The Ozone Weekend Effect

Traffic leaving the Caldecott Tunnel. PHOTO CREDIT: GEORGE BAN-WEISS Traffic leaving the Caldecott Tunnel. (Photo by George Ban-Weiss.)A large part of gasoline and diesel engine pollution consists of two components: soot and nitrogen oxides, or NOx. Soot is made up of tiny carbon particles that hang in the air and dirty it, but NOx, a mix of nitric oxide (NO) and nitrogen dioxide (NO2), has a more complicated story.  NOx feeds chemical pathways that are fueled by sunlight and produce the noxious brew we call smog. In general, reducing NOx tailpipe emissions reduces this pollution. But, as CEE professor Robert Harley has found, there is an exception to this rule that can confound air quality management efforts unless it is understood.

This puzzling exception is known as the ozone weekend effect. In many urban areas, tailpipe emissions of NOx are lower on weekends because the roads carry less diesel truck traffic on those days. This translates into lower levels of airborne soot on Saturday and Sunday and lower levels of secondary pollutants like ammonium nitrate on Sunday and Monday. (There's a one-day “chemistry delay” as NOx, ammonia and other chemicals in the air react to form their smoggy end-products.) But for ozone, the levels are often higher on weekends, when NOx emissions are lower.

Ozone is the poster chemical for air pollution, linked to many respiratory illnesses. There are two main ways to control ozone: either reduce NOx emissions; or reduce levels of volatile organic compounds (VOCs) such as evaporated gas, cleaning product hydrocarbons and methane (responsible for the current controversy over a possible “cow tax” on farmers). NOx emissions have dropped steadily over the past 15 years, thanks to cleaner car engines. And, ever since people in the air quality field began observing the ozone weekend effect, some have wondered whether continued NOx reductions might actually be counterproductive. “There's a big debate,” Harley explains. “Some people say, ‘We don't want to do more NOx control. We should focus on the other precursors to ozone.’”

Caldecott Tunnel sampling equipment, set up in the ventilation tunnel above the traffic. PHOTO CREDIT: GEORGE BAN-WEISS Caldecott Tunnel sampling equipment, set up in the ventilation tunnel above the traffic. (Photo by George Ban-Weiss.)To resolve the question, Harley has focused on a great source of data: the Caldecott Tunnel, which connects Highway 24 between Oakland and Walnut Creek, just two miles south of the UC Berkeley campus. Since 1994, Harley has measured levels of emissions products in the tunnel, including NOx, VOCs, carbon monoxide, carbon dioxide, soot and other particulates. Sensors are placed in both the southernmost bore, which is open to all traffic, and the middle bore, which is off limits to heavy diesel trucks. Running the numbers on the different sources gives Harley a good estimate of how much of the pollution comes from gasoline engines and how much comes from diesel. While some air quality policy has relied on statistical models, the Caldecott offers real data. “It's important to observe reality rather than just simulate it,” Harley says.

The Caldecott observations, along with ozone data from local air districts and other pollution studies, constitute an unusually good regional test bed for deciphering traffic's effect on air quality. “California is a great place to study vehicle emissions and their effects on the atmosphere,” Harley says. “We don't have a lot of other sources. There’s no coal being burned because the power plants here are natural gas and they’re well controlled, and there's a constant inflow of clean air from the Pacific.”

Logging Caldecott data over the years—changes in engines, fuel composition, air policy, population, the economy and everything else—has shown that NOx and soot emissions from car traffic have steadily declined, while diesel NOx emissions have stayed at about the same level. “With gas-engine vehicles, improvements in catalytic converters have more than compensated for population growth,” he notes. “And diesel is now the biggest source of soot and NOx nationwide—not just from truck traffic, but also from farm and construction equipment, railroad locomotives, ship engines, etc.”

Caldecott Tunnel field study team. From left to right: John McLaughlin (UCB B.S. 2008), George Ban-Weiss (UCB Ph.D. 2008), Melissa Lunden (LBNL staff scientist), Tom Kirchstetter (UCB Ph.D. 1998), Andrew Kean (UCB Ph.D. 2002). PHOTO CREDIT: ROB HARLEY Caldecott Tunnel field study team. From left to right: John McLaughlin (UCB B.S. 2008), George Ban-Weiss (UCB Ph.D. 2008), Melissa Lunden (LBNL staff scientist), Tom Kirchstetter (UCB Ph.D. 1998), Andrew Kean (UCB Ph.D. 2002). (Photo by Rob Harley.)By examining the chemistry of ozone formation, Harley found that ozone is generated most readily in air that has a mass ratio of 2.4 parts VOC to 1 part NOx. If the NOx levels are higher than this, reducing them just slightly—as happens on weekends when diesel traffic is reduced—can actually elevate the ozone level. This explains the ozone weekend effect. But if you reduce NOx levels significantly below that ratio, ozone production falls rapidly. On a 3-D graph, it's like having to go over a ridge to make it down to the low areas. Harley argues that, because much of the VOC content in the air now comes from natural sources like vegetation, further reduction of NOx is the best way to reduce ozone.

In terms of policy, this argument has now prevailed, and diesel is the new target for emissions standards. Nationally, all new diesel engines with model year 2007 and later are equipped with particle filters to reduce soot and other particulates. California law will likely be stricter: the state's Air Resources Board is set to pass a rule requiring all older trucks that operate in California to be retrofitted with particle filters (or have their engines replaced) by 2012, and, by 2023, to add catalytic converters for NOx control. Roughly one million existing trucks would be subject to these requirements.

So, even if ozone levels rise short term as a result of these diesel emission measures, we can breathe a sigh of relief that our air will be better later. Following the chemistry curve, as NOx levels decline, weekend ozone levels will plateau and then decline; weekday levels will follow the same path; and, ultimately, both will decline together. As Harley observes, “The weekend effect is a beautiful way of previewing some of the effects of reduced diesel emissions.”