what one panelist referred to as a “weird anomaly of a storm.” Power outages lasted for 2–3 weeks after Sally. Many medically vulnerable, electricity-dependent residents were disproportionately affected by the power outages and reduced transportation options resulting from downed trees and closed roads. Concern was also expressed that as many as 300 jobs would be lost when federal funds, which currently pay for positions such as nurses, security, and childcare, stopped flowing into the counties. Following hurricanes Sally and Zeta, many critical industries had difficulty finding enough workers, explained session participants. Panelists noted that many workers were resentful that they could not work from home and reassessed whether they wanted to continue to be considered “essential workers” while working in professions such as public health. There was agreement among panelists that a pathway to increasing the number of construction industry workers should include a combination of immigration policy reform and improved pay and benefits. Panelists mentioned, however, that these workforce changes were not always negative for the workers: many residents started their own home-based businesses and were able to realize entrepreneurial successes and gain more family time. Panelists spoke about the strong sense of solidarity and willingness to help each other amid the challenges of a pandemic and other disasters. Many panelists noted that cooperation among different parishes, counties, and state and local actors was well coordinated overall for the 2020–2021 disasters. Many agreed that the disasters brought back a sense of community closeness and caring for one another that COVID-19 had taken away. KEY THEMES FROM THE INFORMATION-GATHERING SESSIONS Although the panelists who participated in the information-gathering sessions experienced different disasters in myriad ways in their respective areas, several recurring themes emerged from their remarks. The following sections summarize overarching messages or recurring topics voiced by panelists throughout the sessions and analyzed by the committee. 139 PREPUBLICATION | UNCORRECTED PROOFS 

COVID-19 Complications in Disaster Preparedness, Response, and Recovery Of course the pandemic impacted us. We have to change the way we do business, everybody did. – Galveston County Public Official [The pandemic was a] very prolonged disaster with tentacles reaching into virtually every aspect of life, whether it’s employment, changes in employment status, telework, changing how schools are being operated, kids having to do school from home. Those are all significant impacts on locations and all kinds of operations. – Harris County Public Official A consistent theme in panelists’ remarks was that the COVID-19 pandemic challenged the GOM region’s historical approach to disaster preparation, response, and recovery, illuminating the emergent understanding of the concept and impact of compounding disasters. In each of the six parishes/counties, routine preparations for hurricane season were implemented and coordinated to the extent possible, yet even the best-laid plans required nearly constant adaptation and improvisation to account for the realities on the ground caused by COVID-19. In preparation for the 2020 hurricane season, state and county/parish officials worked with FEMA to determine strategies for effective and safe non-congregate shelter protocols that would accord with COVID-19 social distancing protocols. Panelists recalled that COVID-19 restrictions prevented staff from conducting annual hurricane training run-throughs that typically serve to help unify staff and familiarize them with procedures. One NGO leader asked the room, “How do you manage a hurricane during a pandemic? You’re not supposed to be near each other. . . . Our volunteers went away because they were scared.” Challenges encountered in the preparation phases continued into the response phase during hurricane season. Evacuation COVID-19 increased residents’ reluctance to evacuate during the 2020 and 2021 hurricane seasons or to seek warming centers during Winter Storm Uri. Even if they were willing 140 PREPUBLICATION | UNCORRECTED PROOFS 

to evacuate, social distancing or the fear of contracting or transmitting COVID-19 prevented them from sheltering with family or friends. Fear of infection, as well as negative past evacuation experiences (e.g., restrictions on reentering Calcasieu Parish shortly after 2005’s Hurricane Rita) prompted lower demand for congregate sheltering, with some households preferring to shelter in place (Collins et al., 2021). Following federal guidelines, state and local authorities pivoted from normal transport plans to reduced capacity for buses and other public vehicles used for evacuation, so that more vehicles were needed to move a fixed number of individuals. Panelists noted particular challenges in responding to medically vulnerable households in light of shortages affecting ambulance and emergency medical services. Officials had to update evacuation procedures to include the provision of personal protective equipment (PPE) to bus drivers, evacuation staff, and evacuated residents while in transit. Non-Congregate Sheltering Hurricane shelters used prior to the pandemic had to be reconfigured and procedures had to be updated to account for COVID-19 precautions. Evacuees arriving at shelters went through medical triage procedures to screen for COVID-19. Those who tested positive were isolated. Panelists mentioned complications regarding the functionality of shelters, including difficulty following COVID-19 social distancing and safety procedures, distinguishing COVID-19 from other illnesses, and isolating individuals who tested positive, as well as the insufficient number of volunteers and designated shelters. Increased square footage was allocated per bed space to adhere to the infection control imperative for achieving social distancing, decreasing shelter capacity in some locations by 75 percent. Even as the demand grew for shelter personnel, due to the need for more shelter locations coupled with the predictable losses of shelter staff to COVID- 19 illness, the pandemic sharply reduced the numbers of volunteer personnel who were ready and willing to serve. In contrast, municipalities that customarily use schools as shelters, such as Mobile County, required minimal adaptation of standard sheltering plans because of the innate compartmentalization of these buildings such that groups and families could be safely isolated in individual classrooms. 141 PREPUBLICATION | UNCORRECTED PROOFS 

To increase shelter capacity, and in light of the sharp decline in tourism, local authorities considered reserving blocks of hotel rooms for hurricane evacuees, yet many hotel rooms were already in use to support unhoused people because of COVID-19. Another option considered was using rooms in inpatient mental health and substance use treatment centers. The use of these options was limited, however, because switching from congregate shelters to a multitude of “hotel room shelters” was expensive. Nonprofit leaders described how they were supplementing public-sector efforts to house evacuees by providing resources for hotel room shelters, but noted that doing so became cost-prohibitive for organizations with limited budgets, especially after subsequent disasters and declining donations during the pandemic. Since capacity at many shelters was limited because of COVID-19 restrictions, the city of Houston provided shelter for unhoused families and individuals at the George R. Brown Convention Center during Winter Storm Uri. Separating people who were ill and determining who had been vaccinated (once vaccinations were available) also proved challenging. One innovation to emerge from Winter Storm Uri was to convert buses, hub houses, and nursing homes into warming centers where people could be safely physically distanced. Panelists discussed the multiple trade-off decisions necessitated by the compounding disasters of the COVID-19 pandemic and weather-climate events. The unprecedented nature of the multiple disasters prompted one official from Lake Charles to describe 2020–2021 as “choosing the ‘least worst’ much of the time. There was no right way to do it. There was no handbook.” In some jurisdictions, testing and vaccination sites were reconfigured for the hurricane season, at times in ways that dealt a blow to public health efforts to slow the spread of COVID-19. Preserving the functionality of area hospitals was a key priority. Attempts were made to keep PPE in reserve, but stock was limited, especially for hospitals that were already contending with inadequate supplies of masks, gowns, and hospital beds solely from the pandemic. The need for PPE for frontline disaster response personnel as well as evacuees in the shelters further strained the supply chain. Furthermore, staff had to be attentive in their personal lifestyle choices outside of work to avoid contracting COVID-19 and taking the whole team down. Vaccine mandates reduced staffing capacities as people unwilling to get vaccinated quit or attempted to change jobs, and offices were shut down because of COVID-19 outbreaks. Some employees were forced to choose between performing their emergency response–related jobs and aiding their own families as they contended with stressors such as emergency home repairs, 142 PREPUBLICATION | UNCORRECTED PROOFS 

inadequate finances, childcare concerns, and challenges related to abrupt school closures and virtual schooling. Trade-offs often resulted in loss of staff, institutional knowledge, and expertise. Inevitably, COVID-19 infected staff, forcing staff teams to miss work and quarantine; leave their positions over fears of coming to work; or at worst, pass away, further reducing workforce capacity. Professionals and volunteers alike became ill with COVID-19, and many were unable to perform their response duties. Some NGOs began to see residents who had once been donors or volunteers, now unemployed and needing support themselves. Rental assistance was provided to relieve some of the economic impacts of COVID-19, but also further burdened program management staff tasked with administering these new relief programs. These are just a few of the many trade-off decisions with which people in the GOM region had to contend during the compounding disasters of 2020–2021. The slow and protracted recovery trajectories, impeded by periodic COVID-19 infection surges, extended and aggravated the stress experienced throughout the region, especially among those on the frontlines. FEMA’s own operations in Louisiana were diminished in scope and reach. Routine procedures, such as conducting damage assessments and supporting those completing applications for individual assistance, were shifted from in-person to remote. As a result, not all Louisiana residents were able to receive FEMA funds for which they were eligible. Robust social networks that play such a critical role during recovery were less able to function in that vital capacity because of COVID-19 restrictions and social distancing. As a key example throughout the GOM region, fewer people were attending and donating to faith-based organizations (FBOs) and in turn, the FBOs were less able to fulfill their roles in recovery. Both immediately and in the long-term, disaster resources and other forms of support are often provided by members of the community and faith- and community-based organizations. While these volunteers and organizations are critical during this time, they may be experiencing consequences of the disaster themselves and struggle to assist others. Overreliance on this often volunteer network risks the exhaustion of critical human resource capacity for effective disaster recovery. 143 PREPUBLICATION | UNCORRECTED PROOFS 

Disproportionate Impacts on Vulnerable Populations It’s always the black and brown communities that have been historically disinvested. So, I think the data shows that we have been underserved, and we are the black and brown communities that are always the worst first, and we are the last ones that get any mitigation or anything like that. That’s why we got together to try to do it ourselves. – Harris County NGO Leader I think there needs to be some way we could help the community, especially in lower socioeconomic situations, help them prepare. – Galveston County Public Official Vulnerable populations, including low-income households, undocumented residents, people of color, seniors, medically high-risk patients, and non-native English speakers, were often among those most affected by the compounding disasters, and a topic discussed at length by panelists in the information-gathering sessions. Panelists talked with insight and sensitivity about subsets of the population struggling because of low-paying jobs, unemployment, housing insecurity, and poverty. They described how, for residents living “day to day” and resource constrained because of the pandemic, the impact of a late-season hurricane perpetuated a “cycle of doom.” Many who sustained hurricane damage had no insurance, unreliable working hours, uncertain pay, and limited eligibility for benefits. Residents without flood or homeowners’ insurance and renters, especially low-income households, faced extraordinary hurdles and lacked the resources to repair their damaged homes. Seniors, many of whom lived in homes in disrepair before the storms, saw their homes suffer even more damage. Following the 2020–2021 compounding disaster damage to housing stocks in Louisiana, state officials warn that there has never been a greater need to identify and create affordable housing for vulnerable populations (Louisiana Office of Community Development, 2024). Panelists described these community members as those most affected by disasters, with one noting, “Every disaster impacts vulnerable populations more. We keep seeing that over and over again.” Another added, “It takes 10–15 years to recover from a named storm . . . and only 144 PREPUBLICATION | UNCORRECTED PROOFS 

for those who have the means to recover. We are just trying to keep the people without means from simply drowning.” Another panelist on a federal/state panel described the key findings from a research study that was underway, examining the stark economic disparities in disaster recovery based on socioeconomic status. The official laid out the study results bluntly saying, “The study spelled it right out, if you’re middle class, lower-middle class, it knocks you down. If you’re lower class there’s no hope, you’re never going to come back out of it after the second disaster. . . That’s something we have to fix.” As a result, one panelist described “the desperation and blank stares. People don’t expect to feel better and we’re seeing that a lot among our vulnerable people. Their expectations for what life should be like are so low.” Panelists in all three states visited by the committee mentioned the inequitable distribution of aid among disaster survivors and vary according to race/ethnicity and homeownership status, a point verified by research (Bowdler and Harris, 2022; Hersher and Kellman, 2021; Muñoz and Tate, 2016; Rumbach, 2023) (see also Recommendations 12 and 13 in NASEM [2023] regarding actions to increase equity in federal grant programs). Research confirms that Black or African American and Hispanic residents were approved for recovery assistance at rates of 13 percent and 28 percent, respectively, compared with White individuals (45 percent), and lower-income renters were 23 percent less likely to receive disaster recovery assistance (Hersher and Kellman, 2021). Panelists in Texas noted how disproportionate impacts were felt at multiple levels during and after Winter Storm Uri. Regarding evacuations, customary hurricane preparations would include evacuating vulnerable populations inland, but with winter storms the size of Uri and icy roads, there was no safe zone. Given its expansiveness and impact across all socioeconomic groups, as well as its novelty, Uri presented a scenario in which even households that would normally have been “resilient” struggled. Although the storm’s impacts were widespread, studies have shown that marginalized groups were still at highest risk because of their elevated sensitivity to those impacts and their reduced adaptive capacity (Hedquist et al., 2023; Nejat et al., 2022; Ritchie et al., 2022). Frontline communities were the most affected and most likely to experience sustained power losses; lack of access to food, water, and health care; broken pipes; and suffering during the event (Lee et al., 2022; Tomko et al., 2023). Power-generating plant operators initiated service interruptions to prevent collapse of the entire power grid when power demands surpassed the available supply. These blackouts occurred more frequently and for 145 PREPUBLICATION | UNCORRECTED PROOFS 

longer periods in low-income and racial/ethnic minority census tracts compared with higher- income, nonminority tracts (Lee et al., 2022). On average, 18–19 percent of the Texas population in predominantly White areas suffered a power interruption compared with 33 percent in the high-minority areas. See Figure 3-5 for the average percentage (approximate mean share) of Texas residents affected by blackouts with different ratios of minority populations. FIGURE 3-5 Average percentage (approximate mean share) of Texas population experiencing blackouts among minority quintiles during the winter storm of February 15–18, 2021. This shows that the share of Texas population experiencing blackouts increased with the share of minorities in a Census Block Group. 4 SOURCE: Shah et al., 2023. Throughout the information-gathering sessions, panelists were asked whether subgroups of medically high-risk patients experienced care disruptions and elevated risks during the disaster events. Almost universally, panel members indicated that these individuals faced special challenges. During the peak of the COVID-19 surges in Alabama, for example, patients with cancer who were undergoing active chemotherapy or radiation treatment were not allowed to have family members or even supportive volunteers with them during their therapy sessions 4 A Census Block Group is a geographical unit used by the United States Census Bureau. It is the smallest geographical unit for which the bureau publishes sample data, that is, data that is only collected from a fraction of all households (U.S. Census Bureau, 2022a). 146 PREPUBLICATION | UNCORRECTED PROOFS 

because of fears of contagion. In the aftermath of hurricanes, residents who needed health care struggled to access it. Hospitals were damaged in the storms, and utility outages prevented adequate patient care. Non-native English speakers, particularly in Bayou La Batre, Alabama, who were faced with significant flooding, water and soil contamination, and power outages, encountered equity, trust, language, and access challenges. “Disasters hit them harder,” an NGO panelist explained, “and the recovery is even longer. It creates bottlenecks since their point of trust may be a single person.” The cycle of doom played out in the unhoused population, which ballooned as the limited allocation of resources was further diminished by COVID-19 and the succession of storms in 2020–2021. Many “people just moved away and did not come back,” explained one NGO leader. In the aftermath, with so much unrepaired real estate, the reduced habitable housing stock led to rising rents that were unaffordable to many seeking housing at a time of economic uncertainty, underemployment, and unemployment. The departure of much of the workforce that was needed to rebuild and reconstruct houses and restore and maintain the local economy effectively served to both lengthen the recovery time and further elevate housing prices. Panelists asserted that it is imperative to ensure from the outset that mechanisms are in place to reach disproportionately affected communities and that recovery programs integrate equitable policies and practices to address the stark disparities summarized above. Disaster Mental Health When Nicholas happened, some people just threw up their hands. They were so exhausted from Harvey, Uri, the pandemic. . . . It’s just been one trauma after another. – Galveston County NGO Leader One of the things that we also need to be mindful of is the trauma and the stress that was put on our responders, early responders. They were responding and 147 PREPUBLICATION | UNCORRECTED PROOFS 

responding, and they’re responding again. A lot of them were responding while they were left in shambles. – Louisiana State Official One of the major themes that emerged from the multiple panels during each of the committee’s information-gathering sessions was the primacy of mental health consequences during the compounding disasters of 2020–2021, but also an appreciation for how resilience was revealed. Panelists from multiple sites provided detailed accounts of these mental health consequences, as well as evidence for regional resilience. Panelists described one of the most egregious inequities in the realm of disaster relief with serious mental health implications—the plight of migrant populations, many of whom are undocumented and hence are disproportionately disadvantaged. Their immigration status makes these individuals ineligible for federal or state assistance, leaving them few options for accessing resources. A Harris County NGO leader stated, “There is a whole population of people whose homes can never get fully repaired, and when the next disaster strikes the grant money will be gone.” Many homes with damage from disasters prior to Winter Storm Uri were further damaged by the deep freeze but were denied coverage or reimbursement from their insurance companies or FEMA after the storm struck. In 2020–2021, many GOM residents were still grappling with protracted and incomplete recoveries from hurricanes that had occurred as long as a decade earlier. Reflecting on disasters prior to 2020–2021, panelists described many residents and community leaders as experiencing the mental health consequences of disaster fatigue. A Galveston County official, for example, stated that “response procedures were changing almost daily as COVID-19 protocols evolved.” Others concurred, mentioning that COVID-19 exacerbated existing disaster fatigue and burnout. Individuals in the community experienced anxiety, trauma, fear, and PTSD due to the multiple successive disasters of 2020–2021. As noted, social networks that play a critical role in mental health and well-being during response and recovery were less able to function in that vital capacity because of COVID-19 restrictions and social distancing. A dramatic rise in domestic violence was also reported by panelists and tied to the devastating socioeconomic impacts of compounding and potentially traumatizing events. In addition, suicide rates increased among the elderly. One public health panelist hypothesized that the stress of the disasters, social isolation, and lack of resources potentially contributed to the deaths of many senior citizens, especially 148 PREPUBLICATION | UNCORRECTED PROOFS 

those on fixed incomes. The prolonged and worsening pandemic also took its toll. By the time the wave of the deadly Delta variant arrived in early October 2021, one panelist recalled that it was “as much a mental health impact as a physical impact. . . . People were exhausted and tired after being on response months and months earlier. . . . So, you kind of throw your hands up in the air . . . you’re almost robotic at that point.” Comments from panelists revealed that mental health issues and chronic and acute stress stemming from exposure to compounding disasters are not being adequately addressed. Panelists also discussed the mental health of students in the storm-ravaged counties. Students’ education was disrupted during the pandemic, with classes shifting to online, remote delivery. COVID-19–induced educational challenges were sometimes punctuated by school closures brought about by structural damage and power outages following hurricanes and other severe weather-climate events. Many youths experienced a combination of negative emotional and educational consequences. In Louisiana, youth mental health surfaced as a major theme, in part because of the irony that students returned to in-person school only 1 day before Hurricane Laura hit and caused extensive damage to every educational facility. The storm damaged internet connectivity, severely limiting students’ ability to attend virtual classes. Prevalent anxiety symptoms and pervasive social isolation characterized the students’ experiences. Meanwhile, for parents with frontline response roles, an additional complication was balancing their work demands with managing their children’s roller-coaster educational progress. The ripple effects of multiple disaster impacts throughout the GOM region, exacerbated by COVID-19, depleted the resources and personnel needed for recovery for all affected communities. In turn, the slow and protracted recovery efforts, hobbled by periodic COVID-19 stress, extended and aggravated the stress experienced throughout the region, especially for those charged with response duties. This piling on of pandemic and disaster stressors took a toll. One panel participant stated, “It paralyzed some [workers] as these issues compounded.” Panelists expanded upon how the mental health of the frontline responders was progressively “exhausted,” leaving many of them “burnt out and struggling emotionally.” It was difficult for many frontline workers to ask for help. Many did not know whether—and what—resources might be available to provide support. A panelist highlighted the extreme stress and strain experienced by first responders whose work demanded unrelenting exposures to potentially traumatizing events. All types of frontline 149 PREPUBLICATION | UNCORRECTED PROOFS 

personnel were overwhelmed. One panel member commented, “From a health department perspective, this period of events will break anybody.” When prompted, some panelists were able to identify positive outcomes and evidence for community resilience, citing a strong sense of cohesion and readiness to assist one another. Panelists agreed that the concerted response to the weather disasters reinvigorated a sense of community closeness and care for one another that COVID-19 had taken away. Taken together, the information-gathering discussions in all three states revealed the primacy of mental health consequences for pandemic and storm survivors, the heightened mental health risks for frontline responders, the disproportionate risks for marginalized communities, and the amplification of mental health risks for populations confronted by compounding disaster events occurring simultaneously or in quick succession. SUMMARY OF KEY FINDINGS The weather-climate events and the COVID-19 pandemic exceeded historical benchmarks and fundamentally challenged conventional approaches to disaster preparation, response, and recovery. Throughout 2020–2021, the emergence and spread of COVID-19 transformed the public health risk of compounding disasters in the GOM region by amplifying health-compromising exposures and underlying vulnerabilities while modifying emergency procedures for weather-climate events. With the onset and aftermath of each successive disruptive event, the effects within communities compounded and recovery processes became prolonged. As a result, community resilience and adaptive capacity declined, compassion fatigue set in, and key leaders struggled against complacency in their inability to deliver effective recovery assistance. Socially vulnerable populations, particularly communities of color, those of low socioeconomic status, and non-native English speakers, bore the brunt of compounding disasters. Generally, GOM communities were not prepared to respond or recover from the occurrence of disasters that overlapped in both time and space. The impacts and respective recovery time lines of each event made it difficult, if not impossible, for affected community members to differentiate between or ascribe specific impacts to each disaster. Beyond the immediate damage and loss, chronic and acute stress resulted in widespread mental health 150 PREPUBLICATION | UNCORRECTED PROOFS 

challenges among community members and responders. These effects are not adequately understood or addressed, potentially resulting in the diminishment of adaptive capacity. 151 PREPUBLICATION | UNCORRECTED PROOFS