1. Persistent organic pollutants (POPs) are hazardous chemicals that resist degradation by physical, chemical or biological pathways. Their persistence enables them to be transported by air, water or other means to remote regions where they have never been used. Since they are liable to bioaccumulate, they may concentrate in living tissues of species higher in the food chain and present a risk to the well-being of human populations and wildlife. There has been growing international concern that releases and emissions of POPs may endanger significant parts of the biosphere as well as human populations.

2. In decision 19/13 C on POPs, adopted at its nineteenth meeting, the Governing Council of the United Nations Environment Programme (UNEP) noted the need to develop science-based criteria and a procedure for identifying additional POPs as candidates for future international action. The Council requested the Intergovernmental Negotiating Committee to establish, at its first meeting, an expert group to carry out this work. The Council requested the expert group to work expeditiously, proceeding concurrently with the intergovernmental negotiating process, to develop criteria for consideration by the Committee in the negotiation of a legally binding instrument. The process should incorporate criteria pertaining to persistence, bioaccumulation, toxicity and exposure in different regions and should take into account the potential for regional and global transport, including dispersion mechanisms in the atmosphere and the hydrosphere, through migratory species and the need to reflect possible influences of marine transport and tropical climates.

3. In the same decision, the Council also requested that due consideration should be given to the work currently under way within the United Nations Economic Commission for Europe to develop a regional protocol on POPs under the convention on Long-range Transboundary Air Pollution. Such a protocol will be adopted and signed at a ministerial meeting of the Parties to the Convention in June, in Aarhus, Denmark. In the protocol, reference is made to an Executive Body decision on information to be submitted and the procedures for adding substances to the protocol are described. The decision was adopted at a special session of the Executive Body for the Convention on 19 March 1998.

4. In July 1997, a letter, jointly signed, on behalf of UNEP, by the Director of UNEP Chemicals and, on behalf of the Intergovernmental Forum on Chemical Safety (IFCS), by the Chair of the IFCS Ad Hoc Working Group on POPs, was sent to IFCS national focal points and other participants as well as to UNEP POPs focal points. The letter requested information that would be useful in the development of criteria and a process for identifying further POPs for international action. The information would be collected, compiled and summarized and presented at the first meeting of the POPs Intergovernmental Negotiating Committee. As of 30 April 1998, 29 countries and 11 organizations have responded to this request. The information contained in the responses from Governments and other participants has been used in preparing this document. The responses are also summarized in tabular form in the annex.

5. The following parameters have been identified in the responses as being of primary interest for criteria development to identify further POPs and have been noted in the responses provided to UNEP:

6. In applying the criteria to individual chemicals, climatic factors may need to be considered. When preparing the response to chemicals which fulfil the criteria, social and economic factors must also be considered.

7. Persistence reflects the ability of the substance to resist degradation. It is the basic property that has given its name to this group of compounds. It may be expressed in various ways. In laboratory testing of new and existing chemicals for the assessment of hazard and risk, the Organisation for Economic Cooperation and Development (OECD) Test Guidelines are often used together with the (OECD) Principles for Good Laboratory Practice, as adopted under the OECD Council Act on Mutual Acceptance Data. There are guidelines for testing the easy biodegradability of substances as well as for measuring their inherent biodegradability. These, however, are primarily focused on differentiating between chemicals that are easily biodegradable and those that are not. Persistent chemicals would probably consitute only a small subset of those which are not easily biodegradable, and may not be identified using existing methods.

8. In those responses which include the persistence criterion, the criterion given is the given is the half-life in air, expressed in days. Modelling shows that substantial quantities of a substance that has a half-life of two days or more still remain in air after 8-10 days. During that time the substance may be transported several thousand kilometres.

9. The 12 POPs specified in Governing Council decision 18/32 C all have half-lives in air equal to or exceeding two days. Air persistence data should be considered together with the presence of the substance in remote regions, as demonstrated by the monitoring of biota or other media.

10. Chemicals with long persistence times in water, soil or sediments have a high potential for accumulation in the medium and also for uptake by living organisms. The pattern of environmental releases or application is of great importance. Substances that are applied once annually, for example some pesticides, may have half-lives of several months and still not accumulate in soil over the years, in spite of repeated applications. Most industrial chemicals are not used in this way. Releases and emissions may occur continuously from many sources, including diffuse release from products.

11. Organic chemicals may have a wide range of persistence, from very reactive to extremely persistent. There is no clear demarcation line between persistent and non-persistent chemicals. Responses suggested the use of a half-life in soil, water or sediments of two or six months as a possible criterion to identify POPs. Factors such as temperature, pH and amount and content of biological fraction may greatly influence persistence in the field, as well as processes such as photolysis and hydrolysis.

12. The 12 POPs specified in Governing Council decision 19/13 C all have long half-lives in water and soil. For some the half-lives are several years.

13. Volatility is expressed as the vapour pressure of a substance. Responses indicate that chemicals with a volatility of less than 1,000 pascal might be considered as possible candidates. Nonvolatile chemicals are not likely to be considered unless they are bound to particles in air. Chemicals that have a vapour pressure higher than 1,000 pascal are gases at normal temperatures and are likewise not likely to be considered as candidate POPs. The volatility criterion should be applied together with persistence in air and/or data on presence in remote regions. Other ways to indicate a substance's tendency to volatilize have been suggested, such as Henry's law constants and fugacity calculations.

14. The volatility of the 12 POPs varies by six orders of magnitude. It should be noted that some of these 12 POPs have a very low volatility expressed as vapour pressure.

15. Responses suggest that the best way to establish whether or not long-range transport occurs is through direct measurement of POPs, e.g., in monitoring programmes in remote locations such as the Arctic or Antarctic, isolated islands, or mountain areas. Measurements in biota and human populations also provide data that may be used in risk assessments. Long-range transport can also include regional transport within a continent, such as transport from a densely populated coastal area with intensive agriculture to inland mountainous regions.

16. The potential for long-range transport may be assessed indirectly by persistence times in air, water or soil and by factors such as volatility. The behaviour of persistent chemicals in the environment is dependent, however, on a host of other factors, such as adsorption to particles, soil binding, etc., which makes prediction difficult.

17. There are means of long-range transport of POPs other than through air or water. For example, migratory birds may accumulate substantial amounts of POPs in their northern summer quarters. A significant proportion of the migrating birds die in their winter quarters in southern regions every year, thus transferring their POPs content to these environments. The opposite transport may also occur. There may also be marine transport through currents, or through repeated dissipation and condensation, as well as through migrating marine species.

18. The long-range transport criterion is therefore basically qualitative in character, and would need to be assessed on a case-by-case basis for each POP chemical.

19. Responses suggest that bioaccumulation should be a possible criterion for identifying further candidates for international action. Bioaccumulation is a measure of the potential for a chemical to concentrate in living tissues. While POPs are diluted by dispersion during long-range transport, bioaccumulation counteracts this effect. Bioaccumulation can best be measured in intact organisms in the laboratory or in the field. It is usually expressed as the bioconcentration factor (BCF) or bioaccumulation factor (BAF). Bioaccumulation measured in this way confirms that uptake takes place and integrates accumulation with biodegradation by the organism. Values between 1,000 and 5,000 for BCF in fish have been used as criteria for identifying POPs. Among factors that influence BCF are choice of species, study design, lipid content of the organism and others.

20. Systematic testing of existing chemicals is ongoing in several national, regional and international programmes, but, to date, only a minute proportion of all commercially available chemicals have been studied for bioaccumulating action. In the absence of data from animal testing, the octanol-water partition coefficient (Kow) has been used as a surrogate. It can be easily measured and even calculated on the basis of molecular formula and structure and correlates fairly well with BCF. It must be noted, however, that it should primarily be used as a screening tool, since by itself it will not tell whether a chemical is actually taken up by the organism, or, if taken up, whether it is actually accumulated. Some chemicals with a high Kow have a molecular weight higher than 1,000. Such large molecules are in general not bioavailable. Some chemicals with high Kow values are extensively metabolized and a high Kow (> 10,000) should therefore always be confirmed by testing the BCF in an animal species.

21. As indicated in the responses, there are at present no internationally agreed quantitative toxicity criteria for identifying POPs. A qualitative approach has been used for specific endpoints. While the other parameters are essentially one-dimensional, toxicity is multidimensional and multifaceted. Any assessment of toxicity requires an assessment of dose. Substances of moderate toxicity may cause concern when they are present in significant doses. Chronic and irreversible effects are also assessed differently from acute and transient effects. It is important to relate no adverse effect levels to possible exposures. Toxicity is therefore essentially a qualitative parameter at the screening stage for presumptive POPs. For the 12 POPs presently under consideration for the global negotiations it is suggested that there are significant potential risks to man and the environment at present levels in exposed populations.

22. When applying criteria for identifying further POPs, other factors, such as dispersion mechanisms, patterns of use, influences of marine transport and tropical climates, may be taken into account, as well as the need to conserve biodiversity and to protect endangered species.

23. Any kind of numerical criteria must be applied with judgement. There are several inherent deficiencies in the scientific database that necessitate caution in using single numbers outside their proper context. The uncertainty related to any kind of scientific measurement is of particular importance in cases where the database consists of a single measurement. Similarly, numbers close to the cut-off values must be thoroughly assessed, regardless of whether they are slightly below or slightly above the cut-off. Lack of appropriate methods for predicting what actually happens in the environment is a continuing problem, as well as lack of mutually agreed assessment procedures.

24. Uncertainties in the database frequently give rise to different interpretations between countries or stakeholders. An important part of the criteria development process is to make sure that these divergences are eliminated to the fullest extent possible and that the reasons for any remaining differences are thoroughly understood. A prerequisite for this to happen is a clear understanding of how political considerations, particularly in the field of science policy, influence assessments of hazard and risk. Countries may make various assumptions in their interpretation of scientific data, for example, accepting one well-conducted lifetime study in one species as sufficient evidence of chronic effects, or always taking the "worst" value for an effect parameter. This is generally a reflection of the level of protection desired in the country. Responses suggest that the criteria discussed above should be applied in a stepwise process, which involves risk assessment as a later stage.

25. There are screening methods available to identify probable POPs with a potential for long-range transport. There is still a need, however, for a continuous development of the predictive tools. The criteria process should identify strengths and weaknesses of existing methods and, as appropriate, make recommendations for method improvement. The ultimate proof, however, is the presence of POP chemicals in remote environments, such as the Arctic, as evidenced by monitoring data. For the foreseeable future there is a need for further developing existing monitoring programmes and establishing such programmes in other relevant regions where they do not exist.

26. For some properties, such as bioaccumulation, there are structure activity relationships (SAR) and quantitative structure activity relationships (QSAR) established within certain classes of compounds. While SAR and QSAR can be highly useful as potential indicators or predictors, predictions based on SAR or QSAR should not be used without sufficient corroborative evidence.

27. In applying the criteria to candidate substances, various socio-economic factors should be taken into consideration, such as the different stages of development of countries, the areas of use of substances and the stakeholders involved and, as appropriate, how to build such factors into the process for identifying POP candidates.

28. Other special considerations may also influence the application of criteria, such as the need to conserve biodiversity and the protection of endangered species. It might be necessary to use stricter cut-off values for specific chemicals where the diversity of economically, socially and culturally important biota is at stake, or where individual species are threatened. The principles set out in the Rio Declaration, especially Principle 15, which includes a reference to the precautionary approach, and the provisions of chapter 19 of Agenda 21 should also be taken into account.

29. In assessing whether a chemical meets the criteria for identification as a POP chemical, its distribution and behaviour in all media should be assessed. This includes air, water, sediment, soil and biota.

30. For the 12 POPs already under discussion a concise assessment of hazard and risk has been published by IPCS. For identifying further candidates there are several data sources that might be of use, including the European Union International Uniform Chemical Information (IUCLID) database, the UNEP/European Centre for Ecotoxicology and Toxicology of Chemicals (ECETOC) Inventory of Critical Reviews on Chemicals, the International Register of Potentially Toxic Chemicals (IRPTC) database, IPCS Environmental Health Criteria Documents, INCHEM, the IPCS CD-ROM publication providing information for the sound management of chemicals and EXICHEM, the OECD existing chemicals database. There are several national databases, including product registers, which also contain useful information.

31. Where biological or chemical measurements are not available, surrogate data might be used. Upward changes in national, regional or global production volumes of a potentially persistent chemical may raise legitimate concerns that warrant further testing and/or assessment. Changing use patterns, particularly increases in open and diffuse uses, are other factors that could trigger interest in a potential candidate substance.

32. Total lack of data for a certain parameter should generally be a signal to all involved to supply the necessary data by testing or other means. In cases where this might prove difficult and it is necessary to process the substance further, default values might be used. It should be borne in mind, however, that such data are provisional in character and should always be superseded by real data.