TL: THE DISPOSAL OF NUCLEAR WASTES INTO THE SEABED BY SHORE- ACCESSED FACILITIES SO: Greenpeace (GP) DT: October 1989 Keywords: nuclear civil greenpeace reports ldc ocean waste toxics / PREAMBLE There is currently discussion in international conventions for the prevention of pollution in the marine environment (the London Dumping Convention, the Paris & Oslo Commissions, and the Helsinki Commission), as to whether a shore-based subseabed disposal operation would constitute dumping at sea and pose a risk to the marine environment. In a paper submitted to the 12th Consultative Meeting of the LDC (LDC12/Inf. 17, October, 1989), Greenpeace argued that the LDC was the appropriate Convention for the regulation of such facilities. However, the Contracting Parties to the LDC have been split on the issue, with principally those states that have a growing interest in this option arguing that it should not come under the auspices of the LDC. As a result, the 12th Consultative Meeting agreed that the Fourth Meeting of the Ad Hoc Group of Legal Experts on Dumping should consider: Legal questions related to the disposal of low-level radioactive wastes into a repository, constructed in bedrock either totally or partially beneath the sea, and accessed from shore; i.e., whether these practices should be considered as "dumping at sea" under the terms of the London Dumping Convention. (Circular letter No. 1421) In this paper we restate those arguments which we believe make the LDC the appropriate Convention, and review the current state of development of this form of nuclear waste disposal. We go further and argue that no form of subseabed disposal of nuclear waste of any category can be justified in the light of currently developing criteria relating to transboundary pollution and risks to the global eco-sphere. THE REGULATORY CONTEXT In the Greenpeace submission to the 12th meeting of the LDC, it was argued that this Convention should apply to shore-accessed facilities for the following reasons: - there was nothing in the wording of the LDC that would prevent the Convention applying. Indeed, the express wording relating to "dumping at sea" had already been held to apply to subseabed disposals with access from either ships or platforms; - furthermore, according to the accepted rules of interpretation for international law (1969 Vienna Convention on the Law of Treaties, article 31), a treaty is to be interpreted in good faith and according to the ordinary meaning to be given to terms in their context and in the light of the object and purpose of the treaty, and Greenpeace has thus argued: the plain meaning of "at sea" entails the concept of "on, in, near or by the sea", as the dictionary definition and that any attempt to argue that "at sea" does not include access from shore to the same type of facility that would be covered by the LDC were it accessed by platform or ship, would contravene the express objective of the Convention which is to protect the marine environment from pollution. (DC 12/Inf. 17, October 1989). - it has also been argued that no other global treaty exists which could adequately regulate this potential hazard to the marine environment, and that were the LDC not to apply, then an obvious means of circumventing the present regulatory powers of <$FText of Footnote>the LDC (for any kind of hazardous wastes) would exist. In support of these arguments, Greenpeace noted the past decisions of the LDC with respect to the application of the Convention to the control of subseabed disposal via ships, where a consensus was reached that the LDC was the appropriate international body to consider this form of disposal. The LDC further resolved that no disposal by this method should take place until it is proven to be technically feasible and environmentally acceptable (including a determination that the waste can be effectively isolated from the marine environment), and also until su0ch time as a regulatory mechanism is elaborated in accordance with the provisions of the LDC to govern the disposal into the seabed of such radioactive wastes. There is thus an effective moratorium on the disposal of nuclear waste (in this case most plans relate to the disposal of High Level Waste (HLW) into the seabed via ships or platforms until these matters are resolved. In the case of shore-accessed facilities (where current plans relate to Intermediate Level Wastes (ILW) which although toxic are not heat-generating), we shall show that the risk to the global marine environment is of a similar nature and that there are no sound technical or environmental reasons for excluding this form of disposal from the LDC. In support of this argument Greenpeace has pointed out that other regulatory agencies have already prohibited the subseabed disposal of high level nuclear waste in addition to the uncontained dumping of Low Level Waste (LLW), and hence have not regarded the additional geological barriers as sufficient to warrant exclusion. For example, the United States Environmental Protection Agency (EPA) ruled that subseabed emplacement constituted " dumping" and was thus prohibited by the Marine Protection, Research and Sanctuaries Act of 1972. A UNEP Regional Seas Convention, the South Pacific Regional Environmental Program's (SPREP) Convention contains a ban on dumping which expressly includes HLW, ILW and LLW disposal. Likewise, the Permanent South Pacific Commission's Protocol for the Protection of the South-east Pacific Against Radioactive Pollution (Cali, Colombia, March 1989) bans the "dumping" and "burial" of radioactive wastes on and in the seabed. Furthermore, the Third United Nations Conference on the Law of the Sea (UNCLOS III) expressly includes the sea, its airspace and its seabed, within any framework for addressing issues related to the sea in general, and pollution in particular. In conclusion, Greenpeace has already presented arguments from the standpoint of international law that the LDC is the relevant body to address the issue of shore-accessed seabed disposal. It has been argued, however (see for example, document LDC 12/6/2 and LDC 16/2/Add.1), that shore-accessed disposal options can be excluded from the LDC for the following reasons: @BULLET = they are intended for LLW or ILW and are not covered by the prohibition on HLW disposal at sea; @BULLET = the current moratorium on LLW dumping applies to uncontained disposal, and is in any case not binding and subject to review by the LDC's Inter-Governmental Panel of Experts on Radioactive Waste (IGPRAD); @BULLET = shore-based disposal sites do not present a threat to the marine environment; @BULLET = the repositories would be in internal waters and thus not under the jurisdiction of the LDC, but subject to regional Conventions such as the Oslo (dumping) or Paris (land-based discharges) Conventions. In the following sections we argue that: @BULLET = the classification systems for nuclear waste whereby HLW is categorised as "heat generating" or by source (e.g. first cycle reprocessing waste), do not fully address the toxic potential of the waste for the marine environment, and that LLW and ILW can contain large quantities of extremely hazardous nuclides but nevertheless not generate heat or not originate from the first cycle of the reprocessing industry; @BULLET = the current moratorium resulted from a vote in which the majority of Parties were against disposal, and continue to regard this option as environmentally unacceptable, and it is doubtful whether the IGPRAD review will do anything but confirm original doubts about the long term consequences of LLW disposal; LLW contamination of the marine environment is not acceptable to the majority of Parties to the LDC; and @BULLET = shore-accessed subseabed disposal presents substantially the same type of risk to the marine environment as either uncontained LLW disposal via barrels (which the majority of Contracting Parties to the LDC regard as unacceptable) or HLW subseabed disposal in the deep ocean (currently prohibited), or offshore disposal via platforms. Furthermore, as these risks accrue to all states as a result of the long timescales and global circulation of the oceans, regional Conventions, alone, are not the appropriate bodies for regulation. @14 SW BD UNNP = 2 CLASSIFICATION OF NUCLEAR WASTES FOR DISPOSAL The classification of nuclear wastes as "High, Intermediate (or Medium) and Low" is in all systems, arbitrary as far as the risk to the environment is concerned. The term "high" applies either to the heat generating capacity of the waste, or to its origin in the nuclear fuel cycle (either spent fuel or reprocessed liquid waste from the first cycle). However, a UK Royal Commission pointed out as long ago as 1976 that this classification did not relate to environmental risk, and that large quantities of nuclear waste from the UK classified as "medium" or "low", because they were neither hot or difficult to handle because of penetrating radiation, nevertheless contained sufficient alpha emitters to make them as hazardous as any other type of waste. Indeed, all HLW currently in store will lose its heat generating capacity over a relatively short time scale (say 300 years) as its major short-lived constituents decay and would then be re- classifiable. The classification was developed primarily for operational handling. The long-lived radiotoxic nuclides, such as plutonium, would remain and it is these which dominate the risk profile which peaks well after most of the heat generating waste has decayed. Thus, a waste may be classified as "intermediate", but contain large amounts of extremely toxic long-lived nuclides. Furthermore, wastes not in the form of spent fuel or glass blocks may have other components which increase the hazard, such as material which may encourage gas formation (organic solvents etc.), and in any case will not be as resistant to leaching. It is for these reasons that the current prohibition on HLW should also be extended to any category of nuclear waste intended for subseabed disposal. As a general rule, if it requires isolation in the first place, it represents a significant hazard! @14 SW BD UN = 3 THE INTERNATIONAL "ACCEPTABILITY" OF RADIOACTIVE CONTAMINANTS IN THE MARINE ENVIRONMENT It is our view that the current scientific and legal studies relating to the disposal of LLW to the sea floor, presently the subject of a moratorium, will confirm the unacceptability of any form of radioactive waste disposal in the marine environment. The moratorium has been effective since 1983, and in the intervening years a number of factors relating to the risks have been clarified. In addition, international opinion has espoused the principle of precaution. Here we briefly outline these developments as they relate to the central argument that the LDC is the appropriate global forum for shore-accessed facilities. Firstly, a number of modelling studies have demonstrated that contamination from LLW sites will eventually surface in areas of active fisheries, and that although individual risk levels would be very low (and perhaps regarded by some experts as insignificant) there would be a collective detriment and hence some measure of "harm". There will of course be debate about the significance of levels of "harm" to human health (e.g. the number of expected cancers compared to the natural rate of this disease, the level of individual risk that should be discounted, etc), but it remains the case that no international agreement is likely on the "acceptability" of risk, however low, in view of the disparity of risk and benefit from the activity that generates the waste. Furthermore, it is evident that the risk to the resources and amenities of the oceans is not "dose-related". That is, a marine resource such as fisheries or recreation can be severely affected at extremely low levels of contamination (and hence dose or risk to individuals). This fact has been borne out by the Sellafield leak of 1983 and the Chernobyl releases of 1986. In the former case, Irish Sea fisheries were affected, and in the latter, farming is still blighted in countries far distant from the accident, despite relatively low levels of risk. Recent concern about coastal nuclear developments such as the Dounreay reprocessing plant in Scotland by neighbouring states such as Iceland and Greenland has demonstrated the international issues raised by the potential for transboundary pollution. Even after an environmental assessment (the Dounreay Inquiry, 1986) such states maintained their opposition, in our view quite justified and legitimate in the light of the hazard potential, to this development. These developments demonstrate a trend of increased sensitivity to the damage that even minute amounts of radioactive contamination can cause to marine resources such as fisheries and tourism. Even where, in the case of direct damage to marine biota and of risk to human health, harm has been predicted to be insignificant according to prevailing scientific models and understanding, neighbouring states and populations within the nuclear state in question, have shown a growing reluctance to place their faith in a scientific methodology which is itself coming into question. In this latter regard, an important factor in the development of the precautionary approach to marine protection has been the growing number of examples of scientific models and predictions either failing to identify a risk, or failing to predict the onset of damage. This has been particularly pertinent in the case of dilute and disperse operations which are not now reversible (CFCs, CO2 and SO2 for example). In conclusion, therefore, we see an increasing trend toward distrust of models and predictions in the assessment of consequences and their role in justification of a practice, together with an increasing inability to agree on levels of harm or risk that can be regarded as acceptable. In short, the whole methodology for the assessment of risk in relation to sea dumping that has been developed by such agencies as the IAEA and the NEA, and is currently being used by developers of the shore-access option, has now come under question and may no longer be relevant to future decision making. @14 SW BD UN = 4 THE RISK FROM SHORE-ACCESSED DISPOSAL In this section we demonstrate that all current shore-accessed developments (one has been built in Sweden, one is under construction in Finland and one under active consideration by the UK) rely on a multi-barrier concept of containment and dilution. In all models used to assess the environmental impact of these developments, the marine environment is regarded as an additional level of safety by virtue of the dilution and dispersion factors. In addition these models deal only with the derivation of doses and express these in relation to a criterion of "acceptability". As we have argued above, they therefore rely upon a limited frame of reference and a methodology that is now out-dated. However, we shall review the current state of knowledge in order to demonstrate this fact. Finally, we shall argue that current shore-accessed developments present substantially the same kind of risk to the marine environment as either uncontained LLW dumping or subseabed HLW disposal, both of which are currently prohibited. As a minimum requirement, there should be a moratorium on all future developments, and an active review of current facilities, until such time as international regulations are agreed. @14-SW-BD-L = 4.1 Radioactive discharges into the marine environment following disposal of Intermediate and Low Level Radioactive Waste The following sections demonstrate the reliance upon the apparent diluting capacity of the overlying water body in all current concepts of radioactive waste disposal into offshore and coastal subseabed strata. Heat generating or HLW are excluded from the detail of these arguments, but many of the comments regarding engineered and natural barriers are equally applicable to them. As yet, however, no definite proposals exist to begin disposal of HLW, whereas ILW and LLW are actively being disposed of in the deep repository at Forsmark in Sweden, and several other repositories are either under construction, or site investigation programmes are under way. Disposal of LLW in shallow engineered facilities is also occurring in several countries at present. In addition to these coastal and offshore options, the UK is alone in considering the possibility of disposing of radioactive waste beneath small offshore islands. In many countries, it is now the policy to dispose of all radioactive wastes in deep repositories where isolation from the biosphere for very long periods is attempted. @12-SW-BD-L = 4.1.1 The Multi-Barrier Concept There is now an international consensus within the nuclear industries, and regulatory agencies concerned with radioactive waste disposal, on a model and description of the system of containment that has become known as the Multi-Barrier Concept (NEA 1988). This makes use of both engineered and natural barriers. Within this concept the environment in which the repository exists is divided into several distinct zones: @BULLET = the Near Field consisting of the wasteform itself and the engineered structure of the repository, as well as the area of geology around it which was disturbed during construction of the repository; @BULLET = the Geosphere consisting of the undisturbed geological formations between the repository and the biosphere; and @BULLET = the Biosphere consisting of the surface layers of soil, rivers, lakes and seas, the atmosphere and plant and animal life. Below, we review the repository concepts in relation to the host rocks, potential migration pathways for escaping radionuclides, and the safety assessments. We look particularly at the role that the marine ecosystem plays in these assessments. @12-SW-BD-L = 4.1.2 Repository Concepts In terms of subseabed disposal of ILW/LLW two technical concepts have generally been considered, and are still under study in several countries, particularly the UK, where the body responsible for developing and operating an ILW/LLW repository, UK NIREX Ltd., has yet to rule out adopting either option. Two options are available, a subseabed repository accessed from the shore, or a repository accessed from the sea. a) A subseabed repository accessed from the shore, involving access via shafts and tunnels from a coastal site, with the main body of the repository beneath the sea or ocean. This concept is that adopted in Sweden, where the repository at Forsmark is accessed by tunnels and shafts 1 kilometre from the shore of the Baltic Sea, and is excavated at a depth of 60 metres below the seabed. A similar repository is now under construction at Olkiluoto in Finland. In the UK, both the present candidate ILW/LLW sites, Sellafield and Dounreay, are situated on the coast, although at this time no final designs have yet been produced, such that it is not certain whether the repository itself will actually be situated below the seabed. However, as described below, the principles of containment for a coastal site are very similar to those for a true subseabed repository. It should be pointed out that the proposals for Sellafield and Dounreay involve repositories at considerably greater depths than those in Sweden or Finland, although as argued below, this does not ensure total isolation from the biosphere. b) A subseabed repository accessed from the sea, involving access from a mobile or fixed platform or an artificial island.  This concept has only really been considered in the UK, and was rejected by NIREX during its site selection procedure, due in part to the potential for uncontrolled accidents during operations (UK NIREX 1989). Such a decision reflects the operational history of accidents in the North Sea in particular, and the worldwide off-shore oil industry in general. @12-SW-BD-L = 4.1.3 Potential host-rocks Studies were begun in the last decade on geosphere parameters, such as host-rock type, primarily concerned with HLW. Three principal rock types were identified (salt deposits, claystones and crystalline igneous and metamorphic rocks) as exhibiting suitable properties such as low inherent permeabilities and low flow-through rates: a) Salt Deposits, which may be many hundreds of metres thick, were an obvious target for nuclear waste disposal, because of their supposed dryness and the tendency for fractures and cracks to be rapidly sealed. Their continued existence for millions of years has also been interpreted to suggest lack of groundwater access. Salt exhibits the ability to flow as a plastic, in certain conditions, as opposed to deforming as a solid, and due to being less dense than most surrounding rocks, is capable of upward movement en-masse, to form complex structures known as salt domes or diapirs. Diapirs are common in parts of the southern North Sea, and in the coastal and offshore areas of several countries of Northern Europe, such as West Germany and Holland; b) Claystones are also considered to possess favourable properties for waste disposal, such as low permeability, plastic behaviour and excellent sorptive properties (the ability to slow down movement of impurities). The low permeability of claystones (also called argillaceous rocks) is characteristic, and they are often found surrounding other more permeable rocks in which oil and gas are found. Avoiding such deposits of oil and gas is itself important in repository siting. Whereas argillaceous rocks are considered potentially suitable for ILW/LLW repositories in continental areas (such as at Mol in Belgium) and as sealing layers in shallow sites, few examples exist of their choice for deep coastal or offshore sites. It should be remembered though that deep-sea clays have been identified as potentially suitable for use in deep ocean floor disposal concepts for HLW, technological uncertainties notwithstanding; and c) Crystalline (Hard) Rocks: These are rocks which have either crystallised from cooling magma (igneous rocks) or been formed by alteration of pre-existing rocks by the effects of temperature and/or pressure (metamorphic rocks). In such rocks, flow of water only occurs via fractures, formed either during cooling or by tectonic activity (folding and faulting), and so detailed knowledge of the geological history of a region is necessary. In many countries, crystalline rocks have been chosen as the favoured media for waste disposal. The only active deep repository, at Forsmark in Sweden, is excavated in such rocks. Both potential sites identified in the UK for further investigation (Sellafield and Dounreay) offer hard rock geologies in coastal settings, as does the repository under construction at Olkiluoto in Finland. @12-SW-BD-L = 4.1.4 Research and Development As mentioned previously, in most countries the disposal policy involves the development of a single deep repository. Only in Sweden has such a repository gone into operation, with one currently under construction in Finland. It is significant to note that in several of the countries currently evaluating the deep disposal concept, and engaged in site selection and/or evaluation, coastal or offshore sites feature in these deliberations (UK, Sweden, Finland, Netherlands, Brazil). Although this paper is primarily concerned with disposal of ILW/LLW, it should also be pointed out that considerable international effort has been expended into the examination of proposals to dispose of HLW into deep ocean sediments, by either drilled emplacement or penetrator emplacement (see NEA 1988). @14-SW-BD-L = 4.2 Potential migration pathways As described in the preceding section, the multi-barrier approach depends on the performance of the three component parts: near field; geosphere; and biosphere. The near field is predominantly man-made, and the performance of the barriers needs careful scrutiny in terms of the possible pathways to the biosphere. Several potential pathways by which radioactivity could move away from a repository have been identified: groundwater pathways; gaseous pathways; human intrusion; and natural intrusion. @12-SW-BD-L = 4.2.1 Groundwater It is openly accepted within the radioactive waste management industry that at some stage, water is going to gain access to the waste and its environment (UK NIREX 1988). The near-field barriers are the first line of defence against the outward movement of radioactive leachate, but are essentially designed to slow down the process, not to prevent it. Water is expected to penetrate the backfill, corrode the waste canisters and overpack and slowly dissolve and mobilise the waste elements. Transport to the biosphere will then take place. Obviously, one of the most important characteristics of a rock chosen as a host for a waste repository is that of low permeability. That is why the three rock types described in section 4.1.3 were selected. Unfortunately, it is not always possible to either predict the overall permeability of a rock body or indeed to accurately measure that permeability in boreholes during site investigation. This uncertainty can be broadly quantified, as mentioned by Knill in 1989: @INDENT ITALIC = It would be prudent to assume that consistent permeability values below 10-10 m/s are unlikely to be encountered (in crystalline rock) at shallower depths than 500 to 600m. This observation would tend to suggest that repositories sited at shallower depths (e.g. Forsmark and Olkiluoto) are subject to particular uncertainties regarding even such basic characteristics as permeability and hence flow-through rates. This potential for error in prediction must be borne in mind during safety assessment of repository proposals, outlined below. In the context of this paper, it is important to realise that coastal areas (including small islands) are unique in their groundwater flow systems, due to the interface of shallow, land- derived fresh water overlying deeper, denser, saline water. It has been assumed, particularly by NIREX in the UK and by SKB in Sweden, that the hydraulic gradient (and hence the flow rate) will be very low below coastal and marine areas, and that any flow which does occur will be in an offshore direction and take place over a long time period (NIREX 1987). The potential for error in this assumption was again pointed out by Knill (1989): @INDENT ITALIC = It is apparent that a repository site located in fresh water near the coastal margin will be in an environment within which there would be both an increased hydraulic gradient and a relatively short flow path to the biosphere. @12-SW-BD-L = 4.2.2 Gaseous pathways Gas will be generated due to microbial action on organic material in the waste itself (producing mainly methane and some radioactive carbon dioxide) and due to the corrosive action of the water which enters the repository on concrete and steel structures (producing mainly hydrogen). The production of gas has potentially important consequences. If it is unable to move quickly through the near field and out of the repository, pressure may rise locally, and this could result in damage to the engineered structures or the opening of fissures in the surrounding rocks, which could in turn affect the groundwater movement. A general assumption, which however still remains unsubstantiated, underpins all work on gas migration. This is that a continuous connection exists between the repository and the surface via fractures and capillaries within the surrounding rocks (Rees and Rodwell 1988). Presumably such potential gas migration is to be expected above repositories sited below marine areas. The cover of 60m to the floor of the Baltic at Forsmark (and 70-100m at Olkiluoto) suggests that gas emanation is a distinct possibility during the early post-closure phase of the repository. @12-SW-BD-L = 4.2.3 Human Intrusion The potential for inadvertent intrusion into a radioactive waste repository cannot be overlooked. Most facilities are only planned to have a 50-100 year operational life, and the possibility must therefore exist, over the long timescales envisaged (impacts are typically modelled and calculated over tens of thousands of years), that knowledge of the exact whereabouts, and even function, of the site may be lost. It is for this reason that assessments must be made of the likelihood of the surroundings of the site being seen by future generations as a potential source of minerals. Repositories sited in sedimentary sequences in shallow offshore areas are probably the most vulnerable to possible intrusion by drilling, as these are also favoured areas for oil and gas exploration. @12-SW-BD-L = 4.2.4 Natural Intrusion This pathway includes possible disruption of the geosphere and biosphere due to major natural phenomena such as seismicity and, due to the long timescales involved, consideration must be made particularly of the potential effects of climate change upon the repository site. Although major earthquakes are obviously potentially damaging to surface facilities and shafts, smaller tremors and even microseismicity can be significant, especially in coastal areas, where sedimentation is often rapid, and where subsidence can be locally highly lar. Indeed, many offshore areas are fault-controlled troughs, with generally poor records available of seismic activity. Not only can such changes or events cause damage to the physical structure of the repository, but, more importantly, they can produce changes in the geological environment, affecting both the near and far field, in terms of hydrogeology and structural integrity. The potential for climatic disruption of coastal and near-shore repositories due to changes in sea-level, and hence changes in hydraulic gradient, is likely to be greater than for land-based sites, and those repositories such as Forsmark and Olkiluoto, involving relatively shallow depths of excavation, are particularly vulnerable. At one time it was generally accepted that due to the effects of post-glacial rebound in the Baltic, the site of the Forsmark repository would be dry land within about 1000 years of the present. With global warming now an expectation, a greater degree of uncertainty is brought into the analysis. Certainly the hydrogeological features of the site cannot be predicted with confidence over the timescales in which the waste will remain hazardous. @14-SW-BD-L = 4.3 Safety Assessments The principal objective of safety assessment is to quantify any potential risks to the public that may arise at any time following the abandonment of a repository. Many countries are actively involved in safety assessment studies concerned with their own particular waste management strategy and chosen host-rocks. Many of these studies are, however, generic. That is, they utilise generalised parameters for geological and hydrogeological phenomena, and input these into mathematical models. Many of the data input into these models are not site-specific, due in most cases to the fact that specific sites have not yet been selected. The end result and main purpose of a safety assessment is to show, albeit in fairly general terms, the potential doses to humans likely to occur following the closure of a repository, and its subsequent invasion by groundwater. It is also important to relate this dose to the radiological regulations in force in the particular country. @14-SW-BD-L = 4.4 Representative Safety Assessments @12-SW-BD-L = 4.4.1 Sweden The repository at Forsmark is the only active repository for ILW in the OECD, and is situated 1 kilometre offshore and approx. 60 metres below the bed of the Baltic. At the present rate of post- glacial rebound (6mm/year), the area is expected to be dry land within approx. 1000 years. However, as noted earlier, global warming casts uncertainty on these predictions. Safety analyses were carried out at the site prior to licensing. This is not the place to discuss the validity of that modelling, given the uncertainties highlighted recently by Morner (1989) regarding the structural integrity of the site. What is important in the context of this paper, however, are the assumptions made regarding movement and subsequent discharge of radioactive leachate. The siting below the Baltic means that large volumes of brackish waters are available for dilution. (Carlsson et al 1989). A so-called "salt water period" of up to 2500 years following closure is envisaged before the land is uplifted, due to the effects of post-glacial rebound, during which this sub- marine discharge is expected to occur. @12-SW-BD-L = 4.4.2 Finland Similar safety assessment work has been carried out concerning the planned repository at Olkiluoto, 70-100 metres below the cape of the island adjoining the site of the active reactor, which is expected to be in operation by the end of 1992. A comprehensive site-specific safety analysis was completed in 1986. Here again, the important point was made that the overall measure for safety is the radiological impact to Man. (Peltonen et al 1989). One of the major conclusions of this work was that the groundwater passing the silos ends up in the surrounding marine area. Doses were calculated and compared with the regulatory target, assuming certain "critical groups" who ate large amounts of marine-derived products. No account was taken of the possible direct effects of sub-marine radioactive discharges on the marine ecosystem. @12-SW-BD-L = 4.4.3 United Kingdom Whereas the two studies referred to above were concerned with specific repository locations, both coastal, the UK has yet to identify a site for a repository. As part of the radioactive waste research programme, the United Kingdom Atomic Energy Authority (UKAEA) carried out a series of assessments of several potential repository scenarios, reporting in 1988. The study is usually referred to as the CASCADE project (Comparative Assessment of Concepts and Areas for Deep Emplacement). This was of course based on very little site- specific data, although many of the assumptions and interpretations were later used to justify the choice of the two candidate sites, Sellafield and Dounreay. Of direct consequence to any examination of the coastal and offshore repository concept are the unresolved issues highlighted by this study. These include variations in fluid density between freshwater, seawater and brine and transient effects due to changes in sea level (UKAEA 1988a). These echoed directly the comments made by Knill (1989) mentioned above. Inherent in the calculations of dose are major assumptions concerning the likelihood of discharge of any potentially radioactive leachate which may move away from a repository, into the marine environment. This likelihood is irrespective of the scenario under examination. For example: @BULLET = For a small island site, the report states: In all cases considered the discharge areas are offshore, but usually only a short distance (p15). @BULLET = For seaward-dipping sediments: Some of the pathlengths may be overestimated because they go a considerable distance offshore...before discharging on the seabed (p34). @BULLET = For a low relief hard rock area:...the majority of the flow is seaward...with discharge occurring at or near the coast (p40). As far as dose levels are concerned, the dependence on marine discharge is stated unequivocally: @BULLET = Biosphere dilution is an important element in meeting the regulatory target. (p78). @BULLET = ...all the studies demonstrate the benefit to be gained from having the discharge into a marine environment (p84). @BULLET = ...if any (radionuclides) are released, for coastal or offshore repositories release will be to a marine environment (p86). @BULLET = There are clear advantages in having the discharge from the groundwater pathway into a marine biosphere....A marine biosphere can be guaranteed by having a repository offshore or on the coast (p88). This report has been quoted in some detail to show the basic dependence upon discharge to a marine environment inherent in the UK safety assessment programme. The CASCADE work was used to help justify the selection of two coastal sites for further investigation in March 1989. In 1989, UK NIREX made this prediction: @BULLET = For the first 10,000 years after repository closure it is expected that radionuclides migrating in groundwater would emerge into a marine environment. @14-SW-BD-L = 4.5 The Marine Ecosystem It is transparently obvious from these three studies that one of the major assumptions used in the safety assessments of coastal and offshore repositories is the discharge of migrating radionuclides into the adjacent marine water body. All calculations of dose to "critical groups" are then based on assuming a diet of mainly marine-derived foodstuffs. No assessment is made in any of the analyses of the likely effects on the marine ecosystem in isolation. When the ICRP published its guidelines in 1977 on environmental monitoring of radionuclide levels, the primary source of information on the effects of radiation on aquatic organisms was a 1976 IAEA report. This stressed that although no deleterious effects would be expected at the calculated dose rates used, from current discharge data, only limited data had been collected for many groups, with no data available for any marine mammal or sea- bird. A report produced over 10 years later, in 1988, showed that very little radiological research on such organisms had been carried out (Thompson 1988). This report concludes: @INDENT ITALIC = ...the hypothesis that wildlife species are safeguarded by current discharge limits remains untested. As most of the safety analyses referred to above concern only the effects of submarine discharges on humans, many of these comments remain valid today, and a recent paper by Linsley (1989) in the Bulletin of the IAEA illustrates the generalised approach still being applied. In this paper, the only scenario reviewed involves the impact on the marine environment of direct dumping of LLW from ships, a practice currently subject to a moratorium. It is, however, stressed that: @INDENT ITALIC = ...available information on the effects of radiation on non-human species is limited and the effects of these assessments must be treated with caution. Despite this acknowledged uncertainty, the IAEA has merely concluded that: @INDENT ITALIC = ...at the maximum release rates for the dumping of LLW permitted under the existing definition for the London Dumping Convention, it does seem possible that some environmental impact could result ( Linsley op op. cit.). Given the uncertainties in radiological impact due to radioactive discharges, the IAEA is under some pressure to fulfill its mandate from the 1958 UN Conference on the Law of the Sea, quoted below in full: @INDENT ITALIC = The IAEA should pursue whatever studies and take whatever action is necessary to assist states in controlling the discharge or release of radioactive materials to the sea, in promulgating standards, and in drawing up internationally acceptable regulations to prevent pollution of the Sea by radioactive materials which would adversely affect Man and his marine resources. The IAEA and other regulatory bodies are studying the effects of radiation on other organisms. Woodhead and Pentreath (1989) have noted that: @INDENT ITALIC = ...it is becoming increasingly necessary to assess the complete range of consequences for a variety of different options for managing the disposal of a given radioactive waste, and such assessments should include consideration of the potential effects on populations of wild organisms. At the same time, however, they have concluded that it would be "unrealistic" to believe that information on the doses and effects of radiation to other organisms would ever be available in as much detail as for man. @14 SW BD UNNP = CONCLUSIONS 1. The disposal of ILW/LLW in coastal/offshore environments is being actively considered or even practised in numerous countries. 2. Discharge of migrating radionuclides is assumed to occur into the marine environment in all such scenarios, and is actively promoted as an additional safety feature due to the dilution effect. 3. Little or no account is taken in safety assessment studies of the impact of such discharges on the marine environment. Only the doses likely to be received by humans in so-called "critical groups" in direct contact with the marine biosphere are considered. 4. The potential impact of future discharges, which could occur over timescales of thousands of years, and which would be subject to dispersal to international waters, perhaps on a global scale, has not been assessed by any of the appropriate international Conventions for the control of marine pollution. 5. The case for disposal into any media, as compared with indefinite monitorable and retrievable storage, has not been proven, and even more so with respect to subseabed operations with the potential to contaminate international waters. 6. In view of the nature of the potential impacts upon marine life, fisheries, amenities and human health, it is imperative that shore-accessed radioactive waste disposal is brought under the remit of the appropriate global Convention, in this case, the only global authority being the London Dumping Convention. 7. With regard to the present situation, where there is one operating repository and plans for more in the Baltic and North or Irish Seas, Greenpeace calls for a moratorium on further development, and an assessment of the present operation, in line with the Principle of Precautionary Action. @14 SW BD UNNP = SCIENTIFIC REFERENCES Carlsson et al 1989; Safety assessment of SFR- A Swedish repository for low and intermediate level radioactive waste. In: Proceedings of an International Symposium on the safety assessment of radioactive waste repositories. Paris, October 9th-13th 1989. IAEA 1976; Effects of ionising radiation on aquatic organisms and ecosystems. Technical Report Series No 172. IAEA Vienna. ICRP 1977; Recommendations of the International Commission on Radiological Protection. ICRP Publication 26. Pergamon Press. Oxford. Knill J 1989; Keynote address at: Radioactive Waste Management 2. Proc. BNES Conference Brighton. May 1989. Linsley G 1989; Protection of natural ecosystems: Impact of radiation from waste disposal practices. IAEA Bulletin Vol 31 No.4. Morner NA 1989; In: A critical review of the Swedish Final Repository for radioactive waste (SFR)- "Forsmarkslagret". Publ. The Swedish Society for the Conservation of Nature and The Swedish Antinuclear Movement. Nuclear Energy Agency 1988; Geological Disposal of Radioactive Waste- In situ Research and Investigations. OECD Paris. Nuclear Energy Agency 1988b; Feasibility of disposal of high level radioactive waste into the seabed. Vols 1-8. OECD Paris. Peltonen EK et al 1989; Use of safety assessment for repository systems development in Finland. In: Proceedings of an International Symposium on the safety assessment of radioactive waste repositories. Paris, October 9th-13th 1989. Rees JH and WR Rodwell 1988; Gas evolution and migration in repositories- Current status. NIREX Safety Series. (NSS/G104). Thompson PM 1988; Environmental monitoring for radionuclides in marine ecosystems: Are species other than Man protected adequately? J Environ. Radioactivity 7. UK NIREX Ltd 1987; The Way Forward. (discussion document). UK NIREX Ltd 1988; Presentation of the Nirex Disposal Safety Research Programme. (NSS/G108). UK NIREX Ltd 1989; Deep Repository Project. Preliminary Environmental and Radiological Assessment. (PERA). United Kingdom Atomic Energy Authority 1988a; Comparative Assessment of Concepts and Areas for Deep Emplacement. (CASCADE) (Phase 1) Results. (NSS/A204). United Kingdom Atomic Energy Authority 1988b; Research and Safety Assessment. (NSS/G100). D.S. Woodhead, R.J. Pentreath 1989; Effects of Radioactive Waste Disposal on Marine Organisms. Presented to the Seminar on the Radiological Exposure of the European Community from Radioactivity in North European Marine Waters. Bruges, Belgium, 14-16 June, 1989