Ozone Science: A Canadian Perspective on the Changing Ozone Layer

edited by D.I. Wardle, J. Kerr, C.T. McElroy, and D.R. Francis

The full version of this publication is available in English or French from Environment Canada, or click here.

The Montreal Protocol - the Achievement and the Challenge

The signing of the Montreal Protocol on September 16, 1987, was a remarkable and significant event in modern diplomatic history, one of those rare occasions when individual nations subordinated economic self-interest to the achievement of a common planetary goal. The event was even more remarkable when one considers that it was accomplished in spite of scientific uncertainties about detailed aspects of the depletion process and without immediate evidence of impacts on ecosystems and human health. That an agreement was eventually reached was due not only to an extraordinarily successful collaboration between scientists and policymakers but also to the enormous strides made by the international scientific community in expanding the boundaries of ozone science. The solidity of their achievement can be seen in the very real progress that has been made since 1987 in reducing emissions of ozone-depleting substances

Canada has been concerned about stratospheric ozone depletion since the issue was first raised by scientists in the 1960s and 1970s, and our interest in ozone science goes back even further, to the 1950s, when the first Canadian ozone monitoring programs were established. Over the past couple of decades our involvement in ozone science and contributions to it have been considerable. The Brewer ozone spectrophotometer, now the principal instrument for ground-based ozone measurements, was developed here. Our network of monitoring stations is one of the largest in the world, and, as the home for the World Ozone and Ultraviolet Radiation Data Centre, we are responsible for archiving ozone measurements from around the world. Canada is also proud to have been one of the original parties to the Montreal Protocol and to be among the many countries that have met or exceeded their obligations under the protocol and its amendments.

The present document provides a brief overview of the state of ozone science in 1997, on the 10th anniversary of the Montreal Protocol. Compiled by Canadian scientists, it draws on both Canadian and international research to outline our current understanding of ozone depletion and its effects. It also highlights Canadian research results and data, where appropriate, and emphasizes items of special Canadian concern, such as ozone depletion in the Arctic and the impacts of UV changes on forests and freshwater ecosystems.

The Road to Montreal

The possibility of anthropogenic interference with the ozone layer was raised as early as 1964, when John Hampson of the Canadian Armaments and Research Development Establishment noted the potential for ozone damage as a result of water vapour emissions from rockets and high-flying aircraft. Over the ensuing decade, proposals for the development of supersonic commercial aircraft that would fly in the lower stratosphere brought further attention to the issue, as did the debate over the environmental effects of nuclear weapons. In 1974, however, two articles were published that brought an entirely new dimension to the problem of ozone depletion.The first of these, published in the Canadian Journal of Chemistry by Richard Stolarski and Ralph Cicerone of the University of Michigan, described a process by which chlorine from rocket exhausts could catalyze the destruction of large amounts of ozone in the stratosphere over a period lasting many decades. Independently and almost simultaneously, two University of California researchers, Mario Molina and Sherwood Roland, voiced similar concerns about chlorine-catalyzed ozone loss but suggested the existence of a much larger source of anthropogenic chlorine in the stratosphere. In an article in Nature they argued that many widely used industrial chlorofluorocarbons (CFCs) had the potential to migrate into the stratosphere, where they would eventually break down as a result of exposure to intense ultraviolet radiation and release significant quantities of chlorine.

The combined implications of these articles were disturbing. Society appeared to face a choice between preserving the integrity of the ozone layer, which prevents biologically destructive levels of ultraviolet radiation from reaching the earth's surface, or preserving the economic benefits provided by CFCs, a valuable and otherwise benign group of chemicals that had become essential to a broad range of applications, including refrigeration and the manufacture of foams and electronic components. The potential risks to ecosystems and human health were large and unprecedented, while the costs of abandoning CFCs appeared considerable. In addition, although ozone destruction by chlorine catalysis was highly plausible, there was as yet no empirical evidence of ozone loss in the stratosphere. A further complication was added in 1975 when S.C. Wofsy, M.B. McElroy, and Y.I. Yung of Harvard University showed that bromine, used in fire-retarding halons, was also a potent destroyer of ozone. Not surprisingly, the issue ignited a major controversy, both within the scientific community and beyond.

University, government, and industry scientists in various centres around the world responded by intensifying their research activities. The immediate problem was to validate - or invalidate - the new theories of ozone destruction and assess their implications over a wide range of potential impacts. Given the considerable natural variability of ozone concentrations, both geographically and temporally, detecting the imprint of an anthropogenic disturbance and verifying the processes involved were not easy tasks. As researchers uncovered new information about the chemistry and dynamics of the middle atmosphere, the problem became more complex and uncertainties increased. Modelling of processes affecting ozone amounts proved extremely difficult, partly because of limitations of computer power and partly because of uncertainties about reaction rates and other critical aspects of the depletion mechanisms. Consequently, initial projections of depletion by CFCs showed little consistency. Nevertheless, by the mid-1980s, knowledge of ozone-related processes had expanded considerably. When the United Nations Environment Programme (UNEP) and the World Meteorological Organization (WMO) released their first major international ozone assessment in 1986, the report was able not only to provide a comprehensive analysis of the threat from CFCs 11 and 12 but also to identify additional substances that had the potential to deplete stratospheric ozone. In addition, it drew attention to the fact that many of these substances were extremely powerful greenhouse gases whose presence in the atmosphere also had serious implications for global warming and climate change.

At the national level, reaction to the threat of ozone depletion varied considerably. It was strongest in those countries where media interest and vocal environmental movements ensured a place for the issue on the political agenda. Fact-finding commissions were established and research activities intensified in a number of these countries to provide governments with more information about the nature and implications of the problem. By 1978 a few countries had concluded that it would be prudent to curtail the least essential uses of CFCs until there was greater certainty about the risks involved in their use. Thus, the United States banned the use of CFCs in nonessential aerosol sprays in March of that year, and Canada, Norway, and Sweden shortly followed suit. The European Community, after initially rejecting Dutch and later German proposals for CFC restrictions, agreed in 1980 to a more modest 30% cutback in aerosol use.

The ozone issue, however, was fundamentally an international problem, and it could only be resolved by international action. The initiative for promoting this action was taken at a relatively early date by the United Nations Environment Programme, under the leadership of its executive director at the time, Mostafa Tolba. In 1976 UNEP had called for an international conference to discuss an international response to the ozone issue. The conference, held in Washington in March 1977, drafted a "World Plan of Action on the Ozone Layer," which gave UNEP the responsibility for promoting and coordinating international research and data gathering activities. At the same time, a Coordinating Committee on the Ozone Layer, under UNEP direction, was formed to oversee periodic international assessments of the depletion problem.

The possibility of establishing international controls over the production and use of CFCs was first raised a month later at another international meeting in Washington but attracted insufficient support at the time and on subsequent occasions over the next few years. In April 1981, however, UNEP's Governing Council authorized the organization to begin working towards an agreement to protect the ozone layer. The first step in this process was taken in January 1982, when 24 countries met in Stockholm and agreed to launch an "Ad Hoc Working Group of Legal and Technical Experts for the Preparation of a Global Framework Convention for the Protection of the Ozone Layer." In 1983, the so-called Toronto Group (named after the site of its first meeting and consisting of Canada, Finland, Norway, Sweden, Switzerland, and later the United States) recommended a global ban on nonessential uses of CFC aerosol sprays and proposed that a separate regulatory protocol be developed and adopted simultaneously with the framework convention.

Efforts to complete a framework agreement came to fruition in March 1985 with the signing of the Vienna Convention for the Protection of the Ozone Layer. The participating countries agreed to take measures to protect the ozone layer (although these were not spelled out) and made arrangements for international cooperation in the areas of research, monitoring, and exchange of data on the state of the ozone layer and emissions and concentrations of CFCs and other chemicals. The convention was not accompanied by a regulatory protocol, but under a separate resolution UNEP was authorized to begin negotiations on a legally binding protocol that would be ready in 1987.

Although Vienna was an important milestone, the international consensus needed to support an effective control protocol was still lacking. It was therefore decided to convene two workshops in 1986 to review some of the key economic and scientific issues. Working informally as private citizens rather than as members of national delegations, experts from the UN, governments, industry, universities, and environmental groups met first in Rome and later in Leesburg, Virginia. Though many points of contention remained unresolved, the meetings nevertheless succeeded in building a broader basis of understanding for the negotiations to follow. One of the more important ideas to emerge from these discussions was the concept of an interim protocol - one that did not have to provide a definitive solution to all outstanding problems but allowed for periodic reassessment and revision in the light of changing facts and expanding scientific knowledge.

The actual negotiation of the protocol began in Geneva in December 1986. Subsequent meetings in Vienna and again in Geneva helped to narrow the outstanding differences, but when the delegates convened in Montreal on September 8, 1987, important disagreements remained over such basic issues as the chemicals to be controlled, the use of production or consumption as the basis for restrictions, the extent of the controls and the timing of their implementation, the choice of a base year, and arrangements for developing countries. It was only after further intensive negotiation that agreement was finally reached on the 16th.

The conclusion of such an important and unprecedented agreement owes much to the skill and persistence of those who negotiated it, but a number of other important factors contributed to this achievement. The role of the international scientific community was particularly vital. Although scientists were unable to eliminate many of the uncertainties that surrounded and still surround the stratospheric ozone issue, they were successful in reducing the range of uncertainty and in building a compelling case for action. Chance played a role as well, most notably with the discovery of the Antarctic ozone hole in 1985. The ozone hole did not confirm existing theories about the destruction of the ozone layer - it only raised new questions - but it did create a greater awareness among opinion leaders and the public that something serious was happening to the atmosphere and that precautionary action was necessary. An agreement also became much more likely after 1986, when the American chemical industry, one of the larger producers of CFCs in the world, abandoned its opposition to controls. Finally, it was the very flexibility of the agreement that made it acceptable to many of the parties. It did not attempt the impossible task of solving so complex a problem in one step. Instead, it set up a mechanism for continuing review, so that policy could be refined in the light of new realities and the best available information. Indeed, one of its most important achievements was that it created a mechanism for continuing action. As Mostafa Tolba pointed out, the Montreal Protocol was a starting point, "the beginning of the real work to come."

Mileposts: 1987-1997

The past decade has seen a number of very significant advances in ozone science, both in terms of improvements in research and monitoring capabilities and actual advances in our understanding of ozone depletion and its effects on radiation at the earth's surface. One of the most important developments occurred in the late 1980s, when intensive research efforts unravelled the mystery of the Antarctic ozone hole and uncovered the role of polar stratospheric clouds (PSCs) and hetereogeneous chemistry in its genesis. By 1990, important evidence about ozone trends in other parts of the world was also becoming available, and depletions of about 5% per decade were detected over the northern midlatitudes. After the eruption of Mount Pinatubo in 1991, it became apparent as well that sulphate aerosols injected into the stratosphere by volcanic activity could cause significant depletion. More recently, there has been evidence that significant depletions have been occurring in the Arctic as a result of heterogeneous reactions on PSCs. In 1993, the expected link between ozone depletion and increases in UV radiation at the surface was finally confirmed through the analysis of spectral data. The study, by Environment Canada scientists, has recently been extended to cover an 11-year period ending in 1996. It shows a positive trend of approximately 1% per year in the summer radiation at 300 nm.

Other noteworthy scientific advances of the past decade include:

• a greater understanding of the transport of ozone and other trace gases as a result of the identification of the role of small-scale filamentary structures in mixing processes in the upper troposphere and lower stratosphere
• the application of new techniques for measuring ozone and other atmospheric constituents from the ground, balloons, and satellites
• advances in chemical modelling, including the ability to incorporate chemical processes in general circulation models and forecast models
• a substantial increase in the number of stations making spectral UV measurements and in the length of the data record
• an evaluation and, in some cases, verification by measurements of the effects of CFCs and other ozone-depleting substances on radiative forcing and climate change
• evidence of a possible link between climate change and ozone depletion in the Arctic
• better understanding of the effects of clouds, sulphur dioxide, and albedo on UV irradiance

Improvements in our understanding of depletion processes as well as further evidence of ozone loss has precipitated a significant tightening and extension of the protocol's regulatory regime. The original agreement had required a 50% reduction in CFC use by mid-1998 and a freeze of halon consumption at 1986 levels by 1992, but amendments passed at meetings in London (1990) and Copenhagen (1992) targeted these substances for virtual elimination and then advanced the phase-out dates to the end of 1994 for halons and the end of 1995 for CFCs. In addition, carbon tetrachloride and methyl chloroform, HCFCs and HBFCs (used as substitutes for CFCs and halons), and methyl bromide were brought under regulatory control. Carbon tetrachloride, methyl chloroform, and HBFCs were scheduled for elimination by the end of 1995, while consumption of HCFCs was scheduled to be eliminated by 2030. Further meetings in Vienna (1995) and Costa Rica (1996) led to reductions in the use of methyl bromide and agreement to end its use entirely by 2010.

Present Trends

The greatest and most dramatic ozone loss has occurred during early spring over the Antarctic, where total ozone values have dropped by more than 65% since 1975. Ozone losses over the Arctic for this period have been less severe, because of differences in circulation patterns, but are still in the area of 12%. In the midlatitudes, ozone has been declining at a rate of about 5% per decade.

The accompanying maps show the differences between ozone amounts over Canada at the time of the signing of the Montreal Protocol in 1987 and a decade later in 1997. The decline in ozone values between the two six-month periods is greatest (as much as 12%) in the high Arctic and lowest (about 3%) over the southeast. However, in 1996 Arctic ozone values were briefly about 30% below normal, while during the spring of 1997 they were as much as 45% below normal over the high Arctic and about 7% below normal over the midlatitude regions of the country. The exceptionally large depletion in the Arctic during the past spring is likely the result of unusual upper wind and temperature patterns that may, in turn, be related to the radiative effects of increased concentrations of greenhouse gases.

Meanwhile, as a result of the Montreal Protocol, the rate at which atmospheric concentrations of CFCs have increased has slowed noticeably since about 1990 and, in the case of CFC-11, concentrations have actually begun to decrease. As concentrations of CFCs and other ozone-depleting substances in the stratosphere decline, concentrations of chlorine and bromine should follow suit. If the provisions of the Montreal Protocol are fully obeyed by all the parties, the current stratospheric chlorine concentration of 3.5 parts per billion (ppb) is expected to peak within the next few years and then decrease gradually, returning to its natural level of 1.0 ppb some time after 2100.

Given the expected decreases in concentrations of ozone-depleting substances and stratospheric chlorine and bromine, when can we expect the ozone layer to recover, and when will we be able to detect that a recovery is under way? The accompanying graph shows how ozone has decreased from 1965 to 1996. It then compares, in a simplified fashion, what might happen if our current assumptions about ozone depletion are correct, and the Montreal Protocol and its amendments are fully implemented, with an alternative scenario in which concentrations of ozone-depleting substances remain unchanged at 1997 levels. Although the scientific basis for the graph is minimal, it does illustrate that evidence of a clear trend towards increasing ozone amounts may not emerge until after 2005 or 2010. In reality, however, such evidence may be delayed even further because compliance with the protocol may not be complete and there are still uncertainties in our understanding of the science. These uncertainties also make it difficult to predict confidently when ozone concentrations will finally return to natural levels. As the recent and unexpectedly large Arctic depletions indicate, the ozone issue can still produce surprises.

The Future Agenda

Reliable forecasts of a future ozone recovery ultimately depend on the comprehensive and accurate modelling of atmospheric chemistry and dynamics. At the present time, however, our models generally underestimate the amount of ozone depletion that has actually occurred and cannot accurately simulate all aspects of ozone distribution with altitude, location, and time of year. These limitations suggest that there may be gaps in our knowledge of ozone chemistry and atmospheric dynamics. They may also point to the presence of other as yet unidentified ozone-depleting substances in the stratosphere. In addition, the unexpected extent of recent ozone depletions in the Arctic indicate a need for further study of the Arctic atmosphere and, in particular, of the effects on it of increased concentrations of greenhouse gases. It is now fairly certain that increased concentrations of these gases have caused stratospheric cooling, and a cooler Arctic stratosphere would generally offer a more favourable environment for ozone destruction. Finally, although many of the effects of enhanced UV radiation on biological systems are reasonably well known, our knowledge is far from complete, and it is possible that additional damage mechanisms remain to be identified. Also, the implications of ecological interactions and other environmental stresses in connection with UV effects have yet to be explored in detail.

In spite of the truly remarkable progress that has been made in ozone science over the past decade, the problem of ozone depletion has not yet been solved. Our challenge for the next decade, therefore, is to fill in these gaps in our knowledge of chlorine- and bromine-induced depletion while expanding our understanding of the effect of other factors on ozone amounts. At the same time, we must continue our efforts to monitor and detect the recovery of the ozone layer and expand our work on the biological effects of enhanced UV radiation.

The successful implementation of this agenda over the next 10 years will give us a much better basis for ensuring the ultimate recovery of the ozone layer and for developing effective responses for the protection of human health and ecosystem vitality in the meantime. It will also enhance our understanding of the complex interactions of radiative, dynamical, and photochemical processes that drive the atmosphere and thus give us better tools with which to predict how the atmosphere will respond to other perturbations that may occur in the future.

How Has the Ozone Layer Over Canada Changed Since 1987?

The upper map shows average ozone levels over Canada for January-June 1987, while the middle map shows the same information for January-June 1997. The percentage differences between the two periods are plotted in the lower map, which shows declines ranging from 3% over southern Canada and about 4% over the Prairies up to 12% in the high Arctic. Although these differences are consistent with decadal trends, a comparison of other years, such as 1985 and 1995, would yield different results. The maps have been constructed from both ground-based and satellite measurements.

When Might the Recovery of the Ozone Layer Be Detectable?

The graph shows the global total ozone record from 1964 to 1996 and two hypothetical projections based on different scenarios. Branch (a), which expresses the best case, assumes that the Montreal Protocol and its amendments will be fully implemented and that concentrations of chlorine and bromine will decline according to the projections contained in the 1994 UNEP Assessment. It also assumes that ozone depletion has been due only to known ozone-depleting substances. Branch (b) is based on the assumption that concentrations of all ozone-depleting substances remain at their 1997 levels.

Although these projections are relatively crude, they do illustrate that several years will be needed to detect the start of any recovery and several more to estimate its extent. The actual recovery scenario is likely to be somewhere between these two cases because the protocol may not be fully adhered to by all parties and other factors or as yet unidentified substances may be contributing to depletion.

The graph is derived from a very simple statistical model of seasonally dependent, chlorine-induced depletion, modified by the addition of random noise and the quasi-biennial and solar cycles in ozone.

Summary

Ozone Science: A Canadian Perspective on the Changing Ozone Layer

The tenth anniversary of the 1987 signing of the Montreal Protocol is an event to be celebrated because the protocol has been the basis of a successful effort by the international community to control and greatly reduce the emission of ozone depleting substances into the atmosphere. A serious threat to the global environment has been addressed and appropriate measures have been taken. This report provides a brief overview of the state of ozone science 10 years after the signing. Compiled by Canadian scientists, it draws on both Canadian and international research to outline our current understanding of ozone depletion and its effects. It also highlights Canadian results and data, where appropriate, and emphasizes items of special Canadian concern, such as ozone depletion in the Arctic and the impacts of UV changes on forests and freshwater ecosystems.

THE DISTRIBUTION OF OZONE AND OZONE-DEPLETING SUBSTANCES IN THE ATMOSPHERE AND OBSERVED CHANGES.

The main features of the climatology of total ozone were discovered by G.M.B Dobson by 1930. These were the dependence on latitude and season and the day-to-day changes that are associated with meteorological conditions. A more extensive climatology extending from the ground to all levels in the stratosphere is now available as a result of the use of many new techniques. These measure the distribution of ozone with altitude as well as the total amount both from the ground and from a variety of platforms, including balloons, aircraft, rockets, and satellites. The severe depletion in the Antarctic during the spring, known as the ozone hole, was discovered in 1985, and a few years later midlatitude depletion was recognized. A significant and developing depletion in the Arctic spring has also been observed during the last few years. Atmospheric concentrations of ozone-depleting substances have been measured for some decades and now show definitive evidence of the effect of the Montreal Protocol.

• Canada began total ozone measurements at five stations in the early 1960s and has been operating twelve ground-based ozone observing stations equipped with Brewer spectrophotometers since 1992. The total ozone data are used for trend studies, for forecasting ozone and UV radiation, and for characterization and calibration of satellite measurements.
• The four-station Canadian ozonesonde network was initiated in the mid-1960s. Currently sondes are flown at least weekly from six locations. All the sonde and ground-based data are available from the World Ozone and Ultraviolet Radiation Data Centre (WOUDC) which Environment Canada operates for the WMO.
• The Stratospheric Ozone Observatory at Eureka ( 80°N, 86°W), which is part of the WMO/UNEP Network for Detection of Stratospheric Change(NDSC), was opened in 1992. It houses lidars for measuring ozone and aerosols, Fourier transform spectrometers, and other spectrometers for measuring stratospheric composition. The Ontario Institute for Space and Terrestrial Physics and the Japanese Meteorological Agency and Communications Research Laboratory are partners with Environment Canada in conducting these experiments.
• Annual average ozone values over Canadian stations have declined by about 6% from pre-1980 averages. The decline is greatest in spring and least in the fall. Sonde measurements show that the main loss is in the lower stratosphere.
• Sonde measurements over Canada have shown that free (i.e., unpolluted) tropospheric air has suffered a similar loss of ozone to that in the stratosphere during the past 15 years. However, European sonde data show some increases during the 1980s.
• The ozone values in 1993 were lower than in any other year over southern Canada and over the midlatitudes in general. The decline, which is attributed to the effects of the eruption of Mt. Pinatubo in 1991, occurred in the lower stratosphere and, at least over Canada, throughout the troposphere.
• Ground and balloon-based measurements over the Canadian Arctic from the Eureka NDSC observatory, Resolute, and Alert, as well as satellite measurements, show the development of strong depletion during the past few years, especially in the spring of 1997. However, the lowest Arctic ozone values are still much higher than the low values in the Antarctic.
• The Canadian Meteorological Centre's daily forecast of the total ozone field over Canada is based on measurements of total ozone by the twelve Canadian Brewer spectrophotometers and the output from the Numerical Weather Prediction model. The forecast, based on statistical relationships, has a very low error.
• Measurements show that atmospheric concentrations of CFC-11 at Alert and Point Barrow have begun to decline and that the rate of increase in CFC-12 is much reduced.

THE INFLUENCE OF DYNAMICAL PROCESSES ON OZONE ABUNDANCE.

Transport of ozone by the Brewer-Dobson circulation from its main source region in the equatorial midstratosphere to its regions of greatest abundance, in the lower stratosphere at middle and high latitudes, is a key determining factor in establishing the observed distribution of ozone. The height of the tropopause is also a determining factor for the total ozone amount, since the troposphere is ozone poor while the stratosphere is relatively ozone rich. These are dynamical factors that affect the ozone directly. Different scales of mixing affect concentrations of ozone and other constituents; modelling of chemical reactions may be inaccurate if the spatial resolution of the model is not fine enough to capture the smaller scale effects.

Dynamics act indirectly on the chemistry of ozone by causing changes in temperature, which affect the rates of most chemical reactions (some more than others). Of particular importance is the effect of temperature on the formation of polar stratospheric clouds, and the enhancement that these cause of chlorine concentration and ozone destruction. Understanding of atmospheric dynamics has developed through advances in numerical modelling coupled with increasingly detailed measurements of ozone and other trace constituents, but a number of problems remain.

• The spatial structure of the tropopause directly affects the total abundance and latitudinal distribution of ozone, but our understanding of how this structure is established is mostly qualitative and our ability to characterize, let alone predict, interannual variability and trends is extremely limited.
• Although we have a basic understanding of the origin of the Brewer-Dobson circulation and its seasonal cycle, significant theoretical gaps remain, especially with respect to the tropics. There is considerable potential for interannual variability and decadal-scale trends in the Brewer­Dobson circulation associated with corresponding variability in the driving mechanisms. Such variability places a major limitation on our predictive capability.
• It has recently become evident that the transport and mixing characteristics of the tropics are qualitatively different from those of the midlatitudes. To a first approximation, the tropical stratosphere regime comprises a large­scale systematic upwelling modified by mixing across a permeable subtropical transport barrier. Adequate characterization is not yet possible because of the difficulty in determining the upwelling rate and the current lack of accurate wind measurements in the tropics.
• The lowest part of the tropical stratosphere appears to be a particularly complex region of the atmosphere which plays a critical role in determining the global structure of stratospheric transport and mixing , yet it is a region that remains poorly understood. For example, most current conceptual models completely ignore the effect of longitudinal asymmetries such as the monsoon circulation.
• There is now a good understanding of why wintertime polar ozone depletion is much less and more variable from year­to­year in the Arctic than in the Antarctic. Quantitative evaluation is limited by uncertainties in the cause and magnitude of the polar downwelling (which affects polar temperatures) and in our understanding of planetary wave drag.
• It has been shown recently that simulation of the amount of ozone destruction in the Arctic winter is highly sensitive to an adequate representation of small­scale structure in chemical fields. Similar considerations presumably apply to midlatitudes and represent a major challenge for quantitative modelling.
• The laminae seen in ozonesonde profiles have been clearly linked to filamentation processes associated with planetary wave breaking at the edge of the polar vortex. Dynamical lifetimes of these filaments/laminae are estimated to be at least two weeks, comparable to estimated chemical lifetimes of ozone filaments in midlatitudes.

OZONE CHEMISTRY: SIMULATION AND DEPLETION

Stratospheric ozone originates almost entirely as a result of the action of sunlight on normal diatomic oxygen molecules, mostly at altitudes above about 25 km and in the tropics. Ozone is also produced through reactions involving volatile organic compounds(VOCs) and nitrogen oxides, but these are mainly of significance in the troposphere. The reaction of atomic oxygen with ozone was once thought to be the only ozone loss mechanism in the stratosphere. However, various anthropogenic and natural long-lived gases such as N₂O, CFCs, halons, and methyl bromide and chloride are dissociated in the stratosphere into products that are effective catalytic destroyers of ozone molecules. These reactive molecules include NO and NO₂, Cl and ClO, OH and HO₂, and Br and BrO. Their concentrations are limited by the formation of reservoir species, such as HNO₃, ClONO₂, and HCl, that may dissociate, but not necessarily quickly, to give back the reactive molecules. The relative concentrations of the reactive molecules and the reservoir species which do not destroy ozone determine the effectiveness of ozone destruction. They are strongly influenced by polar stratospheric clouds (PSCs), which facilitate the conversion of reservoir species to reactive halogens as, to a lesser degree, do sulphate aerosols in the midlatitudes. Adequately representing the properties and behavior of PSCs and sulphate aerosols is therefore an important challenge for simulation models. Types of models that are in use range from the zero-dimension, chemistry-only, box models to modified atmospheric general circulation and numerical weather prediction models (AGCMs and NWPs) that incorporate interactive ozone photochemistry. Models serve as tools to diagnose the processes that are occurring in the atmosphere and can be used to make predictions, provided that they are validated and their characteristics are consistent with the known properties of the atmosphere.

• The development of techniques and computer hardware during the past decade has resulted in the ability to carry chemical reaction sets that were once tractable only in box or 1D models into higher dimensional models.

• 3D chemical transport models (CTMs) are models in which the chemical constituents are carried by winds that are generated in AGCMs or NWP models. They do not include feedback from chemical composition to the dynamics. Several advanced AGCMs, including the Canadian Middle Atmosphere Model (CMAM), now incorporate chemistry interactively, so that the feedback from chemistry to dynamics is included.

• NWP models, which are similar to ACGMs but have finer resolution, can also now include interactive chemistry. The Canadian NWP, the Spectral Finite Element Model (SEF), is now carrying gas phase chemistry and heterogeneous reactions are being added. The SEF model, with the addition of new data assimilation methods that are being developed for chemical constituents, is the ideal tool for simulating the Arctic spring stratosphere.

• Current models reproduce the behaviour of atmospheric ozone quite well, but loss rates in the upper stratosphere are overestimated. Also, discrepancies between modelled and observed ratios of active to reservoir chlorine have not yet been resolved.

• Most current models underestimate the ozone loss in the lower stratosphere, which is the major component of the change in total ozone. Some models that give reasonable simulations of the observed loss in total ozone do not reproduce the vertical profile satisfactorily.

• The eruption of Mt. Pinatubo provided a unique opportunity to study the role of sulphate aerosols in ozone chemistry. It is believed that the thermal effects of the enhanced aerosol concentrations may have altered transport patterns so as to enhance ozone destruction and that the conversion of N₂O₅ to HNO₃ on the aerosol surfaces perturbed the chemistry so as to enhance depletion by halogens.

• Simulation of the Antarctic spring depletion by current models appears satisfactory and is somewhat insensitive to the details of the heterogeneous conversion chemistry; this insensitivity is related to the ozone loss being nearly complete in substantial parts of the stratosphere.

• Simulation of the Arctic spring depletion is less satisfactory, the calculated loss being approximately 60% of the observed value according to one recent simulation. The discrepancy may be due to the model's inability to capture small-scale mixing effects. As well, the winter stratospheric temperatures that are used in these simulations may be in error by small amounts that are nevertheless significant for PSC formation.

• Aircraft that fly in the stratosphere and upper troposphere can affect the chemistry there through their emission of nitrogen oxides (NO_x) and water. Both of these will tend to increase the prevalence of PSCs and so enhance ozone destruction. NO_x itself can increase both the loss rate and the production of ozone, and adequate estimation of these competing processes requires characterization of the transport between and within the stratosphere and troposphere of a quality that is not yet available.

• The increased UV-B (290-315 nm) radiation in the troposphere resulting from the depletion of stratospheric ozone could increase tropospheric concentrations of OH. Such an increase would, in turn, increase removal rates for several greenhouse gases, such as methane, CO, and the CFCs. The effects of increased UV-B on tropospheric ozone are complex and uncertain. Ozone concentrations might increase in populated areas with higher concentrations of VOCs and NO_x while decreasing in the less polluted background air.

• The cooling of stratospheric temperatures that is associated with global warming may make PSCs more numerous and thereby counteract the expected recovery of ozone as the amount of stratospheric halogens declines. Evidence of decreasing trends in stratospheric temperatures has been questioned, but in the past few Arctic winters ozone values and stratospheric temperatures have been particularly low. The loss of ozone may itself have augmented the low temperatures and caused the vortex breakup to be delayed.

• Important factors that are not yet treated effectively in atmospheric modelling are: gravity waves and their effect on the middle atmosphere circulation, detailed transport processes in the upper troposphere and lower stratosphere, microphysical aerosol processes that are involved in heterogeneous chemistry in both the troposphere and the stratosphere, and the effects of aerosols on climate through radiation and the formation of clouds.

• It is assumed that, as stratospheric chlorine begins to decline in the next decade, ozone amounts will begin to recover, albeit slowly, shortly thereafter. It is imperative to continue monitoring to determine whether this scenario unfolds in this manner.

THE INFLUENCE OF OZONE AND OTHER FACTORS ON SURFACE RADIATION.

Atmospheric ozone is important as it interacts both with solar radiation and with the thermal radiation that is emitted by the earth's surface and the atmosphere. It screens the surface from solar UV radiation that would otherwise prove harmful to living organisms, causes heating in the stratosphere, and acts as a "greenhouse" gas. The depletion of ozone during the past two decades has allowed more UV-B radiation to reach the ground than would otherwise have happened. The extent of this increase and the variability of ground-level UV radiation and its causes are the focus of current research, since estimation of the biological effects of ozone depletion may require knowledge of both these factors. The effects of UV on people are largely determined by behaviour, and public awareness programs can help people avoid excessive exposure. Ozone-depleting substances are also greenhouse gases, and their presence in the atmosphere and the recent changes in atmospheric ozone affect global warming.

• Changes in the UV are usually expressed as trends in monthly or seasonal averages, although certain biological effects may depend more on peak values during critical stages in life cycles than on the total irradiation. Also, for biological effects that are more complex than skin reddening, there may be problems in representing the effectiveness of UV radiation by simple action-spectra weighted integrals.

• About 100 station years of UV irradiance data can be accessed from the World Ozone and Ultraviolet Radiation Data Centre (WOUDC). Most of the stations have more than five years of data.

• The dependence of clear-sky UV irradiance on ozone is well established. When the sky is overcast, the irradiance throughout the spectral range 300-325 nm is reduced by approximately the same amount at all wavelengths, provided that the transmision is greater than about 15%. Consequently, the ozone amount can be easily derived from the spectrum. The accuracy of one technique developed for this purpose is 3% for daily averages of total ozone.

• With heavily overcast skies, the transmitted spectrum carries the signature of a greater amount of ozone. This is consistent with the enhancement of ozone absorption in the cloud by multiple scattering. Apparent ozone amounts up to 1000 DU have been observed when the cloud transmission is down to about 1%.

• Sulphur dioxide absorbs strongly in the UV-B region. High amounts of absorption by SO₂ are often evident in the UV irradiance spectra from Kagoshima in Japan. This is caused by the plume from the nearby volcano, Sakurajima. The SO₂ absorption associated with the plume can reduce the UV Index by as much as 25%. However, throughout the year the average reduction is only 1-2%. No other stations in the WOUDC exhibit significant absorption by SO₂.

• The effect of ground reflectance on UV radiation has been investigated at seven Canadian locations by comparing data with and without snow on the ground. It was found that the enhancement due to snow cover varies from 8% at Halifax to 39% at Resolute. The range of results reflects differences in terrain around these stations.

• An eleven-year time series of the accumulated spectral irradiation during summer has been derived from spectral irradiance measurements at Toronto that were started in 1986. The irradiation at 324 nm shows a small downward trend, while at 300 nm there is an upward trend of about 1% per year. The ratio of the irradiation at 300 nm to that at 324 nm closely follows the average summer ozone values. This suggests that the main cause of the upward trend at 300 nm is ozone depletion.

• There has been considerable progress in simulating UV irradiance from related atmospheric variables and satellite measurements. These developments should be useful in extending the current climatology of UV and in constructing past climatology.

• The downward thermal radiation at the earth's surface from ozone and several ozone depleting gases has been measured in southern Canada. The results confirm that ozone-depleting substances are making a significant contribution to global warming.

• The UV Index program, which produces daily forecasts of UV radiation and is designed to help people avoid excessive exposure, was started in Canada in 1992. Similar programs are now operated in several other countries. The Ozone Watch program, also started in 1992, provides Canadians with weekly information on the ozone layer.

UV-B EFFECTS

Increased UV flux to the earth's surface has properly heightened concerns over the impact on human health, given that it is well­known to cause a variety of skin cancers, eye conditions, and other problems. Although the potential health impacts may be serious, many aspects of the epidemiology and pathology of UV exposure are relatively well understood; however, the situation with respect to non­human biota is more complex. Ecosytems and biomes have adapted over long periods to a particular range of UV­B, and any change in the regimen must a priori influence their stability, geographical range, and possibly survival. Whether changes wrought by an increase in UV flux will be profound or subtle, minor or dramatic, beneficial or inconsequential depends on many factors, most of which are poorly understood. There is, therefore, a pressing need to measure, monitor, and understand the effects of enhanced UV­B irradiance on biological communities if we are to be able to predict the impacts that will occur in the next 30-40 years and manage the consequences.

Human Health

Within Canada, most UV effects research has centred on skin cancer and eye damage, but only minimal effort has been devoted to studies of immune suppression and infectious diseases.

• Recent Australian and Canadian research has shown that members of lower socioeconomic occupations with substantial outdoor exposure have a lower risk of malignant melanoma and basal cell carcinoma than those engaged in indoor, professional, managerial, and technical occupations, who receive only intermittent exposure to the sun.
• Squamous cell carcinoma appears to be related to cumulative exposure.
• Acquired (non-congenital) skin moles (nevi) are an indicator of elevated risk for malignant melanoma. Australian and Canadian researchers have shown a direct relationship between mole density and solar UV exposure. Generally, the risk of melanoma increases with mole density.
• Persons with light skin colour, red hair, and a propensity to burn in the sun are at greater risk of basal and squamous cell carcinoma.
• UV radiation may suppress the human immune response, increasing the risk of skin cancer and bacterial and viral infection (including the re-activating of smallpox and herpes lesions).
• Acute, long­term, and chronic eye conditions may become more prevalent if UV exposure increases.
• Avoidance of the sun, use of sun screens, and wearing UV­protective glasses are important prophylactic measures.

Ecosystems

Alterations in UV irradiance can affect primary production in all ecosystems, terrestrial and aquatic, natural, managed or exploited with a potential cascade of effects. Current understanding of these processes does not enable confident prediction of the impacts.

There has been little systematic research on the impact of enhanced levels of UV-B on Canadian flora and fauna. A few studies of some agricultural and commercial forest species provide limited insight into the problem, but it is difficult to extrapolate from these to predict impacts on whole ecosystems.

It is important that the effects of both chronic increases in irradiance leading to cumulative doses, and also episodic peaks or events that may coincide with critically vulnerable stages in life cycles, be evaluated. Further, the influence of ozone depletion must also be considered in conjunction with the effects of other stressors such as global warming, acidification, and the presence of toxic chemicals, making it essential that UV-B impact studies be integrated with existing ecological research, monitoring, and assessment programs.

Agriculture

• Ultraviolet radiation damages DNA, cell membranes, and organelles (e.g., chloroplasts). Plants have a capacity to repair damaged DNA, and some can protect themselves by synthesizing UV­absorbing pigments and by modifying key metabolic enzymes. Harm occurs when the radiation dose causes damage beyond a plant's capacity for repair and protection.
• Crop damage is manifest as a decline in yield, reduced fertility with fall in seed and fruit production, drop in marketable quality, and ecological effects such as changes in crop-weed interactions and pasture mixtures.
• While the extrapolation from controlled environments to field conditions remains an issue, research is consistent in identifying that individual varieties differ in their sensitivity to UV-B, as measured by changes in photosynthesis, growth, yield, and reproduction. Most Canadian crops have low or intermediate sensitivity to moderate increases in UV-B radiation. Impacts on forage and vegetable crops have been equivocal. Of over 100 varieties of 12 important crop species, 40% were unaffected by UV-B equivalent to a 20% decrease in the ozone layer and 60% were affected in some fashion. Soybean, tomato, and canola losses may be expected, possibly totalling hundreds of millions of dollars annually. Maize does not appear to be vulnerable to anticipated increases in UV.
• UV-B may accelerate the rate of decomposition of straw and chaff, with potentially beneficial effects in arid environments and under minimum-till conditions.

Forests

• Although forests are of major economic, social, and natural importance, little is known about their susceptibility to enhanced UV-B radiation.
• Trees are long-lived and will be exposed to increased UV-B over decades. Short-term effects have been reported, and there is some indication of cumulative, chronic impacts. Multiyear experiments with long-lived species are desirable.
• The impacts of UV radiation appear to be less serious on species native to high elevations and adapted to greater irradiance.
• Impacts of enhanced UV radiation should be considered in conjunction with the effects of other stressors such as climate change and acidification. Studies of the interactive effects of UV-B and CO₂ enrichment on the growth and physiology of conifer seedlings suggest that future conifer seedling growth and competitive ability will be altered by the changing environment.
• The only field study to have been done so far on UV-B effects on trees in North America attributed sun-scalding of white pine foliage in Ohio and Ontario to elevated ambient UV-B levels. Recent research has also shown that epicuticular wax chemical composition in certain conifer species is affected by UV-B exposure in a manner that inhibits photosynthesis.
• The 50-year window of significant ozone depletion has sufficient potential to affect Canadian forest productivity on a large scale, with far-reaching consequences that are not yet apparent.

Freshwater and Wetland Ecosystems

• Shallow freshwater ecosystems are particularly vulnerable to enhanced levels of UV radiation, showing changes in primary productivity, nutrient cycling, community structure, and modification to the transport and speciation of toxic chemicals in the food-chain.
• UV penetration of surface waters is attenuated by the presence of dissolved organic carbon (DOC), and changes in DOC levels have a greater effect on the vulnerability of freshwaters to UV than changes in stratospheric ozone. In boreal lakes, where global warming and lake acidification have caused a sharp decline in DOC, UV-B penetration into the water column has increased between 22% and 60%.
• Even at current ambient levels, UV-B inhibits some species of zooplankton from frequenting their preferred position in the water column.
• Under experimental conditions, enhanced levels of UV radiation depressed algal production but affected the zooplankton that fed on the algae to a proportionately greater extent. As a result, the accumulated algal biomass increased, even though primary production was reduced. Because a key link in the food chain had been weakened, less production was transferred to higher trophic levels. The implications for aquatic ecosystems are disturbing.
• Concern has been expressed over synergisms between UV-B, climate change, lake acidification, fungal infections, and toxic chemicals that could affect amphibia. However, Ontario amphibians do not seem to have been affected by current levels of radiation, despite the apparent vulnerability of eggs and larvae.

Marine ecosystems

• Direct impacts of UV-B on microbial processes and higher trophic levels have been demonstrated, but the cumulative effects of ozone depletion on marine ecosystems are, as yet, unpredictable. Potential responses include changes in species composition, shifts in food webs (with collateral impacts on fisheries), and possibly climatic changes.
• Increases in UV-B alter the growth, survival, and biogeochemical activities of many microbes, plants, and animals in the sea. Damage to DNA directly influences survival, but UV-B also affects spatial orientation (and hence vertical migration) of phytoplankton, nitrogen metabolism, photosynthesis, larval mortality, and processes ranging from viral infection to hatching success in fish eggs.
• The sensitivity of physiological processes to solar UV-B requires a biological weighting function (BWF, also called an action spectrum) to quantify the effective irradiance. BWFs need to be determined for a greater number of biological functions and estimates need to be made of the range in variation of BWFs for each process, as these appear to be highly variable between species.
• Biological effects of UV-B have been detected tens of metres into the water column, most notably the inhibition of short-term photosynthesis in Antarctic phytoplankton. Quantitative estimates of the inhibition of photosynthesis by UV radiation are converging as research continues.
• Some toxic dinoflagellates show UV-photoprotective mechanisms that might give them a competitive edge, perhaps leading to a greater dominance of toxic or nuisance algae.
• Increases in UV-B may favour species of phytoplankton that produce dimethyl sulphide (DMS), a gas involved in cloud formation, and thus contribute to climate change.
• Interpretations of UV effects in Canadian waters require recognition of the fundamental differences that exist between the Antarctic and waters of the northern hemisphere, including the Arctic. Consequently, basic research on marine ecosystems in Canadian waters is essential.

Materials

UV radiation causes significant and deleterious changes to many materials used in outdoor applications. Any increase in UV flux to the earth's surface will degrade infrastructure and so generate significant costs for repair and replacement.

• Canadian research has addressed the effects of UV on polymers, wood and paper, building materials, paints and coatings, and textiles and clothing, although the main thrust has been on the evaluation of radiation resistance of materials used in outer space and of clothing materials.
• UV-B radiation damages synthetic polymers and other materials, but the mechanisms are not well understood at the molecular level and the combined impacts of both short and long wavelength radiation and other environmental variables adds further complexity to the issue.
• Bleached pulp and paper products made by inexpensive processes are discoloured by UV radiation. Canadian researchers have made significant advances in understanding how this occurs. The ability to reduce discolouration could greatly expand the market for this class of paper.
• Non-plastic building materials such as roofing membranes and outdoor sealants are currently being studied with respect to their resistance to UV but not specifically in the context of enhanced, ozone-related irradiance.