The fire was immediately dramatic. It spread quickly to a railroad car full of chemicals. ChemCentral is owned by a multinational company that makes and distributes thousands of industrial chemicals, so dozens of highly toxic compounds were on-site that day. The blaze spread to a tanker truck. Several tanks caught fire. Each time the fire hit a new vat of chemicals, it exploded again. As it spread, it sent a towering cone of smoke into the sky. The cloud was as black as coal, and, as it hit the higher atmosphere, the winter breeze carried it straight toward downtown. It looked frightening, as though a severe tornado-alley thunderstorm had suddenly formed in otherwise clear skies. It passed over the residential neighborhood of Columbus Park, and as it did, the cloud rained down hot ash and debris.
Firefighters at Station P34 got the call just after 2 p.m. on February 8. They arrived first, but there was little anyone could do. Captain Lisa Malloy ordered her crew to help the evacuation of ChemCentral's employees. Malloy immediately radioed headquarters to ring a second alarm. Battalion 104 arrived as backup. They had no hope of putting out the fire, so the firefighters fell back into defensive positions. With her crew amped on adrenaline, Malloy ordered a "no screaming mode" to keep radios from getting jammed with panicky calls.
Soon, many downtown workers were either glued to live TV coverage or staring out their windows at the approaching cloud. Among the buildings in the path of the cloud was the Environmental Protection Agency's downtown office.
The EPA sent an environmental scientist named Mike Davis to collect samples at the scene. Janice Kroon, an on-scene coordinator for the EPA, stayed behind to field phone calls and to direct the agency's staff.
Like the firefighters, the EPA could do little. The smoke was drifting straight up, hundreds of feet into the air, and wind currents were carrying it southwest. It was impossible for Davis to get close enough to the raging fire to take samples directly from the cloud. The EPA relied instead on five air-quality monitors that it has stationed around Kansas City.
But the air monitors, which measure smog and pollution, can't detect the kind of chemicals that could have been traveling in the cloud.
During a fire, it's the EPA's policy to give firefighters all the latitude they need to protect lives. Kansas City Fire Chief Smokey Dyer took command, overseeing representatives of six government agencies including the EPA, the Federal Bureau of Investigation and the Federal Emergency Management Agency.
All around the city, people were wondering what was in the cloud of smoke. Dyer took control of that, too. He began assuring the media about the safety of the air, even though nobody had any idea what was really in the cloud. "None of these are what those of us in the field would call exotic chemicals," the fire chief told the Associated Press. "None of these have a high rating for toxicity." Dyer continued his assurances to the TV news: "Every indicator we have is that the air is safe — we don't know of any problems."
Dyer suggested that downtown workers and residents shouldn't worry about debris falling from the sky or the ominous cloud now drifting among the high rises. By early evening, cars and homes in Columbus Park and in parts of downtown had become coated with ash. Large pieces of debris, like roofing tiles and something that looked like Styrofoam, were dropping into yards and onto homes. It was unlikely, Dyer said, that anything falling was dangerous.
But Dyer and the federal authorities had no proof to back up these claims. This is made clear in internal EPA and Kansas City Fire Department documents the Pitch obtained through requests for records.
The EPA says lengthy testing after the fire shows that the smoke and debris weren't dangerous. But according to the hundreds of pages of reports and e-mails, officials told the public that everything was safe long before those results were in.
Authorities used the smog detectors as the basis to give the all-clear, but EPA officials knew that the smog indicators couldn't be used to make that determination. The EPA says testing that night found that the cloud did not contain harmful chemicals, but those results did not come in until the following morning — long after the public had been told that the cloud was safe.
As for the debris, testing on the ash and larger chunks didn't begin until the following day. And when it did, scientists found that there may have been traces of toxic chemicals in some ash samples.
Officials also claimed that no water ran off from the plant, but now they acknowledge that thousands of gallons of potentially tainted water ended up in sewer systems, which ultimately flow into the Missouri River.
Recently, the EPA and the Occupational Safety and Health Administration fined ChemCentral a combined $560,260 for shoddy record keeping and failing to maintain safety standards at the plant. ChemCentral failed to notify the EPA of a deadly chemical kept there, the EPA claims, which could have hampered firefighters' ability to fight the blaze.
But in the six months since the fire, federal and local authorities have failed to address their own mistakes.
About an hour after the ChemCentral explosion, a twin-engine plane took off from a small airstrip in Waxahachie, Texas. The ASPECT (Airborne Spectral Photometric Collection Technology) plane was the EPA's best tool to determine whether the cloud was carrying deadly gasses across downtown.
It took two hours for the plane to reach Kansas City. By then, the massive cloud had churned 2,000 feet above the city. It was dusk, and the black smoke melded with the slate-gray sky. Ray Brindle, a former commercial pilot, was in the cockpit that day. He could see the well-defined cloud from 30 miles out. At 5:53 p.m., he steered the plane into a racetrack pattern to circle the smoke. "The fire was pretty spectacular," Brindle tells the Pitch. "Right away, we knew it was pretty serious."
Two sensors jutting out through holes on the belly of the plane began collecting data about the cloud's makeup. The plane's sensors pick up infrared energy, called photons, coming off the Earth's surface. When the photons pass through chemicals in the air, the chemicals interact with the photon beams. Scientists can then figure out what chemicals are present by analyzing the frequencies of energy absorbed in the air.
As the plane circled about 50 feet from the cloud, the data poured into an onboard computer station manned by the EPA's Baron Leger. The computer didn't tell Leger what it was picking up, only that data were coming in properly. The sensors can detect vapors from chemicals in a mile-wide area; on the computer screen, every pixel equals a foot and a half.
The ASPECT plane had been flying for just six years, but its crew had extensive experience collecting such data. After hurricanes Katrina and Rita, this same crew ran dozens of missions across the Gulf Coast to see whether there were dangerous chemicals in the path of flood waters. While flying over the Port of New Orleans, the crew once identified a single drum of chloroacetic acid, a chemical used in pharmaceuticals. The crew was able to alert officials on the ground about the drum before floodwaters engulfed it and contaminated the water with a potentially toxic chemical.
After four or five circles around the cloud, Brindle steered the plane westward. He needed to land the plane so EPA officials could decipher the data they had collected. But the cloud blocked the normal approach to downtown's Wheeler Airport. Brindle brought the plane in low near the West Bottoms, coming in at an angle to make the runway.
Waiting was EPA environmental scientist Mark Thomas, who grabbed the hard drive out of Leger's computer. Thomas brought the information to the EPA's lab in Kansas City, Kansas, where it took the computer 10 minutes to spit out a series of graphs. (The information produced in the EPA lab looked like a stock-market report, with peaks and valleys to help scientists determine a chemical's makeup.) Thomas then matched up the graphs to a book of chemicals, like comparing the lines in a fingerprint.
The process of matching up infrared charts to chemicals isn't perfect, so Thomas sent the charts to the Los Alamos National Laboratory in New Mexico. There, Robert Kroutil, a chemist who invented the ASPECT plane, did the same comparisons Thomas was making in Kansas. They got on the phone to compare notes.
Thomas and Kroutil found that the cloud contained low levels of three chemicals: ammonia, methanol and trimethylbenzene, or TMB. Thomas wasn't sure about TMB. "What is this stuff?" Thomas remembers wondering. Both of them hit their respective libraries to figure it out.
TMB is a chemical often used to make drugs and dyes, they learned. Exposure to it can affect the nervous system and cause fatigue. But based on the EPA's guidelines for exposure to TMB, the levels were low enough that scientists determined the cloud was safe.
By then it was almost 7 p.m. Thomas passed his findings on to EPA officials. The short answer: The plane had found nothing at hazardous levels in the cloud.
This was the first time that officials knew the cloud was safe for people in downtown. But Dyer had been telling the media for hours that there was no reason to fear the smoke.
Asked why he made his statements so soon, Dyer says, "That was based on my 40 years in the business." His experience, he says, led him to believe that the debris was made up mostly of roof material caught in the updraft. And he had been receiving information from the EPA and others about ground readings at the site. "To my mind," Dyer says, "we would have experienced a lot more panic if we had waited."
The EPA didn't correct Dyer's statement, the agency's Kroon says, because it was the fire chief, not the EPA, who made the statement. "When first responders show up, their goal is life safety," Kroon maintains. "We aren't second-guessing them."
EPA officials say Dyer's early comments were based on data they collected from the smog indicators. But even the EPAs final report on the fire, released June 10, indicates the air-monitoring-system sensors "do not measure for specific contaminants" that could have been in the air.
The ASPECT plane made two more voyages around the cloud, once at 8:27 that night and then again at 8:06 a.m. on February 9. After that third flight, the EPA was finally confident that there was nothing dangerous in the air over Kansas City.
Altogether, the plane made 15 passes around the cloud. After the third round of data confirmed the initial findings, the EPA sent the plane home to Texas.
It had been more than 12 hours since Dyer's all-clear announcement. Four hours after the blaze began at 2 p.m., firefighters considered spraying a water-based foam mixture onto the fire. They feared that doing so could cause the chemicals to spread, so they asked for an opinion from federal and state scientists.
Everybody was in agreement: Spraying the fire could cause the chemicals to flow onto neighboring property, into sewers or perhaps straight into the Missouri River. In a report she wrote that night, Mindy McDaniel, an emergency-management specialist with FEMA, indicated that "it would present an environmental hazard to use water on the blaze."
Dyer, speaking to the media that night, cautioned that spraying the plant with water could spread whatever had been kept at the chemical plant. "We are not fighting this fire at all," Dyer said. "The only way to handle this fire is to allow it to burn itself out."
But at 11:20 p.m., Dyer reversed his decision, ordering crews to spray the fire. They quickly unloaded a mixture of 97 percent water and 3 percent foam on the fire. The plant exploded all over again. Flames shot 100 feet into the air. Fireballs rolled into the sky.
Soon, Dyer told his crew to stop. They'd wait until morning to spray again, when they could clearly see what was burning.
Authorities initially agreed that it would be harmful to spray the fire. But now, six months later, they're claiming that the level of toxic damage was negligible.
The night of the fire, authorities — including those at the EPA, the fire department and City Hall — told the media that the water was contained within the plant's property and posed no danger to waterways. But the EPA learned later that the foam mixture ended up flowing directly into Kansas City's Blue River Sewage Treatment Plant.
Lorene Lindsay, manager of Kansas City's Water Services Department laboratory, says plant workers began testing sewer water six hours after the start of the ChemCentral fire and continued for three days afterward. But because the Blue River plant handles 70 million gallons a day, it's unlikely the tests would have shown whether the diluted fire runoff contained toxins.
The foam mixture, after it was treated at the plant, then joined the rest of the treated sewage as it dumped into the Missouri River.
Dyer and EPA officials say the water presented no danger. The EPA's Ken Buchholz, the branch chief of the EPA's Superfund Division, says any contaminated water would have dissipated in the vastness of the sewers. Buchholz claims, "Really, the sewer system is the best place for it."
But because nobody's sure what was in the runoff, it's unclear whether the Blue River plant would've filtered it out. Some chemicals kept at ChemCentral may have gone undetected through the treatment system.
The next morning, the fire was still far from out. Explosions and occasional plumes of smoke and ash continued for much of the day on February 9. Eventually, federal authorities capped sewer drains and put a temporary dike around the plant to prevent further runoff.
But by the time the fire had gone out the following day, runoff had flowed directly into the sewers. The EPA estimates the runoff at 60,000 gallons. Dan Wayne was leaving the Crossroads when he spotted the ChemCentral fire smoke. It was in the direction of his home in Columbus Park, so Wayne headed toward Fifth Street and Harrison. Wayne, a photographer and developer who knows the East Bottoms well, tracked the cloud to its source.
"It was intense down there," Wayne recalls. "There was so much heat."
Wayne got there before most of the fire crews, so he was able to pull up right where the ChemCentral workers were standing. One had a burn on his leg. The workers told Wayne and other bystanders about how they had fled the explosions. "He was just saying that a tank erupted, and if they hadn't gotten out of there, they all would've died," Wayne recalls.
Cops finally kicked Wayne and the other bystanders out of the area, so he went back to his Columbus Park home. "I'm walking around back in my yard," he says, "and shit's just falling from the sky. It was this Styrofoam-type stuff that was all charred." They were shaped like cow patties and light as air. Bits of them fell in a nine-acre vacant field near his home.
At 9:29 p.m., Wayne fired off an e-mail to the EPA, telling the agency about what he had found in his yard. "I have no plan to touch them, but would be glad to help," he wrote. He gave directions to the property and added his phone number.
Wayne never heard back from the EPA. The only record of communication about his note is contained in an e-mail written by one EPA employee, Ina Square, who forwarded it to a co-worker with this comment: "Here's a good one for you."
The EPA's failure to respond to Wayne's request is one of several questionable decisions the agency made during its response to the ChemCentral fire.
The EPA defends its handling of the fire's aftermath, pointing to a series of reports the agency published that claim there was little or no environmental impact from the debris or the smoke.
The day after the fire, the EPA took wipe samples from several schools near the ChemCentral plant. The wipe samples, akin to rubbing a wet paper towel across a surface, are used to detect whether smoke or debris have left behind chemical residues. None of the samples turned up dangerous levels of chemicals.
But all of the schools tested were upwind of either the fire or the smoke — the cloud hadn't passed directly over them. No wipe samples were taken on property that was directly under the smoke.
EPA officials say they took wipe samples only because the Kansas City, Missouri, School District asked them to take extra measures to be certain that schools near the fire were safe. "The only reason we took the wipe samples," Buchholz says, "was at the request of the school district."
Howard Dunn, professor emeritus at the University of Southern Indiana and a retired chemist, became interested in the fire because his daughter was working in Kansas City at the time. Dunn studied the EPA documents. He says he was appalled that no one took wipe samples from areas the cloud passed over. "The long-term health effects — I don't think they can say anything with any certainty," Dunn says.
But Buchholz says wipe samples aren't the best indicator of whether the ash or smoke was harmful. A better gauge, he says, would be four ash samples from the neighborhood west of the plant.
Those ash samples have limited results. They cannot detect harmful levels of several chemicals that were stored at ChemCentral. In the EPA's June 10 final report, scientists acknowledged that the tests showed levels that "slightly exceeded the health-based benchmarks" for chemicals, including the pesticide dibromoethane and a fungicide called trichlorophenol. Both can be harmful, and dibromoethane can be fatal if ingested.
EPA toxicologist Mike Beringer cautions that the test results didn't show that the chemicals were present — only that the tests were inconclusive. Still, Beringer says, if the chemicals were present, they would have been at levels low enough so as not to be harmful.
"It's not just about exceeding the benchmarks," Beringer says. "When we do these analyses, we have to ask ourselves, How nasty are these chemicals and how are people being exposed?" Because residents could have ingested only small amounts of the ash, the EPA determined that the ash did not require widespread cleanup.
Whether the ash was harmful became irrelevant a couple of days after the fire. It rained and snowed, and Wayne recalls watching the debris run into the sewers. After fire investigators determined the explosion began when ChemCentral employees were handling the chemical Indopol, the EPA went back through the plant's records. EPA officials found that ChemCentral had failed to notify the agency that Indopol was stored at the Prospect Avenue plant.
So on July 23, the EPA filed a federal lawsuit against ChemCentral for violating the Clean Air Act.
It had already been a bad month for ChemCentral. The lawsuit followed a July 19 citation filed by OSHA, claiming that ChemCentral had ignored a litany of safety regulations at its plant.
ChemCentral attorney Dan Brennan has told the Pitch that the company intends to help federal authorities, in any way, to clean up the site. But Brennan says he doesn't expect cleanup of the site to be too lengthy. "Was there any chemical that was dangerous to the public? No, there was not. It was a very low level of chemicals that were released that day."
Meanwhile, the only agency to conduct a review of its response to the fire was the Kansas City Fire Department. At 11 a.m. on March 9, the department held a meeting to discuss its response. According to a three-page report from that meeting, fire department higher-ups identified some areas that they could work on, like getting better maps of the area.
But overall, Dyer came to this conclusion about the response: "The incident worked very well."