UCAR > Communications > Quarterly > Fall 1998 Search

Fall 1998

Ozone update

Ridley in Arctic garb in front of the trailer housing the PSE 98 instruments. (Photo by Ann Louise Semner, Purdue University.)

Although ozone loss has been shouldered out of public consciousness by such fresher environmental "crises" as El Niño, it is still a vital area of study for atmospheric chemists. In fact, within the last ten years, a new ozone mystery has been uncovered that Brian Ridley (NCAR Atmospheric Chemistry Division, or ACD) calls "one of the most interesting tropospheric discoveries of the decade." And records of stratospheric ozone depletion now extend back long enough that new conclusions can be drawn about the process and its effects.

Brian Ridley. (Photo by Mike Shibao.)

It's common to think of ozone (O3) as a Dr. Jekyll-Mr. Hyde trace gas. In the stratosphere, where about 90% of atmospheric ozone is located, it's the beneficent Dr. Jekyll, protecting the earth's surface from ultraviolet radiation that can cause skin cancer and other serious illnesses. In the troposphere, it's Mr. Hyde: increasing tropospheric ozone emitted by fossil fuel and biomass burning has been shown to harm human health and agricultural productivity. However, the situation is not that simple, because ozone in the troposphere is also Dr. Jekyll. It's needed to produce hydroxyl radicals that cleanse the atmosphere, breaking down a variety of other trace gases such as carbon monoxide.

Two different groups of NCAR scientists have been focusing on ozone depletion, one in the troposphere and the other in the stratosphere. In the two regions, the processes involved are somewhat similar, but their significance is completely different.

Ground-level Arctic ozone depletion

Canadian Forces Station Alert. (Photo by Ann Louise Sumner.)

About 500 miles from the North Pole, on the northern tip of Ellesmere Island, is a Canadian military base called Alert. Established in the 1950s, it's the Northern Hemisphere's northernmost permanent settlement. About ten years ago, the Canadian Atmospheric Environment Service (AES) started measuring surface ozone at Alert. In springtime, the usual measurement is 30 to 50 parts per billion by volume (ppbv). During some periods, however, the ozone concentration drops to zero.

When this happens, two conditions are always present. First, the ozone-depleted air comes from over the ocean. (Alert is at the edge of the Arctic Ocean, although the water is frozen during most of the year.) Second, the episodes only occur in the spring, which suggests that the depletion reactions require sunlight and a fairly stable atmosphere, without too much heating to create vertical mixing.

"The exact mechanism for the depletion is not known," explains Ridley, "but from the previous campaigns that the AES has organized, from measuring a bunch of hydrocarbons and other chemicals, all the evidence points to radicals of bromine and chlorine atoms destroying the ozone"--the same reactions as in stratospheric ozone depletion. The bromine and chlorine radicals are somehow liberated from more stable compounds in sea salt or perhaps from biological sources in the ocean itself, but the process by which they are released is unknown.

Polar Sunrise 98 Experiment personnel

From NCAR: James Walega, Denise Montzka, Frank Grahek, and Brian Ridley, ACD. From Purdue University: Paul Shepson, Ann Louise Sumner, and Bryan Splawn. From AES: Sandy Steffen, John Deary, and Hacene Boudries. Although they were not on site for PSE 98, Len Barrie, Jan Bottenheim, Kurt Anlauf, Alan Gallant, and Bill Schroeder of AES organized it and the earlier PSE field programs. Ridley's group also helped to collect canister samples for Elliot Atlas (ACD) and Jochen Rudolph's group at York University to study the ratios of hydrocarbons and alkyl nitrates, which can show whether bromine or chlorine is the more important catalyst in the ozone depletion.

Ridley and three other ACD scientists spent last March and April at Alert, hoping to sample some ozoneless air. The program, called the Polar Sunrise Experiment (PSE) 98, also included groups from Purdue University, York University, and the AES. During early March, temperatures ranged from -30 to -40 degrees C (-22 to -40 degrees F) and near darkness; in April, it was -15 to -30 degrees C (5 to -22 degrees F) with 24 hours of sunlight. Ridley calls the program "a mopping-up experiment"; the general working of the depletion process had been ascertained by earlier field programs, and this experiment was an attempt to pin down some of the details.

The NCAR team's mission was to record levels of nitric oxide (NO), one of the most reactive of the atmosphere's nitrogen oxides, along with nitrogen dioxide. The two chemicals are jointly called NOx. A team from Purdue University measured formaldehyde. "For bromine and chlorine to affect ozone, there can't be much NOx and formaldehyde [in the atmosphere]," Ridley says. Bromine and chlorine react easily with NOx and formaldehyde, forming stable compounds. They will react with ozone only when NOx and formaldehyde concentrations are low. Thus "It's important [to know] whether NOx is 30 parts per trillion or 3 parts per trillion. It can make all the difference between bromine being able to be responsible for what's happening and not."

Earlier PSEs had attempted to measure NOx levels, but the instrumentation wasn't stable enough to record very small concentrations. Ridley's group can measure concentrations of NO as low as one-half part per trillion, using a chemiluminescence instrument. The instrument is a flow system at low pressure. A high concentration of ozone is added to the air in the reaction vessel; when ozone reacts with NO, the product emits photons, which are measured by a photomultiplier (a standard detection device). The amount of photons indicates how much NO is in the air.

The Arctic cold didn't affect the trailer-housed instrument. "We had more problems keeping the trailer cool, because there were so many instruments and people inside," Ridley says. "A lot of instruments don't like running at 80 and 90 degrees [F; 26-32 degrees C]." Although the inlets on the outside of the trailer sometimes froze up, on the whole, the system worked well.

The scientists hoped to see three or four ozone-loss episodes over their two-month stay. As it happened, there was only one, lasting 36 hours. "It's hard to base a lot of science on one event," Ridley says, "but the indications are that certainly formaldehyde was low, and in general NOx was pretty low. One of the problems we had was that the base uses 500,000 gallons of diesel fuel a year, and that creates a lot of NOx. We received a lot of pollution from the base, even during the ozone depletion episode. But we should be able to filter that data out."

Ridley emphasizes that this ozone loss, unlike the stratospheric variety, is "99% guaranteed to be a natural process." Signs of the same processes, although not complete ozone depletion, have been measured at the South Pole; it could be that the higher winds in the Antarctic mix the air too well for a total ozone loss. In the Arctic, Ridley says, "The only reason we didn't know about these episodes before is because the measurements were not being made."

In the stratosphere

Ozone warms the stratosphere by absorbing both infrared and ultraviolet radiation; thus ozone loss in that region should lead to cooling. The effect is complicated by such factors as changes in atmospheric dynamics and competing effects of carbon dioxide and other greenhouse gases, but it is well known and appears in climate models.

These graphs show changes over time in column ozone (top curves) and stratospheric temperature near 16 km (bottom curves). Measurements were made in spring over Antarctica (October, left) and the Arctic (March, right). The smooth lines highlight the respective low-frequency, decadal-scale trends. Note the similarity between the sharp temperature drop in Antartica coincident with ozone loss in the early 1980s (the ozone hole) and correlated ozone-temperature drops in the Arctic almost a decade later, in the early 1990s. (Figure courtesy of William Randel.)

Ozone has been closely monitored over the South Pole, by satellite and ground-based systems, since depletion was first noticed in the middle 1980s. During that period, stratospheric temperature records show considerable cooling. But the exact correlation between this cooling and the rapidly decreasing ozone levels has remained an area of study because of large year-to-year variations in both of these factors. Quantifying the causes and effects of those variations requires a decade or more of records.

Bill Randel. (Photo by Carlye Calvin.)

William Randel (ACD) co-chaired "Trends in the Vertical Profile of Ozone," a multiyear assessment of stratospheric ozone trends sponsored by the World Climate Research Program's Stratospheric Processes and Their Role in Climate (SPARC) project, and he is active in SPARC's Stratospheric Temperature Trends Group. These efforts and his earlier collaborative research on how climate models respond to global ozone depletion led him to revisit all existing ozone and temperature data sets to study correlations between variability in the two factors and trends. "Much of the work was getting the data," Randel remarks. Although all of the data he needed are housed at NCAR, he had to sift through balloon-based records from about 700 stations to single out polar stations with long time series at high altitudes.

Randel and ACD colleague Fei Wu have used these data to document remarkably large stratospheric cooling over both polar regions in spring. Their results will be published in the Journal of Climate this fall.

South and North: The same, only different

The Antarctic temperature data show what Randel calls a "near-step-like" drop in springtime temperature from before the ozone hole to after it. Cooling is strongest in the lower stratosphere (about 12-21 km, or 7.5-13 mi, in altitude), with a maximum drop on the order of 6-10 degrees C (11-18 degrees F) in spring (October-December) each year. During the Antarctic summer, smaller but statistically significant losses in temperature and ozone continue; during the fall and winter, no cooling or ozone depletion occurs. The good agreement between these observations and simulations of the ozone hole by general circulation models confirms that ozone loss is the source of cooling in Antarctica.

This extensive springtime stratospheric cooling--unlike the ground-level ozone depletion that Ridley studied--stems from human actions. In fact, Randel points out that it is one of the atmosphere's strongest indicators of global change. "It's much different from the troposphere, where trends are harder to identify," he notes. That's because trends in the troposphere are masked by such complicating factors as the effects of sea-surface temperatures and clouds and natural variations such as the El Niño/Southern Oscillation phenomenon. Furthermore, the change in the stratosphere is larger. "There's been more cooling in the stratosphere in 20 years than warming in the troposphere in 100 years," Randel says.

The AES observatory at Alert includes an ozonesonde station. Here, a sonde is being readied for release. (Photo by Ann Louise Sumner.)

For the Arctic, there are fewer long-term radiosonde and ozonesonde stations, and the records don't extend back as long. However, satellite data show strong ozone loss and colder temperatures during the 1990s. These mesurements are similar to those recorded in the Antarctic, but the change occurred about eight to ten years later in the North. On a graph, the temperature-ozone drops for the two hemispheres look very similar (see figure). One important difference is that the Antarctic ozone loss and cooling occurs mainly in spring, but in the Arctic the changes occur in winter and spring. The reason for this difference is not understood.

In the Arctic, six of the last ten winters have been unusually cold. "A big question is, why is the Arctic so much colder now?" says Randel. "It seems that there are fewer stratospheric warming events; the region becomes colder and less perturbed by dynamical waves. Some speculate that it's because of changes in tropospheric circulation or because of greenhouse gas increases. While these mechanisms may contribute, I believe the cooling is mainly due to ozone depletion, because of the strong similarity in space-time patterns between the Arctic and Antarctic."

Randel notes that there may be an amplification effect, where the cooling that results from ozone loss creates additional ozone loss via chemical depletion occurring at cold temperatures, which then leads to more cooling, and so on. This feedback is likely to be more important in the Arctic than in the Antarctic because it's warmer in the North. Winter temperatures in the Arctic stratosphere historically have hovered around the "cold threshold" for chemical ozone depletion. (At 50 millibars, this threshold is about 195 K, or -78 degrees C.) Any cooling as a result of ozone loss might push temperatures below that threshold, in which case the feedback would come into play. In the colder Antarctic, winter stratospheric temperatures fall well below the cold threshold anyway, and therefore almost all of the ozone is chemically destroyed each year. Additional cooling would have comparatively little effect.

Amplification aside, Randel wants to test his belief that stratospheric ozone depletion is leading to the Arctic cooling, using model sensitivity tests of the stratosphere's response to observed ozone and circulation changes. Tropospheric forcing and ozone levels can be varied in an idealized model to quantify cause and effect. This will be a first step towards better understanding of these large changes in the stratosphere's climate, and their potential effects on the troposphere.

Upcoming experiment seeks collaborators

The return of light and warmth in spring doesn't affect ozone just at the poles; it also is important at the midlatitudes. There, the spring transition revives a range of photochemical processes and relationships that were dormant during the winter, involving ozone, nitrogen oxides, various radicals, volatile organic carbon, and other chemical species.

ACD is planning a field experiment to answer questions about these springtime processes in the troposphere, and it is looking for input from the atmospheric chemistry community. The experiment, called Tropospheric Ozone Production about the Spring Equinox (TOPSE), is currently targeted for the spring of 2000, though that date may change. TOPSE proposes a series of aircraft flights from the Boulder area through Yellowknife, Canada, to about 75 degrees N latitude. On these flights, one or two aircraft will make in situ and lidar measurements of radiation, chemical species, and tracers. These measurements should help answer questions about the rates of ozone production and loss, the evolution of reservoirs of radicals and NOx, the types and sources of volatile organic carbon species, and other issues. The data will also be useful for chemical models.

Guy Brasseur, director of ACD, is seeking input from scientists who would be interested in measurements or modeling related to TOPSE. He hopes to find U.S. and international collaborators. "We are now meeting with university representatives," Brasseur says. "The [TOPSE] proposal is open for discussion." You can contact him at 303-497-1456 or brasseur@ucar.edu.

In this issue... Other issues of UCAR Quarterly

Edited by Carol Rasmussen, carolr@ucar.edu
Prepared for the Web by Jacque Marshall
Last revised: Tue Apr 4 15:03:21 MDT 2000