<<

Agenzia Regionale per la Protezione dell’Ambiente della Lombardia Dipartimento di Sondrio

C. A. Belis - T. Magnani Editors C. A. Belis - T. Magnani C. A. Belis - T.

Monitoraggio dei POPs sul territorio alpino

in the Al p s Monitoring POPs in the M o nit ring POPs - - o in p al o

ri Atti della conferenza o Monitoraggio degli inquinanti organici persistenti sul territorio alpino: uno sguardo sulla situazione europea, italiana e lombarda ul territ ul s PROCEEDINGS OF THE CONFERENCE s Monitoring persistent organic pollutants in the Alps: An overview on the situation in Europe, and Lombardy dei POP dei o

ISBN 978-88-903167-0-8 raggi o nit o

9 788890 316708 > M 31 ottobre 2007 - Bormio (SO) - 31st october 2007 C. A. Belis - T. Magnani Editors

Monitoraggio dei POPs sul territorio alpino

Monitoring POPs in the Alps

Agenzia Regionale per la Protezione dell’Ambiente della Lombardia Dipartimento di Sondrio Copyright © 2007 Agenzia Regionale per la Protezione dell’Ambiente Lombardia

ISBN 978-88-903167-0-8

Finito di stampare nel mese di ottobre 2007 dalla Tipografia Polaris di Sondrio Indice GLI INQUINANTI ORGANICI PERSISTENTI IN UNA PROSPETTIVA NAZIONALE ED EUROPEA Le Alpi e i progetti transnazionali 7

Environmental protection and transnational cooperation in the alpine space: results and perspectives 11

Present and future in the risk assessment of chemicals with PBT properties 13

I RISULTATI DEL PROGETTO MONARPOP Synthesis of the findings of the project MONARPOP 21

Idrocarburi policiclici aromatici negli ecosistemi alpini (progetto MONARPOP) 27

Studio delle concentrazioni di POPs nell’area dolomitica 33

Polybrominated diphenyl ether (pbde) in humus layers in remote forests 39

Distribution of dioxins, polychlorinated biphenyls and chlorinated pesticides in the alpine environment 45

The deposition of pesticides into Alpine forest ecosystems 51

Novel Diagnostic Tools for POPs in Alpine Areas 55

Novel ambient air sampling techniques adapted for the needs of project MONARPOP 59

I POPs NEL TERRIORIO ALPINO LOMBARDO Emissioni atmosferiche di diossine in Lombardia: l’inventario delle emissioni 65

Emissioni atmosferiche di diossine in Lombardia: Il caso di un incendio di un deposito di rifiuti provenienti dalla raccolta differenziata di materie plastiche 73

The emissions of POPs in the agriculture sector: trends and current state 79

The effects of POPs on Alpine organisms and ecosystems 87

Evoluzione temporale e andamenti stagionali di pop in deposizioni d’alta quota in Lombardia e Svizzera 93

Le possibili ricadute sulla salute della presenza ambientale dei POPs 99

GLI INQUINANTI ORGANICI PERSISTENTI IN UNA PROSPETTIVA NAZIONALE ED EUROPEA

THE PERSISTENT ORGANIC POLLUTANTS FROM THE EUROPEAN AND THE NATIONAL PERSPECTIVES

Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 

Le Alpi e i progetti transnazionali

T.Magnani, C.Belis Agenzia Regionale per la Protezione dell’Ambiente Lombardia Dip. Sondrio via Stelvio, 35- 23100 Sondrio

In seguito al veloce progredire della globalizzazione, le Alpi sono state caratterizzate da profondi cambiamenti strutturali, culturali ed ecologici. Il concetto di sviluppo sostenibile e la considerazione della dimensionalità territoriale diventano sempre più importanti per la definizione di strategie politiche in grado di gestire questi cambiamenti. Oggi, le Alpi sono probabilmente la macroregione europea con i contatti interculturali più intensi in relazione allo spazio occupato ed esigono il confezionamento di una soluzione unica per lo sviluppo sostenibile di un territorio che ricopre una superficie di 190.000 Kmq in cui vive una popolazione di 13,6 milioni di abitanti. Rafforzare la cooperazione e contribuire allo sviluppo di una identità e capacità comune di agire nello spazio alpino è stato il primo obiettivo della Convenzione delle Alpi, un accordo di diritto internazionale, stipulato già nel 1991, fra Austria, Svizzera, Germania, Francia, Principato di , Liechtenstein, Italia, Slovenia e Comunità Europea. Tale Convenzione è stata un indubbio precursore dello sviluppo inclusivo e sostenibile dell’arco alpino con il suo approccio integrato includente la dimensione ecologica, economica e sociale per le aree di montagna e valutando attentamente l’interdipen- denza delle differenti dimensioni dello sviluppo. Per individuare soluzioni innovative e sostenibili, data la dimensione transfrontaliera o internazionale, la Convenzione delle Alpi ha posto in primo piano il tema del ricorso alla cooperazione transnazionale come stimolo e sostegno di strategie atte allo sviluppo strutturale dello Spazio Alpino nella nascente Europa unita. In seguito, il tentativo di coinvolgere il territorio affinché esprimesse la propria voca- zione ad inserirsi politicamente, economicamente e socialmente nel contesto europeo ha generato politiche volte a ridurre le criticità derivanti dalla cessazione della condi- zione frontaliera (e delle attività ad esse connessa) e dal perdurare di molti elementi di conflitto (linguistici, infrastrutturali, normativi, etc.). Per far si che il coinvolgimento degli attori locali non rimanesse limitato ai pochi chilometri di confine, dalla seconda metà degli anni Novanta il programma Interreg ha delineato le grandi aree di coo- perazione trasnazionale grazie alle quali il territorio diventa una nuova dimensione della politica europea. Le Regioni si ritrovano in un’area che non coincide più con i confini nazionali, ma che ricerca nel rapporto con le altre istituzioni e nella integra- zione territoriale il proprio punto di equilibrio. Il programma transnazionale “Spazio Alpino” interessa un’area geografica che com- prende circa 70 milioni di abitanti distribuiti in un territorio caratterizzato dalla più grande catena montuosa del continente europeo e dai circostanti territori collinari e  Monitoraggio dei POPs sul territorio alpino di pianura; tale spazio alterna zone di montagna, scarsamente popolate e caratteriz- zate da un ricco patrimonio naturale e culturale, ad insediamenti urbani e metropoli densamente popolate. Le zone ammissibili interessano sette regioni italiane: Valle d’Aosta, Piemonte, Li- guria, Lombardia, Friuli-Venezia-Giulia, Veneto e Trentino-Alto Adige. Sono inoltre coinvolti la Francia (Rhône-Alpes, Provence-Alpes-Côte d’Azur, Franche-Comté, Alsace), la Germania (Oberbayern e Schwaben, Tubingen e Freiburg im Breisgau), la Slovenia e l’Austria (gli interi paesi). Partecipano inoltre al programma, come membri a pieno titolo, due paesi extra-comunitari: la Svizzera e il Liechtenstein. In sintesi, la strategia del programma si articola in quattro assi prioritari. • la priorità 1 “Promozione dello Spazio alpino quale spazio competitivo e vitale nel quadro di uno sviluppo territoriale policentrico nell’Unione europea” che comprende due misure: la conoscenza reciproca e le prospettive comuni; la compe- titività e lo sviluppo sostenibile; • la priorità 2 “Sviluppo di sistemi di trasporto sostenibile, con particolare riguardo all’efficienza, all’intermodalità e al miglioramento dell’accessibilità” che com- prende due misure: le prospettive e l’ analisi; il miglioramento dell’esistente e la promozione di sistemi di trasporto futuri attraverso soluzioni intelligenti di grande e piccola scala come l’intermodalità; • la priorità 3 “Saggia gestione della natura, del paesaggio e del patrimonio cul- turale, promozione dell’ambiente e prevenzione di rischi naturali” che com- prende tre misure: la natura e le risorse, in particolare l’ acqua; la buona gestione e la promozione dei paesaggi e del patrimonio culturale; la cooperazione nel settore dei rischi naturali; • la priorità 4 “Assistenza tecnica” che comprende due misure: l’ attuazione del programma; l’ informazione e la valutazione. Sono moltissimi oggi i progetti realizzati e quelli in fase di realizzazione sullo Spazio Alpino, ma, al di là degli importanti risultati ottenuti, occorre ricordare che l’obiettivo generale del programma è quello di aumentare la competitività e l’attrattività dell’area di cooperazione attraverso lo sviluppo di azioni comuni in settori dove la cooperazione transnazionale è necessaria per arrivare a soluzioni sostenibili. Il Programma tende inoltre ad implementare metodologie per incubare e sviluppare strumenti che influenzino anche i comportamenti e le percezioni sulla necessità, nell’Europa unita, di delineare accordi in un processo lavorativo che promuova e massimizzi la possibilità di un apprendimento comune e condiviso tra i vari partners dei progetti e che apporti un valore aggiunto alla integrazione tra i Paesi. Il significato etimologico della Cooperazione Trasnazionale è di difficile espressio- ne nella sua essenza ed significato . La parola rimanda sicuramente ad una attività collaborativa tra i diversi Paesi ed è atta a descrivere scambi di prassi, di attività, di prodotti, tra i diversi Paesi all’interno della Comunità Europea, ma sottende anche Le Alpi e i progetti transnazionali  a scambi più preziosi quali il dialogo, i know-out, l’integrazione culturale e sociale fondamentale per lo sviluppo di una comune civiltà europea. Il progetto transnazionale “Monitoraggio degli inquinanti organici persistenti (POP’s) ed altri inquinanti nella Regione Alpina”, ad acronimo Monarpop, ha concentrato l’attenzione sulle foreste delle Alpi ritenute un importante recettore per i POP’s trasportanti anche da aree distanti, tematica sulla quale non esistevano finora studi scientifici che, in ogni modo, quantificassero l’entità del problema. Nato da una cooperazione tra Austria, Italia, Germania, Slovenia e Svizzera, il Monarpop, i cui risultati vengono presentati nel presente Convegno, ha coinvolto istituzioni locali, regionali e nazionali, a cominciare dal 2004, con l’obiettivo, tra gli altri, di affidare ai decisori politici le conclusioni tecniche al fine di operare soluzioni strategiche in ottemperanza alla Convenzione sui POP’s. Un ultimo, ma non meno importante, obiettivo del progetto è l’informazione e la comunicazione. La strategia di comunicazione ha previsto una gamma differenziata di elementi (por- tale Internet) e strumenti (banca dati di progetto) per uno scambio sistematico di esperienze e per promuovere non solo le reti telematiche, ma anche, e soprattutto, lo scambio interdisciplinare.

Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 11

Environmental protection and transnational cooperation in the alpine space: results and perspectives

Ivan Curzolo INTERREG IIIB Alpine Space JTS Stadt Rosenheim - Postfach 1209 - D - 83013 Rosenheim - Germany

Abstract The Alpine Space area represents one of the most manifold regions in the heart of the European Union. In fact, this area is a major contact zone, were different European cultures and languages (German, Latin and Slavic) meet. This diversity is also vis- ible in respect with political and socio-economical aspects, as considerable social and economic disparities, often within small distances, are evident. The Alpine Space represents also an important transnational European co-operation area. Its co-operation characteristic has integrated the Alpine tradition for over several decades. It extends from political and administrational co-operations on national or regional level, to co-operations on smaller scale between local authorities and private institutions, some of them developing within the frame of European Community Initiatives and programmes (like INTERREG). Monarpop is one of the 58 operations co-financed by the European Development Fund (ERDF) through the INTERREG IIIB Alpine Space Programme 2000 – 2006. These projects that have developed and are developing their activities according to three thematic priorities: • spatial planning and competitiveness; • fostering the development of sustainable transport systems; • wise management of natural landscapes and cultural heritage as well as promotion of the environment and prevention of natural disasters. The aim of the project Monarpop, is to carry out a survey on persistent organic pol- lutants (POPs) and other pollutants in the Alpine region, as POPs, which belong to the most toxic existing compounds and accumulate in the food webs and in environmental media, endanger the Alpine ecosystems and by that the sustainable development of the Alpine Space. Although the activities started already in 2003, the highly experimental features of the project phases, together with the use of very sensitive and innovative technology which was scarcely tested before, had led the need to extend the time implementa- tion of the project. Results are therefore firstly available and shown in these three international conferences of Venice, Bormio and Salzburg. While the current programming period is fading away with the implementation of 12 Monitoraggio dei POPs sul territorio alpino the approved operations, the new programming period is already a reality on the horizon. The Alpine Space Programme 2007 – 2013 will aim at increasing competitiveness and attractiveness of the cooperation area by developing joint actions in fields where transnational cooperation is required for sustainable solutions. Five EU Countries (Austria, , Germany, Italy and Slovenia) together with the two partner Countries (Liechtenstein and ) have joined again forces to draw up a document that aims at being a synthesis between programmes in the field of regional competitive- ness and employment and programmes fostering cross-border and interregional co- operation. The Operational Programme has been formally approved by the European Commission in September 2007. Programme priorities focus on different fields of action. Beside the promotion of competitiveness and attractiveness, mentioned in the beginning, particular attention will be given to accessibility and connectivity as well as the promotion of natural resources and cultural heritage. Although rather similar in the budget dimension, the new programme will support a different approach when approving projects for co-funding. Activities will have to cover the entire alpine area, be characterised by concrete and tangible results that shall be used also after the end of the public funds support. The involvement of policy makers and political sector shall be granted as well. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 13

Present and future in the risk assessment of chemicals with PBT properties

Ana B. Payá Pérez European Chemicals Bureau, Institute for Health and Consumer Protection, Joint Research Centre, Ispra, Italy 21024 E-mail:[email protected]

Riassunto Questo articolo illustrerà lo stato attuale della valutazione di 127 sostanze pro- dotte in elevate quantità (HPVCs), biocidi e pesticidi che sono state condotte dal Comitato Tecnico della UE sulle sostanze nuove ed esistenti sottogruppo per la valutazione del potenziale PBTs, o vPvBs, o POPs nell’ambito della strategia tem- poranea per la gestione delle sostanze vPvBs e PBTs (Dir. 93/67, Reg. 1488/94, Dir. 98/8/EC e Dir. 91/414/ECC). Verranno inoltre presentati i principali elementi delle linee guida sulla valutazione delle sostanze PBT/vPvB sotto la normativa REACH.

Abstract The paper will illustrate the current status of the evaluation of 127 high production volume substances (HPVCs), biocides and pesticides which has been carried out by the EU Technical Committee on New and Existing Substances subgroup for the evaluation of potential PBTs or vPvBs or POPs in the context of the EC’s Interim Strategy for Management of PBT and vPvB substances (Directive 93/67, Regulation 1488/94, Directive 98/8/EC and Directive 91/414/ECC). The main elements of the guidance developed on how to assess PBT/vPvB substances under REACH Regula- tion will be presented.

Introduction Substances fulfilling PBT criteria are substances that are persistent, bio-accumulative and toxic, while vPvB substances are characterised by a particularly high persistency in combination with a very high tendency to bio-accumulate, but not necessarily proven toxicity. The persistence and bioaccumulation criteria for POPs and vPvB substances are identical. The identification of substances with PBT properties in the EU Risk Assessment process started on year 2003 under the existing substances Regulation 793/93. How these properties are defined by the criteria laid down in Annex XIII of the REACH Regulation (Registration, Evaluation and Authorization of Chemicals) will be described.

 HPVCs are substances or substance mixtures produced/imported in the EU by a particular producer/im- porter in amounts exceeding 1000 t/yr at least once between 1990 and 1994.  OJ L 396, 30.12.2006 p.1; Directive 2006/121/EC and Regulation (EC) No 1907/2006. 14 Monitoraggio dei POPs sul territorio alpino

The European Commission and the Member States are committed to identify and as- sess further POP substances to be included under the international agreements in order to contribute to achieving the 2020 chemicals target adopted at the World Summit on Sustainable Development. The procedure of implementation of a new chemical into the POPs list is determined in the Stockholm Convention. The proposing Party shall provide a statement of the reasons for concern of including, where possible, a comparison of toxicity or ecotoxicity data with detected or predicted levels of a chemical, resulting or anticipated from its long-range environmental transport, and a short statement indicating the need for global control. A very significant argument for implementation of a substance to the POPS list is the need of the global control. For the inclusion of the chemical into the POPs group the most important is the poten- tial risk for human and environment. One of the main sources of information that have triggered action on emerging POPs is the Technical Committee on New and Existing Subtances (TCNES) subgroup on PBTs which currently addresses all pesticides, bio- cides, new chemicals and high production volume chemicals (HPVCs). An update of the conclusions of finalized assessment on various PBTs will be provided.

Background on EU PBT Assessment Process Directive 93/67, Regulation 1488/94 and Directive 98/8 require that an environmen- tal risk assessment be carried out on notified new substances, on priority existing substances and active substances and substances of concern in a biocidal product, respectively. The scope of the Risk Assessment covers emissions and consequent environmental impact and human exposures at each stage of the life-cycle of a priority substance from production, through processing, formulation, use and disposal. Protection goals are established for the environmental compartments - atmosphere, aquatic organisms, sediment and soil dwelling organisms, microorganisms in WWTPs, and mammals and birds exposed via accumulation in the food chain. Exposure of humans from all relevant sources is considered including exposure from consumer products and through air, food, and drinking water and the workplace. Each exposure scenario is assessed individually, and where appropriate, an overall combined exposure is also estimated. The objective of the PBT/vPvB assessment is different from the other risk assessment approaches as it seeks not to estimate safe concentrations but to minimize releases of PBT and vPvB substances and thereby exposure of humans and the environment to substances that cannot be controlled with reasonable effort once they have been released. The PBT assessment was developed to respond to the additional concerns for the potential impact of individual substances on marine ecosystems which may not be adequately addressed by the methodologies used for the inland environmental risk assessment. Present and future in the risk assessment of chemicals with PBT properties 15

These concerns are: a. the concern that hazardous substances may accumulate in parts of the marine environment and that: (i) the effects of such accumulation are unpredictable in the long-term; (ii) that such accumulation would be practically difficult to reverse; b. the concern that remote areas of the oceans should remain untouched by hazardous substances resulting from human activity, and that the intrinsic value of pristine environments should be protected. The detailed methodology is described in the Technical Guidance Document on Risk Assessment for New and Existing Substances (TGD, 2003). The criteria of PBTs and vPvBs are defined in the TGD-2003 as given in table 1.

Table 1.- PBT or vPvB criteria according to the TGD-2003 for implementing Directive 93/67, Regula- tion 1488/94 and Directive 98/8. Property PBT-criteria vPvB-criteria P Half-life > 60 d in marine water or > 40 d in Half-life > 60 d in marine- or freshwater* or half-life > 180 d in marine freshwater or >180 d in marine or sediment or > 120 d in freshwater sediment* freshwater sediment B BCF > 2000 BCF > 5000 T Chronic NOEC < 0.01 mg/l or CMR or Not applicable endocrine disrupting effects * For the purpose of marine environmental risk assessment half-life data in freshwater and freshwater sediment can be overruled by data obtained under marine conditions.

Under REACH a PBT/vPvB assessment is required for all substances manufactured or imported in the EU in amounts of 10 or more tones per year that are not exempted from the registration requirement. An additional guidance has been developed in the REACH Implementation Project (RIP) 3.2-2 Guidance on PBT and vBvP assessment (RIP 3.2-2, 2007) following the Annex XIII of Regulation (EC) No 1907/2006. In practice, the PBT and vPvB assessment comprises of 1 or 3 steps (comparison with the criteria and, if these are met, emission characterization and risk characterization) that are outlined in a guidance document developed as part of the REACH Implementation Projects (RIPs). This guidance for PBT/vPvB assessment shall ensure that all existing information as well as information that can be developed with alternative methods (in vitro, QSAR and read-across) is taken into account in the most appropriate way in order to save costs and reduce the need for conducting new animal experiments.

Results and Discussion A screening-study among the 2682 high production volume chemicals (HPVC) in the IUCLID database (ECB 4/14/02) was performed to identify potential persistent, 16 Monitoraggio dei POPs sul territorio alpino bioaccumulating and toxic substances (PBTs) and very persistent and very bioaccu- mulating substances (vPvBs). A total of 127 substances were selected for evaluation. At present 5 substances (Table 2) have been identified as POPs, 21 substances are fulfilling PBT criteria and 64 have been de-listed, 8 substances have been deferred and 25 are undergoing evaluation. The PBT list and Summary Assessment Reports are published on the European Commission’s European Chemicals Bureau (ECB) Website: http://ecb.jrc.it/esis/index.php?PGM=pbt

Table 2. Substances identified as POPs by the TC NES subgroup for PBT assessment Substance Name CAS Number EC Number Molecular Conclusions formula Hexachlorobuta- 87-68-3 201-765-5 C4Cl6 P/vP; vB and T 1,3-diene very high potential for long- range atmospheric transport. Lindane 58-89-9 200-401-2 C6H6Cl6 POP Hexachlorobenzene 118-74-1 204-273-9 C6Cl6 POP Endosulfan 115-29-7 204-079-4 C9H6Cl6O3S Fulfilling BPT criteria & POP Clofenotane 50-29-3 200-024-3 C14H9Cl5 POP

Conclusions The PBT list and Summary Assessment Reports are published on the European Com- mission’s European Chemicals Bureau (ECB) Website: http://ecb.jrc.it/esis/index. php?PGM=pbt Under REACH the identification of a substance as being PBT/vPvB according to the criteria in Annex XIII or one that other data suggest an equivalent level of concern does not automatically lead to a proposal for inclusion in Annex XIV and the sub- sequent requirement for authorisation. Detailed guidance is provided in RIP 3.2-2 Guidance. In future new POPs will be added to the list reported by The Council of the EU (SEC (2007/341) which contains a total of 12 substances that have been so far proposed to be added in the Stockholm Convention and/or the UNECE Protocol on POPs. These substances have been divided in two groups. First group are substances that are or have been widely used in EU: Commercial pentabromodiphenyl ether (c-pentaBDE; CAS N. 32534-81-9); commercial octa- bromodiphenyl ether (c-octaBDE; CAS n. 32536-52-0); perfluorooctane sulfonate (PFOS), short-chain chlorinated parafins (SCCPs; CAS N. 85535-84-8) andα -lindane (CAS N. 319-84-6) and β-lindane (CAS N. 319-85-7). The second group are substances chlordecone (CAS N. 143-50-0), hexabromodiphe- nyl ether (hexaBDE; CAS N. 36483-60-0); pentachlorobenzene (CAS N. 608-93-5); Present and future in the risk assessment of chemicals with PBT properties 17 hexachlorobutadiene (CAS N. 87-68-3) and polychlorinated naphtalenes. Additional substances i.e. endosulfan (CAS N. 115-29-7), dicofol (CAS N. 115-32-2), trifluralin (CAS N. 1582-09-8), pentachlorophenol (CAS N. 87-86-5) that may exhibit POP characteristics are currently under discussion in the EU TCNES subgroup on PBTs.

Bibliography ECB 4/14/02. Identification of potential PBTs or vPvBs among the IUCLID high production volume chemicals. ECB internal report, 2002. RIP 3.2-2 Guidance for PBT and vPvB Assessment (In prep. 2007). ECB report http://ecb. jrc.it/REACH SEC (2007) 341. Commission Staff Working Document. Community Implementation Plan for the Stockholm Convention on Persistent Organic Pollutants. Brussels, 14 March 2007. TGD (2003). De Bruijn, J., Hansen, B.G., Johansson, S., Luotamo, M., Munn, S.J., Musset, C., Olsen, S.I., Olsson, H., Paya-Perez, A.B., Pedersen, F., Rasmussen, K., Sokull-Kluttgen, B. - Technical Guidance Document in support of Commission Directive 93/67/EEC on Risk Assessment for new notified substances, Commission Regulation (EC) No 1488/94 on Risk Assessment for existing substances and Directive 98/8/EC of the European Parlia- ment and of the Council concerning the placing of biocidal products on the market. Part 1 and Part 2 - EUR 20418/EN/1+2 (2003). Munn S., Allanou R., Aschberger K., Cosgrove O., Pakalin S., Paya Perez A., B. Schwarz- Schulz, G. Pellegrini, S. Vegro - European Union Risk Assessment Report - 2,2’6,6- tetrabromo-4,4’-isopropylidene diphenol (tetrabromobisphenol-A) - EUR 22161 EN, PB/2006/IHCP/3067 Hansen, BG, A van Haelst, K van Leeuwen, P van der Zandt: Priority setting for existing chemi- cals: The European Union risk ranking method (EURAM). Document ECB 4/02/98 Tyle, H, HS Larsen, EB Wedebye, D Sijm, T Pedersen Krog, J Niemelä: Identification of potential PBTs and vPvBs by use of QSARs. Document SHC/TS 2-3/029, Copenhagen 2002-06-27, DK EPA

I RISULTATI DEL PROGETTO MONARPOP

THE RESULTS OF THE PROJECT MONARPOP

Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 21

Synthesis of the findings of the project MONARPOP

Peter Weiss1, Rodolfo Bassan2, Claudio Belis3, Saverio Iozza4, Gert Jakobi5, Manfred Kirchner5, Wilhelm Knoth6, Norbert Kräuchi7, Walkiria Levy‑Lopez5, Wolfgang Moche1, Ivo Offenthaler1, Stefano Raccanelli8, Karl‑Werner Schramm5, Isabella Sedivy7, Primož Simoncˇicˇ9, Maria Uhl1

1 Austrian Federal Environment Agency, Spittelauer Lände 5, 1090 Vienna, Austria, 2 ARPA Veneto: Regional Agency for Environmental Protection of Veneto, Via F. Tomea 5, I-32100 Belluno, Italy, 3 ARPA Lombardia: Regional Agency for Environmental Prevention and Protection of Lom- bardia, Via Stelvio 35, I-23100 Sondrio, Italy, 4 Organical Analytical Chemistry, Basel University, St. Johanns-Ring 19, CH-4056 Basel, Switzerland, 5 GSF, National Research Center for Environment and Health, Ingolstädter Landstraße 1, D-85764 Neuherberg, Germany, 6 UBA-De: German Federal Environment Agency, Wörlitzer Platz 1, 06844 Dessau- Roßlau, Germany, 7 WSL, Swiss Federal Institute for Forest, Snow and Landscape Research, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland, 8 Consorzio I.N.C.A. c/o VEGA Edificio CygnusVia delle Industrie 21/8 30175 Margh- era (VE)Italy 9 Slovenian Forest Institute, Vecna pot 2, 1000 Ljubljana, Slovenia

Riassunto Il progetto MONARPOP ha studiato gli inquinanti alpini quali POPs e altri componenti Organici, tenendo conto delle variazioni regionali ed altitudinali. Cinque nazioni (Austria, Germania, Italia, Svizzera e Slovenia) e 12 partners hanno partecipato a questa rete di monitoraggio che ha ottenuto i fondi dal pro- gramma Alpine Space dell’Unione europea. Sono stati analizzati i campioni di piante, terreno e aria ottenuti da una rete di postazioni situate nelle foreste e 7 profili altitudinali. I campionatori d’aria e di deposizioni atmosferiche associati a sistemi di monitoraggio delle traiettorie delle masse d’aria posizionati nelle cime di tre montagne hanno aiutato a quantificare il trasporto a lungo raggio verso le Alpi e a mappare le regioni sorgente. L’associazione tra la carica inquinante e l’attività detossificante è stata studiata in vitro e nei campioni di piante. Molte componenti hanno mostrato concentrazioni maggiori nelle regioni ai con- fine delle Alpi che nelle parti centrali più riparate. Le concentrazioni variano con l’altitudine. In modo particolare componenti da sorgenti lontane (alcuni pesticidi, PCBs) tendono ad avere concentrazioni maggiori nelle postazioni più alte, mentre in altri (es. PAHs) le concentrazioni maggiori si osservano a quote 22 Monitoraggio dei POPs sul territorio alpino

più basse e sembrano essere influenzate localmente. Alcune componenti (es. PBDE) mostrano il picco alle medie altitudini, suggerendo un un’influenza dello strato rimescolato. Le concentrazioni in aria e le deposizioni sono comparabili a quelli di siti rurali meno remoti, indicando il lungo trasporto di queste componenti al di sopra dello strato rimescolato della troposfera.

Abstract The project MONARPOP (Monitoring Network of Persistant and other Organic Pol- lutants) investigated the alpine pollution with POPs and other organic compounds, including regional and altitudinal variations. Five Countries (Austria, Germany, Italy, Switzerland and Slovenia) and 12 partners participated within this Monitoring Net- work which obtained funds from the Alpine Space programme of the European Union. Plant, soil, and passive air samples from a network of remote forest sites and seven altitude profiles were analysed. Deposition and trajectory related active air sampling on three mountain tops helped to quantify long-range transport to the Alps and map it to source regions. The association between pollutant loads and detoxification activities was investigated in vitro and in plant samples. Several compounds showed higher concentrations in the border regions of the Alps than in the more shielded central parts. The concentrations varied with altitude. Par- ticularly compounds from faraway sources (certain pesticides, PCBs) tended to have higher concentrations at the highest sites, while higher concentrations down in the valleys (e.g. PAHs) seemed to be locally influenced. Some compounds (e.g. PBDE) peaked at mediate altitudes, suggesting an influence of the mixing layer. Concen- trations in air and deposition were comparable to those of less remote rural sites, indicating the long-range transport of these compounds above the mixing layer of the troposphere.

Introduction The Alps are a geographical and meteorological trap for atmospheric pollutants. Bar- rier effects, high precipitation and low ambient temperature further the deposition of airborne compounds. Pilot studies yielded evidence that this might also be the case for semivolatile organic compounds. Pollution at high altitudes adds another burden to ecosystems already subject to various natural and man-made stressors. It poses a potential threat for drinking water resources, food production and ecosystem health in the affected areas. To assess the current situation, five states within the region of the Alps (Austria, Germany, Italy, Switzerland, and Slovenia) have launched the project MONARPOP (Monitoring Network in the Alpine Region for Persistent and other Organic Pollutants) with support from the European Union. Synthesis of the findings of the project MONARPOP 23

Material and Methods Pollutants have been measured in atmosphere, deposition, spruce needles and soil from forest sites. Spruce needles have been chosen as a particularly efficient recep- tor for organic compounds due to their highly absorbing lipophilic cuticles. Forest soils, especially the humus layer, are rich in organic matter and represent a matrix in which the deposited compounds can accumulate over long periods of time. In addi- tion, passive air samplers (SPMD, semipermeable membrane devices) were exposed. As active air samplers Low Volume and High Volume samplers trapped pollutants depending on their volatility. Selected Parameters: MONARPOP’s analytical focus rested on: Polychlorinated dibenzodioxins and -furans (PCDD, PCDF), polychlorinated biphe- nyls (“Ballschmiter” and dioxin-like PCB), 5 hexachlorocyclohexanes (incl. Lindane), hexachlorobenzene, DDT and metabolites, selected other chloropesticides (Aldrin, Chlordan, Dieldrin, Endrin, Heptachlor, Mirex, Pentachlorophenol, Toxaphen), Poly- brominated diphenylethers (PBDEs), chlorinated paraffins, polycyclic aromatic hydro- carbons (PAH), Nitrophenols, Chloroform, Tetrachloromethane, 1,1,1-Trichloroethane, Trichloroethylene, Perchloroethylene, Chlorinated Paraffins and Trichloroacetic acid. Selected Sites Throughout the largest part of the alpine region, 40 sites had been selected in five countries [Figure 1]. In total, 72 needle samples, 53 humus layer samples and 25 mineral soil layer samples were analysed for the selected parameters.

Figure 1: Monarpop sampling sites 24 Monitoraggio dei POPs sul territorio alpino

Altitude profiles The vertical component of alpine pollution was examined with the help of seven local altitude profiles. These profiles combined several sampling sites of increasing elevation. Bioindication Possible biological effects of the accumulated POPs were assessed by measuring detoxifying enzyme activities in Spruce needles and by observing the influence of needle and humus extracts on cell cultures (bioassays). Air and deposition sampling: Active air samplers at three stations beyond the timberline on top of three altitude profiles, (Austria: Sonnblick [Figure 2], Germany: Zugspitze, Switzerland: Weiß- fluhjoch) screened the air for the analysed pollutants An array of filters was used to distinguish between pollutant imports from different regions. The Austrian Central Institute for Meteorology and Geodynamics determined the travel routes (trajectories) of the incoming air masses. Depending on its origin, the air was loaded onto a cor- responding filter. Low volume samplers were used to detect PAHs and halogenated organopesticides, whereas high volume samplers were installed to detect PCDD/F, PCB and PBDE. For the detection of VHC charcoal cartridges were used. Pollutant accumulation in special passive air samplers (SPMD, semipermeable mem- brane device) was compared to that occurring in the living needles and air samples from active sampling. Deposition samplers collected wet and dry deposition at the same sites.

Figure 2: High altitude air and deposition sampling station Sonnblick Synthesis of the findings of the project MONARPOP 25

Results and Conclusions Regional differences: The Alps are an effective barrier for atmospheric pollutant transport. Several compounds showed higher concentrations in the border regions of the Alps than in the more shielded central parts. The advantage of this effect is the interception of a further transport across Europe, its disadvantage the higher pollutant load at the border regions of the Alps. Despite the selection of remote sites, almost all studied compounds (even those banned in Europe since many years ago) were clearly detectable in the sampled matrices (air, deposition, needles, soil). The concentrations of many compounds were correlated indicating a complex mix of pollutants at the studied sites. Altitudinal variation at the height profiles: The concentrations varied with altitude. Par- ticularly compounds from faraway sources (certain pesticides, PCBs) tended to have higher concentrations at the highest sites. Compounds which also have local sources showed higher concentrations down in the valleys (e.g. PAHs). Some compounds (e.g. PBDE) peaked at mediate altitudes, suggesting an influence of the inversion layer. Atmosphere and deposition measurements: The atmospheric concentration of the compounds elaborated so far (PCDD/F, PCB, PBDE) did not show a clear dependence on the geographical origin of the sampled air. This would imply that atmospheric POP pollution of the Alps is primarily influenced by the mere quantitative contribution of a source region to the arriving air masses, i.e. by meteorological factors. Again, concentrations in air and deposition were comparable to those of less remote rural sites, indicating the long-range transport of these compounds above the mixing layer of the troposphere. Biological tests: The remote samples showed dioxin-like activity in the bioassay tests. The signals of the individual samples were correlated to the TEQ concentration of PCDD/F and PCB. MONARPOP was included in the “Global Monitoring Plan” for the “Effectiveness Evaluation” of the “UNEP Stockholm Convention on POPs”. By that, the project results will contribute to the controlling of the success of this important international instrument to reduce the input and load with these compounds.

Acknowledgements The MONARPOP project was funded by the EU INTERREG IIIB Programme “Al- pine Space” and by the participating national institutes and authorities, particularly by the: Austrian Ministry for Agriculture, Forestry, Environment and Water Resources Bavarian State Ministry for Environment, Health and Consumer Protection Swiss Agency for the Environment Italian “Fondo Rotazione”

Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 27

idrocarburi policiclici aromatici negli ecosistemi alpini (progetto MONARPOP)

Belis C A1, Magnani T1, Bassan R2, Jakobi G3, Kirchner M3, Kräuchi N4, Levy W3, Moche W5, Offenthaler I5, Schramm3 K-W, Schroeder P3, Sedivy I4, Simoncˇicˇ P6, Uhl M5, Weiss P5

1 Agenzia Regionale per la Protezione dell’Ambiente Lombardia Dip. Sondrio, via Stelvio, 35- 23100 Sondrio; 2 Agenzia Regionale per la Protezione dell’Ambiente Veneto; 3 GSF-Centro Nazionale di Rricerca per l’Ambiente e la Salute, Germania; 4 Istituto Federale Svizzero per la Ricerca sulle Foreste, la Neve e il Paesaggio; 5 Agenzia Federale per l’Ambiente Ltd., Austria; 6 Istituto Sloveno per le Foreste. E-mail: [email protected]

Abstract The paper describes the levels of Polycyclic Aromatic Hydrocarbons (PAHs) deter- mined in different environmental matrices of alpine ecosystems: humus, needles, mineral soil, ambient air and bulk deposition. Sample were collected in a monitoring network, which includes 40 remote sites, 7 vertical profiles and 3 high altitude sites, implemented within the framework of the project MONARPOP. Data were arranged in maps aiming at describing the spatial distribution of PAHs in the Alps. Measure- ments in high altitude sites were used to estimate the long range transport of these pollutants and the rates of bulk deposition in remote areas.

Introduzione Gli idrocarburi policiclici aromatici (IPA) sono composti formati da due o più anelli aro- matici che derivano prevalentemente dalla combustione incompleta di materia organica come ad esempio carbone, derivati del petrolio, olio o biomassa. Le proprietà fisiche degli IPA, in particolare la pressione di vapore e il coefficiente di partizione tra la fase gassosa e la fase solida variano in funzione del loro numero di anelli e del loro peso molecolare. Gli IPA sono sostanze semivolatili trasportate dalle masse d’aria sia allo stato di gas sia adsorbite sulla frazione solida in sospensione. Queste sostanze sono soggette a deposizione secca e umida e come tanti altri composti organici persistenti tendono ad accumularsi in ambienti con basse temperature ed elevate precipitazioni come le regioni montuose. Trattandosi di sostanze poco polari possiedono una forte affinità per i lipidi e si accumulano nel suolo, nei sedimenti e negli organismi viventi. Inoltre, viaggiano lungo le catene alimentari e si concentrano negli organismi al loro apice, compresi gli esseri umani. Gli IPA sono sostanze di interesse tossicologico in quanto alcune di esse sono considerate probabili o possibili cancerogeni (cIPA) (IARC, 1984) 28 Monitoraggio dei POPs sul territorio alpino

La rete di monitoraggio è stata implementata nell’ambito del progetto Interreg III B Spazio Alpino Monarpop. Lo stesso ha avuto lo scopo di misurare i carichi di IPA in siti remoti, mappare la distribuzione delle loro concentrazioni nell’arco alpino e va- lutare il trasporto a lunga distanza. Per garantire la comparabilità dei risultati ottenuti nei diversi siti di campionamento è stato organizzato un sistema di campionamento, trasporto e preparazione dei campioni che ha consentito di effettuare tutte le analisi in un singolo laboratorio.

Materiali e Metodi Le concentrazioni di 20 IPA (compresi i 16 IPA prioritari indicati dalla US-EPA) sono stati analizzati negli aghi di abete rosso (Picea abies), nell’humus, nello strato minerale del suolo, nell’aria e nelle deposizioni atmosferiche. Le aree di campiona- mento comprendono 40 siti standard e 7 profili verticali distribuiti nell’area Alpina (7° - 16°E; 46° - 48°N) La selezione dei siti di campionamento è stata finalizzata alla predisposizione di un set di siti remoti omogenei dal punto di vista vegetazionale e altitudinale. Ad eccezione dei profili verticali i siti di campionamento sono posti ad una altezza tra 1300 e 1600 m s.l.m. e le loro distanze minime a fonti locali di inquinamento rispettano le linee guida della rete EMEP per la direttiva sulla qualità dell’aria. I siti di campionamento per le matrici solide sono posizionati in appezzamenti forestali di almeno 0.5 ha di superficie e 30 anni di età con una purezza di abete rosso superiore all’80%. In tre siti di alta quota Sonnblick (A), Zugspitze(D) e Weissfluhjoch (CH) rispettiva- mente a 3106, 2650 e 2663 m sl.m. sono stati effettuate misure delle concentrazioni di POPs nell’aria con campionatori attivi e nelle deposizioni atmosferiche (vedere Moche et al., questo volume) Gli IPA sono stati determinati utilizzando spettromettria di massa ad alta resoluzione (HRMS) accoppiata ad uno spettrometro di massa ad alta risoluzione (HRGC) nel laboratorio di analisi dell’I.N.C.A. Le concentrazioni di IPA nelle matrici solide (hu- mus, aghi e suolo minerale) sono riferite al peso secco determinato per essicazione a 105°C. La tossicità equivalente (TEC) della frazione IPA nelle matrici ambientali è stata calcolata utilizzando i potency equivalency factors (PEF) riferiti al Benzo(a)pirene (OEHHA, 1993; Collins et al. 1998).

Risultati e Discussione Nel presente articolo vengono trattate le misure effettuate sulla rete di punti di misura ad altezza standard e i rilievi effettuati nei siti ad alta quota. Per una descrizione delle concentrazioni di IPA nei profili verticali si rimanda a Beliset al. (2007). Le mediane degli IPA totali (Σ20 IPA) nelle matrici solide sono 189 µg kg-1 nell’hu- mus , 495 µg kg-1 nello strato minerale e 19 µg kg-1 negli aghi. Benzo(b)fluorantene, Idrocarburi policiclici aromatici negli ecosistemi alpini (progetto MONARPOP) 29

Crisene, Indeno(c,d)pirene e Fluorantene sono i composti più abbondanti nel suolo mentre nelle aghi di abete gli IPA più abbondanti sono Naftalene e Fenantrene. Nell’humus e nello strato minerale le maggiori concentrazioni di IPA (Σ20 IPA >75° Percentile) si osservano prevalentemente sul versante nord delle Alpi e in parte nel- l’area più occidentale mentre le concentrazioni più basse (Σ20 IPA <25° Percentile) si verificano nelle aree centrale e meridionale delle Alpi (Figura 1). In questi siti si

Figura 1. Siti di campio- namento di humus (A), strato minerale (B) e aghi di abete (C) nelle Alpi . Cerchi blu: Σ20 IPA, cer- chi rossi: concentrazioni di tossicità equivalente (TEC) riferite al Benzo (a) pirene. 30 Monitoraggio dei POPs sul territorio alpino osserva una corrispondenza tra la concentrazione totale di IPA e la tossicità della miscela di questi composti calcolata in base al Benzo(a)pirene. I campioni di aghi con le maggiori concentrazioni totali di IPA (Σ20 IPA >75° Percentile) si verificano prevalentemente nell’area occidentale dell’area di studio mentre quelli con le concen- trazioni più modeste (Σ20 IPA <25° Percentile) si osservano nella zona centrale delle Alpi (Figura 1). A differenza del suolo la distribuzione spaziale della tossicità della miscela di IPA negli aghi non coincide con quella degli IPA totali. I campioni con le maggiori concentrazioni di tossicità equivalente si osservano nella zona orientale dell’area di studio. Questa discordanza tra le diverse matrici può essere spiegata dal fatto che gli aghi di abete rappresentano soltanto l’ultima stagione vegetativa mentre il pool di IPA nel suolo rappresenta l’accumulo di un periodo di almeno un decennio. I rapporti tra alcuni IPA osservati nelle matrici ambientali possono essere utilizzati come marker molecolari per identificare la fonte dalla quale deriva l’inquinamento (Brändli et al., 2006). In particolare i rapporti Fluorantene/Pirene e Indeno(c,d)Pirene/ Benzo(g,h,i)perilene nell’humus e nello strato minerale consentono di escludere l’origine petrogenico degli IPA nei campioni studiati e indicano che la componente di questi composti che deriva dalla combustione di biomasse prevale su quella della combustione di combustibili fossili. I dati raccolti nei tre siti di monitoraggio in alta quota (Figura 2) attrezzati per il campionamento dei POPs nell’aria e nelle deposizioni atmosferiche totali forniscono informazioni sulla circolazione di questi inquinanti a lunga distanza. In generale le concentrazioni di IPA nell’aria e la direzione di provenienza delle masse d’aria più inquinate presentano un discreto carattere stagionale. Nel sito di Sonnblick (A) i livelli medi di IPA sono inferiori a quelli osservati negli altri due siti (Figura 3). In questa postazione le concentrazioni di IPA sono maggiori nelle masse d’aria provenienti da NW e in quelle con provenienze miste (MISTO). A Zugspitze (D) le maggiori concentrazioni di IPA sono associate alla circolazione da NW mentre a Weissfluhjoch (CH) la direzione di provenienza varia da NE in estate a NW e provenienze miste in inverno. A differenza di quanto accade nell’aria ambiente i flussi medi di IPA nelle deposizioni atmosferiche sono superiori a Sonnblick rispetto agli altri due siti di monitoraggio (Figura 4). Queste differenze sono probabilmente determinate dalla maggiore quo- ta del sito austriaco. In questa postazione le temperature più basse provocano una maggiore condensazione degli IPA più volatili promuovendo il passaggio dalla fase gassosa alla fase solida. Le informazioni sinora disponibili indicano che esiste un aumento della tossicità della miscela di IPA a seguito del processo di deposizione. Nell’aria ambiente le con- centrazioni di cIPA rappresentano meno del 1% della miscela totale di IPA, mentre questa percentuale sale nelle deposizioni a valori che oscillano tra il 10% e il 40%. Analogamente gli cIPA negli aghi di abete rappresentano il 10±8 % della miscela mentre nel suolo sottostante le percentuali salgono al 34±5 % nell’humus e al 48±6 % nello strato minerale. Idrocarburi policiclici aromatici negli ecosistemi alpini (progetto MONARPOP) 31

Figura 2. Ubicazione dei siti di mo- nitoraggio in alta quota

Figura 3. Direzione media di prove- nienza degli IPA nell’aria ambiente nei tre siti d’alta quota allestiti nel- l’ambito del progetto Monarpop. (concentrazioni in pg m-3)

Figura 4. Deposizioni atmosferiche di IPA nei siti di alta quota (SON: Sonnblick, ZUG: Zugspitze, WEI: Weissfluhjoch)

Conclusioni Il confronto delle concentrazioni di IPA negli ecosistemi forestali alpini ha messo in evidenza che esistono marcate differenze regionali. Le aree remote maggiormente inquinate da IPA nel suolo e nella biomassa fogliare risultano essere rispettivamente il versante nord e il versante orientale. Per quanto riguarda il versante sud delle Alpi si osserva che le concentrazioni di IPA nei siti sloveni appaiono più elevate rispetto quelli italiani, i quali presentano livelli intermedi di queste sostanze. Tuttavia, una 32 Monitoraggio dei POPs sul territorio alpino valutazione più esaustiva della situazione sul versante sud delle Alpi richiederebbe un maggiore numero di siti di campionamento e una migliore copertura spaziale. La composizione della miscela di IPA nella matrice suolo, in particolare il rapporto tra alcuni composti, conferma l’origine pirogenico di questi inquinanti e indica la combustione di biomasse come probabile fonte principale. I rilievi eseguiti nei siti di alta quota delle Alpi centrali e settentrionali indicano che i livelli di IPA nell’aria ambiente presentano un marcato carattere stagionale sia per quanto riguarda la concentrazione sia per la loro provenienza. Durante il periodo invernale le masse d’aria più inquinate provengono spesso dal settore NW ma sono altrettanto frequenti masse inquinate di IPA con una storia complessa che non consente di stabilire una direzione certa di provenienza. Dalle informazioni raccolte nelle diverse matrici ambientali emerge che il processo di deposizione degli IPA, direttamente o dopo un passaggio attraverso la biomassa fogliare, ha un ruolo chiave nel determinare la tossicità della miscela di IPA in quanto si osserva la tendenza alla concentrazione degli IPA di maggiore peso molecolare tra i quali si trovano quelli probabili o possibili cancerogeni.

Ringraziamenti MONARPOP è finanziato dal Programma Interreg IIIB Spazio Alpino, dal Governo dello Stato Italiano, dal Ministero Federale Austriaco per l’Agricoltura, le Foreste, l’Ambiente e le Acque, dal Ministero Bavarese per l’Ambiente e dall’Ufficio Federale Svizzero per l’Ambiente.

Bibliografia Belis C., R. Bassan, S. Iozza, G. Jakobi, M. Kirchner, N. Krauchi, W. Levy-Lopez, T. Ma- gnani, W. Moche, M. Oehme, I. Offenthaler, K.W. Schramm, H. Schroeder, I. Sedivy, P. Simoncic, G. Thanner, M. Uhl, U. Vilhar, P. Weiss, 2007. PAHs in needles and humus in alpine ecosystem (Project Monarpop). Organohalogen Compounds (in press). Brändli R.C., T.D. Bucheli , T. Kupper, F.X. Stadelmann & J. Tarradellas, 2006. Can sources of environmental contamination with PAHs be identified in recipient matrices by concomitant analysis of molecular markers?. Organohalogen Compounds 68:292 Collins J.F., J.P. Brown , G.V., A.G. Salmon, 1998. Potency equivalent factors for some poly- cyclic aromatic hydrocarbons and polycyclic aromatic hydrocarbons derivatives. Regul. Toxicol. Pharmacol. 28:45 IARC, 1984. Monograph 32, International Agency for Research on Cancer, Lyon Moche W., R. Bassan, C. Belis, S. Iozza, G. Jakobi, M. Kirchner, N. Krauchi, W. Levy-Lopez, T. Magnani, M. Oehme, I. Offenthaler, K.W. Schramm, H. Schroeder, I. Sedivy, P. Simon- cic, G. Thanner, M. Uhl, U. Vilhar, P. Weiss, 2007. Novel ambient air sampling techniques adapted for the needs of project MONARPOP. (this volume). OEHHA. Benzo(a)pyrene as a toxic air contaminant. Air Toxicology & Epidemiology Sect, Berkeley, 1993. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 33

Studio delle concentrazioni di POPs nell’area dolomitica

R. Bassan, C. Bellio, A. Cavinato, S. Ganz, R. Piol, C. Belis., D. Heublein., S. Iozza, T. Jakl, G. Jakobi, M. Kirchner, W. Knoth, N. Krauchi, W. Levy-Lopez, T. Magnani, W. Moche, M. Oehme, I. Offenthaler, R. Perthen-Palmisano, D. Schmid, K. W. Schramm, P. Schroder, H. Schrott, I. Sedivy, P. Simoncˇicˇ, M. Uhl, P. Vannini, U. Vilhar, P. Weiss

Abstract In the context of the MONARPOP project a study on the POPs concentrations in the three sites of the Dolomites has been carried out. In the matrices needles, humus layer, mineral soil and semi permeable membrane devices (SPMD) different POPs groups have been investigated: OCP, PCB, PCDD/F, PBDE, CP and PAH and generally low or very concentrations arose, especially in the site Val Visdende, showing one of the best situation of the whole alpine area.

Introduzione Nell’ambito del progetto MONARPOP relativo alla presenza di inquinanti organici persistenti (POPs) in siti remoti dell’arco alpino (Weiss, 2007), sono stati analizzati campioni provenienti da quattro matrici ambientali: aghi di Picea abies (L.) Karsten (Abete rosso), humus, suolo minerale e aria, quest’ultima attraverso campionatori attivi e passivi. Lo studio qui presentato riguarderà in particolare le prime tre matrici più le membrane semipermeabili per il campionamento passivo dell’aria (Schramm, 2007); per ognuna si analizzerà in dettaglio la situazione nel Veneto attraverso il profilo altimetrico della Val Visdende ed i siti della Val Prampera e del Monte Grappa.

Materiali e Metodi La scelta dei siti di studio è stata condotta con lo scopo di coprire la montagna veneta tra 45° 50’ e 46° 38’ Nord, in modo tale da avere informazioni sia su aree remote af- facciate sulle Alpi Austriache e l’Europa, sia sulla pianura Padana. Con questi criteri informatori, a nord è stata scelta la Val Visdende con la contigua Val Frison, a sud il Massiccio del Grappa, alla testata della Val Stizzon affacciata sulla pianura trevigiana e vicentina; intermedia tra questi due siti è stata individuata la Val Prampera (Dolo- miti Zoldane), inserita nel Parco Nazionale delle Dolomiti Bellunesi, come indicato in Figura 1. La Val Visdende è geograficamente collocata tra il mondo dolomitico a sud, carat- terizzato da valli incise e forte energia del rilievo, e quello alpino a nord, con forme più dolci. La valle si presenta particolarmente ampia per la confluenza di due rami guidati dalla tettonica e modellati dalle forme glaciali e fluvio-lacustri; risulta protetta verso meridione dall’orografia e viene a trovarsi relativamente aperta alle correnti 34 Monitoraggio dei POPs sul territorio alpino

Figura 1. Mappa dei punti veneti di campionamento. Tratta da Google earth 2007 e modificata.

provenienti dal nord delle Alpi. E’ un’area di pregio naturalistico caratterizzata da pochi insediamenti, popolati quasi esclusivamente nel periodo estivo, dall’assenza di attività artigianali e industriali, e dalla presenza di viabilità esclusivamente silvo pastorale con valenza turistica nella buona stagione. Per queste motivazioni la valle è stata scelta per la realizzazione del profilo altimetrico tra 1123 e 1656 m s.l.m. e per il posizionamento delle capannine contenenti i campionatori passivi (SPMD). Nella valle sono stati anche eseguiti campionamenti di aghi di abete, humus e suolo minerale. La Val Prampera costituisce una valle remota di accesso al Parco Nazionale delle Dolomiti Bellunesi dalla valle di Zoldo. Collocata al centro delle Dolomiti Bellunesi tra i Gruppi del Talvena, S. Sebastiano e Cime di Mezzodì, la valle presenta orografia e morfologia particolarmente aspre; è caratterizzata da una notevole energia del rilievo che la rende isolata dai contigui territori antropizzati. La Val Prampera è contraddi- stinta dall’assenza di insediamenti stabili, dalla presenza di viabilità esclusivamente silvo pastorale con valenza turistica prevalentemente estiva. Per il campionamento è stato scelto un sito a quota 1486 m. Il Monte Grappa è un massiccio montuoso che raggiunge i 1775 m. Appartiene alle Prealpi Venete e risulta diviso tra le province di Vicenza, Treviso e Belluno. Nell’am- bito dei siti di progetto costituisce la finestra sulla pianura padano-veneta. Il punto di campionamento è stato scelto a 1345 m in una zona sub pianeggiante ubicata alla testata della bellunese Val Stizzon, caratterizzata anch’essa da scarsissima antropiz- Studio delle concentrazioni di POPs nell’area dolomitica 35 zazione. Il campionamento di aghi di abete rosso, strato di humus e suolo minerale, così come la conservazione ed il trasporto dei campioni, sono avvenuti secondo meto- dologie standardizzate e condivise da tutti i partners di progetto, definite nell’ambito di numerosi incontri tecnici. Le analisi chimiche per la ricerca degli inquinanti organici persistenti sono state quindi svolte presso i laboratori selezionati. Le elaborazioni dei dati analitici svolte presso il Dipartimento Provinciale ARPAV di Belluno, rivolte allo studio della situazione locale, sono state effettuate mediante l’impiego del software statistico Kyplot e fogli di calcolo.

Risultati e discussione Le elaborazioni effettuate verranno di seguito riportate per singola matrice d’indagine. Si considereranno sia i siti standard che i profili altimetrici e nei confronti e calcoli delle variabili statistiche saranno trattati indistintamente. Eventuali valutazioni saranno effettuate nel dettaglio solo per i tre siti veneti, ove presenti i dati. Nella Figura 2 sono rappresentati i valori minimi, massimi e medi per le famiglie di in- quinanti indagate nei siti del Veneto e del resto dell’arco alpino, divisi per matrice. Aghi di abete. Gli aghi delle piante di abete sono stati sottoposti all’analisi di quattro famiglie di POPs: pesticidi organoclorurati (OCP), policlorobifenili (PCB), diossine e furani (PCDD/F) e idrocarburi policiclici aromatici (PAH) (Schroder et al., 2007). Nel profilo altimetrico della Val Visdende per tutti gli OCP i valori di concentrazione più alti si sono osservati a quota 1553 m, in corrispondenza del terzo punto del pro- filo, con concentrazioni in quota o in fondo valle sempre inferiori (Kirchner, 2007). Sebbene tutte le concentrazioni siano comprese nell’intervallo (µ–2σ)

Figura 2. Valori minimi, massimi e medi per inquinanti nei siti veneti e nel resto delle Alpi, per matrice d’indagine. Studio delle concentrazioni di POPs nell’area dolomitica 37 come unico sito quello della Val Visdende. Qui le concentrazioni maggiori di OCP si sono registrate nel punto di fondovalle (1123 m), mentre salendo di quota la situazio- ne ha registrato significativi miglioramenti, fino a raggiungere le concentrazioni più basse di tutto l’arco alpino. Per PCB, diossine e furani si è ripresentato un andamento con un massimo a quota 1553 m, terzo punto del profilo; nel caso dei PCB spesso il valore massimo è risultato notevolmente superiore alle concentrazioni di fondovalle e in quota, mentre minori differenze si sono riscontrate per diossine e furani. I PBDE hanno mostrato andamenti simili ai precedenti composti, ma con concen- trazioni molto basse, prossime ai valori minimi dell’intero campione (Knoth et al., 2007a, Knoth et al., 2007b). La concentrazione di cloroparaffine hanno invece presentato un trend di crescita da valle a monte, con il raggiungimento alle quote più alte (1656 m) di valori molto elevati (Iozza, 2007). Infine i PAH (Bassan & Ganz, 2007; Belis et al., 2007), presenti in concentrazioni sempre inferiori alla media, hanno di contro evidenziato un profilo decrescente con la quota, dove a 1656 m si è presentato il minimo assoluto dell’intero campione alpino. Suolo minerale. Per la matrice suolo minerale i dati attualmente disponibili sono quelli relativi ai soli idrocarburi policiclici aromatici. L’unico campione veneto, relativo alla Val Visdende, ha presentato concentrazioni molto basse vicine ai minimi assoluti. SPMD. I Semi Permeable Membrane Devices (SPMD) sono campionatori passivi dell’aria, posizionati all’interno di colonnine di legno (Levy-Lopez et al., 2007). Anche per questa matrice le analisi disponibili sono quelle relative ai PAH, con campioni provenienti dalla Val Visdende con concentrazioni mediamente basse, al di sotto della media dell’arco alpino.

Conclusioni Dalle analisi effettuate nei siti remoti della provincia di Belluno scelti per il monito- raggio degli inquinanti organici persistenti emerge che la situazione rispetto al quadro generale è fra le migliori. Le concentrazioni sono risultate generalmente basse, spesso inferiori alla media dell’arco alpino, con valori che a volte rappresentano i minimi assoluti dell’intero campione di dati. Il sito meno contaminato è risultato la Val Visdende; a seguire, nell’ordine, la Val Prampera e il Monte Grappa, suggerendo così un gradiente nord-sud tra i tre siti dell’area dolomitica. Interessante è risultato infine l’andamento dei profili verticali in Val Visdende: per OCP negli aghi di abete e PCB, PCDD/F e PBDE nello strato di humus i massimi di concentrazione si sono verificati alla quota intermedia (1553 m), probabilmente a causa delle dinamiche dello strato mescolato nella valle. In altri casi, come ad esempio PAH e OCP nell’humus, i profili sono risultati decrescenti con l’altezza, con valori di concentrazione in quota pari ai minimi assoluti del campione; viceversa profili con massimi di concentrazione in quota, molto alti anche rispetto agli altri campioni dell’ar- co alpino, si sono verificati nel caso, seppur unico, delle cloroparaffine nell’humus. 38 Monitoraggio dei POPs sul territorio alpino

Bibliografia Bassan R. & S. Ganz, 2007. PAHs presence in the investigated matrixes: comparison end evalua- tion. Presenza degli inquinanti organici persistenti nelle valli alpine. Venezia, 5/10/2007. Belis C., R. Bassan, D. Heublein, S. Iozza, T. Jakl, G. Jakob, M. Kirchner, W. Knoth, N. Krauchi, W. Levy-Lopez, T. Magnani, W. Moche, M. Oehme, I. Offenthaler, R. Perthen- Palmisano, K.W. Schramm, H. Schrott, I. Sedivy, P. Simoncic, M. Uhl, U. Vilhar, P. Weis, 2007. PAHs in needles and humus in alpine ecosystem (MONARPOP). Organohalogen Compounds (in press). Iozza S., P. Schmid, P. Oehme, R. Bassan, C. Belis, G. Jacobi, M. Kirchner, I. Sedivy, N. Krau- chi, W. Moche, Offenthaler, I. Simoncic, M. Uhl, P. Weiss., 2007. Chlorinated paraffins in humus layers from the Alps (MONARPOP). Tokyo Dioxin 2007. Kirchner M., G. Jakobi, B. Henkelmann, S. Bernhöft, R. Bassan, C. Belis, N. Kräuchi, W. Moche, I. Offenthaler, P. Simoncˇicˇ, P. Weiss, K. W. Schramm, 2007. Pesticides in soil and spruce in the Alps. Presenza degli inquinanti organici persistenti nelle valli alpine. Venezia, 5/10/2007. Knoth W., R. Bassan, C. Belis, D. Heublein, S. Iozza, T. Jakl, G. Jakobi, M. Kirchner, N. Krauchi, W. Levy-Lopez, W. Moche, M. Oehme, I. Offenthaler, R. Perthen-Palmisano, D. Schmid, K. Schramm, H. Schrott, P. Schroder, I. Sedivy, P. Simoncic, M. Uhl, P. Weiss, 2007a. Results from the Monitoring Network in the Alpine Region for POPs-MONARPOP. Polybrominated diphenyl ether (PBDE) in humus layers in remote forests. International Workshop on Brominated Flame Retardants – BFR 2007 – Amsterdam. Knoth W., R. Bassan, C. Belis, G. Jacobi, M. Kirchner, N. Krauchi, W. Mann, R. Meyer, W. Moche, J. Nebhuth, I. Offenthaler, I. Sedivy, P. Simoncic, M. Uhl, U. Vilhar, P. Weiss, 2007b. Poster: PBDE in humus layers in remote forests. 4th International Workshop on Brominated Flame Retardants – Amsterdam. Levy-Lopez W., B. Henkelmann, G. Pfister, S. Bernhoft, A. Niklauss, M. Kirchner, G. Jakobi, R. Bassan, C. Belis, T. Jakl, N. Krauchi, T. Magnani, W. Moche, I. Offenthaler, R. Perthen- Palmisano, P. Schroder, H. Schrott, I. Sedivy, P. Simoncic, M. Uhl, P. Vannini, U. Vilhar, P. Weiss, K. W. Schramm, 2007. Semipermeables membrane devices (SPMD) as passive samplers: data interpretation regarding exposure time. Tokyo Dioxin 2007. Offenthaler I., 2007. Dioxinlike and other halogenated pollutants in forest ecosystems of the Alps. Presenza degli inquinanti organici persistenti nelle valli alpine. Venezia, 5/10/2007. Offenthaler I., W. Moche, G. Thanner, M. Uhl, K. Van Ommen, R. Bassan, C. Belis, G. Jakobi, M. Kirchner, N. Krauchi, K. W. Schramm, W. Levy-Lopez, I. Sedivy, P. Simoncic, U. Vilhar, P. Weiss, 2007. Dioxin and dioxin-like pollutants in alpine forests. Tokyo Dioxin 2007. Schramm K. W., 2007. Novel diagnostic tools for POPs in alpine area. Presenza degli inquinanti organici persistenti nelle valli alpine. Venezia, 5/10/2007. Schroder P., R. Bassan, C. Belis., D. Heublein., S. Iozza, T. Jakl, G. Jakobi, M. Kirchner, W. Knoth, N. Krauchi, W. Levy-Lopez, T. Magnani, W. Moche, M. Oehme, I. Offenthaler, R. Perthen-Palmisano, D. Schmid, K. W. Schramm, H. Schrott, I. Sedivy, P. Simoncic, M. Uhl, P. Vannini, U. Vilhar, P. Weiss, 2007. Detoxification of organic pollutants by glutathione S transferas in spruce needles alpine forests - Results from a Monitoring in the Alpine Region for POPs (MONARPOP). Chicago Americal Society for Plants Biologists (ASPB). Weiss P., 2007. MONARPOP: starting point, plan and goals of the project. Presenza degli inquinanti organici persistenti nelle valli alpine. Venezia, 5/10/2007. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 39

Results from the Monitoring Network in the Alpine Region for POPs MONARPOP POLYBROMINATED DIPHENYL ETHER (PBDE) IN HUMUS LAYERS IN REMOTE FORESTS

Knoth W1, Bassan R2, Belis C3, Jakobi G4, Kirchner M4, Kräuchi N5, Mann W1, Meyer R1, Moche W6, Nebhuth J1, Offenthaler I6, Sedivy I5, Simoncˇicˇ P7, Uhl M6,Vilhar U7, Weiss P6

1 Federal Environment Agency – Germany 2 Regional Agency for Environmental Prevention and Protection of Veneto 3 Regional Agency for Environmental Protection of Lombardia 4 GSF – National Research Center for Environment and Health – Germany 5 WSL – Swiss Federal Institute for Forest, Snow and Landscape Research 6 Federal Environment Agency Ltd. – Austria 7 Slovenian Forestry Institute

Riassunto I bifenili etere polibromurati (PBDE) sono una famiglia di composti molto utilizzati come ritardanti di fiamma nella produzione di sostanze plastiche e di tessuti. Nel presente lavoro vengono presentate le concentrazioni di PBDE nell’humus dei congeneri più significativi tra cui la miscela PentaBDE (∑BDE 28+47+99+100+ 153+154), l’OctaBDE (BDE 183) e il DecaBDE (BDE 209).

Introduction Polybrominated diphenyl ether (PBDE) are heavily used as flame retardants in plastics and textiles and have therefore become ubiquitous (de Wit 2002, Letscher & Behnisch 2003). There exist three technical products: Pentabromodiphenylether (PentaBDE) mix, octabromodiphenylether (OctaBDE) mix and decabromodiphenylether (De- caBDE). Because of exponentially increasing levels of the congeners of PentaBDE and OctaBDE mix in human blood and milk (Hites 2004), these two products were banned in many countries (BSEF, EU 2003, NCEL) and industry voluntarily ceased production (Tullo 2003). Decabromodiphenylether was not included in these bans because it is believed to have low bioaccumulation (EU 2004) and low long-range air transport potential (Wania & Dugani 2003). About 8000 t DecaBDE were used in EU countries in 2004 (EBFRIP 2005). Industry tries to reduce existing emissions throughout the live-cycle and sponsors a deca-monitoring programme (BSEF). Plant surfaces were already used for sampling of PBDE from air. (Hassanin 2005, Mariussen 2005, Ohta 2002, Okazawa 2004, Schlabach & Gjerstad 2006, Schütz 2004, Zhu & Hites 2006). Especially the rough canopy of coniferous forests combs the bypassing air for pollutants and conifer needles efficiently trap lipophilic organic compounds due to their highly absorbing epicuticular wax. The topsoil humus layers 40 Monitoraggio dei POPs sul territorio alpino in forests with their high organic carbon content accumulate POPs from deposition and fall of needles over a long period (Weiss 2002). For this study humus samples from remote Norway spruce (Picea abies) forests were analysed for PBDE. This paper presents results from the Monitoring Network in the Alpine Region for Persistent and other Organic Pollutants (MONARPOP, Bassan 2005). Its main goal is to investigate the actual contamination of the Alps and to understand the role of high mountains in the global atmospheric transport of POPs (Daly & Wania 2005, Vighi 2006).

Materials and Methods Humus Sampling 56 humus samples from remote Norway Spruce forest sites in the Alps were sampled for PBDE analysis. This was done by collecting the entire humus layer within a 30 x 30 cm metal frame. Ten pits along a 5 x 30 m rectangular grid were collected systemati- cally and mixed to one sample. This yielded up to 60 l of humus per sampling plot. Analytical Method In indoor dust high concentrations of PBDE were detected (Knoth 2003, 2006). There- fore, a clean laboratory environment is essential for the analysis of PBDE, in particular for the analysis of DecaBDE. Polymer material (e.g. red rubber septa) was tested be- fore use. Silica, sodium sulfate and glass wool were extracted with dichloromethane. After evaporation of the solvent, silica and sodium sulfate and all laboratory glassware used for the clean-up procedure were baken out for 16 hours at 450°C. Glass fibre extraction thimbles and filters, Pasteur pipettes and glass wool, which became brittle if heated so long, were immediately cooled down after heating up. After cooling down the glassware was immediately capped with aluminium foil or stored in metal boxes until usage. Sodium hydroxide water solution for the preparation of SiO2-NaOH was extracted three times with dichloromethane. To avoid cross contamination the vapour tubes of the rotary evaporators were changed after each sample. 13 20 g freeze-dried and homogenized humus sample was spiked with the C12-BDE standard mixture (1 ng BDE 28, 47, 99; 2 ng BDE 153, 154, 183 and 5 ng BDE 209) and extracted by Soxhlet extraction (Knöfler-Böhm hot extractor). Residual water (0-11%) was simultaneously distillated with Dean-Stark water separator with tolu- ene. The extract was cleaned by a four column clean-up (1. Multi-layer SiO2-H2SO4, NaOH. 2. Macro alumina. 3. GPC bio-beads S-X3. 4. Mini alumina), spiked with 13 the injection standard (1 ng C12-BDE 138) and reduced to 50 µl. 1 µl was injected on-column (guard column 2 m x 0.32 mm, uncoated, deactivated) and analysed by GC-SIM(EI+)HRMS (TRACE GC-MS MAT 95 XP, ThermoFinnigan, Bremen) using a DB-5MS (15 m x 0.25 mm, 0.1 µm). The two most intense masses of the bromine cluster (Tri- and TeBDE: M+. Te- to DeBDE: M+-2Br) were measured for each homologue group. The identification of PBDE was based on retention time and correct isotope ratio for both fragments recorded. Quantification was performed by 13 means of the C12-labeled internal standards. All congeners except BDE 100 were 13 quantified based on their corresponding C12-labeled analogues used as internal Polybrominated diphenyl ether (pbde) in humus layers in remote forests 41

13 standards. BDE 100 was quantified using the C12-BDE 99 internal standard. PBDE concentrations were recalculated with conversion factors to reduce freeze-dried to oven dry (at 105°C) humus mass. All samples were analysed at the German Federal Environment Agency. The labora- tory took part in the BSEF/QUASIMEME interlaboratory study on brominated flame retardants December 2001 to March 2002 and the interlab trial for ISO/DIS 22032 No- vember 2004 to February 2005. Method blanks were spiked on a plug of glasswool in a Soxhlet extraction thimble and extracted and clean-up processed every four samples. Blank concentrations were calculated on a fictive mean sample weight of 18.3 g d.m.. The method detection limit (MDL) was determined as the mean concentration in the blank plus 3 times the standard deviation of 19 measurements (Table 1).

Table 1. Concentrations of PBDE in humus layers in Norway spruce forests in the Alpine region

Results and Discussion PBDE were detected in all humus samples. The total concentration of six signifi- cant congeners of the technical PentaBDE mix (∑BDE 28+47+99+100+ 153+154) ranges from 190 to 1500 (median 490) and the significant congener of the technical OctaBDE mix (BDE 183) from 14 to 3000 (median 140) ng kg-1 d.m.. Although four multimedia models predicted for DecaBDE (BDE 209) a very low potential to reach remote areas (Wania & Dugani 2003) levels from 610 to 85000 (median 1400) ng kg-1 d.m. were observed. (Table 1 and Fig. 1-3). A contribution of local but until now unknown sources is probable for two very high BDE 209 concentrations (29000 and 85000 ng kg-1 d.m.) (Fig. 3). The dominating congeners in the humus samples are BDE 209 (51-97%), BDE 183 (0.7-31%), BDE 47 and BDE 153 (both 0.2-18%) and BDE 99 (0.1-13%). The con- tribution of the other congeners is of minor importance (BDE 100 0.03-3, BDE 154 0.1-1.5 and BDE 28 0.01-0.4%). BDE 183 is a minor congener in many environmen- tal samples. Its higher percental contribution in humus layers may be due to revola- tilisation of the congeners with less than seven bromine from humus to air (Fig. 4). The dominating congeners in the humus samples are BDE 209 (51-97%), BDE 183 (0.7-31%), BDE 47 and BDE 153 (both 0.2-18%) and BDE 99 (0.1-13%). The contribution of the other congeners is of minor importance (BDE 100 0.03-3, BDE 154 0.1-1.5 and BDE 28 0.01-0.4%). BDE 183 is a minor congener in many environmental samples. Its higher percental contribution in humus layers may be due to revolatilisation of the congeners with less than seven bromine from humus to air (Fig. 4).

Table 1. Concentrations of PBDE in humus layers in Norway spruce forests in the Alpine region concentration [ng kg-1 d.m.] BDE 28 BDE 47 BDE 99 BDE 100 BDE 153 BDE 154 Ȉ 6 BDE BDE 183 BDE 209 n 56 56 56 56 56 56 56 56 56 min 2 93 46 12 7 6 191 14 611 max 19 745 463 110 575 71 1472 3037 84504 5% 2 97 52 14 12 6 207 23 674 10% 3 104 56 15 16 7 220 28 823 25% 4 141 70 19 28 10 313 65 1064 50% 5 252 124 35 39 16 485 139 1414 4275% 7 320 197 55 74 31Monitoraggio 795 dei 297 POPs sul 2550 territorio alpino 90% 10 504 326 82 174 49 1028 789 7483 MDL 1 49 16 4 2 1 73 4 133 Ȉ 6 BDE = BDE 28+47+99+100+153+154

25 12

10 tions 20

8 15

6 10 6 concentrations BDE Ȉ 4

5 2 number of BDE 183 concentra number of 0 0 //

0 0 0 0 0 0 0 0 0 0 00 00 00 00 0 0 0 0 0 00 -2 -3 -4 -5 -60 300 40 600 70 800 900 1100 0-100 0 0 -15 0-100 0- 0- 0- - -1200 00 00 00 00 100-200 20 300- 400-500 50 600- 70 800- 00 1 2 3 40 50 600-70 700-80 800-900900-1000 4 900-10 1000-11 1100-12 1200-13 1300-14 1 1000 11 3000-3100 concentration ranges [ng kg-1 d.m.] concentration ranges [ng kg-1 d.m.]

30 100%

ions 25 80% BDE 209 20 BDE 183

15 60% BDE 154 BDE 153 10 BDE 100 40% BDE 99 5 BDE 47 BDE 28 number of BDE 209 concentrat number 209 BDE of 20% 0 // //

0 0 0 0 0 0 0 0 0 0 0 0 00 0 00 0 00 0 0 0 -30 -50 -6000 -8 2000 90 0 0 0 10 -1 -2 -85 0-1 00-4000 00 0- 0 0 0 0% 00 0 0 00 0 00-11 0 0 0 1000-20002 3 400 5 6000-707 8000-9000 0 90 10 80 40 10 1 2 8 1 6 11 16 21 26 31 36 41 46 51 56 concentration ranges [ng kg-1 d.m.] sample no. Fig. 1-3. Frequency histogram of Ȉ 6 BDE (BDE 28+47+99+100+153+154), BDE 183 and BDE 209 Fig.concentrations 1-3. Frequency in humus histogram layers in of Norway Σ 6 BDE spruce (BDE forests 28+47+99+100+153+154), in the Alpine region. BDE 183 and BDE 209 concentrations in humus layers in Norway spruce forests in the Alpine region. Fig. 4. PBDE congener profiles in humus layers in Norway spruce forests in the Alpine region. Fig. 4. PBDE congener profiles in humus layers in Norway spruce forests in the Alpine region.

Acknowledgement MONARPOP is funded by the EU Interreg III B Alpine Space Programme (Alpine Space) and by the Austrian Federal Ministry of Agriculture, Forestry, Environment and Water Management (lead partner of MONARPOP), by the Bavarian State Min- istry for Environment, Health and Consumer Protection and by the Swiss Federal Office for the Environment (together with the institutions listed above).

References Alpine Space. INTERREG III B Alpine Space Programme.

de Wit CA. An overview of brominated flame retardants in the environment. Chemosphere 2002;46:583. EBFRIP. European brominated flame retardant industry panel. Decabromodiphenyl ether (Deca-BDE) usage in EU countries in 2004 in metric tons. 27 May 2005, 1 p. EU 2003. Directive 2003/11/EC of the European Parliament and of the Council of 6 February 2003 amending for the 24th time Council Directive 76/769/EEC relating to restrictions on the marketing and use of certain dangerous substances and preparations (pentabro- modiphenyl ether, octabromodiphenyl ether). Official Journal of the European Union 2003;L42:45. EU 2004. Update of the risk assessment of bis(pentabromophenyl)ether (decabromodiphenyl ether). Final environmental draft of May 2004. CAS No. 1163-19-5, EINECS No. 214- 604-9, 114pp. Hassanin A, Johnston AE, Thomas GO, Jones KC. Time trend of atmospheric PBDEs inferred from archived U.K. herbage. Environ Sci Technol 2005;39:2436. Hites RA. Polybrominated diphenyl ethers in the environment and in people: A meta-analysis of concentrations. Environ Sci Technol 2004;38:945. Knoth W, Mann W, Meyer R, Nebhuth J. Brominated diphenyl ether in indoor dust. Organo- halogen Comp 2003;61:207. Knoth W, Mann W, Meyer R, Nebhuth J. Humus from coniferous forests a reservoir for PBDE from air and deposition –analysis and quality control. Organohalogen Comp 2006;68:1163 Letcher RJ, Behnisch PA, Ed.. The state-of-the science and trend of BFRs in the environment. Environment international 2003;29:663. Mariussen E, Steinnes E, Gundersen H, Borgen A, Schlabach M. Analysis of polybrominated diphenyl ethers in moss (Hylocomium splendens) from the Norwegian environment. Or- ganohalogen Comp 2005;67:591. MONARPOP. Monitoring network in the Alpine Region for persistent and other organic pol- lutants. NCEL. National Caucus of Environmental Legislators. Bulletins March 4 and May 31, 2005. Ohta S, Ishizuka D, Nishimura H, Nakao T, Aozasa O, Shimidzu Y, Ochiai F, Kida T, Nishi M, Miyata H. Comparison of polybrominated diphenyl ethers in fish, vegetables and meats and levels in human milk of nursing women in Japan. Chemosphere 2002;46:689. Okazawa T, Hanari N, Guruge K, Wzykowska B, Orlikowska A, Yamashita N, Falandysz J. Congener-spezific data of PBDEs in pine needles.Organohalogen Comp 2004;66:3725. Schlabach M, Gjerstad KI. Deposition of airborne BFR around the city of Ålesund, Norway. Organohalogen Comp 2006;68:73 Schütz A. Bestimmung polybromierter Diphenylether (PBDE) in Umweltproben unter be- sonderer Berücksichtigung des DecaBDE. Diplomarbeit, Universität Ulm, Abteilung Analytische Chemie und Umweltchemie 2004;146 pp. 44 Monitoraggio dei POPs sul territorio alpino

Tullo A. Flame-retardant phase out. Chemical & Engineering News November 10, 2003:13. Vighi M, Ed.. The role of high mountains in the global transport of persistent organic pollut- ants. Ecotoxicology and Environmental Safety 2006;63:1. Wania F, Dugani CB. Assessing the long-range transport potential of polybrominated diphenyl ethers: A comparison of four multimedia models. Environ Toxicol Chem 2003;22:1252. Weiss P. Organische Schadstoffe an entlegenen Waldstandorten Sloweniens und Kärntens. Berichte BE-195. Umweltbundesamt GmbH, Wien, September 2002. ISBN 3-85457- 616-1. 121 S. Zhu L, Hites RA. Brominated flame retardants in tree bark from North America.Environ Sci Technol 2006;40:3711. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 45

Distribution of dioxins, polychlorinated biphenyls and chlorinated pesticides in the alpine environment

Ivo Offenthaler1,*, Rodolfo Bassan2, Claudio Belis3, Gert Jakobi4, Manfred Kirchner4, Norbert Kräuchi5, Walkiria Levy‑Lopez4, Teresa Magnani3, Wolfgang Moche1, Karl‑Werner Schramm4, Isabella Sedivy5, Primož Simoncˇicˇ6, Maria Uhl1, Peter Weiss1

1 Umweltbundesamt GmbH, Spittelauer Lände 5, 1090 Vienna, Austria 2 ARPA Veneto, Via F. Tomea 5, I-32100 Belluno, Italy 3 ARPA Lombardia, Via Stelvio 35, I-23100 Sondrio, Italy 4 GSF, Institut für ökologische Chemie, Ingolstädter Landstraße 1, D-85764 Neuherberg, Germany 5 WSL, Abt. Waldökosysteme und ökologische Risiken, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland 6 Slovenian Forest Institute, Vecna pot 2, 1000 Ljubljana, Slovenia; *[email protected]

Riassunto Le concentrazioni di diossine e furani policlorurati, di bifenili policlorurati e di pesticidi organoclorurati sono misurabili negli ecosistemi forestali ancora anni dopo che il rilascio di queste sostanze è stato bandito. Le concentrazioni sono in generale più elevate nell’humus seguite dallo strato minerale del suolo e dagli aghi. Nessuno dei composti è presente su tutta l’area di studio e le concentrazioni presentano differenze longitudinali o latitudinali significative. Le informazioni ricavate da humus e aghi non sempre risultano congruenti. Con poche eccezioni le concentrazioni di inquinanti sono risultate più elevate nelle aree periferiche delle Alpi e si ritiene siano associate alla deposizione di composti legati al materiale particolato. In genere le correlazioni tra diversi inquinanti sono deboli con la sola eccezione delle diossine/furani e i PCB in una delle matrici studiate.

Abstract Polychlorinated dioxins and furans (PCDD/F), polychlorinated biphenyls (PCB) and organochloropesticides (OCP) were detectable in forest ecosystems of the Alps even at remote sites and years after the banned release of some of these substances. Pol- lutant concentrations were generally highest in humus, followed by mineral soil and needles. Neither of the compounds was spread evenly across the investigated region, and concentrations frequently showed significant longi- or latitudinal differences. Screening humus and needles, either with its particular history of exposure, did not always lead to congruent pollution maps. With few exceptions, however, needle and 46 Monitoraggio dei POPs sul territorio alpino humus pollution was higher in the peripheral regions of the alpine range. Pollutant patterns suggested wash- and fallout of particlebound compounds at the fringes of the study area. While in a given matrix there were numerous high correlations between PCDD/F and PCB congeners and/or totals, the majority of investigated pesticides was only weakly associated with the two earlier pollutant classes.

Introduction Part of the Interreg III B project MONARPOP was the investigation of POP loads in forest ecosystems of the Alps, as indicated by pollutant contents in needles of Nor- way spruce (Picea abies [L.] Karst.), humus and mineral soil. Among the numerous compounds included in the survey were polychlorinated dioxins and furans, poly- chlorinated biphenyls and most of the pesticides belonging to the “Dirty dozen” of the Stockholm Convention on Persistent Organic Pollutants. Norway spruce forests were selected for this project as the type of woodland that forms a major part of the investigated area. Moreover, their rough canopy structure and the waxy needle surface is a physical and chemical trap for airborne lipophilic compounds such as POPs. For- est humus at these altitudes mineralises only slowly and remains undisturbed by soil cultivation. Similarly, biomass removal is very limited so that humus and the topmost layer of mineral soil are likely to reflect POP input over the previous years. Needles, in contrast, were chosen as ecologically relevant indicators of the current situation.

Material and Methods Criteria for site selection and extent of the sampling network are described elsewhere in this volume-. Shoots of the current year were harvested from two vital adult Norway spruce (2-3 branches from the top 7th whirl each), transferred in airtight vessels, defoliated in liquid nitrogen and distributed to the laboratories in airtight vi- als. All manipulations between opening of the transport jars and redistributions were conducted in inert N2 atmosphere. After collecting the entire humus layer from ten regularly spaced 0.3×0.3 m pits, a soil core (0–10 cm mineral soil) was taken from each pit. Humus and mineral soil samples were freeze-dried, ground (humus) under cool conditions, homogenised and sieved. PCDD/F, PCB and OCP concentrations were determined with HRGC/HRMS.

Results and Discussion Humus generally showed the highest pollutant concentrations, followed by mineral soil and needles [Figure 1]. This pronounced difference is probably due to the enrich-

 Weiss P., R. Bassan, C. Belis, G. Jakobi, M. Kirchner, N. Kräuchi, W. Levy Lopez, W. Moche, I. Offenthaler, K.-W. Schramm, I. Sedivy, P. Simončič, M. Uhl, 2007. Synthesis of the findings of the project MONARPOP. This volume.  supplemenal information at: http://www.monarpop.at the previous years. Needles, in contrast, were chosen as ecologically relevant indicators of the current situation.

2. Material and Methods Criteria for site selection and extent of the sampling network are described elsewhere in this volume 1,2. Shoots of the current year were harvested from two vital adult Norway spruce (2-3 branches from the top 7th whirl each), transferred in airtight vessels, defoliated in liquid nitrogen and distributed to the laboratories in airtight vials. All manipulations between opening of the transport jars and redistributions were conducted in inert N2 atmosphere. After collecting the entire humus layer from ten regularly spaced 0.3×0.3 m pits, a soil core (0–10 cm mineral soil) was taken from each pit. Humus and mineral soil samples were freeze-dried, ground (humus) under cool conditions, homogenised and sieved. PCDD/F, PCB and OCP concentrations were determined with HRGC/HRMS.

3. Results and Discussion Humus generally showed the highest pollutant concentrations, followed by mineral soil and needles [Figure 1]. This pronounced D differenceistribution of dioxins, is probably PCB and chlorinated due to pesticides the enrichment in the alpine environment of lipophilic, poorly 47 degradable humus compounds in coniferous stands which retain POP contamination from needle litter and throughfall.

sum HCH isomers d.m.

1 − kg

g µ Figure 1: Comparison of HCH concentrations 0 2 4 6 8 10 (sum of alpha…epsilon-HCH) in humus, mineral humus soil needles soil and 0.5 year old Norway spruce needles at 18 remote alpine sites. Figure 1: Comparison of HCHment concentrations of lipophilic, (sum poorly of alpha…epsilon-HCH) degradable humus compounds in humus, in mineral coniferous soil stands which and 0.5 year old Norway spruceretain needles POP atcontamination 18 remote alpine from sites. needle litter and throughfall. After forming longi- or latitudinal groups of comparable site count [Figure 2a], sig- After forming longi- or latitudinalnificant groupsconcentration of comparable differences site were count found [Figure for a number 2a], significant of compounds [Figure concentration differences were2b]. Humus found concentrations for a number and ofalmost compounds all needle [Figureconcentrations 2b]. Humuswere highest in one concentrations and almost all needle concentrations were highest in one of the lateral groups. This suggests that the long-range atmospheric transport of POPs is intercepted by the alpine range.

1 Weiss P., R. Bassan, C. Belis, G. Jakobi, M. Kirchner, N. Kräuchi, W. Levy Lopez, W. Moche, I. Offenthaler, K.-W. Schramm, I. Sedivy, P. Simonþiþ, M. Uhl, 2007. Synthesis of the findings of the project MONARPOP. This volume. 2 supplemenal information at: http://www.monarpop.at Figure 2 a: Latitudinal grouping into northern (N), central (C) and southern (S) sites.

sum PCDD/F Figure 2 a: Latitudinal grouping into northern (N), central (C) and southern (S) sites.

north

central Figure 2 b: Total PCDD/F concentrations in hu- south mus, expressed in TEQWHO. Ranking the sites by this criterium showed a highly significant* imbalance between northern, central and south- 2 4 6 8 10 ern scores. −1 ng TEQWHO kg d.m. * Kruskal-Wallis, α ≤ 0.004, n=10/12/9 for northern/ central/southern sites

Figure 2 b: Total PCDD/F concentrations in humus, expressed in TEQWHO. Ranking the sites by this criterium showed a highly significant3 imbalance between northern, central and southern scores.

PCCD/F and PCB congener profiles differed regionally, and there were continguous zones of similar congener patterns [Figure 3]. For instance, humus from the south did not only have a higher fraction of furans but among these the heavier particlebound congeners dominated [Figure 3]. Note that zones of different congener pattern correspond approximately to the latitudinal bands depicted in Figure 2a. Different deposition regimen (temperature, precipitation, windspeed) and atmospheric loads are a likely cause for the geographical segregation between sites of similar pollutant pattern.

1

0.25

0.20

0.15

0.10

0.05

0.00 TCDF TCDD OCDF OCDD PeCDF HxCDF HpCDF PeCDD HxCDD HpCDD

4

0.20

0.15

0.10

0.05

0.00 TCDF TCDD OCDF OCDD PeCDF HxCDF HpCDF PeCDD HxCDD HpCDD

3 Kruskal-Wallis, α ≤ 0.004, n=10/12/9 for northern/central/southern sites Figure 2 a: Latitudinal grouping into northern (N), central (C) and southern (S) sites.

sum PCDD/F

north

Figure 2 a: Latitudinal grouping into northern (N), central (C) and southerncentral (S) sites. sum PCDD/F south north 2 4 6 8 10

−1 ng TEQ kg d.m. central WHO

48 Monitoraggio dei POPs sul territorio alpino southFigure 2 b: Total PCDD/F concentrations in humus, expressed in TEQWHO. Ranking the sites by this criterium showed a highly significant3 imbalance between northern, central and southern scores.

PCCD/F2 and 4 PCB 6 congener 8 10 profiles differed regionally, and there were continguous zones of −1 similar congenerng TEQ patterns kg d.m. [Figure 3]. For instance, humus from the south did not only have a higher fraction of furansWHO but among these the heavier particlebound congeners dominated [Figure 3]. Note that zones of different congener pattern correspond approximately to the latitudinal bands depicted Figure 2 b: Total PCDD/F concentrations in humus, expressed in TEQ . Ranking the sites by this in Figure 2a. Different deposition WHOregimen (temperature, precipitation, windspeed) and atmospheric criterium showed a highly significant3 imbalance between northern, central and southern scores. loads are a likely cause for the geographical segregation between sites of similar pollutant pattern.

PCCD/F and PCB congener profiles differed regionally, and there were continguous zones of 1 similar congener patterns [Figure 3]. For instance, humus from the south did not only have a higher 0.25 0.20 fraction of furans but among these the heavier particlebound congeners dominated [Figure 3]. Note 0.15 that zones of different congener pattern correspond approximately to the latitudinal bands depicted 0.10 0.05 in Figure 2a. Different deposition regimen (temperature, precipitation, windspeed) and atmospheric 0.00 TCDF loads are a likely cause for the geographical segregation between sites of similar pollutant pattern. TCDD OCDF OCDD PeCDF HxCDF HpCDF PeCDD HxCDD HpCDD

1 4

0.25 0.20 0.20 0.15 0.15 0.10 0.10 0.05 0.05

0.00 0.00 TCDF TCDD OCDF OCDD TCDF TCDD OCDF PeCDF HxCDF HpCDF OCDD PeCDD HxCDD HpCDD PeCDF HxCDF HpCDF PeCDD HxCDD HpCDD

4 Figure 3: Sites of the same number (1…4) have similar PCDD/F congener patterns in humus. The aver- 0.20 3 age congener pattern of clusters 1 (north) and 4 (south) is shown on the right. Kruskal-Wallis, α ≤ 0.004, n=10/12/90.15 for northern/central/southern sites of0.10 the lateral groups. This suggests that the long-range atmospheric transport of POPs Figure 3: Sites of the same numberis0.05 intercepted (1…4) byhave the similar alpine PCDD/Frange. congener patterns in humus. The average congener pattern of clustersPCCD/F0.00 1 (nort and h)PCB and congener 4 (south) profiles is shown differed on the regionally,right. and there were continguous TCDF TCDD OCDF OCDD PeCDF HxCDF HpCDF PeCDD HxCDD zones of similar congenerHpCDD patterns [Figure 3]. For instance, humus from the south PCDD/F and PCB pollution weredid notassociated only have as showna higher by fraction several ofstrong furans and but significant among these correlations the heavier particle- between members of the two pollutant classes [Figure 4]. 3 Kruskal-Wallis, α ≤ 0.004, n=10/12/9 for northern/central/southern sites 3.0 d.m.]

2.5 1 − kg 2.0 g µ [

TE 1.5

PCB 1.0 Σ

0.5

0.1 0.2 0.3 0.4 0.5 0.6 0.7 Figure 4: Highly significant correlation be- −1 Σ PCDD F [µg kg d.m] tween total PCDD/F and content of 12 dioxin- like PCB (rPearson=0.86, α ≤ 0.001) Figure 4: Highly significant correlation between total PCDD/F and content of 12 dioxinlike PCB (rPearson=0.86, α ≤ 0.001)

4. Conclusions Dioxinlike compounds and chlorinated pesticides are traceable even in remote forest sites of the alps. The detection of the highly toxic chemicals in needles shows that these substances are still around – despite rigorous emission control in large parts of Europe and the US (in the case of PCB already since the 1970ies). The degree of pollution and the chemical patterns found in environ- mental samples changes across the investigated area in a way which indicates that the Alps form a climatical and orographical barrier for atmospheric POP transport. As a corollary, mountainous forests can be expected to receive increased inputs of airborne POPs. Even within the narrow range of substances described here, chemicals of different source and properties were closely associated. This demonstrates how POP pollution is an issue of multiple contamination with possibly synergistic amplification of adverse effects.

5. Acknowledgements The project MONARPOP was funded by the EU in the framework of the Interreg III B programme “Alpine space” and received major contributions from the Austrian Ministry for Agriculture, Forestry, Environment and Water Ressources; several Austrian provinces (Burgenland, Niederösterreich, Oberösterreich, Kärnten, Steiermark, Vorarlberg, Wien); the Bavarian State Ministry of the Environment, Public Health and Consumer Protection; the Swiss Federal Office for the Environment; the Italian Fondo Rotazione; together with substantial contributions in money or in kind from the following: Austrian Environment Agency (Umweltbundesamt), German Federal Environment Agency, German National Research Center for Environment and Health (GSF), Italian ARPA Lombardia and ARPA Veneto, Slowenian Forestry Institute, Swiss Federal Institute for Forest, Snow and Landscape Research. Distribution of dioxins, PCB and chlorinated pesticides in the alpine environment 49 bound congeners dominated [Figure 3]. Note that zones of different congener pattern correspond approximately to the latitudinal bands depicted in Figure 2a. Different deposition regimen (temperature, precipitation, windspeed) and atmospheric loads are a likely cause for the geographical segregation between sites of similar pollutant pattern. PCDD/F and PCB pollution were associated as shown by several strong and significant correlations between members of the two pollutant classes [Figure 4].

Conclusions Dioxinlike compounds and chlorinated pesticides are traceable even in remote forest sites of the alps. The detection of the highly toxic chemicals in needles shows that these substances are still around – despite rigorous emission control in large parts of Europe and the US (in the case of PCB already since the 1970ies). The degree of pollution and the chemical patterns found in environmental samples changes across the investigated area in a way which indicates that the Alps form a climatical and oro- graphical barrier for atmospheric POP transport. As a corollary, mountainous forests can be expected to receive increased inputs of airborne POPs. Even within the nar- row range of substances described here, chemicals of different source and properties were closely associated. This demonstrates how POP pollution is an issue of multiple contamination with possibly synergistic amplification of adverse effects.

Acknowledgements The project MONARPOP was funded by the EU in the framework of the Interreg III B programme “Alpine space” and received major contributions from the Austrian Ministry for Agriculture, Forestry, Environment and Water Ressources; several Aus- trian provinces (Burgenland, Niederösterreich, Oberösterreich, Kärnten, Steiermark, Vorarlberg, Wien); the Bavarian State Ministry of the Environment, Public Health and Consumer Protection; the Swiss Federal Office for the Environment; the Italian Fondo Rotazione; together with substantial contributions in money or in kind from the following: Austrian Environment Agency (Umweltbundesamt), German Federal Environment Agency, German National Research Center for Environment and Health (GSF), Italian ARPA Lombardia and ARPA Veneto, Slowenian Forestry Institute, Swiss Federal Institute for Forest, Snow and Landscape Research.

Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 51

The deposition of pesticides into Alpine forest ecosystems

Kirchner M1, Jakobi G1, Levy W1, Henkelmann B1, Bernhöft S1, Niklaus A1, R. Bassan2, C. Belis3, N. Kräuchi5, T. W. Moche6, I. Offenthaler6, B. Perthen-Palmisano4, P Schröder1, I. Sedivy5, P. Simoncˇicˇ7, M. Uhl6, Schramm K-W1, P. Weiss6,

1 GSF-National Research Centre for Environment and Health, Ingolstädter Landstrasse 1, D-85764 Neuherberg, Germany, [email protected] 2 Regional Agency for Environmental Prevention and Protection of Veneto 3 Regional Agency for Environmental Protection of Lombardia 4 Austrian Ministry for Agriculture, Forestry, Environment and Water Resource 5 WSL-Swiss Federal Institute for Forest, Snow and Landscape Research 6 Federal Environment Agency Ltd. - Austria 7 Slovenian Forestry Institute

Riassunto Lo strato di humus e gli aghi di conifera di circa 50 siti remoti distribuiti lungo tutto l’arco alpino di Austria, Svizzera, Baviera, Veneto, Lombardia e Slovenia sono stati studiati per la determinazione del loro contenuto di pesticidi. Nonostante le influenze dovute alle caratteristiche orografiche, vegetazionali e pedologiche e alle differenze meteorologiche, è stato possibile misurare i gradienti regionali della contaminazione del suolo con pesticidi i quali presentano un andamento decrescente dall’estremità nord agli Alpi Centrali. Si osserva inoltre che gli aghi e l’humus si comportano in modo leggermente diverso. Per ultimo è stato riscontrato che alcuni pesticidi mostrano un incremento delle concentrazioni con la quota.

Abstract Humus layer and spruce needles of approximately 50 remote forest sites spread all over the Alpine regions of Austria, Switzerland, Bavaria, Veneto, Lombardy and Slovenia were investigated for their concentrations of pesticides. Despite the numerous influ- ences regarding orographic aspects, forest and soil characteristics and meteorological differences we measured regional gradients of soil contamination with pesticides with increasing concentrations from the Northern fringe to Central Alps. Slightly different features can be observed if the content in spruce needles are considered. For some pesticides a vertical increase from valley to higher is evident. 52 Monitoraggio dei POPs sul territorio alpino

Table 1: Factors influencing the deposition of POPs into mountainous forests forest characteristics physico - chemical landscape forest soil properties properties of pesticide characteristics and meteorological factors orographic roughness substance orographic roughness humus content of terrain of terrain forest type and species gas or particle slope exposition and humus depth inclination position in forest atmospheric residence air temperature organic carbon content (centre or edge) time/half-life-time in atmosphere leaf area index volatility inversion layers water content needle characteristics solubility precipitation pH value wax content biotic and abiotic radiation soil temperature decomposition

Introduction Remote areas like the Arctic and the Alps are likely to serve as a sink of organic per- sistent contaminants. As a consequence of the lipophilic characteristics organochlorine compounds are attracted by plant stands and tend to be retained in the organic material of their leaves and needles. Due to their high perstence they can be stored in humus layer and mineral soil for many years. Furthermore forests are effective filters for all pollutants for their roughness (Table 1). In the framework of the MONARPOP project humus and needles investigations deliver the basis information of ecosystem contami- nation by POPs in Alpine Ecosystems. Whereas spruce forest humus layer content of pesticides delivers the information about the last decades’ deposition, spruce needles as a part of living organisms comprise the information about recent contamination.

Materials e Methods The different sites and vertical profiles were choosen carefully in the way that the high number of possible natural and anthropogenic interfering factors could be minimized. In the case of the soil sampling 10 sub-samples of the humus layer were taken at each of the sites with a frame and composited to one sample. The needle sampling was performed by a tree climber; current year Norway spruce needles were taken from the range of the seventh branch whorl from the top of the trees. As main organochlorine substances have been choosen DDX, HCH’s , HCB, aldrin, dieldrin and mirex to be analyzed in soil and spruce needles. Sampling and analytical procedures have been desribed in Kirchner et al. (2007); the description of different statistical tests to in- dentify differents between the regional groups can be found at Weiss et al. (2000). The deposition of pesticides into Alpine forest ecosystems 53

Figure 1: Distribution of p,p‘-DDT in humus layer in µg/kg

Results and discussion Figure 1 shows the distribution of p,p’-DDT as measured at the singular sites. As the statistical tests reveal, the concentrations of some pesticides in humus layers dif- fer from one to another regional latitudinally and longitudinally defined sub-group. Mainly p,p’DDE and γ-HCH decrease from Northern to Central Alps, longitudinal differences between eastern group and western group could not be identified with statistical significance. Adopting a grouping which considers more climatological aspects the differences between Alpine fringe regions and the more protected centre of the Alps, where precipitation is lower and insolation is higher, may result in steeper gradients.

Table 2: Literature values for humus and needles 54 Monitoraggio dei POPs sul territorio alpino

On behalf of the needle’s concentrations there are tendentially higher concentrations of some DDX in the Southern Alps. It seems that current contamination in Southern Alps may be higher than in Northern Alpine regions. There is a marked increase of some substance, which are transported into the Alps by long-range transport, from bottom of valleys to higher slope levels where input is higher and degradation is lower. The increase is frequent for Bavarian, Austrian and Swiss hight profiles. Italian and slowenian hight profile do not show this increase. The reasons for that are unknown; the remote situation, the complex orographic terrain, different height of inversion layers and possible local sources may be responsible. Generally the lower number of sites in the Southern Alps – only Lombardy, Veneto and Slovenia participated – lead to the result that the representativity of results related to Southern Alps is reduced. A first conclusion of the project regarding the contamination of pesticides is that the content in needles of organo-chlorine compounds may have been reduced in the last decades, which is parallel to the reduction of emissions in Europe, whereas many substances are still stored in the humus layer and possibly present for food chains (table 2).

Literature Kirchner, B., Henkelmann, B., Jakobi, G., Levy Lopez, W., Pfister, G., Niklaus, A., Kotalik, J., Fischer, N., Pandelova, M., Kocsis, G., Bernhöft, S., Schramm, K.-W., 2007: Deposition persistenter schadstoffe im Bayerisch-Südböhmischen Grenzgebirge. Abschlussbericht Bayerisches Staatsministerium für Umwelt, Gesundheit und Verbraucherschutz. Weiss, P., Lorbeer, G., Scharf, S., 2000: Regional aspects and statistical characterisation of the load with semivolatile compounds at remote Austrian forest sites. Chemosphere 40, 1159-1171. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 55

Novel Diagnostic Tools for POPs in Alpine Areas

Levy W1, Henkelmann B1, Pfister G1, Bernhöft S1, Niklaus A1, Kirchner M1, Jakobi G1, R. Bassan3, C. Belis4, T. Jakl7, N. Kräuchi8, T. Magnani3, W. Moche9, I. Offenthaler9, B. Perthen-Palmisano7, P Schröder1, I. Sedivy8, P. Simoncˇicˇ10, M. Uhl9, P. Vannini4, U. Vilhar10, P. Weiss9, Schramm K-W1,2

1GSF-National Research Centre for Environment and Health, Ingolstädter Landstrasse 1, D-85764 Neuherberg, Germany, [email protected] 2TUM-Technische Universität München, Department für Biowissenschaftliche Grundlagen Weihenstephaner Steig 23, D-85350 Freising, Germany, [email protected] 3Regional Agency for Environmental Prevention and Protection of Veneto 4Regional Agency for Environmental Protection of Lombardia 7Austrian Ministry for Agriculture, Forestry, Environment and Water Resource 8WSL-Swiss Federal Institute for Forest, Snow and Landscape Research 9Federal Environment Agency Ltd. - Austria 10Slovenian Forestry Institute

RIASSUNTO Il presente articolo illustra due nuove metodologie che sono state applicate nell’ambito del Progetto MONARPOP per il monitoraggio degli inquinanti org- anici persistenti (POPs) nelle aree alpine.

ABSTRACT Two novel methodologies have been applied for environmental monitoring of per- sistent organic pollutants (POP) in alpine areas. First a modern passive sampling

Figure 1: Experimental setup for deploy- Figure 2: Scheme of the bioassay for the determination of ment of the SPMD at remote field sites dioxin-like compounds in soil and other complex matrices containing 25 SPMD subsequently sam- by determination of CYP1A1 expression. pled and renewed twice a year. Oral contribution MONARPOP Bormio 2007

Organic pollutants as PCDD/F, PCBs and some PAHs are known to be environmentally persistent and to bind to the aryl hydrocarbon receptor (AhR) and thus, induce the cytochrome CYP1A synthesis which then can cause excess oxidative stress in living organisms. This induction can be indirectly measured by the 7-Ethoxyresorufin O-deethlylase (EROD) bioassay (Hofmaier 1999, Schwirzer et al 1996). The ability to elicit response when the cytochrome CYP1A is induced makes this bioassay suitable as a screening tool of dioxin-like chemicals capable of AhR binding. This micro-bioassay evaluates the overall potency as AhR modulators of the sample constituents and consequently is related to the potential toxicity of such chemicals in a complex mixture when an organism is exposed to them. In air and spruce needles the biological activity of dioxin-like activities56 is currently low, but soil exhibits prominent Monitoraggioactivities anddei POPs also sul show territorio some alpino altitude related features. Oral contribution MONARPOP Bormio 2007 Oral contribution MONARPOP Bormio 2007 EROD determinations in humus and mineral soil technique (SPMD)EROD determinations especially in humus suit and- mineral soil 30 able for areas30 without infrastructure EROD 24 hrs EROD 24 hrs Organic pollutants as PCDD/F, PCBs and some PAHs are known to be environmentally such as electricEROD power 72 hrs etc. have been EROD 72 hrs dry soil) dry

Organic pollutants as PCDD/F, PCBs and some PAHs are known to be environmentally -1 dry soil) dry successfully20 employed (Fig. 1). persistent and to bind to the aryl hydrocarbon receptor-1 20 (AhR) and thus, induce the cytochrome CYP1A synthesispersistent which and to then bind can to cause the aryl excess hydrocarbon oxidative receptor stress in (AhR) living and orga thus,nisms. induce This the induction cytochrome Second the question about the pres- CYP1A synthesis which then can cause excess oxidative stress in living organisms. This induction can be indirectly measured by the 7-Ethoxyresorufin O-deethlylase (EROD) bioassay (Hofmaier ence of known and unknown toxic can be indirectly measured by the 7-Ethoxyresorufin10 O-deethlylase (EROD) bioassay (Hofmaier 10 1999, Schwirzer1999, Schwirzer et al 1996). et al The 1996). ability The to ability elicit to response elicit response when thwhene cytochrome the cytochrome CYP1A CYP1A is induced is induced substances has been elaborated makes this bioassay suitable as a screening tool of dioxin-like chemicals capable of AhR binding. by bio-analytical investigations makes this bioassay suitable as a screening tool of dioxin-like chemicals capable of AhR binding. TCDD(ngTE-value kg TE-value (ng TCDD(ngTE-value kg This micro-bioassay evaluates the overall potency as AhR modulators of the sample constituents 0 This micro-bioassay evaluates the overall potency 0as AhR modulators of the sample constituents(EROD) complementary to chemi- 1134 1470 1779 1134 1470 1779 898 1282 1510 898 1282 1510 and consequently is related to the potential toxicity of such chemicals in a complex mixture when cal analysis (Fig. 2).Humus Mineral soil and consequently is related to the potential toxicity of such Humus chemicals in a complex Mineral mixturesoil when an organisman organismis exposed is exposedto them. to In them. air and In airspru andce spru needlesce needles the biological the Altitude biological profile activity (m) activity of dioxin-like of dioxin-like Organic pollutants asAltitude PCDD/F, profile (m) activities isactivities currently is currentlylow, but low,soil exhibitsbut soil exhibits prominent prominentFigure Figureactivities 3: activities 3:EROD and EROD responsealso and responseshow also of the showsome altitude of some altitude the profile altitude altitude related Rauris related PCBs Figure and some 4: ERODPAHs are response known to of the altitude features. features. profilein Austria Rauris in Austria be environmentallyprofile Wechsel persistent in Austria and to bind to the aryl hydrocarbon recep- EROD determinations in humus and mineral soil EROD determinationsEROD determinations in humus and in mineralhumus and soil mineral soil In orderEROD determinations to monitor in humus persistent and mineral organicsoil 20 30 tor (AhR) and thus, induce the cyto- 30 30 pollutants30 in a forested mountainous area, 18 EROD 24 hrsEROD 24 hrs chrome CYP1A synthesis which then EROD 24 hrs EROD 24 hrs 16 humus, mineralEROD soil, 72 hrs spruce needles and air EROD 72 hrs EROD 72 hrs can cause excess oxidative stress in EROD 72 hrs soil) dry 14 dry soil) dry -1 dry soil) dry

-1 20 -1 dry soil) dry 20 were20 sampled and analysed by analytical and -1 20 living12 organisms. This induction bio analytical methods. Quantitative can10 be indirectly measured by the determinations of organic persistent pollutants 8 10 10 7-Ethoxyresorufin O-deethlylase 10 10 were performed by HRGC-HRMS (El-Kady 6 et al. 2007). Then, a total toxic equivalent hours 72 EROD TEQ (EROD) bioassay (Hofmaier 1999,

TE-value (ng TCDD(ngTE-value kg 4 TE-value (ng TCDD(ngTE-value kg

TE-value (ng TCDD(ngTE-value kg Schwirzer et al 1996). The ability to

TE-value (ng TCDD(ngTE-value kg 0 0 value for each sample was calculated using 2 0 898 1282 1510 898 1282 1510 0 1134 1470 1779 1134 1470 1779 elicit response when the cytochrome relative potency factorsHumus and the concentration Mineral soil 0 Humus Mineral soil 898 1282 1510 898 1282 1510 1134 1470 1779 1134 1470 1779 0 2 4 6 8 10 12 14 16 18 20 Humus Altitude profile Mineral (m) soil Humus Altitude profile Mineral (m) soil of the single compounds. These calculated CYP1A is induced makes this bio- Altitude profile (m) Altitude profile (m) WHO-TEQ GC MS Figure 3: EROD response of the altitude values Figure from 4:chemical EROD analysis response were of the then altitude assay suitable as a screening tool Figure 3:profile EROD Rauris response in Austria of the altitude FigurecomparedFigure 4:profile 4:EROD to EROD the Wechselresponse bioassay response of in the Austriaresults. altitude of profileIn the this altitude Wechselway, of dioxin-like chemicals capable of profile Rauris in Austria profilewein Austria determined Wechsel similaritiesin Austria and differences AhR binding.Figure This 5: micro-bioassay TE-EROD values after 72 In order to monitor persistent organic between20 both methods. Based on this, a better evaluateshours the incubation overall potency and as WHO-TEQ AhR values In orderpollutants to monitor in a forested persistent mountainous organic area,20 data18 interpretation is achieved in order to modulators(WHO, of 1998) the sample calculated constitu as- the sum of 16 pollutants humus, in a forested mineral soil, mountainous spruce needles area, and 18 air obtain a sound environmental assessment of ents PCDD/Fand consequently and PCB is related WHO-TEQ to from the humus, mineralwere sampled soil, spruce and analysed needles by and analytical air and16 the 14 samples and to exclude or confirm the analytical determinations of humus 12 the potential toxicity of such chemi- were sampledbio and analytical analysed by methods. analytical Quantitativeand 14 presence of unknown toxic dioxin-like samples from Norway spruce forest sites. 10 cals in a complex mixture when an determinations of organic persistent pollutants12 bio analytical methods. Quantitative compounds.8 EROD values are the mean of 3 replicates. 10 organism is exposed to them. In air were performed by HRGC-HRMS (El-Kady 6

determinations of organic persistent pollutants hours 72 EROD TEQ 8 and spruce needles the biological ac- et al. 2007). Then, a total toxic equivalent 4 were performed by HRGC-HRMS (El-Kady 6

TEQ EROD 72 hours hours 72 EROD TEQ tivity of dioxin-like activities is cur- value for each sample was calculated using 2 et al. 2007). Then, a total toxic equivalent 4 relative potency factors and the concentration 0 rently low, but soil exhibits promi- value for each sample was calculated using 0 2 4 6 8 10 12 14 16 18 20 of the single compounds. These calculated2 nent activities and also show some relative potency factors and the concentration 0 WHO-TEQ GC MS values from chemical analysis were then0 2 4 6 8 10 12 14 16 18 20 altitude related features. of the single compounds. These calculated Figure 5: TE-EROD values after 72 hours incubation compared to the bioassay results. In this way, WHO-TEQ GC MS In order to monitor persistent organ- values from chemical analysis were then and WHO-TEQ values (WHO, 1998) calculated as the we determined similarities and differences Figure 5: TE-EROD values after 72ic pollutants in a forested mountain- compared to the bioassay results. In this way, sum of PCDD/F and PCB WHO-TEQ from the analytical between both methods. Based on this, a better hours incubation and WHO-TEQ valuesous area, humus, mineral soil, spruce we determineddata interpretation similarities is and achieved differences in order to determinationsFigure(WHO, 5:of humus 1998) TE-EROD samples calculated from values Norway as after the spruce sum 72 of forest sites. EROD values are the mean of 3 replicates. needles and air were sampled and between bothobtain methods. a sound Based environmental on this, a assessmentbetter of hours PCDD/F incubation and and PCB WHO-TEQ WHO-TEQ values from the data interpretationthe samples is and achieved to exclude in order or confirm to the (WHO,analytical 1998) calculated determinations as the sumof humus of obtain a soundpresence environmental of unknown assessment toxic dioxin-like of PCDD/Fsamples and from PCB Norway WHO-TEQ spruce from forest the sites. the samplescompounds. and to exclude or confirm the analyticalEROD determinationsvalues are the mean of of 3 replicates. humus presence of unknown toxic dioxin-like samples from Norway spruce forest sites. compounds. EROD values are the mean of 3 replicates. Novel Diagnostic Tools for POPs in Alpine Areas 57

OralOral analysedcontribution contribution by MONARPOPanalytical MONARPOP and Bormio bioBormio analytical 20072007 methods. Quantitative determinations of organic persistent pollutants were performed by HRGC-HRMS (El-Kady et al. 2007). PassiveThen,Passive a sampler,total sampler, toxic so equivalent so called called ‘semi-permeable ‘semi-permeable value for each membranesample membrane was devices’ devices’calculated (SPMD), (SPMD), using were relative were deployed deployed simultaneouslysimultaneouslypotency factors at remoteat remote and mountainthe mountain concentration areas areas for of½ atheandnd 1 single1 ½ ½ years years compounds. (Levy (Levy et al.et Theseal.2006, 2006, Levy calculated Levy et al. et 2007). al. 2007). ActiveActivevalues sampling sampling from measurements chemical measurements analysis were were werealso also performedthen compared at at the the end toend ofthe of the bioassaythe passive passive samplingresults. sampling Inperiods. this periods. 400 400 02.05.05-09.11.05 02.05.05-07.06.06 )

-1 02.05.05-09.11.05 02.05.05-07.06.06 ) -1 300 300

200

200

100

100 Concentration of Phenanthrene of Phenanthrene (ng SPMDConcentration 0 830 1030 1230 1450 1650 2650 Concentration of Phenanthrene of Phenanthrene (ng SPMDConcentration 0 Altitude profile (m a.s.l.) 830 1030 1230 1450 1650 2650 Fig. 6: Phenanthrene concentrations Altitudeat an altitude profile profile (m a.s.l.) for SPMDs ½ year and 1 year exposure Fig. 6: Phenanthrene concentrations at an altitude profile for SPMDs ½ year and 1 year exposure

Fig. 6:6000 Phenanthrene concentrations at an altitude profile for SPMDs ½ year and 1 year exposure Period 1 Period 2 Period 3 Sum of periods 1 + 2

5000 triolein) triolein)

6000-1

4000 Period 1 Period 2 Period 3 Sum of periods 1 + 2

5000 triolein) triolein)

-1 3000

4000 2000

3000 1000 Concentration of 4,4´ DDT (ng kg (ng DDT 4,4´ of Concentration

2000 0 AT- AT- AT- AT- CH- CH- CH- CH- DE- DE- DE- DE- DE- DE- DE- DE- DE- DE- IT- 47-1 47-3 47-5 47-6 01-2 01-5 01-7 01-8 21-1 21-2 21-3 21-4 21-5 21-6 23-1 23-2 23-3 23-4 04-2

1000 Altitude profiles

Concentration of 4,4´ DDT (ng kg (ng DDT 4,4´ of Concentration Figure 7: 4,4´DDT concentration at altitude profiles for SPMD ½ and 1 year exposure. Period 1: ½ Figureyear0 exposure, 7: 4,4´DDT Period concentration 2: ½ year exposure at altitude and Period profiles 3: 1 year for SPMDexposure. ½ and 1 year exposure. AT- AT- AT- AT- CH- CH- CH- CH- DE- DE- DE- DE- DE- DE- DE- DE- DE- DE- IT- Period47-1 1: ½47-3 year47-5 exposure,47-6 01-2 Period01-5 01-7 2: 01-8½ year exposure21-1 21-2 21-3 and21-4 Period21-5 21-6 3: 1 year23-1 exposure.23-2 23-3 23-4 04-2

Concentrations in passive samplers obtainedAltitude by profiles HRGC-HRMS were compared with the atmospheric concentrations for some PAH, PCB and chlorinated pesticides. From the compounds Figure 7: 4,4´DDT concentration at altitude profiles for SPMD ½ and 1 year exposure. Period 1: ½ year exposure, Period 2: ½ year exposure and Period 3: 1 year exposure.

Concentrations in passive samplers obtained by HRGC-HRMS were compared with the atmospheric concentrations for some PAH, PCB and chlorinated pesticides. From the compounds 58 Monitoraggio dei POPs sul territorio alpino way, we determined similarities and differences between both methods. Based on this, a better data interpretation is achieved in order to obtain a sound environmental assessment of the samples and to exclude or confirm the presence of unknown toxic dioxin-like compounds. Passive sampler, so called ‘semi-permeable membrane devices’ (SPMD), were de- ployed simultaneously at remote mountain areas for ½ and 1 ½ years (Levy et al. 2006, Levy et al. 2007). Active sampling measurements were also performed at the end of the passive sampling periods. Concentrations in passive samplers obtained by HRGC-HRMS were compared with the atmospheric concentrations for some PAH, PCB and chlorinated pesticides. From the compounds analysed, those with the lowest affinity to the gas phase were still being sequestered without reaching an equilibrium state between the atmosphere and the SPMD. Key findings with these technique was the increase of selected pesticides such as DDT with altitude (Fig. 6) as a result of long range transport and a decrease of PAH (Fig. 7) with altitude in consequence of the emissions due to traffic and domestic heating at the bottom of the alpine valleys.

Biobliography El-Kady, A., Abdel-Wahhab, M., Henkelmann, Belal, M., Morsi, K., Galal, S., Schramm, K.-W. 2007, Polychlorinated biphenyl, polychlorinated dibenzo-p-dioxins and polychlo- rinated dibenzofuran residues in sediments and fish of the river Nile in the Cairo region. Chemosphere 68, 1660-1668. Hofmaier, A.M. 1999. “Evaluierung eines Testsystems auf zellulärer Basis zur Detektion von Dioxinen und verwandten Verbindungen in Matrizes der metallrecyclenden Industrie sowie die Bestimmung der gentoxischen Wirkung dieser Proben“ PhD Thesis. Technical University Munich. Levy, W., Henkelmann, B., Pfister G., Bernhöft, S., Niklaus, A., Kirchner M., Jakobi, G, Bassan R., Belis, C., Jakl, T., Kräuchi, N., Magnani, T., Perthen-Palmisano B., Schröder, P., Schrott, H., Sedivy, I., Simončič, P., Vannini, P., Vilhar, U., Schramm, K.-W., 2007. Semipermeable membrane devices (SPMD) as passive samplers: Data interpretation regarding exposure time. Organohalogen Compounds 69, 599-603 Levy, W., Henkelmann, B., Pfister, G., Kirschner, M., Jakobi, G., Niklaus, A., Kotalik, J., Bernhöft, S., Schramm, K.-W., 2006. Comparison of PAH concentrations in Semipermeable Membrane Devices, low volume active sampler, and spruce needles. Organohalogen Compounds 64, 45-48. Schwirzer, S.M.G., Hofmaier, A.M., Kettrup, A., Nerdinger, P.E., Schramm, K.-W., Thoma, H., Wegenke, M., Wiebel, F.J., 1996. Establishment of a simple cleanup procedure and bioassay for determining 2,3,7,8-Tetrachlorodibenzo-p-dioxin toxicity equivalents of environmental samples. Ecotoxicology and Environmental Safety 41, 77-82.

Acknowledgements MONARPOP is funded by the EU Interreg III B Alpine Space Programme (Alpine Space) and by the participating national partners. Additionally, we would like to thank the Swiss Federal Office for the Environment (FOEN - BAFU) for financial support. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 59

Novel ambient air sampling techniques adapted for the needs of project MONARPOP

Wolfgang Moche1,*, Rodolfo Bassan2, Claudio Belis3, Saverio Iozza7, Gert Jakobi4, Manfred Kirchner4, Norbert Kräuchi5, Walkiria Levy‑Lopez4, Teresa Magnani3, Ivo Offenthaler1, Karl‑Werner Schramm4, Isabella Sedivy5, Primož Simoncˇicˇ6, Maria Uhl1, Peter Weiss1

1 Umweltbundesamt GmbH, Spittelauer Lände 5, 1090 Vienna, Austria 2 ARPA Veneto, Via F. Tomea 5, I-32100 Belluno, Italy 3 ARPA Lombardia, Via Stelvio 35, I-23100 Sondrio, Italy 4 GSF, Institut für ökologische Chemie, Ingolstädter Landstraße 1, D-85764 Neuherberg, Germany 5 WSL, Abt. Waldökosysteme und ökologische Risiken, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland 6 Slovenian Forest Institute, Vecna pot 2, 1000 Ljubljana, Slovenia, 7 EMPA Materials Science & Technology, Überlandstrasse 129, CH-8600 Dübendorf, Switzerland; * [email protected]

Riassunto Una nuova tecnica di campionamento dell’aria è stata sviluppata nell’ambito del progetto MONARPOP, che offre l’opportunità di attribuire le concentrazioni misurate di diverse POPs a quattro regioni di provenienza predefinite e ritenute rilevanti per il territorio alpino. I campionatori d’aria e di deposizioni atmosferiche sono stati installati in tre siti di campionamento in alta quota: Weissfluhjoch (CH; 2663 m s.l.m.), Zugspitze (D; 2650 m s.l.m.) e Sonnblick (A; 3106 m s.l.m.). Dall’avvio del progetto il campionamento è stato fatto per cinque periodi di durata trimestrale. Per molti dei POPs studiati non è stato possibile identificare le aree di provenienza predominante, tuttavia le differenze stagionali sono evidenti. I livelli di concentrazione nell’aria e le deposizioni si trovano nello stesso intervallo di quelli misurati nelle zone rurali di fondovalle ed indicano il trasporto a lunga distanza di PCDD/F e PCBs verso questi siti.

Abstract A novel ambient air sampling technique has been developed within the project MON- ARPOP, which affords the opportunity to attribute measured concentrations of dif- ferent POPs to four predefined source regions important for the alpine area. Such 60 Monitoraggio dei POPs sul territorio alpino ambient air samplers and in addition bulk deposition samplers have been installed at three high altitude sampling sites Weissfluhjoch (CH; 2663 m), Zugspitze (D; 2650 m) and Sonnblick (A; 3106 m). Since the start of the project sampling was done for five trimonthly periods. For most of the analysed POPs no predominant source region could be detected so far, but clear seasonal differences are obvious. The concentration levels for ambient air and deposition as well are in the same range as those measured in the rural lowlands indicating long-range transport of PCDD/F and PCBs to these sites.

Introduction The evaluation of the influence of long range transport of POPs to the alpine region was one of the main issues of Interreg III B Project MONARPOP. An additional question in this context was to detect which source regions are possibly responsible for POPs immissions in the alpine region. To answer these questions it was planned to carry out ambient air measurements for various POPs with the additional requirement to attribute the measured concentrations to source regions. This meaned relating the sampled air masses to their way to the sampling sites and possible influences by POP emissions during this way.

Material and Methods Ambient air measurements for POPs are a two stage process, a long-time sampling with durations from days to weeks at the sampling site and a more or less complicated chemical analysis in a well equipped laboratory. Therefore in contrast to the onsite and online monitoring of gaseous pollutants like NOx the attribution of POP concen- trations to air masses is not possible after the measurement in most of the cases due to changing weather situations during the long sampling periods. A solution is the predefinition of possible source regions followed by the source region specific sampling of air masses arriving at the sampling site. The region specific sampling allows to correlate between measured concentrations and source regions.Three high altitude measurement sites have been installed at three mountain summits which provide well equipped infrastructures of meteorological stations necessary for the operation of POP samplers. The three sites are Weissfluhjoch (CH; 2663 m), Zugspitze (D; 2650 m) and Sonnblick (A; 3106 m) which are shown in pictures 1, 2 and 3. All these three sites are well-staffed all year round to ensure daily checkup of the sampling equipment and prompt maintenance in the case of malfunctions. Existing sampling techniques for POP had to be modified for the planned investi- gations. Ambient air samplers have been equipped with four filter cartridges each attributed to one source region. The selection of the corresponding filter cartridge was done by remote control based on meteorological trajectory forecasts. All filter Novel ambient air sampling techniques adapted for the needs of project MONARPO 61

1 2

3

Picture 1: Meteorological observatory “Schneefernerhaus” at Zugspitze, 2650m, Germany Picture 2: Snow and avalanche research station at Weissfluhjoch, 2663m, Switzerland Picture 3: Meteorological observatory at Sonnblick, 3106m, Austria cartridges and moving parts had to be heated due to the rough weather conditions at the selected sites. Only 50 days per year show temperatures above 0 °C. In addition to the ambient air samplers deposition samplers for bulk deposition have been installed at the three high altitude sites. The samplers are built according to DIN 19739-1, “Measurement of atmospheric deposition of organic trace substances – fun- nel adsorber method”, but necessarily in a heated version, due to the above mentioned weather conditions. At all three sites the following equipment has been installed: (1) a low volume sampler for the collection of organochloropesticides (OCP) and polyaromatic hydrocarbons (PAH) (2) a high volume sampler for the collection of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/F) and polybrominated diphenylethers (PBDE) (3) 7 identical deposition samplers, each used for the collection of one of these four groups of pollutants completed by chlorinated paraffins (CP), Nitrophenols and trichloroacetic acid. (4) A meteorological cabin for SPMD sampling 62 Monitoraggio dei POPs sul territorio alpino

Picture 4: Sampling site at “Zugspitze”, 1…low volume sampler, 2…high volume sampler, 3…deposition sampler, 4…SPMD sampler

Results and Discussion Since the start of the project sampling was done for five trimonthly periods. For most of the analysed POPs no source region which was predominant in all sampling periods could be detected so far, but clear seasonal differences are obvious. Whether these detected seasonal differences are periodical will be clarified in an intended prolonga- tion of these measurements. The concentration levels both for ambient air and deposition are in the same range as those measured in the rural lowlands, clearly indicating a long-range transport of POPs to these sites and the whole alpine region. I POPs NEL TERRIORIO ALPINO LOMBARDO

POPs IN THE ALPINE AREA OF THE LOMBARDY REGION

Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 65

EMISSIONI ATMOSFERICHE DI DIOSSINE IN LOMBARDIA: L’INVENTARIO DELLE EMISSIONI

A.Giudici, G.Sgorbati, S.Caserini, A.Fraccaroli, A.M.Monguzzi, M.Moretti, ARPA Lombardia Settore Aria - Via Restelli 3/1 20122 Milano

SOMMARIO Il lavoro presenta una stima delle emissioni di diossine in Lombardia, considerando diverse tipologie di sorgenti attive, principalmente l’incenerimento rifiuti, la produ- zione di acciaio, la produzione di alluminio secondario e la combustione nel settore residenziale di legna e combustibili liquidi. Oltre alle metodologie usuali utilizzate nell’ambito degli inventari delle emissioni, che prevedono la raccolta di indicatori di attività (consumo di combustibili, quantità di rifiuti incenerita, ecc.) e di fattori di emissione, per la stima delle emissioni di diossine, vista la grande incertezza nei fattori di emissione disponibili nella letteratura internazionale o nei dati di emissione dichiarati dalle principali sorgenti (es. incene- ritori), è stata effettuata una valutazione delle incertezze e dei possibili intervalli di variazione delle stime. Le variazioni rispetto ai livelli di emissioni di diossine precedentemente stimati confer- mano la riduzione nelle emissioni degli inceneritori di rifiuti, pur a fronte di un sensibile incremento nella quantità di rifiuti bruciati, come in precedenza previsto sulla base del- l’adeguamento degli impianti alle più stringenti condizioni di esercizio previste dal DM 503/97. Un ruolo predominante assumono le emissioni derivanti dalla fusione secondaria di alluminio, anche se l’affidabilità della stima delle emissioni da questa sorgente è limitata dalla disponibilità di poche misure specifiche per la realtà lombarda.

INTRODUZIONE In Regione Lombardia è da anni disponibile un inventario emissioni denominato INEMAR (Regione Lombardia, 2007), che fornisce nella sua versione più recen- te relativa all’anno 2005 le emissioni di 11 inquinanti (SO2, NOx, COV, CH4, CO,

CO2, N2O, NH3, PTS, PM10, PM2.5) e di 3 parametri aggregati quali il potenziale acidificante, il potenziale di formazione dell’ozono e le emissioni totali di gas serra. Le informazioni raccolte in questo database sono le variabili necessarie per la stima delle emissioni: indicatori di attività (consumo di combustibili, consumo di vernici, quantità di rifiuti incenerita, e in generale qualsiasi parametro che traccia l’attività dell’emissione), fattori di emissione, dati statistici necessari per la disaggregazione spaziale e temporale delle emissioni. In questo ambito sono stati condotti due realizzati due inventari delle emissioni delle diossine relative agli anni 1997 e 2001, mentre l’inventario per l’anno del 2005 è in corso di realizzazione. 66 Monitoraggio dei POPs sul territorio alpino

Pur se gli inventari delle emissioni forniscono un importante contributo nel valutare il ruolo delle diverse sorgenti, va ricordato che la rilevanza delle emissioni di diossine non significa necessariamente la rilevanza dell’esposizione umana. In generale l’espo- sizione è determinata, oltre che dalla localizzazione della popolazione, dalle modalità di emissione in atmosfera (altezza del punto di emissione, velocità di innalzamento dei fumi) e dalle caratteristiche meteo-climatologiche delle zone di emissione, parametri che regolano il trasporto e la diffusione e quindi le concentrazioni in atmosfera e i depositi al suolo. Per inquinanti persistenti e bioaccumulabili quali le diossine assumono un ruolo preponderante anche i percorsi indiretti di impatto per ingestione, contatto dermico e dieta alimentare; risulta quindi importante valutare ad esempio la localizzazione delle emissioni in relazione ai luoghi di produzione delle derrate alimentari.

METODOLOGIA Per quanto concerne le diossine, non essendo utilizzati sistemi di misurazioni in con- tinuo, la stima può essere effettuata a partire dalle è necessario ricorrere all’approccio più utilizzato dagli inventari delle emissioni, che effettua la stima sulla base di un indicatore che caratterizza l’attività della sorgente e di un fattore di emissione, spe- cifico del tipo di sorgente, del processo industriale e della tecnologia di depurazione adottata. Questo metodo si basa in altre parole su una relazione lineare fra l’attività della sorgente e l’emissione, secondo una relazione che a livello generale può essere ricondotta alla seguente:

Ei = A ⋅ FEi (1) dove: -1 Ei = emissione dell’inquinante i (g anno ); A = indicatore dell’attività, ad es. quantità prodotta, consumo di combustibile (t anno-1); -1 FEi = fattore di emissione dell’inquinante i (g t di prodotto).

La bontà di questa stima dipende dalla precisione dei “fattori di emissione”, tanto maggiore quanto più si scende nel dettaglio dei singoli processi produttivi, utilizzando specifici fattori di emissione caratteristici della tipologia impiantistica. Per i processi di combustione viene generalmente scelto, come indicatore di attività, il consumo di combustibile, mentre per i processi industriali gli indicatori privilegiati sono la quantità di prodotto o il numero di addetti.

DATI UTILIZZATI Le emissioni in atmosfera di diossine in Lombardia sono state stimate per gli anni 1997 (Caserini e Monguzzi, 2002) e 2001 (Caserini et al., 2006) ; sono in corso di completamento le stime relative all’anno 2005. Sono state considerate le principali tipologie di sorgenti presenti nel territorio regionale: Emissioni atmosferiche di diossine in Lombardia: l’inventario delle emissioni 67

• smaltimento rifiuti: incenerimento rifiuti urbani e ospedalieri, combustioni di biogas in discarica; • combustioni: per la produzione di energia e di calore; • processi ad alta temperatura: produzione cemento, asfalto, calce e vetro; • fusione, lavorazione e nobilitazione metalli: operazioni di metallurgia primaria e secondaria, sinterizzazione di minerali di ferro, produzione di acciaio; • combustioni incontrollate: incendi. Non sono stati considerati i processi biologici e fotochimici (compostaggio, fotolisi) e altre sorgenti di rilascio nell’ambiente (suoli, sedimenti, vegetazione e legno trattato con pentaclorofenolo) sia per la minor rilevanza potenziale delle emissioni che per la maggiore scarsità e incertezza dei dati disponibili per le stime. Le attività considerate sono state classificate sulla base della classificazione SNAP97 utilizzata a livello europeo nei lavori degli inventari Corinair (EEA, 2002) e nell’in- ventario emissioni della Regione Lombardia. Per gli inceneritori di rifiuti le emissioni derivano da dati misurati presso gli impian- ti; per le altre sorgenti la stima è stata effettuata sulla base di indicatori e fattori di emissione, le cui modalità di reperimento sono illustrate in seguito.

A-3.1 Incenerimento dei rifiuti I dati relativi alle emissioni degli inceneritori di rifiuti solidi urbani sono general- mente derivanti dalle dichiarazioni dei responsabili degli impianti nell’ambito del censimento emissioni puntuali dell’inventario emissioni regionale o nelle emissioni misurate durante i controlli periodici. Anche se si tratta di stime derivanti dalle misure effettuate sugli impianti stessi, come tali caratterizzate da un’alta affidabilità, è stato tuttavia calcolato per ogni impianto un intervallo di emissione, con le modalità descritte in precedenza, per tener conto delle incertezze comunque presenti nella caratterizzazione dell’emissione in relazione alla sua variabilià, alle imprecisioni analitiche o dovute ad esempio alla presenza di transitori in cui le emissioni possono discostarsi significativamente dai livelli medi.

A-3.2 Altre sorgenti Per la stima delle emissioni delle altre sorgenti è stata utilizzata la metodologia precedentemente descritta basata sul prodotto fra gli indicatori di attività e i relativi fattori di emissione. Gli indicatori di attività sono stati raccolti nell’ambito dei lavori dell’inventario emis- sioni della Regione Lombardia; i dati derivano sia da indagini di dettaglio effettuate presso i principali grandi impianti industriali, sia da fonti statistiche provinciali e regionali o da associazioni di categoria. I dati relativi al consumo di olio per la produzione di energia hanno una affidabilità elevata in quanto derivanti dalle dichiarazioni degli impianti stessi; viceversa i consumi in ambito residenziale, derivando da statistiche regionali o nazionali sono affetti da 68 Monitoraggio dei POPs sul territorio alpino maggiori margini di incertezza. In particolare si è potuto verificare che il quantitativo di legna effettivamente bruciato si presenta incerto, in quanto l’approvvigionamento della stessa è spesso effettuato in modo autonomo, e quindi molto rilevanti possono essere i quantitativi di legna che sfuggono alle rilevazioni statistiche. Data l’assenza di dati, non è stata considerata la possibilità che parte del legno bruciato sia trattato chimicamente (es. con clorofenolo). La stima in corso di realizzazione per l’anno 2005 è stata notevolmente migliorata per quanto riguarda questa sorgente, sulla base di indagini specifiche. Numerosi sono i lavori di letteratura contenenti fattori di emissione per le diossine; fonti importanti sono la raccolta Air CHIEF versione 10 (US-EPA, 2003) e l’Atmo- spheric Emission Inventory Guidebook, quarta edizione (EEA, 2006). Altre fonti importanti di dati, utilizzate nel presente lavoro, sono: • l’inventario delle emissioni di diossine negli Stati Uniti (US-E.P.A, 1998), in cui sono confrontati i fattori di emissione reperiti nella letteratura internazionale, per le diverse tipologie di sorgenti; • l’inventario europeo delle diossine (LUA-NRW, 2000), che riporta fattori di emis- sione tipici della tecnologie utilizzate nei paesi occidentali ed in particolare in Europa; • gli atti del Workshop dell’Expert Panel per la Combustione e l’Industria (UN-ECE TFEIP, 2002) che ha esaminato le emissioni di POP’s derivanti da grosse sorgenti puntuali. Peraltro, i fattori in letteratura, in studi specifici relativi alla combustione della le- gna (Pfeiffer et al., 2000), alla combustione incontrollata di rifiuti (Lemieux et al., 2000), alla combustione del biogas in discarica (Eduljee e Dyke, 1996) non risultano significativamente differenti da quelli proposti per l’inventario statunitense, e hanno permesso di aumentare il grado di affidabilità dei fattori di emissione rispetto a quanto proposto dall’US-EPA.

A-3.3 Industria metallurgica secondaria A questa categoria appartengono le attività di produzione di acciaio, effettuate preva- lentemente con forni ad arco elettrico, e la fusione secondaria di alluminio. La produzione di alluminio da rottame si caratterizza essenzialmente per due tipologie produttive (ENEA-AIB-MATT, 2002): • produzione di alluminio da rottame di alluminio nuovo e profilati, con impiego di alluminio particolarmente pulito e poco contaminato da sostanze estranee, e consi- ste essenzialmente di una attività di rifusione, generalmente in forni a riverbero, di rottame alluminio già pressoché a titolo; • produzione di alluminio da rottame di varia natura e provenienza (demolizione, raccolta, cadute di lavorazioni meccaniche, ecc.) e da recupero scorie nere, con im- piego di forni rotatori e per l’utilizzo di cloruro di sodio come coprente/scarificante del bagno fuso. Per le emissioni derivanti dall’industria metallurgica secondaria sono stati utilizzati nell’inventario 2001 i fattori di emissioni relativi alla realtà industriale italiana, con- Emissioni atmosferiche di diossine in Lombardia: l’inventario delle emissioni 69

Emissione (g TEQ anno-1) Classif. SNAP Sorgente di emissione Corinair migliore minimo massimo affidabilità stima COMBUSTIONI

1.01.01 Caldaie con potenza termica >= 300 MW - olio 0.08 0.18 0.4 MEDIA

1.01.01 Caldaie con potenza termica >= 300 MW - metano 0.002 0.005 0.01 MEDIA

1.01.01 Caldaie con potenza termica >= 300 MW - gasolio 0.00004 0.0001 0.0002 MEDIA 2.01.03 Combustione residenziale - olio 0.04 n.d. 2.02.02 2.01.07 Pizzerie con forno a legna 0.15 0.5 1.5 BASSA

2.02.05 Combustione residenziale - legno 3.2 10 31.6 BASSA

3.01.03 Combustione industriale - carbone 0.0007 0.002 0.007 BASSA 1.01.02 Caldaie con potenza termica < 300 MW - olio 0.02 0.07 0.2 BASSA 1.01.03 3.01.02 Combustione industiale < 300 MW - olio 0.005 0.01 0.05 BASSA 3.01.03 3.01.03 Combustione industriale - legno 0.03 0.071 0.2 MEDIA 7 Autoveicoli - benzina con Pb 0.1 0.2 0.8 BASSA

7 Autoveicoli - benzina senza Pb 0.02 0.05 0.2 BASSA

7 Autoveicoli - diesel 1.2 3.8 11.9 BASSA ALTRE SORGENTI AD ALTA TEMPERATURA

3.03.11 Forni per produzione cemento 0.5 1.6 5.1 BASSA

3.03.12 Forni per calce 0.02 n.d. 3.03.13 Impianti di produzione asfalto 0.04 n.d.

3.3.14 3.3.15 Produzione di vetro 0.01 n.d. 3.3.17

3.03.21 Cartiere - caldaie recupero soluzione esausta 0.01 0.029 0.07 MEDIA

PROCESSI METALLURGICI 3.03.03 Fonderie di ferro 0.6 n.d. 3.03.07 Fusione secondaria di piombo 0.2 0.36 0.8 MEDIA 3.03.08 Fusione secondaria di zinco 0.9 2.0 4.4 MEDIA 3.03.10 Fusione secondaria di alluminio 26 37 52 MEDIA 4.02.07 Prod. acciaio - forni ad arco elettrico 18 25 36 MEDIA COMBUSTIONI POCO CONTROLLATE ED INCONTROLLATE 11.03.01 Incendio di boschi, cespugli e paglia 0.02 0.1 0.2 BASSA 11.03.02 Incendio di edifici 0.1 n.d. 11.03.02 Incendio di veicoli 0.2 n.d. SMALTIMENTO RIFIUTI 9.02.01 Incenerimento rifiuti (Tab. 2) 3.5 4.9 6.9 ALTA 9.02.02 Incenerimento rifiuti tossici 0.07 0.15 0.3 MEDIA 9.04.05 Combustione di gas da discarica in torcia 0.03 0.1 0.3 BASSA 9.09.01 Forni crematori 0.001 0.004 0.01 BASSA

TOTALE 54 87 152

Tabella A-1: 1: Emissioni Emissioni di diossinediossine stimate stimate per per la Lombardia la Lombardia nel 2001. nel 2001.

4 70 Monitoraggio dei POPs sul territorio alpino

tenuti in uno specifico rapporto “Valutazione delle emissioni di inquinanti organici persistenti da parte dell’industria metallurgica secondaria” a cura di ENEA, Asso- ciazione Industriali Bresciana e Ministero dell’Ambiente e della Tutela del Territorio (ENEA-AIB-MATT, 2002). Dai risultati di questa indagine, basata su numerose misure in diversi tipi di impian- ti, sono stati ricavati dei fattori di emissione medi di emissione di diossine. Pur se caratterizzati da una notevole variabilità qualitativa e quantitativa sia a livello inter- aziendale (tra azienda e azienda) che intra-aziendale (ossia su campionamenti diversi della stessa azienda) (ENEA-AIB-MATT, 2002; Gandellini, 2003), questi dati hanno costituito la base di riferimento per la quantificazione del ruolo di questa sorgente nel 2001; l’alta variabilità registrata nelle misure, la rilevanza di alcuni dati rilevati hanno suggerito la necessità di ulteriori approfondimenti e specifiche misure come base per un aggiornato inventario delle emissioni.

RISULTATI Le stime delle emissioni di diossine in atmosfera, suddivise per sorgente di emissione, sono riportate nella Figura A-1 e nella Tabella A-1. Le emissioni complessive in Regione Lombardia sono stimate in circa 87 gTEQ anno-1 Dall’analisi dei dati di emissione medi è possibile rilevare che le principali fonti emissive di diossine sono costituite dalla fusione secondaria di alluminio, con un valore medio di 37 gTEQ anno-1, seguito dai forni ad arco elettrico con un valore

Incendio di edifici

Incendio di veicoli

Incenerimento rifiuti tossici valore massimo

Caldaie con potenza termica >= 300 MW - olio miglior stima

Fusione secondaria di piombo valore minimo

Pizzerie con forno a legna

Fonderie di ferro

Autoveicoli - benzina con Pb

Forni per produzione cemento

Fusione secondaria di zinco

Autoveicoli - diesel

Incenerimento rifiuti

Combustione residenziale - legno

Prod. acciaio - forni ad arco elettrico

Fusione secondaria di alluminio

100 1000 10000 100000 1000000

Figura A-1: EmissioniFigura minime, 1: Emissioni medie e massime minime, di diossine medie (mg TEQe massime anno-1), suddivise di diossine per le principali (mg TEQ attività, anno in Lombardia-1), suddivise nel 2001. per le principali attività, in Lombardia nel 2001.

5 Emissioni atmosferiche di diossine in Lombardia: l’inventario delle emissioni 71 medio di 25 g TEQ anno-1 . Altre importanti fonti di diossine sono la combustione residenziale di legno con un valore medio di 10 gTEQ anno-1, l’incenerimento rifiuti con un valore di 5 gTEQ anno-1.

CONCLUSIONI Sono risultate principali sorgenti le attività di fusione secondaria di alluminio, i forni ad arco elettrico per la produzione di acciaio, l’incenerimento di rifiuti e la combustione di legno in ambito residenziale. A fronte di una riduzione delle emissioni di diossine dall’incenerimento dei rifiuti, ridu- zione del resto contiunuata negli anni successivi in seguito all’implementazione della normativa sui limiti alle emissioni, gli inventari sino ad oggi disponibili segnalano la rilevanza delle sorgenti legate all’industria metallurgica, che alla luce di dati recen- temente disponibili, assumono un ruolo nettamente predominante in Lombardia. La grande variabilità di queste emissioni, in relazione a possibili interventi tecno- logici di contenimento delle emissioni stesse, suggeriscono la necessità di ulteriori ricerche, sia per valutare l’effettivo ruolo di queste fonti in diversi contesti territoriali e tecnologici, sia per uno studio di dettaglio sui sistemi di riduzione già oggi esistenti e in corso di implementazione, che possono variare anche significativamente le stime fino ad oggi realizzate.

RIFERIMENTI BIBLIOGRAFICI Caserini S., Monguzzi A.M. (2002) PCDD/Fs emissions inventory in the Lombardy Region: results and uncertainties, Chemosphere, Volume 48, Issue 8, 779 – 786. S.Caserini, A.Fraccaroli, A.M.Monguzzi, M.Moretti, A.Giudici, E.Angelino, G.Fossati (2006) Le emissioni in atmosfera dalle combustioni in Lombardia. La rivista dei combustibili, Vol. 60, n. 6, pp. 359-415. Regione Lombardia (2007)INEMAR - Inventario Emissioni in Atmosfera per il 2005 (http://www.ambiente.regione.lombardia.it/inemar/inemarhome.htm) Per gli altri riferimenti citati si veda Caserini et al. (2006) cit.

Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 73

EMISSIONI ATMOSFERICHE DI DIOSSINE IN LOMBARDIA: Il caso di un incendio di un deposito di rifiuti provenienti dalla raccolta differenziata di materie plastiche

PARTEA.G iudici, B Il G caso.Sgorbati, di un incendioS.Caserini, di A. unF raccaroli, deposito diA.M.Monguzzi, rifiuti provenienti M.Moretti, dalla raccolta differenziata diARPA materie Lombardia plastiche Settore Aria - Via Restelli 3/1 20122 Milano

B-1SOMMARIO CAMPAGNA DI MISURA I dati Si presentanoin seguito presentatii risultati disono una stati campagna rilevati nel di misuracorso delle delle campagne concentrazioni ambientali di diossineintorno al in Termovalorizzatore cui è stato Silla 2 di Milano, periodicamente svolte nell’ambito della convenzione AMSA – Comuni – Arpa per il controllo ambientaleosservato dei un microinquinantifenomeno critico nella di zonainquinamento circostante atmosferico, l’impianto di durante termovalorizzazione. il quale le concentrazioni I risultati relative alle misuredegli condotte inquinanti nel periodo misurati ge nnaiosono 2004aumentate – luglio di2005, un espressifattore dain termini 2 a 5 perdi termini il particolato, di tossicità i equivalentemetalli e (I-teq, in glipg/m³), IPA totali,sono sintetizzati di un fattore in Figura compreso B-1 e in fra Tabella 3 e 18 B-1. per il BaP e di un fattore compreso tra 18 e 35 I risultatiper le diossine. mostrano Lacome fonte in questaindicata campagna, come più in ognunoprobabile dei ètre stata siti diidentificata campionamento, in un sono incendio stati rilevati di un valori congruenti, con valori molto superiori a quelli medi ottenuti nelle precedenti campagne in due periodi fra novembre e dicembredeposito 2004, di rifiuti con un provenienti incremento delledalla conc raccoltaentrazioni differenziata per un fattore di circamateriale pari a 30.plastico. Varie osservazioni e verifiche sono state condotte dopo l’acquisizione delle risultanze analitiche che riportavano tali risultati,Campagna al fine di una di loro misura verifica e attribuzione, e determinare la estraneità a fenomeni di artefatto o oscillazione statistica (Sgorbati, 2005). I dati in seguito presentati sono stati rilevati nel corso delle campagne ambientali intorno al Termovalorizzatore Silla 2 di Milano, periodicamente svolte nell’ambito Figura B-1- Concentrazioni di diossine totali (I-teq, in pg/m³) durante la campagna di rilevamento

i-teq (Tossicità equivalente)

pg/m3 10.000

i-teq PM10 PERO 1.000

i-teq PTS PERO

0.100 i-teq PTS RHO

i-teq PTS 0.010 SETTIMO Mi

0.001

Figura 1- Concentrazioni di diossine totali (I-teq, in pg/m³) durante la campagna di rilevamento

6 74 Monitoraggio dei POPs sul territorio alpino della convenzione AMSA – Comuni – Arpa per il controllo ambientale dei microin- quinanti nella zona circostante l’impianto di termovalorizzazione. I risultati relative alle misure condotte nel periodo gennaio 2004 – luglio 2005, espressi in termini di termini di tossicità equivalente (I-teq, in pg/m³), sono sintetizzati in Figura B-1 e in Tabella B-1. I risultati mostrano come in questa campagna, in ognuno dei tre siti di campionamento, sono stati rilevati valori congruenti, con valori molto superiori a quelli medi ottenuti

Tabella 1: Concentrazioni totali di PCDD/F in termini di Indice di tossicità equivalente (I-teq) (in pg/m3 a 25°C e 1013 hPa). Emissioni atmosferiche di diossine in Lombardia: Il caso di un incendio di un deposito... 75 nelle precedenti campagne in due periodi fra novembre e dicembre 2004, con un incremento delle concentrazioni per un fattore circa pari a 30. Varie osservazioni e verifiche sono state condotte dopo l’acquisizione delle risultanze analitiche che riportavano tali risultati, al fine di una loro verifica e attribuzione, e determinare la estraneità a fenomeni di artefatto o oscillazione statistica (Sgorbati, 2005).

Le ipotesi sulle possibili origini del fenomeno osservato Un ipotetica emissione di diossine dall’impianto di Silla 2, di intensità tale da risul- tare nelle concentrazioni rilevate nel corso del periodo con valori più alti misurati in atmosfera, avrebbe potuto essere legata solamente a transitori o condizioni di im- pianto assolutamente eccezionali; tali condizioni eccezionali di impianto sarebbero state rilevate attraverso i sistemi in continua nell’ambito dello SME. Nessuna delle registrazioni in oggetto riporta traccia di alcun fenomeno del genere. Il sistema di monitoraggio in continuo delle emissioni (SME) installato su ogni linea dell’impianto misura le emissioni di acido cloridrico, ammoniaca, COV, ossidi di azoto, anidride solforosa, ossido di carbonio e polveri totali. Sulla base di questi dati e di altri indicatori di funzionamento (n. di linee in funzione, quantità di rifiuti ince- neriti) è possibile ricostruire le principali caratteristiche funzionali dell’impianto nel corso del tempo, ed in particolare nel periodo di interesse. Tra il 9 e il 21 dicembre i parametri di funzionamento dell’impianto sono rimasti totalmente all’interno della normalità. Tutte le osservazioni orientano decisamente le ipotesi per l’attribuzione della causa dell’aumento delle concentrazioni ambientali in una direzione diversa rispetto all’im- pianto di Silla 2. Tale ipotesi troverebbe però una conferma definitiva nell’individua- zione di una differente origine del fenomeno osservato. Nel corso della giornata del 22 novembre 2005, si è sviluppato un incendio in un deposito di imballaggi di materie plastiche, di recupero, sottoposto a sequestro da parte della Magistratura, sito sul territorio del comune di Settimo Milanese. L’incendio ha avuto conseguenze visibili a molti chilometri di distanza. Un’indagine retrospettiva delle concentrazioni delle diossine, effettuata attraverso il particolato raccolto campionatore ad alto volume sito presso la sede di via Juvara, in

Tabella 2: Valori di diossine (Indice di tossicità equivalente) rilevati in via Juvara a Milano. Le concentrazioni sono a temperatura e pressione ambientali. 76 Monitoraggio dei POPs sul territorio alpino Emissioni atmosferiche di diossine in Lombardia: Il caso di un incendio di un deposito... 77

Milano, dell’Arpa, normalmente impiegato per la determinazione giornaliera delle concentrazioni aerodisperse di radionuclidi, ha dimostrato come nel periodo immedia- tamente susseguente a tale evento, si sia verificato un aumento della concentrazione delle diossine, nella città di Milano, di circa 30 volte rispetto alle concentrazioni di fondo attese (vedi Tabella B-2). L’incendio è durato per numerosi giorni, e i fumi contenenti gli inquinanti, ad alta temperatura, hanno raggiunto una quota elevata, e sono stati trasportati verso la città di Milano dalle correnti soprastanti lo strato di inversione, ove sono poi in parte ricaduti. Per quanto concerne gli aspetti tossicologici del fenomeno, un commento è riportato nell’apposito capitolo della relazione . Nel periodo nel quale sono stati osservati gli incrementi di microinquinanti a livello delle stazioni di campionamento relative al monitoraggio previsto in base alla con- venzione AMSA – Comuni – Arpa, qui commentati, la concentrazione misurata nel particolato campionato in via Juvara non ha mostrato innalzamenti delle concentrazioni rispetto ai valori di fondo attesi. Le considerazioni ricavabili dalla situazione descritta sono molteplici, e se ne rias- sumono le più significative, alcune di carattere generale, alcune di carattere specifi- co, legate a plausibili interpretazioni del fenomeno. a) un incendio di materie plastiche, necessariamente contenenti frazioni di polimeri clorurati, può provocare la formazione e la dispersione di microinquinanti in pro- porzione significative, dal punto di vista delle rilevabilità analitica dei microin- quinanti stessi; b) l’impatto di un fenomeno di dispersione di inquinanti, con innalzamento in quota di fumi caldi, come varie esperienze dimostrano, non necessariamente è più signi- ficativo nelle immediate vicinanze del luogo di rilascio; c) nel caso specifico, è possibile, sebbene poco probabile, che, ad una prima fase di rilascio di fumi caldi, con trasporto in quota, sia seguita una fase di rilascio di par- ticolato più freddo e pesante, nelle fasi di intervento e messa in sicurezza condotte nelle settimane successive all’incendio, che possono avere determinato le ricadute locali rilevate dalle stazioni di rilevamento di Pero, Rho e Settimo; d) altrettanto è possibile che si sia verificato, nel periodo delle campagne di misura di novembre e dicembre 2004, un altro episodio, non riconosciuto perché di propor- zioni o evidenza minori, di incendio o combustione e rilascio di microinquinanti, che ha interessato più specificamente l’area.

Conclusioni Nel corso delle campagne di misura delle concentrazioni di microinquinanti nella zona circostante l’impianto di termovalorizzazione di Silla 2 è stato osservato nel novembre – dicembre 2004 un fenomeno critico di inquinamento atmosferico durante il quale le concentrazioni degli inquinanti misurati sono aumentate di un fattore da 2 a 5 per 78 Monitoraggio dei POPs sul territorio alpino il particolato, i metalli e gli IPA totali, di un fattore compreso fra 3 e 18 per il BaP e di un fattore compreso tra 18 e 35 per le diossine. Accurate analisi delle condizioni di impianto, e l’analisi multivariata dei cogeneri delle diossine osservate permettono di escludere che la situazione ambientale verificatasi sia da porre in correlazione a rilasci dell’impianto di Silla 2 (Sgorbati, 2005). Altre condizioni ambientali e fenomeni anomali possono, per altro, essere in grado di provocare una situazione del tipo osservato, come dimostrato dallo studio delle conseguenze di un incendio di un deposito di materie plastiche di recupero che si è verificato in un periodo temporalmente prossimo a quello della campagna che ha rilevato alti livelli, che non si esclude totalmente possa essere la causa del fenomeno osservato. In considerazione delle concentrazioni rilevate, delle incorporazioni estrapolabili per gli esposti e della cinetica dei microinquinanti considerati, dei livelli di riferimento SCF considerati, si esclude che il fenomeno in oggetto sia in alcun modo in grado di provocare effetti di tipo acuto, e si attribuisce al fenomeno la capacità di alterare in modo non significativo l’esposizione rispetto alle condizioni “di fondo” stimate sul medio e lungo periodo.

Riferimenti bibliografici Sgorbati G. (2005) Concentrazioni di diossine relative alla campagne ambientali AMSA del novembre – dicembre 2004. Situazione eccezionale rilevata nel dicembre 2004. Relazione tecnica Arpa Lombardia, Dipartimento di Milano. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 79

The emissions of POPs in the agriculture sector: trends and current state

Christian Schlitt and Angelo Moretto International Centre for Pesticides and Health Risk Prevention Ospedale Luigi Sacco Azienda Ospedaliera - Polo Universitario v. G.B.Grassi 74 - 20157 Milano

Riassunto L’articolo descrive il passato e il presente della produzione, uso e rilascio dei nove pesticidi compresi nella “sporca dozina” (Convenzione di Stoccolma) e di alcuni candidati. Si presenta anche una panoramica sulla presenza di pesticidi in Regione Lombardia e sulle misure adottate per prevenire l’ingresso di alimenti contaminati nel mercato regionale. La presenza nella UE di siti e suoli contaminati con POP e la sussistenza di alcune industrie produttrici di POP, molte delle quali esportano nei paesi in via di sviluppo, indicano che i problemi legati ai POP sono lontani dall’essere risolti. Nell’Italia settentrionale, malgrado 20 anni di legislazione, l’inquinamento di pesticidi POP è ancora in crescita nei laghi subalpini. Viene presentato un esempio dell’influenza delle condizioni climatiche sulla risospensione e il trasporto di suoli e sedimenti contaminati.

Abstract In the present essay the past and current production and use/release are outlined for the nine POP pesticides known as the “dirty dozen” (Stockholm Convention) and a few candidates. Also, a general picture is given on the presence of POP pesticides in the Lombardy Region and on the measures taken to prevent contaminated food to enter the regional market. The presence in EU of POP contaminated sites and soils and the subsistence of a few POP producing industries, most of them exporting to developing countries, show that problems associated with POP persistence are far from being resolved. In North-Italy, in spite of over 20 years of regulation, POP pesticide pollution still has a tendency to increase in large sub-alpine lakes. An example is presented on the influence that weather conditions on contaminated soils and water sediments could have on the re-distribution of POPs. Enhanced release to the atmosphere, air transport and cold condensation are expected to further contribute to POPs’ involuntary mass transfer.

Introduction Among the twelve Persistent Organic Pollutants (POPs) known as the “dirty dozen” (Stockholm Convention), nine are pesticides (used as insecticides in agricultural crops and/or for public health vector control). 80 Monitoraggio dei POPs sul territorio alpino

Chemical properties of POP pesticides include low water solubility, high lipid solu- bility, moderate to medium volatility and medium to high persistence, i.e. resistance to environmental degradation. The risk that these substances may pose to humans and the environment is not only a function of their toxicity, but is strongly related to their specific bioavailability and potential to enter the food chain. Current information indicates that most, if not all, of the nine pesticides in question are still in use in parts of the world where they are considered essential e.g. for ensuring public health, with actual quantities used in specific countries not better known. A summary is here given on the past and current production and use/release of POP Pesticides in the agricultural sector, highlighting the Italian pre-alpine situation.

POP Pesticides’ Use in the EU POP pesticides have been produced in large quantities and were extensively used after World War II. The intensity as well as the period and the types of these pesticides used varied between European Countries although specific data are often not available. In general, their uses were most intensive in the fifties and sixties with a decline in use to the eighties, when most of the substances were banned in the old Member States (EU 15) by the EU Plant Protection Product Directive 79/117/EEC. Production years, typical and last known uses for each of the POP pesticides are sum- marised in Table 1. Data on the use of certain pesticides are difficult to obtain and may be unreliable. The Table nevertheless provides some insight for what purposes they have been or are being used The Table has been extended to new POP pesticide candidates and also takes into account pesticides determined on behalf of the MON- ARPOP monitoring campaigns.

Table 1: POP pesticides: production, typical and last known uses * (including new candidates and Monarpop-analytes)

Production Last known uses (worldwide) POP pesticide Typical uses years EU bans/restrictions according to 79/117/EC Aldrin 1950 – 1990 Applied to soils to kill termites, grasshoppers, corn Against termites and other soil pests, termites at- rootworm, and other insect pests on cotton, seed tacking building materials, in grain storage, and for treatments. vector control. EU: Use banned without exception since 1990. Chlordane 1945 – 1997 Used since 1945 for control of cockroaches, ants, Against termites and other soil pests, termites at- termites and household pests and on a on a range tacking building materials. of agricultural crops. EU: ban on use since 1997. Dichlorodiphenyl- 1940 - 1983 Widely used during World War II to protect soldiers Control of medical and veterinary vectors, such as trichloroethane and civilians from malaria, typhus, and other dis- malaria-transmitting mosquitoes, plague-transmit- DDT eases spread by insects. ting fleas and trypanosomiasis-transmitting tsetse flies. EU: use restricted in 1983 and 1985, all uses pro- hibited in 1990, the use as intermediate for the pro- duction of dicofol is allowed until 2014, production ongoing in Italy and Spain. Dieldrin 1948 – 1979 Used principally to control termites and textile Control of locusts, termites, human disease vec- pests, to control insect-borne diseases and insects tors; ban on use since 1979. living in agricultural soils. EU: ban on use since 1979. The emissions of POPs in the agriculture sector: trends and current state 81

Endrin 1950 - 1985 This insecticide is sprayed on the leaves of crops Formerly used against insects and rodents. No cur- such as cotton and grains. It is also used to control rent or recent uses are known. mice, voles and other rodents. EU: use restricted since 1985, banned since 1990, European production already had ceased in 1982. Heptachlor ? – 1985 Primarily employed to kill soil insects and termites, Against termites and other soil pests, termites at- widely used to kill cotton insects, grasshoppers, tacking building materials. other crop pests, and malaria-carrying mosquitoes. EU: use banned without exception since 1985. Hexachloro- 1945 – 1979 HCB kills fungi that affect food crops. It is also re- Formerly used for seed treatment against fungal benzene leased as a byproduct during the manufacture of diseases, as well as for industrial purposes. No HCB certain chemicals and as a result of the processes current or recent agricultural uses are known. that give rise to dioxins and furans. Also used for EU: banned as pesticide since 1979, derogation fireworks and synthetic rubber. on use as intermediate in Poland, production in Germany. Mirex 1950 – 1990 Mainly applied to combat fire ants (mainly in US) Against leaf-cutting ants, termites in buildings and and other types of ants and termites. Also used as outdoors, and also as a fire retardant and for other fire retardant in plastics, rubber, plastics, paint, pa- industrial purposes EU: all uses prohibited. per, and electrical goods. The substance has never been used in Mediterranean countries Toxaphene 1949 – 1983 This insecticide, is applied to cotton, cereal grains, Control of insect pests in cotton and other crops. (Camphechlor) fruits, nuts, and vegetables. It has also been used EU: use banned without exceptions in 1983. to control ticks and mites in livestock, control of scabies on sheep and cattle and of ectoparasites. The substance has never been used in Mediter- ranean countries

Chlordecone 1951 - 1975 Has been used in various parts of the world for the Legal ban has been reported by Germany, Canada, control of a wide range of pests. It can be used as a the USA and Switzerland. fly larvicide, as a fungicide against apple scab and EU: Listed in the Aarhus Protocol (1998) and An- powdery mildew and to control the Colorado potato nex I to Regulation (EC) No 850/2004 as substance beetle, rust mite on non-bearing citrus, and potato scheduled for elimination without any exception in and tobacco wireworm on gladioli and other plants. production and use; all uses prohibited Chlordecone has also been used in household Apparently there are no commercial reason to products such as ant and roach traps at concentra- maintain stockpiles. Waste containing Chlordecone tions of approximately 0.125%. has to be destroyed if concentration limits of 50 mg/kg are exceeded. Dicofol 1956 – up to Still used as miticide on a large number of crops to Dicofol is produced by certainly one and maybe two date kill crop-feeding mite pests such as the red spider other companies (Spain). For one company (Italy) it mite. Only for pre-harvest application. It has no is known that DDT is produced as an on-site inter- insecticidal activity Also used in combination with mediate in order to make Dicofol. other pesticides such as the organophosphate. EU: Until 1979 severe restriction of Dicofol contain- Dicofol is a mixture of p,p’-dicofol and o,p’-dicofol ing more than 78% p,p*-Dicofol or 1 g/kg of DDT and is produced by hydroxylation of DDT. and DDT related compounds Endosulfan 1954 – up to A broad spectrum contact insecticide and acaricide EU: Authorization for use existing in seven EU- date used on a wide variety of vegetables, fruits, cereal Member States. Non-inclusion in Annex I of grains, and cotton, as well as ornamental shrubs, Dir.91/414/EEC decided in 2005. trees, vines. Also used for the control of ticks and mites, and the control of rice stem borers. Hexachloro- 1940 – 1990 Widely used since early 1940s as insecticide in ag- Ongoing use in control of ectoparasites. In Italy cycohexanes riculture, household, wood and textile protection. Lindane has been largely used as insecticide until α-HCH; 2002. β-HCH; EU: ban of HCH containing less than 99.0% of γ-HCH; γ-HCH according to 79/117/EEC. Use as pesticide Lindane: 99% banned in 1990. γ-HCH Until 2007: technical HCH allowed as intermediate, δ-HCH Lindane restricted to use in public health and veteri- nary as topical insecticide. Until 2006 HCH allowed as professional remedial and industrial treatment of lumber, timber and logs and for indoor applica- tions. Production in France, Italy and Germany Pentachloro- ? – up to date Extensively used as a wood, industrial textile and EU: Restrictions on the marketing by Directive phenol leather preservative (both as insecticide and fun- 91/173/EEC. Ban on the use of pentachlorophenol PCP gicide). Domestic use, such as indoor application and its compounds in a concentration equal to or of wood preservatives and paints based on PCP greater than 0.1% by mass, except in substances or PCP-treated wood resulted in severe Indoor air and preparations intended for use in industrial in- pollution. stallations. * Adapted from: EC (2005), Ritter et al. (1995) and FOOTPRINT (2006) 82 Monitoraggio dei POPs sul territorio alpino

Main objectives of the European POP regulation 2004/850/EC are: • Production and use of hexachlorohexane (HCH) including Lindane should be con- fined to a minimum and ultimately phased out by 2007. • Stockpiles of prohibited substances should be treated as waste. In particular this shall apply for stockpiles which consist of or are contaminated with POPs as soon as possible. • Releases of unintentional by-products of industrial processes should be identified and reduced as soon as possible with the ultimate aim of elimination. • Establish common concentration limits for POP substances and substance groups before 31 December 2005. The mass flow of POP pesticides to the waste regime is mainly due to remaining stockpiles of these substances. Due to different economic systems the issue of stock- piles seems to be more important in the “new” Member States which have reported remaining stocks in a dimension of 5.000 tons of pure POP substances, whereas “old” Member States reported that stocks have already been eliminated. Different pieces of information state that there is still a production of POP pesticides in Europe. Additionally imports in a dimension of 1,000 – 2,000 t/y are mentioned. Both production and imports concern DDT which is used as a pre-product for Dicofol (Italy). The total dimension of the source of the corresponding mass flow is about 3,000 t/y. Also, Lindane (γ-HCH>99%) has been largely used as insecticide until recent years (e.g. stockpile use in Italy up to 31 may 2002) and is still produced in France, Germany and Italy. The amount of identified lindane stocks that are remaining in EU 25 is not high (270 t which, provided a linear stock reduction, corresponds to an annual contribution of 27 t). HCH and HCB are also being classified as industrial chemicals or by-products.

POP Pesticides in the Lombardy Region In spite of over 20 years of regulation, POP pollution still has a tendency to increase in Italian sub-alpine lakes. In particular, a point source of DDT pollution of industrial origin, discovered in 1996 in Lake Maggiore, the second largest (212 km2) lake in Italy, created concern for wildlife and human health due to contaminant levels in fish species exceeding thresholds for human consumption. Subsequent investigations revealed a generalized contamination caused by a chemical plant located in Pieve Vergonte near the Toce river, the second largest affluent of the lake. Extensive researches were started both in Italy and Switzerland and are annually documented (CIPAIS 2002- 2005). Chemical contaminants are monitored on atmospheric precipitation, lake’s water column, lacustrine sediments and indicator organisms. Analyses cover a wide spectrum of POP compounds, comprehending: op-DDD, pp-DDD, op-DDE, pp-DDE, op-DDT, pp-DDT; α-HCH, β-HCH, γ-HCH and γ-HCH; HCB; Heptachlor; Dieldrin; Metoxychlor; Mirex and 13 PCB congeners. Although the area of the chemical plant responsible for DDT contamination is a major reclamation project in Italy (Law n°426/1998), clean up works have just been started. The emissions of POPs in the agriculture sector: trends and current state 83

Results of a recent POPs biomonitoring campaign on the principal Italian sub-alpine lakes has been presented in 2007 by the Department of Biology (Ecology Section) of the University of Milan (Riva et al. 2007). Sampling was carried out on Zebra mussel (Dreissena polymorpha), a common biomarker of chemical contamination, in April 2003 at 15 selected locations on Lakes Maggiore, Garda, Como, Iseo and Lugano. 2003 data showed a moderate overall increase (about 50% averaged on 15 locations) when compared with levels determined in a previous campaign in 1996. Results point out high DDT levels in Zebra mussels from Lake Maggiore, 5-9 times higher than those measured in the other Italian lakes (see Table 2; Riva et al. 2007). Starting from 1996 these concentrations constantly decreased until 1999, followed by a dramatic increase during 2001 - 2002, due to a heavy flood event, with the lake water levels reaching the highest value of the 20th Century (Riva et al. 2007).

Tabel 2: Concentrations of total DDTs (sum of parent and metabolites) in zebra mussels from different rivers and lakes worldwide (Taken from: Riva et al 2007) Study area ΣDDT (ng/g lipids) References Lake Garda, Como, Iseo, Lugano 62 - 224 Riva et al. 2007 Lake Maggiore 730 - 1386 Riva et al. 2007 Rideau River (Canada) 65 - 169 Renaud et al., 2004 Flanders (Belgium) 46 - 471 Covaci et al., 2005 Flanders (Belgium) 49 - 406 Voets et al., 2006 Rhine River (The Netherlands) 272 Hendriks et al., 1998 Meuse River (The Netherlands) 161 Hendriks et al., 1998

Figure 1: Eight-year trend of total DDTs in zebra mussels (ng/g lipids) from Lake Maggiore (Taken from: Riva et al 2007). 84 Monitoraggio dei POPs sul territorio alpino

Food safety POP Pesticides exposure in humans occurs mainly by ingestion of contaminated foods. Italy started to improve monitoring and analytical activities in accordance with the provisions of Council Directive 89/397/EEC on the official control of foodstuffs, and Council Directive 93/99/EC on additional measures concerning the official con- trol of foodstuffs. Sampling methods have been specified in Commission Directive 2002/63/EC establishing Community methods of sampling for the official control of pesticide residues in and on products of plant and animal origin. Directives 2001/201/ EC, 2001/2375/EC, 2002/69/EC, 2003/806/EC and 1986/363/EC set limit values for POP pesticides and PCDD/PCDF in food and feeding stuffs and contain provi- sions for monitoring and control. POP Pesticides covered by the Official Monitoring Program in Lombardy are: Aldrin, Dieldrin, DDEs, DDDs, DDTs, Dicofol, Endrin, HCB, HCHs, Heptachlor, Heptachlor epoxide, alfa-Endosulfan, beta-Endosulfan, Endosulfan-sulfate. No remarkable contamination in foods of plant origin was ob- served in the last years. The limit proposed for pp’DDT in water by the EU Directive 2000/60/EC, which will come into force in 2008, is 0.2 ng/L, four times higher than the average concentration measured in Lake Maggiore waters. Nevertheless, concentrations measured in Lake Maggiore fish were very close and sometimes exceeded the Maximum residue limit (MRLs) settled by the Italian legislation for foods (0.1 mg/kg w.w. for fish containing 5-20% lipid) (Bettinetti et al. 2006).

Conclusions Since the complete ban of principal POP Pesticides in several countries and severe restrictions on their production and use, a general decrease of POP levels in foodstuff and the environment has been observed during the last decades. However, the presence in EU of POP contaminated sites and soils and the subsistence of a few POP produc- ing industries, most of them exporting to developing countries, show that problems associated with POP persistence are far from being resolved. In North-Italy, in spite of over 20 years of regulation, POP pollution still has a tendency to increase in large sub-alpine lakes. The important DDT contamination in Lake Mag- giore is an example on how a chemical that has been banned for use in 1978 (D.M.11 ottobre 1978) can still present problems in public health and environmental sectors. An example is presented on the influence that weather conditions on contaminated soils and water sediments could have on the re-distribution of POPs. Enhanced re- lease to the athmosphere, air transport and cold condensation are expected to further contribute to POPs’ involuntary mass transfer. The emissions of POPs in the agriculture sector: trends and current state 85

Bibliography Bettinetti et al. (2006). pp’DDT and pp’DDE accumulation in a food chain of Lake Maggiore (Northern Italy): testing steady-state condition. In: Environmental Science and Pollution Research. - ISSN 0944-1344. - 13:1(2006). - p. 59-66. CIPAIS (2002-2005). Commissione Internazionale per la protezione delle acque italo-svizze- re. Monitoraggio della presenza del DDT e di altri contaminanti nell’ecosistema del lago Maggiore, Rapporti annuali 2002, 2003, 2004, 2005 C.N.R, I.S.E. Pallanza. EC (2005). Study to facilitate the implementation of certain waste related provisions of the Regulation on Persistent Organic Pollutants (POPs) FINAL REPORT August 2005 Euro- pean Commission, Brussels - REFERENCE: ENV.A.2/ETU/2004/0044. FOOTPRINT (2006). The FOOTPRINT Pesticide Properties DataBase. Database collated by the University of Hertfordshire as part of the EU-funded FOOTPRINT project (FP6-SSP- 022704). http://www.eu-footprint.org/ppdb.html. Law N°426 emitted the 9/12/1998. Nuovi interventi in campo ambientale. G.U. N°291 del 14/12/1998. Regulation No 850/2004 of the European Parliament and of the Council of 29 April 2004 on persistent organic pollutants and amending Directive 79/117/EEC. Official Journal of the European Union 30.4.2004 L158/7 and 29.6.2004 L 229/5 (Corrigendum). Ritter L. et al (1995). A Review of Selected Persistent Organic Pollutants: DDT, Aldrin, Dieldrin, Endrin, Chlordane, Heptachlor, Hexachlorobenzene, Mïrex, Toxaphene, Poly- chlorinated biphenyls Dioxins and Furans. The International Programme on Chemical Safety (IPCS) within the framework of the Inter-Organization Programme for the Sound Management of Chemicals (IOMC). Riva, C. et al. (2007). Evaluation of several priority pollutants in zebra mussels (Dreis- sena polymorpha) in the largest Italian sub-alpine lakes, Environ. Pollut. doi:10.1016/ j.envpol.2007.03.016.

Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 87

The effects of POPs on Alpine organisms and ecosystems

Marco Vighi, Sara Villa and Elisa Bizzotto Department of Environmental Sciences – University of Milano Bicocca Piazza della Scienza 1 – 20126 Milano - Italy

Riassunto Nel presente articolo si presenta una sintesi dei risultati delle campagne di moni- toraggio realizzate dal 2000 al 2007 per misurare la concentrazione di POPs nei corsi d’acqua alpini alimentati da ghiacciai. I dati vengono utilizzati per valutare il potenziale rischio per le comunità acquatiche. I risultati indicano che i POPs possono rappresentare un pericolo per gli ecosistemi acquatici di alta quota.

Abstract A synthesis of the results of monitoring campaigns performed from 2000 to 2007 to measure POP concentrations in Alpine glacial streams is reported. Data are used to assess the potential risk for the aquatic communities. The results indicate that POPs may represent a threat for high mountain aquatic ecosystems.

Introduction Persistent Organic Pollutants (POPs) are known to concentrate in cold environments as a result of volatilisation from warm regions and condensation in colder areas. The role of high mountains as cold condensers was hypothesized by Calamari et al. (1991)

Figure 1. Concentrations of HCHs in the different layers of the Lys glacier ice core, corresponding to a time span from 1950 to 1996, and in new (1) and aged (2 to 4) snow samples collected in June 2000 (modified after Villa et. al., 2003).

20 a-HCH g-HCH 1 2 3 4 20 10 10

5 5 a-HCH g-HCH 2 2

1 1 Detection limit in ice 0.5 0.5

0.2 0.2 Concentration in ice and snow ng/L snow and ice in Concentration 51-60 61-71 72-74 75-77 78-80 81-83 84-86 87-89 90-92 93-95 Snow 2000 Years

Figure 1. Concentrations of HCHs in the different layers of the Lys glacier ice core, corresponding to a time span from 1950 to 1996, and in new (1) and aged (2 to 4) snow samples collected in June 2000 (modified after Villa et. al., 2003). 88 Monitoraggio dei POPs sul territorio alpino and confirmed by many authors (Blais et al., 1998). In particular, Alpine glaciers are a sink for pollutants (Villa et al. 2003, 2006a) and ice layers deposited in the second half of the XX century contain high concentration of POPs (Fig. 1). This burden of pollutants is known to be then released in streams as a result of glacial melting (Blais et al., 2001a; Villa et al., 2006b). As a consequence, it is possible to hypothesize a pollutants flow during summer that may represent a threat to the sur- rounding ecosystems. Most POPs can produce adverse effects at very low concentration, acting as endo- crine disrupting chemicals (Asplund et al., 1999) or as carcinogens (Ahlborg et al., 1995). Moreover, POPs have a high potential for secondary poisoning due to their biomagnification capability. Finally, biological communities of extreme ecosystems are particularly vulnerable, due to their relatively simple structure. Despite these evi- dences, a few studies have focused on high-altitude water samples (Vilanova et al., 2001; Carrera et al., 2001; Blais et al., 2001a, 2001b, Lafreniere et al., 2006). In this paper the results of some monitorig campaigns in Alpine glacial streams are described, to assess the potential risk for aquatic comunities.

Materials and Methods Since the year 2000, POP contamination in Alpine glacial streams was monitored. Five glacial streams were sampled in the Italian Alps (Fig. 2): Lys stream (Lys glacier, massif, Western Alps); Frodolfo stream (Forni glacier, Ortles-Cevedale

Figure 2. Location of the sampling area The effects of POPs on Alpine organisms and ecosystems 89

group, Central Alps); Dora di Veny stream (Miage glacier, Monte Bianco massif, Western Alp), Careser and Noce Bianco streams (Careser glacier and Col De La Mare glacier respectively, both in the Ortles-Cevedale group, Central Alps). Lys and Fro- dolfo streams were sampled in 2000, 2001 and 2002. In 2002 Dora di Veny, Careser and Noce Bianco were added. In 2006 an intensive monitoring was performed on the Frodolfo stream for a more detailed assessment of temporal (daily and seasonal) trends and of transfer in aquatic trophic chains. In 2007 a survey was performed on several glacial streams of Valtellina, in order to estimate the load to River Adda and Lake Como. Analyzed chemicals were DDT isomers and metabolites, HCB, α-, β-, γ-HCH and a selection of PCBs. Details on the monitoring plans and on the sampling and analytical procedures are described in the original papers (Villa et al., 2006b; Bizzotto et al., 2007).

Results and Discussion A synthesis of the major results is reported below. More details on the results can be found in the already quoted original papers. In figure 3 the range of concentrations measured in glacial streams is reported for chlorinated pesticides (sum of DDTs, sum of HCHs and HCB). In Frodolfo and Lys, for which systematic samples are available, the range of concentrations of all chemicals is in good agreement. For Miage the difference is due to the sporadic sampling frequency, however, values fall within the

Figure 3. Range of concentrations of organochlorine pesticides measured in three glacial streams.

Miage 2002 Lys 2000-2002 Frodolfo 2002-2006

Miage 2002 Lys 2000-2002 Frodolfo 2002-2006

Miage 2002 Lys 2000-2002 Frodolfo 2002-2006 DDTs HCHs HCB

1 10 100 1000 10000 Concentration range (pg/L)

Figure 3. Range of concentrations of organochlorine pesticides measured in three glacial streams. 90 Monitoraggio dei POPs sul territorio alpino

Figure 4. Trend of concentrations of total PCBs in the Frodolfo stream in spring-autumn 2006

100000

10000

1000

100

Concentration pg/L Concentration 10

1 may june july september october

range observed in the other streams. Same comments can be made for Careser and

Figure 4. TrendNoce of concentrations Bianco, where of total PCBs data in theeven Frodolfo more stream sporadic. in spring-autumn In the 2006 same geographic areas, some non-glacial streams were also sampled. The concentrations of all chemicals were sub- stantially lower (about one order of magnitude), confirming that the levels measured in glacial streams are a consequence of the accumulation in glaciers. Different comments can be made on PCBs measured in the Frodolfo stream during 2006 (Fig. 4). Levels measured in May are low and comparable with those observed in other glacial streams. In June a concentration increase of about three orders of magnitude is observed. Than, concentration decreases, up to levels comparable to those measured in May. A comparable trend was observed in a non glacial stream sampled in the Frodolfo Valley. A possible explanation of this trend could be a contamination due to snow melting and local emissions instead of long range transport and glacier accumulation. Possible sources of local emissions for PCBs could be hydroelectric power plants. Indeed, a small hydroelectric power plant is present in the Frodolfo Valley, close to sampling sites. About the possibility of ecotoxicological risk, it is reasonable to suppose that the traditional risk assessment approaches, based on data on a few standard organisms and on traditional endpoints, are not suitable for assessing the risk on the Alpine communities due to POPs. Species sensitivity distribution (SSD) has proved a useful approach to predict the sensitivity of entire communities (Posthuma et al., 2002). The basic assumption is that the sensitivity of different species in a community toward a given stressor follows a normal distribution. According to this assumption, the number of species potentially affected by a given concentration of a toxic chemical can be statistically determined. According to the Dutch school, a concentration safe for 95% of the species of a The effects of POPs on Alpine organisms and ecosystems 91

community (HC5: Hazardous concentration 5%) could be assumed as suitable for protecting the ecosystem. Major limitations for the application of SSD, in particular for POPs in Alpine eco- systems, are the following: • due to the lack of information, SSD model are often based on acute toxicity data; chronic data are not frequently available for a large number of species; even less data are available for endocrine disruption endpoints; • SSD does not take into account secondary poisoning, extremely relevant for POPs; • no data are available for developing SSD models for species typical of Alpine eco- systems and nobody knows if they are more sensitive than traditional species. In figure 5, an example of SSD curves, specific for fish and arthropods, is reported for DDT. The estimated HC5 is about two orders of magnitude higher than the upper limit of the range of concentrations measured in glacial streams. However, the value should be reduced by the application of four different security factors due to the difference between acute and chronic toxicity (in this case a factor of 10 is traditionally applied), secondary poisoning, endocrine disrupting effects, higher sensitivity of alpine com- munities. In particular, a reasonable value for the last two factors is totally unknown; however, it is highly probable that measured concentrations in glacial streams could overcome a threshold of risk for aquatic communities.

Figure 5. SSD curves for DDT compared with DDT concentrations measured in glacial streams. To estimate a safe concentration, HC5 should be reduced through a series of application factors (expla- nation in the text). 100 HC5 level Fish

80 Arthropods

60 Water solubility

40 Range of surface water Application factors % species affected species % concentrations Acute to cronic 20 Secondary poisoning Endocrine disruption Higher sensitivity 0 -2 -1 0 1 2 3 4 5 DDT log LC50 ng/L

Figure 5. SSD curves for DDT compared with DDT concentrations measured in glacial streams. To estimate a safe concentration, HC5 should be reduced through a series of application factors (explanation in the text). 92 Monitoraggio dei POPs sul territorio alpino

4. Conclusions In synthesis, the following schematic conclusions can be drawn. • Glacier melting mobilise POPs accumulated during the period of massive global use; therefore. POP concentrations in streams fed by glaciers is substantially higher than in non glacial streams. The pattern can be enhanced by global warming. • Besides global long range transport, Alpine streams can be contaminated by local emissions followed by cold condensation and snow scavenging. In particular, local emissions are likely for PCBs due to the distribution of hydroelectric power plants in the Alpine territory. • Nobody knows if there are significant differences of sensitivity between Alpine and low-land aquatic communities. However, risk from POPs is likely to occur in Alpine streams. Acknowledgements. Paper supported by University of Milano Bicocca, Foundation Lombardia Ambiente and Regional Environmental Protection Agency of Lombardia within the Project RICLIC-WARM (Regional Impact of Climatic Change in Lombardy Water Resources: Modelling and applications).

References Ahlborg U.G., L. Lipworth, L. Titus-Eenstoff, C. Hsieh, A. Hanberg, J. Baron, D. Trichopoulos, H. Adami, 1995. Crit. Rev. Toxicol. 25, 463. Asplund L., A. Athanasiadou, A. Bergman, H. Börjesson, 1999. Ambio 28, 67. Bizzotto E., C. Vaj, S. Villa, M. Vighi, 2007. Chemosphere (submitted). Blais J. M., D. Schindler, D. Muir, L. Kimpe, D. Donald, B. Rosenberg, 1998. Nature 395, 585. Blais J. M., D. Schindler, D. Muir, M. Sharp, D. Donald, M. Lafreniere, E. Braekevelt, W. Strachan, 2001a. Ambio 30, 410. Blais J. M., D. Schindler, M. Sharp, E. Braekevelt, M. Lafreniere, K. McDonald, W. Strachan, 2001b. Limnol. Oceanogr. 46, 2019. Calamari D., E. Bacci, S. Focardi, C. Gaggi, M. Vighi, 1991. Environ. Sci. Technol., 25, 1489. Carrera G., P. Fernández, R. Vilanova, J. Grimalt, 2001. Atmos. Environ. 35, 245. Lafrenière M.J., J. M. Blais, M. Sharp, D. Schindler, 2006. Environ. Sci. Technol., 40, 4909. Posthuma L., G.W. Suter, T.P. Traas (Eds.), 2002. Species Sensitivity Distribution in Ecotoxi- cology. Lewis Publishers, Boca Raton. Vilanova R., P. Fernandez, C. Martinez, J. Grimalt, 2001. J. Environ. Qual. 30, 1286. Villa S., C. Negrelli, A. Finizio, O. Flora, M. Vighi, 2006b. Ecotox. Environ. Safety, 63, 84. Villa S., C. Negrelli, V. Maggi, A. Finizio, M. Vighi, 2006a. Ecotox. Environ. Saf., 63, 17. Villa S., M. Vighi, V. Maggi, A. Finizio, E. Bolzacchini, 2003. J. Atmospheric Chemistry, 46, 295. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 93

EVOLUZIONE TEMPORALE E ANDAMENTI STAGIONALI DI POP IN DEPOSIZIONI D’ALTA QUOTA IN LOMBARDIA E SVIZZERA

Licia Guzzella, Claudio Roscioli, Adolfo De Paolis, CNR-IRSA,Via della Mornera 25, 20047 Brugherio (MI), Italy

Abstract Organochlorine (OC) and PAH compounds were analysed in monthly bulk atmos- pheric deposition samples in two Alpine remote areas: Devero (1600 m) and Robiei (2000 m) stations from 2001 and 2007. Lindane, PCB and PAH annual loads in the Swiss station were greater than those measured in Italian one, while DDT annual load calculated at Devero Alp underlined the proximity of the station to a local source of pollution.

Introduzione Nell’ambito del Progetto Monitoraggio della presenza di DDT ed altri contaminanti nell’ecosistema Lago Maggiore-Quinquennio 2001-2006 finanziato dalla Commis- sione Internazionale per la protezione delle acque italo-svizzere è stato condotto uno studio sulle principali cause di inquinamento del Lago Maggiore, considerando sia gli apporti fluviali dei principali tributari che quelli atmosferici imputabili direttamente alla deposizione di inquinanti sulla superficie lacustre che indirettamente attraverso lo scorrimento delle acque sul territorio del relativo bacino. Nell’ambito di detto Progetto (CIPAIS, 2002; 2003; 2004; 2005; 2006) per gli apporti atmosferici sono stati considerati alcuni composti organici persistenti (POP) legati a fonti di contaminazione presenti nel bacino come per il DDT o i PCB ed altri conta- minati più legati al trasporto atmosferico, come per gli HCH e l’HCB. Nell’ambito del progetto Europeo EUROLIMPACS (Contratto GOCE- CT 2003- 505540) e in parallelo a tale indagine, è stato condotta anche la determinazione dei principali IPA presenti nelle deposizioni atmosferiche al fine di comprendere i principali meccanismi di trasporto e le fonti di contaminazione presenti sul territorio.

Materiali e Metodi Per lo studio del contributo degli apporti atmosferici alla contaminazione da composti organici persistenti nell’arco alpino è stata condotta un’attività di campionamento delle deposizioni bulk in due aree significative ad alta quota che sono state confrontate con due aree a bassa quota. Le località di campionamento selezionate sulla base della disponibilità di stazioni esistenti sono state: in Italia, Pallanza (200 m) e Alpe Devero (1600 m); in Svizzera, Locarno (300 m) e Robiei (2000 m) (Figura 1) . Per ciascuna 94 Monitoraggio dei POPs sul territorio alpino

Figura 1 – La collocazione delle stazioni di campionamento delle deposizioni atmosferiche stazione sono stati preparati i campioni medi ponderati mensili (circa 60 in tutto per stazione) nel periodo da maggio 2001 ad aprile 2007. La raccolta e la preparazione dei campioni medi ponderati è stata condotta dal CNR-ISE per le stazioni italiane e dal UPDA-SPAAS per quelle svizzere. La concentrazione e l’analisi dei campioni (volumi sino ad 2 L in accordo con la quantità campionata) è stata condotta dal CNR-IRSA di Brugherio. Si è proceduto con una unica tecnica di concentrazione, filtrando il campione attraverso un supporto di Bakerbond Speedisk C18 XF 50 mm, seguendo la metodica EPA n.608/8080 per i pesticidi organoclorurati (OC) e i PCB. Tale metodica consente di estrarre e quan- tificare dal campione acquoso sia gli OC associati al particolato che quelli disciolti nella matrice acquosa. Il risultato è espresso come somma delle due componenti. L’analisi è stata condotta utilizzando la tecnica GC-MS/MS (Trace 2000 - PolarisQ della ThermoElectron) per i composti organoclorurati e la tecnica GC-MS con ac- quisizione in SIM per gli IPA. Evoluzione temporale e andamenti stagionali di POP in deposizioni d’alta quota in Lombardia e Svizzera 95

DDT Totale

700,0 5

4,5 600,0 4

500,0 3,5

3 400,0 2,5 mm 300,0 ng/L 2

200,0 1,5 1 100,0 0,5

0,0 0 DIC. DIC. OTT. FEB. FEB. LUG. NOV. LUG. NOV. LUG. NOV. LUG. NOV. APR. LUG. APR. GEN. GEN. GEN. GEN. MAR. MAR. MAR. MAR. MAG. MAG. MAG. MAG. MAG. SETT. SETT. SETT. SETT. GIUG. AGOS.

2001 2002 2003 2004 2005 SETT.-OTT: 2006 2007 campioni

HCH Totale

700,0 20

18 600,0 16

500,0 14

12 400,0 10 mm 300,0 ng/L 8

200,0 6 4 100,0 2

0,0 0 DIC. DIC. OTT. FEB. FEB. LUG. NOV. LUG. NOV. LUG. NOV. LUG. NOV. APR. LUG. APR. GEN. GEN. GEN. GEN. MAR. MAR. MAR. MAR. MAG. MAG. MAG. MAG. MAG. SETT. SETT. SETT. SETT. GIUG. AGOS.

2001 2002 2003 2004 2005SETT.-OTT: 2006 campioni

264,91 PCB110,11 Totale

700,0 97,77 100

90 600,0 80

500,0 70

60 400,0 50 mm ng/L 300,0 40

200,0 30 20 100,0 10

0,0 0

DIC. DIC.

OTT. FEB. FEB.

LUG. NOV. LUG. NOV. LUG. NOV. LUG. NOV. LUG. APR. APR.

GEN. GEN. GEN. GEN. MAR. MAR. MAR. MAR. MAG. MAG. MAG. MAG. MAG.

SETT. SETT. SETT. SETT.

GIUG.

AGOS. 2001 2002 2003 2004 2005 2006

SETT.-OTT: campioni2007

Figura 2 – Concentrazioni (ng/L) e piovosità (mm/mese) dei campioni di Alpe Devero (a sinistra) e Robiei ( a destra). Risultati e Discussione 4.1 Analisi di OC Per quanto concerne l’analisi di DDT (gli isomeri op’ e pp’ di DDT, DDD e DDE), HCH (gli isomeri alfa, beta, gamma e delta di HCH) e PCB (13 congeneri da PCB18 a PCB194) l’andamento temporale è riassunto in Figura 2. Le due stazioni hanno andamenti simili per PCB e HCH con concentrazioni più elevate nelle deposizioni raccolte tra il 2001 e il 2002 e in quelle del 2005 per PCB e nel periodo 2001-2003 per gli HCH. Il Lindano (gamma-HCH) è il composto prevalente tra gli HCH e i con- generi a bassa-media volatilità per i PCB. Per il DDT la stazione di Alpe Devero, più vicina alla fonte di inquinamento (impianto di Pieve Vergante) risulta più contaminata di Robiei soprattutto nel periodo 2001-2002, successivo all’evento di piena del fiume Toce (ottobre 2000). Non si evidenziano relazioni tra le concentrazioni misurate e la piovosità o andamenti stagionali. Confrontando il carico medio annuale di HCH e PCB nelle stazioni considerate (Figura 3), è evidente la maggiore contaminazione di quelle ad alta quota rispetto a quelle di bassa quota (Pallanza e Locarno), confermando il ruolo del trasporto atmosferico dei composti semi-volatili nella contaminazione delle aree remote. 96 Monitoraggio dei POPs sul territorio alpino

4.2 Analisi di IPA DDT DDT Le concentrazioni totali dei DDT 0,5 16 IPA considerati nella sta- 0,5 0,5 zione di Alpe Devero sono 0,4 0,4 risultate relativamente mo- 0,4

2 0,3 deste nel periodo conside-

2 0,3

2 0,3 rato (2004-2005) (Figura 4)

ng /cm 0,2

ng /cm 0,2 2002-01 con valori minori a 40 ng/L ng /cm 0,2 0,1 2003-022002-01 2004-03 in tutti i mesi ad eccezione 0,1 2003-022002-01 2005-042004-03 0,1 2003-02 0,0 2006-052005-042004-03 di marzo ’05, in cui è stata ri- 2007-062006-052005-04 0,0 Pallanza 0,0 Alpe 2007-062006-05 scontrata una concentrazione Pallanza Locarno Robiei DeveroAlpe 2007-06 Pallanza LocarnoMonti DeveroAlpe Robiei pari a 61 ng/L rispetto a quan- LocarnoMonti Robiei Devero Monti to osservato per la stazione 3,4 1,9 HCH 1,5 3,4 1,9 di Robiei. In tutti i campioni HCH 1,5 3,4 1,9 HCH 1,5 la miscela di IPA è domina- 1,2 1,2 ta dai composti a basso peso 1,2 molecolare (3-4 anelli, PM

2 0,9

2 0,9 ≤202 Dalton), tra cui il fe-

2 0,9

ng /cm 0,6 nantrene si è rivelato il più ng /cm 0,6 ng /cm 0,6 abbondante, altri IPA presenti 0,3 2002-01 2003-02 sono risultati essere: antrace- 0,3 2004-032002-01 2003-022002-01 0,3 2005-04 2004-032003-02 ne, acenaftene e fluorene. Le 2006-05 0,0 2005-042004-03 2007-06 2005-04 0,0 Pallanza 2006-05 concentrazioni di IPA nelle Alpe 2006-05 0,0 Locarno Robiei 2007-06 PallanzaDeveroAlpe 2007-06 Pallanza LocarnoMonti stazioni d’alta quota sono DeveroAlpe Robiei LocarnoMonti Robiei Devero Monti piuttosto costanti rispetto alle PCB variazioni osservate nelle sta- PCB PCB zioni di Locarno e Pallanza. 1,5 Questa variabilità è causata 1,5 da un aumento delle con- 1,21,5 1,2 centrazioni di contaminanti

2 1,2 0,9 nelle deposizioni cadute nei 2 0,9 2 0,9 ng/cm 0,6 2002-01 mesi più freddi. L’incremento

ng/cm 0,6 2003-02 ng/cm 0,6 2002-01 delle concentrazioni nei mesi 0,3 2002-01 2004-032003-02 0,3 2003-02 0,3 2005-042004-03 invernali in aree urbanizzate 0,0 2004-03 2006-052005-04 è riconducibile ad un gene- 0,0 Pallanz 2005-04 0,0 Alpe 2007-062006-05 a 2006-05 Pallanz Devero Locarno rale aumento delle emissioni Pallanz Alpe 2007-06 a 2007-06 a DeveroAlpe Locarno derivanti dagli impianti di Devero Locarno riscaldamento, il cui effetto può essere amplificato da fat- tori meteoclimatici in grado Figura 3 – Carichi di OC (ng/cm2/ mese) per le quattro stazioni considerate. di promuovere un accumulo di contaminanti negli strati Evoluzione temporale e andamenti stagionali di POP in deposizioni d’alta quota in Lombardia e Svizzera 97

Figura 1 – La collocazione delle stazioni di campionamento delle deposizioni atmosferiche inferiori dell’atmosfera. Inoltre, le basse temperature portano ad una diminuzione della tensione di vapore e della costante di Henry degli IPA presenti nella fase gas- sosa, favorendone la condensazione e aumentandone, di conseguenza, l’efficienza di rimozione al verificarsi degli eventi di precipitazione. Se si confrontano gli andamenti dei carichi di IPA totali con quelli delle precipitazioni mensili, si può osservare una certa relazione tra le due variabili nelle due stazioni di alta quota: Alpe Devero e Ro- biei. A conferma di ciò, il coefficiente di correlazione lineare risulta pari a 0,83 per i campioni di Alpe Devero (significativo per un livello di confidenza pari a 99%) e a 0,69 per quelli di Robiei (livello di confidenza pari a 95%), mentre non si osservano correlazioni significative nelle due stazioni poste a bassa quota. La presenza di una correlazione tra precipitazioni e carichi di IPA nei siti ad alta quota è coerente con la relativa costanza delle concentrazioni osservata in queste stazioni durante il periodo di campionamento. In questi siti gli apporti di IPA più elevati si osservano, di conse- guenza, nei mesi maggiormente piovosi: agosto ’04 e marzo ’05 a Robiei; agosto ’04 e ottobre ’04 ad Alpe Devero. I carichi annuali risultano pari a 2,14 ng cm-2 anno-1 ad Alpe Devero e a 12,5 ng cm-2 anno-1 a Robiei. Questi valori evidenziano un apporto di IPA più elevato a Robiei rispetto a quanto misurato ad Alpe Devero.

Conclusioni I carichi misurati per OC e IPA nella stazione di Robiei si discostano in modo con- siderevole rispetto a quanto misurato nella stazione di Alpe Devero, ad eccezione della famiglia di contaminanti di origine locale (DDT). Per gli altri contaminanti la 98 Monitoraggio dei POPs sul territorio alpino stazione di Robiei è caratterizzata sia da maggiori carichi medi mensili di inquinanti nelle deposizioni che da una più elevata piovosità annuale.

Bibliografia CIPAIS, 2002. Monitoraggio della presenza di DDT ed altri contaminanti nell’ecosistema Lago Maggiore. Rapporto Annuale Aprile 2001- Marzo 2002, 89 pp. CIPAIS, 2003. Monitoraggio della presenza di DDT ed altri contaminanti nell’ecosistema Lago Maggiore. Rapporto Annuale Aprile 2002- Marzo 2003, 68 pp. CIPAIS, 2004. Monitoraggio della presenza di DDT ed altri contaminanti nell’ecosistema Lago Maggiore. Rapporto Annuale Aprile 2003 – Marzo 2004, 78 pp. CIPAIS, 2005. Monitoraggio della presenza di DDT ed altri contaminanti nell’ecosistema Lago Maggiore. Rapporto Annuale Aprile 2004 – Marzo 2005, 76 pp. CIPAIS, 2006. Monitoraggio della presenza di DDT ed altri contaminanti nell’ecosistema Lago Maggiore. Rapporto Annuale Aprile 2005 – Marzo 2006, 76 pp. Monitoraggio dei POPs sul territorio alpino Copyright © 2007 - ARPA Lombardia ISBN 978-88-903167-0-8 99

Le possibili ricadute sulla salute della presenza ambientale dei POPs

C. Sala1, P. Carrer2, AC Fanetti2 1 ARPA Lombardia/ Environmental Protection Agency of Lombardy 2 Dipartimento di Medicina del Lavoro, Università degli Studi di Milano/University of Milan

Abstract I POPs (persistent organic pollutants) rappresentano un gruppo eterogeneo di sostanze chimiche così definite in quanto in grado di persistere nell’ambiente e bioaccumu- lare, costituendo pertanto un pericolo per la salute umana e ambientale. L’evidenza scientifica suggerisce che alcuni POPs sono in grado di causare effetti avversi sulla salute umana, in particolare effetti sul sistema riproduttore, endocrino, nervoso e sulla cute. Alcune sostanze sono inoltre state classificate dalla IARC come cancerogeni. Un approccio armonizzato a livello internazionale per ridurre il rischio da esposizione a queste sostanze rimane un obiettivo prioritario.

Introduzione I POPs (persistent organic pollutants) rappresentano un gruppo eterogeneo di sostanze chimiche così definite in quanto in grado di persistere nell’ambiente e bioaccumulare, costituendo pertanto un pericolo per la salute umana e ambientale. I POPs sono com- posti organici per lo più di origine antropogenica, caratterizzati da elevata lipoaffinità, semivolatilità e resistenza al degrado. Queste caratteristiche rendono tali sostanze estremamente persistenti nell’ambiente e in grado di essere trasportate per lunghe distanze nelle zone più fredde del globo. In particolare la regione artica è a rischio di inquinamento da POPs; la regione alpina è altrettanto coinvolta nella possibilità di contaminazione. I POPs includono: PCBs; Diossine; Pesticidi (DDT, Aldrin, Chlordane, Chlordecone, Dieldrin, Endrin, Heptachlor, Lindano, Mirex); altre sostanze chimiche (esacloroben- zene, esaclorocicloesano, idrocarburi policiclici aromatici, ecc.). In condizioni ambientali tipiche i POPs tendono alla bioconcentrazione e presentano un processo di biomagnificazione, raggiungendo pertanto concentrazioni potenzial- mente rilevanti sul piano tossicologico. La bio-magnificazione è l’accumulo di quantità crescenti passando dalle prede ai predatori, avviene per via alimentare, più facilmente per composti decisamente lipofili e per i predatori terminali. La via di penetrazione dei POPs nell’organismo umano di gran lunga più importante è attraverso gli alimenti. I POPs nell’organismo umano, una volta accumulati nel tessuto adiposo, sono in grado di essere rilasciati, rappresentando pertanto un pericolo per la salute umana. Inoltre, con la gravidanza e l’allattamento possono trasmettersi alle generazioni successive. L’evidenza scientifica suggerisce che alcuni POPs sono in 100 Monitoraggio dei POPs sul territorio alpino grado di causare effetti avversi sulla salute umana e gli incidenti storici comportanti esposizione a elevate concentrazioni di POPs ne sono un esempio. Tuttavia anche l’esposizione a basse concentrazioni di POPs può rappresentare un pericolo per la salute umana.

PCBs I PCBs sono stati ampiamente utilizzati in Europa soprattutto in dispositivi elettronici come trasformatori. Nell’UE l’utilizzo e il commercio di queste sostanze è stato note- volmente circoscritto a partire dal 1985. Più recentemente l’utilizzo e la commercia- lizzazione dei PCBs è stata completamente proibita (Regulation EC No 850/2004). Nell’uomo gli effetti di esposizioni acute a elevati livelli di PCBs sono ben documen- tati per via della presenza di due incidenti caratterizzati dal consumo di olio di riso contaminato con PCBs avvenuti rispettivamente in Giappone nel 1969 e a Taiwan nel 1979. Segni e sintomi rilevati nel primo incidente in Giappone includevano: ipersecrezione ghiandolare, pigmentazione delle unghie, occasionale comparsa di affaticamento, nausea e vomito, ipercheratosi e iperpigmentazione cutanea, eruzioni acneiformi, spesso associate a sovrainfezione stafilococcica. Non può essere escluso che una parte degli effetti osservati sia da ricondurre a esposizione a diossine per la presenza di questi composti nell’olio di riso. Gli stessi sintomi sono stati registrati nel 1979 a Taiwan nel corso del secondo incidente. I bambini nati tra il 1978 e il 1985 da madri esposte a olio di riso contaminato presentavano iperpigmentazione, ritardo nello sviluppo fetale, sviluppo cognitivo deficitario fino al settimo anno di età, disordini comportamentali e iperattività. Alcuni autori hanno valutato lo sviluppo di bambini nati da 7 a 12 anni dopo l’incidente di Taiwan. I risultati indicano la presenza di un lieve ritardo mentale nei nati. La presenza dell’effetto è verosimilmente da ricondurre alla persistenza di PCBs nel corpo materno anche a distanza di anni dall’esposizione. L’esposizione all’olio contaminato è risultata associata anche a aumento di infezioni (Lu and Wu, 1985), epatomegalia, rash cutanei e acne. Uno studio condotto da Bertazzi et al. hanno studiato la mortalità di 2100 lavoratori impiegati nella produzione di trasformatori nel periodo 1946-1982. Le morti per tu- more erano aumentate (neoplasie del sistema emopoietico e gastroenterico). Studi condotti sull’animale indicano che gli effetti si manifestano principalmente per esposizioni croniche e includono: effetti avversi a livello epatico, dermico, a livello del sistema immunitario, riproduttore gastrointestinale ed endocrino (tiroide). La IARC ha classificato i PCBs come probabili cancerogeni per l’uomo (gruppo 2A).

Le diossine Le diossine sono sostanze chimiche che non vengono prodotte deliberatamente, ma sono sottoprodotti indesiderati di una serie di processi chimici e di combustione. La Le possibili ricadute sulla salute della presenza ambientale dei POPs 101 famiglia include diversi congeneri, tra cui la TCDD (tetraclorobenzoparadiossina). Essendo queste sostanze altamente persistenti, esse permangono nel suolo e nei sedi- menti che diventano veri e propri serbatoi inquinanti. La via principale di esposizione dei soggetti umani alle diossine è l’alimentazione che contribuisce per oltre il 90% all’esposizione complessiva. I prodotti della pesca e altri prodotti di origine animale rappresentano circa l’80% delle fonti di contaminazione. E’ stato ampiamente dimostrato tra gli effetti associati a esposizione a diossina nel- l’uomo la cloracne. Altri effetti riportati includono neuropatie periferiche, affatica- mento, depressione, epatite, epatomegalia, porfiria cutanea, anche se un nesso con l’esposizione a diossine non è sempre stato chiaramente stabilito. Uno studio condotto da Roegner et al., 1991 per la US Air Force sui veterani del Vietnam che applicarono l’agente Orange (TCDD) non ha dimostrato un eccesso di neuropatie, affaticamento, depressione, epatite negli esposti; risultati analoghi dallo studio di Sweeney et al. Studi di esposizione diretta a diossine provengono dai dati relativi a due incidenti (contaminazione di olio di riso in Giappone e Taiwan, cfr parte relativa ai PCBs). La somiglianza strutturale delle diossine coi PCBs non consente tuttavia di trarre precise conclusioni sull’agente causale della sintomatologia manifestatasi. Fingerhut et al. hanno studiato una coorte di lavoratori esposti a 2,3,7,8- TCDD. Gli autori non hanno evidenziato un aumento di incidenza di tumori precedentemente associati a esposizione a diossina (stomaco, fegato, linfoma di Hodgkin e non-Ho- dgkin), ma hanno evidenziato un lieve ma significativo incremento della mortalità per sarcomi dei tessuti molli e tumore del sistema respiratorio. Due studi recenti hanno seguito la popolazione dell’area di Seveso esposta a diossina dopo l’incidente all’ICMESA. Il primo studio (Pesatori et al., 1993) non ha eviden- ziato aumenti statisticamente significativi del rischio di sviluppare tumori nella po- polazione esposta. Il secondo studio ha evidenziato aumenti significativi di incidenza di tumori. In bambini esposti a diossine e/o PCBs durante la fase gestazionale sono stati riscontrati effetti sullo sviluppo del sistema nervoso e sulla neurobiologia del comportamento, oltre a effetti sull’equilibrio ormonale della tiroide. A concentrazioni più elevate di PCBs e a diossine, i bambini esposti per via transplacentare in fase intrauterina (esposizione accidentale o sul posto di lavoro della madre) presentano alterazioni della cute (ad es. cloracne), alterazione della mineralizzazione dentale, ritardo nello sviluppo, disordini comportamentali, riduzione delle dimensioni del pene in fase puberale, riduzione dell’altezza media nei soggetti femminili in età puberale e deficit dell’udito. A seguito della contaminazione da TCDD nell’area di Seveso è stato riscontrato un aumento del numero medio di femmine nate da maschi esposti. Studi condotti sull’animale hanno evidenziato la comparsa di alterazioni a livello del sistema immunitario conseguenti a esposizione a diossina. La IARC ha classificato la TCDD come cancerogeno del gruppo 1 nel 1997. Le altre diossine non sono state classificate come cancerogene (gruppo 3). 102 Monitoraggio dei POPs sul territorio alpino

Pesticidi organoclorurati (DDT) Numerosi pesticidi rientrano nel gruppo dei POPs. Tra questi si annoverano pesticidi organoclorurati tra i quali particolare rilievo assume il DDT. Il DDT è stato per la prima volta sintetizzato nel 1874, tuttavia le sue proprietà inset- ticide sono state scoperte nel 1939. Nei primi anni 70 molti paesi arrivarono a bandire l’utilizzo dell’insetticida. Attualmente il DDT viene ancora utilizzato nei paesi in via di sviluppo per il controllo della malaria. Il DDT è altamente insolubile in acqua ed è solubile nella maggiorparte dei solventi or- ganici. La sua presenza è ubiquitaria e residui sono stati ritrovati anche nell’Artico. Studi condotti su volontari umani non hanno evidenziato la comparsa di effetti av- versi. Uno studio di mortalità condotto su lavoratori impiegati nella produzione di DDT nel periodo 1964-1987 ha evidenziato la comparsa di un aumentato numero di tumori biliari e epatici, tuttavia non statisticamente significativo. Lo studio ha inoltre evidenziato un eccesso di tumori cerebrovascolari anche se il ruolo del DDT o di altri fattori rimane non chiaro. Gli organoclorurati sono stati associati a insorgenza di effetti immunotossici sia nell’animale da esperimento che in studi sulla popolazione umana. Alcuni composti organoclorurati potrebbero esercitare azione estrogenica e potrebbero avere un ruolo nell’insorgenza di patologie ormono-correlate. In partico- lare negli ultimi anni si è verificato un aumento dell’incidenza di tumore del seno. Il miglioramento delle tecniche diagnostiche e la presenza di programmi di scree- ning ampiamente diffusi può in parte spiegare il dato osservato. Inoltre è noto che il principale fattore di rischio del tumore del seno è rappresentato dall’esposizione a estrogeni. Pertanto, è stato ipotizzato che l’esposizione a sostanze chimiche ad azio- ne estrogenica rilasciate nell’ambiente potrebbe essere responsabile dell’aumentata incidenza del tumore del seno. Negli ultimi anni è stato inoltre osservato un aumento dei casi di tumore del testicolo, di criptorchidismo e di ridotta qualità dello sperma. Questi dati portano a ipotizzare la presenza di un alterato sviluppo del sistema riproduttore maschile durante la vita fetale. In particolare è stato ipotizzato che l’esposizione durante la gravidanza a sostanze chimiche ad azione estrogenica potrebbe alterare il normale sviluppo del sistema riproduttore maschile. Allo stato attuale l’evidenza scientifica non consente di giungere a conclusioni definitive sul possibile ruolo svolto da sostanze chimiche ad azione endocrina (endocrine disruption). Alcuni studi confermano quanto sopra riportato, mentre altri lo smentiscono. La IARC ha classificato il DDT come possibile cancerogeno per l’uomo (gruppo 2B).

Considerazioni conclusive I POPs rappresentano un gruppo di sostanze eterogenee, caratterizzate da elevata persistenza nell’ambiente e capacità di spostamento dalla fonte di utilizzo/emissione a zone più fredde del globo. Inoltre queste sostanze sono in grado di biomagnifica- Le possibili ricadute sulla salute della presenza ambientale dei POPs 103 re e bioconcentrarsi, raggiungendo così concentrazioni possibilmente tossiche per l’uomo. Le problematiche relative agli effetti sulla salute umana esercitati da queste sostanze sono molteplici e includono aspetti quali: • Endocrine disruption: la maggior parte dei POPs è in grado di esercitare effetti ormono-simili, potenzialmente in grado di alterare il normale funzionamento del sistema riproduttore e endocrino. In particolare sono possibili effetti avversi durante lo sviluppo fetale, con alterazione del normale sviluppo intrauterino e comparsa di primitive lesioni che nell’adulto potrebbero evolvere in senso neoplastico. E’ stato inoltre avanzata l’ipotesi di una correlazione tra esposizione a POPs e tumore del seno. • Effetti sul sistema immunitario: alcuni studi indicano la possibilità di modulazione del sistema immunitario da parte dei POPs. • Cancerogenesi: i dati a disposizione non sono ancora conclusivi e non consentono di classificare tutti i POPs come sostanze sicuramente cancerogene per l’uomo. La IARC ha classificato la TCDD come cancerogeno del gruppo 1, i PCBs come probabili cancerogeni per l’uomo (gruppo 2A), il DDT come possibile cancerogeno (2B) e le altre diossine come cancerogene del gruppo 3 (non classificate per cance- rogenicità sull’uomo). Non va inoltre sottovalutato il cosiddetto effetto cocktail: l’esposizione contemporanea a più sostanze chimiche è in grado di modificare l’effetto delle singole sostanze, in funzione di effetti additivi e sinergici. L’aspetto deve tuttavia essere tenuto in consi- derazione anche per effetti opposti, che si annullano a vicenda. Molti dei POPs non sono più utilizzati. La necessità di strategie di riduzione del rischio sono tuttavia ancora necessarie, in particolare, esse riguardano l’utilizzo di sostanze alternative e l’adeguato immagazzinamento dei POPs. Le strategie da adottare devono essere coordinate e armonizzate a livello globale.

Bibliografia Bertazzi, P.A., L. Riboldi, A. Pesatori, L. Radice and C. Zochetti. 1987. Cancer mortality of capacitor manufacturing workers. American Journal of Industrial Medicine, 11: 165-176 Chen, P.H. C.K. Wong, C. Rappe and M. Nygren. 1985. Polychlorinated biphenyls, diben- zofurans and quaterphenyls in toxic rice-bran oil and in the blood and tissues of patients with PCB poisoning in Taiwan. Env. Health Persp. 59:59-65 Fingerhut, M.A. et al. 1991a. Cancer mortality in workers exposed to 2,3,7,8-Tetrachlo- rdibenzo-p-dioxin. The New England Joumal of Medicine, 324:212-218. 8 Goto, M. and Higuchi. 1969. The symptomatology of Yusho (chlorobiphyenyls poisoning), in dermatology. Fuoka Act. Med., 60:409-431. Cited in WHO (1993). Gips, T. 1987. Breaking the Pesticide Habit: Alternatives to Twelve Hazardous Pesticides, International Alliance for Sustainable Agriculture, Minneapolis, Minn. 104 Monitoraggio dei POPs sul territorio alpino

Gladen, B.C. et al. 1990. Dermatologic findings in children exposed transplacentally to heat degraded polychlorinated biphenyls in Taiwan. British Journal of Dermatology, 122:799- 808 Holsapple, M.P., et al. 1991. A review of 2,3,7,8tetrachlorodibenzo-p-dioxin-induced changes in immunocompetence: 1991 update. Toxicology, 69:219-255. Hsu, C-C., M-L. Yu, V-C. Chen and W.J. Rogan. 1994. A six year follow-up of behavioural and activity disorders in the Taiwan Yu-cheng children. American Journal of Public Health Okumura, M. and S. Katsuki. 1969. [Clinical observation on Yusho (chlorobiphenyls poiso- rung).] Fukoka Acta. Med., 60:440-446. (in Japanese). Pesatori, A.C., D. Consonni, A. Tironi, M. T. Landi, C. Zocchetti, and P .A. Bertazzi. 1992. Cancer morbidity (1977-86) of a young population living in the Seveso area. In: DIOXIN ‘92. 12th International Symposium on Dioxins and Related Compounds. Organohalogen Compounds, Volume 10. Finnish Institute of Occupational Health. pp. 271-273. Roegner, R.H. et al. 1994. Air Force Health Study: Ranch Hand II. A Long Term Epide- miological Study of Air Force Personnel Occupationally Exposed to Dioxin Containing Phenoxyherbicides in Southeast Asia. 1992. Epidemiology Research Division, Armstrong Laboratory, Human Systems Division (AFSC), Brooks Air Force Base, Texas 78235. Sweeney, M.H., M.A. Fingerhut, J.C. Arezzo, R.W. Hornung, L.B. Connally. 1993. Peripheral neuropathy after occupational exposure to 2,3,7, 8-TCDD. Am. J. Ind. Med.23:845-858. Hayes Jr. and E.R. Laws Jr. (eds.), Handbook of Pesticide Toxicology: Volume 3, Classes of Pesticides. Academic Press, Inc., Toronto, pp.1409-1470.