DU weapons in the balkans.pdf

8/14/2019 DU weapons in the balkans.pdf

1/32


2/32

Report by: David Cullen, Researcher, ICBUWAdditional research by: Doug Weir, Coordinator, ICBUWSpecial thanks to: Naomi Toyoda, Amira Miholjcic (BiH), Ljubica Vujadinovic (Serbia) and AuberonKelmendi (Kosovo).

Unless otherwise stated, photographs by Naomi Toyoda Naomi Toyoda/ICBUWCover: earth moving equipment in use during decontamination project near Borovac, southern Serbia,2007. Photo by Naomi Toyoda.Research and publication funded by: The Government of Norway, Ministry of Foreign Affairs.

Published in September 2010 by ICBUW, Bridge 5 Mill, 22a Beswick Street, Ancoats, Manchester, M47HR, UK. Reprinted with minor textual changes in January 2011.www.bandepleteduranium.orgCopyright ICBUW 2011

Design by ICBUWPrint by: www.calverts.coop

PrefaceWhile this report deals with the legacy of war and the consequences of some of the weapons used, it does not seek to take a

posi on on the con icts in the former Yugoslavia. In highligh ng the e ects caused by some of the methods used in the con ictsby one party or another, some may interpret us as taking sides. We wish to explicitly reject this interpreta on and to stress ourhumanitarian concern for the e ects of uranium weapons on all people.

In con ict, and its a ermath, language itself is o en used to privilege one perspec ve and exclude all others. We have tried in thisreport to choose language which does not do this, and to balance this need with wri ng clearly. If we have got the balance wrongin places, we hope it will be overlooked.

Preparing a report of this kind naturally involves a process of selec on. In order to give the reader an overview, complex subjectshave necessarily been dealt with brie y. In par cular this report should not be taken as a full review of the literature pertaining tothe health e ects of uranium weapons, or their e ect in the Balkans. The legal status of uranium weapons is also a complex ma erthat has only been brie y touched upon. More detailed treatment of these subjects can be found in many of the works cited inthe endnotes.

It should be noted that the views expressed in this report are those of the authors. While every e ort has been made to ensurethat all the informa on in this report is correct, the authors welcome correc ons and clari ca ons from all interested par es. Allvalues expressed in US dollars have been calculated at current exchange rates at the me of wri ng.


3/32

Contents

AcronymsAPC Armoured Personnel CarrierAPM Anti-Personnel MinesBiH Bosnia and HerzegovinaDfID Department for International DevelopmentDU Depleted UraniumEOD Explosive Ordnance DisposalERW Explosive Remnants of WarFBiH Federation of Bosnia and HerzegovinaFRY Federal Republic of Yugoslavia at the time

of the con icts this comprised Serbia &Montenegro

IAEA International Atomic Energy AgencyICBUW International Coalition to Ban Uranium

WeaponsICP-MS Inductively Coupled Plasma Mass

Spectrometer

ICRC International Committee of the Red CrossIHL International Humanitarian LawKFOR Kosovo Force The NATO peacekeeping

force in KosovoMAC Mine Action CentreMoD Ministry of DefenceNATO North Atlantic Treaty OrganisationNBC Nuclear, Biological and ChemicalSAM Surface to Air MissileSFOR The NATO peacekeeping force in BiHUK United KingdomUN United NationsUNEP United Nations Environment ProgrammeUS United States (of America)UXO Un-Exploded OrdnanceWHO World Health Organisation

Executive summary 11.0 Background 32.0 Transparency 4

Release of information 4Assessments by the Federal Republic of Yugoslavia Army 4The fate of targeted vehicles 4

3.0 Environmental assessment 5Identifying contamination in the feld 5Institutional capacity 5Access to equipment 6The current assessment picture 6

4.0 Health consequences of DU use in the Balkans 6The current health picture 6Health studies in post-con ict environments 7 Evaluation of exposure 7Research to date 7Implications and policy concerns 8

5.0 Economic and social consequences 9Case study: the TRZ Hadici site 9Public perception of contamination 10

6.0 Decontamination and monitoring 11 Strike site information in relation to decontamination 11

Site histories 11 The decontamination procedure 12

Cost of decontamination 13Current status of sites 13

Long term consequences of contamination 157.0 Comparison with explosive ordnance disposal 15 International legal status 168.0 Implications for Iraq and other con icts 17 9.0 Recommendations 1710.0 Endnotes 1911.0 Maps 2112.0 Appendix 23


4/32

BackgroundDepleted uranium (DU) is used in armour-piercing tank shellsand bullets because of its extreme density, and because itburns upon impact. It is used in a dart or slug at the core ofthe weapon called the penetrator .

Uranium weapons were deployed in the Balkans by UnitedStates (US) aircra opera ng under NATO auspices in Bosnia& Herzegovina (BiH) in 1994 and 1995, and in Serbia, Kosovoand Montenegro in 1999. The United Na ons EnvironmentProgramme (UNEP) visited a representa ve sample ofsites, and produced three reports between 2000 and 2002.These included lists of recommenda ons for dealing withcontamina on at the sites.

ICBUW visited the region in 2010 to inves gate whetherUNEPs recommenda ons had been carried out, as well aslooking at the problems surrounding the assessment anddecontamina on of sites.

TransparencyAlthough NATO did release lists of strike coordinates in theBalkans, this informa on is s ll not complete, and therewere delays in the release of informa on. In the case of BiH,informa on was not released un l six years a er the warended. Without knowing everything that has happened atthe sites, and what happened to the vehicles that were hit,complete assessment of the risks is not possible .

Environmental assessmentWithout detailed informa on to work from, it is di cult tolocate points of contamina on within the landscape, andwhen red by aircra , penetrators are usually buried in thesoil. Surveying contaminated sites requires exper se andequipment that is specialised and o en very expensive.

Neither may be available to countries emerging from con ict.Furthermore, such e orts need to be highly coordinated, yetthe governance structures to organise such work are unlikelyto be immediately in place a er con ict, or in a newly formedstate. Other administra ve and environmental priori es maycompete for resources and may result in recommenda onslike those of UNEPs not being fully implemented.

In BiH, uranium weapons contamina on has been managedseparately by the two parts of the country. This has historicallycreated unnecessary duplica on of func on and impeded

the sharing of exper se. In Kosovo, the Environment Ministrycurrently lacks the exper se and equipment to carry out soiltes ng and analysis, or decontamina on.

Sites with detectable contamina on in Serbia have been

A Ques on of Responsibility: 1

Executive summary

extensively decontaminated, as has the single site inMontenegro. There are also ongoing monitoring programmes.Decontamina on has only been undertaken at one site in theFedera on of Bosnia & Herzegovina (FBiH). No decontamina onhas been done at the site in Republika Srpska, though therehas been monitoring there in the past. In Kosovo, where mostof the contaminated sites are located and over 70% of the DUwas red, there has been no programme of monitoring sinceUNEPs study in 2001.

While many of the other problems faced by these countriesare a product of their internal history, uranium weaponscontamina on is the result of the ac ons of others and adds tothe burden on countries struggling to emerge from the legacyof war.

Health consequencesThe health consequences of exposure to DU are not clear, but

within the body it is a carcinogen. While several desk studieshave been used to es mate the risks from contamina on,these are not a subs tute for real world studies of the e ects.

There is concern in parts of the region about the use ofuranium weapons. Media reports o en link their use withreports of high rates of cancer. However, studies that couldiden fy the cause have not been done to the extent required.Unfortunately, the circumstances typically found in post-con ict states may impede detailed scien c studies. Aswith environmental assessment, ins tu onal capacity andresources will be in short supply. The public cancer registriesin both BiH and Serbia broke down during the con icts, andKosovo is only now beginning to establish one.

While there have been some studies with interes ng results,the constraints of funding and access to equipment are

limi ng. Again, because uranium weapons contamina onwas not caused by these countries, it should not be their soleresponsibility to prove whether there are any health problemsas a result of it. External funding and access to equipmentcould help facilitate these projects, but when interna onalcommentators call for more evidence on the e ects of uraniumweapons, they must understand the complexity of the workinvolved. Even in the most benign circumstances, conclusiveresults can be elusive, and the legacy of war is such that manypoten al studies are impeded by a lack of data.

Economic and social consequencesThe case of the TRZ Hadii site shows that economic damageand social problems can result from contamina on, even whenthe full extent of the health consequences are unclear. The fearof contamina on can have a major impact on communi esand simply providing more informa on about the risks willnot solve the problem. It would be scien cally unjus edto o er a clean bill of health to places where there are s llelevated levels of uranium in the environment. Furthermore,there is too much distrust a er con ict for authori es to easily


5/32A Ques on of Responsibility: 2

reassure scep cal popula ons. In this context, the use ofuranium weapons is a lightning rod down which old animosityand division can travel, a situa on that is unlikely to change inthe future.

DecontaminationBecause of the dispersal of contamina on, simple point

coordinates are insu cient for site iden ca on andmore detailed informa on should be provided. The lack ofinforma on about whether any clean-up has previously beendone at sites is also problema c. This is par cularly a problemin Kosovo where the role of KFOR (The NATO PeacekeepingForce in Kosovo) is unclear. While Serbian sites with detectablecontamina on have been e ec vely decontaminated, onlysurface decontamina on has been done on one site inFBiH, and none in Kosovo. Again Kosovo, where most of thecontamina on is concentrated, is in the worst posi on.

There are a number of reasons why Serbia has been able tocarry out more extensive decontamina on work than has beendone elsewhere. Firstly it inherited many of the ins tu onsthat dealt with these ma ers in the former Yugoslavia, as theywere based around Belgrade. Serbia is also much larger, so hasmore nancial resources at its disposal and had rela vely fewsites to deal with. Also, although it su ered economic hardshipand aerial a ack during the con icts of the 1990s, it has nothad to deal with the legacy of a ground war.

Decontamina on is di cult work, and it is impossible to fullyremove all the contamina on. It is also very costly: the CapeArza site in Montenegro cost DM 400,000 (almost $280,000 US)and took about 5,000 working person days to decontaminate480 rounds, which in total took around 12 seconds to re.

Given that even a er extensive decontamina on manypenetrators can remain in the ground, sites may requireongoing tes ng of groundwater. In some circumstances,es mates of how long this may need to be done run intocenturies, and again the tes ng is very expensive. This is oneof the reasons ICBUW advocates a precau onary approach todecontamina on, and to the use of uranium weapons.

Explosive ordnance disposal & uraniumweapons contaminationThe presence of mines and unexploded ordnance (UXO)complicates dealing with DU and vice versa. While thereare demining standards in place for DU, in prac ce themanagement of uranium weapons contamina on was nota primary focus for the BiH or Kosovo Mine Ac on Centres.At the TRZ Hadii site in BiH, a project that dealt with bothwas undertaken, and mines and UXO were detonated in situbecause they represented the more immediate risk.

The presence of uranium weapons together with UXO meantthat those planning demining ac vi es had to weigh up therela ve risks of losing a limb or developing cancer later in life.

International legal statusIn contrast with explosive remnants of war (ERW), whichare the subject of both interna onal humanitarian law andspeci c treaty law, the norms governing the use of uraniumweapons or other toxic remnants of war are derived solelyfrom interna onal humanitarian law. Several of these normsshow that there is a clear legal case that states should observeprecau on both in the use of uranium weapons, and indecontamina ng a ected areas. The moral case is boosted bythe considerable barriers to e ec ve decontamina on.

Unfortunately, and despite these exis ng legal regimes, stateusers do not seem to have placed any restric ons on theuse of the weapons, using the unresolved scien c issuessurrounding the long-term impact of contamina on as a jus ca on. While the harm posed by ERW is more direct, andeasy to understand, this cannot be allowed to jus fy inac onon uranium weapons.

Implications for Iraq and other con ictsThe use of A-10s in the two Balkan con icts was actually verylimited. If it had not been for speci c condi ons, the use ofDU in the con ict would have been much greater. This wasthe case in Iraq, where more than 57 mes as much DU wasused. Although the environmental situa on is very di erent,the challenges discussed in this report are likely to also beexperienced by Iraq, but in much greater magnitude. Extensive

eldwork and risk reduc on programmes with interna onalassistance are urgently required.

RecommendationsThe circumstances that always surround the use of DU (i.e.con ict) mean that we should never presume that a ectedstates will be able to deal with assessment, monitoring ordecontamina on alone. In both the use of uranium weapons,and decontamina on, a precau onary approach should prevail.There is a clear need for transparency over the use of uraniumweapons, and for technical assistance with decontamina on.Interna onal assistance should be targeted to increase capacity

in the region and strengthen links between researchers.

There is s ll a need for further health studies in the region toassess the health consequences of uranium weapons use.

While the immediate need is for transparency and technical

assistance, states should consider whether there is a case forspeci c interna onal measures that address the par cularcharacteris cs of uranium weapons. Considera on shouldalso be given to the best way to ensure that capacity exists toundertake marking, monitoring and clearance, including the

crea on of a semi-permanent capacity for such work to dealwith both exis ng contamina on and long-term monitoring.



Depleted uranium is a by-product of the process used to enrichuranium to make nuclear weapons or fuel for power sta ons.

Because of its extreme density, and because it burns uponimpact, it is used in armour-piercing tank shells and bullets.The uranium is used in a dart or slug at the core of the weaponcalled the penetrator .

Depleted uranium may also be manufactured fromreprocessed nuclear fuel, in which case it generally containstrace amounts of reactor waste, such as plutonium: the termuranium weapons is used to describe conven onal weaponsthat contain uranium from any source. Unlike nuclear weapons,uranium weapons do not cause damage by using radioac ve

ssion. Instead they rely on their high density, which allowsthem to penetrate armour when red at high velocity.

Independent analysis has shown that samples of the uraniumweapons stock used in the Balkan con icts appear to havebeen manufactured from post-reactor material. Because ofthis, the terms uranium weapons and depleted uranium (DU)are both used to refer to material of this type in this report.

Uranium weapons were deployed in the Balkans by UnitedStates (US) aircra opera ng under NATO auspices. In Bosnia& Herzegovina (BiH), they were deployed as part of Opera on

Deny Flight on two occasions in August and September 1994,and as part of Opera on Deliberate Force on a further 17occasions in August and September of 1995. At least 1,271kg ofDU was red during these 19 sor es. As part of Opera on AlliedForce , uranium weapons were red on at least 112 occasionsbetween 6 th April and 11th June 1999 at targets in southernSerbia and Kosovo as well as one site in Montenegro. At least5,723kg of DU ammuni on was red during this me.1

In all of these cases, the uranium-based round was the PGU-14/B, which is red against targets on the ground by the largeGAU-8 Gatling-type rotary cannon on the A-10 Thunderbolt IIaircra . Despite being used by only a single model of aircrafor a single military purpose, and in limited military opera ons,this s ll resulted in signi cant environmental contamina on.

Nonetheless, the quan es involved are rela vely smallcompared to those expended in Iraq. During the two con ictsin 1991 and from 2003 onwards, at least 404,000 kg wasexpended, more than 57 mes as much. Although theenvironmental condi ons and the type of warfare involved arenot iden cal, the experience of the Balkan countries providesan invaluable insight into the e ects of uranium weapons,with implica ons for their use in other con icts. In par cular,the issues surrounding environmental surveying, conduc nghealth studies and decontamina on procedures are likely tobe common to all countries recovering from con ict.

Following growing concerns about its use in the 1991 GulfWar, the use of uranium weapons was already a controversialtopic by the me of the 1999 Kosovo con ict. Researchersbased in Serbia tested recovered NATO ammuni on andcon rmed that it contained uranium, 2 and signi cant mediainterest followed the Pentagons public statement that theyhad used uranium weapons. 3 It was against this backgroundthat the recently launched Post-con ict Assessment Unit ofthe United Na ons Environment Programme (UNEP) beganto look into the ma er. A desk study was produced in late1999,4 and it became apparent that for a proper assessmentof contamina on it would be necessary for NATO to producedetails of targeted loca ons. Following a request by theUN Secretary-General, NATO provided some non-speci cinforma on and, following a second request, they produced alist of 112 strike sites. Subsequently NATO also con rmed theuse of uranium weapons in BiH.

Using this informa on UNEP visited a representa ve sample

of 11 sites in early 2000, and produced a report in early2001. To some extent the report was reassuring ratherthan the widespread contamina on of the region that somehad feared, contamina on was localised. As the uranium inweapons is mostly comprised of the isotope uranium 238, itprimarily emits alpha radia on. Alpha radia on cannot passthrough the skin, which means that the major concern forhuman health is if it should nd its way into the body. Uraniumis also chemically toxic, a further reason to suspect it couldbe harmful to health. From UNEPs ndings, it seems that thelikely risks were that the uranium at the sites could nd its wayinto drinking water sources, or that uranium dust at the sitescould become resuspended and inhaled.

To reduce these risks, UNEP produced a list ofrecommenda ons for dealing with contamina on at the sites,and for sites that it had not visited. A World Health Organisa on(WHO) mission also visited several sites in Kosovo althoughthey did not carry out any sampling work in the eld.5 Subsequently, UNEP reports were produced looking at sites inSerbia and Montenegro in 2002, and BiH in 2003. A summaryof the recommenda ons is produced in the appendix on pages23 and 24 of this report.

In 2009, ICBUW secured funding from the Norwegian Ministryof Foreign A airs to undertake a research trip to BiH, Kosovo andSerbia in order to examine whether UNEPs recommenda onshad been carried out. ICBUW also wanted to learn more aboutthe e ects of the contamina on on people living nearby andto assess any research that had been done into possible healthconsequences. An interna onal team from ICBUW visited theBalkans in March and April 2010. While our remit was notto undertake primary research, we were able to speak with

many of the individuals and government agencies who havebeen involved in both research and remedia on. We were alsoable to build up a picture of the current status of some sites,and look at the problems surrounding the assessment anddecontamina on of contaminated sites more generally.

1.0 Background


7/32

The degree to which uranium weapons contamina on hadbeen monitored and dealt with varied greatly across theregion. Extensive decontamina on had been undertaken onthe sites in Serbia where there was detectable contamina on.In BiH, limited decontamina on had been done at one ofthe worst a ected sites, where there is also an ongoingmonitoring programme. However, the capacity for such workis very limited in Kosovo and there is no organised ongoingprogramme, despite the fact that this is the loca on for mostof the contamina on in the region. This report exploressome of the factors underlying this disparity, as well as thecircumstances that impede health studies being undertakento establish the health consequences of uranium weaponscontamina on.

Release of informationAlthough NATO did release lists of strike coordinates a er twointerven ons by the UN Secretary-General, this informa on iss ll not complete. For six of the sites in BiH, there is neitherinforma on about the number of rounds expended, nor thestrike coordinates. The only informa on available is that thestrikes were in the vicinity of the capital, Sarajevo.6 In the listof strikes for the Kosovo con ict, a further 23 entries do notdetail how many rounds were red, and some coordinatesappear to be invalid. 7 The data was compiled using missionreports to es mate coordinates, and records from each unit ofthe amount of ammuni on expended.

Even where this informa on has been disclosed, there isconsiderable confusion about what propor on of bullets redby the A-10s were PGU-14/B rounds with a DU penetrator,and what propor on were PGU-13/B high explosive rounds.Informa on held by the Kosovo Ministry of the Environmentsuggests that PGU-14/B made up 4/5 of the overall number, 8 the propor on that was used in the 1991 Gulf War. 9 Informa on provided by KFOR (the NATO peacekeeping forcein Kosovo) to the WHO also cited the same ra o.10 However,for the Kosovo con ict, the correct ra o is more likely to bethat used by UNEP and the UK MoD (United Kingdom Ministryof Defence): DU ammuni on made up 5/8 of the total. 11 In theearlier interven ons in BiH, the situa on is less clear. UNEPmade the case that one should assume the same ra o wasused in both con icts,12 but their calcula ons for the weightof DU expended at each site appear to assume that all rounds

red at sites in BiH were PGU-14/B.13 Needless to say, knowinghow many uranium rounds were likely to have been used at asite is of fundamental importance for the proper assessmentof contamina on, and NATO should clarify the situa on assoon as possible.

Authori es in Serbia tasked with decontamina on detectedcontamina on at one loca on that was not on the list providedby NATO on a hill at Pljakovica near Vranje where a TVtransmi er was based. Contamina on at the site was latercon rmed by UNEP.14 This discrepancy, and the di culty ofrecovering uranium penetrators in the eld, has led to concernsthat the NATO data is not correct, either in the loca ons listed,or the number of rounds red at each loca on. 15 Concernsabout the accuracy of the coordinates are not limited to theSerbian authori es. An inves ga on by the UK MoD into strikesites in the UK KFOR zone concluded that:the accuracy of themap coordinates for the loca ons where DU was used are only

accurate to plus or minus one nau cal mile. 16

The delay in the release of informa on was also problema c.In the case of BiH, con rma on that uranium weapons were

red did not come into the public domain un l six yearsa er the cessa on of hos li es. In the case of one sitevisited by ICBUW, this delay meant that workers at the site

cleared contaminated war debris by hand without any formof protec on, 17 and that some demining of a contaminatedsite was undertaken by teams who had no knowledge of thecontamina on. 18 While this example is an extreme case, it ishighly probable that at numerous sites throughout the region,civilians were unnecessarily exposed to hazards when theprompt disclosure of informa on could have allowed the localauthori es to warn the populace and take remedial ac on.

Assessments by the Federal Republic of

Yugoslavia ArmyAlthough primary responsibility for the disclosure of strikesites naturally rests with the users of uranium weapons, somewithin the Federal Republic of Yugoslavia (FRY) Army duringthe years of con ict were aware that uranium weapons hadbeen used, and did not release this informa on into the publicdomain. Possibly as early as 1994,19 and certainly by 1996,20 teams had visited the Han Pijesak site in BiH and con rmedthe presence of DU contamina on. However this informa onwas not made available to the civil authori es un l a er NATOhad con rmed its use several years later.

In the run-up to the Kosovo con ict, a project to iden fy thesites where uranium weapons had been used within Serbia wasdevised, and researchers were able to visit various strike sitesduring the con ict itself. This project iden ed several siteson NATOs list of strike sites in southern Serbia.21 Although thisinforma on would have had li le relevance for the other sitesin BiH or those within Kosovo, it presented an opportunityto inform civil society and the interna onal community ofthe use of uranium weapons. It was an opportunity that wasunfortunately missed.

The fate of targeted vehiclesOne issue that ICBUW pursued during its visit was that ofvehicles that were hit with uranium weapons. Some studies


2.0 Transparency



have suggested that much of the contamina on in thissitua on remains within the vehicle. 22 Therefore the lack ofvehicles at most of the sites visited by UNEP suggests thatsigni cant sources of contamina on remained unexamined.

These presumably posed a risk to anyone who came intocontact with them. Aside from the case of the TRZ Hadiisite (dealt with in detail later in the report), we were not

able to get a detailed answer to this ques on in BiH. It seemslikely that at most of the other sites, any vehicles that weresalvageable were removed by the Serbian forces, though noneof our interviewees could say what was done with them.

Certainly in Kosovo, very few destroyed vehicles were le atstrike sites when a US assessment team visited the area in theweeks and months a er the con ict. However, it has beensuggested this was simply due to NATOs lack of success intarge ng Serbian forces on the ground. 23

UNEP found wreckage at more than one site during their

study, but none that tested posi ve for DU.24

The wreckagethat was le behind was apparently dealt with by KFOR,25 butit has not been possible to ascertain precisely how.

For any vehicles within Serbia, the armys Nuclear,Biological and Chemical (NBC) Branch was responsible fordecontamina on, and the civilian authori es were not able toprovide details.26 Unfortunately we were unable to secure amee ng with the Serbian Army during our visit, and thereforewere unable to establish what decontamina on procedureshad been undertaken.

During UNEPs study in Serbia and Montenegro they wereable to examine an Armoured Personnel Carrier (APC) that hadbeen hit with uranium weapons, and recommended that thevehicle be cleaned and the crew provided with medical careand health monitoring. 27

Identifying contamination in the eldWithout quite speci c informa on to work from, it is di cultto locate points of contamina on within the landscape. Theradioac ve signature of DU is mainly comprised of alphapar cles, and this type of radia on is not easily detectablefrom a distance. Instead, equipment for detec ng beta andgamma radia on is used in the eld. DU only produces a smallamount of this type of radia on, meaning that it is di cult todetect.

Penetrators are usually buried in the soil, where the radia onis not detectable from the surface, except for the soilimmediately around the point of impact. While this uraniumcould in me come to pollute water sources, these factorsmake the task of environmental assessment extremely di cult.

Teams inves ga ng contamina on have been forced to relyon local knowledge, or visual indicators of the a ack in orderto locate contamina on. 28,29 It has been suggested that usingmetal detectors may help to locate buried penetrators, 30 butothers found that the amount of shrapnel in the ground meantthat they were of limited use. 31 Because of these limita ons,there is no realis c possibility of loca ng the six sites in thevicinity of Sarajevo for which there are no target coordinates.Local residents and the authori es have no choice but to livewith this situa on.

Institutional capacityEven in circumstances where contamina on points canbe located, countries that are emerging from con ict facesigni cant impediments to performing environmental surveywork. The exper se and equipment required to inves gateuranium contamina on are highly specialised, and may not beavailable. This is par cularly true of the equipment that canreliably di eren ate between natural and depleted uranium.Furthermore, such e orts need to be highly coordinated, wellsupported and funded. Without well developed governancestructures, academic exper se in the relevant eld, radia onand environmental protec on legisla on and poli cal focus,they are unlikely to progress in a su ciently sustainedfashion.

While there can be no doub ng the professionalism anddedica on of those working in this eld in the region, it isclear that many face signi cant impediments to their work.Before the breakup of Yugoslavia much of the work in this

eld, as well as many others, was done in or around Belgrade,which had been the capital of Yugoslavia, and where nuclearsciences research began as early as 1948. 32 One of our (non-Serbian) interviewees referred to it as the real capital of theregion because of this legacy. To re-create all of these func onsin a newly independent country is a huge task, and takes aconsiderable length of me. This is well illustrated by onena onal ministry in BiH, which began work with only four or vesta who had transferred over from the ministry in Belgrade the rest of the sta began with no direct experience.

The compromises necessary for peace can also bequeathins tu onal frameworks that are ill-suited to smoothgovernance during peace me. The Dayton Agreement splitBiH into two dis nct poli cal en es where most governancefunc ons were devolved, with few exercised at the state level.This means that for most of the me since uranium weaponswere used, the issue of contamina on has been managedseparately by the two en es, crea ng unnecessary duplica onof func on and impeding the sharing of exper se. While aradiological protec on agency has recently been formed atthe state level, Republika Srpska has only had radiologicalprotec on legisla on in place since 2007, and did not have anyradioac ve waste storage capacity. 33 The joint agency now hasaccess to a central site that serves the whole country. 34

3.0 Environmentalassessment



As BiH had been a former republic of Yugoslavia, withconsiderable autonomy prior to the war, it was perhaps be erplaced in this regard than Kosovo, which had its autonomousstatus within Serbia rescinded at the end of the 1980s. Whilethere has been an established Environment Ministry theresince 2002, it s ll lacks the exper se and equipment to carryout soil tes ng and analysis or decontamina on. The ministryis reliant on KFOR to no fy them if DU contamina on is aproblem. 35

Access to equipmentThe lack of funds and the most precise equipment is notpar cular to Kosovo but is a shared problem throughout theregion. Environmental tes ng in the en ty of Republika Srpskarelied upon equipment borrowed from the Interna onalAtomic Energy Agency (IAEA).36 The tes ng of poten allyexposed civilians in southern Serbia was restricted by cost. 37

Lack of access to the most advanced equipment in BiH hasalso hampered detailed follow-up of some of UNEPs work.One of the major ndings of UNEPs study in BiH was that DUcould be detected in one or possibly two water samples takenat the TRZ Hadii site. While the levels are below the WHOs15 g/litre provisional guideline value,38 ongoing monitoringof the water at the site is essen al to see whether there hasbeen any change in the level of DU in the water source.

While the authori es in BiH have been taking both soil andwater samples at the site, they have not found signi cantlyincreased levels of uranium contamina on in the water.However, with the equipment available to them it is di cultto reliably dis nguish between the isotopic signatures ofnatural and depleted uranium. 39 For such a test one wouldneed to use an ICP-MS (Induc vely Coupled Plasma MassSpectrometer), rather than an Alpha Spectrometer. At the

me of wri ng the cost of a new machine was in the regionof 120,000 (US$180,000),40 with independent laboratoriescharging hundreds of pounds to test individual samples. So,while it is possible to say that uranium contamina on has notincreased greatly since 2002, it is impossible for them currentlyto dis nguish the presence of DU from natural uranium.

The current assessment pictureBecause of the factors described above, knowledge about thestatus of contamina on di ers between the a ected countriesin the region. The sites with detectable contamina on inSerbia have been extensively decontaminated, 41 as has thesingle site in Montenegro. 42 Some surface decontamina onhas been performed at the TRZ Hadii site, and there isongoing monitoring.43 Other sites in the Federa on of Bosnia& Herzegovina (FBiH) are not currently being monitored,nor is the site in Republika Srpska, although there has beenmonitoring there in the past.

In Kosovo, where most of the contaminated sites are located,there has been no visible programme of monitoring since

UNEPs study in 2001, which looked at only 11 of 85 sites.44 As the known environmental problems in Kosovo include thewaste legacy from communist-era heavy industry, air pollu onfrom lignite electricity genera on, and untreated sewage inwater courses, 45 it is unsurprising that the authori es are in nohurry to add further to this list.

Facing all of these impediments, it is clear that countries

emerging from con ict cannot be expected to conductenvironmental surveying work on uranium weapons withoutnancial and technical support from the interna onal

community.

While many of the other problems faced by these countriesare a product of their internal history, uranium weaponscontamina on was a result of speci c military decisions,mandated by the UN in the case of BiH and by NATO memberstates in the case of Kosovo. Their use is an addi onal andunwelcome burden for countries struggling to emerge from thelegacy of war: a new headache , as one of our intervieweesdescribed it.

The current health pictureThe most pressing and controversial ques on regarding the useof uranium weapons is whether they have any nega ve impacton human health. Since the issue came to prominence, therehave been a number of desk studies that have sought to assessthe risk from these weapons by compiling the exis ng researchon the issue. However, with a lack of signi cantly sized studieson exposed civilians, there are ongoing uncertain es over therisks posed by the ba le eld use of uranium weapons.

A full review of the evidence concerning the health risks ofuranium weapons contamina on is beyond the remit of this

report. Nevertheless, while there is a spectrum of opinion onthis ques on, there can be li le doubt that within the bodythe alpha radia on produced by DU is a carcinogen.46 This is aproperty, amongst others, that is also exhibited by uraniumsheavy metal toxicity.47

Without a comprehensive picture of the extent of thecontamina on, it is very di cult to assess the risk to popula onsin the Balkans. Although many of those who have worked onmeasuring the contamina on at sites were of the opinion thatthe general risks from the loca ons they had inves gated are

fairly low at the current me,48

in some situa ons the riskcould be signi cant.49 However, studies to assess the e ects oncivilians of chronic long term exposure to ba le eld residuesof depleted uranium have not been undertaken. 50

4.0 Healthconsequences of DUuse in the Balkans



Health studies in post-con ictenvironmentsThere is considerable concern in some parts of the regionthat the use of uranium weapons has resulted in rising ratesof cancer for some popula ons. Media reports o en link thehigh rates of cancer in southern Serbia and northern Kosovowith the use of uranium weapons. Unfortunately these claimshave not been subjected to detailed scien c analysis. Therehave also been reports that the popula on who moved fromthe Hadii region to Bratunac in Republika Srpska have higherrates of cancer and greater cancer mortality than the localpopula on.51

For greater scien c certainty, what is required are studiesthat can both show that individuals have been exposedto uranium weapons contamina on, and that there is asta s cally signi cant correla on between this exposure and agiven health outcome. As with the ques on of environmentalassessment, for a country recovering from con ict, conduc ng

such studies presents a considerable challenge.

To carry out epidemiologic studies assessing all risk factors,including DU, a reliable registry of diseases in that area isneeded, especially for cancer. Epidemiologic studies requirethe kind of uninterrupted and high quality healthcareprovision which is prac cally unheard of in war me. Largescale popula on movements are common in con icts, as isthe breakdown of governance structures, scarcity of fundsand extreme stress on remaining healthcare systems. Keepingme culous records of the type required for later scien c

work is unlikely to be a priority.The public cancer registries in both BiH,52 and Serbia,53 broke

down during the con icts, in spite of the fact that Serbia didnot endure a ground war. The registry in BiH underwent a12 year hiatus, only being re-established in 2004. Kosovo iscurrently undergoing a large scale reform of its health system.This covers cancer registra on, public health laws, medicalprac ce systems, and health insurance. The rst stage of thisprocess runs from 2010 to 2014, and as such a central cancerregistry is yet to be established. 54

O en the a ermath of war can bring further disrup on. TheDayton Agreement in BiH, which divided the country intotwo en es along ethnic lines, presaged further popula onshi s.55

Another example is the hospital in North Mitrovica in Kosovo,which used to cater for all the inhabitants of the region butnow nds itself within an ethnic Serb enclave in a divided city.Those in the south of the city (predominantly ethnic Albanians)travel to Prish na for medical a en on. Although the hospitalis thought to also cater for Serbs from outside the immediate

region, there is no o cial census and popula on gures arehighly poli cised. This has rendered it impossible to makeauthorita ve statements on the incidence of disease in thispopula on. Although the hospitals physicians have informallinks with their medical colleagues in the rest of Kosovo, the

fractured state of medical cover rules out any study whichwould seek to analyse medical records in order to look at ratesof illness before and a er the con ict. 56

In one case, the disrup on to medical records was the directresult of uranium weapons contamina on. When the TRZHadii site was discovered to be contaminated with DU, itspost-war business collapsed and it was unable to con nue

paying taxes for a me. As a consequence, the workers, manyof whom may have been exposed to DU contamina on, losttheir governmental health cover for this period. 57

Evaluation of exposureOne major issue amongst the many which complicateepidemiological research is iden fying those most likely tohave been exposed. Using simple geographic proximity, suchas living in the same town as a strike site, for example, isunlikely to be a su ciently sophis cated method of iden fying

subjects.Notwithstanding, those who have either lived or worked

very close to an area which is contaminated are likely to beof interest to researchers. Ques onnaires can then be usedto re ne this search further but self-repor ng brings itsown challenges. One unexpected problem men oned by aninterviewee was that the high pro le of the DU issue and thesubsequent notoriety from having been present at an a ackcan make self-repor ng by subjects somewhat unreliable. Itwas suggested that recrui ng a trusted local intermediary tohelp with subject iden ca on was desirable in such cases. 58

One of the standard tests to see whether an individual hasbeen exposed to DU, is to test their urine to establish the level ofuranium. The ra o of uranium isotopes then indicates whetherit has a natural or man-made source. A study on workers andthose living near to a uranium weapons manufacturing facilityfor DU products in New York State showed that exposurecan leave a trace in urine for up to 20 years a er the event. 59 However, even if a urine test is nega ve, it is not possibleto rule out any exposure. Nonetheless, it remains the mostcommon approach for detec ng human DU exposure.

Research to dateThe majority of urine tes ng to determine exposure in thosewho may have been in contact with uranium contamina onin the region has been conducted by NATO member states ontheir peacekeepers. This tes ng was in response to media andmilitary concerns that they may have been contaminated. 60 Unfortunately, while these studies achieve the objec veof showing that the peacekeepers in ques on had nodetectable contamina on, they tell us li le about whethercivilians have been put at risk as a result of uranium weaponscontamina on.

One study of 12 residents of BiH and Kosovo appeared to showthe presence of DU in the urine of all subjects. 61 However these



Anja and Sanin Haveric, researchers who have undertaken chromosomal aberration studies on exposed workers at the Hadici facility, INGEB - Institute forGenetic Engineering and Biotechnology, Sarajevo BiH

ndings were ques oned by a later study that took several ofthe subjects as part of a cohort of 24 residents of Kosovo andSerbia, and found no detectable DU. 62 Unfortunately the costs

associated with ICP-MS tes ng of samples means that largescale tes ng of subjects in this way is prohibi vely expensive,and ICBUW knows of no studies that have involved largecivilian cohorts.

In BiH, a major focus of research has been in the region ofHadii, where three sites were targeted with uranium weapons.One study of interest looked at chromosome aberra ons (aspeci c biomarker for radia on damage to DNA) in workersfrom the TRZ Hadii site. The study included control groupsfrom other regions, including one that had been exposed to

some of the e ects of the con ict, but not DU. The group ofworkers were found to have more aberra ons, but the groupsize was not su cient for sta s cally signi cant ndings. Inaddi on the groups were not evenly matched for lifestyleand other factors. Unfortunately funding for larger and morecomprehensive studies was not forthcoming. 63

The ins tute of Public Health in FBiH has also done a pilot studyon 40 residents of Hadii to see whether there were unusualrates of several diseases, including cancer. No signi cant

ndings were observed, but monitoring will con nue. 64,65

A larger study was considered, comparing the municipalityof Hadii to another municipality, but it was ruled out ongrounds of cost. 66

Another study compared a number of possibly exposedgroups, examining both chromosome aberra ons and also

the uranium content and isotope ra o in urine using alphaspectrometry. A number of subjects from contaminatedareas in Serbia appeared to have an isotopic ra o consistent

with DU contamina on, and were part of the study groupwith a sta s cally signi cant increased risk of chromosomeaberra ons. While signi cant, this was lower than a controlgroup who were occupa onally exposed to X-rays as partof their work.67 Again, the original study was limited in thefunds and equipment that were available at the me, but theresearchers would be willing to conduct more comprehensivework if funds became available.68

Implications and policy concernsNotwithstanding the recognised limita ons to these studies,it is clear that well designed studies with su cient fundingand equipment should be able to iden fy poten ally exposedpopula ons and assess their exposure. Unfortunately,researchers from these countries have been le to pursuethese lines of inquiry with li le help from other na ons, whohave in the main restricted their research to ascertainingwhether peacekeeping troops have been contaminated.

ICBUW has long called for further research into the e ectsof ba le eld uranium weapons contamina on on civilians,

but it is vital that when interna onal commentators call formore evidence on the e ects of uranium weapons in the eld,they understand the complexity of the work required. Externalfunding and access to equipment could help facilitate thiswork. Even with this kind of assistance, producing evidence


12/32

force in BiH) who removed a box of spent penetrators, andwarned them to stay away from any others that were found. 70 UNEP visited in 2002 and made a full assessment of the site,prior to issuing recommenda ons on how to reduce the risksof exposure (the site is named Hadzici Tank Repair Facility intheir report). The site was heavily mined during the con ict,but has since been demined, and periodic sampling of soil andwater for DU has been undertaken by the FBiH authori es.Although the risks to the general popula on are not thoughtto be great, people performing work in the contaminated hotspots could be at risk of signi cant exposure.71

Economically, the damage has been signi cant. A er repairingtanks in Sarajevo during the war, they returned to a factorythat was stripped of equipment. In order to avoid laying o their sta , the business diversi ed into work that did notrequire specialised equipment. When news broke about thecontamina on, they were manufacturing military helmets.Because of the s gma associated with contamina on, ordersdried up. The company su ered badly, and at this me theworkers lost their health insurance as the company was not

able to pay its taxes. As one of the few pro table businessesin the area, with more than 100 employees, the e ects wouldhave been felt well beyond the factory walls. However theywere able to keep paying wages by taking on work repairingroads and other ac vi es that did not involve products that

su ciently compelling to count at the interna onal table is atall order, and the legacy of war is such that many poten alstudies simply lack the data that would be required.

At present policy-makers in a country a ected by uraniumweapons contamina on face an unappealing dilemma. Dothey invest scarce public resources into studies that may yieldinconclusive results or jus fy expensive decontamina on work

that may fail to calm worried popula ons? Or do they divertfunds elsewhere in the hope that the contamina on does notrepresent a major problem and risk being accused of coveringup the issue?

In the absence of interna onal assistance with research andpost-con ict management of uranium weapons contamina on,governments of contaminated countries are le with fewother op ons.

Case study: the TRZ Hadici siteAn instruc ve illustra on of the problems caused bycontamina on is the case of the TRZ Hadii site, BiH. Priorto the con ict the facility hosted a business repairing tanks.During the war, the business moved to Sarajevo and undertookmaintenance work for vehicles belonging to the armed forcesdefending the city. The facility was occupied by the army ofRepublika Srpska, and the strategically important site with itsvehicles and equipment came under sustained a ack fromNATO during September 1995.

Under the Dayton Agreement the site became part of theFBiH, and the Republika Srpska soldiers withdrew. When theworkers returned to the site in March 1996 all equipment,

ngs, machines and working vehicles had been taken bythe retrea ng forces. Sta found a great deal of rubbish onthe site, some of it contaminated with DU, and although all

operable vehicles had been taken, the tanks that had beenrendered immobile were le behind. However, the workerswere unaware of the possibility of contamina on, and fromthe Director down they cleaned up by hand using brooms.As they were not aware that the rubbish required specialhandling, it was simply dumped in land ll.69

Some me later, they became aware of rumours that somekind of unconven onal weapon had been used. These centredupon reports of higher rates of cancer in the refugee popula onfrom Hadii that had moved to Bratunac on the Serbian

border. However, at the me it was dismissed as a rumourspread by the Serbians to cause fear and alarm. It wasnt un lan interna onal journalist came to the site in 2001 that theywere made aware that DU had been used. This was con rmedby a German Commander from SFOR (the NATO peacekeeping


5.0 Economicand socialconsequences

Zijad Fazlagic, Director of the TRZ facility


13/32

came into contact with people. At the me of our visit, theywere manufacturing again but requested that we did notreveal the nature of their business for fear that the associa onwith the site become known. 72

Many of the workers have been with the company their wholelives. A number have died unusually young the Directormen oned one who kept a penetrator as a souvenir because it

released sparks when struck with a metallic object. He had diedaround 45 of lung cancer. While the Director knew that it wasimpossible to say for sure whether there was a connec on, hesaid that the workers live with a constant sense of uncertainty.Many of them have to deal with concerns in their family aboutworking at the site but are unwilling give up their income. Itwas described to us that any illness or ailment, whether it beasthma, kidney disease or a bad back, is accompanied by thethought in the back of their minds that it might be connectedto the contamina on. When they rst heard about thecontamina on a er the war, it did not seem to be too much of

a threat as there were many other things to worry about, butas me has gone by, their concern has increased. 73

While we were informed by the Director that medicalexamina ons have apparently been done by SFOR and bythe Federal authori es on around 150 of the workers, noinforma on had found its way back to them. This has beena further source of anxiety. 74 A similar issue was highlightedby another researcher concerning southern Serbia. There,residents have seen researchers come and go without beinggiven feedback about whether their land was safe, what resultshave been found and what precau ons they should take. 75


Public perception of contaminationThe lack of informa on for those living with contamina on canbe par ally addressed by following up academic research withsubjects, as well as risk educa on programmes (recommendedby UNEP in all three reports).76 However, without a largerbody of research into the poten al health consequences,and interna onal assistance in order to ensure that sites aredecontaminated, fear and exaggerated concern will con nueto dominate. The Finance Director at the facility recounted astory about an experience she had shortly a er the news aboutthe contamina on became well known. Visi ng the bank todeposit some funds, there was a large queue of between 50and 100 people, as banking facili es were s ll very scarce.However, when she was recognised a whisper went aroundthe room: Shes irradiated , and everyone moved out of theway to avoid her. 77 While the anecdote is comic, the day-to-day e ects of s gma sa on within the community wouldhave been far from amusing.

The Director of the facility said that if they had money, theywould consider a lawsuit against NATO. They were grateful toNATO, and believed that the interven on had helped to end thewar, but found it hard to accept that NATO is also responsiblefor a threat to their survival. 78

While it may be imagined that such issues can be easilycountered by proper educa on about the risks, this fails toappreciate both the uncertain es over the long term healthimpact of uranium weapons and the nature of con ict. Althoughthe authori es would like to reassure their popula ons, it

would be scien cally unjus ed to o er a clean bill of healthto places where there are s ll elevated levels of uranium inthe environment, so any assurances would be necessarilyequivocal.79

Authori es in both Serbia and Kosovo complain aboutsensa onal coverage of the issue in the press, which hadincreased public fear. 80 While chemical toxicity may well be ofgreater concern medically speaking, fear of radioac vity is anatural human reac on to an invisible and deadly threat. AtRadoniq/Radonjick Reservoir in Kosovo, visited by the ICBUW

team as part of our research, leaching could occur from thearea contaminated by DU close to the lake. UNEP assessmentsconcluded that the sheer volume of water in the lake issu cient to dilute any contamina on to the extent that itdoes not pose a risk. 81 Nevertheless, the lake supplies drinkingwater for much of southern Kosovo, and whatever the actualdangers, the poten al for such situa ons to cause public alarmis clear.

During our visit, it was abundantly clear that concern aboutthe use of uranium weapons is more keenly felt in the Serbianparts of the region, with a clear dis nc on between thosewho supported the NATO interven ons and those who didnot. It is nave to expect that in the future those who havebeen targeted with uranium weapons will accept that thereis li le risk - especially considering the ongoing uncertainty

Steel peg marking impact hot spot on the tarmac apron at the TRZ facility,gamma reading is 40x background, April 2010.



about the long term health impact. For this reason, the useof uranium weapons is always likely to be accompanied bysitua ons such as those in Tuzla Canton in BiH, where SFORrepresenta ves struggled to persuade a scep cal popula onthat the destruc on of weapons near the village of Seljubljewas not connected with a local cluster of cancers. 82

During our visit to Kosovo, we were told that even a vaccina on

campaign for Swine Flu had been perceived as a conspiracyagainst a par cular ethnic group. 83 A distrust of authority andgovernments was common throughout the region, and is to beexpected in the a ermath of con ict. In this context, the use ofuranium weapons is a lightning rod down which old animosityand division can travel, and this is unlikely to change in thefuture. The irony of using such weapons under a humanitarianbanner was noted by many of our interviewees.

Strike site information in relation todecontaminationUNEPs recommenda ons for the sites that they visitedinvolved the removal of penetrators and jackets (the casing forthe penetrator within the projec le), as well as contaminatedsoil where feasible (the full recommenda ons are summarised

on pages 23 & 24). They also recommended periodical samplingof groundwater for contamina on, measures to inform thoseliving near to contaminated sites of the risks, and inves ga onsinto claims of health problems. UNEP also recommended thatthe sites where they were unable to visit be surveyed and, ifnecessary, decontaminated.

Aside from the issues already raised regarding the early releaseof strike site informa on, the experiences of those involved indecontamina on illustrate the limita ons of the informa onprovided by NATO. US Army informa on suggests that in atypical A-10 stra ng run, 90% of the rounds will not hit theirtarget. Instead they will be spread across an area of 500m 2.84 This gure is heavily in uenced by the ight path of the plane,as the GAU-8 cannon is mounted below the cockpit.

The prac ce of Serbian decontamina on teams was to try toguess the planes angle of approach, and locate the area withthe highest radioac ve signature. They would then explore aconical area from this point. However, they frequently foundthemselves moving the fence around their area of work asother points of contamina on were iden ed. 85 Similarly,UNEP made use of military experts to iden fy the probableangle of a ack in order to help them locate contamina on. 86

When providing informa on to the authori es dealing withunexploded ordnance (UXO), on the probable loca ons ofcluster bomblets within Kosovo, NATO provided informa on

with rectangular es mated strike footprints, with inner andouter areas marked. These rectangles related to probabledimensions of bomblet dispersal area around the desired

loca on and probable dispersal including addi onal errorsassociated with weapon delivery from aircra . Although thecluster strike informa on was more detailed, even this wasdescribed as inadequate by the former Programme Managerfor the Mine Ac on Programme in Kosovo, who called foraddi onal informa on, such as the direc on of ight of theaircra , to also be included. 87 This stands in contrast to thesingle point coordinates provided to UNEP for DU strikes.

Although NATOs informa on about sites in Kosovo, Serbiaand Montenegro did not include the actual target (such asbuilding, APC etc.), this informa on was made availableto the UK MoD and would certainly be of use in the eld.88 Interes ngly, NATOs strike data for BiH did include this targetinforma on, as well as lis ng several coordinates for sometargets. In the nal analysis, any technical informa on related

to the original ring of the weapons is likely to be of use indecontamina on, and informa on provided should alwaysbe compiled in such a way as to maximise the possibility ofe ec ve decontamina on.

Site historiesThere are signi cant gaps in the informa on regarding what, ifany, remedia on ac vi es have been undertaken at individualsites. In BiH, where informa on about the use of uraniumweapons was not put into the public domain un l six yearsa er the con ict ended, it is very di cult to ascertain whatoccurred at the sites during this me. A er the informa onbecame known, a German SFOR commander visited theTRZ Hadii site and removed a box of penetrators.89 DuringUNEPs visit they struggled to establish what had been donewith the penetrators, and for a long me assumed thatthey had been destroyed at an ammuni on detona on site.Subsequent requests to NATO resulted in internal enquiriesand an assurance that they were removed to a US radioac vewaste storage site. 90

Sites in Serbia were iden ed early on, fenced o and markedwith radia on signs and are likely to have remained undisturbedun l UNEPs visit. However, one of our interviewees suggestedthat FRY military teams may have removed some penetratorsduring their inves gatory work in 1999.91 The number ofpenetrators removed from a site is signi cant: as mostpenetrators red in an A-10 a ack are thought to remainintact, most of the contamina on at the Balkan strike sitesis s ll present in the form of intact penetrators. 92 Withoutknowing how many are s ll present in the eld, it is impossibleto assess the success of decontamina on work.

In Kosovo, as the Environment Ministry has been unable toundertake any work on contaminated sites, any ac vi es atthese sites are likely to have been done either by local peopleor KFOR. Unfortunately KFOR were not willing to meet with

6.0 Decontaminationand monitoring



our team, though they did con rm subsequently throughcorrespondence that a declara on was made in 2001 thatKFOR would no longer be involved in:detec ng, markingand removing spent Depleted Uranium (DU) in Kosovo. 93 Thiswas apparently in response to UNEPs report in March of thatyear, though it is hardly in line with UNEPs recommenda ons,indeed the recommenda ons for Kosovo speci cally said thatKFOR:should be fully involved in these tasks owing to thesecurity risks posed by mines and unexploded ordnance. 94

Informa on passed from KFOR to the Kosovo EnvironmentMinistry is that no systema c clean-up of sites was undertaken,although one KFOR brigade collected 36 penetrators at oneloca on.95 In fact inves ga ve work was done on all theuranium weapon strike sites in the UK Zone of Kosovo andan Italian study team also inves gated some target sites. 96 At the Gjakove/Djakovica garrison site, major demoli on andclearance work was being carried out by Italian KFOR troopsun l penetrators were uncovered. 97 At the Rikavac site in

Kosovo, UNEP recommended that penetrator impact holesbe repaired. 98 When ICBUW visited the site in 2010 this hadbeen done, possibly by KFOR, although we were informedthe work had been done by local people. Unfortunatelywithout comprehensive and detailed informa on aboutwhat interac ons KFOR troops have had with individualsites, priori sa on of environmental survey work in Kosovowill be impossible, with long term implica ons for e ec vedecontamina on.

Unfortunately, our experience that KFOR was reluctant to

engage with this issue is not unique. One of our intervieweesapproached KFOR troops regarding a site near to where he livedin northern Kosovo that was rumoured to be contaminatedwith uranium weapons. This was an issue of local concern,but he was rebu ed. 99 It is to be hoped that in the futureKFOR will feel willing to publically state that it welcomesenvironmental survey and decontamina on work and willprovide informa on and logis cal support to assist with this.This would enable the ful lment of UNEPs recommenda ons,and ensure that the local popula on is not needlessly at riskfrom uranium weapons contamina on. 100 The need for posi veengagement by KFOR on the issue is par cularly necessary asthe Kosovo Environment Ministry has been reliant on them forinforma on as to the appropriate ac on to take with regard toDU contamina on.101

The decontamination procedureSeveral sites in Serbia had been iden ed by a projectundertaken during the Kosovo con ict itself. Howeverremedia on work on them did not begin un l 2002, a erthe publica on of UNEPs report. Decontamina on work

on the single site in Montenegro at that me a republicwithin the Federal Republic of Yugoslavia alongside Serbia began in 2001, and UNEP was able to observe the results ofthe rst phase of opera ons. The work was nanced by theMontenegrin government, and involved sta from the Centre

for Ecotoxicological Research of Montenegro, the Faculty ofScience at the University of Montenegro and the Vina Ins tuteof Nuclear Sciences in Belgrade. NBC o cers from the FRYArmy were also involved. Following the project, UNEP declaredthe site to be prac cally clean (meaning that there was nocontamina on that was detectable with eld measurements).Consequently, they recommended that a nal check should becarried out and signs placed in case penetrators were found at

the site in the future. 102

In six of the 10 coordinates in Serbia iden ed by NATO, theFRY authori es did not locate any contamina on.103 This wascorroborated by UNEP at Bukurevac.104 Two of the remainingfour coordinates were very close together and were treated as asingle site in the decontamina on programme. The Pljakovicasite, which did not appear on the NATO coordinates, was alsoincluded. At the Borovac site, a second area of decontamina onwas located during the rst stage of decontamina on and wasadded to the programme. 105

The rst stage of work was a surface scan of the site with alpharadia on detectors, a er which the soil was turned over to adepth of 1m using tractors. Dosimetric measurements weretaken, and any solid pieces of penetrator and contaminated soilwere removed. Throughout this me, air quality was measuredwith vacuum pumps to see if any DU was being disturbed.Teams would separate out metal and contaminated soil, whichwas then taken to the waste repository at the Vina ins tutein Belgrade for inde nite storage. The process of measuring,

nding contaminated soil and removing it was repeated, and

the procedure was likened to ne archaeological research.Access to the site was limited, and workers were restricted tospending six hours a day on the site, and only working for twoweeks at a me. They wore protec ve clothing and underwenthealth checks before and a er the work. 106

The structure of the soil at each site determined the paths ofthe penetrators in the ground, and how deep it was necessaryto dig. At Pljakovica, the ground was rocky and very hard towork with machines and they had to resort to manual tools.They found penetrators lodged in rocks at depths of up to30cm. At Borovac 2, the penetrators were found up to a depthof 2.5m. In some cases the path of the penetrator under thesoil was far from straight it was described to us variously asa corkscrew ,107 or an arabesque .108 Only when the penetratorshad hit hard surfaces were they to be found above ground.The thoroughness of the Serbian decontamina on projectswork can be demonstrated by the fact that UNEP could notdetect any surface contamina on at Borovac, 109 even thoughthe Serbian authori es subsequently removed 49 penetrators,36 jackets and 1500 kg of contaminated soil. 110

In total the work in Serbia involved surveying 29,724m2 ofland, and removing almost 10,000kg of soil to a radioac vewaste repository. 111 Work ran in Bratoselce from September toNovember 2002 and the same months the following year; inPljakovica from July to November 2004; in Borovac from July



to December 2005 and June to July 2007; and in Reljan fromSeptember to December 2006 and April to June 2007. 112

At the site in Montenegro, the work involved surveying a totalof 45,000m2, of which 18,000m2 was found to be contaminated.In total 242 whole penetrators and 49 fragments (equivalentto 16 whole penetrators) were removed. Around 200 kg ofhighly contaminated soil with ac vity between 104 and 3.5x106

Bq/kg was removed, and about 6m3

of low radioac ve material(about 7000kg) was removed and buried in a bunker near tothe site. The work lasted 220 working days with a team of 10-15 professionals (seven civilian and eight military) and around10 more in logis cs.113

It was clear that even this detailed work was not su cientto completely remove contamina on from the sites. Manypenetrators remain unaccounted for, many of them possiblymore than 2m under the ground, and therefore undisturbedby the work. On average, across all sites where gures areavailable, only 6% of penetrators known to have been redhave been removed although this average masks greatvariability. On sites in Serbia, BiH or Kosovo where only theremoval of penetrators on the surface has been done, thenumber of penetrators removed is generally fewer than 10.However, where decontamina on work has been done inSerbia and Montenegro, in some cases just over 50% of thepenetrators have been removed 328 penetrators in the caseof Bratoselce.114

The Serbian authori es are clear that it is impossible tocompletely decontaminate the areas, and they work on the

assump on that current e orts have not been su cient. 115 With regards to the Cape Arza site in Montenegro, UNEP notedthat most penetrators were probably intact somewhere onthe site. 116

As noted in a 2001 brie ng on DU from Switzerlands SpiezLaboratory, which was involved in tes ng samples takenby UNEP, the use of uranium weapons: leaves behind along-las ng contamina on on the ba le elds, which is

not compa ble with civil radia on protec on norms. This

argument holds independently whether or not - objec vely -

there is a danger to man and the environment. 117

Cost of decontaminationThe total number of individual PGU-14/B DU projec les redat sites in Serbia was around 3,000. The GAU-8 gun on the A-10

res at a speed of 3,900 rounds per minute, so mixed togetherwith the high explosive rounds, this would represent 65seconds of con nuous ring.118 In total, the decontamina onwork at these sites cost 116m Dinar 119 (equivalent to about$1.479 million US), and took approximately 41,000 workingperson-hours. 120

The decontamina on of the Cape Arza site in Montenegrocost DM 400,000 (almost $280,000 US) and took about 5,000working person-days. This was to clean up 480 high explosive

and DU rounds, which in total took less than eight seconds tore. The site was situated in an important area for tourism, but

at the me of UNEPs report, nancial support was required tocomplete the decontamina on programme. 121

By comparison, the budget for the surface decontamina onat the TRZ Hadii site was 138,000 KM (about $96,000 US), tocover training in decontamina on, prin ng informa on for the

public, equipment, personnel and travel costs.122

Current status of sitesFollowing this work, the Serbian authori es have an ac venetwork of detectors that measure radia on in the airthroughout the country. There are ongoing plans to monitorwater, soil and air in order to ensure that there is minimal riskfrom the sites. 123

The six sites in Serbia, such as Bukurevac, where nocontamina on has been found raise some interes ng ques ons.It is possible that a small amount of contamina on is present,but below the limit of detec on; in which case there is li lecause for concern. However, it is also possible that a signi cantnumber of penetrators are buried at the sites, but are notdetectable from the surface. Another explana on is that theNATO coordinates are incorrect, and the contamina on iselsewhere.

In BiH, no subsurface decontamina on work has been done,but at the TRZ Hadii site, removal of contaminated materialand covering of impact hot spots has been undertaken.

However, the ICBUW team was s ll able to detect the radioac vesignature of uraniums decay products on the concourse usinga handheld gamma detector at one point above an impacthole this was more than 40 mes the background radia on.Soil around the Hadii TRZ is s ll known to be contaminated,124 and there are thought to be more than 900 penetrators buriedat the site. 125 Once the Hadii Ammuni on Storage Depotis cleared of mines and UXO, decontamina on work may berequired, as NATO gures suggest that between 1187 and1900 DU rounds were red at the site. 126 There is an ongoingmonitoring programme for the groundwater in and aroundthe TRZ Hadii site but the authori es are less concernedabout Rosca and Pjelugovici, because fewer rounds were redthere. 127

In spite of extensive work, the six sites around Sarajevo haves ll not been iden ed,128 and it is of great importance thatNATO produces detailed informa on about these sites. TheHan Pijesak site in Republika Srpska is s ll a military site, thoughit is not thought to currently be in use. Previous a empts toundertake decontamina on there were frustrated by the lackof a radioac ve waste storage facility.

As far as ICBUW is aware, no systema c decontamina on hasbeen undertaken on any sites in Kosovo, and as detailed above,there is very li le capacity to study the current status of thesites. This is par cularly problema c, as Kosovo accounts for



Serbian decontamination team sweeping soil for radioactive hotspots with the help of earth moving equipment at Borovac in 2007.

Serbian decontamination team removing the jacket of a 30mm DU round at Borovac in 2007. Contaminated materials and soil from the site were removedto the Vinca Nuclear Institute near Belgrade for vacuum packing and inde nite storage in their low-level waste repository.


18/32

Agency notes, there is: much debate on the problemsassociated with exis ng models and li le consensus on how

chemical reac ons and eld parameters should be determined

for eld applica ons. 135

Even with ongoing monitoring, there are uncertain esregarding the safe levels of uranium in drinking water. Themost recent WHO provisional guideline amount gure of 15

g/litre is based on the calcula on of a 60kg adult drinkingtwo litres of water a day. However, as there is insu cient dataon the carcinogenicity of uranium in either children or adults,this gure is based on a sub-chronic 91-day study done onrats, and is of ques onable relevance to a human exposed touranium in drinking water over a longer me period.136

When the uncertainty over the long term impact ofcontamina on is coupled with its socioeconomic impact,it becomes clear that the best possible approach whenimplemen ng post-con ict management and decontamina onprocedures is a precau onary one. In prac ce, this meansthat decontamina on work should aim to remove asmuch contamina on from the environment as is feasible.Interna onally, states should be transparent and open aboutthe use of uranium weapons, as well as sharing exper se andproviding technical assistance for decontamina on.

At numerous sites, the presence of An -Personnel Mines(APM) and UXO is a complica ng factor in the assessment ofuranium weapons contamina on. In 11 of the 26 sites visitedin BiH, Kosovo and Serbia and Montenegro, UNEP was unableto make a full assessment due to the presence of mines and/orUXO.137 During our visit we met with the na onal Mine Ac onCentres (MAC) in both BiH and Kosovo. While both wereextremely helpful, it was clear that managing uranium weaponscontamina on was not a priority for either organisa on,although it was recognised as being a complica ng factor forsome sites by the BiH MAC. In the Kosovo MAC, the issue wasbelieved to have been dealt with by KFOR.

At least one of the contaminated sites in BiH, the HadiiAmmuni on Storage Depot, is s ll to be cleared of mines andUXO,138 however the TRZ Hadii site has undergone mine andUXO clearance. As previously men oned, some work was doneon the site prior to the release of informa on about the use ofuranium weapons, which undoubtedly exposed the deminingteam to needless risk.

Where the presence of DU is known or suspected, a technicalnote is available, as part of the Interna onal Mine Ac onStandards, to guide demining teams and to help them tominimise risks.139 However, this technical note is not binding,


7.0 Comparisonwith explosive

ordnance disposal

the vast majority of contaminated sites in the region: 85 sites(represen ng around 72% of the uranium weapons rounds

red in the Balkans), as opposed to 10 in Serbia (9% of therounds), 11 in BiH (18%) and one in Montenegro (1%).129

During ICBUWs me in Kosovo we visited four sites, anddid not locate any surface contamina on although we werenot equipped for a proper survey, as this was not the purpose

of our trip. We visited three sites that UNEP had previouslyvisited: Gjakove/Djakovica garrison, Radoniq/Radonjick Lakeand Rikavac. ICBUW also visited coordinate points that to thebest of our knowledge had never been surveyed, in a eldadjacent to the Rikavac site.

Although the Kosovo Environment Ministry assured us thatthey have taken steps to warn local people about the hazardsof contamina on, none of the people we spoke to on our visitto these sites seemed to have been warned. Neither wereany of these sites marked (although marking sites whereappropriate was one of UNEPs recommenda ons) 130 and ahazard sign that was visible in pictures of Radoniq/RadonjickLake in UNEPs report was no longer there.131 A eld directlynext to the coordinates beside Rikavac was being used to growle uces, and we were told that local people had lled in holesin the road where there were visible signs of the a ack atRikavac. Unfortunately, without proper environmental surveywork, it is impossible to tell whether these ac vi es put thelocal popula on at risk.

Long term consequences ofcontaminationIn the long term, li le is known about the consequences ofuranium weapons contamina on. Although at some sites therisk of resuspension of uranium is considered to be of greaterconcern in terms of human health, for most sites in the regionthe long term issue is likely to be groundwater contamina on.Transport of DU in soils and groundwater is in uenced bynumerous environmental factors. 132

The mescales involved can be extremely long calcula onsby UNEP for sites in Kosovo for the me taken for uraniumcontamina on to reach nearby wells give gures that rangebetween one day at the lower band, and 312 years at theupper band. Given such uncertain es, it is clear that uraniumweapon contamina on saddles countries with a burden ofenvironmental monitoring that could last several life mes. 133 This burden is increased by the cost of technology, such as ICP-MS, that can reliably di eren ate between natural uraniumand DU in the environment.

The standard prac ce is to try and es mate the possibleconsequences of contamina on using computer models ofuranium transport in soil, taking into account soil composi onand other factors. In some cases where this has been done foractual strike sites, the results have suggested that signi cantcontamina on of groundwater could occur a er some mehas elapsed. 134 However, as the US Environmental Protec on


19/32

the use of uranium weapons in areas where civilians may beat risk of exposure could breach the principle of dis nguishingbetween combatants and civilians (codi ed in Ar cle 51 ofAddi onal Protocol I to the Geneva Conven ons). Anotherrelevant obliga on is Ar cle 57 of Addi onal Protocol I, underwhich par es are required to take all feasible steps to minimizethe e ects of military ac vi es on civilians.

The Interna onal Commi ee of the Red Cross (ICRC) hasiden ed one of the rules of customary IHL as being that allfeasible precau ons should be taken to minimise incidentaldamage to the environment, and that a lack of scien c certaintydoes not absolve a party from taking these precau ons.Taken with Ar cle 58 of Addi onal Protocol I, which obligespar es to take all necessary steps to protect the civilianpopula on against the dangers resul ng from militaryopera ons, there is a clear legal case that states shouldobserve precau on both in the use of uranium weapons, andin decontamina ng a ected areas. 145 Considering the cost of

decontamina on, and the barriers to carrying it out e ec vely,the moral case is even stronger.

Unfortunately, despite these exis ng legal regimes, stateusers of uranium weapons do not seem to have placed anyrestric ons on their use, ci ng the unresolved scien cissues surrounding their long-term impact as jus ca on.A er con ict, as the experience of the Balkan states shows,there has been li le interna onal focus on assis ng states toquan fy the problem, study any e ects and to remediate sites.Contamina on has apparently been le to individual states to

resolve, as one of many other compe ng priori es. There isan immediate need for transparency over the use of uraniumweapons, and technical assistance with decontamina on. Inthe medium term states should consider whether there is acase for focused interna onal measures that address thespeci c characteris cs of uranium weapons.

While some may contend that ERW cons tute a greater riskthan uranium weapons contamina on, and this jus es thedisparity between the two legal regimes, this posi on is hardto support, either scien cally or morally. From the experienceof the Balkan countries it is quite clear that one cannot withany con dence es mate how many people have been exposedto uranium weapons contamina on, or quan fy the risk tothese people. Whatever the rela ve risks, it is clear that allfeasible steps to reduce harm and to protect civilians and theenvironment have not been taken.

While the threat posed by ERW is more direct, this shouldnot prevent ac on on uranium weapons. The poten al forharm, and the di cul es of decontamina on, necessitatethat the interna onal community engage more fully with theissue. Countries a ected by DU contamina on can scarcelybe expected to leave poten ally hazardous contamina onuntreated because of a lack of scien c clarity.


and is not clear how well versed actors in the demining worldare in dealing with uranium weapons contamina on. In thetechnical note, it makes it clear that in each country theNa onal Mine Ac on Authority is responsible for warning allmine ac on agencies of any con icts that have taken place andany history of the use of DU.140 Without the users of uraniumweapons making informa on available, it is hard to see howthis could occur.

In the BiH Mine Ac on Standards the procedure for mineand UXO clearance for the whole country, there is no men onof DU, and the document clearly states that all mines and UXOare to be destroyed in situ unless there are compelling reasonsto do otherwise. 141 This clearly raises the possibility thatcontaminated material could become resuspended in the air.When the demining of the TRZ Hadii site was being planned,this was a ma er of some discussion, and it was decided thatthe more immediate risk was from moving mines or UXO.142 ICBUW has no reason to query this judgement, but obliging

deminers to face a choice between the risk of losing a limb andof increasing their risk of developing cancer later in life (as thisquandary was characterised by one of our interviewees) 143 isfar from ideal.

International legal statusIn a legal context, a comparison between uranium weaponsand Explosive Remnants of War (ERW) is instruc ve. Whileuranium weapons are not subject to any speci c interna onalregula ons beyond the general obliga ons of interna onalhumanitarian law (IHL), ERW are the subject of variousinterna onal norms, and obliga ons and procedures are wellunderstood.

Although primary responsibility for providing informa on andcoordina ng demining ac vi es falls to belligerents and thosecontrolling territories containing explosive remnants of war,procedures exist for ensuring that the UN can establish mineac on centres if necessary. There is also an established bodyof NGOs and private contractors with a wealth of experienceand sta . Furthermore, many in the interna onal community

fund Explosive Ordnance Disposal (EOD) ac vi es, whether ornot they were involved in the original con ict. As such, thereis an understanding that a ected countries are not le to dealwith this legacy of con ict themselves.

Interna onal treaty regimes apply to the use of APM, clusterbombs, and other explosive devices that may become post-con ict hazards. These include the O awa and Oslo trea es andProtocols II and V of the Conven on on Certain Conven onalWeapons. 144 Their use is also governed by IHL, as is the casewith all weapons. While these norms are not su cient to fully

protect civilians from the e ects of such weapons, they doserve to reduce the risks inherent in their use.

By contrast, the norms regarding the use of uranium weaponsor other toxic remnants of war are those found within IHL,rather than trea es that cover them speci cally. For example,


20/32

follows that the challenges faced by Iraq in environmentalsurveying, conduc ng health studies and decontamina on,are likely to be much greater than anywhere in the Balkans.By almost any measure, the situa on of uranium weaponcontamina on in Iraq is many mes worse than that of theBalkans, and the need for interna onal assistance many mesgreater.

While the experience of the Balkans may be instruc vewhen considering Iraq, it is worth bearing in mind that thecontamina on picture from tank based uranium weapons maybe very di erent from A-10 re. Instead of penetrators driveninto the ground, much more contamina on may be on thesurface, or may become aerosolised. Similarly, the di erentsoil types, land use and climate in Iraq could mean that thecharacteris cs of contamina on are very di erent. For thatreason, extensive eldwork, and risk reduc on programmeswith interna onal assistance are urgently required. While theUK Department for Interna onal Development (DfID) funded

a UNEP run programme to train Iraqi na onals to assess theextent of contamina on, this was limited in scope, and UNEPhas called for the interna onal community to fund futureprojects by the Iraqi Environment Ministry.151

Because of the number of unanswered scien c ques onsrela ng to the poten al impact of uranium weapons, ICBUWhas long advocated a precau onary approach to their use. The

ndings of this report, which highlight the almost completelack of regula ons governing their use and post-con ictmanagement, serve only to strengthen that call.

It is clear from the experience of the Balkan countries that,more than a decade on from the use of uranium weapons, muchs ll needs to be done to address the legacy of contamina onand reduce the poten al risk to civilians. Again, precau onis the principle that should be applied in decontamina on.The recommenda ons made by UNEP in their three reportsremain the most authorita ve standard for reducing risk. Asfar as ICBUW has been able ascertain, nowhere have theybeen implemented in full. It is to be hoped that a renewedfocus on the issues iden ed in this report can help to remedythis situa on.

Health studiesComprehensive and well designed health studies are ofparamount importance in order to establish the actual healthconsequences of uranium weapons use. The type of studyrecommended by UNEP in their report on BiH is one possibleapproach. 152 This report has men oned some promising early


9.0 Recommendations

Had NATO forces been employed in greater numbers, orhad they engaged in a ground war, the uranium weaponscontamina on in the Balkans would almost certainly have beenfar greater. As it was, the A-10s GAU-8 cannon was used oncompara vely few occasions. The reasons for this are speci cto these con icts, and therefore no reason for complacencyabout the scope of contamina on during future con icts.

In the case of Kosovo, where the ostensible purpose ofmilitary ac on was to a ack Serb units suspected of humanrights viola ons in Kosovo, US Air Command favoured targe nginfrastructure in the rest of Serbia instead. E orts to locate

Serb ground units within Kosovo the type of engagementfor which the A-10 was designed were almost completelyine ec ve , and gave li le opportunity for the GAU-8 to beused. 146

Furthermore, due to an ongoing threat of Serbian surfaceto air missiles during the con ict and the perceived po

DU weapons in the balkans.pdf

Documents

Transcript of DU weapons in the balkans.pdf