How EPA's Design Values Exclude Air Pollution Days

UPDATE 1/14/2012: edited, included quotes from the news articles.

The Fairbanks North Star Borough recently announced the 24-hour PM 2.5 design value for 2011 was 38, significantly lower than in recent years.

We are at 38 [98th percentile design value]…. Actually this last year [2011], we had more exceedences than we had the year before [2010].”
Audio Track 1 (2:51-23:28) FNSB Assembly Finance presentation by Dr. Jim Conner, Air Quality Program 1/5/2012

Does a design value represent the day with the highest PM 2.5 concentration of the year? Does it consider most of the days of the year? Does it include local hot zones known to have the highest PM 2.5 concentrations, such as near Watershed Charter School or North Pole’s Rectangle of Death? [no to all]

Design values are the statistics reported to EPA for comparison to the National Ambient Air Quality Standard to determine compliance with the Clean Air Act. Calculating the design value is not difficult to understand or perform. However, two recent news stories muddled the facts by promoting design value as the year’s ”the highest 24-hour average readings.” Residents are given the impression the air pollution problem is nearly solved.

• Fairbanks borough to discuss air quality progress with public FDNM 1/7/2012
• Fairbanks borough air quality is better than expected for 2011 FDNM 1/6/2012

By definition, a design value is NOT the highest PM 2.5 concentrations that year. In Fairbanks, the EPA-approved method of calculating a design value excludes more than two-thirds of the actual days. Further, in reporting the “good news,” these news articles overlooked the simultaneous announcement of the growing number of days exceeding the state and federal 35 µg/m³ standard. The problem is not nearly solved but… a little wind can make it look that way.

Residents deserve to know how we are being harmed by high concentrations of fine particulate air pollution. Instead, the severity and frequency of high PM 2.5 concentration days are minimized by agency methodology and local reporting.  Misleading information dangerously risks our health and the health of those we love. Repeating inaccurate information is like hammering down a new roof with bent nails.

Ironically, the news articles came about because of a staff briefing presented to the Borough Assembly. The briefing was to prep the Assembly for consideration on whether to approve an $800,000 grant from the Federal Highway Administration for an Air Quality Educational Outreach Program (page 115). Describing the need for the program Dr. Conner stated,

The latest survey that we had shows that about 30% of the people in Fairbanks still don’t believe that there is an air quality problem in the winter.”

Now that gap has widened, increasing the need for this costly educational program.

To understand how the PM 2.5 24-hour 98th percentile design value is calculated and how it is used locally to exclude more days of data than it includes, review the federal regulations in 40 CFR 50 appendix N or follow this layperson summary below.

First, only the Federal Reference Method monitor located in downtown Fairbanks at the State Office Building (675 7th Ave, Fairbanks, AK 99701) is used to collect data for the design value calculation. The FRM collects one day of data every three days, resulting in approximately 120 days of PM 2.5 data each year. 245 “skipped” days during the year are not considered. Higher concentrations measured at other sites or by other monitors are also not considered.

Second, exclude invalid data (due to monitor malfunctions). FRM malfunctions become more frequent at temperatures below minus 40ºF. Severe cold, associated with a lack of wind, temperature inversions, and higher space heating demands, is when our highest winter PM 2.5 concentrations occur.

Third, “exceptional events” may be excluded also. For example, high PM 2.5 concentrations caused by wildfires have been excluded during previous years.

Fourth, array all remaining days of PM 2.5 24-hour (midnight to midnight) averages for the calendar year from highest concentration to lowest. EPA’s reliance on daily averages hide higher subdaily peaks which may have a greater association with acute cardiac and pulmonary adverse effects, see Johnson 2011.

Last, exclude two percent of the days with the highest 24-hour concentrations. The next highest concentration day is the 98th percentile design value for that year. In Fairbanks, with only 120 FRM sampled days each year, the top 2 concentrations are excluded, and the 3rd highest concentration is the design value for that year. (If we had 351 or more days of data, the 7 highest concentrations are excluded, and the 8th would be the design value.)

EPA uses a three-year average to compare to the standard of 35 µg/m³. In 2009 Fairbanks’ design value was 51. In 2010, it was 51.8. And now in 2011, it was 38. Thus, Fairbanks’ 2009-2011 average 98th percentile design value is 47.

The deadline for meeting the federal PM 2.5 35 µg/m³ three-year average design value standard is December 2014. Design values from 2011, 2012, and 2013 will be averaged and reported to EPA. Because 2011 exceeded the standard, the design values for the next two years must be lower than 35 µg/m³ to meet the standard.

A recent News-Miner editorial included a half-measure correction of the previous misleading news articles, Action needed: Officials update efforts to improve Fairbanks air FDNM 1/10/2012. The editorial notes the exclusion of the two days of data but seems unaware that fully two-thirds of each year, 245 days, are skipped over by the design value methodology. The editorial addressed the issue solely from a regulatory perspective and failed to connect any relevance to the life and safety of residents.

Children, elders, vulnerable individuals, and even our healthy adults and athletes are at risk from breathing here just one day. More likely than not, any given day of smoke pollution we breathe will not even be counted by regulators.

About these ads