Skip to main content
PMC home page
GeoHealth logoLink to GeoHealth
. 2019 May 29;3(5):122–126. doi: 10.1029/2019GH000197

Extreme Weather, Chemical Facilities, and Vulnerable Communities in the U.S. Gulf Coast: A Disastrous Combination

Susan C Anenberg 1,✉, Casey Kalman 1
PMCID: PMC7038901  PMID: 32159036

Abstract

Many chemical facilities are located in low‐lying coastal areas and vulnerable to damage from hurricanes, flooding, and erosion, which are increasing with climate change. Extreme weather can trigger industrial disasters, including explosions, fires, and major chemical releases, as well as chronic chemical leakage into air, water, and soil. We identified 872 highly hazardous chemical facilities within 50 miles of the hurricane‐prone U.S. Gulf Coast. Approximately 4,374,000 people, 1,717 schools, and 98 medical facilities were within 1.5 miles of these facilities. Public health risks from colocated extreme weather, chemical facilities, and vulnerable populations are potentially disastrous and growing under climate change.

Keywords: climate change, extreme weather, chemical safety, public health

Key Points

  • Natural hazards can trigger explosions, fires, and chemical releases at industrial facilities, endangering public health and safety

  • Approximately 4,374,000 people live within 1.5 miles of 872 highly hazardous chemical facilities along the U.S. Gulf Coast

  • Public health risks from colocated chemical facilities, people, and natural hazards are understudied, yet potentially large and growing

Without adequate preventative measures, industrial disasters can be triggered by natural hazards, such as hurricanes, floods, lightning, and earthquakes (Krausmann et al., 2011; Salzano et al., 2013). A recent example occurred during Hurricane Harvey, when over three feet of flooding disabled the refrigeration system at the Arkema plant in Crosby, TX, causing organic peroxides to spontaneously combust. As a result, 21 people sought medical attention and 200 people within 1.5 miles of the facility evacuated and could not return home for a week (U.S. Chemical Safety and Hazard Investigation Board, 2018). Like the Arkema Crosby facility, which was located within the 100‐ and 500‐year flood zones, many chemical facilities in the United States are located in low‐lying coastal areas, leaving them vulnerable to damage triggered by hurricanes, flooding, and erosion, which are increasing as a result of climate change‐related sea level rise (Cruz & Krausmann, 2013; Kopp et al., 2014; Wuebbles et al., 2014). Texas in particular is projected to lead the nation in climate change‐related flood damage (U.S. Environmental Protection Agency, 2015) and has many chemical facilities surrounded by vulnerable communities (Tinney et al., 2016; U.S. Chemical Safety and Hazard Investigation Board, 2016).

Extreme weather, such as hurricanes and accompanied storm surge flooding, can have direct and indirect effects on public health, safety, and welfare. For example, Hurricane Maria led to nearly 3,000 fatalities in Puerto Rico, though initial government reports had estimated only 64 deaths (Santos‐Burgoa et al., 2018). Downstream effects of extreme weather can include storm surge flooding, power outages, reduced capacity of healthcare facilities, and road closures limiting access to medical care. Natural hazards can also trigger industrial disasters from chemical explosions, fires, and releases, as exemplified by the Arkema explosions after Hurricane Harvey (Cozzani et al., 2010; Nascimento & Alencar, 2016; World Health Organization, 2018). These natural hazard‐triggered technical disasters, often referred to as “natech” events, are the subject of a growing area of research. Natural hazards can trigger industrial disasters in various ways, including by washing out chemicals into floodwaters, damaging storage vessels and pipes containing chemicals, lightning strikes igniting flammable materials, and damage to power supply that can upset processes or affect temperature and pressure of stored chemicals. As natural hazards often also damage the capacity, effectiveness, and functioning of public services, the health consequences of natech events may be particularly difficult to manage during the event itself. A preventative approach that avoids vulnerability to natech events would be more effective, especially given potential increases in natural hazard events as the climate changes.

Explosions, fires, and chemical releases occur at industrial facilities frequently, subjecting the communities surrounding these facilities to public health and safety hazards including injury from debris, inhalation or dermal exposure to chemicals and smoke, and burns. We obtained access to the U.S. Chemical Safety Board's “incident screening database,” which logs fires, explosions, and chemical releases at industrial facilities in the United States, via a Freedom of Information Act request in October 2018. From 2001 to 10 October 2018, the database captured 9,406 incidents, an average of ~1.5 incidents per day. Of these, 978 incidents (10%) occurred in Texas, and 1,890 (20%) occurred in the five Gulf Coast states (from most to least: Texas, Florida, Louisiana, Mississippi, and Alabama). Incidents involved chemicals, food, metal, petroleum (refineries), and other types of facilities (Gomez et al., 2008). Approximately 40% of industrial incidents in the database led to injury or fatality among workers and/or the public.

The number of these types of industrial disasters that were triggered by natural hazards in the United States and globally is not well understood. One estimate indicates that 2–5% of incidents resulting in the release of hazardous substances were triggered by natural hazard events (World Health Organization, 2018). However, other studies suggest a higher frequency of natech events. A 2008 study estimated that 500–800 natech events occur throughout the United States each year (Steinberg et al., 2008). More recently, studies have estimated that up to 450 incidents annually involving on‐shore hazardous liquid pipeline systems are triggered by natural hazards (Girgin & Krausmann, 2016), over 600 hazardous material releases from gas installations and offshore oil facilities and pipelines were triggered by Hurricanes Rita and Katrina (Cruz & Krausmann, 2009), and 16,600 hazardous material spills caused by natural hazards occurred from 1990–2008 in the United States (20% of which were due to hurricanes; Sengul et al., 2012). Floods were the second largest contributing cause of natechs in the EU in 2010, behind only lightning, and ahead of low temperature, rain, storm/wind, landslide, heat, and earthquake (Krausmann, 2010) These estimates vary but consistently indicate that natech events, including those triggered by floods, are relatively common. Without adequate mitigation measures, the frequency and impacts of these types of events may be exacerbated in the future as industrialization, urbanization, and hydrometeorological hazards associated with climate change increase simultaneously (Bernier et al., 2017; Piatyszek et al., 2017; World Health Organization, 2018).

Public health impacts from these natural hazard‐triggered industrial disasters range from acute explosions, fires, and large chemical releases to longer‐term, more chronic exposure to chemicals leaking slowly into the air, water, and soil. Facility damage may also cause extended shutdowns or closures, often leading to job losses and economic damages. Thus, the potential impacts of these types of events can be disastrous for individuals, families, and communities. Following Hurricane Harvey, the New York Times found that 1,400 chemical sites across the United States are in areas at highest risk of flooding (Tabuchi et al., 2018, February 6). However, there is limited public information available about where the facilities are, the chemicals stored there, the number of people and community buildings nearby, and the degree of changing risks from extreme weather.

To help understand the magnitude of the potential risk to public health and safety posed by highly hazardous chemical facilities in locations vulnerable to natural hazards, we calculated the number of people, schools, and hospitals near such facilities throughout the U.S. Gulf Coast. We used locations of facilities registered in the U.S. Environmental Protection Agency's (EPA) Toxic Release Inventory from the EPA's Facility Registration System Database and filtered for facilities within 50 miles of the coast to capture those potentially most exposed to flood hazards. We defined “highly hazardous chemical facilities” as those with EPA Risk Screening Environmental Indicator (RSEI) scores ≥ the median score of 415 (excluding facilities with a score of zero). RSEI scores reflect “risk‐related impacts on chronic human health” and account for the magnitude and toxicity of stored chemicals and the population living nearby. We then calculated the number of people, schools, and hospitals within a 1.5‐mile buffer around each facility (the size of the Arkema evacuation zone; Figure 1). We used census block group population from 2016 from the American Community Survey and locations of medical and educational facilities from the U.S. Geological Survey's National Structures Dataset (Manson et al., 2018; U.S. Geological Survey, 2018).

Figure 1.

Figure 1

Open in a new tab

Illustration of the analytical steps taken in ArcGIS to determine population, educational facilities, and medical facilities within 1.5 miles of a highly hazardous chemical facility: (1) Draw 1.5‐mile buffer; (2) divide buffers by census block boundaries; (3) dissolve buffers by GeoID to eliminate overlap; and (4) calculate percentage of each census track located within the buffer region. This percentage was then multiplied by the population of the census block, assuming the population was distributed evenly throughout the block group. We then performed a spatial join between the dissolved buffer layer and point locations of medical and education facilities. RSEI = Risk Screening Environmental Indicator.

We identified 2,545 facilities located within 50 miles of the coast throughout Florida, Mississippi, Alabama, Louisiana, and Texas (Figure 2). Of these, 872 facilities had a RSEI score ≥ 415; for comparison, the Arkema Crosby plant had a score of 16. Approximately 14% (4,374,000 people) of the coastal population of these states, 1,717 educational facilities, and 98 medical facilities are located within 1.5 miles of these facilities. Half (50%) of the highly hazardous chemical facilities in this region were located in Texas, 23% in Florida, 19% in Louisiana, 5% in Alabama, and 3% in Mississippi (Table 1). The size of the nearby vulnerable populations in each of these states followed a similar pattern, with the largest number of nearby people, educational facilities, and medical facilities in Texas, followed by Florida and Louisiana. This preliminary analysis indicates that there is a sizable population that would be vulnerable should fires, explosions, or chemical releases occur at nearby facilities. These communities include schools and hospitals, indicating that the surrounding population also include particularly susceptible subgroups. Further, using worst‐case scenario predictions from the National Oceanic and Atmospheric Administration's Storm Surge Maximum of the Maximum data set (National Oceanic and Atmospheric Administration National Hurricane Center, 2019), massive flood events could occur following category five hurricanes, with 63 of these facilities potentially experiencing >20 feet of flooding. Given rising storm surge levels with climate change, this is a public health risk that is likely to grow in the future.

Figure 2.

Figure 2

Open in a new tab

Locations of highly hazardous chemical facilities (Risk Screening Environmental Indicator score ≥ 415) within 50 miles of the U.S. Gulf Coast overlaid on census block group population size for 2016.

Table 1.

For Each State Along the U.S. Gulf Coast, the Number of Highly Hazardous Chemical Facilities Within 50 Miles of the Coast and the Number of People and Percent of Coastal Populations in 2016 (Rounded to Hundreds), Number of Educational Facilities, and Number of Medical Facilities Within 1.5‐Mile Buffer Region Surrounding Those Coastal Highly Hazardous Chemical Facilities

Category Subcategory Texas Florida Louisiana Alabama Mississippi All five states
Highly hazardous chemical facilities No. facilities within 50 miles of coast 949 1,045 388 106 57 2,545
No. coastal facilities with RSEI ≥415 432 200 166 45 29 872
Percent of state coastal facilities with RSEI ≥415 46% 19% 43% 42% 51% 34%
Population Population within 50 miles of coast 8,468,000 19,492,000 3,187,200 693,700 511,900 32,352,400
Coastal population within 1.5 mile buffer region 2,181,000 1,626,000 417,000 84,600 64,500 4,373,800
Percent of state coastal population that is within 1.5 mile buffer region 26% 8% 13% 12% 13% 14%
Population density within 1.5 mile buffer region of individual highly hazardous chemical facilities in people per mi2 [mean (min‐max)] 1,483 (3–9,656) 1,619 (6–6,865) 539 (5–4,495) 444 (16–2,361) 528 (15–1,551) 1,250 (3–9,656)
Educational facilities No. within 50 miles of coast 2,839 5,804 1,382 287 224 10,536
No. within 1.5 mile buffer region 731 672 217 57 40 1,717
Percent of state coastal facilities that are within 1.5 mile buffer region 26% 12% 16% 20% 18% 17%
Medical facilities No. within 50 miles of coast 160 285 119 17 15 596
No. within 1.5 mile buffer region 43 27 21 3 4 98
Percent of state coastal facilities that are within 1.5 mile buffer region 27% 9% 18% 18% 27% 16%
Open in a new tab

Note. RSEI = Risk Screening Environmental Indicator.

Awareness of natural hazard‐triggered industrial events has been growing over the last decade (Cruz et al., 2006; Nascimento & Alencar, 2016; Steinberg et al., 2008), particularly in Europe (Nascimento & Alencar, 2016). In the United States, the industry is already aware of these risks to their operations and has pursued a $12 billion, 60‐mile sea wall along the Gulf coast to protect refineries from sea level rise (Weissert, 2018, August 22). However, natech events in the United States remain poorly understood among the public and infrequently studied compared with other environmental health issues, such as air pollution and water quality. They are also not adequately regulated at the federal, state, or local level. The Chemical Safety Board's investigation of the Arkema incident found that the Occupational Safety and Health Administration and EPA consider flooding and extreme weather hazards to be part of a normal Process Hazard Analysis under Occupational Safety and Health Administration's Process Safety Management standard and EPA's Risk Management Program but that neither agency explicitly requires such extreme events to be addressed (U.S. Chemical Safety and Hazard Investigation Board, 2018), a regulatory gap that puts public health and safety at risk.

We focused this article on acute disasters, such as fires, explosions, and major chemical releases, that could endanger public health and safety. Slower, more chronic leakage of chemicals into floodwaters, air, and soil is also a relatively unexplored, and perhaps more pervasive, public health risk. Colocation of extreme weather, chemical facilities, and vulnerable communities is a potentially disastrous combination that is likely to worsen with climate change. When the next hurricane hits, will we be prepared?

Data Availability

All data used in this analysis are publicly available from the referenced sources.

Conflict of Interest

The authors declare no actual or perceived conflicts of interests.

Acknowledgments

Support was provided by the George Washington University Milken Institute School of Public Health. We appreciate helpful discussions with Kristen Kulinowski, David Michaels, Jim Neumann, and Veronica Southerland.

Anenberg, S. C. , & Kalman, C. (2019). Extreme weather, chemical facilities, and vulnerable communities in the U.S. Gulf Coast: A disastrous combination. GeoHealth, 3, 122–126. 10.1029/2019GH000197

References

  1. Bernier, C. , Elliott, J. R. , Padgett, J. E. , Kellerman, F. , & Bedient, P. B. (2017). Evolution of social vulnerability and risks of chemical spills during storm surge along the Houston Ship Channel. Natural Hazards Review, 18, 04017013(4). 10.1061/(ASCE)NH.1527-6996.0000252 [DOI] [Google Scholar]
  2. Cozzani, V. , Campedel, M. , Renni, E. , & Krausmann, E. (2010). Industrial accidents triggered by flood events: Analysis of past accidents. Journal of Hazardous Materials, 175(1‐3), 501–509. 10.1016/j.jhazmat.2009.10.033 [DOI] [PubMed] [Google Scholar]
  3. Cruz, A. M. , & Krausmann, E. (2009). Hazardous‐materials releases from offshore oil and gas facilities and emergency response following Hurricanes Katrina and Rita. Journal of Loss Prevention in the Process Industries, 22(1), 59–65. 10.1016/j.jlp.2008.08.007 [DOI] [Google Scholar]
  4. Cruz, A. M. , & Krausmann, E. (2013). Vulnerability of the oil and gas sector to climate change and extreme weather events. Climatic Change, 121(1), 41–53. 10.1007/s10584-013-0891-4 [DOI] [Google Scholar]
  5. Cruz, A. M. , Steinberg, L. J. , & Vetere‐Arellano, A. L. (2006). Emerging issues for natech disaster risk management in Europe. Journal of Risk Research, 9(5), 483–501. 10.1080/13669870600717657 [DOI] [Google Scholar]
  6. Girgin, S. , & Krausmann, E. (2016). Historical analysis of U.S. onshore hazardous liquid pipeline accidents triggered by natural hazards. Journal of Loss Prevention in the Process Industries, 40, 578–590. 10.1016/j.jlp.2016.02.008 [DOI] [Google Scholar]
  7. Gomez, M. R. , Casper, S. , & Smith, E. A. (2008). The CSB incident screening database: Description, summary statistics and uses. Journal of Hazardous Materials, 159(1), 119–129. 10.1016/j.jhazmat.2007.07.122 [DOI] [PubMed] [Google Scholar]
  8. Kopp, R. E. , Horton, R. M. , Little, C. M. , Mitrovica, J. X. , Oppenheimer, M. , Rasmussen, D. J. , Strauss, B. H. , & Tebaldi, C. (2014). Probabilistic 21st and 22nd century sea‐level projections at a global network of tide‐gauge sites. Earth's Future, 2, 383–406. 10.1002/2014EF000239 [DOI] [Google Scholar]
  9. Krausmann, E. (2010). Analysis of Natech risk reduction in EU Member States using a questionnaire survey.  Retrieved from (23 April 2019) http://publications.jrc.ec.europa.eu/repository/bitstream/JRC61931/jrc_eu_natech.pdf [Google Scholar]
  10. Krausmann, E. , Renni, E. , Campedel, M. , & Cozzani, V. (2011). Industrial accidents triggered by earthquakes, floods and lightning: Lessons learned from a database analysis. Natural Hazards, 59(1), 285–300. 10.1007/s11069-011-9754-3 [DOI] [Google Scholar]
  11. Manson, S. , Schroeder, J. , Van Riper, D. , & Ruggles, S. (2018). PUMS National Historical Geographic Information System: Version 13.0 [Database]. Minneapolis: University of Minnesota. 10.18128/D050.V13.0 [DOI] [Google Scholar]
  12. Nascimento, K. R. D. S. , & Alencar, M. H. (2016). Management of risks in natural disasters: A systematic review of the literature on NATECH events. Journal of Loss Prevention in the Process Industries, 44, 347–359. 10.1016/j.jlp.2016.10.003 [DOI] [Google Scholar]
  13. National Oceanic and Atmospheric Administration National Hurricane Center (2019). Storm surge maximum of maximum (MOM). Retrieved from (19 March 2019) https://www.nhc.noaa.gov/surge/momOverview.php
  14. Piatyszek, E. , Tardy, A. , Lesbats, M. , & Cruz, A. M. (2017). Natech events triggered by floods: When floods cause technological accidents In Floods (pp. 73–87). Oxford, UK: Elsevier; 10.1016/B978-1-78548-268-7.50005-5 [DOI] [Google Scholar]
  15. Salzano, E. , Basco, A. , Busini, V. , Cozzani, V. , Marzo, E. , Rota, R. , & Spadoni, G. (2013). Public awareness promoting new or emerging risks: Industrial accidents triggered by natural hazards (NaTech). Journal of Risk Research, 16(3‐4), 469–485. 10.1080/13669877.2012.729529 [DOI] [Google Scholar]
  16. Santos‐Burgoa, C. , Sandberg, J. , Suárez, E. , Goldman‐Hawes, A. , Zeger, S. , Garcia‐Meza, A. , Pérez, C. M. , Estrada‐Merly, N. , Colón‐Ramos, U. , Nazario, C. M. , Andrade, E. , Roess, A. , & Goldman, L. (2018). Differential and persistent risk of excess mortality from Hurricane Maria in Puerto Rico: A time‐series analysis. The Lancet Planetary Health, 2(11), e478–e488. 10.1016/S2542-5196(18)30209-2 [DOI] [PubMed] [Google Scholar]
  17. Sengul, H. , Santella, N. , Steinberg, L. J. , & Cruz, A. M. (2012). Analysis of hazardous material releases due to natural hazards in the United States. Disasters, 36(4), 723–743. 10.1111/j.1467-7717.2012.01272.x [DOI] [PubMed] [Google Scholar]
  18. Steinberg, L. J. , Sengul, H. , & Cruz, A. M. (2008). Natech risk and management: An assessment of the state of the art. Natural Hazards, 46(2), 143–152. 10.1007/s11069-007-9205-3 [DOI] [Google Scholar]
  19. Tabuchi, H. , Popovich, N. , Migliozzi, B. & Lehren, A. (2018, February 6). Floods are getting worse, and 2,500 chemical sites lie in the water's path. New York Times. Retrieved from (23 April 2019) https://www.nytimes.com/interactive/2018/02/06/climate/flood-toxic-chemicals.html
  20. Tinney, V. A. , Denton, J. M. , Sciallo‐Tyler, L. , & Paulson, J. A. (2016). School siting near industrial chemical facilities: Findings from the U.S. Chemical Safety Board's Investigation of the West Fertilizer explosion. Environmental Health Perspectives, 124(10), 1493–1496. 10.1289/EHP132 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. U.S. Chemical Safety and Hazard Investigation Board (2016). Investigation Report: West Fertilizer company fire and explosion. Retrieved from (23 April 2019) https://www.csb.gov/file.aspx?DocumentId=5983 [Google Scholar]
  22. U.S. Chemical Safety and Hazard Investigation Board (2018). Investigation report: Organic peroxide decomposition, release, and fire at Arkema Crosby following Hurricane Harvey flooding. Retrieved from (23 April 2019) https://www.csb.gov/file.aspx?DocumentId=6068 [Google Scholar]
  23. U.S. Environmental Protection Agency (2015). Climate change in the United States: Benefits of global action. Retrieved from (23 April, 2019). https://www.epa.gov/cira/downloads-cira-report [Google Scholar]
  24. U.S. Geological Survey (2018). USGS National Structures Dataset. data.gov. Retrieved from (14 December 2018). [Google Scholar]
  25. Weissert, W. (2018, August 22). Oil industry wants government to build seawall to protect refineries from climate change effects. The Oregonian. Retrieved from (23 April 2019) https://expo.oregonlive.com/news/erry-2018/08/88ce31f2fa4310/oil-industry-wants-government.html
  26. World Health Organization (2018). Chemical releases caused by natural hazard events and disasters: Information for public health authorities. Retrieved from (23 April 2019) http://apps.who.int/iris/bitstream/handle/10665/272390/9789241513395-eng.pdf?ua=1 [Google Scholar]
  27. Wuebbles, D. , Meehl, G. , Hayhoe, K. , Karl, T. R. , Kunkel, K. , Santer, B. , Wehner, M. , Colle, B. , Fischer, E. M. , Fu, R. , Goodman, A. , Janssen, E. , Kharin, V. , Lee, H. , Li, W. , Long, L. N. , Olsen, S. C. , Pan, Z. , Seth, A. , Sheffield, J. , & Sun, L. (2014). CMIP5 Climate Model Analyses: Climate extremes in the United States. Bulletin of the American Meteorological Society, 95(4), 571–583. 10.1175/BAMS-D-12-00172.1 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

All data used in this analysis are publicly available from the referenced sources.

Articles from GeoHealth are provided here courtesy of Wiley

RESOURCES