Introduction

ROLF MÜLLER

Institute for Energy and Climate Research (IEK-1), Forschungszentrum Jülich, 52425 Jülich, Germany

1.1 The Stratospheric Ozone Layer

1.1.1 Early Observations of Stratospheric Ozone

Regular measurements of stratospheric ozone started in 1924, when G. M. B. Dobson designed a spectrograph to measure the total ozone column that was suitable for routine outdoor use. Dobson's new instrument allowed regular measurements to be made over extended time periods. The success of the measurement program initiated by Dobson is obvious by the fact that today the unit for the total atmospheric column of ozone is called the Dobson unit (DU). Dobson's first observations at Oxford in 1924–1925 showed a marked annual variation of ozone and a strong day-to-day variability that was closely connected to meteorological conditions. The Dobson instrument network was extended to worldwide measurements until end of 1927. In the USSR, the first measurements of total ozone commenced in 1933; later, in 1959, the M-83 ozonometer was developed, which became the basis of the USSR ozone station network. Up to today, ground-based total ozone measurements are essential for long-term monitoring of the ozone content of the atmosphere.

The early total ozone measurements, however, did not allow estimates of the altitude profile of the stratospheric ozone concentration to be made, so that in the late 1920s, the ozone layer was still assumed to be located in the upper stratosphere. However, in 1933 Götz et al., based on so-called "Umkehr" measurements with a Dobson spectrograph, realized what is well known today, namely that "the average height [of the ozone in the atmosphere] at Arosa now appears to be about 20 km". Soon thereafter, the first balloon-borne measurements of the solar UV spectrum in the stratosphere and the ozone profiles deduced from these measurements independently confirmed the Umkehr observations. The first measurements above balloon altitudes were made by a UV spectrograph mounted on a rocket in 1946.

1.1.2 The Chemistry of Stratospheric Ozone

When Dobson and Harrison published their first report on their column ozone measurements in 1926, the formation mechanism of stratospheric ozone was still unclear. Today it is well established that stratospheric ozone is produced by the photolysis of molecular oxygen (O2) at ultraviolet wavelengths below 242 nm,

R1: O2 + hv -> 2O;

where hv denotes an ultraviolet photon. The atomic oxygen (O) produced in reaction R1 reacts rapidly with molecular oxygen to form ozone (O3)

R2: O + O2 + M -> O3 + M;

where M denotes a collision partner (N2 or O2) that is not affected by the reaction. Ozone is photolyzed rapidly

R3: O3 + hv -> O + O2:

The dissociative absorption of short-wave solar radiation by ozone is not only relevant for photochemistry but also constitutes the dominant source of heating in the stratosphere.

Through reactions R2 and R3, ozone and O establish a rapid photochemical equilibrium. Therefore, instead of considering ozone and atomic oxygen as separate species, the sum of ozone and O is often considered and referred to as "odd oxygen", or, alternatively, as the "odd oxygen family". It is denoted by the symbol Ox. This concept is useful because the sum of the family members is produced and destroyed much more slowly than the individual members. Throughout the stratosphere (up to about 50 km altitude) ozone constitutes the vast majority of odd oxygen. Through the reaction

R4: O + O3 -> 2O2

both an O atom and an ozone molecule are lost. Because O and ozone are in rapid photochemical equilibrium, the loss of one oxygen atom effectively implies the loss of an ozone molecule, i.e., R4 destroys two molecules of odd oxygen. Reactions R1-R4 were proposed by Chapman in 1930 as the first photochemical theory for the formation of ozone and are therefore referred to as the "Chapman reactions".

However, destruction of ozone by reaction R4 alone cannot explain the observed ozone abundances in the stratosphere. Today it is established that in the mid-latitudes and in the tropics the stratospheric ozone production through reaction R1 is largely balanced by the destruction in catalytic cycles of the form

[FORMULA OMITTED]

where the net reaction is identical to reaction R4. It is important to note that the catalyst X is not used up in the reaction cycle. The most important cycles of this type in the stratosphere involve reactive nitrogen (X = NO), originally proposed by Crutzen, and hydrogen (X = H, OH) radicals, originally proposed by Bates and Nicolet with a mesospheric focus and by Hampson with a stratospheric focus. Stolarski and Cicerone introduced the possibility of chlorine-catalyzed ozone loss (via cycle C1 with X = Cl).

Because of their great reactivity, the species X and XO are referred to as "active" and are commonly considered together as a so-called "chemical family", in analogy to the odd oxygen family Ox. Thus, NO + NO2 is referred to as active nitrogen (NOx), H + OH + HO2 as active hydrogen (HOx), and Cl + ClO as active chlorine (ClOx). Under polar winter conditions, the dimer of ClO, Cl2O2 is also part of the active chlorine family so that ClOx = Cl + ClO + 2 Cl2O2. However, most of the atmospheric nitrogen and chlorine not tied up in very long-lived gases (like N2O and chlorofluorocarbons) is not prevailing in active form as NOx or ClOx. Rather, most nitrogen and chlorine is bound in so-called "reservoir species"; HNO3 and N2O5 in the case of nitrogen and HCl, ClONO2, and HOCl in the case of chlorine. The sum of NOx + HNO3 + 2 N2O5 is referred to as NOy and the sum of ClOx + HCl + ClONO2 + HOCl as Cly or "total inorganic chlorine".

Because of the strong increase of O with altitude, the rates of the catalytic ozone loss cycles increase strongly between 25 and 40 km; the same is true for the rate of photolytic ozone production through reaction R1. The relative importance of the cycles for ozone loss varies considerably with altitude. Between 25 and 40 km the NOx cycle is the dominant ozone loss process, whereas above 45 km HOx-catalysed ozone loss dominates (Figure 1.1). Gas- phase reactions causing ozone loss through the ClOx cycle (which also depends on the stratospheric chlorine loading) peak at 40 km. The HOx catalyzed ozone loss dominates below about 25 km (Figure 1.1) where the concentration of O deceases strongly because a HOx-catalyzed cycle exists (C2) which only involves ozone and does not require O to be present.

[FORMULA OMITTED]

Two different chemical regimes exist for stratospheric ozone: the upper stratosphere and the lower stratosphere. In the upper stratosphere, the ozone distribution is largely determined by the balance between production from the photolysis of molecular oxygen (R1) and destruction via the catalytic cycles involving hydrogen, nitrogen and halogen radical species discussed above (see Section 1.1.2). In the upper stratosphere, a reduction in temperature slows the destruction rate of ozone (Figure 1.2). The rate of both the ozone destruction cycles and of ozone production via reaction R1 is substantially faster in the upper stratosphere than in the lower stratosphere; therefore, chemical equilibrium is reached rapidly in the upper stratosphere, which is not the case in the lower stratosphere (Figure 1.2).

In the lower stratosphere, in addition to gas-phase chemistry, reactions on aerosol and cloud particles (i.e., heterogeneous reactions) become important. Throughout the lower stratosphere, a layer of aerosol particles exists which consist of sulphuric acid and water, the so-called Junge layer. In the polar stratosphere, in winter, polar stratospheric clouds (PSCs) form. (Chapter 4). The distribution of the radicals (and the partitioning of the nitrogen, hydrogen and halogen species between radicals and the reservoir species which do not destroy ozone) are affected by heterogeneous chemistry. In the mid-latitudes, reactions on aerosol surfaces convert active nitrogen to the HNO3 reservoir, making mid-latitude ozone less vulnerable to active nitrogen (X = NO in cycle 1), but increase the efficiency of chlorine-catalysed (X = Cl in cycle 1) ozone loss. Heterogeneous reaction at low temperatures are of particular importance in the chemical mechanisms causing polar ozone loss (see Section 1.3.3 and Chapter 4).

1.1.3 The Distribution of Ozone in the Stratosphere

The distribution of ozone in the stratosphere is governed by three processes: photochemical production, photochemical destruction by catalytic cycles, and transport. Transport processes are typically divided into large-scale advection and mixing processes on smaller scales. The large-scale circulation of the stratosphere, with rising motion at low latitudes followed by poleward motion and descent at high latitudes, systematically transports ozone poleward and downward (Figure 1.3). This circulation in the stratosphere is referred to as the "Brewer–Dobson circulation" because such a circulation was originally suggested by Brewer based on water vapour measurements in the stratosphere and by Dobson et al. based on column ozone measurements. Because of the short photochemical lifetime of ozone in the upper stratosphere (see Section 1.1.2 above), the Brewer–Dobson circulation has little effect on the ozone distribution there (Figure 1.3). However, in the lower stratosphere, the photochemical lifetime of ozone is long (several months or longer) so that transport processes dominate the distribution of ozone.

Maximum ozone mixing ratios occur in the tropics between 30 km (Figure 1.4). In this altitude region, ozone is very short-lived and is essentially in photochemical equilibrium; that is, rapid photochemical loss is balanced by rapid production (see Figure 1.2). Under these conditions, transport timescales are slow compared to chemical timescales. Therefore, the altitude region 30–40 km in the deep tropics cannot serve as a source region for the extra-tropical stratosphere. The region where the ozone mixing ratios that are exported to the extra-tropics are determined, is the transition region between the area of chemical control and the area where ozone is controlled by transport.

Transport of ozone from the high latitudes poleward is important in the extra-tropical lower stratosphere, where ozone can accumulate on the time scale of a season. Variations in the ozone concentration in this region, between the tropopause and 20–25 km altitude, control changes in total column ozone abundance (Figure 1.3). Below about 25 km, ozone concentrations and mixing ratios in the extra-tropics are greater than in the high latitudes and (Figures 1.3 and 1.4). Thus, the ozone column at high and mid-latitudes are greater than in the tropics (Figure 1.5).

The Brewer–Dobson circulation is driven by planetary wave activity which is strongest in winter. Further, because of the asymmetric distribution of the topography and land-sea thermal contrasts that force planetary waves, planetary wave activity is stronger in the northern hemisphere than in the southern hemisphere. The stronger planetary wave activity in the northern hemisphere causes the Brewer–Dobson circulation to be stronger during the northern hemisphere winter than during the southern hemisphere winter. Therefore, ozone builds up in the extra-tropical lower stratosphere during winter and spring, with a greater build-up occurring in the northern hemisphere. The ozone then decays photochemically during the summer when transport is weaker and strong NOx-driven ozone loss occurs at the poles. The development of total ozone over the year as a function of latitude (Figure 1.5) reflects the seasonality of the Brewer–Dobson circulation.

1.2 Anthropogenic In?uence on the Stratospheric Ozone Layer

The first concern about the impact of anthropogenic activities on the ozone layer was formulated in the late 1950s: the possible impact of nuclear weapons tests on the ozone layer. Later, in the early 1970s, attention was focused on the effect a planned fleet of hundreds of supersonic aircraft might have on the stratospheric ozone layer. Research programs directed at assessing the impact of supersonic transport on stratospheric ozone greatly improved knowledge about stratospheric processes and paved the way for research on the question of the impact of anthropogenic halogen emissions on stratospheric ozone, which was first raised by Molina and Rowland in 1974.

1.2.1 Increase in Halogen Source Gases in the Atmosphere

Human activities result in the emission of a variety of halogen source gases that contain chlorine and bromine atoms and that have no natural sources. Important examples of anthropogenic halogen source gases are chlorofluorocarbons (CFCs), once used in almost all refrigeration and air- conditioning systems, and halons, used as fire-extinguishing agents. Because the halogen source gases have been identified as the major cause of the observed ozone depletion in the stratosphere they are also referred to as ozone-depleting substances. Production of most of these substances (and practically all for dispersive uses) has ceased because of the provisions of the "Montreal Protocol on substances that deplete the ozone layer", which was signed in 1987, and its subsequent amendments and adjustments. Nonetheless, emissions continue because halogen source gases are still present in existing equipment, chemical stockpiles, foams etc.; halogen source gases not yet released to the atmosphere are referred to as "banks".

Without the Montreal Protocol, production, consumption, and thus emission, of ozone-depleting substances would have continued with an annual growth rate of about 3%. As a result, the stratospheric halogen loading would have increased by 2030 by about a factor of ten compared to only natural sources (Figure 1.6). Model studies suggest that, had this happened, stratospheric ozone would have been strongly depleted globally with erythemal UV radiation more than doubling in the mid-latitudes in Northern hemisphere summer by the middle of this century (Section 1.4.2). However, both the Montreal Protocol itself and the later London amendments in 1990 would only have slowed the growth of the stratospheric halogen burden. Not until the amendments and adjustments signed in Copenhagen in 1992 do the projections of the future halogen loading indicate the decrease (Figure 1.6) required for a true recovery of the stratospheric ozone layer.

The most abundant naturally emitted chlorine source gas is methyl chloride. Methyl chloride is present in the troposphere in globally averaged concentrations of about 550 ppt and accounts for about 16% of the chlorine loading of the stratosphere today (Figure 1.7). At the end of the 21st century, when the abundance of anthropogenic chlorine source gases (e.g. CFCs) will have been greatly reduced as a consequence of the Montreal Protocol, methyl chloride is expected to account for a large fraction of the remaining stratospheric chlorine.

The emission of halogen source gases to the atmosphere ultimately leads to stratospheric ozone depletion. The first step is the photochemical breakdown of the source gases. Because of the great chemical stability of most source gases, this breakdown only occurs at appreciable rates in the upper stratosphere, where there is high-energy solar radiation. The chlorine atoms released in this way from the source gases are mostly converted to reservoir species, the most important being HCl and ClONO2. Because the reservoir species themselves do not cause ozone depletion, for ozone depletion to occur chlorine must be liberated from the reservoirs and converted into an active form. This occurs through gas-phase processes in the upper stratosphere and through heterogeneous chemistry in the polar lower stratosphere in winter.

An important measure of the potential for ozone depletion in the stratosphere due to the presence of halogen-containing (ozone-depleting) source gases in the stratosphere is the so-called "equivalent effective stratospheric chlorine". Equivalent effective stratospheric chlorine values are calculated by summing over adjusted amounts of all chlorine and bromine source gases. The adjustments are designed to account for the different rates of decomposition of the source gases and the greater per-atom effectiveness of bromine in depleting ozone compared to chlorine (Chapter 2).

In the latter half of the 20th century up until the 1990s, equivalent effective stratospheric chlorine values increased steadily and rapidly (Figure 1.7). As a result of the regulations of the Montreal Protocol, the long-term increase in equivalent effective stratospheric chlorine slowed, reached a peak in the mid-1990s and began to decrease thereafter. With a delay of a few years, the stratospheric abundance of halogen source gases follows the changes observed in the troposphere. The concentrations of all major chlorine-containing source gases decrease now in the troposphere, the longest-lived source gas, CFC-12, reached its peak approximately in 2000. Although measurements in the stratosphere are much sparser, declining growth rates of CFC-12 in the stratosphere have been reported. The start of a reduction in equivalent effective stratospheric chlorine values means that, as a result of the Montreal Protocol, the total stratospheric concentration of ozone-depleting halogen and thus the potential for stratospheric ozone depletion has begun to decrease.