Juan Flores and his family live in Galena Park, Texas, which is bordered on three sides by pipeline terminals, oil refineries, fertilizer plants and rail yards.
Flores has lived in the town of about 11,000 people just east of downtown Houston since he was 4 years old. For a while, he even served on the City Council.
After all these years, he is accustomed to the rhythms of life among the industrial plants. Strange smells and occasional warnings to shelter in place don't bother him too much. "I live so close to [one] company that I can hear their alarms," he says. "The thing is, you hear it so much you get immune to it, and it's like background noise."
But there are also times when he takes notice. "If I smell something out here, it's bad," he says, "and I can tell you during Harvey, it smelled real bad."
Hurricane Harvey caused industrial facilities in Texas to release an extra 5.98 million pounds of pollution into the air, according to the most recent analysis by the Environmental Defense Fund: the pollutants benzene and toluene, both carcinogens, as well as a brew of other chemicals that can irritate eyes and exacerbate respiratory problems.
But for days after the flooding began, the residents of Galena Park and other neighboring communities had little or no information about the air they were breathing. Air monitors scattered across the region were taken out of service, to protect them from storm damage, officials say.
The Environmental Protection Agency, university teams and environmental groups did some air testing while the monitors were down, but the limited effort produced far less information than the permanent air monitoring network would have.
The EPA opted to wait until the majority of the monitors were back up and running — about a week — to begin releasing statements to the public about air quality.
"We released our data as it became available," says acting EPA Regional Administrator Sam Coleman. "We were working with both the state and others to make sure we could access all areas safely and make sure our data collection would be done in a consistent and statistically valid manner. So it took a little bit of time to make sure all those details were worked out."
All the while, Flores says, the smell of gasoline was so strong in Galena Park that his eyes were watering. Flores does community outreach for the Houston air quality nonprofit AirAlliance, so he knew enough about pollution to be concerned. He turned off his air conditioner, trying to keep to keep the noxious air out of his house, and especially away from his toddler.
"Some people left Galena Park," he remembers. "They're like, 'Man, I can't take this.' "
On Sept. 3, eight days after Harvey hit Houston, the EPA put out a press release. The section on air quality read:
"One of the many preparations for Hurricane Harvey included EPA, TCEQ, and other monitoring entities temporarily removing approximately 75 percent of the stationary air monitoring equipment from the greater Houston, Corpus Christi, and Beaumont areas. Since then, state and local authorities are working to get the systems up and running again.
"As of Saturday, September 2, over 70 percent of the monitors are up and working again; and authorities expect that the network will be fully operational again by next week. Of the available air monitoring data collected from August 24-September 2, 2017, all measured concentrations were well below levels of health concern. Monitors are showing that air quality at this time is not concerning and local residents should not be concerned about air quality issues related to the effects of the storm."
The statement simultaneously acknowledged the lack of information about what was in the air and reassured the public that the air was safe.
Chris Sellers, who studies environmental history and the EPA at Stony Brook University, says this type of statement is common. "That is a tried and true [reaction], not just for environmental agencies but for public health agencies in general," he explains. "They feel like they cannot admit uncertainty about dangers, particularly in the face of public panic."
A toxic brew
Sellers says the urge to reassure the public that the air is safe can be dangerous. For example, after the World Trade Center attacks on Sept. 11, 2001, public health officials, including the EPA, reassured people that the air around the collapsed buildings in Manhattan was safe, despite the fact that there was very little information about air quality in the area.
Years later, long-term health studies found thousands of workers had gotten sick from the air. New York City eventually reached a multimillion-dollar settlement with 10,000 first responders and others who were exposed to dust and smoke.
A 2003 report from the EPA inspector general found the agency's response to Sept. 11 had not been based on data, and it called on the EPA to work more closely with state and local health and environmental officials, as well as other federal agencies, to prevent misleading public safety announcements after disasters.
Coleman says that collaboration did happen during Harvey. For example, all press releases after the storm were jointly released by the EPA and the Texas Commission on Environmental Quality.
"I think this response was an excellent example of cooperation at all levels of government," says Coleman, who previously worked on the EPA's emergency response to Hurricane Katrina. "After Katrina, we found that there were often some circumstances where EPA and the state could not communicate data quickly and accurately, and we feel like we corrected that in this response by getting the information out in just a matter of days after it was collected."
But Sellers says collaboration hasn't fixed the underlying problem. He says the EPA isn't set up to handle chemical disasters, especially ones that involve the release of a mixture of chemicals across large areas.
"Most of the EPA's infrastructure for monitoring is based on low-level chemical exposures, and exposures you can kind of single out," he says, measuring each individual chemical in the air, and analyzing the levels over time. "When you have a whole sort of brew of chemicals that's thrust into the air — as in 9/11 or as in Harvey — that's what the EPA is very ill-equipped to handle."
Coleman disagrees. "While there are some limitations to the instrumentation, these methods are the acceptable ways that this data should be collected," he says of the air monitoring system. "It's how we report this information to the public throughout the entire country."
"You can't hide that."
To Flores, the phrasing of EPA statements after Harvey felt like a slap in the face. He had been smelling gasoline in his neighborhood for days when the agency told residents there was nothing to worry about.
"I mean, God, we smelled it; why try to hide it? I mean, we're not idiots," he says angrily, sitting on his living room couch. "You can tell us all you want, 'Oh, you guys are good, nothing to worry about,' " he continues, and curses in frustration. "We're the ones smelling it! You can't hide that."
The mystery was solved when Magellan Midstream, a pipeline and petrochemical storage company in Galena Park, disclosed that nearly half a million gallons of gasoline had leaked out of its storage facility on Aug. 31. It turned out to be the largest single petrochemical leak reported to be caused by Harvey.
The company had first reported the leak, and the evacuation of employees from the facility, to the National Response Center spill hotline just after midnight on Sept. 1., two days before the EPA press release saying the air was safe.
Flores worries about the health effects from inhaling gasoline that evaporated into the air. Inhaling gas can cause respiratory problems. In larger amounts it can damage organs or contribute to cancer.
"Believe me, I think about it every day," Flores says. "I'm like, I wonder how many years of life I'm losing because I live out here. But it's home."
Yet, he says, if he has to leave his home to keep his family safe, he will.