In the aftermath of the Chevron refinery fire nearly three weeks ago, state and federal investigators are trying to determine the cause of the original fuel leak and evaluate whether the response inside the refinery was appropriate.
A similar independent probe should be conducted to analyze the Bay Area Air Quality Management District's poor response. In such emergencies, the district must provide useful and accurate information to help guide the emergency response. It did neither.
Instead, as I reported two weeks ago, the district put out misleading information the day after the fire that its air samples showed levels of toxic pollutants to be under state standards and "not a significant health concern."
But none of those test results measured smoke particulates in the air, a likely cause of respiratory problems. Rather, the results measured gaseous pollutants. And it turns out that one reading did show levels above state standards.
On Thursday, 2½ weeks after the fire, the district released its findings on smoke particulates. As I predicted, it was useless information.
That's because the district measures for particulates at just six Bay Area locations. One happens to be in San Pablo, about three miles from the refinery. But the samples are only collected for a 24-hour period once every six days.
The next scheduled collection for the San Pablo location happened to begin at midnight, about six hours
After finally completing analysis of that sample, the district reported "slightly elevated levels of elemental carbon, a marker for combustion."
What does that mean? "The particulate results are what we expected to see given that the monitoring began at midnight after the fire was out," said Dr. William Walker, Contra Costa's health director. "These results, however, do not suggest there were not health impacts experienced by residents in the immediate area."
Walker's comments were contained in a two-page news release in which air district officials could not bring themselves to clearly state the obvious: They don't have any meaningful data about smoke particulates the night of the fire.
It's not just that the district has consistently overstated the significance of its information; it has also misrepresented the data. The day after the fire, the district news release claimed gaseous samples from six locations were tested for 23 compounds, and none exceeded state standards.
Only after I questioned them did they acknowledge that a sample taken at the top of the El Cerrito hills had levels of Acrolein, a respiratory irritant, above the state standard. But, in a news release admitting the error, the district downplayed its significance.
The district said Acrolein was "commonly" found above that level in the Bay Area. The statement also said the measurement was within the "background level" found in Bay Area air.
Dr. Paul Blanc, chief of the Division of Occupational and Environmental Medicine at UC San Francisco, disputes that. He says the Acrolein reading was well above the 90th percentile, and perhaps much higher, for periodic air sampling readings.
The Acrolein reading was 3.2 parts per billion. The district news release suggested this was a midrange reading, stating that "levels routinely range between 1 and 4.5 parts per billion." But that's not what the state data shows.
As Blanc notes, the median in the 2011 state data is 0.6 ppb, and the 90th percentile was 1.1 ppb. Thus the reading of 3.2 ppb was quite high, certainly not a midrange reading.
Eric Stevenson, a chemical engineer responsible for air district testing, defends his agency's conclusion. He says the state data averages wide fluctuations in short-term readings. Hence, he says, the Acrolein reading should be compared to those broader numbers.
What's clear is that the air district officials are making their calls based on their interpretation of the data without consulting outside experts. With the health of Bay Area residents at stake, it's time for independent review of practices and standards to ensure the agency doesn't fail us again.