setting limits: using air pollution thresholds to protect ... · setting limits: using air...

esa

Published by the Ecological Society of America

esa

Setting Limits: Using AirPollution Thresholds to Protect

and Restore U.S. EcosystemsMark E. Fenn, Kathleen F. Lambert, Tamara F. Blett, Douglas A. Burns,

Linda H. Pardo, Gary M. Lovett, Richard A. Haeuber, David C. Evers,

Charles T. Driscoll, and Dean S. Jeffries

Fall 2011 Report Number 14

Setting Limits: Using AirPollution Thresholds to Protect

and Restore U.S. Ecosystems

Issues in EcologyIssues in Ecology

© The Ecological Society of America • [email protected] esa 1

ISSUES IN ECOLOGY NUMBER FOURTEEN FALL 2011

Setting Limits: Using Air Pollution Thresholdsto Protect and Restore U.S. Ecosystems

SUMMARY

More than four decades of research provide unequivocal evidence that sulfur, nitrogen, and mercury pollution havealtered, and will continue to alter, our nation’s lands and waters. The emission and deposition of air pollutants harmnative plants and animals, degrade water quality, affect forest productivity, and are damaging to human health. Many air qual-ity policies limit emissions at the source but these control measures do not always consider ecosystem impacts. Air pollutionthresholds at which ecological effects are observed, such as critical loads, are effective tools for assessing the impacts of air pol-lution on essential ecosystem services and for informing public policy. U.S. ecosystems can be more effectively protected andrestored by using a combination of emissions-based approaches and science-based thresholds of ecosystem damage.

Based on the results of a comprehensive review of air pollution thresholds, we conclude:

l Ecosystem services such as air and water purification, decomposition and detoxification of waste materials, climate regu-lation, regeneration of soil fertility, production and biodiversity maintenance, as well as crop, timber and fish suppliesare impacted by deposition of nitrogen, sulfur, mercury and other pollutants. The consequences of these changes maybe difficult or impossible to reverse as impacts cascade throughout affected ecosystems.

l The effects of too much nitrogen are common across the U.S. and include altered plant and lichen communities,enhanced growth of invasive species, eutrophication and acidification of lands and waters, and habitat deteriora-tion for native species, including endangered species.

l Lake, stream and soil acidification is widespread across the eastern United States. Up to 65% of lakes within sensi-tive areas receive acid deposition that exceeds critical loads.

l Mercury contamination adversely affects fish in many inland and coastal waters. Fish consumption advisories formercury exist in all 50 states and on many tribal lands. High concentrations of mercury in wildlife are also wide-spread and have multiple adverse effects.

l Air quality programs, such as those stemming from the 1990 Clean Air Act Amendments, have helped decrease airpollution even as population and energy demand have increased. Yet, they do not adequately protect ecosystemsfrom long-term damage. Moreover they do not address ammonia emissions.

l A stronger ecosystem basis for air pollutant policies could be established through adoption of science-based thresh-olds. Existing monitoring programs track vital information needed to measure the response to policies, and couldbe expanded to include appropriate chemical and biological indicators for terrestrial and aquatic ecosystems andestablishment of a national ecosystem monitoring network for mercury.

The development and use of air pollution thresholds for ecosystem protection and management is increasing in the United States,yet threshold approaches remain underutilized. Ecological thresholds for air pollution, such as critical loads for nitrogen and sulfurdeposition, are not currently included in the formal regulatory process for emissions controls in the United States, although they arenow considered in local management decisions by the National Park Service and U.S. Forest Service. Ecological thresholds offer ascientifically sound approach to protecting and restoring U.S. ecosystems and an important tool for natural resource managementand policy.

Cover photo credit: Loch Vale in the Colorado Rocky Mountains. Photo by SteveB in Denver (http://www.flickr.com/people/darkdenver/) and used in thispublication under a Creative Commons Attribution license.

© The Ecological Society of America • [email protected] esa

Introduction

Natural ecosystems have been altered in vari-ous ways by nitrogen, sulfur, and mercurydeposited in rain, snow, or as gases and parti-cles in the atmosphere. Through decades ofscientific research, scientists have documentedhow local, regional, and global sources of airpollution can produce profound changes inecosystems. These changes include acidifica-tion of soils and surface waters, harmful algalblooms and low oxygen conditions in estuar-ies, reduced diversity of native plants, highlevels of mercury in fish and other wildlife,and decreased tolerance to other stresses, suchas pests, disease, and climate change.Advancing our understanding of the linkagesamong pollutant deposition rates or concen-trations, ecosystem effects, and associated pol-icy decisions is a priority in policy-relevantscience in the U.S.

Air pollutants that affect human health andecosystems are primarily emitted from electricpower generation, industrial, transportation,and agricultural activities. The benefits andnecessities of these activities must be consid-ered in light of the often detrimental effects ofatmospheric emissions on human health, visi-bility, ecosystems, and on the services pro-vided to society by these ecosystems (Table 1).The 1990 Clean Air Act Amendments andother air quality regulations have led tomarked declines in emissions of nitrogen, sul-fur and mercury. Some emissions from powergeneration and other sources have decreasedby over 50% since the 1970s, even as popula-tion and energy demand have increased. Asthe emissions and deposition of most pollu-tants have declined, some impacted ecosys-tems have started to recover. In many parts ofthe country, however, ecological conditionsare still declining due to the increase in otherforms of pollution such as ammonia (NH3),the long term accumulation of sulfur and

nitrogen compounds in soils, and the ongoingbiomagnification of mercury in food webs.

The purpose of this report is to distilladvances in the science of air pollutionthresholds and to describe their use to assess,protect and manage the nation’s ecosystemsand the vital services they provide. We focushere on the environmental impacts of nitro-gen, sulfur, and mercury and refer to connec-tions to climate change. The discussion drawson the published research of hundreds of sci-entists over the past several decades with afocus on U.S. ecosystems and lessons fromCanada and Europe.

Air Pollution Thresholds

Thresholds of air pollution in the U.S. havebeen widely discussed in the scientific litera-ture since the 1970s, when research estab-lished that sulfur deposition was above levelsat which damage occurs in many sensitiveecosystems in the eastern U.S. More recently,nitrogen deposition has been shown to impactsensitive ecosystem components and processesthroughout the United States. Defining thespecific concentration or deposition input ofan air pollutant that will cause adverse or sig-nificant ecosystem effects has been the subjectof much scientific research. Pollutants canaccumulate with little noticeable impact onplants or animals until major changes occur asa tipping point is reached (Box 1). Thesechanges are measured by scientifically deter-mined chemical or biological indicators (Box2). Such environmental changes might elimi-nate a single sensitive species, or a broad shiftmay occur in biodiversity throughout anecosystem. Once a species or ecosystem haspassed a tipping point, a return to the previousstate may not be possible.

Air pollution thresholds can be definedbased strictly on scientific research (ecologicalthresholds) or based on a balance of policy con-

Setting Limits: Using Air Pollution Thresholdsto Protect and Restore U.S. Ecosystems

Mark E. Fenn, Kathleen F. Lambert, Tamara F. Blett, Douglas A. Burns, Linda H. Pardo, Gary M. Lovett,

Richard A. Haeuber, David C. Evers, Charles T. Driscoll, and Dean S. Jeffries




siderations spanning law, economics, ecologi-cal effects, human health and risk assessment(policy thresholds) (Box 1) (Figure 1). One toolincreasingly used to integrate the science andpolicy of air pollution thresholds for ecosystemprotection and management is critical loads(Box 4).

Advances in the Science ofAir Pollution Thresholds

Based on research over the past decade, astrong scientific foundation exists for definingair pollution thresholds using critical loadsapproaches (Box 4). In the following sectionswe synthesize the state of the science relatedto the ecological effects, key indicators, andcritical loads approaches for acidifying deposi-tion, nitrogen pollution and mercury contami-nation.

1. ACIDIFYING DEPOSITIONA. Effects of Acidifying Deposition

Acidifying deposition (or “acid rain”) iscaused by emissions to the atmosphere of sul-fur dioxide (SO2), nitrogen oxides (NOx), andother acidifying compounds such as ammonia(NH3)(see Box 3 for definition of chemicalnames and symbols). These pollutants returnto Earth in rain, snow, fog, mist and gases informs such as nitric and sulfuric acids andammonium (NH4

+) and can have long-termnegative impacts to terrestrial and aquaticecosystems. Ecosystems in the western U.S.have not been greatly affected by acidificationbecause acidifying deposition is relatively lowin much of the region and because in manyarid or semi-arid regions the soils are relativelyinsensitive to acid inputs. Some high eleva-tion streams in the Colorado Rockies and the

Box 1. DEFINITION OF TERMS

ACIDIFYING DEPOSITION. Deposition of substances from the atmosphere as rain, snow, fog, or dry particles that have the potentialto acidify the receptor medium, such as soil or surface waters. Emissions of sulfur and nitrogen oxides and ammonia are the most com-mon sources of acidifying air pollutants.

ACID NEUTRALIZING CAPACITY. A measure of the ability of a solution to neutralize inputs of strong acids, commonly applied to sur-face water or soil solution. The acronym ANC is widely used in referring to acid neutralizing capacity.

ATMOSPHERIC DEPOSITION. The transfer of air pollutants from the atmosphere to the Earth’s surface. Atmospheric depositionoccurs as wet (e.g., rainfall, fog, or snow) and dry deposition (e.g., gaseous or particulate deposition).

BASE SATURATION. The fraction of exchangeable cations in soil which are nonacid forming cations (Ca+2, Mg+2, K+ and Na+), alsoreferred to as ‘base cations’. The higher the amount of exchangeable base cations in soil, the more acidity can be neutralized.

BIOACCUMULATION. The increase in concentration of a contaminant in an individual organism relative to the surrounding environ-ment or medium (e.g., water, sediment).

BIOMAGNIFICATION. The increase in concentration of a contaminant from lower trophic levels to higher trophic levels in thefood chain.

CRITICAL LOAD. The quantitative estimate of an exposure to one or more pollutants below which significant harmful effects on spec-ified sensitive elements of the environment do not occur according to present knowledge.

ECOLOGICAL THRESHOLD. The dose of a pollutant at which a measurable change occurs in the response of some component of anecosystem (e.g., NO3

– leaching at nitrogen deposition of 8 kg/ha/yr).

ECOSYSTEM SERVICES. Benefits to society from a multitude of resources and processes that are supplied by natural ecosystems(e.g., clean drinking water).

ENDPOINT. The ultimate ecological, biological or human condition or process to be protected from harm. Two examples of endpointsare human health and forest sustainability.

INDICATOR. A measurable physical, chemical, or biological characteristic of a resource that may be adversely affected by a change inair quality (e.g., ANC).

NITROGEN SATURATION. Syndrome of effects occurring in an ecosystem caused by an overload of nitrogen, usually from long termatmospheric nitrogen deposition.

POLICY THRESHOLD. A quantitative value of desired ecological condition established by policy and selected based on a balancingof science and land management or policy goals.

SENSITIVE RECEPTOR. The indicator that is the most responsive to, or the most easily affected by a type of air pollution.

TARGET LOAD. The acceptable pollution load that is agreed upon by policy makers or land managers. The target load is set below thecritical load to provide a reasonable margin of safety, but could be higher than the critical load at least temporarily.

TIPPING POINT. The point at which an ecosystem shifts to a new state or condition in a rapid, often irreversible, transformation.



Table 1. Linking air pollution impacts to ecosystem services, indicators and thresholds. Ecological thresholdsgiven are typical values that can vary depending on ecological and environmental conditions.

Impact Ecosystem Ecological Response Ecosystem Services Indicator Ecological ThresholdImpacted

Sulfur and Nitrogen Deposition

Acidification Terrestrial 1. Decreased forest 1. Timber production Ca: Al+3 ratios in soil 10 – low risk

2. Increased 3. Biodiversitysusceptibility to 4. Resilience to Soil percent base 30% - low risk

Foliar chemistry



Sierra Nevada Mountains do experience acidicepisodes when pollutants retained in the snowpack over the winter are released into soils andstreams during snowmelt. In the eastern UnitedStates, depletion of available calcium and mag-nesium pools and acidification of forest soils iswidespread and well documented in theAppalachian Mountains, including the Cats-kills and the Adirondacks, and in the Shenan-doah Mountain region of West Virginia.

Mountain forests of the northeastern andsoutheastern United States receive high ratesof acidifying deposition due to frequent expo-sure to acidic clouds, fog, rain and snow.Changes associated with acidifying depositionhave reduced the ability of some tree speciesto cope with the cold temperatures commonto these mountain environments. This effectcontributed to large-scale red spruce deaths inthese regions in the 1980s and 90s, andremains a problem today. In eastern U.S. hard-wood forests at lower elevations, many sugarmaple, white ash, flowering dogwood, andother trees have high calcium requirementsand therefore are also sensitive to acidifica-tion. Tree declines have negative conse-quences for forest productivity and ecosystemservices, including timber production and cli-mate regulation (lower productivity means lessremoval of carbon dioxide from the atmos-phere). Research has attributed sugar mapledeclines in western Pennsylvania to acidifica-tion acting in concert with insect outbreaks,and research in New Hampshire has shownimproved growth and reproduction of sugarmaple, and less frost damage to red spruce,when calcium was added to an acidified forestfor experimental purposes.

Acidification of sensitive surface waters hasresulted in well documented adverse effects onfish, zooplankton, aquatic insects, microorgan-isms, and other aquatic biota. In many sensi-tive areas receiving elevated acidifying deposi-tion, surface waters are too acidic to supportany fish species. The reduction in the numberof aquatic species and in the number of fishsupported diminishes biodiversity and recre-ational fishing opportunities. Long-termresearch on acidification impacts on forests,lakes and streams has produced a wealth ofdata, from which are drawn the most com-monly applied indicators for assessing acidifi-cation status and effects (Table 1). Althoughterrestrial and aquatic indicators are treatedseparately below, recognition should be givento the connection of soil acidification toaquatic acidification.

B. Indicators - Acidifying Deposition

Indicators of Soil Acidificationand Forest Health

One way to assess the risk to acid sensitivetree species such as red spruce and sugar mapleis by tracking chemical indicators in the soiland in the leaves and needles of plants (i.e.,

Box 2. INDICATORS AND AIR POLLUTION THRESHOLDS

Just as physicians use a range of diagnostic measurements to monitor humanhealth, scientists track chemical and biological indicators to monitor ecosystemhealth. When many different studies confirm an association between a pollutantamount and an ecosystem response, threshold pollutant levels can often beidentified for indicators that signal likely problems. Chemical indicators are oftenused as surrogates for biological effects because chemical indicators are typi-cally simpler and less expensive to measure. Chemical indicators are imperfectsurrogates since accurate prediction of just how plants and animals will respondto chemical changes in their environment is not always possible.

Box 3. CHEMICAL NAMES AND SYMBOLS, AND UNITSOF MEASURE

Chemical Names and Symbols:Sulfur dioxide, SO2Nitrogen dioxide, NO2Nitrogen oxides, NOxSulfur oxides, SOxAmmonia, NH3Ammonium, NH4

+

Mercury, HgMethylmercury, MeHgSulfate, SO4

-2

Nitrate, NO3-

Dissolved organic carbon, DOCPhosphorus, PNitrogen, NCarbon, CAluminum, Al+3

Calcium, Ca+2

Magnesium, Mg+2

Sodium, Na+

Potassium, K+

Calcium to aluminum ratio, Ca:Al pH, a measure of acidity or hydrogenion concentrationNutrient ratios (e.g., N:P, N:Ca, C:N)

Units of Measure:Equivalents per hectare per year,eq/ha/yrKilograms per hectare per year,kg/ha/yrParts per million, ppmMicroequivalents per liter, µeq/LMilliequivalents per square meter peryear meq/m2/yrMicrograms per liter, µg/L

Figure 1. Conceptualrepresentation of how ecologicaland policy thresholds may bedeveloped. Both lines showestimates of ecosystemdegradation as pollutantsincrease in ecosystems. Line“A” represents a gradual declinein ecosystem condition, wheremanagers, policy makers, andregulators can set policythresholds at any number ofdifferent points depending ongoals (for example, A1, atbeginning of decline or A2, atmidpoint of decline). Line “B”represents a rapid decline inecosystem condition, with aclearly identified, ecologicalthreshold at which a tippingpoint occurs (B1).

A

B

A1 A2 B1Deposition of Air PollutionS

upp

ly o

f E

cosy

stem

Ser

vice

s



foliage) (Table 1). Three elements naturallypresent in soils, calcium (Ca+2), magnesium(Mg+2), and aluminum (Al+3), influence theextent to which trees and other plants may beadversely affected by acidifying deposition.Calcium and magnesium are nutrients neededfor a variety of plant functions and their sup-ply helps neutralize acid inputs to soils,whereas Al+3 can be harmful to plants at highconcentrations when present in the readilyavailable exchangeable form. Acid depositionslowly removes readily available exchangeableCa+2 and Mg+2 from soils and replaces themwith exchangeable Al+3 and hydrogen ion (oracidity), setting off a cascade of adversechanges.

In general, greater availability of Ca+2 andMg+2 and low Al+3 provides favorable conditionsfor many acid-sensitive tree species such as sugarmaple and red spruce. Calcium to aluminum ratio(Ca:Al) in soils and soil solutions is one indica-tor used to assess the health risk to acid sensitivetree species such as red spruce and sugar maple.Soil percent base saturation is another useful indi-cator for assessing sensitivity and extent of acid-ification. Scientists generally concur that wheresoil percent base saturation is low there is a highrisk of damage to the vitality of sensitive treespecies due to nutritional deficits resulting fromacidification. The risks to forest vegetation asso-ciated with a range of Ca:Al ratios and soil per-cent base saturation values are shown in Table 1.Other studies have focused on the concentra-tion of exchangeable Ca+2 and Mg+2 as a usefulindicator since soils can have widely varyingamounts of these nutrients that are essential tothe health of forest vegetation. Concentrations of

Ca+2 and Mg+2 in the leaves and needles ofplants (foliage) have recently been identified asvaluable indicators for evaluating acid deposi-tion impacts. For example, low concentrationsof these nutrients have been identified as limit-ing the growth of sugar maple (Table 1).

Indicators of Acidification inAquatic Ecosystems

Indicators of acidification in lakes and streamsare generally based on changes in water chem-istry. Water chemistry strongly affects thenumbers and types of aquatic organisms thatare present in a water body. The indicatorsmost commonly used to track changes in sur-face water acidification are ANC, pH, and/orconcentrations of key elements.

Acid neutralizing capacity (ANC) is a com-monly used chemical indicator of lake orstream sensitivity to acidification. ANC, mea-sured in microequivalents per liter (µeq/L; SeeBox 3 for a list of chemical units of measure),characterizes the ability of water to neutralizestrong acids including those introduced byatmospheric deposition. ANC is a good gen-eral indicator of acidity-related water qualitybecause values are typically strongly correlatedwith pH, Al+3 concentrations, and Ca+2 con-centrations. Specific concern levels have beenidentified and are used to estimate criticalloads (Table 2). The diversity of fish speciesdeclines precipitously with decreases in ANCin Adirondack Lakes (Figure 2). In Shenan-doah National Park (Virginia) streamsresearchers found that one fish species, onaverage, is lost for every 21 µeq/L decline inANC. Recent studies have demonstrated thatanother useful chemical indicator is base cationsurplus. Low values indicate that the soil hasbecome sufficiently acidified to enable toxicforms of aluminum to be transported from thesoil into streams at concentrations of concern.

The pH value of a water body is a funda-mental measure of acidity or the hydrogen ionconcentration. A pH of 7 is neutral, and pHvalues below 7 are increasingly acidic whilevalues above 7 are increasingly basic or alka-line. Like ANC, decreases in pH are associ-ated with decreases in the richness of aquaticspecies (Table 1). Studies have shown that inlakes of the Adirondack Mountains of NewYork and the White Mountains of NewHampshire, one fish species is lost for everypH decline of 0.8 units as values decrease from6 to 4. Few fish species can survive at pH val-ues of 4 or less (Figure 3).

Figure 2. Number of fish speciesper lake as a function of acidneutralizing capacity (ANC) inAdirondack lakes. The data arepresented as the mean of speciesrichness for every 10 µeq/L ANCclass. Lakes are also classifiedinto five descriptive categoriesranging from low to acuteimpacts. (Adapted from: Sullivan,T.J. and others 2006. Assessmentof the Extent to WhichIntensively-Studied Lakes areRepresentative of the AdirondackMountain Region. Final Report06-17. New York State EnergyResearch and DevelopmentAuthority. Albany, NY).



Decreases in pH and ANC are often paral-leled by changes in element concentrationsincluding increases in Al+3 concentrations anddecreases in Ca+2. High dissolved Al+3 concen-trations can have toxic effects on many types ofaquatic biota, and at extreme levels few aquaticspecies can survive (Table 1). Organic forms ofAl+3 are much less toxic than inorganic forms.Emerging research suggests that Ca+2 concentra-tions in streamwater are also an important bio-logical indicator. Acidifying deposition hasaccelerated the leaching of Ca+2 from soils tosurface waters gradually decreasing the avail-able pool of Ca+2 in soils and lowering Ca+2

concentrations in runoff. This soil depletiontogether with decreases in leaching associatedwith declines in acidifying deposition is con-tributing to decreases in surface water Ca+2.Many lakes in the boreal forest of the CanadianShield now have Ca+2 concentrations that areconsidered sub-optimal for water fleas, crayfishand other crustaceans and may be limiting thespecies richness of lakes in this region.

C. Critical Loads – AcidifyingDeposition

Critical loads represent the deposition ratethat can occur without surpassing tippingpoints for a given species or ecosystem basedon established indicators and effect levels.The critical load for a specific pollutant orgroup of pollutants will vary depending on dif-ferences in landscape sensitivity and in theendpoints for which the critical loads are cal-culated (e.g., forest soils, lake chemistry).

Advances in understanding of chemical andbiological indicators of acidification have sup-ported the development of critical loads forsulfur and nitrogen in parts of the U.S. andCanada.

Table 2. Expected ecological effects and concern levels in freshwater ecosystems at various levels of acid neu-tralizing capacity (ANC). (Source: USEPA)a.

Category Label ANC level (µeq/L) Expected Ecological Effects

Low Concern >100 Fish species richness may be unaffected. Reproducing brook trout populations are (No Effect) expected where habitat is suitable. Zooplankton communities are unaffected and

exhibit expected diversity and distribution.

Moderate 50-100 Fish species richness begins to decline (sensitive species are lost from lakes). BrookConcern trout populations are sensitive and variable, with possible sub-lethal effects. Diversity(Minimally and distribution of zooplankton communities begin to decline as species that are sensi-Impacted) tive to acid deposition are affected.

Elevated 0–50 Fish species richness is greatly reduced (more than half of expected species are Concern missing). On average, brook trout populations experience sub-lethal effects, including (Episodically loss of health and reproduction (fitness). During episodes of high acid deposition, brook Acidic) trout populations may die. Diversity and distribution of zooplankton communities

declines.

Acute Concern



Forests

U.S. researchers use models to develop criticalloads for forest soil acidification (Box 4). Arecent study estimated the critical acid loadsfor forest soils across the conterminous U.S.The critical acid loads for S and N throughoutthe Appalachian Mountain Range and Floridaare estimated to be less than 1,000 eq/ha/yr(critical loads for combined sulfur and nitro-gen are expressed in terms of ionic charge bal-ance as equivalents per hectare per year). Thisstudy estimated that about 15% of U.S. forestsoils exceed their critical acid load by at least25% including much of New England, WestVirginia, and parts of North Carolina. Bycomparison, critical load modeling in Canadaestimated that 30 to 40% of upland forestareas in Canada are in exceedance of the criti-cal load for acidification, while more than50% are in exceedance in eastern Canada(Ontario, Quebec, New Brunswick, NovaScotia and Newfoundland).

Surface Waters

Regional critical loads for surface waters havebeen developed for acidifying deposition ofsulfur and nitrogen in sensitive regions of theAdirondack Mountains of New York and inthe central Appalachians of Virginia and WestVirginia. The median critical load for a targetANC of 50 µeq/L is 129 milliequivalents persquare meter per year (meq/m2/yr) in theAdirondacks and 45 meq/m2/yr in the centralAppalachians with values ranging from less

than 0 to over 1,000 meq/m2/yr in relativelyinsensitive ecosystems. The number of aquaticecosystems exceeding the critical loads is stillquite high, but has declined with decreases inacid deposition from the early 1990s to thelate 2000s (Figure 4). Currently, 44% ofAdirondack lakes evaluated exceed the criti-cal load and in these lakes recreationally valu-able fish species such as trout are missing dueto acidification. In the Shenandoah area, 85%of streams evaluated exceed the critical loadresulting in losses in fitness in fish species suchas the blacknose dace. The persistence of criti-cal load exceedances despite significantdecreases in SO2 emissions is related to con-tinued high inputs of acidifying NOx, low ini-tial ANC conditions, and soil depletion ofnutrient cations (Ca+2 and Mg+2) that haveleft many watersheds more sensitive to aciddeposition over time.

A similar study of 2053 lakes in six north-eastern states and four eastern Canadianprovinces estimated critical loads for acidify-ing deposition of sulfur and nitrogen for a tar-get ANC of 40 µeq/L. Results show that 28%of the lakes studied have a critical load in thecategories of ≤20 and 20–40 meq/m2/yr, sug-gesting vulnerability to acidification with rela-tively moderate atmospheric deposition. It isestimated that the critical load is exceeded in12% of the study lakes, based on depositionlevels in 2002. These studies point to theimportance of long-term monitoring andresearch for assessing the impact of emissionscontrol programs on deposition and ecologicalrecovery (Box 5).

Box 4. UNDERSTANDING THE CRITICAL LOADS APPROACH

Critical loads, and other approaches that use models or empirical observations to link deposition with effects, provide tools that enableresource managers and policymakers to evaluate tradeoffs between the costs of more stringent emissions controls and the benefits ofecosystem services provided by healthy ecosystems.

A critical loads approach can be used to synthesize scientific knowledge about air pollution thresholds that cause adverse impactsor ecosystem change. Describing air pollutant effects on ecosystems in critical load terms quantifies estimates of “cause and effect” ina way that allows researchers to communicate science to air quality regulators and natural resource managers. Critical loads are mostcommonly applied to evaluate the effects of nitrogen and sulfur pollutants and their associated acidity or the eutrophying effects ofnitrogen. When critical loads are exceeded there is increased risk for a range of problems including ecosystem acidification, excessnitrogen effects, declines in forest health, and changes in biodiversity.

Critical loads are typically expressed as deposition loading rates of one or more pollutants in amount per area per year (e.g., kilo-grams per hectare per year (kg/ha/yr)). Critical loads are based on changes to specific biological or chemical indicators such as speciescomposition of a given ecosystem (e.g., grassland) or biotic community (e.g., understory plants or tree-dwelling lichens) or acid neu-tralizing capacity (ANC) in soils, streams or lakes. Because different sensitive receptors (e.g., forest soils, high elevation lakes, speciesof lichen) or species may have varying sensitivities to air pollutant loads, multiple critical loads can be used to describe a continuum ofimpacts with increasing deposition at a given location (See Figure 5).

In addition, even for the same organism, multiple critical loads may be associated with biological thresholds for different negativeeffects, such as stunted growth, reduced reproduction, and increased mortality. Several different threshold levels may therefore beincluded in a critical load assessment. The policymaker, air regulator, or land manager can assess all the critical loads (science-drivenecological thresholds) and select target loads (policy thresholds) based on the level of ecosystem protection desired, economic con-siderations, and stakeholder input at a given location.



2. NITROGENA. Effects of Excess Nitrogen onEcosystems

The nitrogen gas that makes up most of theEarth's atmosphere is inert, with little impacton ecosystems. Nitrogen converted to its reac-tive forms such as NH3 and NOx, however,can cause profound biological changes.Activities such as fertilizer manufacturing,intensive livestock production and the burn-ing of fossil fuels convert nitrogen to thesereactive forms which can then enter andpotentially over-fertilize ecosystems. This canlead to problems such as algal overgrowth inlakes, reduced water quality, declines in foresthealth, and decreases in aquatic and terrestrialbiodiversity by favoring “nitrogen loving”species at the expense of other species withlow nitrogen preferences. For example, mostestuaries and bays in the Northeast U.S. andMid-Atlantic regions experience some degreeof eutrophication (where excess nutrients pro-mote a proliferation of plant life, which candeplete oxygen in the deeper waters), as aresult of nutrients from atmospheric deposi-tion and agricultural, urban and industrialrunoff. Excess nitrogen can also changespecies composition. In Waquoit Bay,Massachusetts elevated nitrogen allows tallcord grass to thrive but not eelgrass, whichdecreases critical fish habitat.

Adding nitrogen to forests whose growth istypically limited by its availability may appeardesirable, possibly increasing forest growth and

timber production, but it can also haveadverse effects such as increased soil acidifica-tion, biodiversity impacts, predisposition toinsect infestations, and effects on beneficialroot fungi called mycorrhizae. As atmosphericnitrogen deposition onto forests and otherecosystems increases, the enhanced availabil-ity of nitrogen can lead to chemical and bio-logical changes collectively called “nitrogensaturation.” As nitrogen deposition from airpollution accumulates in an ecosystem, a pro-gression of effects can occur as levels of biolog-ically available nitrogen increase (Figure 5).Because of the multiple potential effects ofnitrogen deposition in terrestrial and aquaticecosystems, the ecosystem services affectedvary depending on the sensitive receptorsfound within a given ecosystem and the levelof atmospheric deposition. Prominent exam-ples of affected ecosystem services in forestsinclude timber production, climate regulation,recreational use, and biodiversity loss. In

Figure 4. Percentage of lakes inexceedance of the critical loadfor sensitive eastern US surfacewaters in the Adirondacks (169lakes in NY) and the centralAppalachians (92 streams in VAand WV). The percent exceedingthe critical load has declined asemissions and deposition havebeen decreasing (Source: JasonLynch- USEPA).

Box 5. THE ROLE OF LONG-TERM MONITORING AND RESEARCH

Long-term studies measure baseline ecosystem conditions and trends and can show how ecosystems respond when atmosphericdeposition decreases below a threshold that was previously exceeded. The trajectory of recovery is not always consistent with modelsimulations, illustrating the importance of long-term monitoring and research to improve the capabilities of simulation models. A num-ber of regional- and national-scale air, water, soil, and biota monitoring networks collect high-quality data that are useful in assessingecosystem thresholds. However, current efforts are not enough to provide continuous data at sites across the country, and often lackthe coordination needed to effectively combine datasets for maximum benefit. We recommend that existing monitoring and researchprograms be continued, expanded and better integrated. Some examples of federal monitoring programs include:

• Federal agency air pollution monitoring programs such as the Interagency Monitoring of Protected Visual Environments(IMPROVE) http://vista.cira.colostate.edu/improve/ and the National Atmospheric Deposition Monitoring Program (NADP),http://nadp.sws.uiuc.edu/, and the Clean Air Status and Trends Network (CASTNET) http://www.epa.gov/castnet/

• The U.S. Forest Service's Forest Inventory Analysis and Forest Health Monitoring (FIA/FHM)

• The Environmental Protection Agency’s Temporally Integrated Monitoring of Ecosystems/Long-term Monitoring (TIME/LTM)network http://www.epa.gov/airmarkt/assessments/TIMELTM.html and National Surface Water Surveys

• The U.S. Geological Survey's National Water-quality Assessment Program (NAWQA) http://water.usgs.gov/nawqa/ andBiomonitoring of Environmental Status and Trends (BEST) programs

• The U.S.Forest Service’s wilderness area surface water-monitoring programs http://www.fs.fed.us/waterdata/

• The NSF-sponsored National Ecological Observation Network (NEON) http://www.neoninc.org/ and Long-Term EcologicalResearch (LTER) network - http://www.lternet.edu/

Appalachians

Adirondacks

% W

ater

Bo

die

s In

Exc

eed

ance

1989

-199

1

2006

-200

8

1989

-199

1

2006

-200

8

100

90

80

70

60

50

40

30

20

10

0



freshwaters affected ecosystem servicesinclude recreational fishing, other forms ofrecreation, and provision of high qualitydrinking water (Table 1).

B. Indicators – Nitrogen Pollution

Nitrogen impacts on ecosystems can be identi-fied by examining changes in biota, or by mea-suring chemical indicators. Ideally, a chemicalindicator provides a warning suggesting thatsensitive biota are at risk before biologicalharm occurs.

Nitrate Leaching

One of the most notable symptoms of nitrogensaturation is increased leaching of nitrogenfrom soils into lakes, streams and groundwater,primarily in the form of nitrate (NO3

–).Streamwater NO3

– concentration is a usefuland simple indicator of the nitrogen status of acatchment because this measure integratesmany nitrogen cycling processes that occurwithin the catchment, including the process-ing of atmospheric nitrogen deposition. Innitrogen limited ecosystems in the westernUnited States, U.S. Forest Service land man-agers have set policy thresholds of 20µeq/L as a concern level indicating potentiallyover-enriched systems. Studies in Europe andthe northeastern U.S. show that nitrogenleaching begins to increase in forests receiving

levels of atmospheric nitrogen depositiongreater than 8-12 kilograms of nitrogen perhectare per year (kg N/ha/yr), although not allforests receiving those levels of depositionshow NO3

– leaching, due to land disturbancehistory, the presence of wetlands and othercharacteristics.

Nutrient Ratios

Other commonly used chemical indicators ofnitrogen enrichment include nutrient ratios infoliage such as nitrogen:phosphorus (N:P),nitrogen:calcium (N:Ca+2), or carbon:nitrogen(C:N). C:N ratios in organic or mineral hori-zons of the soil also indicate ecosystem nitro-gen status or the predisposition to nitrogensaturation. The C:N ratio of the soil and thegrowth rate of the forest influence nitrogenleaching. Forests with soil C:N ratios less than20-25 (indicating high N availability) aremore likely to exhibit nitrogen leaching thanforests with higher C:N ratios.

Biological Indicators

Biological indicators of nitrogen over-enrich-ment or eutrophication include shifts inspecies or biological communities, enhancedestablishment of invasive species, or othermeasures of biodiversity change (Box 6). Localextinction of sensitive species or functionalgroups can also occur. The most sensitiveorganisms exhibiting such changes in responseto nitrogen enrichment in freshwater ecosys-tems such as lakes are small single-celled algaeknown as diatoms. Even a small amount ofadditional nitrogen deposition from air pollu-tion that is transported to waters can inducemajor shifts in the species of diatoms. Eachdiatom species has specific patterning in theshell, so by studying lake sediment cores fromwestern lakes that extend back 100 years ormore, researchers have been able to documentwhether recent diatom species shifts corre-spond with changes in nitrogen deposition.Tree-inhabiting lichens (epiphytes) are highlysensitive indicators of nitrogen air pollution inforests and woodlands. The lowest criticalloads for nitrogen effects are typically based ondiatom or lichen community responses, mak-ing them good early warning indicators forecosystem changes. Nitrogen also affects thebiodiversity of herbaceous or grassland plantcommunities at relatively low levels (Box 6).

Nitrogen effects on these biological indica-tors are linked to changes in ecosystem func-

Figure 5. Continuum of nitrogendeposition impacts as

demonstrated from pastobservations and potential future

effects in Rocky MountainNational Park in Colorado. As

ecosystem nitrogenaccumulation continues,additional acidification or

eutrophication impacts occur tovarious ecosystem receptors.Note that the trajectory line is

conceptual even though theeffects below the current

nitrogen deposition level havebeen documented. Similar

trajectories of additionalecosystem effects as nitrogenaccumulates in the ecosystem

likewise occur in other ecologicalregions. (Figure courtesy of Ellen

Porter, National Park Service).

Synthesis: Continuum of Impacts in Rocky Mountain National Park



tions including alteration of food webs,increased risk of fire, and reduction in impor-tant nursery habitat for commercial fisheries.For example, among the nitrogen sensitivelichen species are those that are vital foragespecies for deer, elk, and the northern flyingsquirrel; the latter is the major prey of the fed-erally endangered spotted owl. Nitrogen depo-sition effects on plant biodiversity can resultin major impacts on ecosystems, includingenhanced invasion by exotic grasses, increasedfire danger, vegetation type change, and disap-pearance of biological species that depend ondeclining native plant species and communi-ties (Box 6). Finally, negative impacts to eel-grass beds from the over-enrichment of coastalwaters can diminish the quality of importantnursery, habitat, and feeding grounds for com-mercially important fish and shellfish in theeastern U.S., such as scallops.

C. Nitrogen Critical Loads

Air pollution thresholds at which excess nitro-gen effects on ecosystems occur can be deter-mined using field studies or estimated by mod-eling. Nitrogen critical loads can also beextended over wide geographic areas or pre-dicted through time by use of models. Critical

loads based on field observations across spatialgradients of varying air pollution exposure orfrom field experiments are known as empiricalcritical loads. Empirical critical loads for nitro-gen deposition effects in selected ecoregions ofNorth America are presented in Table 3.Recent studies demonstrate that exceedanceof empirical critical loads for nitrogen is com-mon across the U.S.

By comparing modeled and empirical criti-cal load values to current and future deposi-tion data and estimates, policymakers canassess current ecosystem condition, set goalsfor ecosystem recovery, and track improve-ment. This information can also aid decisionmaking processes for air pollution controls ormitigation programs for damaged ecosystems.For example, the low end of the critical loadrange in Mediterranean California mixedconifer forests (3 kg N/ha/yr) describes thepoint where impacts begin in the most sensi-tive parts of these ecosystems, specifically,changes in lichen communities. Such a lownitrogen critical load provides a ‘canary in thecoal mine’ threshold that is indicative of ini-tial ecosystem responses to added nitrogen.Ecoregions in the United States where lichencommunities are likely affected by nitrogen airpollution based on nitrogen deposition in

Box 6. BIODIVERSITY

Biological diversity (or “biodiversity”) may simply be defined as the species richness of a geographic area. Biodiversity loss has accel-erated in modern times due to land use change, the introduction of invasive species and other disturbances. Climate change and airpollution also contribute to changes in plant community composition and biodiversity. In polluted regions, the occurrence of sensitivespecies may decrease and lead to replacement by pollution-tolerant species. When air pollution alters the biodiversity or the composi-tion of biological communities, detrimental effects on the provision of valued ecosystem services can occur. The implications for biodi-versity shown by long-term studies of acid deposition and nitrogen pollution are highlighted in the case studies below.

Acid Rain: Diminishing Aquatic Diversity in the NortheastAquatic organisms vary in sensitivity to acidity with sensitive species showing limitations at pH 6.0 and many organisms declining inabundance and richness at pH levels of 5.5 and lower. As acidity increases, sensitive species or sensitive life history stages of specieseither die or seek refuge in less-acidified habitats leaving the original habitat less productive and diverse. The impacts are most severein sensitive high elevation ecosystems that have experienced chronic deposition. Of the 53 fish species recorded in lake surveys in theAdirondack Mountains of New York, half are absent from lakes with pH less than 6.0. Recreational fishes, such as Atlantic salmon, tigertrout, bluegill, walleye and alewife, are among those absent from low-pH lakes. These acidity effects can extend further down the foodchain. In lakes of the Adirondacks and the White Mountains (New Hampshire) an average of 2.4 zooplankton species are lost with eachpH unit decrease. Long-term monitoring of acidifying deposition and surface water chemistry confirm that decreased emissions ofacidic pollutants have resulted in lower deposition and some recovery in pH. The long and complex process of biological recoveryincluding the restoration of soil base saturation is only just beginning.

Nitrogen Pollution, Plant Communities and Biodiversity in CaliforniaBiodiversity of plant communities is sensitive to N added by air pollution. Nitrogen-loving species are often favored and increase inprominence as ecosystem nitrogen availability increases. Forests and woodlands in many regions of the world show large changes inepiphytic lichen communities in response to chronic atmospheric nitrogen deposition. These lichen community impacts occur at nitro-gen pollution thresholds as low as 3-6 kg/ha/yr. Ecologically important lichen species have been eliminated from forests over largeareas of California. Similarly, in coastal central California, native serpentine grassland plant communities are exterminated when the Ndeposition load is 6 kg/ha/yr or greater and are replaced by exotic grasses and other native plant species. These changes result in theloss of the plant species needed for reproduction and survival of the endangered Bay Checkerspot butterfly, and has caused local pop-ulation extinctions of the butterfly. Likewise, nitrogen from air pollution at levels around 8 kg/ha/yr in desert and scrub vegetation com-munities of California favors the growth of exotic annual grasses, crowding out native species and increasing fire risk in some areas.



exceedance of the critical load for licheneffects are shown in Figure 6.

Recent research has shown that by stimulat-ing increased growth of non-native grasses,nitrogen deposition may increase the fre-quency of wildfires in southwestern U.S. desertareas because these grasses provide fuel to sus-tain the spread of fire in areas with little or noprevious fire history. Simulation models haveestimated that the lowest threshold level ofnitrogen that initiates these changes inpinyon-juniper ecosystems is about 3 kgN/ha/yr of nitrogen deposition. Fire riskincreases exponentially above this level toabout 5.7 kg N/ha/yr, at which level grasses aregenerally fully established. This provides anexample of how policymakers or stakeholderscould select a variety of policy thresholds (3.0,5.7 kg/ha/yr, or levels in between) dependingon risk tolerance and goals for protectingnative vegetation in different types of ecosys-

tems. Nitrogen deposition currently exceedsthese ecological thresholds in many southwesternU.S. ecosystems, indicating the importance ofthis information in air pollution policy or landmanagement decision making in this region.

Another example of nitrogen depositioneffects relative to ecological thresholds occurs inthe Colorado Rocky Mountains. In this region,alpine vegetation has begun to shift toward ahigher proportion of grasses. This shift occursat a threshold of around 4 kg N/ha/yr, whichapproximates current nitrogen deposition lev-els in areas of the Rockies most influenced byagricultural and urban emissions. Theresponses of individual plant species to nitro-gen (critical load of 4 kg N/ha/yr) is a muchmore sensitive indicator of nitrogen effectsthan soil acidification (critical load of 10-15 kgN/ha/yr). Nitrogen deposition also has initi-ated shifts in diatom species to those that favorhigher nitrogen levels in some high-elevation

Table 3. Critical Loads (CL) of Nitrogen Deposition for Effects on Selected NorthAmerican Ecosystems

Ecosystem Chemical or Biotic Response CL for N Deposition(kg N/ha/yr)

Arctic Tundra Plant community change; grass growth 1-3

Arctic Tundra Changes in shrub, bryophyte, lichen cover 6-11

Boreal Shrublands Decrease in shrub cover; increase in grass cover 6

Northern Forests Change in soil community structure 5-7

Northern Hardwood Increased surface water nitrate leaching 8and Coniferous Forests; Eastern Temperate Hardwood Forests

Northwest Forested Lichen community change from oligotrophic 3-5Mountains and to eutrophic species dominanceMediterranean California Mixed Conifer Forests

Alpine Changes in herb and grass species 4-10composition

Great Plains Tall Grass Change in biogeochemical N cycling, plant and 5-15Prairie insect community shifts

Mediterranean California Increased surface water nitrate leaching 17Mixed Conifer Forests

Mediterranean California Native herbs replaced by annual grasses; 6Serpentine Grasslands loss of checkerspot butterfly habitat

Rocky Mountain Western Freshwater eutrophication 2Lakes

From: Pardo, L.H., Robin-Abbott, M.J., and Driscoll, C.T., eds. 2011. Assessment of nitrogen deposition effects and empiricalcritical loads of nitrogen for ecoregions of the United States. Gen. Tech. Rep. NRS-80. Newtown Square, PA: U.S. Departmentof Agriculture, Forest Service, Northern Research Station.



lakes at levels of 1.4 to 1.5 kg N/ha/yr as wetdeposition in national parks and Class IWilderness areas in Colorado, the greaterYellowstone Ecosystem, and the eastern SierraNevada Mountains. This threshold is amongthe lowest identified for any ecosystem changesresulting from nitrogen deposition, makingdiatoms ideal early warning indicators ofdecline in aquatic ecosystem condition fromnitrogen air pollution. Current nitrogen depo-sition levels are higher than these thresholdsfor most of the western U.S.

3. MERCURY POLLUTIONA. Effects

Mercury is a naturally occurring metal and alocal, regional and global pollutant. Mercuryemissions from electric utilities, incinerators andindustrial manufacturing are among the largestsources of mercury to the environment in theU.S. Mercury in the air and water are not directpublic health risks at levels commonly found inthe U.S. The risk to human and ecologicalhealth typically occurs through consumption ofmercury-contaminated fish and other biota.Inorganic mercury (Hg) is deposited to thelandscape, transported from soils to wetlandsand surface waters, and converted by bacteria tomethylmercury (MeHg) – the organic form ofmercury that is readily absorbed by fish andother organisms. Once ingested, mercury canbioaccumulate in organisms and biomagnifythrough the food web to elevated concentra-tions in fish and other organisms that are con-sumed by people and wildlife. Fish contamina-tion by MeHg poses a widespread problem infreshwater, and in coastal and marine recre-ational and commercial fisheries.

As mercury sampling in lakes and rivers hasexpanded, the extent of waters known to beimpaired by mercury pollution has increased. In2008, all 50 states, one U.S. territory, and threeNative American tribes issued mercury advi-sories for human fish consumption covering16.8 million lake acres and 1.3 million rivermiles. That was a 19% increase in lake areaunder advisory and a 42% rise for rivers com-pared to 2006. The number of statewide mer-cury advisories for coastal waters increased from12 in 2004 to 15 in 2008. These increases likelydo not reflect increases in Hg deposition, butrather increases in measurements documentingthe widespread nature of mercury contamina-tion. In addition to contamination of fish,MeHg poses risks to fish-eating wildlife such asloons, mink, eagles and otter. MeHg concentra-

tions can also be elevated in organisms thatfeed on aquatic insects, such as songbirds andbats, and in organisms that dwell in wetlandsand upland environments such as the Bicknell’sthrush – a migratory songbird that breeds in theforested mountains of the Northeast.

B. Indicators - MercuryIndicators for Human and EcologicalHealth

Mercury concentrations in fish and other ani-mals routinely exceed human and wildlifehealth levels. Human health indicators formercury are based on the concentration ofMeHg in fish tissue that is considered safe forthe average consumer (Figure 7). The U.S.Environmental Protection Agency has recom-mended a human health criterion of 0.3 partsper million (ppm) in fish tissue which repre-sents the maximum advisable concentration ofMeHg in fish and shellfish that protects theaverage consumer among the general popula-tion. Particularly sensitive groups of peopleincluding women in child-bearing years andchildren under 12 years of age are advised tolimit consumption to fish low in mercury.Many states have set even more stringent

Figure 6. Lichen Based CriticalLoad Exceedance Map. Areasshown in red and orangereceived atmospheric nitrogendeposition at levels deleteriousto communities of epiphytic(tree dwelling) lichens. This mapshows that these effects occurin over half of the forested landarea, including urban forests, ofthe continental U.S. Levels ofcertainty in the critical loadexceedance estimates varyamong ecoregions dependingon the amount of availablelichen community data. (FromPardo, L.H. and others 2011.Effects of nitrogen depositionand empirical nitrogen criticalloads for ecoregions of theUnited States. EcologicalApplications,doi: 10/1890/10-234.1).

Exceedance of Critical Loads of N Uncertainty

Below CL Reliable

At CL Fairly Reliable

Above CL min Expert Judgement

Above CL max



human health protection levels. Maine andMinnesota use 0.2 ppm as the human healththreshold, as does Canada.

Ecological effects thresholds for mercury aregenerally based on the concentrations of mer-cury in the tissue, blood, or diet (often fish) ofan organism that are associated with adverseimpacts. Adverse impacts to biota from mer-cury exposure include reduced reproductivesuccess, decreased egg incubation time andother behavioral changes, and neurologicalproblems such as the loss of movement knownas ataxia. Several ecological effects thresholdshave been defined in the literature and manyare lower than human health effect thresh-olds. For example, reproductive effects in fish-eating birds are reported at levels of 0.16 ppmin their prey fish. Significant adverse repro-ductive impacts on loons are commonly citedat the threshold of 3.0 ppm in loon blood.

These indicators of human and ecologicaleffects can be used to assess and communicaterisk, to determine the presence of biologicalmercury hotspots where concentrationsexceed established thresholds, to establish tar-gets for critical loads estimates, and to assessthe effectiveness of mercury emissions reduc-tions on target species.

Chemical Indicators of FreshwaterSensitivity to Mercury

There is large variation in the degree to whichmercury deposited onto the landscape will betransported to lakes and streams throughdrainage waters and converted from inorganicmercury to MeHg by bacteria in soils, wet-lands and lake and river sediments. The rateof methylation by these bacteria is affected bypH, sulfur and dissolved organic carbon(DOC) concentrations and, in Canadianshield lakes, was found to increase with

increasing water temperatures. Once con-verted to the MeHg form, mercury can bioac-cumulate in individual organisms and biomag-nify in the food web. As a result, total mercuryand MeHg concentrations in surface watersmay not correlate well with mercury concen-trations in biota, such as fish. In areas wheremercury deposition is low or moderate, levelsin fish and wildlife may be disproportionatelyhigh if conditions are conducive to MeHg pro-duction and bioaccumulation. This has beenobserved in some Alaskan ecosystems, such asNoatak and Gates of the Arctic NationalParks, in Kejimikujik National Park of NovaScotia, and large areas of the Northeastincluding the Adirondacks.

In recent years scientists have turned theirattention to understanding the specific condi-tions that make an ecosystem sensitive to mer-cury. Sensitive systems may be more efficientat converting inorganic mercury to MeHg ormore efficient at bioaccumulating mercury ateach level in the food chain. In general, acidicecosystems with low productivity and high sul-fate and DOC tend to be sensitive to mercuryinputs and to exhibit higher fish mercury con-centrations. Several of these chemical indica-tors are influenced by inputs of acidifyingdeposition leading to interactive effectsamong two or more atmospheric pollutants. Astudy of freshwater ecosystems in the north-eastern U.S. used the following chemical indi-cators of mercury sensitivity: total phosphorus,DOC, ANC, and pH (Table 4). These waterchemistry indicators provide managers with ameans for evaluating where MeHg concentra-tions in fish are likely to be high and can helpprioritize monitoring and assessment efforts.

Watersheds particularly sensitive to mercuryare more commonly found in the southern andeastern U.S., Great Lakes, and isolated areasin the western U.S. The sensitive regionsshown in Figure 8 were identified based on thephysical and chemical characteristics of awatershed that cause it to convert inorganicmercury to MeHg at a higher rate than otherwatersheds.

Recovery from mercury deposition has beenstudied in watersheds where emission controlshave been implemented for large sources suchas municipal waste incinerators. Studies fromsouthern New Hampshire suggest that eventhough mercury is persistent and bioaccumu-lates in the environment, decreased inputsfrom local sources have been accompanied bydecreased concentrations in top predatorsincluding fish and loons. In New Hampshire

Figure 7. Fish tissue mercuryconcentration ( ppm; same asµg/g) across the U.S. All data

standardized to 14 inchlargemouth bass, skin-off fillets.(Figure is derived from a modeland national dataset described

in: Wente, S.P., 2004. U.S.Geological Survey Scientific

Investigation Report2004-5199, 15 p.).

Fish-tissue MercuryConcentration (ppm)

states

1.0



mercury emissions upwind of a biological mer-cury hotspot declined by 45% between 1997and 2002. Mercury concentrations in yellowperch and loon blood in the region declined32 and 64% respectively between 1999 and2002 – rates much greater than observed else-where in the region. These results suggest thatreduced emissions and deposition of mercuryfrom local and regional sources are needed torestore healthy wildlife and safe fisheries andthat return to levels consistent with humanhealth criterion is likely to occur withindecades, not centuries. However continuedmonitoring is essential in light of increasingcontributions of mercury from global sourcesand the need to better understand potentialinteractions with other pollutants and withclimate change.

C. Mercury and Critical Loads

Science-based air pollution thresholds andcritical loads for mercury are not as well estab-lished as those of sulfur, nitrogen and acidity.Efforts are underway to develop and refinecritical loads for mercury by investigating thelinkage between atmospheric deposition lev-els, methylation processes and chemical andbiological thresholds for human and ecologicalhealth. Critical loads for mercury and othertrace metals have been estimated for parts ofEurope using a set of very general assumptionsrelating atmospheric Hg to Hg concentrationsin groundwater, food crops, and aquaticorganisms. Available data for calculatingexceedances are quite limited. However theresults suggest that a large part of the land-scape (approximately 50%) exceeds mercury

critical loads for ecosystem effects. Current understanding of the links between

mercury emissions and deposition and biologi-cal responses in humans and ecosystems ishampered by a lack of consistently collectedlong-term data on mercury levels in waters,soils, and biota. Mercury monitoring effortsvary widely among states and are difficult tointegrate and synthesize to establish responsepatterns. In 1996 the Mercury DepositionNetwork (MDN) began measuring mercurydeposition in precipitation (wet deposition)(http://nadp.sws.uiuc.edu/mdn/) and currentlyincludes 115 sites in the United States andCanada as part of the National AtmosphericDeposition Program (NADP). In 2009, theAtmospheric Mercury Network (AMNet) alsojoined NADP and includes 21 sites that trackthe concentration of different forms of mer-cury in precipitation. There is further need fora comprehensive environmental mercurymonitoring network.

USING AIR POLLUTIONTHRESHOLDS IN POLICY &MANAGEMENT

Ample evidence exists to advance the wideruse of air pollution thresholds in policy, man-agement and regulatory issues. Policy and reg-ulatory decisions in response to air pollutantemissions are based on many economic, politi-cal, human health, environmental, scientificand sociological considerations and tradeoffs.Air pollution thresholds can be used to helpevaluate and monitor these tradeoffs and sev-

Table 4. Indicators of Surface WaterSensitivity to Mercury. The followingthresholds are associated with con-centrations of methylmercury in yel-low perch greater than 0.3 ppm inlakes in the northeastern U.S.

Indicator Threshold

Total phosphorus 4 mg/L

Surface water pH

16 esa © The Ecological Society of America • [email protected]


cussed in the recent US-Canada ProgressReports, which detail progress achieved inimplementing the US-Canada Air QualityAgreement. These assessments rely on ecosys-tem element cycling models from which criti-cal loads can be estimated and the effects ofthe Clean Air Act and other emissions reduc-tions laws and policies can be evaluated.Long-term measurements have generallyshown improvements in some surface-waterindicators of acidification, such as sulfate con-centrations and pH, in the Northeast U.S.over the past 30 years. Most of these watershave not recovered to pre-acidification condi-tions, and many remain in excess of criticalload thresholds.

Dynamic ecosystem modeling can be usedto simulate likely ecosystem responses in aspecific year to future emissions reductions orincreases. Recent modeling work simulatesthat future decreases in SO2 and NOx emis-sions of greater than 50% (relative to clean airlaws implemented as of 2005) will be neces-sary to decrease the number of chronicallyacidic lakes in the Adirondacks by one-thirdto one-half by the year 2050. Despite theseimprovements, model results indicate thatmany of the currently chronically acidic lakeswill improve only to an episodically acidic sta-tus, so the net change in the sum of chronicplus episodically acidic lakes is likely to stayabout the same or improve slightly dependingon the extent to which emissions decrease.This work shows how models can simulatewhether emissions policy goals for ecosystemrecovery are likely to be met within a specifictime frame.

Finally, critical loads can inform the devel-opment of national air quality standardsknown as the “secondary standards” that areaimed at protecting environmental resourcesfrom air pollution. The Clean Air Act requiresEPA to set national air quality standards for sixcriteria pollutants (nitrogen oxides, sulfuroxides (SOx), particulate matter, ozone, carbonmonoxide, and lead) based on primary (health-based) and secondary (welfare-based) consider-ations. “Welfare” includes consideration ofenvironmental harm. The law also requiresEPA to periodically review the scientific crite-ria upon which the standards are based.

While EPA generally reviews criteria andstandards for each of the six criteria pollutantsindividually, EPA decided to jointly examineNOx and SOx compounds in a recent review ofthe secondary standards. In a policy assess-ment completed as a part of the NOx/SOx sec-

eral examples of their effective applicationexist in the U.S., Europe and Canada.

1. U.S. Policy Use of Thresholds

Although the 1970 U.S. Clean Air Act man-dates protection of human health and welfare(which includes ecological effects), neitherthe Clean Air Act nor its 1990 amendmentsspecifically mandates a critical loads approachfor addressing air pollution. Class I areas aredesignated federal wilderness areas that weregiven special protection from degradation byair pollution under the Clean Air ActAmendments of 1977. It is becoming increas-ingly evident that critical loads for effects onterrestrial and aquatic ecosystems areexceeded in many Class I areas, even thoughthe human health-based standards for NO2and SO2 are rarely exceeded in these areas.State and federal environmental and regula-tory agencies and multi-stakeholder organiza-tions are increasingly turning to critical loadsas a type of threshold that can aid in thedevelopment of air quality standards, theassessments of emissions regulations, and otherpolicies aimed at protecting or improvingecosystem condition. In Rocky MountainNational Park in Colorado, the critical load fornitrogen deposition impacts on aquatic diatomcommunities provides the basis for a nitrogendeposition goal to achieve resource protection.The National Park Service, the State ofColorado Department of Public Health andEnvironment, the U.S. EnvironmentalProtection Agency, and interested stakeholderscollaborate in the Rocky Mountain NationalPark Initiative to develop strategies to reduceair pollutant emissions that contribute tonitrogen deposition in the Park.

The EPA recently started using critical loadsto describe threshold effects in its annual AcidRain Progress Reports. In the 2009 Acid RainProgress Report, critical loads for acid deposi-tion were calculated for over 1,300 lakes andstreams in the Northeast and Mid-AppalachianHighlands regions of the eastern United States.By comparing critical loads to deposition databefore and after implementation of the AcidRain Program, it was determined that 37% oflakes and streams in those regions where atmos-pheric deposition was in exceedance of the crit-ical load in the 1989-91 period were no longerreceiving sulfur and nitrogen deposition loadsin 2007-2009 that threatened the health ofthese ecosystems.

Critical loads are also presented and dis-

ondary standard review, EPA staff found thatalthough the current secondary standardsserve to protect vegetation from direct damageassociated with exposures to gaseous SO2 andNO2, “currently available scientific evidenceand assessments clearly call into question theadequacy of the current standards with regardto deposition-related effects on sensitiveaquatic and terrestrial ecosystems, includingacidification and nutrient enrichment”(USEPA 2011). They further conclude that“consideration should be given to establishinga new ecologically relevant multi-pollutant,multimedia standard to provide appropriateprotection from deposition-related ecologicaleffects of oxides of nitrogen and sulfur on sen-sitive ecosystems with a focus on protectingagainst adverse effects associated with acidify-ing deposition in sensitive aquatic ecosystems”(USEPA 2011). Finally, the policy assessmentrecommends the use of a critical loadsapproach in establishing and monitoring thissuggested ecologically relevant standard.

The conclusions of the EPA staff policy doc-ument were supported in a review by the inde-pendent Clean Air Science AdvisoryCommittee (CASAC), which advises EPA onscientific issues related to Clean Air Actimplementation. Similar to the CASAC find-ings, the data and information presented heresupport the advance of secondary air qualitystandards to enhance recovery of sensitiveecosystems from acidifying atmospheric nitro-gen and sulfur pollution. In particular, sec-ondary standards for NOx and SOx could bedeveloped using a critical loads approach tolink pollution concentrations in the air withdeposition and ecological effects based onestablished indicators for surface water chem-istry and biology. In addition, the wealth ofresearch-scale data on thresholds and criticalloads becoming available for U.S. ecosystemssupport the exploration of other collaborativepolicy and management approaches to strate-gically reduce pollution in areas where demon-strated impacts exist. Such a process was usedin Colorado, as discussed above.

2. European Policy Use of Thresholds

European scientists and policymakers haveused a critical loads approach for addressingnitrogen and sulfur effects in ecosystems since1994. The air pollution abatement strategiesunder the European Union Convention onLong-range Transboundary Air Pollution andunder the European Commission’s Thematic

Strategy on Air Pollution are linked to acidityand nitrogen critical loads as the basis fornegotiating national emissions maxima.Meanwhile, ecosystem monitoring has demon-strated that pollution abatement efforts havedecreased acidification, and to a lesser degree,nitrogen levels in ecosystems. Once scheduledemissions controls are completely imple-mented, European areas that exceed acidifica-tion critical loads will be reduced from 93 mil-lion hectares in 1990 to an estimated 15million hectares.

In Europe, application of critical loads hasconnected science to policy by providingmethodologies for using scientific evidence todefine pollution limits and to assist in settingemission control targets within a broad multi-nation policy framework. As a result, nitrogencritical loads have been developed for NO3

–

leaching from forests and for changes toEuropean plant species diversity for mostmajor vegetation types found in Europe.Similarly, acidification critical loads havebeen developed to protect terrestrial ecosys-tems and thousands of lakes and streams.Critical loads for mercury, lead and cadmiumhave also been developed in Europe based onendpoints (e.g., human health and ecosystemfunction) and indicators (metal concentrationsin soil, food crops, and biota).

3. Canadian Policy Use of Thresholds

Soil nutrient declines caused by acidifyingdeposition have resulted in extensive timber



Figure 9. Equipment usedto monitor air quality atGreat Smoky MountainsNational Park.

productivity loss in Atlantic Canadian forests.Fish declines in lakes and rivers of easternCanada have significantly impacted theCanadian recreational fishing industry.Concern about acidification impacts toecosystem services (Table 1) motivatedCanadian policymakers to apply critical loadswhen establishing regional and, later, nationalemissions reduction policies. Sulfur-based crit-ical loads that initially served as the basis foremission control policy have been loweredover time as understanding of air pollutioneffects on forests and surface waters expanded.Canada periodically reviews its critical loadsto ensure that they remain consistent with thelatest scientific information and policyrequirements. Since the 1990s both S and Nare considered in critical load analyses,although airborne S pollutants continue to bethe predominant, anthropogenic acidifyingagent in Canada.

The national emission control policy agreedto in 1998 by the federal and all provincialand territorial governments is called theCanada-Wide Acid Rain Strategy for Post2000. It seeks to meet the environmentalthresholds of critical loads for acid depositionacross Canada; decreasing SO2 emissionswhere needed to meet critical loads, and mini-mizing growth in emissions of both SO2 andNOx where acid deposition is currently belowcritical loads. As of 2008, the application ofthis policy has decreased SO2 emissions inCanada 63% below1980 levels. Based on criti-cal loads projections indicating that acid rainwill continue to damage sensitive ecosystemseven after full implementation of currentCanadian and U.S. control programs, furtheremission controls will be needed.

4. Global Policy Use of Thresholds

Fifty-one European countries, Canada, andthe U.S. participate in the United NationsConvention on the Long-Range Transport ofAir Pollutants (CLRTAP). Under this treaty,the Gothenburg Protocol to AbateAcidification, Eutrophication and Ground-level Ozone went into effect in 2003, and estab-lished more stringent emissions targets for SO2,NOx, NH3, and other pollutants. UnderCLRTAP, European countries compare criticalloads of S and N deposition for acidification andeutrophication to estimates of current deposi-tion as a means of evaluating the effectivenessof emissions control measures. Countries areencouraged to work together to reduce harmful

impacts to surface waters, soils, and vegetation.In addition to Europe, both Canada and theUnited States have ratified CLRTAP based onan understanding of the global nature of air pol-lution transport and the need to reduce air pol-lution impacts.

There is currently no mechanism in placefor decreasing global sources of mercury.However, the February 2009 United NationsEnvironment Program Governing Council,including the U.S., China, India and 138other countries established a process to createa legally binding agreement to control globalmercury pollution by 2013. Negotiationsstarted in June 2010, and will address a broadset of issues such as decreasing mercury emis-sions from human sources, improving manage-ment of mercury in the waste stream and atstorage sites, curbing demand for mercury inproducts, decreasing mercury mining, remedi-ating contaminated sites, and enhancingglobal monitoring. Once the framework isestablished, participating nations must eachratify the treaty in order for it to go into effect.The impact of this framework will be assessedover time based in part on the extent to whichfish tissue mercury concentrations havedeclined to below human health thresholds.

ADVANCING THE USE OF AIRPOLLUTION THRESHOLDS INPOLICY

Nitrogen, sulfur, and mercury pollution arealtering the Earth’s ecosystems. Coupled withother stressors including land use changes,invasion of exotic species, and climate change,these pollutants are threatening the provisionof ecosystem services such as air and waterpurification, waste detoxification, climate reg-ulation, soil regeneration, biodiversity mainte-nance, and production of crops, timber, andfish. Ongoing and future projected climatechange is likely to become an increasing envi-ronmental stressor in coming decades.However, little research to date has quantita-tively explored the interaction of climatechange with the effects of nitrogen, sulfur, andmercury pollutants in ecosystems.

In the face of large-scale global change, nat-ural resource managers, air regulators, andpublic stakeholders need to know whetheremissions controls are effective in producingthe ecosystem benefits that were anticipated.While scientists can often determine an airpollution threshold where ecological change islikely to occur and define the nature and

© The Ecological Society of America • [email protected]


18 esa

degree of change, decision makers must weigha number of scientific and societal considera-tions in deciding which ecosystem changes areof concern and what level of protection isdesired to address these concerns. Further-more, long-term monitoring programs fortracking trends in air pollution, soil and waterchemistry, and aquatic and terrestrial biota area critical component for protecting naturalresources and for the development, refine-ment, and application of ecological and policythresholds.

This interaction between science and deci-sion-making often proceeds iteratively. Onceecosystem components that should be pro-tected (sensitive receptors) have been identi-fied, scientists can identify response thresholds(e.g., critical loads) for those components, landmanagers and stakeholders can determinedesired protection levels (policy thresholds suchas target loads), and environmental regulatorscan evaluate tradeoffs to determine whetheremissions controls are warranted to achievethese goals. Field studies and modeling canhelp by linking potential threshold responses(ecological thresholds) to stressors. Moni-toring is essential to determine whether thedesired response is achieved.

The National Ambient Air Quality Standards(NAAQS) for air pollutants such as NOx andSOx are based on concentrations of these pollu-tants in ambient air rather than on depositionlevels experienced by ecosystems. Scientificprogress has improved our ability to relate ambi-ent air concentrations to atmospheric deposi-tion inputs and effects through the estimation ofcritical loads. The secondary standards provide aframework for addressing these issues and ampleevidence exists for applying existing researchand modeling to the case of acidifying deposi-tion impacts on sensitive aquatic ecosystems.Similar applications of secondary standardstoward protection of terrestrial ecosystems fromthe effects of nitrogen and sulfur pollutionwould also be of great benefit. Nitrogen asammonia and ammonium (NH3 and NH4

+), areincreasingly important sources of nitrogen airpollution, but are not regulated by EPA as crite-ria pollutants in the NAAQS.

Air pollution thresholds based on scienceprovide a mechanism for evaluating the extentto which ecosystem services have been com-promised and for restoring impaired ecosys-tems. Establishing priorities such as the levelsat which various ecosystem services should bemaintained will require the mutual engage-ment of public stakeholders, policymakers,

and scientists. Use of ecological thresholds forassessing the impacts of air pollution on essen-tial ecosystem services and for informing pub-lic policy is gaining ground. These ecologicalthresholds provide a strong basis for develop-ment of policy thresholds and offer a scientifi-cally sound approach to protecting and restor-ing U.S. ecosystems.

For Further Reading

Aber, J.D., C.L. Goodale, S.V. Ollinger, M.-L.Smith, A.H. Magill, M.E. Martin, R.A.Hallett, and J.L. Stoddard. 2003. Is nitrogendeposition altering the nitrogen status ofnortheastern forests? BioScience. 53: 375-389.

Burns, D.A., T. Blett, R. Haeuber, and L.H.Pardo. 2008. Critical loads as a policy toolfor protecting ecosystems from the effects ofair pollutants. Frontiers in Ecology and theEnvironment. 6: 156-159.

Driscoll, C.T., G.B. Lawrence, A.J. Bulger, T.J.Butler, C.S. Cronan, C. Eagar, K.F.Lambert, G.E. Likens, J.L. Stoddard, andK.C. Weathers. 2001. Acidic deposition inthe northeastern United States: Sourcesand inputs, ecosystem effects, and manage-ment strategies. BioScience. 51: 180-198.

Driscoll, C.T., et al. 2007. Mercury Matters:Linking mercury science with public policyin the northeastern United States. HubbardBrook Research Foundation. Science LinksTM

Publication. Vol. 1, no. 3, Hanover, NH.Dupont, J., T.A. Clair, C. Gagnon, D.S.

Jeffries, J.S. Kahl, S.J. Nelson, and J.M.Peckenham 2005. Estimation of criticalloads of acidity for lakes in northeasternUnited States and eastern Canada.Environmental Monitoring and Assessment.109: 275–29.

Evers, D.C., Y.-J. Han, C.T. Driscoll, N.C.Kamman, W.M. Goodale, K.F. Lambert,T.M. Holsen, C.Y. Chen, T.A. Clair, T.J.Butler. 2007. Biological mercury hotspotsin the northeastern United States andsoutheastern Canada. BioScience. 57: 1-15.

Fenn, M.E., J.S. Baron, E.B. Allen, H.M.Rueth, K.R. Nydick, L. Geiser, W.D.Bowman, J.O. Sickman, T. Meixner, andD.W. Johnson. 2003. Ecological effects ofnitrogen deposition in the western UnitedStates. BioScience. 53: 404-420.

The Heinz Center. 2010. Indicators of EcologicalEffects of Air Quality. The H. John Heinz IIICenter for Science, Economics and theEnvironment. Washington, D.C. Availableonline at: http://heinzctr.org/Programs/Reporting/Air_Quality/index.shtml

Lovett, G.M., and T.H. Tear. 2008. Threatsfrom Above: Air Pollution Impacts on



Ecosystems and Biological Diversity in theEastern United States. The Nature Con-servancy and the Cary Institute of Eco-system Studies.

Pardo, L.H., M.J. Robin-Abbott, and C.T.Driscoll (eds.). 2011. Assessment of Nitro-gen Deposition Effects and EmpiricalCritical Loads of Nitrogen for Ecoregions ofthe United States. Gen. Tech. Rep. NRS-80. Newtown Square, PA: U.S. Departmentof Agriculture, Forest Service, NorthernResearch Station.

USEPA. 2009. 2008 Biennial National Listingof Fish Advisories. EPA 823-F-09-007.http://water.epa.gov/scitech/swguidance/fishshellfish/fishadvisories/upload/2009_09_16_fish_advisories_tech2008-2.pdf

USEPA. 2009. Acid Rain and Related Pro-grams: 2008 Environmental Results. http://www.epa.gov/airmarkt/progress/ARP_3/ARP_2008_Environmental_Results.pdf

USEPA. 2011. Policy Assessment for theReview of the Secondary National AmbientAir Quality Standards for Oxides ofNitrogen and Oxides of Sulfur. EPA-452/R-11-005a. U.S. Environmental ProtectionAgency, Office of Air Quality Planning andStandards Health and EnvironmentalImpacts Division, Research Triangle Park,North Carolina. 364 p. http://www.epa.gov/ttnnaaqs/standards/no2so2sec/data/20110204pamain.pdf

Acknowledgements

Funding for this project was provided byPurchase Order EP09H001132 from the USEnvironmental Protection Agency to theEcological Society of America.

About the Scientists

Mark E. Fenn, USDA Forest Service, PacificSouthwest Research Station, Riverside, CA92507Kathleen F. Lambert, Harvard University,Harvard Forest, Petersham, MA 01366Tamara Blett, National Park Service, AirResources Division, Denver, CO 80225 Douglas A. Burns, U.S. Geological Survey,Troy, NY 12180Linda H. Pardo, USDA Forest Service, North-ern Research Station, Burlington, VT 05403Gary M. Lovett, Cary Institute of EcosystemStudies, Millbrook, NY 12545 Richard Haeuber, U.S. EnvironmentalProtection Agency, Clean Air Markets

Division, Washington, DC 20460David C. Evers, BioDiversity ResearchInstitute, Gorham, ME 04038Charles T. Driscoll, Department of Civil andEnvironmental Engineering, SyracuseUniversity, Syracuse, NY 13244Dean S. Jeffries, Environment Canada,National Water Research Institute,Burlington, ON L7R4A6

Science Writing and Layout

Mark Schrope, Science writerBernie Taylor, Design and layout

About Issues in Ecology

Issues in Ecology uses commonly-understood lan-guage to report the consensus of a panel of sci-entific experts on issues related to the environ-ment. The text for Issues in Ecology is reviewedfor technical content by external expert review-ers, and all reports must be approved by theEditor-in-Chief before publication. This reportis a publication of the Ecological Society ofAmerica. No responsibility for the viewexpressed by the authors in ESA publicationsis assumed by the editors or the publisher.

Jill S. Baron, Editor-in-Chief,U.S. Geological Survey and Colorado StateUniversity, [email protected].

Advisory Board of Issues inEcology

Charlene D'Avanzo, Hampshire College Robert A. Goldstein, Electric Power ResearchInstitute Noel P. Gurwick, Union of ConcernedScientists Rachel Muir, U.S. Geological Survey Christine Negra, the H. John Heinz III Centerfor Science, Economics, and the Environment Louis Pitelka, University of Maryland -Center for Environmental Science Sandy Tartowski, USDA - AgriculturalResearch Service, Jornada Experimental RangeDavid S. Wilcove, Princeton University Kerry Woods, Bennington College

Ex-Officio Advisors

Jayne Belnap, U.S. Geological Survey Clifford S. Duke, Ecological Society of AmericaRobert Jackson, Duke University Richard Pouyat, USDA - Forest Service

© The Ecological Society of America • [email protected]


20 esa

ESA Staff

Clifford S. Duke, Director of Science ProgramsJennifer Riem, Science Programs Coordinator

Additional Copies

This report and all previous Issues in Ecologyare available electronically for free at

www.esa.org/issues.

Print copies may be ordered online or by con-tacting ESA:

Ecological Society of America1990 M Street NW, Suite 700Washington, DC 20036(202) 833-8773, [email protected]



setting limits: using air pollution thresholds to protect ... · setting limits: using air...

Documents