REGIONAL ATMOSPHERIC POLLUTION AND TRANSBOUNDARY AIR QUALITY MANAGEMENT*

7 Oct 2005 17:38 AR ANRV256-EG30-01.tex XMLPublishSM(2004/02/24) P1: KUV10.1146/annurev.energy.30.050504.144138

Annu. Rev. Environ. Resour. 2005. 30:1–37doi: 10.1146/annurev.energy.30.050504.144138

Copyright c© 2005 by Annual Reviews. All rights reserved

REGIONAL ATMOSPHERIC POLLUTION AND

TRANSBOUNDARY AIR QUALITY MANAGEMENT∗

Michelle S. Bergin,1 J. Jason West,2 Terry J. Keating,3

and Armistead G. Russell11Georgia Institute of Technology, Department of Civil and Environmental Engineering,Atlanta, Georgia 30332; email: [email protected]@ce.gatech.edu2Princeton University, Program in Atmospheric and Oceanic Sciences and WoodrowWilson School of Public and International Affairs, Princeton, New Jersey 08540;email: [email protected]. Environmental Protection Agency, Office of Air and Radiation, Washington,District of Columbia 20460; email: [email protected]

Key Words ozone, particulates, mercury, POPs, climate

■ Abstract The regional nature of several important air pollutants, which includeacids, ozone, particulate matter, mercury, and persistent organics (POPs), is widely rec-ognized by researchers and decision makers. Such pollutants are transported regionallyover scales from about 100 to a few 1000s of kilometers, large enough to cross state,provincial, national, and even continental boundaries. Managing these regional pol-lutants requires overcoming political, economic, and cultural differences to establishcooperation between multiple jurisdictions, and it requires recognition of the linkagesbetween pollutants and of impacts at different geographic scales. Here, regional dy-namics of the pollutants are discussed, addressing them individually and as a tightlylinked physical and chemical system. Collaborative efforts to characterize and manageregional pollution are presented, along with potential directions for future efforts.

CONTENTS

REGIONAL AIR POLLUTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2GENERAL CHARACTERISTICS OF REGIONAL ATMOSPHERIC

TRANSPORT AND CHEMISTRY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3REGIONAL AIR POLLUTANTS AND EFFECTS . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Acid Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Tropospheric Ozone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Particulate Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Mercury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

∗The U.S. Government has the right to retain a nonexclusive, royalty-free license in and toany copyright covering this paper.

1543-5938/05/1121-0001$20.00 1

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.

7 Oct 2005 17:38 AR ANRV256-EG30-01.tex XMLPublishSM(2004/02/24) P1: KUV

2 BERGIN ET AL.

Persistent Organic Pollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16Regional Climate Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18A Unified View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

MANAGEMENT OF REGIONAL TRANSBOUNDARY AIR POLLUTION . . . . . 19Examples of Transboundary Air Quality Management Regimes and Approaches . 21Achieving Successful Regimes for Transboundary Air Pollution . . . . . . . . . . . . . . 23Assessment of Existing International Regional Regimes . . . . . . . . . . . . . . . . . . . . 25

SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

REGIONAL AIR POLLUTION

Current estimates are that a quarter of the world’s population is exposed to un-healthy concentrations of ambient air pollutants (1), with nearly 6.4 million yearsof healthy life lost owing to long-term exposure to ambient particulate matteralone (2). Research results indicate further harmful health effects from ozone, air-borne metals (e.g., mercury), and organics (e.g., benzene, dioxins). Additionally,air pollution damages crops, built structures such as buildings and historic mon-uments, ecosystems, and wildlife. Outdoor air pollution largely results from thecombustion of fossil fuels for transportation, power, and other human activities (1),although natural sources play important roles in determining the ultimate impactof those anthropogenic emissions. Without further actions to control emissions,air pollution problems will only worsen as populations and fossil-fuel combustionincrease.

Although stratospheric ozone depletion and climate change have been identi-fied as global issues, most air pollution has historically been viewed and treatedas local or urban scale. Now both the scientific and political communities in-creasingly recognize that several important air pollutants have regional distribu-tions. Attention has shifted toward regional air pollution for a number of reasons.First, shifting demographic and development growth patterns decentralized indus-trial and transportation-related emission sources over regional scales. Second, pasturban-scale air quality management has effectively addressed some of the most ap-parent urban problems [such as smoke, lead, CO, and total suspended particulates(TSPs)], allowing regional problems to come to the forefront. Similarly, controlof urban emissions has often made the regional contribution a growing fraction ofthe total pollutant concentrations in an urban area. Third, management efforts havefound that some controls, such as those on vehicles, are best applied over largerareas because of economies of scale and because such sources move betweenpolitical jurisdictions. Finally, scientific research, in part through ground-basedmeasurements, satellite observations, and modeling studies, has shown that urbanpollutants such as ozone and particulate matter are transported long distances, andresearch has highlighted the importance of the impacts of those pollutants on hu-man health and ecosystems. A recent report from the National Research Councilin the United States listed pollutant transport across multistate regions, nationalboundaries, and continents as one of the seven major air quality challenges facing

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


REGIONAL AIR POLLUTION 3

the United States in the coming decades (3), and this challenge is important tomany other nations as well.

Regional-scale air pollution is defined in this review as being on the orderof 100 to a few 1000s of kilometers. Over this scale, geographic patterns oftransport are dependent on the pollutant considered, the location of emissions,and meteorology, which may themselves be affected by changes in topography.This scale is often large enough to include multiple political jurisdictions, i.e.,states, provinces, or nations, and the geographic patterns of pollutants often donot coincide with the boundaries of political jurisdictions. Consequently, impactsof air pollution in one political jurisdiction may be caused, in part, by emis-sions from another jurisdiction, making cooperation necessary to achieve pollutioncontrol.

In this review, we first present an introduction to regional atmospheric trans-port and chemistry as well as to the impacts, chemistry, and physics of severalsignificant regional air pollutants of current and historical interest. We presenteach of these pollutants individually [acid deposition, tropospheric ozone, par-ticulate matter, mercury, and persistent organic pollutants (POPs)], then discusssome of their important linkages with regional climate change and with eachother. After providing this scientific background, we describe efforts to cooper-atively characterize and manage these regional pollution problems and discusspotential directions for future policy efforts. Although there is still much to belearned about the regional nature of different air pollutants, our current scientificunderstanding recognizes that transboundary air pollution is common and sug-gests that future management efforts must recognize both the linkages across ge-ographic scales and linkages between pollutants (3), making political cooperationnecessary.

GENERAL CHARACTERISTICS OF REGIONALATMOSPHERIC TRANSPORT AND CHEMISTRY

Air pollutants are broadly categorized as either primary or secondary. Primarypollutants are directly emitted, such as inorganic carbon particulates (e.g., dieselsoot). Secondary pollutants, such as tropospheric ozone and much of the fineparticulate mass, are formed through chemical reactions in the atmosphere. Primarypollutants can be managed directly by reducing their emissions, although sourcesmay be hard to identify or reduce. Management of secondary pollutants can bemore complex because their relationships to emission sources may not be readilyapparent and response to emission reductions may not be proportional.

Whether a pollutant has a regional impact is determined by its rate and methodof emission and/or formation, method of transport, and removal and/or destructionprocesses—many of which are bound by the governing processes of the atmosphereand natural emission patterns. The atmosphere is far from equilibrium, and itschemical nature, transport properties, and other characteristics all play a role in the

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


4 BERGIN ET AL.

fate of air pollutants. In particular, the atmosphere is both oxidative and turbulent,and these two characteristics dominate the evolution of pollutants after emission.

In a regional context, pollutant transport occurs in the lower atmosphere, ortroposphere, which is approximately 10 km thick. The troposphere includes theplanetary boundary layer (PBL) and the free troposphere above it. The PBL isthe most directly influenced by warming and cooling of the earth’s surface, andits thickness varies generally between a few hundred to a couple thousand metersduring the day. Most regional air pollutants are contained in this boundary layer.Within the PBL is the mixed layer, which is the region where the most rapidturbulent mixing occurs.

Turbulence in the mixed layer is caused by ground-layer heating and mechanicalforcing, which drives vertical transport. Vertical transport and mixing has a largeeffect on the ultimate distance pollution is carried and on the dilution and chemicalprogression of the initial air parcel. Ground-layer heating warms air parcels next tothe surface, causing them to rise, whereas colder air parcels from above (from adi-abatic expansion) fall, creating an unstable atmosphere with thermal eddies, whichefficiently disperse pollutants throughout the mixed layer. This mixing can diluteground-level emissions upward and fumigate elevated plumes, e.g., from powerplants, downward. On a sunny day, thermal mixing dominates vertical pollutantdispersion, with pollutants diffused up to hundreds of meters in a few minutes.On less sunny days, or at night, mixing still occurs owing to mechanical turbu-lence generated by the wind, which interacts with a rough surface, although thisprocess is slower. When there is little or no sunlight, the atmosphere can becomestable as cooling at the earth’s surface traps cold air below warmer air. In suchcases, mechanically generated turbulence can also be inhibited, leading to highconcentrations of ground-level pollutants trapped very near the surface (4).

Horizontal transport is dominated by advection, generally following customarylocal and regionally seasonal wind currents. One estimate suggests that the clima-tologically averaged mean transport velocity is around 400 km per day, althoughthis depends on episodic meteorology and can vary from a few tens of kilometersto about 1000 km. Strong winds can spread emissions and pollutants over a wideregion, transporting persistent pollutants, increasing background concentrations,and causing occasional severe pollution episodes (i.e., dust storms). However,many of the more severe pollution episodes occur when the air is more stagnant,exacerbating the impact of local sources.

Transport and pollution patterns are influenced by physical changes on theearth’s surface, such as land-ocean interfaces, mountain ranges, and variations insurface terrain. Pollutant distribution is also affected strongly by the location ofemission sources, including the dispersion of megacity emissions from highly in-dustrialized regions, and the geographical distribution of land-use changes thataffect both anthropogenic and natural emissions (such as forest versus desertemissions).

Emissions result from natural sources, such as soil, plants, volcanoes, and light-ning, and from anthropogenic sources, such as power plants, industrial processes,

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



cars, and farm equipment. These sources can be affected by meteorological andhuman activity pattern changes (e.g., daily, weekly, seasonal) and other variables,including the altitude at which the source operates. For example, mobile sourcesoften have higher organic emissions at high altitudes. Once precursors or pollutantsare emitted, the distances they travel are largely governed by the rates of chemicalreaction and deposition processes.

Oxidation (such as the transformation of NO to NO2) often initiates the chemi-cal reactions of emissions that lead to secondary pollutant formation (4a). In fact,oxidation by the hydroxyl radical (HO•) is key to the formation and fate of manyregional pollutants (as described below). Radiative energy from the sun also fa-cilitates various chemical reactions and alters the rates, spatial distributions, andchemistry of many emissions (e.g., biogenic and evaporative emissions), so pol-lution follows daily, seasonal, and latitudinal patterns. The emission sources andchemistries of many regional pollutants are closely linked by these factors.

Secondary formation can occur rapidly or over a long period of time depend-ing on the chemical pathways, the presence of other precursor species or catalysts,and whether meteorological conditions are conducive to production. Major chemi-cal pathways for secondary pollutant formation include oxidation, photochemistry(reactions requiring energy from the sun), and heterogeneous chemistry (generallyoccurring on or in liquid droplets in clouds or fog). These processes can occur indi-vidually or in conjunction with each other. Chemical transformations can involvethe formation of intermediate species, which may be stable under some condi-tions and then react under other meteorological conditions or in the presence ofother chemical species. The formation of intermediate species can contribute tothe long-range transport of pollution.

Together with the chemical destruction of pollutants, deposition to the earth’ssurface determines the lifetime of pollutants and the extent of their transport. De-position may occur with precipitation (i.e., wet deposition) or without (i.e., drydeposition). In dry deposition, gases and small particles turbulently diffuse in themixed layer to very near the surface, where Brownian diffusion and residual turbu-lence move the pollutants into contact with surfaces to which they can physicallyadhere (i.e., particles) or be consumed chemically or bonded (i.e., gases). Depend-ing on variables, such as the pollutant species (if a gas), the particle size (if anaerosol), the surface, and the prevailing meteorology, the rate of deposition canvary widely. For example, nonreactive gases, like CO, have a deposition velocityon the order of 0.0003 m/s or less, leading to a characteristic lifetime due to de-position of weeks or more. Conversely, reactive gases, like nitric acid, can have adeposition velocity as high as 0.04 m/s, leading to a lifetime of hours (5).

Wet deposition occurs when the pollutant becomes associated with a droplet(fog, rain, cloud, or snow) that deposits to a surface. Pollutant gases and particlesdiffuse to the droplet surface and are absorbed, or particles impact a droplet. Insome cases, the pollutant actually seeds the formation of the droplet (serves as anucleation site), affecting climate as discussed further below. Timescales for wetdeposition are highly variable.

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


6 BERGIN ET AL.

REGIONAL AIR POLLUTANTS AND EFFECTS

The previous section discussed general processes that affect the geographic dis-tributions of pollutants. However, each pollutant has unique impacts, causes, andchemistry. This section discusses individual regional-scale air pollutants and theireffects, including (a) acid deposition, which destroys or damages aquatic and ter-restrial ecosystems and manufactured materials, (b) tropospheric ozone, whichharms the health of humans, wildlife, crops, and ecosystems, and is a green-house gas, (c) particulate matter, which also harms health, degrades visibility,and changes precipitation and temperature patterns, (d) mercury, which is toxicto humans, plants, and animals, and bioaccumulates in the food chain, and (e)POPs, which are also toxic, bioaccumulate, and remain in the environment formany years. Regional climate change is also discussed, which both impacts and isimpacted by regional air pollution.

Acid Deposition

Deposition of acidic compounds is one of the best-known regional air pollutionproblems and, as discussed below, has been the focus of several transboundaryagreements. Acids are deposited to the earth’s surface through both wet and dryprocesses. Acid rain in a polluted area often has a pH in the range of 3 to 4, andeven lower values have been measured (6). For comparison, the natural pH ofprecipitation is around 5.6 (slightly acidic) owing to reactions of carbon dioxidewith water in the atmosphere (5). Most aquatic insects, algae, and plankton cannottolerate water with a pH below 5.0. These organisms are at the base of the foodchain, and their survival affects the survival of all other aquatic organisms. ApH below 5.0 can also result in reproductive failures in fish and amphibians (6).Acid deposition also harms terrestrial ecosystems and manufactured surfaces andmaterials, including historic buildings and monuments (7).

Harmful effects are caused directly by acidification and/or indirectly by theaddition of poisonous metals, such as aluminum and mercury, that are mobilizedwhen acidified runoff leaches the metals from soils and clay. A body of water canbe “shocked” by a large acid load from snowmelt runoff in the spring. Effects onhuman health are uncertain, but there are likely effects from ingesting fish thathave bioaccumulated leached mercury and/or aluminum, and effects are known tooccur from respiring related acidic aerosol particles.

The natural pH of water bodies and soils varies considerably depending on thesoil composition of a watershed, including its buffering capacity. These differencesexplain why some areas are highly susceptible to damage from acid deposition(e.g., lakes in the northeastern United States and the Black Forest in Germany),whereas other areas, such as those with soils rich in lime, are relatively resistant.Additionally, meteorology and emission patterns cause some areas to experiencemuch higher concentrations of deposited acids than others. Extensive damage fromacid deposition has been found in eastern North America, northern Europe, Japan,

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



China, and Southeast Asia (Figure 1). These patterns of deposition clearly crosspolitical boundaries.

The vast majority of acid deposition pollution is caused by emissions of sulfurdioxide (SO2) and oxides of nitrogen (NO and NO2, together referred to as NOx),which are primarily from fossil-fuel combustion in power plants and vehicles (6).High emissions of both precursors are in southeastern Asia, eastern North America,and western Europe (Figures 2 and 3), although biomass burning, which contributesabout one third of the global NOx, is the dominant source of anthropogenic NOx inthe Southern Hemisphere (5). Fossil-fuel combustion is the overwhelming sourceof SO2 emissions (5). There are also natural sources of these precursors, such asvolcanoes (NOx and SO2) and vegetation, soil, and lightning (NOx). In additionto acid deposition, atmospheric nitrogen pollution (including both oxidized andreduced forms) is responsible for impacts on human health, visibility, crop damage,and regional eutrophication of soils and waters, as well as on climate change andstratospheric ozone depletion (e.g., N2O emissions) (8a–8c). In a cycle termed thenitrogen “cascade,” an atom of emitted nitrogen can participate in many pollutioneffects before being immobilized or converted back to a benign form (e.g., N2).Nitrogen emissions from transportation, energy generation, and industrial andagricultural activities are found to contribute significantly to the total nitrogenpollution cascade (8b, 8c).

Reactions of SO2 with peroxides (e.g., H2O2) in clouds or with HO• in airlead to sulfuric acid (H2SO4) (Figure 4). Sulfuric acid is highly soluble, so it isincorporated by clouds, fog, and raindrops. NO2 is also oxidized by HO• and bynighttime heterogeneous reactions involving ozone, resulting in the formation ofnitric acid (HNO3). Nitric acid is easily deposited to surfaces through both wet anddry processes. The HO• and HO2

•, which cycle between each other, are central tothe chemical transformation of virtually all species (4a). HO• is formed primarilyfrom the photolysis of ozone followed by reaction with water.

In general, NOx is converted to nitric acid within a few days or less after beingemitted, and SO2 is converted to sulfuric acid within several days, suggesting thatmean transport distances of SO2, NOx, and their oxidation products are 200 to1200 km before deposition (5, 10). In addition to direct transport, NOx can formsemistable compounds (organonitrates) that can travel hundreds to thousands ofkilometers within the cooler regions of the troposphere before converting back toNOx and then to nitric acid.

In the 1950s and 1960s, the regional nature of acid deposition became apparentwhen a relationship was demonstrated between emissions of sulfur in continentalEurope and acidification in Scandinavian lakes (10). Cooperative monitoring andmodeling studies of regional acid deposition in Europe began in the early 1970sunder the auspices of the Organization for Economic Cooperation and Develop-ment and led to the 1979 Convention on Long-Range Transboundary Air Pollution(discussed in more detail below). Awareness of the regional nature of acid depo-sition in North America followed and, in the United States, led to the formationof the National Atmospheric Deposition Program in 1978 and the National Acid

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


8 BERGIN ET AL.

Figure 4 Schematic of the unified chemistry of regional ozone, particulate matter,acids, and mercury. Particulate phase species are boxed.

Precipitation Assessment Program in 1981 (11). Recognition of the large-scalenature of acid precipitation in eastern North America led to the development of aregional emission trading program in the United States (discussed below) and toemission reduction commitments contained in the 1991 Canada-United States AirQuality Agreement. Regional acid deposition was recognized in East Asia in theearly 1990s, and eventually led to the creation of the Acid Deposition MonitoringNetwork in East Asia (EANET) in 1998 (http://www.adorc.gr.jp/).

Tropospheric Ozone

Ozone (O3) is a highly reactive state of oxygen, and concentrations between 10 and40 parts per billion (ppb) occur naturally in a clean troposphere through both strato-spheric injection and chemical reactions of natural emissions in the troposphere.Ozone concentrations in polluted urban areas reach between 100 and 400 ppb(5). Tropospheric ozone was first recognized as an urban air pollution problem,notably in Los Angeles in the 1940s, and for decades was perceived and treatedas an urban-scale problem. During the 1980s, measurement and modeling studies

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



demonstrated the more regional nature of ozone pollution, in part as a result of thesuccess in reducing extreme urban ozone episodes.

Anthropogenic ozone is responsible for damage to human health, crops, naturalvegetation, and wildlife, and it is a potent greenhouse gas that contributes to globalclimate change (12–14). Ozone causes both acute and chronic health problems,especially related to lung functions, asthma, and pulmonary infection (15), as wellas other health end points including daily hospital admissions (16). Ozone may alsocause chronic health effects, although no link with long-term mortality has beendemonstrated (16, 17). Although people with preexisting respiratory disease andchildren are most sensitive to ozone damage (18), much evidence indicates thatozone can adversely affect the respiratory systems of any individual dependingon the conditions of exposure (19). In 2002, over 50% of the U.S. populationlived in counties where monitored air pollution was found to exceed the humanhealth based National Ambient Air Quality Standards (NAAQS); the vast majorityof this pollution was due to ozone and particulate matter (7, 20). Many urbanareas outside of the United States also often experience ozone levels far exceedingthe current U.S. NAAQS, and other nations have set even lower concentrationstandards.

In addition to human health impacts, air pollution affects agricultural produc-tion (21), accounting for an estimated several billion dollar crop loss every year inthe United States alone (22). Estimates are that 10% to 35% of the world’s grainproduction occurs in regions where ozone pollution likely reduces crop yields(23). Ozone may also cause short- and long-term damage to the growth of foresttrees (24) and alters the biogenic hydrocarbon emissions of vegetation. In addi-tion, research is examining ozone effects on sensitive wildlife populations. Evenaccounting for planned reductions in precursors from some regions, concentrationsof ozone are predicted to increase over the next decades (13, 25). As discussedbelow, tropospheric ozone also impacts climate.

Although a small amount of ozone is transported from the stratosphere to thetroposphere, the only significant process that forms ozone in the lower atmosphereinvolves reactions between volatile organic compounds (VOCs) and NOx in thepresence of sunlight (Figure 4). Photolysis of NO2 is followed by the rapid reactionof the released oxygen atom with O2, which is quickly reversed by the reaction ofO3 with NO. With no other removal process for NO (and most NOx is emitted asNO), reactions with the released NO would rapidly deplete ozone, and there wouldbe no net ozone formation—as is the case at night with near-surface NOx emissions.VOCs provide the energy needed to oxidize NO to NO2, with some reactive VOCsactually cycling NO to NO2 many, many times over. Ozone removal is largelydependent on photolysis reactions, kinetic reactions, and deposition (5). Photolyticloss depends on water vapor concentration, so it is highest at low altitudes in thetropics, which have high radiation and high humidity (10). Loss to depositiontakes a few days over land, and longer over water. Ozone and its precursors canbe transported over large regions (25), making it difficult or even impossible forsome areas to manage their own concentrations through local emission controls

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


10 BERGIN ET AL.

(M.S. Bergin, J.-S.Shih, J.W. Boylan, J.G. Wilkinson, A.J. Krupnick, A.G. Russell,unpublished results).

As noted above, NOx emissions, which lead to increased ozone concentrations,are mainly produced during fossil-fuel combustion in the Northern Hemisphereand biomass burning in the Southern Hemisphere (Figure 2). Globally, most VOCemissions are biogenic, emitted by plants and trees, although anthropogenic VOCsin heavily polluted areas are instrumental in the formation of high ozone con-centrations. Anthropogenic VOCs arise from many sources, including exhaustand evaporative emissions from vehicles, solvents, surface coating activities, foodcooking, and dry cleaning. There are hundreds of different compounds classifiedas VOCs, and their ozone formation potentials (termed reactivity) vary widely(27, 28).

In urban areas, ozone formation may be limited by emissions of either NOx

or VOCs, depending mainly on the relative concentrations of the two. Becauseozone formation requires time, and NO emissions initially decrease ozone, peakozone concentrations are often observed downwind of large emission sources (i.e.,industrial or urban areas), and correspondingly, emission reductions may benefitdownwind areas much more than local areas (Figure 5). Because NOx is relativelyshort lived in the atmosphere, and because of the availability of (mainly biogenic)VOCs in many rural regions, ozone formation on a regional scale is generallylimited by the availability of NOx (14).

In the United States, the regional nature of ozone formation was describedby Wolff & Lioy (29) as a “river of ozone” flowing along the east coast of theUnited States from the District of Columbia to New England. Through the 1980sand early 1990s, a number of monitoring and modeling studies were performedto examine the regional nature of ozone, including the San Joaquin Valley AirQuality Study (30), the Lake Michigan Ozone Study (31), the Regional OzoneModeling for Northeast Transport, and the Southern Oxidant Study (SOS) (32).The SOS demonstrated that the river analogy used to describe regionality in theNortheast was not applicable in the Southeast. Instead, the SOS found that urbanareas and power plants tended to add fresh emissions to a ubiquitous sea of VOCs,leading to widespread and simultaneous formation of ozone across the Southeastand creating a “rising tide of ozone” (32). These ozone studies formed the basisfor regional management strategies implemented in the late 1990s, as describedfurther below. Some recent modeling studies examined the sensitivity of ozone in19 individual U.S. states to emissions from other states to support policy decisionsfor air pollution control (33; M.S. Bergin, J.-S.Shih, J.W. Boylan, J.G. Wilkinson,A.J. Krupnick, A.G. Russell, unpublished results) and to evaluate cost-optimizedemission reduction strategies (J.-S. Shih, M.S. Bergin, A.J. Krupnick, A.G. Rus-sell, unpublished results). Ozone concentrations in most states were found to besensitive to emissions from other states (Figure 5).

Europe also began addressing ozone as a regional problem during the 1980sand 1990s. Cooperative research funded through the European Experiment on theTransport and Transformation of Environmentally Relevant Trace Constituents

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



in the Troposphere over Europe (EUROTRAC) made significant contributions tounderstanding tropospheric ozone (35).

More recently, the potential for intercontinental ozone transport was demon-strated first by monitoring studies of the export of ozone from the east coast ofNorth America (36) and Asia (37–39) and by modeling studies of trans-Atlantic(40, 41) and trans-Pacific transport (42, 43). Historical measurements and model-ing studies further suggest that anthropogenic NOx emissions continue to increaseglobal tropospheric ozone concentrations (44, 45), thereby raising the backgroundupon which urban and regional concentrations build.

Particulate Matter

Ambient aerosols (particulate matter) are chemically complex, dynamic mixturesof solids and liquids suspended in the atmosphere. Particulate matter is harmful tohuman and environmental health, degrades visibility, damages built structures, andalters climate. Some of the health impacts were recognized early (10), especially inhighly polluted cities. Most regulations initially focused on controlling the mass oftotal suspended particulate (TSP). Restrictions could not be more specific becauseparticles can rapidly change in both size and chemical composition as humidity,temperature, and other surrounding ambient concentrations change, making theaerosol composition difficult to sample and characterize.

As technical capabilities advanced, urban aerosols were found generally to forma trimodal lognormal size distribution in the atmosphere (by mass). The smallestmode, the nuclei mode, is composed of particles with aerodynamic diameters (Dp)less than about 0.1 µm (ultrafines). The accumulation mode is 0.1 < Dp < 2 µm,and the coarse mode is Dp > 2 µm. The nuclei and accumulation modes are calledfine particles.

Regulations soon targeted coarse particles (PM10—particles which have Dp <

10 µm), rather than TSP. Currently, many regulations target particles with Dp <

2.5 µm (PM2.5). Smaller particles have been emphasized in regulations for severalreasons. First, smaller particles are more efficient at penetrating deep in the lungs,although there is evidence that particles in the 2.5 to 10 µm range also have healthimpacts (46). Larger particles also settle out rapidly, reducing their range of impact,and they have a smaller impact on visibility per unit mass.

Fine particulate matter (alone or in combination with other air pollutants) hasbeen shown to cause a large number of adverse health effects, including prematuredeath (47, 48) and an estimated global loss of nearly 6.4 million years of healthy lifedue to long-term exposure (2). Evidence of premature mortality comes from bothdaily time series studies (49) and long-term cohort studies that suggest particulatematter causes chronic health effects (17). Exposure to larger particles is primarilyassociated with aggravation of respiratory conditions, such as asthma, and withpremature death (7, 50). A threshold has not been identified under which particulatematter has no adverse health effects (2). Some recent studies indicate that theultrafine particles may also be pivotal in health and environmental impacts (51,

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


12 BERGIN ET AL.

52); however the effects are difficult to assess because the methods and tools neededhave not yet been fully developed and are not widely available (51). Research isnow also addressing whether the chemical composition of particles, in addition tosize, affects health (50).

In addition to human health effects, soils, plants, water, and materials can also bedamaged when particles deposit on them. The inorganic fraction of fine particulatematter, particularly sulfate, is responsible for much of the acid deposited and thuscauses most of the impacts discussed above. Fine particulates are the most efficientsize for scattering light in the atmosphere, causing large degradation of visibilityand altering climate. In many parts of the United States, visual range has beenreduced 70% from natural conditions by air pollution. The average visual range ineastern parks has decreased from 90 miles to 15–25 miles, and in western parksfrom 140 miles to 35–90 miles (53). Light-scattering properties of aerosols, theireffects on clouds, and light-absorbing properties of some aerosols (i.e., elementalcarbon) alter global and regional climate, as discussed below in the RegionalClimate Change section.

Many of these aerosol effects depend on particle size, which is the result of manyvariables such as particle sources (both primary and secondary), the chemicalcomposition of the local atmosphere, and meteorology. Particles of given sizeshave differing emission, formation, and removal processes. Fine particles are bothdirectly emitted into the atmosphere (by combustion processes) and formed in theatmosphere through secondary processes, such as condensation and heterogeneouschemical reactions (Figure 4). Coarse particles are generally primary (emitted)solids such as crustal materials (e.g., wind-blown dust), sea salt, pollen, and organicdetritus, as well as fugitive dust from agricultural activities, unpaved roads, andindustries.

Coarse particles tend to deposit rapidly through sedimentation, with atmo-spheric lifetimes of minutes to days, usually resulting in transport of less than tensof kilometers (5) and in local impacts. However, large dust storms and biomass-burning events can carry coarse (and fine) particles very long distances (as dis-cussed further below). Although dry deposition can also be important (54), ac-cumulation mode particles (fine particulate matter) are removed primarily by wetdeposition (both in-cloud and below-cloud scavenging) and thus have atmosphericlifetimes of days to weeks, resulting in potential travel distances of hundreds tothousands of kilometers (5). Particles in the nuclei mode grow into the accumu-lation mode by coagulation and condensation. For these reasons, fine particulatesare dominant in most regional-scale aerosol pollution events. In addition to size,hygroscopic properties of the aerosol (which depend on chemical composition)also determine travel distance.

Regional aerosol pollution patterns depend on the particle type and regionalmeteorology and topography. Regional and intercontinental aerosol pollution canbe caused by transport of urban particles, biomass burning, and dust storms, whichare mostly natural but can be increased by anthropogenic desertification. In indus-trialized areas such as much of Europe, North America, Japan, and large cities in

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



Asia, particles are mainly generated by gasoline, diesel, and coal combustion andfrom condensed organic and metal vapors (2). Particles from urban and industrialareas tend to contain sulfate, nitrate, ammonium, and organic carbon, with smalleramounts of metals and crustal materials. As previously discussed, sulfate (SO4)is formed when sulfur dioxide (SO2, often emitted from coal-fired power plantsand large industries) either reacts through an aqueous path with hydrogen perox-ide (H2O2) or is oxidized in the atmosphere by the HO• to form sulphuric acid(H2SO4), which has a low vapor pressure. Ammonia (NH3) reacts with H2SO4

to form ammonium sulfate. Nitrate aerosol is formed from gaseous nitric acid(HNO3) reacting with gaseous NH3 to form condensed phase ammonium nitrate(NH4NO3) or reacts with existing alkaline aerosol. Nitric acid is formed when NO2

is oxidized by the HO• (as discussed above and in Figure 4). Ammonia is emittedfrom livestock manure and other agriculturally related processes, such as spread-ing natural and synthetic fertilizers, and from vehicle exhaust. Organic carbonparticulate can be primary or formed by the condensation of organic gases. Thereare many sources of organic gases (discussed above). In less developed areas, suchas parts of Africa and Asia, biomass burning is the principal source of regionalaerosol pollution, resulting largely in elemental and organic carbon particles andnitrate. Dust storms largely contribute mineral particles.

Many measurement and modeling studies, as well as satellite measurements,have now shown that aerosols can cover large geographic areas. Dust storms fromas far as the Sahara and Gobi deserts have been observed to impact western Europe(55, 56), various sites across the United States (57–61), and western Canada (62–64). Observations in Europe have even recorded the effects of Asian dust thattraveled across the Pacific, North America, and then the Atlantic (65). Episodes ofdust can be striking because dust during these storms can far exceed anthropogenicemissions (66). At least one modeling study suggests that transported Asian dustalso carries with it significant anthropogenic particulate matter (67), suggesting,for example, that Asian anthropogenic emissions could have a significant influenceon particulate matter concentrations in western North America.

Biomass burning also creates potent pollution events. Large forest fires inCanada were observed to impact air quality in the United States (68), including asmoke plume from wildfires in Quebec, Canada, measured as far south as Virginia(69). Fires in Mexico and Guatemala were also found to impact air quality in theUnited States (70, 71), and smoke from Central American and southern Mexicanforest fires was captured in satellite images traveling across the United States andinto Canada (Figure 6). During this air pollution event, visibility in many Texascities was less than one mile (72). Satellite images can track and quantify impor-tant properties of biomass burning and dust storm events (73–75), and informationfrom these images can be verified and enhanced by comparison with ground-basedmeasurements, which provide more detail in chemical composition and in agingand mixing effects (69).

Other studies have demonstrated the regional extent of PM2.5 and PM10 fromurban sources, such as across state boundaries in the United States (32, 33, 77–79;

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


14 BERGIN ET AL.

M.S. Bergin, J.W. Boylan, J.G. Wilkinson, J.-S. Shih, M.T. Odman, A.J. Krupnick,A.G. Russell, unpublished information), across North American nations such as theUnited States and Canada (68, 80), and between the nations of Europe (81). A recentmodeling study calculated the sensitivity of PM2.5 concentrations in 19 easternU.S. states to SO2 emissions from other states and found that on average 75% ofPM2.5 in most of these states was sensitive to reductions in regional SO2 emissions(Figure 7) (M.S. Bergin, J.-S.Shih, J.W. Boylan, J.G. Wilkinson, A.J. Krupnick,A.G. Russell, unpublished results). Another U.S. study used back trajectories andfine sulfate measurements and found roughly half of the sulfate in New York stateoriginated from regional sources (82). A global modeling study (83) found thattransboundary transport of particulate matter and its precursors dominated naturalsources in the United States and that up to 30% of the transboundary particulatematter was due to transport from Asia. Each of these studies highlights the needto address regional transport in developing efficient aerosol pollution reductionplans.

Mercury

Although other toxic metals (e.g., lead) are known to be transported long distancesand introduced to the environment through atmospheric pollution, mercury (Hg)is proving to be the most serious of these metals. Mercury, particularly in theform of methylmercury (CH3Hg) or (MeHg), is highly toxic to humans and otheranimals, and evidence has shown that mercury threatens the health of humansand wildlife even in areas that are not obviously polluted (84). Concentrations arehighest near emission sources, but gaseous elemental mercury can circulate ona global scale, and anthropogenic emissions have resulted in worldwide chronic,low-level environmental exposures (85). Even at very low atmospheric depositionrates in remote locations, the bioaccumulation of mercury can result in toxic effects(84). Although mercury is dispersed globally, differences in sources, emissiondensities, deposition patterns, aquatic and terrestrial environments, and exposure-related behaviors lead to varying impacts in different regions.

Mercury affects the immune system, alters genetic and enzyme systems, anddamages the nervous system (84). Developing embryos are five to ten times moresensitive to damage by MeHg, which rapidly crosses the placenta, than are adults(84), and young children are especially susceptible to mercury’s neurotoxic effects(86). For these reasons, pregnant and nursing woman and young children are themost sensitive human populations to poisoning (87). Humans are exposed to MeHgalmost entirely by eating contaminated fish and animals that are at the top of aquaticfood chains (84). In the United States, 45 states issued mercury advisories in 2003for limiting the consumption of certain fish from freshwater lakes and rivers, coastalwaters, and marine environments (87). Even with these advisories, it is estimatedthat over 60,000 children are born in the United States each year at risk for adverseneurodevelopmental effects caused by in utero exposure to MeHg (85). Althoughdifficult to prove, studies clearly indicate that ingesting contaminated fish also

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



poisons many aquatic birds and other wildlife (84). Predators such as large fishand aquatic birds can have mercury levels more than a million times higher thanthose in the surrounding environment (86).

Some mercury occurs naturally in the environment; however, atmospheric de-position is the dominant source of mercury over most of the landscape (84). Once inthe atmosphere, mercury can disperse widely and circulate for years after emissionthrough a series of “hops” via atmospheric transport, deposition, and evaporationback to the atmosphere (84, 88, 89) (Figure 8). Sedimentation or entry to the foodchain will eventually remove it from this cycle.

Most mercury is emitted as nonreactive, inorganic mercury vapor, Hg0, al-though a substantial fraction is in the form of reactive gaseous mercury, Hg2+. Inthe atmosphere, Hg0 is oxidized to form HgO (or Hg2+) by both gas phase andaqueous reactions (90–94). Hg2+ rapidly deposits out of the atmosphere. HgO hasa relatively low vapor pressure, so it may exist in the gas phase or may condense onpreexisting particulate matter. This form is soluble in water and can be absorbedin cloud droplets or rain, where it can react to form Hg(OH)2 and Hg(Cl)2; thelatter is preferred if much HCl is present. Transfer to land and water occurs bothby wet and dry deposition, with wet deposition dominating (95). The lifetime ofHg0 in the atmosphere is generally estimated to be approximately 1.2–1.5 years(95), although new findings strongly indicate a much smaller lifetime of about 1or 2 months (96). Some studies suggest that this more rapid oxidation may be dueto reactions with halides (at the poles and temperate marine boundary layer) andurban pollutants, most likely ozone (96). This shorter lifetime suggests that depo-sition occurs much more rapidly than previously estimated, and therefore currentemissions are likely higher than estimated. One such underreported source may begold mining activities, as discussed below. Typical background concentrations ofHg0 are about 1.2–1.6 ng/m3. Hg+ is found in the pg/m3 range, showing greaterspatial variability than Hg0 because it is more reactive.

The most toxic form of mercury, CH3Hg, bioaccumulates through the aquaticfood chain. Mercury reaches aquatic systems through deposition, runoff, or directdischarge (e.g., during gold mining). Transformation of inorganic mercury vaporto MeHg proceeds either indirectly via atmospheric reactions, with subsequenttransformation in terrestrial organisms, or directly by bacteria in aquatic systemsthat normally process sulfate. The bacteria containing MeHg can then be consumedor may excrete the MeHg to the water, where it can adsorb to plankton, which canthen also be consumed (84). Certain types of wetlands (e.g., coastal), dilute low-pHlakes and newly flooded reservoirs, and coastal wetlands are known to favor theproduction of MeHg (84). Once the mercury has entered the food chain throughbacteria, it bioaccumulates in higher animals.

There is a great deal of variation in estimates of the mercury contributionsto the atmosphere from anthropogenic, natural, and reemitted mercury sources,which may have originated from natural or anthropogenic sources (97). The U.S.Environmental Protection Agency (EPA) currently estimates that natural sourcescontribute about a third of current global mercury air emissions, and anthropogenic

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


16 BERGIN ET AL.

emissions make up the remaining two thirds (98). Seigneur et al. (95) estimatethat 2143 and 1049 Mg [Mega grams (106)] of new anthropogenic and biogenicemissions are released per year, respectively, and that another 3201 Mg/yr arereemitted emissions. Of this, 2134 Mg/yr are likely reemitted anthropogenic mer-cury, in essence doubling the effective rate of emission. About three quartersof the total anthropogenic emissions of mercury in 1995 are attributed to com-bustion of fossil fuels (100), particularly in coal-fired power plants (85, 98). Itwas estimated that Asian countries contribute over 50% of the total emissions(100), chiefly from coal combustion in China, India, and South and North Korea(101) (Figure 9). Though firm numbers are not yet established, recent estimatesindicate that artisanal gold mining activities release between 800 and 1000 Mgper year into the air, water, and soil combined (102), which is much larger thanprevious estimates (e.g., 300 Mg/year) (89). These releases generally result inheavy local deposition (103), and this source category will have more signif-icant regional (and global) impacts than previously estimated. This potentiallylarge adjustment to estimates of emissions highlights the underlying uncertaintiesin source attribution and the difficulties of identifying the best controls for thispollutant.

The potential for regional and long-range transport of mercury emissions isdependent on the chemical form in which the mercury is emitted and how fastit is transformed. As we gather more observations of the speciation of anthro-pogenic emissions and their oxidation (and reduction) rates, our understandingof the relative importance of global, regional, and local sources is evolving. In1997, the U.S. EPA estimated that anthropogenic emissions in the United Stateswere responsible for 60% of deposition in the 48 contiguous United States, and40% of deposition was from natural sources, human activities outside the UnitedStates, and U.S. emissions returning via global cycling. Approximately two thirdsof U.S emissions are transported outside the United States (104). Other studieshave suggested that anthropogenic U.S. emissions contribute less than 20% of thedeposition observed throughout most of the United States, with maximum localcontributions of up to 80% in a few locations (95). Similar estimates have beenmade for European countries (105).

Persistent Organic Pollutants

Persistent organic pollutants (POPs) are toxic organic substances that widely dis-perse, persist in the environment, bioaccumulate through the food web, and posea risk to human health and the environment (106, 107). Concentrations can mag-nify in fatty tissues by up to 70,000 times background levels (108). POPs enterhumans chiefly through animal-derived food (fish, poultry, beef, eggs, and dairyproducts) and have been linked to causing, among other effects, cancer, damage tothe nervous system, reproductive disorders, behavioral disorders, and disruption ofthe immune and endocrine systems (109). Because of their fat-seeking properties,POPs magnify in infants through both in utero transmission and breast-feeding.

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



Human breast milk concentrations of some monitored POPs have been increasingexponentially, and these concentrations serve as one indicator of past human ex-posures and environmental conditions as well as conduits that magnify concentra-tions in the next generation (110). Previously unrecognized POPs (i.e., brominatedbiphenyl ethers) have been identified owing to increasing concentrations found bysystematic monitoring of chemicals in human breast milk in Sweden after theybanned the use of seven other POPs in the 1970s (109).

The most commonly recognized POPs are partially oxidized and/or halogenatedorganics, often including one or more aromatic structures. Although there are hun-dreds of POPs currently in the environment, 12 have gained particular interestand have been tagged as “the dirty dozen”: DDT, aldrin, dieldrin, endrin, chlor-dane, heptachlor, mirex, toxaphene, hexachlorobenzene (HCB), polychlorinatedbiphenyls (PCBs), dioxins, and furans. These compounds have been used as pes-ticides (e.g., DDT, aldrin, dieldrin, endrin) in past consumer or industrial applica-tions, such as coolants, flame retardants, lubricants, and sealants (e.g., PCBs), or aregenerated unintentionally as by-products of various combustion processes, suchas medical waste incineration (e.g., polycyclic aromatic hydrocarbons (PAHs),dioxins, and furans).

Common characteristics of POPs are their semivolatility and/or partial watersolubility. These characteristics lead to their ability to be transported in the at-mosphere both in the gaseous and condensed phases (associated with particulatematter), deposited, and then reemitted through a repeated (and often seasonal) cy-cle of evaporation and deposition (similar to mercury, as described above.) Withlifetimes from a few years to decades long, which are increased by halogenation,POPs released in one part of the world may circulate regionally and globally (101)via the atmosphere, oceans, and other pathways, traveling far from their originalpoint of emission (111).

Atmospheric monitoring data for POPs are sparse, and global emission dis-tributions are poorly characterized. Much of what has been learned about POPtransport and impacts in the United States has come from the study of contamina-tion of the Great Lakes region, which, for many years, was a receptor of industrialand agricultural pollution from within the Great Lakes’ watershed and beyond. All12 of the dirty dozen POPs have been identified in Great Lakes’ fish and wildlife,even though some of these compounds were never used in significant quantities inthe region. As a result of the contamination, consumers of the regions’ fish haveup to eight times the body burden of POPs as the general public. The regionaltransport and deposition of POPs (and heavy metals) in Europe have been demon-strated through modeling (e.g., Figure 10) (111a). Further evidence of long-rangetransport comes primarily from the detection of POPs in remote locations, such asremote marine or polar environments where there are no emission sources. POPsare transported to the poles via episodic transport from industrialized source re-gions or through the cycle of deposition and reemission described above, whichacts as a global distillation process. Once in the colder environments at the poles,semivolatile POPs condense and accumulate (112).

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


18 BERGIN ET AL.

Recognizing the need for addressing the POP problem across boundaries, thedirty dozen are targeted to be banned or restricted by the Stockholm Conventionon POPs, which was signed in May 2001 by over 100 countries (113), althoughDDT will continue to be used in some countries that have high malaria risks until asuitable alternative is developed. Additionally, new regulations under considerationin the European Union (E.U.) (REACH) (114) would restrict the use of otherPOPs in chemicals manufactured in or imported to E.U. nations. This program isdiscussed below.

Regional Climate Change

Concern over global climate change derives from observed increases in the at-mospheric concentrations of long-lived greenhouse gases, which are known toabsorb heat in the form of infrared radiation, as well as from observed increases inglobal mean temperature (13). However, although there is substantial evidence forconcern about climate change, projecting future climate remains highly uncertainowing to uncertainties in emissions, the influence of pollutants, and feedback pro-cesses, which may enhance or counteract warming. Even though anthropogenicaerosols and tropospheric ozone are known to influence the climate system, someof the uncertainties associated with their influence are among the most importantin our current understanding (13).

Particulate matter has multifaceted impacts on climate. Most fine particles, suchas sulfate, scatter light very efficiently, causing solar radiation to be redirectedaway from the Earth, resulting in a cooling effect on ambient temperature. Blackcarbon particles (soot) absorb light and therefore may have a warming effect (115).However, because these aerosols have different spatial distributions (and differentdistributions than greenhouse gasses), their effects do not simply offset each other.Particles also serve as cloud nuclei, increasing the number of cloud droplets butdecreasing the average cloud droplet size. This causes an increase in cloud albedo(cooling) and less precipitation (116), which can have manifold effects. Aerosoleffects on temperature include altering regional atmospheric stability and verticalmotions, impacting both large-scale circulation and the hydrologic cycle, all withsignificant regional climate effects (117). The situation is complicated by the factthat the impact of aerosols depends on ambient chemical and physical propertiesof the particles that are not yet well characterized.

Ozone also has a complex relationship with climate. Tropospheric ozone ranksonly behind carbon dioxide and methane in anthropogenic contributions to theglobal average radiative forcing and is expected to increase the global averagetemperature over the next decades (13). As climate changes, the resulting changesin regional climate are expected to increase urban ozone concentrations in manyareas, particularly during highly polluted episodes (118).

Natural climate patterns clearly impact air pollution formation and transport (asdiscussed above), and evidence strongly indicates that air pollution also impactsclimate. Political bodies must address questions such as whether to act under the

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



current understandings and uncertainties, and if so, how to coordinate or prioritizemitigation efforts and who is responsible for reducing emissions. Although muchattention has focused on the global scales of climate change, regional-scale impactshave also been identified and may become more important in the future.

A Unified View

Each pollutant has distinct regional characteristics, yet they are intimately linkedby the chemistry and physics of the atmosphere, as well as by common sourcesand possible controls. Their linked chemistry and common sources present strongarguments for addressing them simultaneously to identify the most effective controlpolicies. For one example, consider relationships between acids, ozone, mercury,and particulate matter—four important regional pollutants (Figure 4).

First, they have common sources. Sulfuric and nitric acids are due, respectively,to the oxidation of sulfur dioxide and nitrogen oxides. These precursors are alsoresponsible for a large fraction of aerosol pollution, and NOx is a main precursor oftropospheric ozone. The major anthropogenic source of sulfur dioxide is electricitygeneration, primarily from coal-fired utility boilers, which also emit mercury.Power plants are also a major source of NOx, along with mobile sources and fires,both of which also emit organic and elemental carbon particles and VOCs.

Second, NOx (as NO2) and SO2 are oxidized in large part by the hydroxylradical, HO•, which also initiates the VOC reactions that drive much pollution, suchas ozone and aerosol formation. Ambient HO• levels are particularly important inaffecting the O3 formation rate in the presence of NOx because reaction with HO•

is a major (and in many cases the only) process that causes most VOCs to react.The HO• is also instrumental in the transformation of other air pollutants, such asoxidizing Hg0 to Hg2+ and initiating atmospheric destruction on POPs. Further,most of the HO• comes from ozone photolysis.

Third, particulate matter and ozone both impact climate (as discussed above).Obviously, these regional pollutant systems—acids, ozone, and particulates—and,to a lesser extent, mercury and POPs are tightly linked chemically and are, indeed,virtually inseparable.

MANAGEMENT OF REGIONAL TRANSBOUNDARYAIR POLLUTION

Whereas meteorological and emission patterns, and other physical causes createdifferent spatial distributions of pollutants, management of air quality is boundedby politics, economics, and cultures. Regional pollution can cross state, provincial,national, and even continental boundaries, and each of these entities can have verydifferent air quality goals and management strategies. Air pollution often has arelatively clear upwind-downwind dimension, in which emissions from a sourcenation (or state) affect air quality in a receptor nation (or state) much more strongly

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


20 BERGIN ET AL.

than vice versa. Part of the downwind nation’s pollution problem, therefore, may bebeyond its jurisdiction to control. This situation is conceptually similar to the caseof a single polluting industry affecting a population, which may be resolved throughgovernment regulation. On an international level, however, a higher governmentauthority is often lacking. A fundamental issue in regional transboundary air qualitycontrol, whether across state or national boundaries, is how to encourage or coerceupwind jurisdictions to reduce emissions, even when they do not perceive it to bein their self-interest.

Most national governments regulate ambient air pollution through the creationand enforcement of air quality standards. In some nations, such as the UnitedStates, an agency of the federal government sets standards and determines andenforces compliance, and state or provincial governments are responsible for im-plementing actions to meet those standards. The goal for state and local air qualitymanagers, then, is to achieve compliance with standards for several pollutants atlow cost (both economically and politically). Air quality goals are balanced againstother considerations, including transportation and energy needs and business andconsumer interests.

Regional air pollution raises problems for such a state-focused system. Upwindstates may have their own motivations for reducing emissions, or they may not. Astate may be influenced so strongly by emissions from other states that even thecomplete elimination of in-state emissions may not be sufficient to achieve stan-dards, especially for secondary pollutants such as ozone and some PM2.5. Suchupwind-downwind disparities can be reinforced by differences in geography, eco-nomic well-being, vested economic interests, and even culture and philosophy.Not surprisingly, densely settled, affluent downwind jurisdictions in violation ofstandards often emphasize their rights to breathe clean air and not be affectedby upwind emissions. Jurisdictions that are upwind, protective of industry, andhave smaller, more dispersed populations tend to emphasize their rights to emit(119). Downwind areas may have already exhausted the most cost-effective con-trols, while many options may remain untapped upwind. An overall most cost-effective strategy might then include upwind controls (J.-S. Shih, M.S. Bergin,A.J. Krupnick, A.G. Russell, unpublished results). As will be discussed below,similar issues arise for international communities, although cooperation is muchmore difficult to attain between areas without a unifying government structure.

A political jurisdiction that is impacted by transboundary air pollution can at-tempt to address the sources of pollution either by taking unilateral actions, byforming a cooperative regime, or (if available) by appealing to a higher govern-mental authority. Unilateral action may take the form of research and educationprograms designed to change awareness, perceptions, and behaviors in the up-wind jurisdiction. Alternatively, a downwind jurisdiction may use the leverageof trade regulations to affect source behavior in upwind jurisdictions. For ex-ample, the European Commission’s proposed chemical legislation REACH mayhelp reduce global sources of certain POPs by restricting access to a large por-tion of the global market, forcing manufacturers to develop alternatives. In most

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



cases, however, decreasing upwind emissions will require the development ofsome type of cooperative regime (120). Here, we use the term regime to mean “so-cial institutions consisting of agreed-upon principles, norms, rules, procedures,and programs that govern the interactions of actors in specific issue areas” (121).That is, a regime encompasses a whole process of interaction, including both sci-entific and political activities, with or without formal agreements. Examples ofsuch cooperative regimes range from technology transfer and research efforts tothe establishment of formal emission reduction obligations or emission tradingmarkets.

In the next section, we present three examples of management programs or co-operative regimes from the United States and Europe where this collective actionproblem has been (at least partially) successfully addressed. We then more gener-ally consider the conditions under which regimes for transboundary air pollutionare most likely to succeed and assess the capabilities and potentials of existingregimes for addressing regional air pollution around the world.

Examples of Transboundary Air Quality ManagementRegimes and Approaches

EMISSION TRADING OF SO2 FOR ACID DEPOSITION CONTROL Arguably one of themost effective regional air pollution control programs to date was also one of themore controversial at the time of inception. Trading of SO2 emissions was intro-duced as part of the 1990 U.S. Clean Air Act Amendments (CAAA) to deal withthe major precursor of acid deposition (as well as of sulfate aerosol) and is a corecomponent of the Acid Rain program. As opposed to the more historic “commandand control” approach by which the government specified the type of controls to beused by individual plants, a market was designed with a fixed number of emissionallowances that could be bought, sold, or traded. Companies that could reduceemissions inexpensively could reduce their emissions to less than their initial allo-cation and sell the extra allowances. Industries that had higher control costs couldbuy these extra allowances, reducing their control needs.

As designed, and as realized in practice, the market seeks the lowest cost ap-proach to achieving the desired emission reductions. The program was designed tocut SO2 emissions from power plants to half of 1980 levels by 2010. By 2003, theprogram had achieved a 38% reduction from 1980 levels (122). Initial estimatessuggested that the cost of the reductions would be between $4 and $8 billion peryear; however, recent estimates suggest that the costs of the program have beenabout $1 billion per year (http://www.etei.org/handbook/chapter 2/09.html). Thereductions achieved through the Acid Rain program will be expanded through theimplementation of the recent Clean Air Interstate Rule, designed to achieve an ad-ditional 67% reduction over 2003 levels (http://www.epa.gov/cair/). The successof this program has spawned similar approaches to dealing with NOx and VOCemissions for ozone control. An emission trading program in Los Angeles uses

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


22 BERGIN ET AL.

trading zones (upwind and downwind) to assure that trading tends to decreaseemissions where reductions are most beneficial (http://www.aqmd.gov/reclaim/).

REGIONAL NOX CONTROL FOR OZONE IN THE UNITED STATES Regional controlof NOx to reduce ozone in the eastern United States is an example of both successand failure. By the early 1990s, the scientific community generally understoodthat ozone could move between states and that controlling emission of NOx wasmost important for addressing transported ozone (14). However, this understandingwas not necessarily shared or considered credible by all federal and state govern-ment actors (124). The Ozone Transport Commission (OTC) was formed in the1990 CAAA to consist of 12 east coast states (Maine to Virginia) and ultimatelyachieved success in cooperatively reducing interstate NOx emissions. Because theozone nonattainment region stretched continuously across many of these (mostlysmall) states, and most states were both upwind and downwind of nonattainmentareas, the need for cooperative action was apparent. The OTC states, with technicalsupport from the EPA, agreed to cap their collective emissions of NOx from largeelectric power plants and to achieve the emission reductions through an internalemission trading program. Reaching this agreement required overcoming concernsabout the effect of a trading system on the locations of emissions and the conse-quent effects on both ozone concentrations and on the economies of the differentstates. The OTC succeeded because it created a framework by which states weremutually assured that all states would be subject to similar controls and that theoverall costs of emission reductions would be reduced through trading. Achiev-ing this agreement took almost a decade and required the formal structure andfrequent technical and political meetings that the OTC provided, which generatedtrust among the participating states (119).

As the OTC efforts were taking shape, additional analysis revealed that emis-sions from states upwind of the OTC states had important effects on ozone in OTCnonattainment areas. In 1995, the Ozone Transport Assessment Group (OTAG),comprised primarily of representatives of the EPA and the 37 states in the easternhalf of the United States, was created to conduct the largest assessment at the time oftransboundary ozone. The main objective of OTAG was to create an opportunity forthe states to cooperatively develop a strategy for addressing regional ozone trans-port, which was considered preferable to the alternative of direct EPA mandates.Ultimately, the differences between upwind and downwind states and the lack ofpolitical leverage to encourage upwind states to reduce emissions were not over-come, and OTAG was able only to reach a weak consensus, recommending a widerange of emission reductions (0% to 80% control) to address the problem. Soon af-ter OTAG ended in 1997, the U.S. EPA exerted its authority to control interstate airpollution transport by requiring 22 states to reduce NOx emissions, primarily fromelectric power plants. Only through the authority of a higher governmental bodycould a regional strategy encompassing more than the OTC states be implemented(119, 124). Recently, the U.S. EPA extended its regional NOx controls in the easternUnited States through the Clean Air Interstate Rule (http://www.epa.gov/cair/).

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



COOPERATIVE AIR POLLUTION CONTROL IN EUROPE The 1979 Long-Range Trans-boundary Air Pollution (LRTAP) Convention is one of the most successful interna-tional regimes addressing environmental problems. Arising out of concern aboutacid deposition in Europe and initially shaped by Cold War politics, its signato-ries include all of the nations of Europe (including the former Soviet Bloc), theUnited States, and Canada. Eight protocols have been successfully negotiated un-der the Convention, including agreements to support scientific cooperation andobligations to reduce emissions related to acidification, tropospheric ozone, POPs,heavy metals, and eutrophication. Recently, LRTAP expanded to include severalof the former Soviet republics in Central Asia.

The success of LRTAP can be attributed, at least partially, to several factors.First, LRTAP constructed a large technical and analytical support structure, includ-ing the Cooperative Program for Monitoring and Evaluation of the Long-RangeTransmission of Air Pollution in Europe (EMEP). This cooperation helped createa shared understanding of the sources and effects of transboundary air pollution,which has been critical to reaching agreements on emission reductions (125). Sec-ond, leadership by individual nations has been important in creating pressure forother nations to follow (125). More recently, the expansion of the European Unionto include more nations in Eastern Europe has added to the atmosphere of coop-eration, as countries that wish to become part of the European Union have soughtopportunities to cooperate with the current E.U. members (126). LRTAP’s successhas taken significant effort and time to achieve. The first several protocols nego-tiated under LRTAP mainly identified problems, established cooperation, and seteasily attainable goals. It was not until the protocols developed in the 1990s thatmore ambitious obligations for emission reductions were agreed upon (127).

Achieving Successful Regimes for TransboundaryAir Pollution

As the previous examples illustrate, forming cooperative international or interstateregimes that include obligations to reduce emissions is an intensive, lengthy pro-cess. This lengthy process is unavoidable because the amount of human resourcesor money is not the deciding factor in reaching an agreement, but “generation ofthe coordination, cooperation, and trust” is needed to create effective management(128).

Although cooperative international or interstate regimes do not always fol-low set patterns, their evolution can often be described in two phases. The firstphase frames the problem—defining the issues, relevant knowledge, and generalrelationships between the parties responsible for and affected by the problem. Co-operative scientific research plays a leading role during this phase by providing acommon understanding of the problem’s physical aspects and by creating a cooper-ative atmosphere. Given that environmental problems are often complex, problemframing also defines the interests and positions of parties, as well as the differ-ences in positions that must be overcome. These differences may be so intractable

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


24 BERGIN ET AL.

that no agreement will result, or the regime may agree to accommodate differ-ent perspectives while maintaining communications. Consequently, agreementsreached within this first phase are typically general, such as setting terms for fur-ther cooperation or codifying positions or laws that parties already hold or arelikely to adopt (127).

The second phase builds on the institutions created in the first phase, al-lowing for the creation of increasingly complex and substantive policy agree-ments, which may include emission reduction obligations. During this phase,emphasis is placed less on international dimensions of the problem and moreon analysis of collectively beneficial, efficient alternatives. By participating in thislengthy process of technical analysis and negotiation, parties may change their per-ceived self-interests, with consequent changes in internal policies and negotiatingpositions (127).

Experience with past regimes for transboundary environmental problems illu-minates several characteristics of successful regimes. As mentioned earlier, coop-eration among technical experts helps foster a shared framing of the problem andbuilds trust. Effective regimes often develop scientific cooperation before politicalnegotiations and build the technical capacity that supports timely analyses. Suchanalyses often emphasize consideration of control scenarios and are most effectivewhen considered to be credible (competent and appropriate for decision needs), le-gitimate (the assessment process is transparent and impartial), and salient (relevantand timely for decision-making needs) (128).

Regimes with fewer parties and with parties that are economically, culturally,or geographically similar tend to be more successful than large heterogeneousregimes. Over time, however, regimes are unlikely to succeed unless they in-volve all major parties contributing to a problem. Parties that share borders or aregional identity often have other reasons for cooperation, unrelated to the envi-ronmental issue at hand. Because the generation of trust is critical for a successfulregime, parties that have a history of cooperating on other issues may be morelikely to succeed. Initially, regimes require the leadership of a sponsor, often aparty affected by upwind emissions who will invest the resources to establish andmaintain institutions. These sponsors have lasting influence on how problems areframed.

Alternatively, large cultural and economic differences can make communica-tion difficult and hinder success. International pollution transport involving bothindustrial and developing nations may expose differences in social priorities, capa-bilities to support emission reductions, technical capacity, and government regula-tory abilities. Participating in regimes may stretch the limited technical resourcesoften found in developing nations, exposing an unfair balance in the negotiationprocess. The most important short-term action in working with developing nationsis to provide support for the technical and management capabilities of these nationson domestic air pollution issues. Doing so will increase the capacities of develop-ing nations to participate meaningfully in international regimes and to address theimportant air pollution issues existing in their own growing urban areas. In many

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



cases, awareness and motivation to address air pollution already exist, but the lackof technical and government capacity limits prospects for emission reductions.Supporting these capacities would allow developing nations to address their worstair pollution problems, with consequent benefits such as reduced long-range airpollution transport (127) and reduced greenhouse gas emissions (135).

Because trust is so important, successful regimes develop both interpersonaltrust between key actors and institutional trust. Institutional trust can be developedthrough leadership that is perceived as impartial, that promotes decision making in aprocess perceived as fair, and that allows for the participation of all relevant parties.Ensuring the accountability of parties in fulfilling commitments, such as throughreporting systems and implementation review bodies, is also vital for maintainingtrust. Management and coordination are also important, and institutions with strongcentral secretariats or administrators are more likely to succeed, as are financiallystrong and technically capable regimes. Institutions should also be able to adapt tochanges in understanding or environmental or political conditions. When naturalor political events or scientific discoveries focus public and political attention onan issue, effective regimes are capable of acting quickly to take advantage of suchwindows of opportunity (127).

Assessment of Existing International Regional Regimes

Although regional transboundary air pollution is also relevant within nations, thissection addresses the major international regimes addressing regional transbound-ary air pollution. We first consider the current status and prospects for futuregrowth of existing regimes in Europe, North America, and Asia and then explainhow regimes can address problems of related transboundary air pollution beyondregional scales.

EUROPE For twenty years, LRTAP has been the dominant forum for addressingtransboundary air pollution in Europe. Since 1999, however, policies set by theEuropean Commission have driven cooperation in Europe. As the European Unioncontinues to expand (http://europa.eu.int/pol/enlarg/overview en.htm), moreLRTAP parties will be subject to policies formed in Brussels. In the meantime,LRTAP will continue to become less relevant as a policy-making institution for theEuropean region.

Given this political reality and the growing scientific evidence that many ofthe pollutants addressed by LRTAP are affected by flows beyond the current geo-graphic scope of the Convention, the LRTAP Executive Body acted in December2004 to form a new EMEP task force to examine the evidence for hemispherictransport of air pollution (www.htap.org). This task force, to be chaired by theUnited States and the European Commission, will seek to build the scientificfoundation for considering future policy initiatives that would involve additionalcountries outside the current geographic scope of the Convention.

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


26 BERGIN ET AL.

NORTH AMERICA The United States and Canada have a long history of coop-eration addressing transboundary air pollution, which began in the 1980s withconcerns over acid rain and led to the 1991 Canada-United States Air QualityAgreement and 2000 Ozone Annex. Similarly, the United States and Mexico haveworked to address issues of air pollution mainly along the border (focusing less onregional transboundary pollutant flows) through the 1983 La Paz Agreement andthe Border 2012 program, which includes agreements and mechanisms to worktoward broad environmental goals and includes funding for pilot projects to reduceemissions in targeted areas. The North American Commission on EnvironmentalCooperation comprises the United States, Canada, and Mexico, and although ithas helped coordinate the exchange of information and continent-wide analyses,it has not served as a forum for negotiating emission reductions. NARSTO (for-merly the North American Strategy on Tropospheric Ozone) was established in1995 with the goals of coordinating policy-relevant scientific research and as-sessment between Mexico, the United States, and Canada and of addressing localand regional air pollution management (129, 130). NARSTO has coordinated andsupported many scientific endeavors, such as NARSTO-NE, EPA Supersites, andSOS.

In North America, emissions in Mexico have little effect on air quality in Canadaand vice versa. Binational agreements between the United States and both Canadaand Mexico will continue to be the focal point of regional air pollution agreements,and we expect these agreements to expand to include more ambitious reductionsand more pollutants. Future efforts may also seek cooperation with the nations ofCentral America and the Caribbean area and may include efforts to build awarenessand capacity in these nations on regional air quality issues. The United Stateshas become increasingly concerned over the effects of forest fire emissions fromCentral America on air quality in the United States.

ASIA Asia is clearly an area of concern regarding the future growth in air pollu-tant emissions. Technical knowledge and experience in managing air pollution isgenerally weak but improving in Asia, and cooperative regimes to address regionalair pollution issues are in their early stages. Three regimes currently exist in eastAsia and south Asia, focusing mainly on supporting rural air quality measure-ments to provide an understanding of regional air pollution. The Association ofSoutheast Asian Nations, composed of 10 nations, developed a 1995 CooperationPlan on Transboundary Pollution. This was followed by a 2002 Agreement onTransboundary Haze Pollution, which was motivated by the prevalence of haze inthis region from large and uncontrolled forest fires. Second, EANET, founded in1998, involves the participation of 12 east Asian nations. Like LRTAP, it mainlyfocuses on measurements and analysis related to SO2 emissions and regional acid-ifying deposition. The United Nations Environment Programme (UNEP) serves acoordinating function, and Japan, which lies downwind of major Asian industrialcenters, provides substantial support. Finally, the Male Declaration on Control andPrevention of Air Pollution and Its Likely Transboundary Effects for South Asia

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



also began in 1998, involving eight nations in South Asia. Again, the focus of theMale Declaration to date has been in capacity building and support of rural airquality measurements.

Although the existence of these three regimes is very encouraging, none ofthem have yet succeeded in finalizing agreements on emission reductions. ManyAsian countries are focused on air pollution problems at the local or urban scale,such as through the World Bank’s Clean Air Initiative for Asian Cities, but trans-boundary air pollution may be considered a low priority. However, these regionalregimes both increase the awareness of transboundary air pollution issues and in-crease general cooperation among these nations. Already, scientific findings fromseveral recent measurement campaigns in Asia, including the Indian Ocean Ex-periment (131), Aerosol Characterization Experiment (ACE)-Asia (38), and ChinaMap (39) have brought a great deal of attention to the “atmospheric brown cloud”and regional air pollution in Asia. Although these studies were enlightening, theywere conducted with little participation from Asian nations and resulted in somecaution and distrust in Asia about the motivations of western scientists. Addition-ally, western scientists have found that the differences in the capacity and interestsof political and scientific communities in some Asian countries obstruct cross-cultural collaborations. As science has recently improved the understanding ofregional transport in Asia, there has been somewhat more of an atmosphere ofblame than of cooperation regarding emission responsibilities.

REGIMES TO ADDRESS POLLUTANT TRANSPORT BEYOND THE REGIONAL SCALE Suc-cessful environmental regimes exist for problems that are clearly identified asglobal, such as the 1992 United Nations Framework Convention on ClimateChange (UNFCCC) with the resulting 1997 Kyoto Protocol and the 1985 ViennaConvention on the Protection of the Ozone Layer with the resulting 1987 MontrealProtocol. None of these agreements, however, addresses the air pollutants discussedin this review (with the exception of methane, which affects tropospheric ozone),even though some have impacts beyond the regional scale. However, two globalregimes been recently been established to address POPs and mercury. The 2001Stockholm Convention on Persistent Organic Pollutants was signed by 151 nationsto bring reductions in emissions of 12 POPs, and the Global Mercury Program wasinitiated by UNEP following the Global Mercury Assessment (132). The UNFCCCand Kyoto Protocol could serve as future forums to address tropospheric ozoneand particulate matter, which affect climate (133). The Intergovernmental Panel onClimate Change, which produces scientific assessments related to the UNFCCC,has begun to focus on the linkages between traditional air pollutants and climatechange.

Institutions are lacking for dealing with the long-range transport of air pollu-tants, such as ozone and fine particles, at the hemispheric or global scale. This gapmay be filled by (a) geographically extending the participation of existing regimes,perhaps by having existing regimes negotiate or combine together, or (b) creat-ing new, larger scale regimes (134). Creating new regimes or expanding existing

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


28 BERGIN ET AL.

ones will take time and progress may be quicker if existing relationships or in-stitutions, such as UNEP, World Meteorological Organization (WMO), or WorldHealth Organization (WHO), can be used as a foundation to begin the building oftrust.

Control of air pollution beyond the regional scale must address emissions fromdeveloping nations. In addition to Asian emissions, emissions from the MiddleEast, Africa, and Latin America are of concern, and the formation and develop-ment of cooperative international regimes in these areas should be encouraged.New regimes have recently been developed in southern Africa (the Air PollutionInformation Network Africa and its Harare Resolution) and in the Middle East(the Bahrain Code of Environmental Conduct for the Middle East).

SUMMARY AND CONCLUSIONS

Although there is still much to be learned about the regional nature of differentair pollutants, our current scientific understanding recognizes that transboundaryair pollution is common and its damage extensive, making political cooperationnecessary, and suggests that future management efforts must integrate the link-ages between pollutants across large geographic scales. Strong linkages betweenthe chemistry of regional ozone, fine particles, mercury, and acid deposition havebeen described above, as well as some common sources and precursors of thesepollutants. For example, the combustion of fossil fuels is a major source of NOX,SO2, anthropogenic VOCs, primary particulates, mercury, and greenhouse gases.Regional air pollution issues are also linked to problems at the local scale, in-cluding “hot spots” near emission sources, and at the global scale, includinglinkages to climate change. These relationships between pollutants across geo-graphic scales offer opportunities to cost-effectively address multiple problemssimultaneously. An important pollution management goal is to develop integratedcontrol strategies incorporating the multiple linkages between pollution problems,the multiple scales over which pollutants cause impacts, the multiple benefits ofalternative technology and behavioral choices, and the linkages between air pol-lution and economic development, energy, agriculture, transportation, and tradepolicies.

Tremendous progress has been made in understanding air pollution issues and,in many industrialized countries, mitigating the most severe impacts; however,significant scientific uncertainties and challenges still remain. These uncertaintieslimit our abilities to quantify the frequency and magnitude of transport events on airquality downwind. Some of the most significant scientific uncertainties could be re-duced by improving emission characterization, especially in developing countries;further elucidating the chemistry and physics of secondary pollutant formation,such as organic aerosol formation; and improving our ability to model source-receptor relationships on regional and global scales. Recent advances in satellite

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



monitoring of air pollutants promise to revolutionize the information available toatmospheric scientists, particularly by improving information in areas remote fromground-based measurements (136).

Prioritizing pollutants for control will require developing a better understandingof the various impacts of pollutants on human health and the environment, whichinclude identifying specific toxicological modes of action and synergistic effectsof multiple exposures. Long-term risks, such as the increasing concentrations ofPOPs in humans (passed through generations), may not be recognized until thedamage is extensive and difficult to mitigate. Historically, ecosystem impacts havereceived less attention than human health, although there are often feedbacks tohuman health and welfare from ecosystem damage. Similarly, the risks associatedwith potential effects of climate change are now driving science and policy makersto better understand relationships between air pollutants and climate. Improvingour ability to quantify the multiple benefits of air pollution control, developingcontrol technologies, and determining cost-optimized control plans will aid policymakers in balancing investments in air pollution control against other economicand development needs.

As industrialized nations in Europe and North America strive to reach stricterdomestic air quality standards, the contributions of upwind sources become rela-tively larger. Consequently, intercontinental transport and increasing global back-grounds of some pollutants will gain attention (137), as will the growing emissionsfrom developing nations. Industrialized nations can be expected to increasinglyengage developing nations in efforts to reduce emissions overseas (127, 134).However, industrialized nations must first help build fundamental air quality man-agement capacity in developing countries, so they can effectively participate inregional cooperative regimes.

There are a number of mechanisms available for addressing regional air pollu-tion. For example, educating citizens and consumers creates an awareness of theneed for cooperative solutions, strengthening the ability of governments to finan-cially commit to control efforts. Educating government leaders about the benefitsof pollution control, as well as about the need to invest in building and maintainingcooperative regimes to achieve their own government’s environmental goals, mayalso be affective. Most importantly, experience both within a nation and internation-ally has demonstrated that the formation of regimes is often necessary for resolvingtransboundary pollution issues. The process of forming regimes and negotiatingagreements often requires significant time and resources. This regime-buildingprocess can be smoothed through scientific cooperation between nations, whichleads to a shared understanding of the transboundary problem. Building fundamen-tal air quality management capacity in developing countries is necessary so theycan effectively participate in regional cooperative regimes. Improving the accessi-bility and sharing of information on emissions, mitigation options, and impacts ofair pollution around the world also aids nations in participating in environmentaldiscussions. Full participation by all affected parties is critical for the success of

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


30 BERGIN ET AL.

addressing a regional air quality problem within a regime. Strengthening existingregional and global international institutions, such as the WMO or UNEP’s regionalcenters, may provide a foundation for future cooperative air pollution efforts.

ACKNOWLEDGMENTS

The authors would like to acknowledge support from the EPA STAR graduate fel-lowship program under grant number U915779; from EPA STAR research grantsR825821, R831076, and R830960; from a National Aeronautics and Space Admin-istration (NASA) New Investigator Program grant NAG5-12542 to D. Mauzerall;and by award NA17RJ2612 from the National Oceanic and Atmospheric Ad-ministration (NOAA), U.S. Department of Commerce. Any opinions, findings,conclusions, or recommendations expressed in this article are the authors and donot necessarily reflect the views of the U.S. EPA, NASA, or NOAA.

The Annual Review of Environment and Resources is online athttp://environ.annualreviews.org

LITERATURE CITED

1. World Health Organ. 2004. Chil-dren’s environmental health, outdoor airpollution. http://www.who.int/ceh/risks/cehair/en/

2. World Health Organ. 2003. Health as-pects of air pollution with particulatematter, ozone and nitrogen dioxide, Rep.WHO Work. Group, Jan. 13–15, WorldHealth Organ., Bonn, Ger. http://www.euro.who.int/document/e79097.pdf

3. Natl. Res. Counc. 2004. Air QualityManagement in the United States. Wash-ington, DC: Natl. Acad.

4. Natl. Res. Counc. 2003. Managing Car-bon Monoxide in Meteorological and To-pographical Problem Areas. Washing-ton, DC: Natl. Acad.

4a. Prinn RG. 2003. The cleansing capacityof the atmosphere. Annu. Rev. Environ.Resour. 28:29–57

5. Seinfeld JH, Pandis SN. 1998. Atmo-spheric Chemistry and Physics. NewYork, NY: Wiley

6. Am. Meterol. Soc. 2003. Acid deposi-tion. http://www.ametsoc.org/POLICY/aciddepo 2003.html

7. US Environ. Prot. Agency. 2002. Air

trends. http://www.epa.gov/air/airtrends/highlights.html

8. Rodhe H, Dentener F, Schulz M. 2002.The global distribution of acidifying wetdeposition. Environ. Sci. Technol. 36:4382–88

8a. Galloway JN, Aber JD, Erisman JW,Seitzinger SP, Howarth RW, et al.2003. The nitrogen cascade. BioScience53(4):341–56

8b. Erisman JW, Grennfelt P, Sutton M.2003. The European perspective on ni-trogen emission and deposition. Envi-ronment International 29:311–25

8c. Driscoll CT, Whitall D, Aber J, BoyerE, Castro M, et al. 2003. Nitrogen Pol-lution in the Northeastern United States:Sources, Effects, and Management Op-tions. BioScience 53(4):357–74

9. Natl. Instit. Public Health Environ.(RIVM), Bilthoven, Neth. 1995. NOx

and SO2 from anthropogenic sources in1995, dataset EDGARV32. http://www.rivm.nl/edgar/model/ims/

10. Jacobson MZ. 2002. Atmospheric Pollu-tion: History, Science, and Regulation.Cambridge, UK: Cambridge Univ. Press

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



11. Cowling E. 1982. Acid precipitationin historical perspective. Environ. Sci.Technol. 16:110A

12. US Environ. Prot. Agency. 1996. AirQuality Criteria for Ozone and OtherPhotochemical Oxidants. Washington,DC: US EPA, Off. Res. Dev.

13. Intergov. Panel Clim. Change (IPCC).2001. Climate change. http://www.ipcc.ch/

14. Natl. Res. Counc. Comm. TroposphericOzone Form. Meas. 1991. Rethinkingthe Ozone Problem in Urban and Re-gional Air Pollution. Washington, DC:Natl. Acad. 524 pp.

15. Bell ML, McDermott A, Zeger SL,Samet JM, Domenici F. 2004. Ozoneand short-term mortality in 95 US urbancommunities 1987–2000. J. Am. Med.Assoc. 292(19):2372–78

16. Levy JI, Carrothers TJ, Tuomisto JT,Hammitt JK, Evans JS. 2001. Assess-ing the public health benefits of reducedozone concentrations. Environ. HealthPerspect. 109:9–20

17. Pope CAI, Burnett RT, Thun MJ, CalleEE, Krewski D, et al. 2002. Lung cancer,cardiopulmonary mortality, and long-term exposure to fine particulate air pol-lution. J. Am. Med. Assoc. 287:1132–41

18. Cent. Disease Control Prev. 1995. Chil-dren at risk from ozone air pollution—United States, 1991–1993. J. Am. Med.Assoc. 273(19):1484–85

19. Horvath SM, McKee DJ. 1994. Acuteand chronic health effects of ozone. SeeRef. 138, pp.39–83

20. US Census Bur. 2002. Population esti-mates. http://www.census.gov/popest/

21. Mauzerall DL, Wang X. 2001. Protect-ing agricultural crops from the effects oftropospheric ozone exposure. Annu. Rev.Energy Environ. 26:237–68

22. Tingey DT, Olszyk DM, Herstrom AA,Lee EH. 1994. Effects of ozone on crops.See Ref. 138, pp. 175–206

23. Chameides WL, Kasibhatla PS, YiengerJ, Levy H. 1994. Growth of continental-

scale metro-agro-plexes, regional ozonepollution, and world food-production.Science 264(5155):74–77

24. McLaughlin SB, Downing DJ. 1995. In-teractive effects of ambient ozone andclimate measured on growth of matureforest trees. Nature 374(6519):252–54

25. Derwent R, Collins W, Johnson C,Stevenson D. 2002. Global ozone con-centrations and regional air quality.Environ. Sci. Technol. 36(19):A379–82

26. Deleted in proof27. Carter WPL. 1994. Development of

ozone reactivity scales for volatile or-ganic compounds. J. Air Waste Manag.Assoc. 44:881–99

28. Bergin MS, Russell AG, Carter WPL,Croes B, Seinfeld JH. 1998. VOC reac-tivity and urban ozone control. In Ency-clopedia of Environmental Analysis andRemediation, ed. RA Meyers, pp. 3355–83. New York, NY: Wiley

29. Wolff GT, Lioy PJ. 1980. Developmentof an ozone river associated with synop-tic scale episodes in the eastern UnitedStates. Environ. Sci. Technol. 14:1257–60

30. Ranzieri AJ, Thullier R. 1991. SanJoaquin Valley air quality study (SJVA-QS) and atmospheric utility signatures,predictions, and experiments (AUS-PEX): a collaborative modeling pro-gram. Presented at 84th Annu. Meet. Ex-hib. Air Waste Manag. Assoc. Vancouver

31. Dye TS, Roberts PT, Korc ME. 1995.Observations of transport processes forozone and ozone precursors during the1991 Lake Michigan Ozone Study. J.Appl. Meteorol. 34(8):1877–89

32. Chameides WL, Cowling EB. 1995. Thestate of the Southern Oxidants Study(SOS): policy-relevant findings in ozonepollution research 1988–1994, South-ern Oxidants Study. Coll. Forest Resour.N.C. State Univ., Raleigh, NC. 136 pp.

33. Odman MT, Boylan JW, WilkinsonJG, Russell AG, Mueller SF, et al.2002. SAMI Air Quality Modeling Final

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


32 BERGIN ET AL.

Report. South. Appalach. Mt. Initiat.,Ashville, NC. http://environmental.gatech.edu/SAMI/Documents/Reports/finalreport.pdf

34. Deleted in proof35. Borrell P, Builtjes P, Grennfelt P, Hov O,

eds. 1997. Photo-oxidants, Acidificationand Tools: Policy Applications of EURO-TRAC Results. Vol. 10. New York, NY:Springer

36. Parrish DD, Holloway JS, Trainer M,Murphy PC, Forbes GL, Fehsenfeld FC.1993. Export of North America ozonepollution to the North Atlantic Ocean.Science 259:1436–39

37. Jacob D, Crawford JH, Kleb MM, Con-nors VS, Bendura RJ, et al. 2003. Trans-port and chemical evolution over thePacific (TRACE-P) aircraft mission:design, execution, and first results. J.Geophys. Res. Atmos.108:1–19

38. Huebert BJ, Bates T, Russell PB, ShiGY, Kim YJ, et al. 2003. An overviewof ACE-Asia: strategies for quantifyingthe relationships between Asian aerosolsand their climatic impacts. J. Geophys.Res. Atmos. 108:8633

39. Chameides WL, Yu H, Liu SC, BerginM, Zhou X, et al. 1999. Case study ofthe effects of atmospheric aerosols andregional haze on agriculture: an oppor-tunity to enhance crop yields in Chinathrough emission controls? Proc. Natl.Acad. Sci. USA 96(24):13626–33

40. Li QB, Jacob DJ, Bey I, Palmer PI, Dun-can BN, et al. 2002. Transatlantic trans-port of pollution and its effects on surfaceozone in Europe and North America. J.Geophys. Res. 107(D13): Artic. 4166

41. Derwent RD, Stevenson DS, CollinsWJ, Johnson CE. 2004. Intercontinentaltransport and the origins of the ozone ob-served at surface sites in Europe. Atmos.Environ. 38:1891–901

42. Jacob D, Logan JA, Murti PP. 1999. Ef-fect of rising Asian emissions on sur-face ozone in the United States. Geo-phys. Res. Lett. 26(14):2175–78

43. Yienger JJ, Galanter M, Hollaway TA,Phadnis MJ, Guttikunda SK, et al. 2000.The episodic nature of air pollutiontransport from Asia to North America.J. Geophys. Res. 105(26):26931–45

44. Fiore AM, Jacob DJ, Bey I, YantoscaRM, Field B, et al. 2002. Backgroundozone over the United States in sum-mer: origin, trend, and contribution topollution episodes. J. Geophys. Res.107(D15): Artic. 4275

45. Prather M, Gauss M, Berntsen T, IsaksenI, Sundet J. 2003. Fresh air in the 21stcentury? Geophys. Res. Lett. 30(2):Ar-tic. 1100

46. US Environ. Prot. Agency. 2004. Airquality criteria for particulate matter.Washington, DC: US EPA, Off. Res.Dev. http://cfpub.epa.gov/ncea/cfm/partmatt.cfm

47. US Environ. Prot. Agency. Off. AirRadiat. 2002. Health and environmentaleffects of particulate matter. http://www.epa.gov/ttn/oarpg/naaqsfin/pmhealth.html

48. Pope CA. 2004. Air pollution andhealth—good news and bad. New Engl.J. Med. 351(11):1132–34

49. Stieb DM, Judek S, Burnett RT. 2003.Meta-analysis of time-series studies ofair pollution and mortality: update in re-lation to the use of generalized addi-tive models. J. Air. Waste Manag. Assoc.53:258–61

50. Kaiser J. 2000. Evidence mounts thattiny particles can kill. Science 289:22–23

51. Guodong Y. 2004. Environmental nano-materials: occurrence, syntheses, char-acterization, health effect, and poten-tial applications. J. Environ. Sci. Health.A. 39(10):2545–48

52. Pekkanen J, Kulmala M. 2004. Exposureassessment of ultrafine particles in epi-demiologic time-series studies. Scand. J.Work Environ. Health. 30:9–18

53. US Environ. Prot. Agency. Off. Air. Ra-diat. 2004. Visibility. What is visibility?

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



http://www.epa.gov/air/visibility/what. html

54. Brasseur GP, Orlando, John J, TyndallGS, eds. 1999. Atmospheric chemistryand global change. In Topics in Envi-ronmental Chemistry, ser. ed. JW Birks.New York, NY: Oxford Univ. Press.654 pp.

55. Aymoz G, Jaffrezo JL, Jacob V, ColombA, George C. 2004. Evolution of organicand inorganic components of aerosolduring a Saharan dust episode observedin the French Alps. Atmos. Chem. Phys.4:2499–512

56. Gerasopoulos E, Andreae MO, ZerefosCS, Andreae TW, Balis D. 2003. Clima-tological aspects of aerosol optical prop-erties in northern Greece. Atmos. Chem.Phys. 3:2025–41

57. Kim E, Hopke PK. 2004. Improvingsource identification of fine particles ina rural northeastern US area utilizingtemperature-resolved carbon fractions.J. Geophys. Res. Atmos. 109(D9) Artic.D09204

58. Liu W, Hopke PK, VanCuren RA. 2003.Origins of fine aerosol mass in the west-ern United States using positive matrixfactorization. J. Geophys. Res. Atmos.108(D23): Artic. 4716

59. Perry KD, Cahill TA, Eldred RA,Dutcher DD, Gill TE. 1997. NorthAfrican dust to the eastern United States.J. Geophys. Res. 102:11225–38

60. Prospero JM. 1999. Long-term measure-ments of the transport of African mineraldust to the southeastern United States:implications for regional air quality. J.Geophys. Res. Atmos. 104(D13):15917–27

61. DeBell LJ, Vozzella M, Talbot RW, DibbJE. 2004. Asian dust storm events ofspring 2001 and associated pollutantsobserved in New England by the Atmo-spheric Investigation, Regional Model-ing, Analysis and Prediction (AIRMAP)monitoring network. J. Geophys. Res.Atmos. 109(D1): D01304, 13 pp.

62. Jaffe DA, Anderson T, Covert D,Kotchenruther R, Trost B, et al. 1999.Transport of Asian air pollution toNorth America. Geophys. Res. Letts.26(6):711–14

63. Husar RB, Tratt DM, Schichtel BA,Falke SR, Li F, et al. 2001. The Asiandust events of April 1998. J. Geophys.Res. 106:18317–30

64. VanCuren RA, Cahill TA. 2002. Asianaerosols in North America: frequencyand concentration of fine dust. J. Geo-phys. Res. 107(D24):4804–21

65. Grousset F, Ginoux P, Bory A, Bis-caye P. 2003. Case study of a Chi-nese dust plume reaching the FrenchAlps. Geophys. Res. Lett. 30(6):1277–80

66. Penner JE, Andreae M, Annegarn H,Barrie L, Feichter J, et al. 2001.Aerosols, their direct and indirect ef-fects. In Climate Change 2001: TheScientific Basis, Intergov. Panel Clim.Change (IPCC), ed. JT Houghton, YDing, DJ Griggs, M Noguer, PJ van derLinden, et al., pp. 289–348. Cambridge,UK: Cambridge Univ. Press

67. Takemura T, Uno I, Nakajima T,Higurashi A, Sano I. 2002. Model-ing study of long-range transport ofAsian dust and anthropogenic aerosolsfrom East Asia. Geophys. Res. Lett.29(24):2158, 4 pp.

68. Wotama G, Trainer M. 2000. The influ-ence of Canadian forest fires on pollutantconcentrations in the United States. Sci-ence 288:324–28

69. DeBell LJ, Talbot RW, Dibb JE, MungerJW, Fischer EV, Frolking SE. 2004. Amajor regional air pollution event in thenortheastern United States caused by ex-tensive forest fires in Quebec, Canada. J.Geophys. Res. Atmos. 109:D19305, 16pp.

70. Tanner RL, Parkhurst WJ, Valente ML,Humes KL, Jones K, Gilbert J. 2001.Impact of the 1998 Central Americanfires on PM2.5 mass and composition in

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


34 BERGIN ET AL.

the southeastern United States. Atmos.Environ. 35:6539–47

71. Rogers CM, Bowman KP. 2001. Trans-port of smoke from the Central Ameri-can fires of 1998. J. Geophys. Res. 106D(22):28357–68

72. US Environ. Prot. Agency. 2003.Latest findings on national air qual-ity, 2002 status and trends. EPA 454/K-03-001. US EPA Off. Air Qual. Stand.,Research Triangle Park, NC. http://oncampus.richmond.edu/∼cstevens/EnviroWeb/Documents/USAirQuality2002.pdf

73. Hillger DW, Ellrod GP. 2003. Detectionof important atmospheric and surfacefeatures by employing principal com-ponent image transformation of GOESimagery. J. Appl. Meteorol. 42(5):611–29

74. Falke SR, Husar RB, Schichtel BA.2001. Fusion of SeaWiFS and TOMSsatellite data with surface observationsand topographic data during extremeaerosol events. J. Air. Waste Manag. As-soc. 51(11):1579–85

75. Husar RB, Tratt DM, Schichtel BA,Falke SR, Li F, et al. 2001. Asian dustevents of April 1998. J. Geophys. Res.Atmos. 106(D16):18317–30

76. Deleted in proof77. Ames RB, Malm WC. 2001. Com-

parison of sulfate and nitrate particlemass concentrations measured by IM-PROVE and the CDN. Atmos. Environ.35(5):905–16

78. Sister JF, Malm WC. 2000. Interpreta-tion of trends of PM2.5 and reconstructedvisibility from the IMPROVE network.J. Air. Waste Manag. Assoc. 50(5):775–89

79. Baumgardner RE, Isil SS, Bowser JJ,Fitzgerald KM. 1999. Measurements ofrural sulfur dioxide and particle sul-fate: Analysis of CASTNet data, 1987through 1996. J. Air. Waste Manag. As-soc. 49(11):1266–1279

80. Irving PM, ed. 1991. Acidic deposi-

tion: state of science and technology.Summary Report of the U.S. NationalAcid Precipitation Assessment Program.Washington, DC: US Gov. Print. Off.

81. Querol X, Alastuey A, Ruiz CR, Arti-nano B, Hansson HC, et al. 2004. Speci-ation and origin of PM10 and PM2.5 in se-lected European cities. Atmos. Environ.38(38):6547–55

82. Dutkiewicz VA, Qureshi S, Khan AR,Ferraro V, Schwab J, et al. 2004. Sourcesof fine particulate sulfate in New York.Atmos. Environ. 38(20):3179–89

83. Park RJ, Jacob DJ, Field BD, Yan-tosca RM, Chin M. 2004. Natural andtransboundary pollution influences onsulfate-nitrate-ammonium aerosols inthe United States: implications for pol-icy. J. Geophys. Res. 109(D15):D15204,20 pp.

84. US Geol. Surv. 2000. Mercury in theenvironment, in fact sheet 146-00. http://www.usgs.gov/themes/factsheet/146-00/

85. Natl. Res. Counc. 2000. ToxicologicalEffects of Methylmercury. Washington,DC: Natl. Acad.

86. Little M. 2002. Reducing mercury pol-lution from electric power plants. Is-sues Sci. Technol. (Summer) http://www.issues.org/issues/18.4/

87. US EPA Off. Water. 2004. National list-ing of fish advisories, fact sheet EPA-823-F-04-016. http://www.epa.gov/waterscience/fish/advisories/factsheet.pdf

88. Environ. Can. 2004. Mercury and the en-vironment, environment and health, bio-geochemistry. http://www.ec.gc.ca/MERCURY/EH/EN/eh-b.cfm?SELECT= EH

89. UN Eviron. Programme (UNEP). 2003.Global mercury assessment report.http://www.chem.unep.ch/mercury/

90. Pal B, Ariya PA. 2004. Gas-phase HOcenter dot-initiated reactions of elemen-tal mercury: kinetics, product studies,and atmospheric implications. Environ.Sci. Technol. 38(21):5555–66

91. Shia RL, Seigneur C, Pai P, Ko M,Sze ND. 1999. Global simulation of

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



atmospheric mercury concentrations anddeposition fluxes. J. Geophys. Res.Atmos. 104(D19):23747–60

92. Seigneur C, Vijayaraghavan K, LohmanK, Karamchandani P, Scott C, Levin L.2003. Simulation of the fate and trans-port of mercury in North America. J.Physique. 107:1209–12

93. Ryaboshapko A, Kullock R, EbinghausR, Ilyin I, Lohman K, et al. 2002. Com-parison of mercury chemistry models.Atmos. Environ. 36(24):3881–98

94. Sommar J, Gardfeldt K, Stromberg D,Feng XB. 2001. A kinetic study of thegas-phase reaction between the hydroxylradical and atomic mercury. Atmos. En-viron. 35(17):3049–54

95. Seigneur C, Vijayaraghavan K, LohmanK, Karamchandani P, Scott C. 2004.Global source attribution for mercury de-position in the United States. Environ.Sci. Technol. 38(2):555–69

96. Renner R. 2004. Rethinking atmosphericmercury. Environ. Sci. Technol. (OnlineNews, Science News) http://pubs.acs.org/subscribe/journals/esthag-w/2004/oct/science/rr rethinking.html

97. Agency Toxic Subst. Disease Regist.(ATSDR). 1999. Toxicological Profilefor Mercury. Atlanta, GA: US Dep.Health Hum. Serv./Public Health Serv.

98. US EPA. 2004. Mercury. http://www.epa.gov/mercury/index.html

99. Deleted in proof100. Pacyna JM, Pacyna EG, Steenhuisen

F, Wilson S. 2003. Mapping 1995global anthropogenic emissions of mer-cury. Atmos. Environ. 37(Suppl. 1):109–17

101. Munthe J, Palm A. 2003. Atmosphericcycling of mercury and persistent or-ganic pollutants. Overview of subprojectMEPOP. In Towards Cleaner Air for Eu-rope: Science, Tools and Applications,http://www.gsf.de/eurotrac/publications/et2 fin rep/fr 10 mepop.pdf

102. Veiga MM, Baker R. 2004. Protocolsfor environmental and health assess-

ment of mercury released by artisanaland small scale miners. See Ref. 139,170 pp

103. Hinton J, Veiga M. 2004. Technicaland socio-economic profiles of globalmercury project sites. See Ref. 139,56 pp.

104. US Environ. Prot. Agency. 1997. Fateand transport of mercury in the envi-ronment, Vol. III. In Mercury Study Re-port to Congress, 376 pp. US EPA Off.Air Qual. Planning Stand./Off. Res. Dev.http://www.epa.gov/mercury/report.htm

105. Meteorol. Synth. Cent.—East (EMEP).2004. Detailed reports for each EMEPcountry. http://www.msceast.org/countries/index.html

106. UN Environ. Programme. 2004. Chemi-cals. Persistent organic pollutants (PO-Ps). http://www.chem.unep.ch/pops/

107. Hooper K, McDonald TA. 2000. ThePBDEs: an emerging environmentalchallenge and another reason for breast-milk monitoring programs. Environ.Health Perspect. 108(5):387–92

108. World Bank. 2004. Persistent organicpollutants. http://lnweb18.worldbank.org/ESSD/envext.nsf/50ByDocName/WhatArePOPs

109. Solomon GMS. Cent. Health Glob.Environ., Harvard Med. School. 2001.Persistent organic pollutants: humanhealth issues, ed. S Mahalingaiah. http://www.med.harvard.edu/chge/textbook/toxic/organic/transcript.htm

110. Hooper K. 1999. EnvirBreast milk mon-itoring programs (BMMPs): world-wideearly warning system for polyhalo-genated POPs and for targeting stud-ies in children’s environmental health.Environ. Health Perspect. 107(6):429–30

111. US Environ. Prot. Agency. 2004.Pesticides—persistent organic pollu-tants (POPs). http://www.epa.gov/oppfead1/international/pops.htm

111a. Meteorol. Synth. Cent.—East (EMEP).2004. Depositions and transboundary

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


36 BERGIN ET AL.

fluxes of B[a]P. Detailed informationfor France for 2004. http://www.msceast.org/countries/France/index.html

112. Rodan BD. 2002. The Foundation forGlobal Action on Persistent Organic Pol-lutants: A United States Perspective.Washington, DC: US Environ. Prot.Agency/Off. Res. Dev.

113. UN Environ. Programme. 2001. UNEPChemicals. Stockholm convention onPOPs. http://www.chem.unep.ch/pops/POPs Inc/dipcon/sto-eng.pdf

114. Eur. Comm. 2004. REACH homepage.http://europa.eu.int/comm/enterprise/reach/index.htm

115. Chameides WL, Bergin M. 2002. Cli-mate change—soot takes center stage.Science 297(5590):2214–15

116. Chameides WL, Luo C, Saylor R, StreetsD, Huang Y, et al. 2002. Correlationbetween model-calculated anthropoge-nic aerosols and satellite-derived cloudoptical depths: indication of indirecteffect? J. Geophys. Res. Atmos. 107-(D10): Artic. 4085

117. Menon S, Hansen J, Nazarenko L, LuoY. 2002. Climate effects of black car-bon aerosols in China and India. Science297(5590):2250–53

118. Hogrefe C, Lynn B, Civerolo K, KuJY, Rosenthal J, et al. 2004. Simulat-ing changes in regional air pollutionover the eastern United States due tochanges in global and regional climateand emissions. J. Geophys. Res. Atmos.109(D22): D22301

119. Farrell AE. 2001. Multi-lateral emissiontrading: lessons from inter-state NOx

control in the United States. Energy Pol-icy 29(13):1061–72

120. Ostrom E. 1990. Governing the Com-mons: The Evolution of Institutionsfor Collective Action. New York, NY:Cambridge Univ. Press

121. Young OR, ed.1999. The Effectiveness ofInternational Environmental Regimes:Causal Connections and BehavioralMechanisms. In Global Environmen-

tal Accord: Strategies for Sustainabilityand Institutional Innovation, ser. ed. N.Choucri. Cambridge, MA: MIT Press.326 pp

122. US Environ. Prot. Agency. 2004. Acidrain program 2003 progress report. EPA430-R-04-009. Clean Air Mark. Div.,Off. Air Radiat., Washington, DC

123. Deleted in proof124. Farrell AE, Keating TJ. 2002. Trans-

boundary environmental assessments:lessons from OTAG. Environ. Sci.Technol. 36(12):2537–44

125. Levy MA. 1993. European acid rain:the power of tote-board diplomacy. InInstitutions for the Earth: Sources ofEffective International EnvironmentalProtection, ed. PM Haas, RO Keo-hane, MA Levy, pp. 75–132. MIT Press:Cambridge, MA

126. Van Deveer SD. 2005. European pol-itics with a scientific face: framing,national participation, and capacity inLRTAP. In Assessments of Regionaland Global Environmental Risks: De-signing Processes for Effective Useof Science in Decisionmaking, ed. AFarrell, J Jaeger. Washington, DC: RFF.In Press

127. Keating TJ, West JJ, Farrell A. 2004.Prospects for international managementof intercontinental air pollutant trans-port. In Intercontinental Air PollutionTransport, ed. A Stohl, pp. 295–320.Berlin: Springer-Verlag

128. Clark W, Dickson N, eds. 2001. Learn-ing to Manage Global EnvironmentalRisks: A Functional Analysis of SocialResponses to Climate Change, OzoneDepletion, and Acid Rain. Vols. 1, 2.Cambridge, MA: MIT Press

129. NARSTO. 2000. An assessment of tro-pospheric ozone pollution. EPRI Rep.100040, Palo Alto, CA

130. NARSTO. 2003. Particulate matter sci-ence for policy makers. EPRI Rep.1007735, Palo Alto, CA

131. Lelieveld J, Crutzen PJ, Ramanathan

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.



V, Andreae MO, Brenninkmeijer CAM,et al. 2001. The Indian Ocean exper-iment: widespread air pollution fromSouth and Southeast Asia. Science. 291:1031–36

132. UN Environ. Programme. 2002. Chem-icals Program: Global Mercury Assess-ment Geneva, Switz.: UNEP

133. Rypdal K, Berntsen T, Fuglestvedt S,Aunan K, Torvanger A, et al. 2005.Ozone and aerosols in future climateagreements: scientific and political chal-lenges. Environ. Sci. Policy. 8(1):29–43

134. Holloway T, Fiore AM, Hastings MG.2003. Intercontinental transport of airpollution: Will emerging science lead toa new hemispheric treaty? Environ. Sci.Technol. 37:4535–42

135. West JJ, Osnaya P, Laguna I, Martinex J,Fernandez A. 2004. Co-control of urbanair pollutants and greenhouse gases in

Mexico City. Environ. Sci. Technol. 38:3474–81

136. Natl. Res. Counc. 2001. Global AirQuality: An Imperative for Long-termObservational Strategies. Washington,DC: Natl. Acad.

137. Stohl A, ed. 2004. IntercontinentalTransport of Air Pollution. New York:Springer-Verlag

138. McKee DJ, ed. 1994. TroposphericOzone: Human Health and AgriculturalImpacts. Boca Raton, FL: CRC Press/Lewis. 333 pp.

139. UN Ind. Dev. Organ. (UNIDO). 2002.Removal of Barriers to Introductionof Cleaner Artisanal Gold Mining andExtraction Technologies. Global Mer-cury Project EG/GLO/01/G34.Vienna,Austria: Glob. Environ. Facil./UN Dev.Programme. http://www.unites.uqam.ca/gmf/intranet/gmp/documents/documents.htm

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.

REGIONAL AIR POLLUTION C-1

Figure 1 Observed values of volume-weighted annual average pH of precipitation.Reproduced with permission from Reference 8.

Figure 2 NOx from anthropogenic sources in 1995. Reproduced with permissionfrom Reference 9.

HI-RES-EG30-01-Bergin.qxd 10/7/05 5:56 PM Page 1

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.

C-2 BERGIN ET AL.

Figure 3 SO2 from anthropogenic sources in 1995. Reproduced with permissionfrom Reference 9.

Figure 5 Regional ozone and sensitivity to emissions. (a) Simulated one-hour aver-aged ozone concentration and (b) its sensitivity to a 30% reduction in surface NOxemissions from Tennessee (M.S. Bergin, J.-S. Shih, J.W. Boylan, J.G. Wilkinson,A.J. Krupnick, A.G. Russell, unpublished results).


Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


Figure 6 Satellite image from the Earth Probe Total Ozone Mapping Spectrometerof smoke/dust (absorbing aerosols) over North America on May 15, 1998 (72).Image developed by the Earth Observatory team, GSFC/916, and reproduced withpermission.


Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.

C-4 BERGIN ET AL.

Fig

ure

7 R

egio

nal p

artic

ulat

e m

atte

r an

d se

nsiti

vity

to e

mis

sion

s. (

a) S

imul

ated

24-

hour

ave

rage

dPM

2.5

conc

entr

atio

n an

d (b

) its

sen

sitiv

ity t

o a

30%

red

uctio

n in

SO

2em

issi

ons

from

Ten

ness

ee(M

.S.

Ber

gin,

J.-S

. Sh

ih,

J.W

. B

oyla

n,J.

G.

Wilk

inso

n,A

.J.

Kru

pnic

k,A

.G.

Rus

sell,

unpu

blis

hed

resu

lts).


Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


Fig

ure

8

C

once

ptua

l bi

ogeo

chem

ical

m

ercu

ry

cycl

e.

Rep

rodu

ced

wit

h pe

rmis

sion

fr

omR

efer

ence

88.


Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.

C-6 BERGIN ET AL.

Fig

ure

9

Est

imat

ed g

loba

l em

issi

ons

of m

ercu

ry i

n 19

95.

Rep

rodu

ced

with

per

mis

sion

fro

mR

efer

ence

100

.


Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.


Figure 10 Modeled benzo[a]pyrene (BaP, a PAH) depositions from emissions inFrance in 2001, g/km2/year. Reproduced with permission from Reference 111a.


Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.

P1: JRX

September 27, 2005 15:29 Annual Reviews AR256-FM

Annual Review of Environment and ResourcesVolume 30, 2005

CONTENTS

I. EARTH’S LIFE SUPPORT SYSTEMS

Regional Atmospheric Pollution and Transboundary Air QualityManagement, Michelle S. Bergin, J. Jason West, Terry J. Keating,and Armistead G. Russell 1

Wetland Resources: Status, Trends, Ecosystem Services, and Restorability,Joy B. Zedler and Suzanne Kercher 39

Feedback in the Plant-Soil System, Joan G. Ehrenfeld, Beth Ravit,and Kenneth Elgersma 75

II. HUMAN USE OF ENVIRONMENT AND RESOURCES

Productive Uses of Energy for Rural Development,R. Anil Cabraal, Douglas F. Barnes, and Sachin G. Agarwal 117

Private-Sector Participation in the Water and Sanitation Sector,Jennifer Davis 145

Aquaculture and Ocean Resources: Raising Tigers of the Sea,Rosamond Naylor and Marshall Burke 185

The Role of Protected Areas in Conserving Biodiversity and SustainingLocal Livelihoods, Lisa Naughton-Treves, Margaret Buck Holland,and Katrina Brandon 219

III. MANAGEMENT AND HUMAN DIMENSIONS

Economics of Pollution Trading for SO2 and NOx , Dallas Burtraw,David A. Evans, Alan Krupnick, Karen Palmer, and Russell Toth 253

How Environmental Health Risks Change with Development: TheEpidemiologic and Environmental Risk Transitions Revisited,Kirk R. Smith and Majid Ezzati 291

Environmental Values, Thomas Dietz, Amy Fitzgerald, and Rachael Shwom 335

Righteous Oil? Human Rights, the Oil Complex, and Corporate SocialResponsibility, Michael J. Watts 373

Archaeology and Global Change: The Holocene Record,Patrick V. Kirch 409

ix

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.

P1: JRX

September 27, 2005 15:29 Annual Reviews AR256-FM

x CONTENTS

IV. EMERGING INTEGRATIVE THEMES

Adaptive Governance of Social-Ecological Systems,Carl Folke, Thomas Hahn, Per Olsson, and Jon Norberg 441

INDEXES

Subject Index 475Cumulative Index of Contributing Authors, Volumes 21–30 499Cumulative Index of Chapter Titles, Volumes 21–30 503

ERRATA

An online log of corrections to Annual Review of Environment andand Resources chapters may be found at http://environ.annualreviews.org

Ann

u. R

ev. E

nvir

on. R

esou

rc. 2

005.

30:1

-37.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Mos

cow

Sta

te U

nive

rsity

- S

cien

tific

Lib

rary

of

Lom

onos

ov o

n 10

/07/

13. F

or p

erso

nal u

se o

nly.