Depleted Uranium in Bosnia and Herzegovina
Post-Conflict Environmental Assessment

United Nations Environment Programme (UNEP) 10mar03

This file contains excerpts from the UNEP Report.
For the full report at http://postconflict.unep.ch/publications/BiH_DU_report.pdf
280 page 12.2 MB PDF file

Foreword

In 2001, the United Nations Environment Programme (UNEP) published the findings from the first-ever assessment on the environmental impact of the use of depleted uranium (DU) originating from a real conflict situation. This work was conducted in Kosovo in 2000 and followed-up one year later in Serbia and Montenegro. Since then, UNEP has become a reference in the scientific community regarding the impacts of DU when used in a conflict situation. When, in the summer of 2002, the Council of Ministers of Bosnia and Herzegovina (BiH) requested UNEP to conduct a similar assessment in BiH related to the use of DU ordnance in 1994-95, UNEP was naturally ready to initiate action.

In this new study, we learn that more than seven years after the end of the conflict it is still possible to detect DU in soil and sensitive bio-indicators at sites where DU had been used. A large number of contamination points (holes were DU penetrators hit the ground), as well as loose contamination, including DU penetrators, fragments and jackets/casings were found. UNEP could confirm local DU contamination around impact points, although the levels were low and no significant level of radioactivity could be measured.

Importantly, for the first time during an assessment in the Balkans, it was possible to detect DU contamination in drinking water. The contamination, however, was very low and remained below the World Health Organization’s (WHO) reference value. Finally, DU was also detected in several of the air samples where it had been unexpected to find any DU particles in the air so long after the end of the conflict. Again, detected levels remained below international safety limits. However, for precautionary purposes, confirmation of DU contamination inside some buildings leads UNEP to recommend to the local authorities decontamination and clean-up measures.

The mission also analysed the handling and storage conditions of radioactive sources within BiH. The representative from the International Atomic Energy Agency (IAEA) provided valuable analysis on these issues. During this challenging work, our cooperation with BiH authorities has been excellent. The government shared their scientific and health expertise with UNEP, as well as their important civil protection and mine clearance experience. NATO/SFOR co-operated with UNEP throughout the study, and UNMIBH, as our local UN partner, helped make this work possible in many ways.

All of the scientific members on this mission were experienced from earlier UNEP assessments. I want to congratulate these scientists not only for a work well done, but also for producing new and valuable information on the behaviour of DU. Close cooperation with our colleagues from the IAEA and the WHO was a success. Health related information was presented and reviewed by the WHO during meetings with hospitals and government health officials. The WHO assessment, as the competent United Nations agency on health issues, is included in this report.

This work could never been conducted in such an efficient manner without the professional work by the national institutes of Greece, Italy, Russia, Sweden, Switzerland, the United Kingdom and the United States, ensuring the highest quality discussion and results. Above all, my gratitude goes to the governments of Italy and Switzerland that provided UNEP with experts, laboratory assistance and generous financial support. Following this third DU assessment in the Balkans, the collective information from these reports can now be used to minimize any health and environmental risks from depleted uranium. These studies confirm that the behaviour of DU is a complex issue, and that DU can be found in soil, vegetation, water and air in certain conditions many years after the conflict.

For this reason, UNEP strongly encourages further studies in the areas where risks could be higher than in the Balkans.

Klaus Töpfer United Nations Under-Secretary-General Executive Director of the United Nations Environment Programme

**********************

Introduction

The question on environmental and health impacts originating from the use of depleted uranium (DU) ammunition has, after several conflicts, become a much debated issue. Since there has been very little scientific fieldwork with proper measurements as well as laboratory work outside of the military community, until recently it has been difficult to come to any significant conclusions.

In the autumn of 2000, UNEP carried out the first-ever international assessment on the environmental behaviour of DU following its use in a real conflict situation. In March 2001, UNEP published the report, entitled Depleted Uranium in Kosovo - Post-Conflict Environmental Assessment.

To reduce the uncertainties about DU's environmental impacts, a second phase was carried out in Serbia and Montenegro with a field mission in October 2001. This study investigated six sites, as well as - for the first time - one targeted military vehicle, which was studied in detail. UNEP subsequently published the report, entitled Depleted Uranium in Serbia and Montenegro - Post-Conflict Environmental Assessment in the Federal Republic of Yugoslavia in March 2002.

In Serbia and Montenegro, authorities had already conducted some decontamination and clean-up operations, which were in line with the findings of the UNEP field studies. Both the Kosovo and the Serbia and Montenegro reports were well received by local stakeholders, as well as by the international scientific community. These reports helped alleviate some of the

public concerns with respect to DU by scientifically demonstrating the low contamination levels and providing recommendations to reduce future risks at affected sites.

The request by the Bosnia and Herzegovina (BiH) authorities to conduct similar studies over seven years after the use of DU was a new challenge for the scientists in UNEP's team. 15 international experts comprised the UNEP mission to BiH, which took place on 12 - 24 October 2002.

UNEP had selected 15 sites to be visited during the mission. One of the sites was unfortunately inaccessible due to the heavy presence of mines. For the remaining 14 sites, the presence of mines and other unexploded ordnance (UXO) was a factor that occasionally restricted the work to a degree. Five of these fifteen sites were areas where NATO had reported using DU munitions. The remaining 10 sites were areas where the local population or authorities were concerned that DU might have been used.

The possible health risks and questions for safe storage of radioactive waste were integrated into the tasks of this mission. Therefore, experts from the relevant UN agencies - the WHO and the IAEA - participated on this mission. The valuable contributions and recommendations made by these experts are included in this report.

A total of 128 samples were collected during the mission: 46 surface soil, 5 smear, 2 scratch, 3 soil profiles of 60 cm, 19 water, 24 air, and 29 vegetation samples. Both the Swiss Spiez Laboratory and the Italian APAT Laboratory conducted sample analyses. Of the 14 sites investigated, three clearly showed DU contamination, confirming the earlier use of DU ordnance. These sites correspond to the information on DU targets provided by NATO.

Four new and significant findings are contained within this report. First, detailed laboratory analyses of surface soil samples revealed low levels of localized ground contamination. At most, local ground contamination could be detected around contamination points at distances below 200 meters, but usually much closer. None of the sites showed 'pure' widespread contamination, meaning a contamination over large surfaces in the range of a couple of hundred meters. Ground surface DU contamination detectable by portable beta and gamma radiation detectors was typically limited to areas within 1 - 2 meters of penetrators and localised points of contamination caused by a penetrator impact.

Second, penetrators buried near the ground surface and recovered by UNEP had decreased in mass by approximately 25% over 7 years. Based on this finding, correlated with those penetrators studied in UNEP's earlier studies, a DU penetrator can be fully oxidized to corrosion products (e.g. uranium oxides and carbonates) in 25 to 35 years after impact. Following that time period, no more penetrators – metallic DU – will be found buried in the Balkans soil. In contrast, penetrators lying on the ground surface showed significantly lower corrosion rates.

Penetrators and fragments fully corrode 25 - 35 years after impact

Third, for the first time, DU contamination of drinking water could be found at one site. DU could be clearly identified in one drinking water sample. A second drinking water sample from a well also showed traces of DU, but was detectable only through the use of mass spectrometric measurements. Contamination of the well water may be due to the fact that the well is positioned in what would have been the line of attack by planes. The concentrations are very low and the corresponding radiation doses are insignificant for any health risk. This is also true considering the toxicity of uranium as a heavy metal. However, because the mechanism that governs the contamination of water in a given environment is not known in detail, it is recommended that water sampling and measurements should continue for several years, and that an alternative water source should be used if DU is found in the drinking water.

Finally, the presence of DU in air was found at two sites, including air and certain surface contamination inside two buildings at two different sites. Resuspension of DU particles due to wind and/or human activities from sources such as contamination points, corroded penetrators or fragments laying on the surface are the most likely cause. The concentrations were very low and resulting radiation doses are minor and insignificant. However, as some of these buildings are currently under use by the civil population or by military, UNEP considers exposure to such a source unnecessary. Therefore, precautionary decontamination and cleanup steps for these buildings are recommended.

In addition to these key findings, some important remarks must also be added. Throughout the mission, the UNEP team observed that workers and civilians, as well as military and mine clearance personnel with access to sites where DU presence was confirmed, were unaware of or misunderstood the risks and issues surrounding DU ammunition. Awareness raising activities should be considered, including information about DU in general, associated risks, handling and storage and contact information for relevant authorities. A flyer or leaflet, like the ones used to advocate mine safety, could be produced and distributed.

The importance of having correct locations and coordinates for DU-affected sites and of obtaining access to these sites for the purpose of conducting surveys and measurements is essential. The longer the elapsed time since the date of the attack, the more difficult it is to implement countermeasures, including decontamination, if necessary. As 6 coordinates of confirmed attack sites are still missing according to the NATO web page, these coordinates should be disclosed without delay.

Another important issue related to information on what had happened to the radioactive material that had previously been collected and stored in BiH. During the assessment study, UNEP wanted to confirm the whereabouts of a box containing DU penetrators collected earlier from Had•ici. The information received from NATO confirms that NATO/SFOR military authorities have properly stored it outside of BiH.

UNEP also visited certain ammunition destruction sites to confirm that DU had not been included among detonated ammunition, as well as to analyse another environmental aspect; the contamination by heavy metals as a result of such destruction activities. Selected water and soil samples were analysed for their heavy metals content. High surface soil contamination of heavy metals was measured at three sites. Such contamination could represent a future health risk. Results indicate that past ammunition production, as well as current ammunition destruction activities, have produced heavy metal contamination of the soil. Ammunition destruction sites should therefore not be situated in areas where secondary contamination could occur, for example, contamination of the groundwater and any animals grazing nearby.

Overall, the findings of this study are consistent with the findings of UNEP's earlier DU studies. The levels of DU contamination are not a cause for alarm, but some uncertainty remains with respect to future potential groundwater contamination from penetrator corrosion products. Both general and site-specific recommendations are included in this report for follow-up and implementation.

This study is UNEP's third contribution to the scientific debate on the environmental risks and the behaviour of DU. UNEP is committed to working with other UN organisations to extend DU studies to other post-conflict regions where the long-term effects of DU contamination should be studied. As part of this commitment, UNEP was invited in Spring 2002 by the IAEA to participate in a DU mission to Kuwait.

I would like to extend my genuine thanks and appreciation to all the national and international experts that worked so hard to contribute to the success of this study. All scientists have made excellent contributions. I would like to extend my gratitude to three in particular. Jan Olof Snihs, from the Swedish Radiation Protection Authority (SSI), has been the Scientific Leader of all three UNEP DU studies in the Balkans. His role in keeping the scientific quality of these reports at a high level has been exceptional. Gustav Åkerblom, also from SSI, has been the Technical Leader of each mission. Based on his experience, appropriate methods for finding and measuring DU ammunition have been developed by UNEP. Finally, Mario Burger, from Spiez Laboratory in Switzerland, has been a key scientist in all three missions in the Balkans and, for the Bosnia and Herzegovina assessment, acted as UNEP's Project Coordinator. Without their respective dedicated and professional work, the UNEP assessments on depleted uranium would not have been possible.

Based on its work in the Balkans, UNEP strongly encourages further assessments to be undertaken in other regions and climate zones where DU has been used in earlier conflicts in order to reduce any uncertainties about its potential environmental impacts in the longer term.

Conflicts and wars are never good news. I believe that the findings of this study will contribute both to conflict-prevention and to the protection of human health and environment debate during times of conflicts.

Pekka Haavisto Chairman, UNEP Depleted Uranium Assessment Team Geneva, 10 March 2003

********************

Background

2.1 UNEP'S ROLE IN POST-CONFLICT ENVIRONMENTAL ASSESSMENT

UNEP's Post-Conflict Assessment Unit first emerged in May 1999 as a joint UNEP/UNCHS (Habitat) 'Balkans Task Force' with the aim of producing an overall assessment of the consequences of the Kosovo conflict on the environment and human settlements. Its particular focus was on the Federal Republic of Yugoslavia (Kosovo, Montenegro and Serbia).

As part of this work, an international expert group - the 'Depleted Uranium Desk Assessment Group' - was appointed to "assess the potential health and environmental impact of depleted uranium (DU) used in the Kosovo conflict". However, the use of DU in Kosovo had not been officially confirmed at that time and no information was available on the locations of sites possibly targeted by DU munitions.

Thus, the work was carried out, inter alia, by:

• collecting background information on the potential effects of DU on human health and/ or the environment, the quantity and quality of depleted uranium used in the conflict, and the locations of affected sites;

• assessing the medium- and long-term potential health and environmental impacts of DU used in the Kosovo conflict by means of a scenario-based desk study;

• undertaking a fact-finding mission to Kosovo to make preparations for a possible future sampling campaign; and

• analysing information in order to both quantify problems 'on the ground' in potentially affected areas and to provide qualitative answers concerning the possible risks to human health and the environment.

The fact-finding mission did not encounter elevated levels of radiation, either in and around the wreckage of destroyed military vehicles, or on/ alongside roads. Based on these preliminary measurements, UNEP concluded that there was no evidence or indication of the presence of DU at the locations visited. However, it was stressed that any further investigations could only be meaningful if and when confirmation was received that DU ammunition had been used and, if so, where.

Such confirmation arrived in July 2000. Following approaches from the United Nations Secretary-General, NATO made available a detailed list of sites where DU had been used. Operating under the newly formed 'Balkans Unit', UNEP then moved quickly to assemble a team of international experts to prepare a scientific mission to Kosovo from 5-19 November 2000. In March 2001, UNEP published the findings in the report 'Depleted Uranium in Kosovo' (see section 2.2 for a summary of these findings).

With the continued help of NATO and local authorities, a similar mission to Serbia and Montenegro (FRY) took place from 27 October to 5 November 2001. Building on the Kosovo report, this mission also took air samples in addition to water, soil and lichen samples. The report, 'Depleted Uranium in Serbia and Montenegro', was published in March 2002 (see section 2.2).

The success of these missions created a need for UNEP to expand its scope beyond the Balkans. In late 2001, the 'Balkans Unit' became the Post-Conflict Assessment Unit (PCAU) in order to "extend the work to other areas of the world where the natural and human environment has been damaged as a consequence of conflict".

In mid-2002, UNEP received an official invitation, this time from Bosnia and Herzegovina (BiH), to make a third DU assessment focusing on the use of depleted uranium during air attacks against armoured vehicles, tanks and artillery positions in 1994 and 1995. The benefits of this project would also extend beyond the boundaries of BiH, as important new information would be discovered on the environmental behaviour of DU more than seven years after its use in combat. In order to assess the feasibility and safety of such an assessment so long after the end of the conflict, a fact-finding mission was undertaken from 5-14 September 2002. Following the success of this undertaking, the third DU assessment mission took place from 12-24 October 2002.

2.2 SUMMARY OF FINDINGS FROM THE KOSOVO AND SERBIA/MONTENEGRO MISSIONS

In Kosovo, the mission did not find any widespread contamination of the soil or ground surface, though some localized points of contamination were identified at some of the sites where the use of DU had been reported. The major part of ground contamination was found in the upper 10-20 cm directly below a penetrator. No DU contamination of water or domesticated cow milk was found during the mission and subsequent laboratory testing, and there was no evidence to suggest any immediate health problems. However, it was concluded that there could be future risk of DU contamination of groundwater. Analyses of bio-indicators (i.e. lichen, bark, moss and grass) at four sites indicated that DU had been used at these sites, but did not uncover any conclusions about the aerosolisation of DU or airborne contamination.

In the subsequent Serbia and Montenegro mission, all the sites investigated had previously been visited, cleaned, fenced-off and assessed by the FRY authorities. This had not been the case in Kosovo. UNEP could not find any significant contamination of the ground surface or the soil except at localized points of concentrated contamination. Nine penetrators and 13 contamination points were identified. The penetrators were removed and the contamination points marked for later decontamination by the FRY authorities. However, laboratory analyses of soil samples enabled contamination to be detected several metres from contamination points. DU contamination was found in some soil samples within the fenced areas (i.e. the target areas). With the exception of Cape Arza, none of the soil samples collected outside the fenced areas showed any DU contamination. Thus, there was no indication that DU had spread outside the fenced areas or over a large distance. Importantly, however, the contamination levels inside the fenced areas were of such a low level that they were considered insignificant from the human health point of view.

In terms of groundwater contamination arising either from DU at contamination points or from more widespread ground contamination, the possible consequences in Serbia and Montenegro were insignificant. The general conclusion for the five sites investigated in Serbia was that there were no penetrators remaining on the surface in the areas that were searched by UNEP. However, at some sites there were indications that penetrators (and contamination points) were present outside the searched areas and might be present outside the areas fenced-off by the FRY authorities in Serbia. There were good reasons for believing that most of the DU rounds fired against targets at the investigated sites did not fragment, but instead entered the ground more or less intact. In this case, the buried penetrators constitute a source of uranium that might, in the future, influence the concentration of uranium in drinking water. During the mission and subsequent laboratory testing, there was no detectable DU in any of the water samples.

Two of the sites showed a clear indication of DU in the air sampled. However, digging for penetrators was undertaken at the same time as the operation of the filters used for air sampling, making it difficult to find an unequivocal explanation for this finding.

As was found in the 2000 Kosovo mission, lichen appeared to be a reliable indicator of airborne DU contamination. Of the lichen samples taken in Serbia and Montenegro, only those obtained from four sites showed any significant indication of DU.

2.3 DEPLETED URANIUM

What is depleted uranium?

Depleted uranium is a by-product from the process used to enrich natural uranium ore for use as fuel in nuclear reactors and nuclear weapons. It is distinguished from natural uranium by differing concentrations of certain uranium isotopes. Natural uranium has a uranium-235 (abbreviated as U-235 or 235U) content of 0.7 per cent, whereas the content of U-235 in DU is reduced to about one-third of its original content (0.2-0.3 per cent). The U-235 content in DU used in DU ammunition in the Balkans was found to be 0.2 per cent (UNEP 2001; UNEP 2002).

Like naturally occurring uranium, DU is an unstable, radioactive heavy metal that emits ionizing radiation of three types: alpha, beta and gamma. Because of its radioactivity, the amount of uranium in a given sample decreases continuously but the socalled 'half-life' (the period required for the amount of uranium to be reduced by 50 per cent) is very long - 4.5 billion years in the case of the isotope uranium-238 (U-238 or 238U). Therefore, the level of radioactivity does not change significantly over human lifetimes. The unit of measurement for radioactivity is becquerel (Bq), 1 Bq being the disintegration of one atom per second.

When uranium decays, another nuclide or isotope is created, which in turn is also radioactive, leading to a long chain of radionuclides (uranium daughter products) being produced (see Appendix O 'Data on Uranium'). DU is roughly 40 per cent less radioactive than natural uranium and, consequently, less radiotoxic. This is because during the industrial process by which uranium ore is converted to uranium metal, uranium is chemically separated from all its daughter products beyond U-234, i.e. radium, radon and others.

In the enrichment process used for the production of nuclear fuel, the uranium concentration of the isotope U-235 is enriched from 0.7 per cent in natural uranium to roughly 4 per cent in the uranium destined for fuel in nuclear reactors.

The by-product is uranium with a lower concentration of U-235, i.e. depleted uranium (DU). The U-235 concentration in the DU produced is usually 0.2-0.3 per cent. In enrichment plants, U-235, which is slightly lighter in mass than U-238, is used to separate the two isotopes, allowing the enrichment process of U-235.

Since U-234 is an even lighter isotope, its concentration is correspondingly higher in fuel uranium and lower in DU when compared with natural uranium. The fact that DU has lower concentrations of U-235 and U-234 than natural uranium also explains why DU is less radioactive than natural uranium. Data on the specific activity of DU are given in Table 2.1.

Uranium occurs naturally in all rock, soil, water and biota. The typical concentration of activity - expressed as specific activity (activity per mass unit) - of U-238 in the Earth's crust is 5 to 125 becquerels per kilogram (Bq/kg), equivalent to 0.5-10 mg/kg (1 mg/kg = 1 ppm = 1 gram/tonne). Typical values for U-235 are around 0.2 to 5 Bq/kg. The specific activity of U- 238 in uranium ore of good quality (0.5 to 30 per cent uranium) is 0.6· 105 to 3.6· 106 Bq/kg. The specific activity of pure uranium metal in radioactive equilibrium with its immediate decay products is 50.23· 106 Bq/kg (50.23 Bq/mg natU). Details on the specific activity of uranium in soils, rocks, water and air are given in Appendix O (Tables O.18 and O.19).

The overwhelming part of the radiation emitted from the nuclides in the U-238 series is emitted from the isotopes that follow after U-234. Compared with the sum of the energy of alpha radiation emitted per transformation from all isotopes in the U-238 series, the isotopes that follow after U-234 emit about 89 per cent of the alpha energy, roughly 58 per cent of the beta radiation energy and about 98.6 per cent of the gamma radiation energy (Appendix O, Table O.4).

If reprocessed uranium from a nuclear reactor is used (fully or partially) as feed material in the enrichment process of uranium, or if this was the case during earlier runs of the technical facilities of the enrichment plant, the DU may contain tiny traces of fission products, uranium isotopes and transuranic elements that are specific to reprocessed reactor fuel. In DU penetrator material found during earlier UNEP missions to the Balkans region (UNEP 2001; UNEP 2002), traces of U-236 and Pu-239/240 could be identified. U-236 was analysed around 0.003 per cent (mass per cent), and Pu-239/240 contamination of the DU was around 20 Bq/kg (10-2 micrograms per kilogram), which is equivalent to the very low content of one plutonium atom per 100 billion uranium atoms. This indicates that the DU found in the Balkans came into contact with reprocessed uranium at some point during its fabrication process. The concentration of contaminating nuclides is indeed so low that their contribution to the total radiation dose of DU is insignificant and can be neglected in assessing risk to humans or the environment.

Uranium occurs naturally in the +2, +3, +4, +5, and +6 valence states, but it is most commonly found in the hexavalent form at the Earth's surface. In nature, hexavalent uranium is commonly associated with oxygen as the uranyl ion, UO2 2+. The different uranium isotopes are chemically identical and thus exert the same chemical and toxicological effects.

Metallic DU reacts chemically in the same manner as metallic uranium, which is considered to be a reactive material. The general chemical character of uranium is that of a strong reducing agent, particularly in aqueous systems. In air at room temperature, solid uranium metal oxidizes slowly and first assumes a golden-yellow colour. As oxidation proceeds, the colour darkens and at the end of three to four weeks, the metal appears black (Blasch et al., 1970).

Metallic DU, particularly as a powder, is a pyrophore, which means that it spontaneously ignites in air at temperatures of 600-700ºC. When DU burns, the high temperatures oxidize the uranium metal to a series of complex oxides, predominantly triuranium octaoxide (U3O8), but also uranium dioxide (UO2) and uranium trioxide (UO3) (RAND, 1999).

Upon oxidation, uranium metal first forms UO2. A typical oxidation rate for massive uranium metal would be penetrations of 0.005 mm/day (0.19 mg/cm2 per day) at 175ºC. Significant oxidation of UO2 does not occur except at temperatures above 275ºC (Bennellick, 1966). Uranium oxides are sparingly soluble in water but in a moist environment will gradually form hydrated oxides. Under such conditions, the addition of 0.75 per cent titanium to DU metal used in penetrators appears to slow the oxidation rate by approximately a factor of 16 (Erikson, 1990).

Microbial action can speed the corrosion rate of uranium. The corrosion rate is controlled by several variables, including the oxygen content, presence of water, size of metal particles, presence of protective coatings and the salinity of any water present. The principal factor controlling corrosion is the size of the particles and hence, surface area. Thus, small particles of uranium metal, produced by abrasion and fragmentation, corrode rapidly, whereas large masses of uranium metal usually corrode very slowly. In the long term, all uranium metal will oxidize to U4+ and U6+ (US AEPI, 1994). Studies carried out on penetrators collected by the UNEP DU mission to Kosovo in 2000 showed that impact on the ground causes numerous fine cracks in penetrators (UNEP, 2001). This favours increased rates of corrosion and dissolution. Rapid corrosion was further confirmed by studies made on penetrators collected during the 2001 UNEP mission to Serbia and Montenegro (UNEP, 2002).

DU can expose people to radiation both from the outside (external radiation) and from the inside (internal radiation) if DU enters the body by inhalation or ingestion. The harmful effect of such radiation is mainly an increased risk of cancer, with the magnitude of risk depending on the part of the body exposed (particularly exposure of the lungs through the inhalation of insoluble compounds) and on the radiation dose.

Like naturally occurring uranium and other heavy metals, DU is also chemically toxic. The toxic effect depends on the amount ingested by the body and the chemical composition of the uranium. Depleted uranium's toxicity is normally the dominant risk factor to consider in the case of ingestion.

For complete, specific data on uranium and depleted uranium see Appendix O. The military uses of DU are summarized in Appendix N.

2.4 ASSESSING THE RISKS

The concept of risk, its meaning and application are discussed in detail in Appendix A 'Risk Assessment'. The following is a summary, intended to equip readers with the necessary background for interpreting the Overall Findings, Conclusions and Recommendations presented respectively in Chapters 4, 5 and 6 of this report.

'Risk' can either refer to the probability, sometimes possibility of occurrence of a given event, or to the consequences of an event if it occurs. A third possibility is a combination of both probability and consequence. Regardless of how the term is used, it is clear that scientific quantification of a given risk has to be expressed clearly and concisely so that appropriate judgements and responses can be made.

The effects of being exposed to DU are both radiological (i.e. due to radiation) and chemical (i.e. as a result of biochemical effects in the human body). Corresponding health consequences may, depending upon the dose or intake, include cancer and malfunction of body organs, particularly the kidneys.

In order to avoid consequences developing from day-to-day procedures in which radioactive and toxic materials are used, a range of applicable standards have been established, including limits for exposure to radiation and toxic materials. However, although such limits and standards exist, these do not imply that if these values are surpassed that there will automatically be severe or adverse consequences, such as serious illness. Wide safety margins are built in before any unconditional or high probability of serious illness could occur. But, from a safety point of view, such a situation would be unacceptable.

A potential way to judge the consequences of events or circumstances where DU exposure may have occurred is to compare findings, measurements or assessments with natural levels, and with given 'safety' limits or standards (see Appendix O). In this report, the consequences are those that might be caused by intake of DU through ingestion or inhalation and/or through external radiation exposure to DU.

The consequences of radiation may be expressed directly in terms of the radiation dose, which is measured in millisieverts (mSv) or microsieverts (µSv). Comparisons can be made with natural levels and with established limits and action levels. Consequences of radiation, in this report, are considered insignificant for doses less than 1 mSv per year (or per infrequent event), and significant for doses higher than 1 mSv. Because there is an assumption of a linear non-threshold relationship for biologically detrimental effects of ionizing radiation, there is also a decreasing probability of occurrence with decreasing radiation doses. Therefore, an insignificant radiation dose means, in reality, a low and insignificant probability of getting a serious illness from that dose as compared with the overall probability of contracting that same illness from all other potential sources.

With respect to chemical toxicity, consequences are treated as insignificant in this report for concentrations or total intakes below applicable health standards or guidelines, and significant for those above.

In the site-specific findings in Chapter 7, judgements of risk are made on the basis of measured DU ground contamination and measurements of possible DU contamination of drinking water and air. The relationship between measurements and risks are discussed in Appendix A 'Risk Assessment'. There is also a summary of risk assessment in relation to a given situation (known as the Reference Case and taken from the 1999 UNEP DU Desk Assessment Report). This assumes ground-surface contamination of 10 g DU per square metre, hereafter referred to as the Reference Level.

Some levels of exposure lead to significant risks (consequences, radiation doses, intakes, as compared with chemical toxicity standards), others to insignificant risks. If ground contamination is less than 0.1 to 1 g/m2, the consequences are normally insignificant. In the current report, the risks considered and assessed - in terms of significance or insignificance of consequences for the environment and human health - are the following:

• If there is widespread measurable contamination of the ground surface by DU, there is a risk that some DU will become airborne through wind action and subsequently be inhaled by people. There is also a possibility of contamination of food (fruit, vegetables, meat, etc.) and drinking water.

• If there are localized points of concentrated contamination (referred to in this report as 'contamination points'), there is a risk of contamination of hands and/or of direct ingestion of contaminated soil. There is also a possible risk of airborne contamination and contamination of drinking water.

• Solid pieces of DU lying on the ground surface - either fragments of or complete penetrators - can be picked up by persons completely unaware that they are handling uranium. Consequently, there is a risk of being exposed to external beta radiation and to internal radiation (i.e. from inside the body) if corroded DU dust or DU fragments enter the body.

• A large percentage of DU rounds that hit soft targets, or missed the intended target completely, will have penetrated into the ground and become corroded over time (to a widely varying degree, depending on site-specific environmental conditions). As a result, there is a risk of future contamination of groundwater and nearby wells used to supply drinking water. There is also a risk that DU fragments will be brought up to the surface through reconstruction activities.

As more than seven years had elapsed since the attacks with DU munitions in BiH (1994- 1995), the conditions influencing the environmental consequences have changed and, thereby, the risks to people. For instance, the risks of airborne contamination from resuspension of DU dust on the ground surface should decrease over time due to the expected dispersion into the ground by dissolution in water, as well as an increasing cover of grass, leaves, etc. On the other hand - and for the same reasons - the probability of water contamination increases over time as DU from surface dust and corroded penetrators enters the water table.

Furthermore, over the aforementioned seven-year period, people may have been exposed to any of the risks described in Appendix A. The possible health consequences of such exposures need to be taken into account by the relevant competent bodies within BiH.

The risks of contamination from touching a penetrator on the ground increase, given the possibility of hands or clothes becoming contaminated by corroded DU and the risk of subsequent internal contamination through ingestion. However, this increased risk may be offset by the decreased probability of finding a penetrator that is hidden by vegetation. In conclusion, and as discussed in further detail in Appendix A, the overall risks from DU decrease with time.

******************

Appendix J

DU in Surface Deposits and Special Studies on Surfaces

J.1 DU IN SURFACE DEPOSITS

Background

When penetrators impact on the ground surface, a portion of its DU mass is transformed into aerosols or fine particles and thrown into the surrounding air. The quantity of DU dispersed into the air mainly depends on the hardness of the surface where the impact takes place. The quantity dispersed is greater on hard surface impacts than on softer surfaces. Consequently, aerial dispersion of small DU quantities is expected after impact on soft soil (soil without stones), higher quantities on hard ground surfaces (stony soil, rock, concrete, asphalt), and the highest quantities when the impact occurs on the heavy armour of a tank or APC.

These aerosols and fine particles are normally deposited in measurable quantities on the surrounding ground or on other surfaces within about 100 m from impact. After initial deposit, it is possible that fine DU dust particles are resuspended into the atmosphere together with soil-dust by wind or human activities, leading to secondary air contamination. These particles are then deposited once more on the surrounding ground and other surfaces. If the deposition takes place on surfaces other than soil that are exposed to rain and other meteorological phenomena, the surface deposit will be partly washed off. Surface deposits on soil will penetrate into the topsoil layer with time.

If the deposition of DU aerosols and fine particles takes place on surfaces which are not affected by rain and other meteorological phenomena (e.g. inside a building), surface deposits will accumulate and remain undisturbed over a longer time and can later be collected and analysed by taking smear or scratch samples.

Measurements of surface deposits

During the mission to Bosnia and Herzegovina (BiH), special samples were taken inside a wooden storage barn at the Han Pijesak Artillery Storage and Barracks site. Inside this building, shot holes on the concrete floor indicated that the building had been attacked and hit by DU rounds. After the attacks, the DU contaminated building was repaired and used once again to store army material, such as cannons and instruments in wooden boxes. However, the detailed history of the building’s management is not known.

One scratch sample was taken from the edge of the concrete floor against the wall; a second one was collected from the horizontal surface of a wooden beam at a height of about 1 m above the floor surface. Two smear samples were taken from smooth, painted horizontal surfaces of army material that had been stored in the barn: one from a cannon, another from a wooden box. The description of the samples and the analyses of results are summarised in Tables J.1 to J.7.

The scratch sample from the rough concrete floor surface consisted of sand and dust.

DU content of uranium in scratch samples exceeded 99% 23.8g of the material was collected from a surface of 420 cm2. The uranium concentration of the material was 1.89 milligram per gram, representing a surface contamination of 107 µg U/ cm2, or 1.07 g/m2. This uranium concentration is approximately 1 000 times higher than the natural uranium content of soil. The isotope composition shows that the uranium consists to almost 100 % of DU.

The scratch sample from the rough surface on a wooden beam consisted of sand and dust. 2.38 g of the material was collected from a surface of 200 cm2. The uranium concentration of the material was 92 µg/g, representing a surface contamination of 1.1 µg U/cm2 or 11 mg/ m2. This uranium concentration is approximately 100 times higher than the natural uranium content of soil. The isotope composition again shows that the uranium consists of almost 100 % depleted uranium.

These two scratch samples mainly represent the primary deposition of debris and dust from the impact of the DU penetrators on the concrete floor inside the building. It is unlikely that this coarse, sandy material was resuspended inside the building at a later time. The measuring results for these samples show that inside a building, the primary surface contamination from impacts of DU penetrators can, as expected, be higher than in the open field. As it is indoors, no influence of weathering effects can occur and the initial superficial contamination will remain preserved at the ground surface, if the floor is not cleaned.

Both smear samples from smooth painted surfaces consisted of fine brown dust. The measured loose surface contamination from the cannon was 5.9 ng U/cm2, or 59 µg/m2. The measured loose surface contamination from a wooden box was 27 ng U/cm2, or 270 µg/m2. The isotope composition for both smear samples shows that the uranium consists of almost 100% depleted uranium.

These two smear samples represent the secondary deposition of resuspended contaminated dust, resulting from the impact of the DU penetrators on the concrete floor inside the building. Because the detailed history of the building management is unknown, it is indeed not possible to discern the length of time elapsed since the dust was deposited on the sampled surfaces. Measurement results from these samples show that – inside a building – the secondary deposition of resuspended dust from contaminated ground surfaces can lead to a DU contamination of objects’ surfaces that were brought into the building only after the attack. Indeed, the DU concentration was found to be about 1 000 times less than the primary contamination on the ground surface. There is no influence from weathering effects, and the superficial contamination will accumulate and remain preserved on the surface of the objects for a long time.

p222

An important observation is the detection of DU in a smear sample taken from a metal surface at a distance of about 400 m from the attacked barrack (sample code NSI-smr-07-16).

Assuming that:

• DU contamination of this surface is 5 times higher (by using the above-mentioned ratio between the scotch tape and total density of the surface contamination), i.e. – 0.16 mg m-2;

• Deposition velocity of released DU particles at this distance is 0.01 m s-1;

• Air dilution factors for a distance of 400 m for weather stability class D is 5 10-7 s m-3 and 4 10-6 s m-3 for stability class F

In that case, the DU release in the atmosphere as a result of an attack can range from 4-33 kg for weather stability class F and D respectively. If it is also assumed that about 10% of DU can suspend in air, it can be estimated that the total DU used in the attack at the Han Pijesak Artillery Storage and Barracks was 40-330 kg, or 150-1100 penetrators. This assessment is thought to be quite reasonable.

Conclusions

• The use of the Inspector device gives the possibility to find places with DU surface contamination densities higher than 0.1 g m-2.

• The median value of DU contamination density of the concrete floor inside the attacked storage barn (Han Pijesak site) was 0.7 g m-2.

• The part of collected contamination on scotch tape is equal to approximately 20 % of the total density of the surface contamination.

• The smear sampling method permits establishing the contamination of surfaces by DU at a level of 3 ng m-2.

• Detection of DU fallout at a distance of about 400 m from the storage barn has allowed to evaluate the total mass of DU used in the attack at a level of 40-330 kg (i.e. 150-1100 penetrators).

******************

Results p.227

This following results are based on the Dutch target and intervention values (see above).

Vogosca Ammunition Production Site

In all selected soil samples from this site (NUC-02-030-001 to –005), high concentrations of chromium (280 – 408 mg/kg) and nickel (179 – 330 mg/kg) were recognized. The target values for these metals were exceeded by several factors. Additionally, in most of the samples the intervention values were already reached. A future detailed assessment of the situation concerning the heavy metals for this site could be considered based on these results.

Kalinovik Ammunition Destruction Site

The soil sample NUC-02-031-002 showed high contamination of zinc (1 900 mg/kg), arsenic (90 mg/kg), cadmium (6 mg/kg) and lead (1 000 mg/kg). The target values for these metals (based on the Dutch target and intervention values for soil) were exceeded by several factors. Moreover, in most of the samples the intervention values were already reached. The other soil sample (NUC-02-031-003) showed an indication of contamination by heavy metals of the neighbouring environment. The water sample also showed heavy metals contamination.

The overall picture from this analysis suggests a detailed assessment be carried out concerning the heavy metals present at this site, the more so since it is situated in a karstic region and might be the source of streams and rivers supplying drinking water.

Kalinovik region in general

Results from the other soil samples taken at the Kalinovik water reservoir indicate that naturally high levels of Fe, Mn, Ni, Cu, Pb, etc. already exist. However, too few samples were taken to give more than an indication or come to any definite conclusion. The soil in the area is a thin layer on the limestone, residual in nature, and is formed by in situ weathering. In such soils, the concentration of metals is known to be frequently (much) higher than in the underlying limestone.

Bjelasnica Plateau – Ammunition Destruction Site

In both samples from this site (NUC-02-035-001 and -002), a high contamination of copper (in the range of 2000 mg/kg), zinc (~460-1650 mg/kg) and lead (~290-600 mg/ kg) could be measured. The situation is similar to the one mentioned above for the Kalinovik ammunition destruction site. However, what is alarming in the results for this site – in addition to the lead contamination - is the very high copper concentration in the samples taken.

Specific results are presented in the tables on the following pages.

****************

L.2 BACKGROUND

What diseases might be associated to DU, and how strong is the evidence? What health impacts could we expect to find in BiH if it were confirmed that there had been relevant exposure to DU?

Health effects would depend on the route and magnitude of exposure (ingestion, inhalation, contact or in wounds) and the characteristics of the DU (such as particle size and solubility). The potential effects of DU on human health could be due to its chemical form that enters the body, which could lead to both chemical and radiological effects.

In terms of chemical toxicity, uranium can cause kidney damage in experimental animals, and some studies in humans also suggest that long-term exposure may result in pathological damage to kidneys. The types of damage that have been observed are nodular changes to the surface of the kidney, lesions to the tubular epithelium and increased levels of glucose and protein in the urine.

Radiological toxicity comes from DU decay, mainly through emission of alpha particles. These particles do not have the ability to penetrate the skin. However, if ingested or inhaled, they may have an effect on lung or gut epithelium. Exposure to alpha and beta radiation from inhaled insoluble DU particles may, in principle, lead to lung tissue damage and increase the probability of lung cancer. Similarly, absorption into the blood and retention in other organs, in particular the skeleton, is assumed to carry an additional risk of cancer in these organs. In all such cases any additional risk of cancer will depend on the severity of radiation exposure. At low levels of exposure to radiation, the additional risk of cancer is thought to be very low.

• Depleted uranium and uranium are essentially the same, except that the content of 235U is three times lower in DU. Consequently, DU is less radioactive than natural uranium and, thus, a radiation dose from it would be about 60% lower than that from purified natural uranium with the same mass. It is assumed that prior knowledge from scientific (experimental, clinical and epidemiological) studies on uranium can be applied to DU.

• Up to now no adverse health effects of DU have been established in the limited epidemiological studies that have been undertaken. DU may, in principle, cause both nephrotoxic effects and internal exposure to radiation (through inhalation, or wounds contaminated with DU). However, these have not yet been confirmed.

• No consistent or confirmed adverse chemical effects of uranium have been reported for the skeleton or liver. No reproductive or developmental effects have been confirmed in humans.

• In a number of studies on uranium miners, an increased risk of lung cancer was demonstrated, but this has been attributed to exposure from radon decay products. Because DU is only weakly radioactive, very large amounts of dust (in the order of grams) would have to be inhaled for the additional risk of lung cancer to be detectable in an exposed group.

• Risks for other radiation-induced cancers, including leukaemia, are considered to be very much lower than for lung cancer.

• However, evidence is inadequate to completely dismiss an association with lymphatic and bone cancer, even though most studies have shown no effect. Veterans from the 1991 Gulf war who have had DU fragments in their soft tissues since the Gulf war are excreting raised uranium concentration, but neither increased rates of lung and bone cancers nor of leukaemias have been detected among them.

As the current debate on the possible adverse effects of potential DU contamination has focused on cases of leukaemia in the military, it is important to assess the known facts regarding leukaemia and DU. While ionising radiation is known to cause leukaemia, the risk is proportional to the level of radiation exposure. Such exposure from DU is calculated to be low.

Even in war zones under extreme conditions and shortly after the impact of penetrators, the inhalation and ingestion of DU contaminated dust, as determined by the amount of dust that can be inhaled, has been calculated to result in a radiation exposure of less than 10

millisieverts, which represents around half the annual dose limit for radiation workers. Such an exposure is thought to result in only a small proportional increase in the risk of leukaemia, of the order of 2% over the natural incidence of the disease. This increase in the incidence rate is so low that it is fully covered by the annual fluctuation of the background (natural) occurrence of this disease. Furthermore, no increase in leukaemia could be observed in uranium miners, or in workers milling uranium for nuclear reactor fuel elements. Finally, a minimum of ten years is usually needed between exposure to ionising radiation and a clinical manifestation of cancers (i.e. a longer period than the time since the conflict in BiH).

From the existing knowledge of DU and its health impacts, and assuming that a large enough group of the population may have had sufficient exposure to DU, it appears unlikely that any significant increase in cancers and leukaemia would be observed in the elapsed time since the armed conflict.

*************

Appendix N

MILITARY USE OF DU

N.1 MILITARY SOURCES OF DEPLETED URANIUM

Depleted uranium has multiple uses by military forces. One of its uses, as in the civilian sector, is to serve as counter-ballast in both aircraft and missiles. However, not all counterballasts are made of depleted uranium. Because of its high density (19.0 g/cm3) and resistance to penetration by anti-armour munitions, depleted uranium can also be used in the armour of tanks, although not all tanks have depleted uranium armour.

Depleted uranium is also used in anti-armour munitions and has several properties that make it ideal for this purpose. For example, when a depleted uranium penetrator hits armour, the rod begins to self-sharpen, thereby enhancing its ability to pierce the armour. Since DU is pyrophoric, during this self-sharpening the depleted uranium forms an aerosol, creating fine DU particles that may burn. The amount of depleted uranium which forms as an aerosol will depend upon the munition, the nature of the impact, and the type of target (i.e. whether it is an armoured vehicle or not). Both tanks and aircraft can fire depleted uranium munitions, with tanks firing larger calibre rounds (105 mm and 120 mm) and the aircraft firing smaller calibre rounds (25 mm and 30 mm).

Many of the world’s armies possess, or are thought to possess, DU munitions (RAND, 1999). Depleted uranium munitions are conventional weapons and have been used in warfare. As such, these munitions are readily available on the open market to other armies. Munitions containing DU were used in Iraq during the 1991 Gulf War, as well as in Bosnia and Herzegovina BiH) in 1994-1995. In the 1999 Kosovo conflict, NATO A-10 aircraft also used 30 mm DU munitions at targeted sites, and depleted uranium munitions were fired at sites in southern Serbia and Montenegro. NATO confirmed that over 30 000 rounds of DU had been used in Kosovo, more than 2,500 rounds in Serbia and 300 rounds in Montenegro (UNEP, 2000). In BiH, NATO information states that the numbers of DU rounds fired at any one target range from 120 to 2400, with a total number of confirmed rounds standing at 6 230 although the exact number remains unknown (see Appendix P). According to NATO/KFOR information provided to UNMIK, the mixture comprised 5 DU rounds per 8 fired (KFOR, 2000).

Nothing indicates that a different mixture was used in BiH. The effectiveness of DU in kinetic energy penetrators (the rods of solid metal used as munitions) has been repeatedly demonstrated at various test ranges and in actual military conflicts. Kinetic energy penetrators do not explode but, if they hit an armoured (hard) target, they may form an aerosol of fine particles. Since uranium metal is pyrophoric, the DU particles ignite and burn, forming small particles of uranium oxides due to the extreme temperatures generated on impact greater than 1000°C). Most of the contamination remains inside any vehicle that has been struck and penetrated, although some of the dust will be dispersed into the air and deposited on the ground of the surrounding environment. Importantly, DU hits on “soft” targets (e.g. non-armoured vehicles) do not generate significant amounts of dust. Most DU dust from hard target impacts remains within roughly 100 metres of the target, 90 % of which is expected within 50 metres of the target (CHPPM, 2000).

Most penetrators that hit non-armoured (soft) targets will pass right through the target and, in most cases, remain intact. A penetrator that hits the ground usually also remains intact and will continue down into the soil. The depth depends on the mass of the penetrator, the flight angle of the round, the speed of the tank or plane, and the type of soil. In clay, penetrators used by the NATO A-10 aircraft have been reported to reach more than two metres in depth. Penetrators hitting hard objects, such as stones, may ricochet and may thus be found on the surface of the ground several metres from the attacked target. As 7 years had elapsed between the conflict and the DU assessment mission, the major interest of the UNEP mission to BiH was to examine the possible risks from DU to ground, water, biota and populations near the impact sites after such a period of time.

The type of DU munition that the NATO A-10 aircraft uses has a conical DU penetrator. Its length is 95 mm and the diameter at the base 16 mm. The weight of the penetrator is approximately 300 grams. The penetrator is fixed in an aluminium ‘jacket’ (also called ‘casing’), with a diameter of 30 mm and a length of 60 mm. The penetrator and jacket fit tightly through the cylindrical bore of the barrel of the A-10’s Gatling gun and the jacket assists the round in flying straight. When the penetrator hits a hard object, e.g. the side of a vehicle, the penetrator continues through the metal sheet, but the jacket does not usually penetrate.

The NATO A-10 aircraft is equipped with one Gatling gun. This gun can fire 3 900 rounds per minute. A typical burst of fire occurs for 2 to 3 seconds and involves 120 to 195 rounds. The shots will hit the ground in a straight line and, depending on the angle of the approach, the shots will hit the ground from 1-3 m apart and occupy an area of about 500 m2. The number of penetrators hitting a target depends upon many factors, including the type and size of the target. On average, not more than 10% of the penetrators hit the target (CHPPM, 2000). It is important to note that not all A-10 attacks are done with DU munitions. These planes also carry bombs and are used for bombing runs independently of any DU attacks.

UNEP has no information that depleted uranium was used in the cruise missiles fired by NATO forces, or that DU tank munitions were fired during the conflict in BiH. The Yugoslavian Authorities say that the FRY forces did not use depleted uranium. In a letter to the journal Health Physics, the U.S. Department of Defense Directorate of Deployment Health Support stated that “Tomahawk [cruise] missiles used in combat and Apache [helicopter] 30 mm rounds do not contain depleted uranium”(Kilpatrick, 2002). This letter was in response to a previously published article in Health Physics, which stated that DU in the warhead of the Tomahawk was “an assumption that was probably not correct” (Durante and Pugliese, 2002); however, the journal editors allowed the assumption to stand as an extreme worst case.

N.2 THE FATE OF DU IN THE ENVIRONMENT

Normally, 10-35% (and a maximum of 70%) of the penetrator becomes an aerosol on impact (with a “hard” target such as a tank or an armoured personnel carrier (APC)), or when the DU dust accumulated inside the hit target catches fire (RAND, 1999). The DU concentrations, rather than the number of DU particles in various particle sizes, will drive the resultant human intakes and internal doses. A DU hit should be confirmed with a radiation detection instrument since other types of kinetic energy penetrators, such as tungsten alloys, may also leave a black dust cover. The use of colour of the dust cover (i.e. light yellow) is an unreliable indicator for the presence of DU and depleted uranium oxides.

After an attack with depleted uranium munitions, DU will be deposited on the ground and other surfaces in the form of DU metal in pieces, fine fragments and dust. If the DU has caught fire, DU will be deposited as dust of uranium oxides. Most of the DU dust around the targets on the U.S. Nellis Air Force Range is reported to have been deposited within a distance of 100 m of the target (Nellis, 1997). These have been used for a long time period as U.S. Air Force training targets.

Most of the penetrators that impact on soft ground (e.g. sand or clay) will probably penetrate intact more than 50 cm into the ground and remain there for a long time. Penetrators that hit armoured vehicles form an aerosol upon impact or ricochet. Bigger fragments and pieces of DU are initially deposited on the ground surface. Through weathering, the smaller fragments and dust will gradually be transported into the upper soil layer by water, insects and worms. Wind, rainwater or water that flows on the ground may also transport the smaller fragments and DU dust. Depending on soil composition, some of the dust particles will adsorb onto soil particles, mainly on clay and organic matter (iron-oxyhydroxides and/or carbonates), and thus be less mobile.

Due to the fluctuating chemical properties of different soils and rocks, the fate of DU in the environment varies. Penetrators that are buried in clay will remain intact and will not affect the surrounding soil and groundwater. If penetrators are buried in quartz sand, they will weather relatively fast and may migrate to nearby groundwater. Weathering of penetrators buried in residual soils depends on the type of bedrock. If the soil consists of weathered granite or acid volcanic rock, the environment will be acidic and the weathering will be fast. Acid rain will accelerate the weathering, since uranium is acid soluble.

Once located, penetrators and large pieces of DU can be collected. Otherwise, the DU is removed by gradual leaching from rain and melting snow. This weathering process of DU is principally by corrosion into hydrated uranium oxide [U(VI)] that is soluble in water. Other possible uranium compounds have a lower solubility in water. However, various adsorption processes in soil may slow the migration of uranium through soil by several orders of magnitude, essentially making it immobile. Accordingly, it will take many years, perhaps several hundred years, before DU contamination migrates from the site.

source: http://postconflict.unep.ch/publications/BiH_DU_report.pdf (280 page 12.2 MB PDF file)

If you have come to this page from an outside location click here to get back to mindfully.org