Recovery of ACIDIFIED European Surface Waters

RICHARD F. WRIGHT

THORJØRN L ARSSEN NORWEGIAN INSTITUTE FOR WATER

RESEARCH

LLUIS CAMARERO CENTRE OF ADVANCED STUDIES (SPAIN)

BERNARD J. COSBY UNIVERSITY OF VIRGINIA

ROBERT C. FERRIER

R ACHEL HELLIWELL MACAULAY INSTITUTE (U.K.)

MARTIN FORSIUS FINNISH ENVIRONMENT INSTITUTE

AL AN JENKINS CENTRE FOR ECOLOGY AND HYDROLOGY

(U.K.)

JIRI KOPÁ CEK HYDROBIOLOGICAL INSTITUTE (CZECH

REPUBLIC)

V L ADIMIR MAJER CZECH GEOLOGICAL SURVEY

FILIP MOLDAN SWEDISH ENVIRONMENTAL RESEARCH

INSTITUTE

MA XIMILIAN POSCH DUTCH NATIONAL INSTITUTE FOR PUBLIC

HEALTH AND THE ENVIRONMENT

MICHEL A ROGOR A CNR INSTITUTE OF ECOSYSTEM STUDY

(ITALY)

WOLFGANG SCHÖPP INTERNATIONAL INSTITUTE FOR APPLIED

SYSTEMS ANALYSIS (AUSTRIA)

To abate acid rain, policy makers can use results of

acidification models to predict future recovery.

Acid rain! Dead fish! Forest dieback! In the 1980s and 1990s, these headlines ap-

peared frequently in environmental news coverage in Europe and North America.

Air pollutants from highly industrialized regions had caused widespread damage

to pristine ecosystems far downwind. The victims—people living in regions such

as eastern Canada and Norway—pressured the polluters. In Europe, 30 countries engaged

in tough negotiations that finally resulted in international treaties to reduce the emissions

of sulfur and nitrogen oxides. Acid deposition has now declined by ~60%, and some lakes

and streams have begun to recover. Will all waters recover, or must emissions be reduced

even more? And how long will recovery take? In this article, we try to answer these ques-

tions by using models to predict future acidification of surface waters in 12 acid-sensitive

regions in Europe.

A short historyIn the 1970s, the situation was bleak indeed. Acidified lakes and streams were report-

ed from many regions in Europe, including the uplands of the United Kingdom, southern

Fennoscandia, the low mountain ranges in Germany and the Czech Republic, and alpine ar-

eas in central and southern Europe. Norway and Sweden were especially hard hit, with trout

populations lost in thousands of lakes and salmon stocks eradicated from many major rivers

(1). And the situation was getting worse; every year brought reports of new areas affected.

© 2005 American Chemical Society FEBRUARY 1, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY ■ 65A

RICH

ARD

F. W

RIG

HT

Many years of hard scientific detective work were required to explain why fish were dying in re-mote areas, such as southern Norway. The break-through came when scientists realized that the key to understanding surface-water acidification lay in the amount and source of the accompanying an-ion rather than in the acid itself. Two factors proved necessary to explain surface-water acidification: The water must be acid-sensitive, and the area must receive sufficient amounts of acid deposition (2).

Acid-sensitive lakes and streams are found throughout the world in catchments with weather-ing-resistant bedrock, such as granite and quartzite, and young, often poorly developed, podzolic, and organic-rich soils (3). In these waters, the dominant anion is usually the weak-acid anion bicarbonate (HCO3

–), whose source is CO2 from decomposition of organic matter, respiration by plant roots, and dis-solution in soil water. The base cations calcium and magnesium generally accompany HCO3

–. The concen-trations of all three ions depend on how easily weath-ering breaks down the soil minerals. Acid-sensitive waters have low concentrations of all these ions.

The second factor is acid deposition. The most sensitive waters are affected when the rain is more acidic than pH ~4.7 and sulfate (SO4

2–) concentrations exceed ~20 microequivalents per liter (µeq/L). In acidified waters, the strong-acid anion SO4

2– is usually the dominant ion and replaces HCO3

–. Acidified, SO42–-

rich waters often have pH <5 and elevated concentra-tions of inorganic aluminum species (Aln+). The acid and inorganic aluminum species are toxic to fish and other aquatic organisms.

Acid neutralizing capacity (ANC) is a chemi-cal criterion widely used to describe the acid–base status of water (4). ANC is defined as the equiva-lent sum of base cations (Ca2+, Mg2+, Na+, K+) minus the equivalent sum of strong-acid anions (SO4

2–,Cl–,

NO3–). For example, HCO3

–-dominated waters have positive ANC, whereas acidified waters have nega-tive ANC. ANC level is a good indicator of biological status; empirical dose–response relationships have been developed for fish (5), macroinvertebrates (6), and diatoms (7). An ANC threshold of 20 µeq/L marks water quality sufficient for protecting most key indicator organisms.

Establishing protocolsIn Europe, acid rain is clearly a problem that calls for an international solution, because the air pollutants are carried long distances and across national bor-ders. In 1979, negotiations to reduce the emissions of air pollutants began under the auspices of the UN Economic Commission for Europe with the estab-lishment of the Convention on Long-Range Trans-boundary Air Pollutants (CLRTAP) (8, 9). Work under CLRTAP has produced a series of protocols in which countries agreed to reduce emissions of sulfur and nitrogen compounds.

The latest protocol, signed in 1999 in Gothenburg, Sweden, is a multi-pollutant, multi-effect measure in which acidification of surface waters was one of several effects considered. (As of November 25, 2004, 49 countries had signed the protocol, but only 14 had ratified it. Sixteen ratifications are required before the protocol goes into effect.) The protocol uses the critical load concept and seeks to mini-mize—if implemented as proposed by 2010—the number of ecosystems in which the critical load will be exceeded (8). Specifically, the protocol calls for ~80% reduction in sulfur emissions and 50% reduc-tion in nitrogen emissions by 2010, relative to 1980 levels.

Emissions of acidifying gases in Europe peaked in the 1970s, just before protocols were implement-ed. In the 1990s, surface waters in Europe showed the first signs of recovery in response to lower levels of acid deposition; SO4

2– concentrations decreased, pH levels and ANC concentrations increased, and concentrations of Aln+ decreased (10, 11; Figure 1). By 2000, sulfur deposition had decreased by >50% and nitrogen by ~20% (12). Further decreases are predicted for the next 20 years if the Gothenburg Protocol and other national legislation are imple-mented (Figure 2). The waters are becoming less toxic for fish and other aquatic organisms, but com-plete recovery is a long way off.

Modeling and mechanismsIf the Gothenburg Protocol is implemented, will acidified waters recover or are additional reduc-tions required? And how long will the recovery take? Models for acidification of soil and water are required to address these questions (12). Two such models are Model for Acidification of Groundwater in Catchments (MAGIC; 13) and Simulation Model for Acidification’s Regional Trends (SMART; 14). These models are consistent in structure and out-put (15) and have been extensively tested and eval-uated (16). Researchers involved with two large EU projects, European Mountain Lake Ecosystems: Regionalisation, Diagnostic and Socio-economic

1880

70

60

50

40

30

20

10

1910 1940

Year

1970 2000 2030

Mt/y

r

F I G U R E 1

Sulfur and nitrogen emissions in Europe (1880–2030)The rise and fall of emissions in Europe over the period 1880–2030 as estimated by Schöpp et al. (12 ). Units are megatonne/yr of SO2 (black), NO2 (green), and NH3 (pink).

66A ■ ENVIRONMENTAL SCIENCE & TECHNOLOGY / FEBRUARY 1, 2005

Evaluation (EMERGE; 17) and Predicting Recovery in Acidified Freshwaters by the Year 2010, and Be-yond (RECOVER:2010; 18) have recently used these models to predict the future status of acid-sensitive surface waters in 12 European regions (Table 1; see next page).

The models were first calibrated to present-day deposition and water and soil chemistry at each site. Deposition sequences for sulfur and nitro-gen compounds from 1860 to 2020 were used to scale through time. For predictions through 2016, the reductions in deposition of sulfur, oxidized ni-trogen species (NOx), and reduced nitrogen spe-cies (NHy) were calculated using the RAINS model (19) on the 150 × 150 km Cooperative Programme for Monitoring and Evaluation of the Long-Range Transmission of Air Pollutants in Europe (EMEP) grid (Table 1; 12). The predictions assumed that the Gothenburg Protocol and other current legislation will be implemented by 2010, as agreed. For predic-tions beyond 2010, modelers assumed that deposi-tion would remain constant.

As expected, the model simulations showed that in 1860, prior to the onset of acid deposition, very few waters were acidic and most waters had ANC concen-trations >20 µeq/L (Figure 3; see p 69A). Extensive historical and paleolimnological records for fish and other acid-sensitive aquatic organisms also indicate that acidification was not an environmental problem before the early 1900s—even in extremely sensitive regions, such as southern Norway (1).

In the regions that received acid deposition ex-ceeding the threshold pH 4.7, model simulations show that waters became increasingly acidified dur-ing the 1900s. By 1980, >25% of the waters in 8 of the 12 regions were acidified to ANC < 20 µeq/L and in 5 of these 8 regions to ANC < 0 µeq/L. Conditions were clearly harmful for fish; again, this agrees well with the widespread reports at the time of damage to fish and other organisms. In contrast, northern Sweden, northern Scotland, and the Pyrenees had relatively low levels of acid deposition, with ANC < 20 µeq/L at only a few sites (Figure 3).

For the period 1980–2000, extensive measure-ments document the state and geographical extent of freshwater acidification and the onset of chemi-cal recovery as sulfur deposition began to decrease (10, 11). By the year 2000, only a few of the sites mod-eled here still had a large fraction of waters with ANC < 20 µeq/L.

The model simulations predict that recovery will continue and that by 2016 most waters, except those in southernmost Norway, will have ANC > 20 µeq/L. However, these reductions do not mean that the waters will have returned to their pre-acidifica-tion state. After 2010, small but significant deposi-tion of sulfur and nitrogen will remain. In addition, decades of acid deposition have reduced the soils’ ability to neutralize acid deposition.

Two mechanisms change surface-water ANC during the acidification and recovery phases. The first mechanism is a concentration effect; changes in cation concentrations are required to balance the changes in concentrations of strong-acid anions.

The more acidic the soil (i.e., the lower the percent base saturation; %BS), the greater will be the frac-tion of acid cations (4, 20). The second mechanism is related to the size of the pool of exchangeable base cations. Base cation inputs from weathering and atmospheric deposition add to this pool, and net uptake by vegetation and loss to runoff deplete the pool. An increase in the flux of SO4

2– in runoff will boost the flux of base cations out of the soil. If these base ions are not replenished, the soil pool of

1850

(a)

(b)

(c)

(d)

250

200

150

100

50

01900 1950 2000 2050 2100

Sulfa

te (

eq/m

2 yr)

1850

1850 1900 1950 2000 2050 2100

300

250

200

150

100

50

0

100

50

0

–50

–100

–150

–200

–250

1900 1950 2000 2050 2100

Sulfa

te (

eq/L

)AN

C (

eq/L

)

1850

181614121086420

1900 1950 2000 2050 2100

%BS

F I G U R E 2

An example of acidification and recoveryThe MAGIC model (solid line) was used to reconstruct the acidifica-tion history and to predict the future recovery given implementation of the Gothenburg Protocol as it applies to Birkenes, a small acid-sensitive stream in southern Norway. Measurements (points) car-ried out since 1972 confirm significant recovery as of the mid-1980s in (a) sulfate deposition, (b) streamwater sulfate, (c) streamwater acid neutralizing capacity, and (d) soil percent base saturation.

FEBRUARY 1, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY ■ 67A

exchangeable base cations will decrease over the long term. Depleting and rebuilding of the base cat-ion pool takes decades and occurs only when the rate of base cation weathering and deposition input exceeds the rate of depletion of the base cation pool due to the leaching of strong acids.

The model simulations indicate that, as expect-ed, the %BS decreased during the long period of acidification of 1860–1980. Between 1980 and 2000, the large reductions in sulfur deposition appeared in most cases to be sufficient to stop the decrease in %BS but still insufficient to allow %BS to recover. And the prognosis for the future is also bleak: Little or no recovery of base saturation in the soil is expected, and in the Tatra Mountains in Slovakia the soil continues to acidify (Figure 4; see p 70A).

Other factorsOf course the picture is not that simple, and several important caveats accompany these general con-clusions. First, although sulfur deposition has caused most surface-water acidification across Europe, ni-trogen is also an important component. On a mo-lar basis, nitrogen deposition in Europe is nearly as large as that of sulfur and has become more im-portant because sulfur emissions have decreased much more rapidly than nitrogen emissions (12). In contrast to sulfur, nitrogen is usually strongly re-tained in terrestrial ecosystems; typically <10% is leached in runoff, mostly as NO3

–, which can acidify soil and water just like SO4

2–. A real danger exists that chronic nitrogen deposition may saturate terrestri-al ecosystems, a condition manifested by increased leaching of NO3

– to surface waters (21). Thus far, there are no signs of widespread regional increas-

es in NO3– concentrations in sensitive freshwaters

in Europe, but nitrogen continues to accumulate in catchment soils and vegetation (22). The processes leading to nitrogen saturation thus appear to require many decades of chronic deposition. However, in the Tatras and the Italian Alps, two of the European re-gions studied here, NO3

– concentrations are already significant (23, 24). For these two regions, we in-creased the degree of nitrogen saturation in the fu-ture scenario. For all the other regions, we assumed a constant percent of retention.

Second, future global climate change may in-troduce more uncertainty to these recovery pre-dictions: Changes could either accelerate or hinder recovery of acidified freshwaters. Predicted chang-es include increased temperature and precipitation, both of which might increase NO3

– leaching to sur-face waters. Large-scale experiments have shown that increased temperature can accelerate mineral-ization rates of soil organic matter, with subsequent release of NO3

– to runoff (25). Global change might also alter precipitation and runoff amounts, growth and uptake of nutrients by vegetation, and frequen-cy and severity of storm inputs of sea salts and at-mospheric dust.

In addition, other environmental factors, such as changes in forestry practices and in emissions and deposition of base cations, can affect soil and water acidification. The harvesting of forests may export as many base cations as leaching of base cations due to sulfur deposition. Our studies focused on the chemi-cal recovery of surface waters, because this is a pre-requisite for biological recovery. Just as the response of soils and waters lags behind the changes in acid deposition, so will the biological response lag be-

TA B L E 1

Modeling acid-sensitive surface waters in EuropeRegional mean nonmarine sulfur deposition in 1980 is given, and percent reductions in 2000 and 2016 (relative to 1980) used in the model simulations are also shown (12 ). Vegetation types: F, forest; H, heathland; M, moorland; A, alpine.

Region

No. of

sites Vegetation

Sulfate deposition

Ref.1980

(µeq/m2 yr)2000

(% reduction)2016

(% reduction)

Finland 36 F 66 73 77 48Northern Sweden 30 F, H 52 55 66 49Southern Sweden 35 F, H 78 55 73 49Central Norway 19 F, H 54 66 77 50Southern Norway 60 F, H 105 64 77 51Northern Scotland 30 H 52 42 65 52Southwestern

Scotland 54 F, H 228 68 82 53Wales 95 F, M 73 70 85 53Southern Pennines,

England 59 M 328 68 86 53Tatra Mountains,

Slovakia 31 A 146 57 65 54Southern Alps, Italy 13 F, A 95 34 70 55Pyrenees, Spain 79 A 66 49 65 56

68A ■ ENVIRONMENTAL SCIENCE & TECHNOLOGY / FEBRUARY 1, 2005

hind improvements in water quality. Organisms re-quire time to migrate back into restored habitats, and populations must adjust to new arrivals. Therefore, several generations may be necessary before the pop-ulation stabilizes. Characteristic lag times will prob-ably differ widely between groups; for example, algae can be expected to react within a few years, where-as fish populations may require more than a decade (26). False starts may also occur, because episodes of acidic water may wipe out newly re-established pop-

ulations. Thus far in Europe, reports of biological recovery are scattered (27).

Some success in EuropeFor our analysis, we intentionally chose the most sensitive waters rather than a statistically repre-sentative sample of all European freshwaters. The spirit of the work follows CLRTAP: Efforts are being made to reduce the total ecosystem area in Europe in which the critical load of acidity is exceeded.

1860 1980

2000 2016

F I G U R E 3

Simulated ANC concentrations in surface waters in 12 regionsThree acid neutralizing capacity (ANC; µeq/L) classes correspond to the probability of viable populations of brown trout and other key indicator organisms. Red: ANC < 0, barren of fish; yellow: ANC 0–20, sparse popula-tion; blue: ANC > 20, good population. Data from the acidification model MAGIC (SMART in Finland). Four key years are shown: 1860, pre-acidification (no simulations for Finland, because the SMART model was initiated in 1960); 1980, maximum acidification; 2000, present; and 2016, complete implementation of emission reduction protocols.

FEBRUARY 1, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY ■ 69A

Given the emission ceilings of the Gothenburg Pro-tocol, ~8% of Europe will still exceed the critical load of acidity in 2010 (28). Our sensitive freshwa-ters generally lie within this area. Should further re-ductions in emissions be sought such that recovery of these acidified ecosystems extends over a greater area more quickly? The prediction of future recov-ery of these sensitive waters is key scientific infor-

mation for political decisions regarding revisions and potential new protocols to CLRTAP.

The model results indicate that even after com-plete implementation of the Gothenburg Protocol and other current legislation, acidification with com-mensurate adverse biological effects will contin-ue to be a significant problem in southern Norway, southern Sweden, the Tatras, the Italian Alps, and

1980

2016

2000

F I G U R E 4

Simulated base cation poolsIncremental changes [(yearn + 1 − yearn )/yearn ] of exchangeable base cation pools in catchment soils in each of 12 regions are estimated by application of the models. Red: depletion by >0.1%/yr; yellow: negligible change (<0.1%/yr); green: replenishment by >0.1%/yr.

70A ■ ENVIRONMENTAL SCIENCE & TECHNOLOGY / FEBRUARY 1, 2005

the Southern Pennines in the United Kingdom. More than 5% of the ecosystems in each of the regions evaluated here will not meet the ANC criterion to protect sensitive aquatic organisms. Additional miti-gation measures will be required in these regions to meet long-term European policy objectives. Further reduction of emissions of sulfur and nitrogen is one possible measure.

The Gothenburg Protocol will be up for revision one year after it enters into force. The models used to establish the protocol were steady-state, and the time for recovery was not considered. The dynamic models used here (MAGIC and SMART) add time as a variable. Work under way to advance the next revi-sion of CLRTAP is now taking time into account. The new concept is the target-load function: What are satisfactory levels of sulfur and nitrogen, and when must deposition be further decreased to obtain a satisfactory water quality (i.e., ANC limit) at a given time in the future? (29). In Europe, this work is orga-nized by the International Cooperative Programme (ICP) for Waters (30) and the ICP for Modelling and Mapping (31). The first European data set on target-load functions for waters and soils is currently be-ing collated under the auspices of the CLRTAP.

These results are relevant in a wider context for other EU policy. The EU Water Framework Directive (WFD; 32) calls for member states to develop plans for remedial measures to achieve “good ecological status” by 2016. Acidification is one of many pollution factors currently causing degradation of water qual-ity and thus nonachievement of the WFD. Further, international conventions need the results to protect the sea, because a portion of the nitrogen deposited on land and lost to surface waters as NO3

– ultimately reaches the sea. Nitrogen losses predicted by acidifi-cation models can be used to estimate future loading of the nutrient nitrogen to coastal marine ecosys-tems. Relevant EU policy includes management of coastal zones (33), the convention to protect the North Sea (Oslo and Paris Commissions; 34), and the Baltic Sea (Helsinki Commission; 35).

Progress in North America?This saga of the rise and fall of acid deposition in Europe has been running in parallel to similar events in eastern North America. Surface-water acidifica-tion became widespread in the 1970s in large regions of southeastern and eastern Canada and the north-eastern United States, because of acid deposition on acid-sensitive terrain. Since the mid-1980s, sulfur deposition has decreased by ~40% as a result of legis-lation implemented in Canada (Eastern Canada Acid Rain Programme; 36) and the United States (Clean Air Act Amendments; 37) and the 1991 Canada–U.S. Air Quality Agreement (38). Surface-water chemistry has begun to recover (11, 39–42). Additional reduc-tions in sulfur deposition may occur in the future. If implemented, the Canada-Wide Acid Rain Strategy for Post-2000 (43) and the proposed U.S. Clear Skies Act (44) would reduce sulfur emissions in 2010 by 50% from 2000 levels.

Predictions for future recovery of surface waters—given this scenario—have been made using MAGIC

and other models for southeastern Ontario (45), the Atlantic Provinces (46), and the Adirondack region of New York (47). These predictions point to contin-ued recovery in response to the reductions in sulfur deposition, but a significant percentage of sites will still be acidic in the future. In many regions of east-ern North America, as in Europe, further measures appear to be necessary to help acidified surface wa-ters fully recover.

Richard F. Wright is a senior research scientist and Thor-jørn Larssen is a research scientist at the Norwegian In-stitute for Water Research. Lluis Camarero is a research scientist at the Centre of Advanced Studies (Spain). Ber-nard J. Cosby is a research professor at the University of Virginia. Robert C. Ferrier is head of catchment man-agement and Rachel Helliwell is a research scientist at the Macaulay Institute (U.K.). Martin Forsius is head of the Research Programme for Global Change at the Finnish Environment Institute. Alan Jenkins is the sci-ence director for the Water Programme at the Centre for Ecology and Hydrology (U.K.). Jiri Kopácek is a research scientist at the Hydrobiological Institute (Czech Repub-lic). Vladimir Majer is a research scientist at the Czech Geological Survey. Filip Moldan is a research scientist at the IVL Swedish Environmental Research Institute. Maximilian Posch is a senior researcher at the Coordi-nation Center for Effects at the Dutch National Insti-tute for Public Health and the Environment (RIVM). Michela Rogora is a research scientist at the Italian Na-tional Research Council’s Institute of Ecosystem Study. Wolfgang Schöpp is a research scientist at the Interna-tional Institute for Applied Systems Analysis (Austria). Address correspondence regarding this article to Wright at richard.wright@niva.no.

AcknowledgmentsThis work was funded in part by the European Commission under the EMERGE (EVK1-CT-1999-00032) and RECOV-ER:2010 (EVK1-CT-1999-00018) projects; the Natural Envi-ronment Research Council (U.K.); the U.K. Department for Environment, Food and Rural Affairs (Contract No. EPG 1/3/194); the Scottish Executive Environment and Rural Affairs Department; the Academy of Finland; the Span-ish Interministerial Committee on Science and Technolo-gy (REN2000-0889/GLO); and the Norwegian Institute for Water Research.

References (1) Overrein, L.; Seip, H. M.; Tollan, A. Acid precipitation—

Effects on forest and fish; Final report of the SNSF-proj-ect 1972–1980; Report No. FR 19-80; Norwegian Forest Research Institute: Oslo-Ås, Norway, 1980.

(2) Wright, R. F.; Henriksen, A. Chemistry of small Norwe-gian lakes with special reference to acid precipitation. Limnol. Oceanogr. 1978, 23, 487–498.

(3) Skjelkvåle, B. L.; Wright, R. F. Overview of areas sensitive to acidification in Europe; Acid Rain Research Report No. 20/1990; Norwegian Institute for Water Research: Oslo, Norway, 1990.

(4) Reuss, J. O.; Johnson, D. W. Acid Deposition and the Acid-ification of Soils and Waters; Springer-Verlag: New York, 1986; pp 1–119.

(5) Lien, L.; et al. A critical limit for acid neutralizing capa-city in Norwegian surface waters, based on new analy-ses of fish and invertebrate responses. Sci. Tot. Environ. 1996, 177, 173–193.

(6) Raddum, G. G.; Skjelkvåle, B. L. Critical limits of acidi-fication to invertebrates in different regions of Europe. Water, Air, Soil Pollut. 1995, 85, 475–480.

FEBRUARY 1, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY ■ 71A

and Mapping of Critical Loads and Levels and their Air Pollution Effects, Risks and Trends (ICP Modelling and Mapping), www.unece.org/env/wge/mapping.htm.

(32) The EU Water Framework Directive: Integrated river basin management for Europe, http://europa.eu.int/comm/environment/water/water-framework/index_en.html.

(33) Global Web Service on Oceans, Coasts, and Islands, www.globaloceans.org.

(34) OSPAR Commission, www.ospar.org/eng/html/background.htm.

(35) Baltic Marine Environment Protection Commission, www.helcom.fi.

(36) Eastern Canada Acid Rain Programme, www.ec.gc.ca/acidrain.

(37) U.S. Clean Air Act Amendments, www.epa.gov/air/oaq_caa.html.

(38) 1991 Canada–U.S. Air Quality Agreement, www.ijc.org/rel/agree/air.html.

(39) Jeffries, D. S.; et al. Assessing the Recovery of Lakes in Southeastern Canada from the Effects of Acid Deposi-tion. Ambio 2003, 32, 176–182.

(40) Jeffries, D. S.; et al. Assessment of changes in lake pH in southeastern Canada arising from present levels and expected reductions in acidic deposition. Can. J. Fish. Aquat. Sci. 2000, 57 (Suppl. 2), 40–49.

(41) Bouchard, A. Recent lake acidification and recovery trends in southern Quebec, Canada. Water, Air, Soil Pol-lut. 1997, 94, 225–245.

(42) Driscoll, C. T.; et al. Chemical response of lakes in the Adirondack region of New York to declines in acidic de-position. Environ. Sci. Technol. 2003, 37, 2036–2042.

(43) The Canada-Wide Acid Rain Strategy for Post-2000, www.ec.gc.ca/acidrain/strat/strat_e.htm.

(44) U.S. EPA Clear Skies Act, www.epa.gov/clearskies. (45) Aherne, J.; Dillon, P. J.; Cosby, B. J. Acidification and re-

covery of aquatic ecosystems in south central Ontario, Canada: Regional application of the MAGIC model. Hy-drol. Earth Syst. Sci. 2003, 7, 561–573.

(46) Clair, T. A.; Dennis, I. F.; Cosby, B. J. Probable changes in lake chemistry in Canada’s Atlantic Provinces under proposed North American emission reductions. Hy-drol. Earth Syst. Sci. 2003, 7, 574–582.

(47) Church, M. R.; van Sickle, J. Potential relative future ef-fects of sulfur and nitrogen deposition on lake chemis-try in the Adirondack Mountains, United States. Water Resour. Res. 1999, 35, 2199–2211.

(48) Posch, M.; et al. Modelling the recovery of acid-sensi-tive Finnish headwater lakes under present emission reduction agreements. Hydrol. Earth Syst. Sci. 2003, 7, 484–493.

(49) Moldan, F.; et al. Modelling acidification and recovery of Swedish Lakes. Water, Air, Soil Pollut. Focus 2004, 4, 139–160.

(50) Wright, R. F.; Cosby, B. J. Recovery of acidified moun-tain lakes in Norway as predicted by application of the MAGIC model. J. Limnol. 2004, 63, 101–110.

(51) Wright, R. F.; Cosby, B. J. Future recovery of acidified lakes in southern Norway predicted by the MAGIC model. Hydrol. Earth Syst. Sci. 2003, 7, 467–483.

(52) Helliwell, R. C.; Kernan, M. R. Modelling hydrochemical and ecological trends in acid sensitive surface waters in the Scottish highlands. J. Limnol. 2004, 63, 111–122.

(53) Helliwell, R. C.; et al. Modelling the recovery of surface water chemistry and the ecological implications in the British uplands. Hydrol. Earth Syst. Sci. 2003, 7, 456–466.

(54) Kopácek, J.; et al. Modelling reversibility of Central Euro-pean mountain lakes from acidification: Part II—The Tatra Mountains. Hydrol. Earth Syst. Sci. 2003, 7, 510–525.

(55) Rogora, M.; Marchetto, A.; Mosello, R. Modelling the ef-fects of atmospheric sulphur and nitrogen deposition on selected lakes and streams of the Central Alps (Ita-ly). Hydrol. Earth Syst. Sci. 2003, 7, 540–551.

(56) Camarero, L.; et al. Application of MAGIC to Lake Redó (Central Pyrenees): Assessment of the role of climate-driven changes in atmospheric precipitation, base cat-ion deposition, and weathering rates on lake water chemistry. J. Limnol. 2004, 63, 123–132.

(7) Juggins, S.; Ormerod, S. J.; Harriman, R. Relating criti-cal loads to aquatic biota. In Critical Loads of Acid Depo-sition for UK Freshwaters; Battarbee, R. W., Ed.; Institute of Terrestrial Ecology: Edinburgh, U.K., 1995.

(8) Bull, K. R.; et al. Coordinated effects monitoring and modelling for developing and supporting internation-al air pollution control agreements. Water, Air, Soil Pol-lut. 2001, 130, 119–130.

(9) Convention on Long-Range Transboundary Air Pollu-tion, www.unece.org/env/lrtap.

(10) Evans, C. D.; et al. Recovery from acidification in Eu-ropean surface waters. Hydrol. Earth Syst. Sci. 2001, 5, 283–298.

(11) Stoddard, J. L.; et al. Regional trends in aquatic recov-ery from acidification in North America and Europe 1980–1995. Nature 1999, 401, 575–578.

(12) Schöpp, W.; et al. Long-term development of acid de-position (1880–2030) in sensitive freshwater regions in Europe. Hydrol. Earth Syst. Sci. 2003, 7, 436–446.

(13) Cosby, B. J.; et al. Modelling the effects of acid deposi-tion: Refinements, adjustments and inclusion of nitro-gen dynamics in the MAGIC model. Hydrol. Earth Syst. Sci. 2001, 5, 499–518.

(14) de Vries, W.; Posch, M.; Kämäri, J. Simulation of the long-term soil response to acid deposition in various buffer ranges. Water, Air, Soil Pollut. 1989, 48, 349–390.

(15) Posch, M., Hettelingh, J.-P., Slootweg, J., Eds. Manual for dynamic modelling of soil response to atmospheric deposition; Report No. 259101012/2003; WGE ICP-M&M Coordination Center for Effects, RIVM: Bilthoven, The Netherlands, 2003.

(16) Beier, C.; Moldan, F.; Wright, R. F. Terrestrial ecosystem recovery—Modelling the effects of reduced acidic in-puts and increased inputs of sea salts induced by glob-al change. Ambio 2003, 32, 275–282.

(17) Environmental Change Research Centre, www.geog.ucl.ac.uk/ecrc/themes/mountain.stm.

(18) Predicting Recovery in Acidified Freshwaters by the Year 2010, and Beyond, www.macaulay.ac.uk/recover/index.html.

(19) The RAINS model, www.iiasa.ac.at/~rains/doc/iiasa.htm.

(20) Reuss, J. O.; Cosby, B. J.; Wright, R. F. Chemical process-es governing soil and water acidification. Nature 1987, 329, 27–32.

(21) Aber, J. D.; et al. Nitrogen saturation in northern forest ecosystems. Bioscience 1989, 39, 378–386.

(22) Wright, R. F.; et al. Trends in nitrogen deposition and leaching in acid-sensitive streams in Europe. Hydrol. Earth Syst. Sci. 2001, 5, 299–310.

(23) Kopácek, J.; Veselý, J.; Stuchlík, E. Estimation of sul-phur and nitrogen fluxes in the Bohemian Forest and Tatra Mountains during the industrial revolution (1850 to 2000). Hydrol. Earth Syst. Sci. 2001, 5, 391–406.

(24) Rogora, M.; Marchetto, A.; Mosello, R. Trends in the chemistry of atmospheric deposition and surface wa-ters in Lake Maggiore watershed. Hydrol. Earth Syst. Sci. 2001, 5, 379–390.

(25) van Breemen, N.; et al. Impacts of elevated carbon di-oxide and temperature on a boreal forest ecosystem (CLIMEX project). Ecosystems 1998, 1, 345–351.

(26) Yan, N. D.; et al. Developing conceptual frameworks for the recovery of aquatic biota from acidification. Ambio 2003, 32, 165–169.

(27) Wright, R. F.; Lie, M. C. Workshop on Models for Biolog-ical Recovery from Acidification in a Changing Climate; Document No. 55/02; Norwegian Institute for Water Research: Oslo, Norway, 2002.

(28) Hettelingh, J.-P.; Slootweg, J.; Posch, M. Critical Loads and Dynamic Modelling Results; Report No. 259101014/ 2004; RIVM: Bilthoven, The Netherlands, 2004.

(29) Jenkins, A.; et al. Assessing emission reduction targets with dynamic models: Deriving target load functions for use in integrated assessment. Hydrol. Earth Syst. Sci. 2003, 7, 609–617.

(30) International Cooperative Programme on Assessment and Monitoring of Acidification of Rivers and Lakes (ICP Waters), www.unece.org/env/wge/waters.htm.

(31) International Cooperative Programme on Modelling

72A ■ ENVIRONMENTAL SCIENCE & TECHNOLOGY / FEBRUARY 1, 2005