Trace Gases

a. Ozone

       1) Continental United States

 The NOAA Climate Monitoring and Diagnostics Laboratory (CMDL), in cooperation with several other organizations, operates a network of 16 Dobson ozone spectrophotometers spread across the continental U.S., as well as at Barrow, Alaska, Mauna Loa, Hawaii, and American Samoa. These instruments are part of a global network of approximately 60 such instruments that are calibrated against the World Standard Dobson Spectrophotometer maintained by NOAA/CMDL. At four stations spanning the middle of the United States between 37° and 40°N (Fresno, CA; Boulder, CO; Nashville, TN; and Wallops Island, VA), total column ozone amounts during 1999 (red curve) remained well below pre-1980 levels (Fig. 65). The low 1999 ozone amounts are comparable to those observed during most of the 1990s, but contrast with the relatively high amounts observed in some years in association with the quasi-biennial oscillation (QBO). Overall, average total column ozone has declined by about 5% over the past two decades (blue trend line) at these stations, with the rate of decline slowing in the past five years.

At the Mauna Loa Observatory (20°N) in Hawaii, total column ozone concentrations exhibit a strong annual cycle (Fig. 66), although there is also some variability related to the QBO and to the long-term trend. Prior to 1977, there was no discernible trend in ozone amounts at this site. However, ozone has decreased at a rate of 0.6% per decade since that time, which is notably less than the average ozone decrease of 2.5% per decade observed at the midlatitude U. S. sites.

       2) Southern Hemisphere Ozone

Total column ozone data were obtained from the National Aeronautics and Space Administration Solar Backscatter Ultraviolet Spectrometer (SBUV) instrument on Nimbus-7 (1979 through 1988), and the SBUV/2 instruments on NOAA-11 (January 1989 through August 1994), NOAA-9 (September 1994 through June 1997), and NOAA-14 (beginning in July 1997). Data from the SBUV instruments are only available during daylight viewing conditions, which results in an absence of data over the polar latitudes during winter. Other sources of ozone data include Dobson spectrophotometer readings and measurements from balloon-borne ozonesondes, both of which are obtained from the NOAA CMDL.

The ozone hole, denoted by total column ozone concentrations less than 220 Dobson units (DU), typically reaches its peak areal extent in early October. During the last 20 years the average size of the ozone hole (Fig. 67) has increased from 1. 5 x 10⁶ km² in 1982 to the record value of 16.7 x 10⁶ km² in 1998. During 1999, the average size of the ozone hole decreased to 16.2 x 10⁶ km², slightly less than the 1998 value.

Daily measurements indicate that the maximum areal extent of the ozone hole during 1999 was 22 x10⁶ km² (Fig. 68a), which is less than the record value of more than 25 x 10⁶ km² observed during 1998. In recent years the ozone hole has also exhibited a tendency to last later into the year, as was observed during November–December 1999 when several daily records of maximum areal extent were established. The ozonesonde data also indicate that 29 September 1999 was the earliest date for the minimum total ozone of the season. In previous years, the minimum ozone has always occurred between October 4–15. Thus, the overall duration of the ozone hole was longer during 1999 than has been observed in past years.

Total ozone concentrations are closely coupled to lower stratospheric temperatures through photochemistry. Extremely low stratospheric temperatures (below –80°C) contribute to the formation of polar stratospheric clouds (PSCs), which enhance the production and lifetime of reactive chlorine, thereby leading to ozone depletion (WMO/UNEP 1994). The areal extent of the ozone hole is also closely related to the polar stratospheric vortex, which isolates and concentrates the chemicals that destroy ozone at low temperatures. From early October through early December the areal extent of this vortex was larger than the long-term average (Fig. 68b), but considerably smaller than the record values observed during 1998.

The evolution of conditions accompanying the 1999 ozone hole was obtained from 68 ozonesondes flown at the South Pole (Fig. 69a). Total ozone for each flight is shown (red curve) along with the average temperature in the 20–24 km layer (black curve). Following sunset on March 20, stratospheric temperatures began dropping steadily from –50°C as the polar vortex developed over the Antarctic continent and began to isolate the stratospheric air from warmer midlatitude intrusions. By mid-June, temperatures in the 20–24 km layer had dropped to below –90°C, which is well below the threshold for polar stratospheric cloud formation, and subsequently remained at this level until early September. In fact, a near record minimum temperature of -94.7°C (178.5 K) was measured at 21.5 km on 6 August 1999. At these temperatures the chlorine activation reactions occur on the PSC surfaces, producing easily-photolyzed forms of chlorine compounds.

At sunrise on September 23 the free chlorine was released, which initiated the catalytic reactions that destroyed ozone. Severe ozone depletion was then evident from late September into early December 1999. The primary breakup of the ozone hole then occurred in early December (Fig. 68a) when the overall stratospheric polar vortex began to dissipate and air of midlatitude origin moved into the polar region.

Ozone concentration profiles from three ozonesondes during 1999 show the pre-ozone hole conditions on 28 July (Fig. 69b), minimum total ozone conditions on 29 September (Fig. 69c), and post-ozone hole conditions on 23 December (Fig. 69d). The pre-ozone profile indicates a maximum in ozone between 15–20 km, which is the primary contributor to the total column ozone of 255 DU. In contrast, the 29 September profile shows essentially complete ozone depletion in this layer, with total ozone of 90 DU on that day approaching the record low value of 88 DU observed in 1993. After ozone recovery, the maximum in ozone became established between 18–23 km (Fig. 69d). Thus, it is evident that the ozone hole reflects the annihilation of the primary ozone maximum that normally occurs between 15–23 km. It is also evident that stratospheric chlorine levels still remain at a "saturation" level in the main ozone depletion region above the South Pole.

Initial signs of Antarctic ozone recovery are expected within the next 10 years (Hofmann et al., 1997) when stratospheric chlorine concentrations begin returning to pre-1980 levels. This recovery should first appear as a decrease in both the ozone loss rate and in the extent of the ozone depletion near the top of the main depletion layer.

b. Carbon Cycle Greenhouse Gases

The three major carbon cycle gases are carbon dioxide (CO₂, methane (CH₄), and carbon monoxide (CO), with each having its own impact on climate. Methane and CO₂ have direct effects on climate due to their significant absorption of terrestrial IR radiation. While CO does not absorb terrestrial IR strongly, it and CH₄ indirectly affect climate through their atmospheric chemistry. Both species are removed from the atmosphere predominantly by chemical reaction with hydroxyl radical (OH). The OH concentration depends on the amounts of CH₄, CO, NO_x, O₃, and nonmethane hydrocarbons in the atmosphere. Changes in the atmospheric burdens and distributions of these compounds can affect OH concentrations and, therefore, the residence times of CH₄ and other greenhouse gases which are treated under the Kyoto protocol [such as the hyrdofluorocarbon compounds (HFCs)]. Oxidations of CH₄ and CO also affect the concentration and distribution of tropospheric ozone, another strong greenhouse gas.

      1) Carbon Dioxide

After water vapor, CO₂ is the most important infrared absorbing (i.e., greenhouse) gas in the atmosphere. Since the late 1800s atmospheric CO₂ levels have increased approximately 30%, due primarily to emissions from combustion of fossil fuels and to a lesser extent from deforestation (Keeling et al. 1995). The total sink and the partitioning into marine and terrestrial components varies significantly from year to year (Conway et al.1994; Ciais et al. 1995). A better understanding of the processes that remove CO₂ from the atmosphere and how these processes respond to climate fluctuations will enable better predictions of future CO₂ levels, which will in turn decrease the uncertainty associated with models of future climate.

Current modeling approaches to understanding the carbon budget depend on atmospheric measurement data either as a constraint (carbon cycle models) or as input (inversion models). The latitudinal variation of atmospheric CO₂ determined from the NOAA/CMDL Global Air Sampling Network for 1989–98, with the record from the South Pole shown in red (Fig. 70), conveys the long-term CO₂ increase, the interhemispheric difference, and the seasonal variations of atmospheric CO₂. Note that even though CO₂ in the Northern Hemisphere is on average 3–4 ppm higher than in the Southern Hemisphere (because 95% of fossil fuel CO₂ emissions occur in the Northern Hemisphere), the mean CO₂ gradient is reversed during the northern summer because of photosynthetic uptake by plants. In contrast, the mean north-south gradient is strongest in spring when respired CO₂ has accumulated in the atmosphere. Carbon cycle models attempt to reproduce this measured distribution by combining source/sink functions with atmospheric transport models, while inversion techniques combine the data with a transport model to deduce sources and sinks.

       2) Methane

Methane contributes about 20% of the direct radiative forcing attributed to those long-lived greenhouse gases which are directly affected by human activities. Changes in the burden of methane also feed back into atmospheric chemistry, and indirectly affect climate by influencing other greenhouse gases such as tropospheric O₃, stratospheric H₂O, and the concentration of OH. These indirect effects are estimated to add ~40% to the direct climate effect of methane (Lelieveld et al. 1993). It is suggested that reducing methane emissions to the atmosphere would decrease its potential effect on climate. However, before reasonable policies can be developed to reduce CH₄ emissions, its budget of sources and sinks must be better understood.

High-precision measurements of atmospheric methane indicate that approximately 70% of CH₄ emissions are in the Northern Hemisphere. As a result, methane concentrations are also higher in that hemisphere, with the largest concentrations observed at higher latitudes (Fig. 71). There is a strong seasonal cycle to methane concentrations throughout the globe, with a particularly regular seasonal cycle observed in the high latitudes of the Southern Hemisphere (see inset Fig. 71). In the Northern Hemisphere middle and high latitudes (30°–90°N), the seasonal cycle features a peak in CH₄ concentrations from mid-January–mid-February (Fig. 72a, blue curve) and a minimum concentration in April.

There is also an upward trend in CH₄ concentrations through the globe, with a de-seasonalized trend in the Northern Hemisphere extratropics of 8.0 ppb yr1 evident between 1984–98 (black curve, Fig. 72a) . However, the instantaneous growth rate of methane in this region, determined as the derivative of the trend line (Fig. 72b), has decreased by 0.9 ppb yr^-1. Superimposed upon this long-term decrease in growth rate are significant interannual variations. One outstanding feature is the large increase in growth rate during 1998, possibly related to increased CH₄ emissions from wetlands due to anomalously warm temperatures in that year. This suggests that changes in climate could have a large impact on CH₄ emissions, and, therefore, on the contribution of methane to earth’s radiative budget.

       3) Carbon Monoxide

Unlike CO₂ and CH₄, carbon monoxide is not a strong absorber in the terrestrial IR spectral region, yet it still has an impact on climate through its chemistry. The chemistry of CO affects OH (which influences the lifetimes of CH₄ and HFCs) and tropospheric O3 (itself a greenhouse gas), so emissions of CO can be considered equivalent to emissions of CH₄ (Prather 1996). Current emissions of CO may contribute more to radiative forcing over decadal time scales than do emissions of anthropogenic N2O (Daniel and Solomon 1998).

The lifetime of CO is on the order of weeks to months, depending on location and season. As a result, CO has large spatial gradients (Fig. 73). At the South Pole, CO varies from about 30 ppb during summer to 60 ppb in winter. However, CO values are much larger in the Northern Hemisphere where emissions are largest. For example, at Barrow, Alaska, CO varies from 200 ppb during winter to 90 ppb during summer. During most of the 1990s, globally-averaged CO amounts decreased by approximately 2% yr^-1, possibly due to decreased emissions as a result of catalytic converters on automobiles (Bakwin et al. 1994). As with CH₄ and CO₂, 1998 was anomalous in that a significant global increase in CO was observed.

c. Chlorofluorocarbons

Tropospheric concentrations of total organic equivalent effective chlorine (EECl) continued to decline through the late 1990s (Fig. 74). EECl takes both stratospheric release and the increased weighting of bromine into consideration and essentially represents the ability of the chlorine and bromine in tropospheric gases to contribute to stratospheric ozone depletion. In the past few years, CFC–11, CFC–113, and CCl₄ all declined at rates of less than 1% per year. CFC–12, which had been increasing in recent years due to its longer lifetime and to continued emission from automobile air conditioners, had a 1999 growth rate near zero in the Northern Hemisphere but continued to increase in the Southern Hemisphere. The growth of CFC–12 concentrations in the Southern Hemisphere results mainly from interhemispheric transport.

The primary gas driving the decrease in total organic chlorine is the solvent methyl chloroform (CH₃CCl₃) (Fig. 75). The Montreal Protocol banned production of CH₃CCl₃ in developed countries in 1996, along with that of the CFC’s and carbon tetrachloride (CCl₄), another industrial solvent and a precursor to CFC production. Methyl chloroform has a much shorter lifetime than the CFC’s (5–6 years versus 50–120 years), and is therefore being removed more rapidly from the atmosphere at a rate of more than 15% yr^-1. This rate will decrease in the future as CH3CCl3 is depleted from the atmosphere (Montzka et al. 1999).

Increases in halon concentrations, particularly for H–1211 (CBrClF₂) (Butler et al. 1998; Fraser et al. 1998), are also of concern and underscore the need for adherence to the Montreal Protocol guidelines if a decline in total chlorine is to continue. The bromine in these gases has the potential to outweigh the effects of decreasing chlorine in the atmosphere. The atmospheric burden of HCFC’s increased during 1999 at rates similar to those reported for the early 1990s (Montzka et al. 1996)

In 1999 new information was reported demonstrating that the CFCs’, major chlorocarbons, and halons presently in the atmosphere were absent entirely in the early 20th century, as shown for air collected in consolidated snow (firn) in the Greenland Ice Cap (Fig. 76). The gas concentration profiles show that at this site beyond a depth of 69 m, which relates to CO₂ age related concentrations present in 1929, there was no evidence of any of the CFC’s, chlorocarbons, and halons listed in Figure 76. Measurements of these gases in air trapped at high Northern and Southern Hemispheric latitudes showed further that models of anthropogenic emissions could explain the atmospheric histories of these gases (Butler et al. 1999). This work confirmed that the often-invoked volcanic and biospheric emissions are inconsequential or nonexistent in their effects upon atmospheric budgets of these gases, and that the presence of these gases in the atmosphere is due almost entirely to human sources.