toward the next generation of air quality monitoring: persistent organic pollutants

at SciVerse ScienceDirect

Atmospheric Environment xxx (2013) 1e8

Contents lists available

Atmospheric Environment

journal homepage: www.elsevier .com/locate/atmosenv

Toward the Next Generation of Air Quality Monitoring: PersistentOrganic Pollutants (POPs)

Hayley Hung a,*, Matthew MacLeod b, Ramon Guardans c, Martin Scheringer d,Ricardo Barra e, Tom Harner a, Gan Zhang f

aAir Quality Processes Research Section, Environment Canada, 4905 Dufferin Street, Toronto, Ontario M3H 5T4, CanadabDepartment of Applied Environmental Science, Stockholm University, Svante Arrhenius väg 8, SE-11418 Stockholm, SwedencMinistry of Agriculture, Food and Environment, Alvarez de Castro 12, Madrid 28010, Spaind Institute for Chemical and Bioengineering, ETH Zurich, Wolfgang-Pauli-Str. 10, Room HCI G 127, CH-8093 Zürich, SwitzerlandeCentro de Ciencias Ambientales EULA-Chile, Universidad de Concepcion, ChilefGuangzhou Institute of Geochemistry, Chinese Academy of Sciences, 511 Kehua Street, Wushan, Tianhe District, Guangzhou, GD 510640, China

h i g h l i g h t s

� Global air monitoring of POPs should provide consistent and comparable data.� Emission can be used to assess regional environmental performance concerning POPs.� Using air measurements, reverse-modeling can be used to estimate emissions.

a r t i c l e i n f o

Article history:Received 21 February 2013Received in revised form21 May 2013Accepted 28 May 2013

Keywords:Air monitoringPersistent Organic Pollutants (POPs)Transport and environmental fate modelsEmission inventoryEnvironmental Performance Index (EPI)

* Corresponding author. Tel.: þ1 416 739 5944; faxE-mail address: [email protected] (H. Hung).

1352-2310/$ e see front matter Crown Copyright � 2http://dx.doi.org/10.1016/j.atmosenv.2013.05.067

Please cite this article in press as: Hung, H.,Atmospheric Environment (2013), http://dx

a b s t r a c t

Persistent Organic Pollutants (POPs) are global pollutants that can migrate over long distances andbioaccumulate through food webs, posing health risks to wildlife and humans. Multilateral environ-mental agreements, such as the Stockholm Convention on POPs, were enacted to identify POPs andestablish the conditions to control their release, production and use. A Global Monitoring Plan wasinitiated under the Stockholm Convention calling for POP monitoring in air as a core medium; howeverlong temporal trends (>10 years) of atmospheric POPs are only available at a few selected sites. Spatialcoverage of air monitoring for POPs has recently significantly improved with the introduction andadvancement of passive air samplers. Here, we review the status of air monitoring and modeling ac-tivities and note major uncertainties in data comparability, deficiencies of air monitoring and modelingin urban and alpine areas, and lack of emission inventories for most POPs. A vision for an internationally-integrated strategic monitoring plan is proposed which could provide consistent and comparable moni-toring data for POPs supported and supplemented by global and regional transport models. Keyrecommendations include developing expertise in all aspects of air monitoring to ensure data compa-rability and consistency; partnering with existing air quality and meteorological networks to leveragesynergies; facilitating data sharing with international data archives; and expanding spatial coverage withpassive air samplers. Enhancing research on the stability of particle-bound chemicals is needed to assessexposure and deposition in urban areas, and to elucidate long-range transport. Conducting targetedmeasurement campaigns in specific source areas would enhance regional models which can beextrapolated to similar regions to estimate emissions. Ultimately, reverse-modeling combined with airmeasurements can be used to derive “emission” as an indicator to assess environmental performancewith respect to POPs on the country, region, or global level.

Crown Copyright � 2013 Published by Elsevier Ltd. All rights reserved.

: þ1 416 739 4281.

013 Published by Elsevier Ltd. All

et al., Toward the Next Gener.doi.org/10.1016/j.atmosenv.2

1. Introduction and objective

Persistent Organic Pollutants (POPs) are an internationalconcern due to their resistance to degradation, ability to be trans-ported over long distances from sources by air and ocean currents,and potential to bioaccumulate through terrestrial and aquatic food

rights reserved.

ation of Air Quality Monitoring: Persistent Organic Pollutants (POPs),013.05.067

mailto:[email protected]

www.sciencedirect.com/science/journal/13522310

http://www.elsevier.com/locate/atmosenv

http://dx.doi.org/10.1016/j.atmosenv.2013.05.067



H. Hung et al. / Atmospheric Environment xxx (2013) 1e82

webs to levels that may result in adverse health effects for animalsand humans. Due to these concerns, multilateral environmentalagreements (MEAs) have been enacted to control the release, pro-duction and use of POPs, including global conventions dealing withtoxics and waste (the Basel and Rotterdam Conventions), theStockholm Convention (SC) on POPs, and the POPs Protocol underthe United Nations Economic Commission for Europe (UNECE)Convention on Long-Range Transboundary Air Pollution (CLRTAP).

Air is the major route of long-range transport through theenvironment for many POPs (UNEP, 2009, 2013). In areas wherethere are no sources of POPs, atmospheric transport drivescontaminant levels. Thus, both SC and CLRTAP have providedframeworks to monitor POPs in the atmosphere. These include theGlobal Monitoring Plan (GMP) for SC (which has identified air asone of the two core media for monitoring), the European Moni-toring and Evaluation Programme (EMEP) and the Working Groupon Effects (WGE) for CLRTAP. Various national and regional moni-toring programs, and surveillance-type studies, provide valuabledata on POP concentrations in air. However, standard environ-mental indicators for POPs to interpret these concentrations and toassist in policy-making are still lacking (UNEP, 2012).

The objective of this article is to propose a systematic structurefor collecting and applying information about the atmosphericconcentrations of POPs, in order to develop and evaluate potentialindicator(s) for assessing environmental performance at curtailingPOPs at national, regional and global scales. Such indicators couldbe incorporated into the Environmental Performance Index (EPI),which ranks countries according to how close they are to estab-lished environmental policy goals (http://epi.yale.edu/).

Our approach is to first identify gaps and deficiencies in theexisting ambient air monitoring and measurements for POPs byassessing spatial coverage, temporal resolution and measurementtechniques. We then develop a vision for a strategic monitoring planthat integrates measurements andmodeling, and would provide anempirical basis to develop standard air quality indicators for bothshort- and long-term environmental performance assessments.

The development of air quality indicators for POPs presentssignificant challenges distinct from other priority air pollutants dueto several specific characteristics:

(1) Health impacts of POPs are not immediate; and usually resultfrom chronic, cumulative, and long-term exposure to one ormore substances. For example, POPs do not usually cause res-piratory health effects.

(2) POPs can partition to multiple environmental media (air, soil,water, sediment, plants, animals etc.) and the major exposureroute is non-atmospheric; e.g. through ingestion and bio-accumulation. For many POPs, the atmosphere serves as anentry-point into the environment where they are carried fromsources to receptors.

(3) Sources of POPs vary spatially and temporally depending onuse and transport. Primary emission sources may be agricul-tural [for organochlorine pesticides (OCPs)], industrial [poly-chlorinated biphenyls (PCBs), poly- and perfluorinatedcompounds (PFCs), flame retardants (FRs)], health-related ap-plications [dichlorodiphenyltrichloroethanes (DDTs) andlindane], or unintentional releases due to combustion [poly-chlorinated dibenzo-p-dioxins/furans (PCDD/Fs) and polycyclicaromatic hydrocarbons (PAHs)]. As atmospheric concentra-tions of legacy POPs decline, reservoirs that have accumulatedin oceans, land and snow/ice can become ‘secondary sources’that re-emit POPs to the atmosphere. The profile of POPs in airat any given location is unique; representing a combination oflocal and regional primary and secondary emissions coupledwith transport from more distant regions mainly through air

Please cite this article in press as: Hung, H., et al., Toward the Next GenerAtmospheric Environment (2013), http://dx.doi.org/10.1016/j.atmosenv.2

but also via water or even biovectors (e.g. migrating birds, fish,marine mammals).

(4) Toxic effects of POPs manifest themselves in multiple end-points and are not well defined for mixtures of POPs. Someendpoints may not even be recognized yet.

(5) There are sampling and analytical challenges for POPs due totheir very low ambient air concentrations [typically in pg m�3

(1 ppq) or lower], and the large number and variety of POPs.Real-time analysis of POPs is not yet feasible with existing in-strument sensitivity.

(6) New POPs are continually being added to national and inter-national control initiatives (Scheringer et al., 2012). In manycases, there are high uncertainties regarding the physico-chemical properties and environmental fate and behavior ofthese new substances. The increasing number of POPs presentsresource (including financial) and technical challenges forexisting monitoring programs.

Recently, several extensive assessments have identified the de-ficiencies in the air monitoring and measurements of POPs, andmade recommendations for improvements, including UNEP (2007,2012, 2013), UNECE (2010) Part C, and AMAP (2010). Viewpointarticles (e.g. Klánová et al., 2011) and special issues in the peer-reviewed literature have also addressed the role of science insupporting international efforts to regulate and monitor POPs [At-mospheric Pollution Research vol.3 (4), 2012 and Trends in AnalyticalChemistry vol. 46, 2013]. Here, we summarize and build on theinformation given in these documents and translate them into aform that is usable by decision-makers for developing indicatorsspecifically for POPs.

2. Overview of Current monitoring systems, modeling andindicators

2.1. Current monitoring networks and sampling techniques

Two main types of air sample collection techniques are used forPOPs: active sampling (involving pulling air through a trap with anelectric pump, e.g. high- or low-volume air sampling) and passiveair sampling (PAS) (mainly diffusion-based, trapping chemicals onsorbents without the use of electricity). Concentrations of POPs inair derived from these two techniques are not always directlycomparable; but they complement each other by providingdifferent information. Active sampling provides quantitative con-centrations of POPs in both gas and particle phases over short timeintervals (several hours to 1 week) whereas PAS provides semi-quantitative data over longer periods (typically 1e3 months up to1 year) (Harner et al., 2004; Wania et al., 2003). Recent de-velopments include a passive flowthrough sampler (FTS) thatcaptures both gas and particle phase chemicals in relatively large-volume air samples for use in remote locations (e.g. the Arcticand the Tibetan Plateau) (Xiao et al., 2007, 2012). Active sampling ismore expensive to operate, more labor intensive and requirespower, therefore it is well-suited for intensive monitoring at well-equipped stations. PAS is cheaper, easy to deploy and, thus, better-suited for remote monitoring sites or for developing a largenetwork of sites. More information can be found in UNECE (2010)and UNEP (2013).

Fig. 1 shows a map of currently-operating monitoring sitesand networks. Most networks using active sampling techniquesstarted in the 1990s (circles in Fig. 1) and those using PASs startedin the mid to late-2000s (squares in Fig. 1). Sampling sites thatwere previously in operation and may be restarted to assess con-centration changes, are given in Fig. A1. Additional network infor-mation can be found in UNECE (2010). Following the establishment


http://epi.yale.edu/

Fig. 1. Currently-operating air monitoring sites and networks (network acronyms abbreviations are given in Table A1). Symbol shapes represent sampling types (passive -, active� or both 4) and symbol colors represent program/network. Shaded area indicates the regional active sampling network of XVPCA (see Table A1) with number of sites (30)given on top.

H. Hung et al. / Atmospheric Environment xxx (2013) 1e8 3

of the GMP, new networks and stations were initiated to increasespatial resolution and to fill data gaps (Fig. 2). Better spatialcoverage is still needed in the Caucasus region of Eurasia, MiddleEast, Russia and the Pacific Islands.

2.1.1. Uncertainties and challenges in the monitoring of POPs in airMajor uncertainties in POP air measurements originate from

different sample preparation and analytical techniques used bydifferent laboratories for the same chemicals. Two large-scale in-ternational interlaboratory comparison studies (Su and Hung, 2010;Schlabach et al., 2012) assessing analytical differences among lab-oratories analyzing air samples for POPs showed good agreementsamong laboratories when analyzing standards; but much largervariability (up to a factor of two difference between laboratories)when actual air samples were analyzed, likely related to samplecleanup processes. Temporal and spatial trend data from a givenmonitoring program are likely to be internally consistent, butcaution is warranted when datasets from different monitoringprograms are compared.

Moreover, adding new chemicals to MEAs may generate sam-pling and analytical challenges, increasing the costs and technicalexpertise required to operate air monitoring programs.

2.2. Existing indicators for characterizing POPs in air

Existing indicators used to identify chemicals that are likely tohave the undesirable properties which define a chemical as a POP,


as given under the SC and CLRTAP POPs Protocol include persistence(P), bioaccumulation (B) and toxicity (T). An indicator that is espe-cially relevant to assessing the behavior of atmospheric organicmicropollutants is L, long-range transport potential. Most POPsidentified in the SC and CLRTAP POPs Protocol combine high P, Band L with at least some degree of T. It is notable that none of theexisting indicators were designed to address the measurement ofair quality directly. These indicators generally estimate the poten-tial of chemicals to affect environmental quality in regions far awayfrom sources and a long time after their use and release hasstopped.

2.3. Transport and environmental fate models

Several models that have been developed specifically todescribe the behavior of POPs in the global environment aredescribed in the review by Scheringer and Wania (2003). Themodels describe the interplay of emissions, partitioning betweenair, water and soil, degradation and transport of POPs on the globalscale. Owing to a lack of emission data and limited spatial andtemporal coverage of monitoring data, case studies that evaluatethe performance of themodels against observationswere restrictedto only a few, relatively well-studied POPs (UNECE, 2010; Beckeret al., 2011). In general, comparisons between modeled andmeasured concentrations of POPs in the atmosphere agree withinfactors of 3 for the well-studied POPs, and there is good agreementamong the different models.


Fig. 2. New networks and monitoring stations (network acronyms abbreviations are given in Table A1). Symbol shapes represent sampling types (passive - or active �) and symbolcolors represent programs/networks. Shaded areas show regional passive sampling networks with number of sites given on top.


3. Considerations on air monitoring for POPs in differentregions

Global monitoring can only be achieved by balanced planningand international cooperation. Region-specific aspects must beconsidered to produce meaningful results with limited resources.Large geographical subunits of the world called regional groups ofParties are specified in MEAs. The regional groups are countries orcountry-groups (e.g. EU) that have ratified and are thus obliged to aparticular MEA. Within a regional group, there are geographicaland social configurations relevant to describing air quality andlevels of POPs which we refer to as representative zones that arecommon across regions. A good selection of air monitoring sites inrepresentative zones specific to each regional group is essential tocharacterize the status of air pollution.

In representative zones, various sites are considered: (1) near-source zones that include (a) urban/industrial sites and (b) agri-cultural sites where POPs are used and may be produced; (2) alpineand highland zones with high-altitude sites which in some casescan be densely populated and lastly, (3) polar/remote zones withremote sites which are far from sources and have low populationdensities, making them most suitable for observing global burdenchanges and long-range atmospheric transport (LRAT).

3.1. Near-source zones

Measuring concentrations of POPs in air at sites near sources(urban, industrial or agricultural) is important for understanding:(1) the exposure routes to humans, the food webs and the envi-ronment, (2) the impact on POP concentrations due to changes in


local regulations and activities, and (3) the role of these regions as asource for LRAT.

3.1.1. Urban and industrial regional sitesRecently, concerns raised by the potential effects of chronic

exposure to POPs in air have driven an interest in monitoring inurban settings, where air concentrations are elevated (Ross andBirnbaum, 2003). However, an adequate risk assessment cannotbe performed since air concentration data of POPs within cities arelimited.

Emission inventories of POPs are still lacking for many regions.However, regulating POPs under SC (with 178 parties thus far) hasfacilitated the development of inventories for legacy POPs; espe-cially for the highly toxic PCDD/Fs, released as unintentional by-products of many industrial processes and practices, such as openburning of wastes (Estrellan and Iino, 2010), common in thedeveloping world.

Urbanization and industrialization patterns observed decadesago in developed countries are now repeated in countries witheconomies in transition and in developing countries. Rapid ur-banization in Asian and Latin American megacities with >10million inhabitants is causing significant pollution problems linkedto the increase in air-borne particle concentrations (UNEP, 2012),having implications on inhalation and deposition of particle-boundPOPs (e.g. most flame retardants, PAHs and PCDD/Fs). Many coun-tries are transitioning from regulating PM10 to PM2.5 with morestringent limit values, aiming to better protect human health. Fineparticles, which may transport further without depositing and mayprotect substances from degradation en route, play an importantrole in determining the fate of POPs in the atmosphere; yet, there



are insufficient laboratory-based studies to characterize partition-ing of POPs to fine particles. The distribution of POPs on particles ofdifferent sizes in ambient environment is virtually unknown due tomajor challenges in obtaining adequate quantifiable amounts ofPOPs on each particle size range. Gas-particle partitioning corre-lations used in models are subject to high uncertainty and cannotbe assumed to adequately describe observed ambient values at aparticular place or time due to high variability.

Significant changes in global emission patterns of POPs can beexpected in the urbanization/industrialization process. Industriesthat tend to emit POPs or their precursors have largely migratedfrom developed countries in Europe and North America to Asiancountries (including China, India, Vietnam and Thailand), Russiaand Brazil (UNEP, 2012). Polluting industries located in cities moveto rural/remote locations to accommodate high-tech and serviceindustries, with potential technological upgrades and improvedemission/waste management during the move. Development ofnew energy channels (e.g. gas pipelines and ultra-high voltagepower grids) and modern transportation systems (e.g. cross-country highways and high-speed railways) would significantlyalter the emission patterns of combustion-derived POPs.

Electronic-waste is of great concern in developing countries.Bogdal et al. (2013) reviewed recent global PAS measurements ofPOPs and found relatively high air concentrations of PCBs in Africaand South-East Asia, where PCBs have never been extensively usedor produced, which can be linked to waste from industrializedcountries. E-waste processing (dismantling, burning and acidictreatment) can release significant amounts of PBDEs, PCBs andPCDD/Fs to the environment, and pose negative effects on humanhealth on site (Chen et al., 2011).

Both the global shift of industries and e-waste could potentiallyincrease the POP emissions per unit of production, as the produc-tion is carried out and the products and e-waste are exposed athigher environmental temperatures in subtropical-tropical regions(Tian et al., 2011; Breivik et al., 2011).

3.1.2. Agricultural regional sitesAgricultural regions are important sources of current-use pes-

ticides (CUPs) and legacy pesticides, e.g. lindane and DDT, andcombustion by-products (e.g. dioxin emissions from agriculturalburning), accounting for in-situ human exposure and potentialLRAT. Pesticide sales and use patterns may be used to interpretmeasured air concentrations (Tuduri et al., 2006).

3.2. Alpine and highland zones

High-altitude sites share several features with high-latitude (orpolar) sites which influence the atmospheric fate of POPs, includinglow temperatures, snow, ice and high winds, despite differences intheir cold-trapping mechanisms (Wania and Westgate, 2008; Dalyand Wania, 2005). In addition, high-altitude sites can be denselypopulated (e.g. the Andes) and might be relatively close to sources(e.g. the Rocky Mountains in North America and the Alps inEurope).

Enhanced precipitation drives deposition along altitudinal gra-dients (Tremolada et al., 2008;Wania andWestgate, 2008); makingmountains a transient storage area for POPs and other pollutants(Bogdal et al., 2010). Distance to source is a key consideration inunderstanding high alpine regions being deposition traps for POPs.

Millions of people around the world live in highland areas withsources of POPs in close proximity (for example in Colombia,Bolivia, China, Italy and France). Due to enhanced deposition inalpine regions, highland inhabitants may be exposed to differentquantities and types of POPs as compared to those in lowlands.Unfortunately, measurements of POPs in air in highly populated


mountain areas are scarce (Liu et al., 2010; Levy et al., 2009) andthere are only a few models that explicitly address mountain areas(Daly et al., 2007; Wegmann et al., 2006).

3.3. Polar/remote zones

POP concentrations in Arctic and Antarctic air are usually verylow (in pgm�3). However, once entered into the foodweb, POPs canaccumulate to high levels in top predators which are a food sourcefor local populations. There is also concern about increasedresource exploitation and shipping emissions predicted for theArctic which may result in higher POP emission in polar regions.

Polar regions and other low-population, non-industrializedremote regions act as a sink for many POPs that were never usedthere but arrived and deposited there by LRAT.

International collaboration on monitoring POPs in remote re-gions in past decades provides a global-scale understanding ofLRAT of POPs. The experience and competence developed tomonitor POPs in remote regions, such as the Arctic, can be veryhelpful for building monitoring strategies in other regions.

3.4. Perspectives on regional air monitoring

Different monitoring capabilities exist in areas with differenteconomic situations and scientific infrastructure development,resulting in differing monitoring approaches in different regions. Aglobal overview of air pollution by POPs is only made possiblerecently by the existence of effectiveness evaluation within the SC,involving many developing countries (de Boer and Fiedler, 2013).Many sources and types of air pollution have common character-istics. Significant synergies can be found by dealing with them in anintegrated and consistent manner.

4. ‘State of the Art’ monitoring, indicators, and futureresearch needs

Here, we propose a vision for an internationally-integratedstrategic monitoring plan that includes (1) robust and long-termmonitoring linked to (2) effective and transparent data manage-ment and quality assurance/quality control (QA/QC); and (3)effective tools (including models) and procedures to synthesize,assess, and interrogate the data for answers to policy-relevantquestions. We propose a strategy to develop these three compo-nents in a balanced manner, based on existing monitoring efforts toensure effectiveness of this plan; generating consistent and com-parable data that can be applied globally, providing the foundationfor developing and validating indicators.

4.1. Air monitoring of POPs and data management

Existing atmospheric monitoring programs provide at least onesite located in each of the three zones in most regions of the world(Figs. 1 and 2); especially with the establishment of new networksand monitoring stations, particularly in East Asia and in theSouthern Hemisphere where data were previously lacking. Only afew established programs (Fig. 1) can provide time series >10 yearswhich is considerably shorter than for other air-quality-relatedpollutants. Monitoring must continue to assess long-term timetrends to demonstrate the effectiveness of control measures. TheUNEP Guidance Document for the GMP of the SC (UNEP, 2013,2007) provides technical guidance for monitoring core media(including air and human tissues) and is an important resource forregions where monitoring is deficient and for existing monitoringnetworks facing the challenge of including new POPs in theiranalytical lists.



The key to improving spatial coverage in all regions is tomaintain a sufficient and viable number of sites in each represen-tative zone, including a few remote sites, in each regional group.Improved coverage would be possible using passive air samplers.

Costing of monitoring efforts is essential for effective planningin future air monitoring. However, allocating exact costs is difficultas many steps are shared with other processes and depends onexisting infrastructures (e.g. monitoring stations and analyticallaboratories), whichmay be highly variable among regional groups.

Effective data management procedures, from data recording ininstruments to local and global QA/QC, reporting, modeling andassessment, must be accounted for during capacity building,cooperation and costing. These tasks require entirely differentcompetencies and modes of work from sampling and chemicalanalysis. UNEP with funding from the Global Environment Facility(GEF) and Strategic Approach to International Chemicals Manage-ment (SAICM) has conducted a capacity-building program to pro-vide training in sampling and analysis in developing regions (Leslieet al., 2013). Economic and technical support of threshold countrieswith financial constraints needs to continue and be extended toinclude data management to ensure the sustainability of a globalmonitoring network. The supporting systems could be managedand administrated through UNEP, GEF and SAICM.

Great efforts have been undertaken to archive and share POPmonitoring data in large-scale databases [e.g data centers for theAMAP and EMEP, EBAS atmospheric database operated by theNorwegian Institute for Air Research (NILU), a recent database inthe GMP network (http://www.pops-gmp.org/index.php)], whichcan facilitate information flows and be used as a foundation forderiving air quality indicators based on measured values.

Developing a regional framework for air monitoring of POPs thatuses existing monitoring facilities for air quality and meteorologywould leverage synergies between monitoring programs andreduce cost. Several POP air monitoring stations are co-located atGlobal Atmospheric Watch (GAW) stations (Table A2). The GMP isincorporated into the Global Earth Observation System of Systems(GEOSS) 2012e2015workplan. It is recommendable to partner withGAWand GEOSS and strive to expand such co-deployment to coverall regions. The goals are to develop baseline POP air concentrationinformation in all regions, to better understand adverse effects ofPOPs and thresholds of concern, to develop and compare long-range transport and regional models, to associate existing emis-sion inventories of other pollutants to suitable indicators for POPs(e.g. relationships between black carbon and certain POPs in air),and provide a foundation for new tools for air quality monitoring,modelling and management.

4.2. Emission as the environmental performance indicator for POPs

The most relevant indicator for the relative burden of POPs thata country or region contributes to the global atmosphere is theemission rate of POPs to the air in the region or country. Unfortu-nately, assembling emission inventories for POPs is difficultbecause the application and emission patterns of POPs are highlydiverse and poorly characterized; and because market and usagedata are very limited and sometimes undisclosed. Global-scaleemission information that can be apportioned at the countrylevel is only available for a few POPs, namely PCBs (Breivik et al.,2007) and technical hexachlorocyclohexane (HCH) (Li et al., 2000,2003; Wöhrnschimmel et al., 2012), and may involve large un-certainties (up to a factor of 10). More recently, a global griddedinventory for two congeners of pentabrominated diphenyl ether(PeBDE) (BDE 47 and BDE99) for 2005 has been compiled andevaluated with air monitoring data around the world (Li et al.,2010); however, this inventory is not yet publicly available. An


Obsolete Pesticide Inventory (up to 2006) is available from the UNFood and Agriculture Organization but the inventory is reported astotal mass per country and not compound specific.

Under the SC and other MEAs, extensive information on POPreleases and emissions to air and other media has been compiled.This repository has limitations and gaps but is a relevant bench-mark. Each party under the SC is required to complete a NationalImplementation Plan (NIPs), which includes POP inventories and,whenever possible, their estimated releases into the environmentcalculated using specific toolkits. Party reports and their reviews arepublicly available (http://chm.pops.int/Countries/NationalReports/SecondRoundofPartyReports/tabid/1315/Default.aspx).

Several cooperative processes are underway focusing onimproving the assessment of releases. These are described in theSC website under “Implementation” (http://chm.pops.int/Home/tabid/2121/mctl/ViewDetails/EventModID/7595/EventID/322/xmid/7598/Default.aspx). Although not yet standardized, effortstoward harmonization of POP release inventories would help toestablish emission as an indicator on a party/country level.

To complement and expand upon existing emissions informa-tion and include additional POPs, we suggest that a POP emissionindicator could be derived by means of chemical fate and transportmodels that are used to back-calculate POP emissions frommeasured POP concentrations in air. This approach has beenapplied at the regional scale to lindane from the Great Lakes-St.Lawrence River region in North America (Ma et al., 2003) and to arange of chemicals that were measured in Zurich, Switzerland, assummarized by Bogdal et al. (2012), and it can be transferred toother locations and expanded to other spatial scales.

We envision that POP measurements and models could be usedin combination to estimate emissions based on studies at at leasttwo spatio-temporal scales. First, campaign-based measurementsover several days to a week in agricultural regions and urban andindustrial centers of a range of countries would provide sufficientempirical basis to apply the regional-scale reverse-modelingapproach that has been pioneered in the studies in Zurich (Bogdalet al., 2012). Such studies would provide per-crop area, or per-capita emission estimates for POPs that could be extrapolated toregions with similar characteristics as the study sites.

Second, POP measurements obtained by expanding passive airsampling networks to increase spatial coverage, together withstrategically sited active air sampling sites to provide greater time-resolved concentration data, including particle-bound fractions,can be evaluated and compared against results from global-scalefate and transport models driven by emission estimates derivedfrom the regional campaigns. Where disagreements betweenmodel results and monitoring data are found, there will be evi-dence of either an incomplete understanding of the processescontrolling the fate and transport of specific POP(s), or of de-ficiencies in the emission estimates (Becker et al., 2011). Combiningmonitoring and modeling will thus provide directions for furtherinvestigation of possible unidentified sources.

Adopting emissions determined by this combined monitoringand measurement approach as an EPI with respect to POPs willrequire confronting and overcoming the following conceptualchallenges:

- Difficulties in accounting for e-waste emission of products usedin one country and POPs being emitted during disassembly anddisposal in another. In many cases the country of origin fore-wastes may not be traceable.

- Similarly, assigning responsibility for POP emissions in prod-ucts for export to another country is difficult. For instance,Asian-made furniture treated with PBDEs is exported to Africawhere buyers never requested PBDE treatment.


http://www.pops-gmp.org/index.php

http://chm.pops.int/Countries/NationalReports/SecondRoundofPartyReports/tabid/1315/Default.aspx

http://chm.pops.int/Countries/NationalReports/SecondRoundofPartyReports/tabid/1315/Default.aspx

http://chm.pops.int/Home/tabid/2121/mctl/ViewDetails/EventModID/7595/EventID/322/xmid/7598/Default.aspx




5. Recommendations and path forward for POP airmonitoring and modeling

By combining a global air monitoring network withcontinuously-improving global- and regional-scale POP models, itwill be feasible to establish modeled emission rates as an EPI forPOPs applicable at the country, region, or global level.

Key recommendations include:Investing in developing expertise in all aspects of air moni-

toring (including air sampling, chemical analysis, data manage-ment, international and regional interlaboratory comparisons, dataQA/QC and interpretation) is important in ensuring the delivery ofcomparable and consistent scientific data. Such data is needed forbuilding models, understanding global and regional transport,assessing changes in emissions over time and evaluating theeffectiveness of control measures.

Co-locating POP air monitoring at existing monitoring sta-tions for airquality andmeteorologywould reduce costs and allowfor a better understanding of the relationship and transport prop-erties of air quality co-pollutants (e.g. aerosols, ozone and POPs).

Facilitating data sharing with atmospheric pollutant dataarchives which include air monitoring data of POPs (e.g. AMAP/EMEP database), providing a foundation for model and indicatordevelopment.

Increasing spatial coverage of air measurements with passiveair samplers. Such measurements should be sustained andexpanded. Passive samplers capable of sampling particle-phasecompounds should also be explored.

Enhancing research into the stability of particle-boundchemicals is required. This is particularly relevant in urban areasof developing regions where significant increases in air-borneparticles leads to increased human exposure (via inhalation) anddeposition of POPs. The association of POPs with atmosphericparticles may also increase their persistence in air by protectingthem against atmospheric oxidants and result in enhanced atmo-spheric transport potential (Zhou et al., 2012).

Supplementing existing monitoring programs with targetedmeasurement campaigns in designated source areas that,together with appropriate regional-scale modeling, will provideemission estimates that can be extrapolated to similar regions.Emission estimates derived from these campaigns should be usedto drive global models. Results from these models can then becompared to long-term monitoring data to identify research needsand constrain emission rates and their spatial and temporal trends.

Using “Emission” as an EPI for POPs which allows countriesto be compared based on their contribution to the burden ofPOPs in the global atmosphere. Due to the deficiencies in emis-sion inventories for most POPs, we recommend that reverse-modeling be used to back-calculate emissions from measured airconcentration data. There is a need to build regional models thatappropriately reflect the geographical and social configurations ofdifferent regions to reduce uncertainty in such emission estimates.

Acknowledgment

The authors would like to acknowledge NASA, Columbia Uni-versity, and Yale University for organizing the Next Generation ofAir Quality project and workshop to develop this paper. Thanks toJulie Narayan, Environment Canada, for compiling the maps forglobal air sampling stations (Figs. 1 and 2 and A1).

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.atmosenv.2013.05.067.


References

AMAP, 2010. AMAP assessment 2009-Persistent Organic Pollutants (POPs) in theArctic. Science of the Total Environment Special Issue 408, 2851e3051.

Becker, L., Scheringer, M., Schenker, U., Hungerbühler, K., 2011. Assessment of theenvironmental persistence and long-range transport of endosulfan. Environ-mental Pollution 159, 1737e1743.

Bogdal, C., Müller, C.E., Wang, Z., Scheringer, M., Niedermann, N., MacLeod, M.,Gerecke, A.C., Schmid, P., Zennegg, M., Hungerbühler, K., 2012. Back to thesource: measurements and modeling reveal atmospheric emissions of fluori-nated, brominated, and chlorinated persistent organic pollutants. Organo-halogen Compounds 74, 359e362.

Bogdal, C., Abad, E., Abalos, M., van Bavel, B., Hagberg, J., Scheringer, M., Fiedler, H.,2013. Worldwide distribution of persistent organic pollutants in air, includingresults of air monitoring by passive air sampling in five continents. Trends inAnalytical Chemistry 46, 150e161.

Bogdal, C., Nikolic, D., Lüthi, M., Schenker, U., Scheringer, M., Hungerbühler, K., 2010.Release of legacy pollutants from melting glaciers: model evidence and con-ceptual understanding. Environmental Science and Technology 44, 4063e4069.

Breivik, K., Sweetman, A., Pacyna, J.M., Jones, K.C., 2007. Towards a global historicalemission inventory for selected PCB congeners d a mass balance approach 3.An update. Science of the Total Environment 377, 296e307.

Breivik, K., Gioia, R., Chakraborty, P., Zhang, G., Jones, K.C., 2011. Are reductions inindustrial organic contaminants emissions in rich countries achieved partly byexport of toxic wastes? Environmental Science and Technology 45, 9154e9160.

Chen, A., Dietrich, K.N., Huo, X., Ho, S., 2011. Developmental neurotoxicants in e-waste:an emerging health concern. Environmental Health Perspectives 119 (4), 431e438.

Daly, G.L., Wania, F., 2005. Organic contaminants in mountains. EnvironmentalScience and Technology 39, 385e398.

Daly, G.L., Lei, Y.D., Teixeira, C., Muir, D.C.G., Castillo, L.E., Wania, F., 2007. Accu-mulation of current-use pesticides in neotropical montane forests. Environ-mental Science and Technology 41, 1118e1123.

de Boer, J., Fiedler, H., 2013. Persistent organic pollutants. Trends in AnalyticalChemistry 46, 70e71.

Estrellan, C.R., Iino, F., 2010. Toxic emissions from open burning. Chemosphere 80,193e207.

Harner, T., Shoeib, M., Diamond, M., Stern, G., Rosenberg, B., 2004. Using passive airsamplers to assess urban-rural trends for persistent organic pollutants. 1. Pol-ychlorinated biphenyls and organochlorine pesticides. Environmental Scienceand Technology 38, 4474e4483.

Klánová, J., Diamond,M., Jones, K., Lammel, G., Lohmann, R., Pirrone, N., Scheringer,M.,Balducci, C., Bidleman, T., Bláha, K., Bláha, L., Booij, K., Bouwman, H., Breivik, K.,Eckhardt, S., Fiedler,H., Garrigues, P.,Harner, T.,Holoubek, I., Hung,H.,MacLeod,M.,Magulova, K., Mosca, S., Pistocchi, A., Simonich, S., Smedes, F., Stephanou, E.,Sweetman, A., �Sebková, K., Venier, M., Vighi, M., Vrana, B., Wania, F., Weber, R.,Weiss, P., 2011. Identifying the research and infrastructure needs for the globalassessment of hazardous chemicals ten years after establishing the StockholmConvention. Environmental Science and Technology 45, 7617e7619.

Leslie, H.A., van Bavel, B., Abad, E., de Boer, J., 2013. Towards comparable POPs dataworldwide with global monitoring data and analytical capacity building inAfrica, Central and Latin America, and the South Pacific. Trends in AnalyticalChemistry 46, 85e97.

Levy, W., Henkelmann, B., Pfister, G., Bernhöft, S., Kirchner, M., Jakobi, G., Bassan, R.,Kräuchi, N., Schramm, K.W., 2009. Long-term air monitoring of organochlorinepesticides using Semi Permeable Membrane Devices (SPMDs) in the Alps.Environmental Pollution 157 (12), 3272e3279.

Li, Y.F., Scholtz, M.T., van Heyst, B.J., 2000. Global gridded emission inventories ofalpha-hexachlorocyclohexane. Journal of Geophysical Research e Atmospheres105 (D5), 6621e6632.

Li, Y.F., Scholtz, M.T., Van Heyst, B.J., 2003. Global gridded emission inventories of b-hexachlorocyclohexane. Environmental Science and Technology 37, 3493e3498.

Li, Y.-F., Tian, C., Yang, M., Jia, H., Ma, J., Li, D., 2010. Global gridded emission in-ventories of pentabrominated diphenyl ether (PeBDE). Geophysical ResearchAbstracts 12. EGU2010e13059.

Liu, W., Chen, D., Liu, X., Zheng, X., Yang, W., Westgate, J.N., Wania, F., 2010.Transport of semivolatile organic compounds to the Tibetan Plateau: spatial andtemporal variation in air concentrations in mountainous western Sichuan,China. Environmental Science and Technology 44, 1559e1565.

Ma, J., Li, Y.-F., Poissant, L., 2003. Estimation of emissions of g-hexa-chlorocyclohexane from the Great Lakes-St. Lawrence ecosystem using acoupled efficient Kalman filter-atmospheric transport-soil model. Environ-mental Modeling and Assessment 8, 71e84.

Ross, P.S., Birnbaum, L.S., 2003. Integrated human and ecological risk assessment: acase study of persistent organic pollutants (POPs) in humans and wildlife.Human and Ecological Risk Assessment: An International Journal 9, 303e324.

Scheringer, M., Wania, F., 2003. Multimedia models for global transport and fate ofpersistent organic pollutants. In: Fiedler, H., Hutzinger, O. (Eds.), Handbook ofEnvironmental Chemistry, Persistent Organic Pollutants, vol. 3/O, pp. 237e269.

Scheringer, M., Strempel, S., Hukari, S., Ng, C.A., Blepp, M., Hungerbuhler, K., 2012.How many persistent organic pollutants should we expect? AtmosphericPollution Research 3, 383e391.

Schlabach, M., Farag-Clement, R., Hung, H., Kallenborn, R., Su, Y., Aas, W., 2012.AMAP/EMEP/NCP Inter-laboratory Study for POP Analysis 2010. EMEP/CCC-Report 7/2011. Norwegian Institute for Air Research, Kjeller, Norway, p. 133.




http://refhub.elsevier.com/S1352-2310(13)00440-8/sref1














































































































Su, Y., Hung, H., 2010. Inter-laboratory comparison study on measuring semi-volatile organic chemicals in standards and air samples. Environmental Pollu-tion 158, 3365e3371.

Tian, M., Chen, S.-J., Wang, J., Zheng, X.-B., Luo, X.-J., Mai, B.-X., 2011. Brominatedflame retardants in the atmosphere of e-waste and rural sites in southernChina: seasonal variation, temperature dependence, and gas-particle parti-tioning. Environmental Science and Technology 45, 8819e8825.

Tremolada, P., Villa, S., Bazzarin, P., Bizzotto, E., Comolli, R., Vighi, M., 2008. POPs inmountain soils from the Alps and Andes: suggestions for a ‘Precipitation Effect’on altitudinal gradients. Water Air Soil Pollution 188, 93e109.

Tuduri, L., Harner, T., Blanchard, P., Li, Y.-F., Poissant, L., Waite, D.T., Murphy, C.,Belzer, W., 2006. A review of currently used pesticides (CUPs) in Canadian airand precipitation: part 2: regional information and perspectives. AtmosphericEnvironment 40, 1579e1589.

UNEP, 2007. Guidance on the Global Monitoring Plan for Persistent Organic Pol-lutants. United Nations, Geneva, p. 145.

UNEP, 2009. Global monitoring report under the global monitoring plan for effec-tiveness evaluation. In: UNEP/POPS/COP.4/33 Conference of the Parties of theStockholm Convention on Persistent Organic Pollutants Fourth Meeting,Geneva, 4e8 May 2009, p. 20.

UNECE, 2010. Hemispheric Transport of Air Pollution (HTAP) 2010 Part C: PersistentOrganic Pollutants. In: Air Pollution Studies No. 19. United Nations, Geneva,Switzerland, p. 236. þxix.

UNEP, 2012. Chapter 6 chemicals and waste. In: GEO-5 Global EnvironmentalOutlook 5 Assessment Report. Progress Press Ltd, Malta, pp. 168e192. http://www.unep.org/geo/geo5.asp.


UNEP, 2013. Guidance on the Global Monitoring Plan for Persistent Organic Pol-lutants. UNEP/POPS/COP.6/INF/31. United Nations, Geneva. http://chm.pops.int/default.aspx.

Wania, F., Shen, L., Lei, Y.D., Teixeira, C., Muir, D.C.G., 2003. Development and cali-bration of a resin-based passive sampling system for monitoring persistentorganic pollutants in the atmosphere. Environmental Science and Technology37, 1352e1359.

Wania, F., Westgate, J.N., 2008. On the mechanism of mountain cold-trapping oforganic chemicals. Environmental Science and Technology 42, 9092e9098.

Wegmann, F., Scheringer, M., Hungerbühler, 2006. First investigations of moun-tainous cold condensation effects with the CliMoChem model. Ecotoxicologyand Environmental Safety 63 (1), 42e51.

Wöhrnschimmel, H., Tay, P., von Waldow, H., Li, Y.F., MacLeod, M., Hungerbühler, K.,2012. Comparative assessment of the global fate of ae and behexa-chlorocyclohexane before and after phaseeout. Environmental Science andTechnology 46, 2047e2054.

Xiao, H., Hung, H., Harner, T., Lei, Y.D., Johnston, G.W., Wania, F., 2007. A flow-through 415 sampler for semivolatile organic compounds in air. EnvironmentalScience and Technology 41, 250e256.

Xiao, H., Hung, H., Lao, R., Sablic, E., Sverko, E., Lei, Y.D., Wania, F., 2012. Fieldevaluation of a flow-through sampler for measuring pesticides and brominatedflame retardants in the Arctic atmosphere. Environmental Science and Tech-nology 46, 7669e7676.

Zhou, S., Lee, A.K.Y., McWhinney, R.D., Abbatt, J.P.D., 2012. Burial effects of organiccoatings on the heterogeneous reactivity of particle-borne benzo[a]pyrene(BaP) toward ozone. Journal of Physical Chemistry A 116 (26), 7050e7056.































http://www.unep.org/geo/geo5.asp

http://www.unep.org/geo/geo5.asp

http://chm.pops.int/default.aspx

http://chm.pops.int/default.aspx


































toward the next generation of air quality monitoring: persistent organic pollutants

Documents