Identifying Chemicals That Are Planetary Boundary ThreatsMatthew MacLeod,† Magnus Breitholtz,† Ian T. Cousins,† Cynthia A. de Wit,† Linn M. Persson,‡

Christina Ruden,† and Michael S. McLachlan*,†

†Department of Applied Environmental Science (ITM), Stockholm University, 10691 Stockholm, Sweden‡Stockholm Environment Institute, Linnegatan 87D, Box 24218, 10451 Stockholm, Sweden

ABSTRACT: Rockstrom et al. proposed a set of planetaryboundaries that delimit a “safe operating space for humanity”.Many of the planetary boundaries that have so far beenidentified are determined by chemical agents. Other chemicalpollution-related planetary boundaries likely exist, but arecurrently unknown. A chemical poses an unknown planetaryboundary threat if it simultaneously fulfills three conditions:(1) it has an unknown disruptive effect on a vital Earth systemprocess; (2) the disruptive effect is not discovered until it is aproblem at the global scale, and (3) the effect is not readilyreversible. In this paper, we outline scenarios in whichchemicals could fulfill each of the three conditions, then usethe scenarios as the basis to define chemical profiles that fiteach scenario. The chemical profiles are defined in terms of thenature of the effect of the chemical and the nature of exposure of the environment to the chemical. Prioritization of chemicals incommerce against some of the profiles appears feasible, but there are considerable uncertainties and scientific challenges thatmust be addressed. Most challenging is prioritizing chemicals for their potential to have a currently unknown effect on a vitalEarth system process. We conclude that the most effective strategy currently available to identify chemicals that are planetaryboundary threats is prioritization against profiles defined in terms of environmental exposure combined with monitoring andstudy of the biogeochemical processes that underlie vital Earth system processes to identify currently unknown disruptive effects.

■ INTRODUCTION

Rockstrom et al.1,2 introduced the concept of planetaryboundaries that delimit a “safe operating space for humanity”.The concept emerged from a recognition that human activitiescan lead to impacts at a planetary scale that threaten the vitalEarth system processes that allow humanity to thrive. Vital Earthsystem processes were defined by Rockstrom et al. as“biophysical processes of the Earth System that determine theself-regulating capacity of the planet”. One of the nine planetaryboundaries identified by Rockstrom et al. is “chemical pollution”,and four other planetary boundaries they identified are alsogoverned by chemical agents. We recently postulated that“chemical pollution” is not a single category in the planetaryboundary framework, but rather a placeholder for currentlyunrecognized planetary boundaries governed by chemicalagents.3 We dubbed chemical-related problems that coulddestabilize the Earth system and that we are currently ignorantabout chemical pollution planetary boundary threats. Chemicalpollution may pose unacceptable risks to human health or theenvironment on local, regional, and global scales, and societyattempts to mitigate known risks and respond to new ones withcontinually evolving national and international regulations andguidelines. However, confronting currently unknown planetaryboundary threats from chemical pollution is an importantsocietal task that remains unaddressed. By highlighting theimportance of vital Earth system processes as effect end points

for chemical pollutants, the planetary boundary conceptaddresses a dimension of chemicals management that isneglected in current chemicals policy.In a previous paper,3 we identified three conditions that must

be simultaneously met for chemical pollution to pose a planetaryboundary threat (Table 1). The first condition (Condition 1) isthat the chemical pollution has a disruptive effect on a vital Earthsystem process. Importantly, we must recognize and accept thatwe are currently ignorant of this effect, as we are likewisecurrently ignorant of the planetary boundary. The secondcondition (Condition 2) is that the disruptive effect is notdiscovered until it is, or inevitably will become, a problem at theplanetary scale. If the disruptive effect is discovered before it is aproblem at the planetary scale, then it is not a planetary boundarythreat since it would be possible to take action to control thechemical pollution before the planetary boundary is exceeded.The third condition (Condition 3) is that the effects of thepollution in the environment cannot be readily reversed. If theycan be readily reversed, then exceedance of the planetaryboundary could be corrected, and the problem does not fit ourdefinition of a planetary boundary threat. In summary, chemical

Received: April 17, 2014Revised: August 28, 2014Accepted: September 2, 2014Published: September 2, 2014

Policy Analysis

pubs.acs.org/est

© 2014 American Chemical Society 11057 dx.doi.org/10.1021/es501893m | Environ. Sci. Technol. 2014, 48, 11057−11063

pollution is a planetary boundary threat if it impacts a vital Earthsystem process, if this impact is not discovered until the planetaryboundary is exceeded, and if the effects that result from theexceedance cannot be readily reversed.Confronting chemical pollution planetary boundary threats

requires that the causative agents be identified so that action canbe taken to control pollution before the planetary boundary isexceeded. If society is rational and capable of responding tothreats, then once the potential for a disruptive effect on a vitalEarth system process is identified action will be taken to preventexceedance of the planetary boundary. Experience with theplanetary boundaries that have so far been identified indicatesthat society will not always act quickly and decisively to deal withknown planetary boundary issues, but this failure to act is aseparate problem from the challenge of identifying unknownplanetary boundary threats.In this paper, we develop a set of profiles of chemicals that are

potential planetary boundary threats. Using the three conditionsfor chemical pollution to pose a planetary boundary threat as astarting point, we define chemical pollution scenarios that lead tothe fulfillment of each of the three conditions. Then, for eachscenario we define chemical profiles in terms of the currentlyunknown effect of the chemical on a vital Earth system process,environmental fate properties, and characteristics of use andemission. Finally, we discuss the prospects for using the profilesto prioritize chemicals according to their potential to be chemicalpollution planetary boundary threats.Scenarios for Chemical Pollution to Fulfill the

Conditions to Pose a Planetary Boundary Threat. Ourpoint of departure to develop profiles for chemicals that arepotential planetary boundary threats is to identify scenarios inwhich a chemical could fulfill each of the three conditions forbeing a planetary boundary threat. A useful point of reference is ascenario that has characterized much of our experience withchemical pollutants and that has strongly influenced currentchemical regulation. In this “reference scenario”, illustrated inFigure 1, society introduces a chemical into commerce withoutfully understanding its effects on the environment, andconcentrations in “local” environments where the chemical isreleased begin to rise (solid green line and primary axis in Figure

1). Eventually, the concentration in the environment crosses acritical threshold, and effects are observed in regions where thechemical is released locally (dashed green line and secondary axisin Figure 1). If the effects are unacceptable, society responds bycurtailing emissions of the chemical, but this action requires acertain response time, ΔtRESP, during which concentrationscontinue to rise and effects become more prevalent. AfterΔtRESP,concentrations begin to decline in response to society’s actions,and the prevalence of the effect of the chemical in the affectedareas disappears.In the reference scenario illustrated in Figure 1 the chemical

pollution is hardly dispersed into the global environment, andchemical concentrations fall rapidly after society acts to curtailemissions. A related scenario reflects effective management ofpersistent organic pollutants (POPs) that are identified underinternational regulations (Figure 2). The POPs scenario assumesthat the potential for unacceptable effects at the global scale as aresult of long-range transport is recognized such that the rise inconcentrations can be halted before the critical thresholdconcentration is exceeded on the global scale, but in this case

Table 1. Three Conditions That Must Be Simultaneously Fulfilled for Chemical Pollution to Pose a Planetary Boundary Threat,Scenarios for Chemicals to Fulfill Each Condition, And Profiles of Chemicals That Fulfill Each Scenario

Condition Scenario Chemical Profile

C1 unknown disruptive effect on a vital Earthsystem process

C1-1 unknown disruptive effect on avital Earth system process

C1-1 unknown, but less likely for chemicals that have been used and emitted for many years ifproduction/emissions and environmental concentrations are not increasing.

C2 disruptive effect is not discovered until itis, or inevitably will become, a problem at aplanetary scale

C2-1 rapid development of nearlyhomogeneous pollution levels atthe global scalea

C2-1 emissions dispersed globally combined with moderate mobility, or volatile andpersistent chemicals that are rapidly distributed in the atmosphere.

C2-2 rapid global distribution ofeffects (independent of exposure)

C2-2 unknown, but an example would be a chemical that causes genetic changes in organismsfollowed by distribution and proliferation of the affected organisms.

C2-3 time delay between exposureand effects

C2-3 unknown, but an example would be a subtle effect on long-lived organisms, or atransgenerational epigenetic effect.

C2-4 effects only observable on aglobal scale

C2-4 unknown, but could arise as a result of global exposure to persistent chemicalsdistributed in the oceans or persistent and semivolatile chemicals distributed by acombination of atmospheric and oceanic transport.

C3 disruptive effect is poorly reversible C3-1 poor reversibility of exposurea C3-1 emissions from a technology that society is highly dependent on or from multiplesources that are costly to eliminate, or the chemical is persistent in the exposureenvironment.

C3-2 poor reversibility of effects C3-2 effects that are permanent or cause a regime shift to a new stable state.aThe chemical agent causing disruptive effects may be a persistent, immobile compound formed from the transformation of a mobile compound,which in turn could be the transformation product of an immobile compound. Therefore, transformation products must be considered whenchemicals are evaluated against these chemical profiles.

Figure 1. Reference scenario in which pollution occurs in areas withlocal chemical emissions, unacceptable local effects occur, and there iseffective chemical management after a response time ΔtRESP.

Environmental Science & Technology Policy Analysis

dx.doi.org/10.1021/es501893m | Environ. Sci. Technol. 2014, 48, 11057−1106311058

concentrations in the environment do not respond quickly tosociety’s actions.Below, we explore scenarios that deviate from those illustrated

in Figures 1 and 2 such that one of the three conditions forchemical pollution to pose a planetary boundary threat is fulfilled.

None of the following scenarios alone describes a planetaryboundary threat since that requires fulfilling at least one scenariofor each of Condition 1, Condition 2, and Condition 3.

Scenarios Fulfilling Condition 1: Unknown DisruptiveEffect on a Vital Earth System Process. In the referencescenario, society begins to use chemicals and they enter theenvironment as pollutants, then previously unknown disruptiveeffects are discovered at the local scale. Condition 1 is satisfiedwhen the unknown effect disrupts a vital Earth system process. Ifwe recognize and accept our prior ignorance of the effect, then itis apparent that classifying possible disruptive effects intoscenarios is not possible. Therefore, there is only one relevantscenario for Condition 1:

Scenario C1-1: Chemical Pollution Has an UnknownDisruptive Effect on a Vital Earth System Process. ScenariosFulfilling Condition 2: Disruptive Effect Is Not Discovereduntil It Is, Or Inevitably Will Become, A Problem at aPlanetary Scale. In the reference scenario, the unwanted effectof chemical pollution is discovered at the local scale and remainsspatially contained. Condition 2 is satisfied if the disruptive effectis not discovered until it occurs, or inevitably will occur, on thescale of the entire planet. There are at least four scenarios thatcould lead to this condition being satisfied:

Figure 2. A scenario for effective management of a persistent organicpollutant under current international regulations.

Figure 3. Scenarios for chemical pollution to fulfill Conditions 2 and 3 for being a planetary boundary threat. A, Concentrations are nearly homogeneouson a global scale (C2-1) and exposure is poorly reversible (C3-1). B, Effects are rapidly distributed globally (C2-2) and are poorly reversible (C3-2). C,Time delay between exposure and effects (C2-3). D, Effects are only manifested at the global scale (C2-4). Another variation of Panel B is conceivablethat combines C2-2 with poorly reversible exposure (C3-1), but rapid global distribution of effects independently of the global distribution ofconcentrations implies that the effect is poorly reversible.

Environmental Science & Technology Policy Analysis

dx.doi.org/10.1021/es501893m | Environ. Sci. Technol. 2014, 48, 11057−1106311059

Scenario C2-1: Concentrations Are Nearly Homogeneouson a Global Scale. The entire planet can be exposed to chemicalpollution that is rapidly distributed in the global environment. Inthis case the difference between concentrations of the chemical atthe local scale and at the global scale is small enough that an effectcan occur at the global scale without society having sufficienttime to take action to ensure the critical concentration is notexceeded at the global scale (Figure 3A).Scenario C2-2: Effects Are Rapidly Distributed Globally.

Although the pollution is largely constrained in space, onceeffects occur on the local or regional scale they are rapidlydistributed around the planet. In analogy to Scenario C2-1, theresponse time of society is longer than the time needed for theeffect to be distributed around the planet, causing a global scaleproblem (Figure 3B).Scenario C2-3: There Is a Time Delay between Exposure

and Effects. A time delay in the manifestation of effects at thelocal scale, ΔtDELAY, combined with the time delay for society torespond to the effect results in global concentrations exceedingthe critical threshold before chemical management actions canarrest the increase in environmental concentrations (Figure 3C).Scenario C2-4: Effects Are Only Observed on a Global Scale.

The adverse effect is not manifested until the chemical isdistributed at a planetary scale. Even though there is a stronggradient in chemical concentration in the environment thedisruptive effect is not manifested in more highly contaminatedlocal areas near sources until the concentrations on a planetaryscale exceed the threshold required to disrupt the vital Earthsystem process (Figure 3D).Scenarios Fulfilling Condition 3: Disruptive Effect Is

Poorly Reversible. In the reference scenario society decides totake action to reduce the concentrations of chemical pollution inthe environment and, after a response time ΔtRESP, theprevalence of the effect associated with the chemical pollutiondeclines in concert with declining concentrations. Condition 3 issatisfied when the disruptive effect is poorly reversible. There areat least two scenarios that could lead to Condition 3 beingsatisfied:Scenario C3-1: Exposure to the Chemical Pollution Is Poorly

Reversible. The emissions of the chemical cannot be readilyreduced, or the environmental contamination is not rapidlyreversible. Exposure to the chemical pollution may be poorlyreversible if society is highly dependent on a technology thatreleases the chemical, if the emissions come from long-livedproducts that cannot be rapidly removed from society, or if thechemical has high environmental persistence (Figure 3A).Scenario C3-2: The Effects of the Chemical Pollution Are

Poorly Reversible. The disruptive effect itself is persistent evenafter levels of the chemical pollution have been reduced. Effectsmay be poorly reversible if the pollution has caused a regime shift,that is, a change to a new, stable state. If a regime shift occurs,reducing pollutant levels will not mitigate the effect (Figure 3B).Chemical Profiles. Chemicals must have a certain profile to

fulfill each of the scenarios described above. The profile may bedefined by properties of the substance, or by the way thesubstance is used. Chemical profiles that fulfill each of thescenarios are summarized in Table 1, and belowwe give examplesof chemicals or combinations of chemical properties and usepatterns that match the profiles. Note that chemicals named asexamples do not necessarily represent planetary boundarythreats since at least one scenario from each of the threeconditions must be fulfilled for a chemical to pose a planetaryboundary threat. As noted in the footnote of Table 1,

transformation products must be considered when chemicalsare evaluated against some of the chemical profiles.

Profile C1-1: Unknown Disruptive Effect on a Vital EarthSystem Process. Confronting our ignorance requires us torecognize that we cannot exclude any chemicals from potentiallyfulfilling Scenario C1-1. Therefore, the most precautionaryapproach is to assume that all chemicals that are in theenvironment as a result of human activity fulfill this scenario.However, there is also a range of other possible approaches toprioritize chemicals according to their likelihood to fulfill thescenario that apply different levels of precaution. For instance,one might assign a lower priority to chemicals that have been inuse for a long period of time without having disrupted vital Earthsystem processes, provided that levels in the environment are notincreasing. Further lowering the level of precaution, low prioritymight also be assigned to chemicals that have structures that aresimilar to chemicals that have been nondisruptive, provided thattheir anticipated concentrations in the environment are nothigher.

Profile C2-1: Nearly Homogeneous Environmental Expo-sure at the Global Scale. Chemicals that are volatile andpersistent in the atmosphere, such as carbon dioxide, becomenearly well-mixed in the troposphere of the northern andsouthern hemisphere from a single emission source withinmonths, and globally in just a few years.4 However, chemicals areunlikely to be emitted from a single source. It is more likely thatemissions occur from many sources distributed on the majorcontinents. This emission pattern, combined with moderatemobility in the environment can lead to a situation in which thereare not strong gradients in contamination levels in theenvironment, and Condition 2 could be satisfied. Examples ofchemicals that fit this profile are lighter (mono- to tetrachlori-nated) PCBs and hexachlorobenzene (HCB). A combination oftheir widespread historical use and mobility in the atmospherehas led to atmospheric levels of lighter PCBs and HCB beingsimilar across the northern hemisphere.5 Levels of HCB andlighter PCBs in global background surface soils are also similarwhen normalized to soil organic content.6

Profile C2-2: Rapid Global Distribution of Effects.The natureof the disruptive effect caused by the chemical is a definingcharacteristic of the chemical profile for this scenario, and bydefinition the effect is unknown. However, fulfilling this scenariorequires that the effect be local or regional in nature, but with thepotential to become a global scale problem due to mobility. Inaddition to mobility, the effect must be self-amplifying in order toexpand its sphere of influence from the local/regional to theglobal scale. Organisms that have been modified by chemicals arecandidates for such effects. The development of antibioticresistant bacteria as a result of exposure to chemical pollution isan illustrative example of a chemical-induced effect that initiallymay have a local or regional impact but could become a globalscale problem.7,8 Current use practices for antibiotics havepromoted antibiotic resistance that is now considered a globalhealth concern.8 Since antibiotics are released to the environ-ment via wastewater (e.g., during manufacturing and followinguse), the development of antibiotic resistance in the environmentis also of concern.9 Resistance traits can be exchanged betweenbacteria, and global trade and travel allows resistant bacteria tospread rapidly,8 implying that the disruptive effects could be self-amplifying.

Profile C2-3: Time Delay between Exposure and Effects. Asfor Profile C2-2, the chemical profile for Scenario C2-3 is closelylinked to the unknown effect of the chemical on a vital Earth

Environmental Science & Technology Policy Analysis

dx.doi.org/10.1021/es501893m | Environ. Sci. Technol. 2014, 48, 11057−1106311060

system process, but the requirement that the manifestation of theeffects must be delayed compared to exposure introducesconstraints. The time delay of the effect is linked to theenvironmental transport and fate properties of the chemical sincethe chemical must transition from being a local or regionalpollutant to being a global pollutant during the delay phase. Shortdelays in the manifestation of effects (i.e., of the order of severalyears) are unlikely to be sufficient to allow a chemical that is notalready highly mobile in the environment (and thus alreadyfulfills scenario C2-1 (nearly homogeneous environmentalexposure on a global scale)) to make this transition. It is morelikely that a time delay of the order of a decade or more isrequired. Chemicals capable of making the transition fromregional-scale to global-scale pollutants on the time-scale ofdecades are persistent and semivolatile chemicals that aredistributed in the atmosphere but partition reversibly to water,soil and vegetation, and persistent water-soluble substances thatare transported in the global oceans.10 One example of delayedeffects that occur on the time scale of decades is reproductiveimpairment in long-lived organisms that are only manifested inthe second or third generation of offspring.Profile C2-4: Effects Are Only Manifested on a Global Scale.

This profile differs from profile C2-3 in that the effects occur withlittle or no time delay, but they are not manifested as anunwanted effect until they are global in scale. Thus, Profile C2-4is not dependent on the time scale for global distribution of thechemical, and time scales of a century or longer are conceivable.Chemicals that combine high persistence with some degree ofmobility in the atmosphere or oceans fit this profile. For example,perfluorinated alkyl acids (e.g., perfluorooctanesulfonic acid,PFOS and perfluorooctanoic acid, PFOA) are persistent, strongsurfactants that are becoming globally distributed in the world’soceans11 following more than 60 years of global emissions.12 Iffluorinated surfactants have an effect on a vital Earth systemprocesses, the nature of the effect is currently not known.However, it is known that organic surfactants present in theoceans can be transferred onto sea spray aerosols, significantlyincreasing the concentration of cloud condensation nuclei(CCN) and thus cloud albedo.13 An increase in cloud albedoleads to more efficient absorption of solar radiation, andcontributes to global warming. Hypothetically, if fluorinatedsurfactants contribute to increases in global CCN concen-trations, the effect would only manifest itself once fluorinatedsurfactants are distributed on a global scale.Profile C3-1: Environmental Exposure Is Poorly Reversible.

The chemical profile that fulfills this scenario can be determinedby use characteristics of the chemical, or by intrinsic properties.Environmental exposure will be poorly reversible if emissions ofthe chemical cannot be readily reduced because society isdependent on technological services that cause emissions of thechemical, or because it is released from products that have a longusage lifetime. Two examples illustrate these two variations ofpoor reversibility of emissions. The first is carbon dioxide formedby the combustion of fossil fuels. Society is highly dependent onthe combustion of fossil fuels, and this technological dependencyhas so far made reducing emissions very difficult. Polychlorinatedbiphenyls (PCBs) in sealants and other materials common inbuildings are an example of a chemical in products with a longusage lifetime. In this case the poor reversibility of PCB emissionsresults from the multitude of diffuse sources and the high cost ofcontrolling each of them.14,15

Environmental exposure is also poorly reversible for chemicalsthat have high overall environmental persistence. Examples of

chemicals that fit the profile of having poor reversibility ofenvironmental contamination due to high environmentalpersistence are the aforementioned, extremely persistentfluorinated surfactants (e.g., PFOS and PFOA) which willremain in circulation in the environment for centuries to come.

Profile C3-2: Effects Are Poorly Reversible. Effects can bepoorly reversible if they cause permanent damage, or if the vitalEarth system process responds very slowly to the change inemissions. Effects can also be poorly reversible if they result in ashift to a new stable regime. In the case of a regime shift, reducingcontaminant levels would not result in a return to the originalregime. Two well-known chemically induced cases of regimeshifts are (i) greenhouse gas-driven global warming causing lossof sea and land ice, and possibly disrupting ocean currents drivenby thermohaline circulation and (ii) eutrophication, which is theecosystem response in a water body to the addition of excessivenutrients, especially nitrates and phosphates. Freshening of thewater in the Arctic Ocean due to increased riverine inputs andmelting ice is expected to irreversibly affect ocean circulation16

and cause regime shifts in ecosystems,17 whereas eutrophicationhas been shown to irreversibly alter ecosystem structure.18 As weare, by definition, ignorant of the nature of the effects in aplanetary boundary threat context, further refining the chemicalprofile for this scenario is difficult.

Research Priorities to Identify Chemical PollutionPlanetary Boundary Threats. All of the three conditionsmust be simultaneously met for a chemical to be a planetaryboundary threat. Hence, one approach to screening for potentialplanetary boundary threats would be to evaluate chemicalsagainst the chemical profiles developed for the different scenariosthat fulfill the three conditions. If a chemical fits one (or more)chemical profile for each condition, then it would have thepotential to pose a chemical planetary boundary threat.Chemicals that fit one or more profiles but cannot beconclusively categorized as fitting a profile for each of the threeconditions would be candidates for more detailed study,monitoring, and for precautionary substitution or restrictionson use.Before we can begin prioritizing chemicals as planetary

boundary threats, criteria must be defined for the chemicalprofiles. The chemical profiles are defined in terms of the natureof the effect of the chemical on a vital Earth system process andthe nature of the exposure of the environment to the chemical.Given that the effect of the chemical is, per definition, unknown,it is more difficult to screen chemicals against the chemicalprofiles that are defined in terms of the effect than against thosedefined in terms of environmental exposure.Clearly the most challenging chemical profile to evaluate

chemicals against is Profile C1-1, since it is solely defined in termsof the unknown effect of the chemical on a vital Earth systemprocess. While one could take the most precautionary approachand classify all chemicals introduced to the environment byanthropogenic activities as fulfilling C1-1, experience indicatesthat only a very small fraction actually have disruptive effects on avital Earth system process. Hence, developing criteria forprioritizing chemicals for their potential to disrupt a vital Earthsystem process is an important challenge in the context ofchemical pollution planetary boundary threats. For instance,chemicals might be assigned a low potential of fulfilling ProfileC1-1 when society has long experience with the chemical,concentrations in the environment are not increasing, and nodisruptive effects have been observed. The prioritization mightthen be extended to substances with similar chemical structures

Environmental Science & Technology Policy Analysis

dx.doi.org/10.1021/es501893m | Environ. Sci. Technol. 2014, 48, 11057−1106311061

using read-across techniques, provided they are also notincreasing in concentration in the environment. The applicabilityof such approaches could be increased by expanding the numberof chemicals in commerce that are monitored in the environ-ment.However, such approaches are not sufficient. We must

recognize the possibility that any prioritization of chemicalsagainst Profile C1-1 could be rendered invalid by the discovery ofa currently unknown disruptive effect on a vital Earth systemprocess. The history of identification of the chemical pollution-related planetary boundaries originally presented by Rockstromet al. provides some guidance about what is needed to identifyunknown effects of chemicals on vital Earth system processes. In1985, a totally unexpected ozone hole was discovered overAntarctica,19 and the hole was eventually shown to be caused byCFC-catalyzed ozone depletion within polar stratosphericclouds.20 Another example is emissions of carbon dioxide fromfossil fuel burning, which were long thought to not pose a threat,but are now recognized to be causing climate change and oceanacidification.21 In both these cases the chemical agents causing anunknown planetary boundary threat were only identified after adisruption of a vital Earth system process was discovered. Historythus suggests that understanding the function of key Earthsystem processes andmonitoring their status are vital activities toconfront the challenge posed by chemical planetary boundarythreats. Vigilance and an improved understanding of processesvital to the Earth system will also contribute to a reduction in theresponse time of society, ΔtRESP, thereby reducing the likelihoodof scenarios C2-1, C2-2, and C2-3 occurring (Figure 3).Profiles C2-2, C2-3, C2-4, and C3-2 all include characteristics

of the unknown effect on an Earth system process. However, ineach case there are restrictions on the nature of the effect or thenature of the exposure that causes the effect that can be exploitedto prioritize chemicals. As outlined above, for Profile C2-2 (rapidglobal distribution of effects), only a subset of possible disruptiveeffects will be mobile. The unique challenge for evaluatingchemicals against C2-2 will be identifying possible scenarios formobile effects. Similarly, only a subset of disruptive effects will bepoorly reversible, and thus fulfill Profile C3-2. Above, weidentified two different effect classes: irreversible effects, or aregime shift. There may be some possibility to assess thelikelihood that a chemical will cause either irreversible effects orregime shifts, but research will be required to elucidate this. BothProfile C2-3 (time delay between exposure and effects) andProfile C2-4 (effect only manifests at global scale) require thatglobal exposure to the chemical pollution occurs over long timescales. This requirement provides a basis to prioritize chemicalsbased on a combination of persistence, mobility and geographicextent of emissions.Profiles C2-1 and C3-1 are defined only in terms of the nature

of environmental exposure to chemical pollution, and criteria andmodels can likely be developed to screen chemicals against theseprofiles. For instance, Profile C2-1 (nearly homogeneousenvironmental exposure on a global scale) can be assessed ifthere is an understanding of how the chemical will be emitted tothe environment, and of its mobility and persistence. Likewise,Profile C3-1 (poor reversibility of exposure) is related to thepersistence of the chemical in the environment, and chemical fateand transport models developed in recent years provide aconceptual basis for assessing reversibility of chemical con-tamination, provided reliable and representative data aboutdegradation rates in air, water and soils can be obtained.22,23 Thisis not to say that defining methodologies to evaluate chemicals

against C2-1 and C3-1 will be trivial. On the contrary, thedevelopment of efficient and robust procedures is certain topresent interesting research challenges. Experience with theexisting hazard assessment indicators long-range transportpotential (LRTP) and chemical persistence (P) will inform thisprocess, but they do not generally address all aspects of C2-1(e.g., the influence of spatial distribution of emissions onhomogeneous environmental exposure) or C3-1 (e.g., thereversibility of emissions) so new methodologies will berequired. Furthermore, there are difficult issues that environ-mental managers must confront. For instance, what is the speedof reversibility of chemical pollution that society would require inorder to feel confident in its ability to mitigate a planetaryboundary issue?Confronting the challenge of chemical planetary boundary

threats should not be confused with or detract from chemicalsmanagement motivated by protection of human and ecosystemhealth at the local and regional scale. There are some commonconsiderations, but the challenge posed by planetary boundarythreats from chemical pollution is unique. Although there areconsiderable uncertainties, prioritizing chemicals in commerceagainst some of the chemical profiles defined in this paperappears achievable from the perspective of our existingexperience with chemical assessment and regulation. For otherchemical profiles a more substantial research effort may berequired to make them operational. Given the high uncertaintiesassociated with evaluating chemicals against Profile C1-1,monitoring and studying the biogeochemical processes thatunderlie vital Earth system processes is probably the mosteffective strategy to identify chemicals with unknown disruptiveeffects. Chemicals that fit one or more of the profiles for C2 andC3 should receive high priority for exploratory research aboutpotential effects on vital Earth system processes, and should becandidates for precautionary substitution or phase out, especiallywhen concentrations in the environment are increasing.

■ AUTHOR INFORMATIONCorresponding Author*Phone: +46 8 674 7228; fax: +46 8 674 7638; e-mail: michael.mclachlan@itm.su.se.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was financed by faculty funding from the Departmentof Applied Environmental Science, Stockholm University,Sweden. Astrid Safron (www.astridsafron.eu) created the TOCArt for this paper.

■ REFERENCES(1) Rockstrom, J.; Steffen, W.; Noone, K.; Persson, A.; Chapin, F. S.;Lambin, E. F.; Lenton, T. M.; Scheffer, M.; Folke, C.; Schellnhuber, H.J.; et al. Planetary boundaries: Exploring the safe operating space forhumanity. Supplementary Information. Ecol. Soc. 2009, 41.(2) Rockstrom, J.; Steffen, W.; Noone, K.; Persson, A.; Chapin, F. S.;Lambin, E. F.; Lenton, T. M.; Scheffer, M.; Folke, C.; Schellnhuber, H.J.; et al. A safe operating space for humanity. Nature 2009, 461, 472−475.(3) Persson, L. M.; Breitholtz, M.; Cousins, I. T.; de Wit, C. A.;MacLeod, M.; McLachlan, M. S. Confronting unknown planetaryboundary threats from chemical pollution. Environ. Sci. Technol. 2013,47, 12619−12622.(4) Jacob, D. J. Introduction to Atmospheric Chemistry; PrincetonUniversity Press: Princeton, NJ, 1999.

Environmental Science & Technology Policy Analysis

dx.doi.org/10.1021/es501893m | Environ. Sci. Technol. 2014, 48, 11057−1106311062

(5) Pozo, K.; Harner, T.; Lee, S. C.; Wania, F.; Muir, D. C. G.; Jones, K.C. Seasonally resolved concentrations of persistent organic pollutants inthe global atmosphere from the first year of the GAPS study. Environ. Sci.Technol. 2009, 43, 796−803.(6)Meijer, S. N.; Ockenden, W. A.; Sweetman, A.; Breivik, K.; Grimalt,J. O.; Jones, K. C. Global distribution and budget of PCBs and HCB inbackground surface soils: Implications for sources and environmentalprocesses. Environ. Sci. Technol. 2003, 37, 667−672.(7) Johnson, A. P.; Woodford, N. Global spread of antibioticresistance: The example of New Delhi metallo- -lactamase (NDM)-mediated carbapenem resistance. J. Med. Microbiol. 2013, 62, 499−513.(8)WHO.WHOFact Sheet no 194, updatedMay 2013: Antimicrobialresistence, 2013.(9) Pruden, A.; Larsson, D. G. J.; Amezquita, A.; Collignon, P.; Brandt,K. K.; Graham, D.W.; Lazorchak, J. M.; Suzuki, S.; Silley, P.; Snape, J. R.;et al. Management options for reducing the release of antibiotics andantibiotic resistance genes to the environment. Environ. Health Perspect.2013, 121, 878−885.(10) Wania, F. Assessing the potential of persistent organic chemicalsfor long-range transport and accumulation in Polar regions. Environ. Sci.Technol. 2003, 37, 1344−1351.(11) Yamashita, N.; Kannan, K.; Taniyasu, S.; Horii, Y.; Petrick, G.;Gamo, T. A global survey of perfluorinated acids in oceans.Mar. Pollut.Bull. 2005, 51, 658−668.(12) Prevedouros, K.; Cousins, I. T.; Buck, R. C.; Korzeniowski, S. H.Sources, fate and transport of perfluorocarboxylates. Environ. Sci.Technol. 2005, 40, 32−44.(13)Moore, M. J. K.; Furutani, H.; Roberts, G. C.; Moffet, R. C.; Gilles,M. K.; Palenik, B.; Prather, K. A. Effect of organic compounds on cloudcondensation nuclei (CCN) activity of sea spray aerosol produced bybubble bursting. Atmos. Environ. 2011, 45, 7462−7469.(14) Diamond, M. L.; Melymuk, L.; Csiszar, S. A.; Robson, M.Estimation of PCB Stocks, Emissions, and urban fate: Will our policiesreduce concentrations and exposure? Environ. Sci. Technol. 2010, 44,2777−2783.(15) Robson, M.; Melymuk, L.; Csiszar, S. A.; Giang, A.; Diamond, M.L.; Helm, P. A. Continuing sources of PCBs: The significance of buildingsealants. Environ. Int. 2010, 36, 506−513.(16) Hatun, H. Influence of the Atlantic Subpolar gyre on thethermohaline circulation. Science 2005, 309, 1841−1844.(17) Duarte, C. M.; Agustí, S.; Wassmann, P.; Arrieta, J. M.; Alcaraz,M.; Coello, A.; Marba, N.; Hendriks, I. E.; Holding, J.; García-Zarandona, I.; et al. Tipping elements in the Arctic marine ecosystem.AMBIO 2012, 41, 44−55.(18) Carpenter, S. R.; Lathrop, R. C. Probabilistic estimate of athreshold for eutrophication. Ecosystems 2008, 11, 601−613.(19) Farman, J. C.; Gardiner, B. G.; Shanklin, J. D. Large losses of totalozone in Antarctica reveal seasonal ClOx/NOx interaction. Nature1985, 315, 207−210.(20) Solomon, S. Progrees towards a quantitative understanding ofAntarctic ozone depletion. Nature 1990, 347, 347−354.(21) IPCC. Climate Change 2013: The Physical Science Basis.Contribution of Working Group I to the Fifth Assessment Report of theIntergovernmental Panel on Climate Change; Stocker, T. F., Qin, D.,Plattner, G.-K., Tignor, M.; Allen, S. K., Boschung, J., Nauels, A., Xia, Y.,Bex, V., Midgley, P. M., Eds.; Cambridge University Press: Camridge,United Kingdom, 2013.(22) Stroebe, M.; Scheringer, M.; Hungerbuhler, K. Measures ofoverall persistence and the temporal remote state. Environ. Sci. Technol.2004, 38, 5665−5673.(23) Choi, S.-D.; Wania, F. On the reversibility of environmentalcontamination with persistent organic pollutants. Environ. Sci. Technol.2011, 45, 8834−8841.

Environmental Science & Technology Policy Analysis

dx.doi.org/10.1021/es501893m | Environ. Sci. Technol. 2014, 48, 11057−1106311063