On February 20, people on the Swinomish reservation started smelling something strange. Some described the odor as burning tires or rotten eggs. Mike Cladoosby, 81, told KING 5 that breathing in the odor "burned," and that he spent the night in the hospital because of respiratory problems.

The Swinomish reservation sits on a little piece of land situated immediately south of the Shell refinery near Anacortes. After people reported the smell, Shell said that maintenance work resulted in a release of "sulfur-based compounds," but that no harmful levels of chemicals were detected. But by mid April, the Northwest Clean Air Agency (NCAA) had received 63 complaints regarding the incident. Officials are still investigating what happened.

On Thursday, the NCAA released two reports about the smell filed by Shell. The first report, dated March 30, detailed emissions for the month of February and maintenance on something called the East Flare line. The second, dated April 10, outlined emissions specifically related to the February 20 incident.

The reports read like technical gobbledygook, and the NCAA didn't interpret what they mean. To wrap my head around the numbers Shell reported, I called Dan Jaffe, professor of atmospheric and environmental chemistry at the University of Washington-Bothell and a Fulbright winner specializing in air pollution.

Jaffe agreed to help. "Give me 45 minutes," he said. Then, 45 minutes later, he had this:

Using a tool from the National Oceanic and Atmospheric Administration that takes into account weather patterns, Jaffe created a model of how hydrogen sulfide—the smelliest and most toxic compound described in Shell's report—likely swept over the Swinomish reservation between 12:50 p.m. and 4:18 p.m. that day in February. Jaffe used Shell's report of 2.3 pounds of emissions. (Shell's reports show that other compounds were also emitted, and in greater quantity than the hydrogen sulfide, but Jaffe said a toxicologist would have to figure out whether those emissions were significant enough to impact people.)

"I think what we can say is that it's totally consistent with the reports people had, and it shows how an industrial accident can influence nearby communities very quickly," Jaffe said about the hydrogen sulfide release. According to KING 5, the community learned about the source of the smell two hours into the incident, when Shell issued a press release.

We still have to wait for the NCAA to issue a conclusion about the size of the different emissions, their impact, and what went wrong. Shell's self-reported numbers don't result in large concentrations of hydrogen sulfide, but as Jaffe's model shows, it doesn't take much for these compounds to affect people. "Any time you have industrial facilities located near people, this is going to happen," Jaffe said. "I think we have to hold industries accountable to do the best job they can, and a part of it ought to be notification."