environmental assessment of mercury pollution in two artisanal gold

technical report for partnership africa canada’s just gold project

BUTUZI , SOUTH KIVU SOME, ITURI

Environmental assessment of mercury pollution in two artisanal gold mining sites in eastern Democratic Republic of the Congo

First published in November 2016 by the United Nations Environment Programme© 2016, United Nations Environment ProgrammeUnited Nations Environment ProgrammeP.O. Box 30552, Nairobi, KENYATel: +254 (0)20 762 1234Fax: +254 (0)20 762 3927E-mail: [email protected]: http://web.unep.org/

This publication may be reproduced in whole or in part and in any form for educational or non-profit purposes without special permission from the copyright holder provided acknowledgement of the source is made. No use of this publication may be made for resale or for any other commercial purpose whatsoever without prior permission in writing from UN Environment. The contents of this volume do not necessarily reflect the views of UN Environment, or contributory organizations. The designations employed and the presentations do not imply the expressions of any opinion whatsoever on the part of UN Environment or contributory organizations concerning the legal status of any country, territory, city or area or its authority, or concerning the delimitation of its frontiers or boundaries.

Cover Image: A piece of ‘sponge’ gold produced on burning of mercury-gold amalgam

Photo Credits: 3,4,5,7,8,15,16,17,18, 19,21,33,34 © Paul Cordy

Cover photo, Front Inset, 1,2,6,9,10,11,12,13,14,20,22,23,24,25,26,27,28,29,30,31,32,35,36,37,38, Back Inset © Hassan Partow/UN Environment

All the photos in this report were taken in Butuzi, South Kivu and Some, Ituri, except for photos 14 and 19 which were taken in Mongbwalu, Ituri.

Design and layout: Antonia Mendes and Cristina Poiata

Artisanal gold mining is degrading landscapes in eastern Democratic Republic of Congo

TechnicalReportforPartnershipAfricaCanada’sJustGoldProject

2

AcknowledgementsThisreportwaspreparedbyPaulCordybasedonfieldworkcarriedoutineasternDRCin2016.The

studywas coordinated by Hassan Partow (UNEP/Post-Conflict andDisasterManagement Branch),

whoalsocontributedsubstantiveinputsandeditedthereport.

Fieldworkwas carriedout in collaborationwith theMinistryof Environment,NatureConservation

andSustainableDevelopment(MECNDD),theCongoleseEnvironmentAgency(ACE),theMinistryof

Mines (SAESCAM) and in consultation with local authorities and mining cooperatives. Sample

analysiswasconductedbySpiezLaboratoryinSwitzerland.PartnershipAfricaCanada(PAC),theUN

peacekeepingmission inDRC (MONUSCO)and theUNDevelopmentProgramme (UNDP)provided

logisticalsupportforthefieldwork.

Specialthanksareextendedtothefollowingpersonsfortheirassistanceinfacilitatingthefieldwork

including:DanielRuiz,M'handLadjouzi,RosemaryKioni,AlainMulimbi(MONUSCO);JamieGraves,

Zacharie Bulakali (Partnership Africa Canada); Félicitér Kabonwa, Michel Bagalwa (Kaziba Local

Authority);MuhsoshereGanywamulume(KazicomCooperative);ClaudiaKamanji(UNEP);andJosef

Debeus(UNDP).

UNEP would also like to extend its sincere appreciation to Spiez Laboratory for carrying out the

laboratory sample analysis including in particularMario Burger,Marc Stauffer, JasminOssola and

AndréPignolet.

UNEPwouldliketoacknowledgethepersonswhoreviewedthereportincluding:JeanClaudeEmene

Elenga,PapyMola(AgenceCongolaisede l’Environnement);JoanneLebert,ZuziaDanielski,Gerard

vanderBurg (PartnershipAfricaCanada); LudovicBernaudat,KennethDavis (UNEPChemicalsand

Waste Branch); Muralee Thummarukudy, David Jensen, Pauliina Upla, Andrew Morton, Sophie

Brown(UNEPPost-ConflictandDisasterManagementBranch);andUsmanTariq(UNITARChemicals

andWasteManagement Programme). Sincere thanks to AntoniaMendes and Cristina Poiata for

finalizingthereportlayoutandpublication.

EnvironmentalassessmentofmercurypollutionineasternDemocraticRepublicoftheCongo

3

Tableofcontents

Summary..............................................................................................................................................4

1. Introduction..................................................................................................................................8

Artisanalgoldmining...........................................................................................................................9

GoldminingineasternDRC...............................................................................................................14

ButuziAGMsite,SouthKivu...............................................................................................................14

SomeAGMsite,Ituri..........................................................................................................................24

2. Mercuryassessmentmethods....................................................................................................30

Sedimentsampling.............................................................................................................................30

Dustsampling.....................................................................................................................................31

Mercuryinventory.............................................................................................................................32

3. Analysisresults...........................................................................................................................34

Sedimentanalysis...............................................................................................................................38

Dustanalysis.......................................................................................................................................38

Mercuryinventory..............................................................................................................................38

4. Mercuryreduction......................................................................................................................40

Actionplanformercuryreduction.....................................................................................................42

5. Butuzi:Slopestabilizationandriskmitigation.............................................................................44

6. Some:Forestdegradationandbiodiversityloss..........................................................................50

7. Miningexploration.....................................................................................................................54

8. Occupationalsafety....................................................................................................................56

9. Barrierstoformality,andtheartisanalminingpovertytrap........................................................58

10.Conclusionsandrecommendations.............................................................................................61

Appendix1:Laboratoryanalysesinterpretation.................................................................................63

Appendix2:Laboratoryanalysesresults.............................................................................................69


4

SummaryThisreportisanenvironmentalassessmentofmercurypollutionintwoartisanalgoldmining(AGM)

sitesintheeasternDemocraticRepublicoftheCongo(DRC).ItwasconductedbytheUnitedNations

Environment Programme (UNEP) based on fieldwork in Butuzi, South Kivu and Some, Ituri. The

purpose of this assessment is to provide technical advice to Partnership Africa Canada’s (PAC)

pioneering ‘JustGold’project,whichaimstobring legal,conflict-free,andtraceablegold fromthe

two aforementioned artisanal mine sites to international markets. In this respect, technical

assistanceonmercuryreductionresultinginincreasedgoldproductionisalsoanintegralcomponent

oftheincentive-basedmodelthatPAC’sprojectaimstodeveloptoencourageartisanalproducersto

selltheirgoldthroughlegalchannels.

This study represents the first field assessment of mercury use, emissions and releases to the

environment by the artisanal goldmining sector in the DRC. Themethodology for preparing the

inventoryofemissionsand releasesat the twoartisanal goldmining sites complieswithguidance

developedby theUNEPGlobalMercuryPartnership for theMinamataConventiononMercury. In

particular, thebaselineestimates fromthis studyshouldhelp informthepreparationof theDRC’s

Minamata Initial Assessment (MIA) and National Action Plan on Artisanal and Small Scale Gold

Mining (NAP)1. In addition, the report addresses other environmental problems and associated

healthandoccupationalrisksatthetwoaforementionedAGMsites,aswellassomeofthebroader

governancechallengesfacingtheartisanalgoldminingsector.

In addressing themultiple environmental challenges at the twoAGM sites, a progressive strategy

taking intoaccount investmentcostsand thecomplexityofmeasures isproposed. It comprisesof

targetedtechnicalinterventionsto:

i)reducemercuryuseandpollution;

ii)helpmitigateslopedestabilization,deforestationandbiodiversityloss;and

iii)improveoccupationalsafetyintheminingsites.

1TheDRCMIA-NAPwillbeimplementedwithtechnicalassistancefromUNEPandUNITAR,andfinancialsupportfromtheGlobalEnvironmentFacility.Projectactivitiesareplannedtostartinend2016.

SnapshotofArtisanalGoldMiningineasternDRC

Ø 80%ofallartisanalminersineasternDRCdigforgold

Ø Anestimated160,000-200,000goldminers

Ø ~1,000AGMsitesineasternDRC(IPIS:857;BGR:1305)

Ø AGMproductionestimatedat8-10tonnesperannum

Ø Officialexportsaround~200Kgs(i.e.98%smuggled)

Ø GoldisthemainresourcefinancingarmedandcriminalgroupsineasternDRC,accountingfor50%of

profits

Source:IPIS(2015):MappingConflictMineralsinEasternDRC;UNEP-MONUSCO-OSESG(2015);Experts’backgroundreportonillegal

exploitationandtradeinnaturalresources


5

MercurypollutionMercury use by artisanal goldminers iswidespread in easternDRC, and there is little knowledge

amongusersaboutthetoxicityandenvironmentalhazardsitcanproduce.Extensiveuseofmercury

stems from its quick, easy and effective separation of gold from the ore. Significant mercury

concentrations can be found in soil and rivers near mine sites, as well as inside homes and

businesseswheremercury is burned. The greatest health risk forminers,women, and children is

fromopenamalgamburninginsmallcommunitiesandkitchens,aswellasinurbancentres.

Fewof thoseencounteredduring the fieldassessment (miners, governmentofficials, goldbuyers)

seemedtohavepreviousawarenessofthehealthdangersofmercury.Moreover,noonewasaware

that gold buyers in cities releasemercurywhen smelting andpurifying the gold.Mercury is likely

evaporatedwhenthespongegoldismeltedintoingots.

Atmosphericmercuryemissionsmakeup29%of total releases frombothAGMsites.Mostof the

mercuryusedislostintheminewastes(71%),andpotentiallyuptohalfofthatmaybecarriedby

run-offwiththesuspendedsediments.Theresultingmercurycontaminationmaynegativelyimpact

fluvialbiotaformanykilometresdownstreamofsiteswherewastedischargesintostreams.

Laboratoryanalysisconfirmedthepresenceofmercury ineverycontext inwhich itwasexpected,though concentrations are relatively low compared to artisanal mining communities in othercountries inAsiaandSouthAmerica.This reflects the fact that localminersdonotburnamalgamindoors, nor usemercury in excessive amounts of the gold they can recover. Artisanalmining ineasternCongocreatesavast,dispersed,lowgradewebofmercurycontaminationandmethylationhotspots.Thiscanimpactthelongtermhealthofecosystemsandpeopleacrossbroadswathesofpoor, ruralareas. Inaddition,arsenicconcentrations10to20 foldaboverecommendedstandardsweredetecteddownstreamoftheButuziminingsites,whichcancauseserioushealthproblems.

Table1:AveragevaluesformercuryinventorymeasurementsinButuziandSome.

Inventorymeasurement AverageN=11

Standarddeviation

Mercuryapplied 4.09g 3.98

Amalgam 0.50g 0.42

RecoveredMercury 3.62g 3.73

Goldsponge 0.23g 0.19

Mercurylost 1.00g 1.14

Mercury:Gold 4.3:1 2.74

%losttoair 29 0.10

%losttoland 71 0.10

Apositivefindingisthatsomeminersproducerelativelymodestandconfinedmercuryreleasesby

concentratingthegoldintheiroreusingsimplesluicesbeforeapplyingmercury.Thisrelativelygood

practice, however, is typically offset by the large quantities of tailings and primary material

processed usingwhole ore amalgamation (inwhichmercury is applied to all of the primary ore).

Preliminary mercury inventories of the sites suggest that the ratio of mercury losses to gold

produced lies in the range between 2:1 and 7:1 (averaging 4.5:1), which is consistent with

assessmentsofotherAfricanartisanalminingsiteswithsimilarworkpractices.Thisratioofmercury


6

to gold use is high and inefficient by international standards but low in comparison with more

mechanizedpracticesintheAmericasandAsia.

ActionplanforreducingmercuryemissionsArtisanalmining sites shouldbeorganizedso thatallmineralprocessing,mercuryuseand tailings

storageoccur1km (or at least500m)away fromall naturalwater sources, villages,markets, and

restaurants.Thesitemustbeonhighgroundtopreventinundationandalldrainagefromthearea

must go through settling ponds to reduce suspended sediment emissions. The process of

amalgamationofreprocessedtailingsmustbereplacedbybetterprimaryprocessingoforesothat

allrecoverablegoldthatcanbeconcentratedoramalgamatedisextractedinthefirstpass.Further

recommendations for testing include using a blue bowl gold concentrator to process tailings and

recovermercuryandgold,andvaryingthesluicewaterflowrate,amongothers.

Mercuryusecanbegreatlyreducedoreliminatedcompletelybychanginghowtheoreisprocessed.

Thestrategyshouldbetoprioritizeincreasingpre-concentrationbeforeamalgamating.Efficientpre-

concentrationincursgoldlosses,soprocessinginallpartsofthechainneedstobeimproved,from

explorationandplanningofdepositextractionandmineworkings,choiceandpre-processingofore,

efficient grinding for gold liberation, as well as concentration andmonitoring of gold in ore and

waste.Overall,highergoldrecoverycanbeachievedwithoutmercury.Technologyandtrainingare

thesolution,assumingnotonlyformalitybutaccesstoinvestmentandservices.

Aprogressiveplanofactiontoreducemercuryuseandpollutionisproposedbelow.Theplanstarts

with changes requiring the least investment and training, followed by progressively costlier and

more complex changes that can be added when the miners are ready to take the next step2.

Consultationswithminersandlocalcommunitieswillneedtobeorganizedtoobtaintheirviewsand

buy-inontheproposedplan.

Immediateaction:

1. Promote use of retorts or fume hoods to recover mercury and reduce exposure and

inhalationoftoxicmercuryvapours.Minersmuststopusingleavesasaretort.

2. Promoteconcentrationanddirectsmeltingasanalternativetowholeoreamalgamation.

3. Improvewater supply andwastemanagement using impermeable reservoirs, refilling old

minepits,andreorganizationofprocessingactivities.Thiswill isolatemercuryreleasesand

reduceinfiltrationofwaterintolandslideinitiationzones.

Mediumtermobjectives:

4. Analyze goldoremineralogy to guide changes inprocessingmethods forbothmilling and

gravityconcentration.

5. Optimize mill feed, speed, duration, and grain size control, also using participatory

controlledtrials(includinginstallationandcomparisonofexistingmillingequipment).

6. Improve existing sluice practices by optimizing flow velocity, feed consistency and

turbulencethroughparticipative,controlledtrialsonsite.

2DetailedinformationandguidanceonthesolutionsoutlinedbelowisgiveninthefollowingUNEPdocument:ReducingMercuryUseinArtisanalandSmall-ScaleGoldMining:APracticalGuide


7

7. Fieldtestdifferentmechanicalconcentrationdevicesonsite,beginningwithshakertables,

spirals,bluebowls,andjigs,usingshorttermtrialsoflabscaleequipment(whichcanalsobe

usedtofurtherupgradeprimaryconcentratestoenabledirectsmelting).

Longtermgoals:

8. Implementbestpracticewetmillingandconcentrationprocesses.Potentialoptionsinclude

usingtheArtisanalGoldCouncilBurkinaFasomicroscaleprocessingplantasamodel3.

9. Usetrenchingandpanning(basic)andpotentiallyalsomoderngeophysicalinstrumentation

(suchaselectromagneticsurveysorother)tostudygoldreservesandplanminingactivities.

Theaboveplanwillimprovegoldrecoverywhileprogressivelyreducing,andeventuallyeliminating,

mercuryuse.Itappliesequallytohardrockandalluvialminingsites.

OtherenvironmentalissuesInadditiontomercurycontamination,othermajorenvironmentalimpactsatthetwoAGMsitesare

deforestation(Some/Kafiawemaarea),increasedriskoflandslides(Butuzi),andfluvialdegradation.

Inaddressingthesechallenges,asystemsapproachmaybeadoptedtocapitalizeonsupplementary

benefits that could be derived by linking transparency, land use planning, and pollution control,

where feasible. For example, sustainable forest management and timber transparency initiatives

wouldbeanaturalcorollaryactivity,withbenefitstocommunitiesandlocalecosystemsthatcome

fromlanduseplanningandtechnologicaltransformation.

Land use planning in the context of gold production requires basic exploration and ore body

estimationinordertotightlyconstraintheexcavationsandsitetailingsdisposalandcamplocations

with minimal impact and maximum regenerative potential. Combined mine and forest planning

shouldbedirectedtowardoffsettingsocialandfinancialrisksofformalizationandnewproduction

technologies. Addressing these problems in isolation would ultimately overlook important

unanticipatedconsequencesandmisspotential leveragepoints.Forexample, itwouldnotbewise

orsustainabletosolelyfocusondivertinggoldwealthfromarmedgroupswithoutconsiderationof

the rapid irremediable devastation to large patches of contiguous Central African equatorial

rainforest.Neitherisitpossibletoignorethepossibledeathsandinjuriesofhundredsofpeoplein

landslides and catastrophic flooding induced by uncontrolledmountain slope destabilization from

artisanalmining.ThesethreatsriskreputationaldamagetoPAC’spioneeringpilotprojectandmay

undermine its impact. It is therefore important to establish a risk mitigation plan demonstrating

proactive mitigation actions will be taken. In the Butuzi case, a number of simple measures are

recommendedthatcoulddiminishslopedestabilizationandexposureofpeopletothehazardarea,

butthethreatwillremain.

3SeeAGCBurkinaFasoUpdatereportandvideofilmEmploiel‘OrArtisanalpourl’AideauDéveloppement.


8

1. IntroductionThisreportisanenvironmentalassessmentofmercurypollutionintwoartisanalgoldmining(AGM)

sitesintheeasternDemocraticRepublicoftheCongo(DRC).ItwasconductedbytheUnitedNations

EnvironmentProgramme(UNEP)basedonfieldwork inButuzi,SouthKivu(11-15March2016)and

Some, Ituri (18-22 March 2016). The specific purpose of this assessment is to provide technical

adviceandsupporttoPartnershipAfricaCanada4(PAC)intheimplementationofitspioneering‘Just

Gold’ project, which aims to bring legal, conflict-free, and traceable gold from the two

aforementionedartisanalminesitestointernationalmarkets.

Theprimaryfocusofthisstudyisonassessingmercuryuseandreleasesintotheenvironmentand

theidentificationofcontaminatedhotspots.Forthispurposearapidmercuryinventoryinthetwo

AGMsiteswascarriedoutbasedonguidancedevelopedbytheUNEPGlobalMercuryPartnership

for the Minamata Convention on Mercury; namely the Minamata Initial Assessment (MIA) and

National Action Plan on Artisanal and Small Scale Gold Mining (NAP). Environmental samples

(soil/sedimentanddust)werecollectedandshippedtoSpiezLaboratoryinSwitzerlandforanalysis.

In addition, the report addresses other environmental problems and associated health and

occupationalrisksatthetwoaforementionedAGMsites,aswellassomeofthebroadergovernance

challengesfacingtheartisanalgoldminingsector.

In addressing themultiple environmental challenges at the twoAGM sites, a progressive strategy

takingintoaccountinvestmentcostsandthecomplexityofmeasuresisproposed.Itiscomprisedof

targetedtechnicalinterventionsto:

i)reducemercuryuseandpollution;

ii)helpmitigateslopedestabilization,deforestationandbiodiversityloss;and

iii)improveoccupationalsafetyintheminingsites.

Furthermore, it is importanttoemphasizethatreducingandeliminatingmercuryuse isnotonlya

health and environmental protection measure. As technological improvements for mercury

reduction leads to increased gold yields, it is also an integral component of the incentive-based

modelthattheJustGoldprojectaimstodeveloptoencourageartisanalproducerstoselltheirgold

through legalchannels.Awarenessraising includingpracticaldemonstrationtrials isequallycritical

forpromotingmercuryreductionmeasuresandtheiradoptionbyartisanalminers.

The technical assistance provided in this report is part of a collaboration between PAC, the UN

peacekeepingmissionintheDRC,MONUSCO,andUNEPtosupporttheformalizationoftheartisanal

goldminingsectorbyhelpingcontroltheextractionandtradeingold.

4PartnershipAfricaCanadaisfocusedondevelopinginnovativeapproachestostrengthennaturalresourcegovernanceinconflict-affectedandhigh riskareas.PAC isanofficialpartnerorganizationof the InternationalConferenceof theGreatLakes Region (ICGLR), and receives funding from a broad range of governments and institutions to address the issueconflictmineralsintheGreatLakesregionofAfrica.


9

ArtisanalgoldminingArtisanal miners use rudimentary methods to

extract minerals due to lack of access to

appropriateequipmentandresources,causedby

poverty and tenure insecurity. Mining refers to

theextractionofthegoldbearingmaterial(“ore”)

from the ground, which is the most dangerous

andlaboriouspartofaminingoperation.Mineral

processing separates gold from the ore. In hard

rock (ore “lode”) mining, one first has to crush

and grind the rocks to liberate the gold. Alluvial

minersusuallyonlyexcavatetheoreandseparate

the gold as it occursmostly as free nuggets and

dust. Both alluvial and lode gold miners

concentratetheiroretomakeseparationfeasible.

With the right equipment it is often possible to

extractthemajorityoftheliberatedgoldusingacombinationofdifferentconcentrationtoolsthat

progressivelyupgradetheconcentrateandcapturegoldlosses.

Chemical separationusingmercuryor cyanide isoftenapplied togoldconcentrates toextract the

portionofthegoldthatinevitablycannotbeextractedusingconcentrationmethodsalone.Mercury

picks up gold particles as itmixes and filters through the grains of ore, forming an “amalgam”of

roughlyequalpartsmercuryandgold.Theamalgamcanthenbeevaporatedtoleaveonlythegold

behind.Mercury iseasilyrecoveredbypanning,thoughsignificantamountsofmercurywillalways

adhere to the ore. These contaminated wastes (or “tailings”) must be disposed of durably and

isolated from bodies of water. Metallic mercury used by miners is insoluble. Under certain

conditions, however, elemental mercury can be transformed into more toxic methyl-mercury

species that can dispersemore easily and bioaccumulate in the food chain.Mercury in the food

supply(especiallyfish)isthemainmechanismofhumanexposuretomercuryforpeoplenotdirectly

involved in mining. Translucent tea coloured water, typical of tropical environments, is rich in

organic compounds and likely has oxidation-reduction potential (Eh) and pH conditions that are

favourableformethylatingmercury.Miningeffluentsalsotendtocontainfinesuspendedsediments.

The nanostructure of suspended clays consist of stacks of atom thick sheets that provide many

excellentbindingsites formercury,andmaycatalyzetransformation intoreactivespeciesthatare

precursorsinthemethylationprocess.

Manyartisanalminersarenotsufficientlyawareoradeptatgoldconcentration,andinsteadapply

mercury to all of their ore, for example by adding it to grinding or concentration devices. This is

known as whole ore amalgamation, and it is perhaps themost important of the worst practices

identifiedbytheMinamataConventiononMercury.Minerswhoconcentrategoldinasluiceusually

emithalfofthemercurycomparedwiththosewhopracticewholeoreamalgamation,becausemuch

lessmaterialcomesintocontactwithmercury.Promotingmoreefficientminingpracticescanhelp

motivatesuchbehaviouralchanges,astheyproducemoregoldwhileusinglessmercuryornoneat

all.

Photo1:Anamalgamofgoldandmercurybeforeburning


10

Mercuryrecyclingisanotherimportantbehaviourchangethatcanreduceemissionsandcosts.Small

mercurydistillationdevicescalled“retorts”condensemercuryvapours thatareemittedwhen the

amalgam is burned, and larger condensers can be built to trap mercury emissions in advanced

refiningsteps.

Map1:MapoftheDemocraticRepublicofCongo,showingthemainfieldsites.GoogleEarthimagery


11

Map2:SpatialcontextofButuziminesiteanditslocationvis-à-visthemainpopulationcentreofKazibavillage.Google

Earthimagery

Floodplain

Flow

dire

ction

RelicForest


12

Map3:OverviewofKaziba/Butuziarea(SouthKivu);samplinglocationsandcodesareexplainedinthemethodologyand

results(Appendix1).GoogleEarthimagery


13

Map4:Some/Kafiawema(Ituri)areashowingdeforestationpatches;samplinglocationsandcodesareexplainedinthe

methodologyandresults(Appendix1).GoogleEarthimagery


14

GoldminingineasternDRCArtisanalgoldmininghasbeenwaxingandwaning ineasternDRCover thepastcentury.The two

artisanal goldmines examined here have only been operational for 10 to 15 years or less,which

reflects the unpredictable, migratory nature of artisanal mining in this region. In terms of

environmental impacts, thesecasesstandoutnotonly for theirmercuryemissionsbutequally for

the landscape disturbance and forest degradation they produce. Nevertheless, it is important to

reduce mercury at these sites to provide a test case for the elimination of mercury use, as the

cumulative emissions of hundreds of such operations across the DRC amounts to a serious

contaminationproblem.Understandingthenatureoftheseoperationsiskeytomodifyingactivities

forthereductionofenvironmentalandhumanhealthrisks.

ButuziAGMsite,SouthKivuTheButuzisiteliesinaruggedhighlandterrainintheMitumbaMountainrangealongthewestern

edgeoftheAlbertineRiftValley.NaturallydominatedbyAfro-montaneforest,theregionhasbeen

almostcompletelydeforestedformanydecades.Administratively,theButuzisiteislocatedinKaziba

Chefferie (Walungu Territory, South Kivu Province),which has an estimated population of around

45,000.Travelling the~55Kmdirt road from theprovincial capitalBukavu toButuzi takesaround

threehoursbyfour-wheeldrivevehicle.

Photo2:ViewofKazibavillagefromthehillsnearButuziAGMsite


15

ThemaineconomicactivityinKazibaChefferieissubsistenceagriculture(cassava,maize,potatoes,

beans)andtraditionallivestockrearing.Inthepast,cashcropping(coffee,quinquina)waspracticed

but appears to be largely neglected. Woodland plantations (cypress, pine, eucalyptus) are

extensivelygrownfortheproductionofpolesandplanks.AmajorsupplycenterforBukavu,wood

productionreportedlyprovidesthemainsourceofcashincomeforthelocalpopulation.

Butuziartisanalmining

Artisanalminingtakesplaceonthesteepermountainslopesthataretoodifficulttocultivate.Inthe

Butuzisite,miningistakingplacealongthefaceofaverysteepmountainslope(>30°withafallof

~200meters) that runs down into the Luzinizi river valley. The crest of the Butuzi hill reaches an

altitudeofaround2,200metersdroppingtoaround2,000intheLuzinziRiverbelow.Minersthatlive

inKazibahikeuptotheButuziminesiteinunderonehour.

EnteringtheButuzisite,onecanseetheuncontrolleddestabilizationoftheslopesupcloseaswell

as in the middle distance and successive ridges in the background. All have created a massive,

imminent landslide hazard by removing anchoring vegetation, and digging unlined reservoirs as a

water supply.Water ismanually pumped into hand dug open channels and/or hoses usingwash

basins, and this feeds the earthwork ‘sluices’ burrowed on themine face.Water percolating into

joints, fracturesandtheslideplanethroughthewatercollectionpitsandhaphazardlyconstructed

water ditches will further weaken the slope over time. Cracks are evident in adjacent slopes,

suggestingthatactivelandslideinitiationzonesareundertensionandreadytocollapse.Minerssend

bulkdirt(eitherbarrenoramalgamatedtailings)aswellaslargeboulderstumblingdowntheslope,

whichcreateswasteslidesthatundermineslopestabilityinalonglinedowntotheriver.Themines

themselvescreateanotherstabilityproblem,asunmappedminetunnelpassagesriskbotheroding

the slope from within and by cave-ins from unexpected intersection of workings. Not all mine

entrancesarebuttressedasinphoto4,andtheinsideissparselyreinforced,ifatall.

SouthKivuProvince

SurfaceArea:69,130km2

Population(2005estimate):3.9million

ProvincialCapital:Bukavu

Populationdensity:60persons/km2

Agriculture:70%ofprovincialGDP

Artisanalminers:>100,000(85%ingoldmining)

Source:MONUSCO(2015),StratégieProvincialdeStabilisationetPland’ActionpourlaProvincedeSudKivu;PNUD(2009)Provincedu

SudKivu:ProfilRésuméPauvretéetConditionsdeVie;IPIS(2014):AnalysisoftheinteractivemapofartisanalminingareasinEastern

DRCongo


16

Photo 3: Butuzimine entrance.Note the grass fringed rain collection pits visible in the far left of the photo, at the top

marginofthedisturbance,andthehosethatbringswaterbygravitytothesluicesbelow.Abovethepersoninredarethree

amalgamationpondsfullofyellowmud.Greyandredwastechutesarevisiblerightofcentre,andlandslidecracksleftof

centre.Theriverbelowisfullofalluvialminerseveryday,allyear

Photos4and5:Rudimentarymineentrancesandtunnelsaresometimesbuttressedbywoodenbeams,butthesesupports

arefewandtunnelsarepronetocollapse.Minerspanoreinshovelstotestthegoldcontentofore


17

Photo6:Impreciseweightsusedinlocalgoldmarketsappeartofavourthebuyers

The Butuzi miners are exploiting primary rock deposits, but only work the unconsolidated near

surfacesaprolitedepositswherethegoldhasalreadybeenliberatedandconcentratedbychemical

weathering.Thissuggeststhata largeamount,perhapsthemajority,ofthegoldintheirdeposit is

containedinthehardquartzrichblocksofrockthattheysendrollingdownthehillastheydigfor

goldbearingsand.Minerssearchouttheyellowandredsands inbetweenharderblocksofrocks,

makingcircuitoustunnelsastheychasethehighgradesandsdeeperintothemountain.Thissearch

isdirectedbyperiodicartisanal“assays5”inwhichminerspanasampleoftheoreinashoveltosee

iftheycanseeparticlesofgoldinside.This“assay”isalsousedtodeterminehowmuchmercuryis

tobeapplied.

Butuzi miners that sluice tend to work in larger teams that coordinate the digging, sluice

construction,watermanagement, and concentration amalgamation roles. It is important to study

andunderstandtheseexistingorganizationalstructuressothattheycanbeadaptedandexpanded

toincludenewrolesdemandedbytheimplementationofbestpractices.Asagroup,sluicingminers

pay about 10 US dollars per day for daily water costs (~5000 FC for water and 4000 FC for pipe

rental).

Butuziminers typicallywork6daysperweekallyear long foranaveragedailygoldproductionof

0.5-1 grams. They arepaid between34-37USdollars per gram.However, themethodsofweight

measuringare impreciseand fromspot checkvisits to the local goldmarketappear to favour the

buyer.Thatis,ifthebuyerweighsthegoldas1gram,itinfactmayweighmorethanthat,whichis

one reasonwhy buyers can afford to apparently give nearmarket price for the deflatedweights

determinedbythem.

5Inmining,anassayreferstoaphysicalanalysisofanore,metaloralloy.


18

Photo7:Notethewebofcracksinthecoppermineontheadjacentslope,andthegreywasterockdumpslidestretching

halfwayacrosstheriverbed.Sinuouspatternscreatedbywalledchannelsbuiltbythealluvialminerscanalsobemadeout

below

TheLuzinziriverbedbelowthisminesiteiscompletelyalteredbytheaccumulationofminewaste,

inducedlandslidesandtheactionofhundredsofminersthatworkalongtheriverdaily.Theminers

areconstantlyexposedtothelandslidehazardabove.

They also restructure the riverbed to create sluices, basins, and cobble walls that protect

manufacturedchannelsandholdbackwastesandthattheyhavealreadyworked.Thesesandsare

repeatedlyreworkedovertimebecausetheirmethodsareso inefficientthatthere isalwaysmore

goldtoextractinanyminewaste.Periodicfloodsprobablyalsohelptomobilizeandconcentratethe

gold.


19

ButuzimineralprocessingButuziminersdonotcrushorgrindtheirore, insteadconcentratingfreegoldfoundinloosesands

inside their tunnels. Some use sluice channels dug out of the slope and lined with banana bark.

Sometimes they also putmercury in the banana bark,which inevitably leads to losses. Individual

minerstendnottousesluices,andinsteadpanconcentratetheirorebyhand.Theymixinmercury

into the whole ore using plastic wash basins in small amalgamation ponds. Eachminer puts one

~50kgricesackoforeintotheplasticbasin,aswellasseveralgramsofmercury,dependingonthe

richnessoftheore.Theamalgamationpondsareunlinedandnotsufficientlycontainedtoprevent

groundwater infiltration or overflow emissions of mercury bearing suspended sediments during

heavyrainsorinundation.

Photo8:Wallandchannelstructureofheavilyworkedriverbed

Wastedischarges fromhardrockmining increeksandrivers,aswellas thedisturbanceofalluvial

miners working in the larger rivers, completely alter the fluvial morphology of areas they work.

Siltationdestroysfluvialecosystems,renderingformerlyfishbearingriversbarrenandeliminatingan

importantsourceoflocalprotein.

Mercuryuseisusuallycontrolledbytheheadofanartisanalminingteam(so-calledPDG,President

Directeur-General, i.e. CEO), who apportions inputs free of cost according to the potential gold

contentintheoreasassessedusingtheirartisanal“assay”methodofpanningasampleoftheorein

asmallgoldpan. ThePDGisusuallyveryconservativeinhisapplicationofmercury, incontrastto

operationsinAsiaandLatinAmericawhichgenerallyapplymercuryfarinexcessofwhatisoptimal,

withthefalseexpectationthatthiswillensuremaximumpossiblerecovery.Minersthenthoroughly

mixthemercuryinwiththeoreorconcentrateusingbarehandsinaplasticwashbasin.Aftermixing,

theoreispannedtograduallyreducethemassoforeinthebasinuntilalmostallisdischargedand


20

they can easily recover the liquidmercury. Thismercury is then squeezed through a cloth, taking

caretocatchtheexcessmercuryinthewashbasin.ThatmercuryisreturnedtothePDG.Next,the

resulting amalgam is burnedover a charcoal fire. InButuzi themercury is dispensed fromaneye

dropper,andinSometheyusesmallsyringes.Mostmercuryislosttothetailings,thoughthemost

immediatepublichealthriskresultsfromopenamalgamburning.

Photo9:Openairburningofgoldamalgamwithatablespoononcharcoalstove


21

Photos10and11:Bananabarkmakesaneffective,ifinefficient,goldconcentrator(left).Propercarpetsandasmooth,10

slopedsluicebedwouldbemoreeffectiveincapturinggold.Plasticwashbasinsmakerelativelypoorgoldconcentrators,

andpropergoldpanswouldbemoreeffective

Photo12:Mercurydropletsvisibleonthebananabark


22

Oreextraction

Amalgamation

Sluice

Ore

Sluiceconcentrate

Burning

Tailings

Amalgam

MercuryVapour Gold

Diagram1:FlowdiagramformineralprocessinginButuzi.Sluicesareonlyusedbygroupsofminers.

Independentminersusewholeoreamalgamationeachtimetheyextractonesackofore.


23

Photos13and14:ThePDGmanagesthemercuryforthesite,whichishandedoutforfreetominers,andthencollectedin

thesamebottleorsyringeaftertheexcessissqueezedoutoftheamalgam

OntheLuzinziriver,belowButuziandadjacenttoKaziba,sluicesaredugintotheriverbedandlined

withlargerocksonthesidesandbananabarkasaconcentrationmedium.Minersscoopsandsand

gravelsintoshovelsandwashthembydippingthemintothewatercurrentatthetopofthesluice

andthrowingthemintotheairandcatchingthemagainseveraltimesinsuccession.Thisseparates

the finesediments thatcontainthesoughtaftergolddust that isconcentratedas they flowdown

thesluice.Thewashedgravelsaretossedintowastepiles.

Photo15:Repeatedreworkingofsandsrightattheedgeofthevillageagriculturalfields


24

SomeAGMsite,IturiSomeislocatedinIturiprovince,inaflatequatorialrainforest(<900meters)whichisstillrelatively

intact but under increasing deforestation pressure. The AGM site is adjacent to one of theDRC’s

most importantprotectedareas,theOkapiWildlifeReserve(OWR),which isalsoaUNESCOWorld

HeritageSitethatisonits‘DangerList’since1997.Situatedwithina25-30kilometredistancefrom

theORW,Somefallswithinthedesignated50-kmbufferzonesurroundingtheentireReserve.

Miners in the Someareaof Ituri province practice amix of alluvial andhard rockmining, though

apparentlyonlythehardrockminersusemercury.Thereisadistinctpossibilitythatalluvialminers

will realize that mercury can also improve alluvial gold production. Therefore, it is important to

monitoralluvialoperationsandprovidethemwithmoreproductivemethodssothattheyhavegood

alternativestomercury.Alluvialminersandhardrockminersleavelargeholesintheforestunfilled

andbarrenofgoodsoil.Thiscreatesapockmarkedlandscapeofstagnantponds,gravellysoils,and

disturbedwatercourses.

Kafiawema,themainSomeareaminesiteinvestigatedinthisstudy,islocated30minutessouthof

Mambasa,mostly on good dirt roads. Although there is aminer’s camp located at themine site,

manyminers live innearbySome,which is about30minutesawayon foot.Mostof Some’s~400

inhabitantsaredirectlyinvolvedinmining,althoughloggingofhighvaluehardwoods(oftenillegally),

andsomeagriculture(mostlysubsistence)arealsopracticed.

IturiProvince

SurfaceArea:65,658km2

Population(2005estimate):4.2million

Populationdensity:65persons/km2

ProvincialCapital:Bunia

Artisanalmining:almostentirelygold

Artisanalproduction:~2tonnes

Source: SiteOfficielde laProvincede l’Ituri (2016):http://provinceituri.info/; IPIS (2014):Analysisof the interactivemapofartisanal

miningareasinEasternDRCongo


25

Kafiawemaartisanalmining

Theorebodiesusedby theKafiawemaminersconsistof thick,hard,andcontiguousquartzveins.Thismaterialisbrokenapart,hauledout,andcrushedto“millable”chunks,allbyhand.

In Kafiawema, the family of the present concession holder discovered a gold bearing vein behind

their house, so theentire camp is beingmoved to allowexploitationof thenewvein.Apparently

that person now has legal provincial title to mine aggregate (stones, gravels, sand) at the site.

Seemingly, provincial authorities are using quarrying licenses to also cover gold mines (whose

mandate normally lies with the central government). Moving their camp to expand their mining

operation,meansclear-cuttinganotherpatchofclimaxequatorialrainforest.

Photo16:Openpitminescanonlyoperateas longasthewater ispumpedoutandpitwallsarestabilized.This imposes

limitsonthedepthofexploitation,andexpandstheareathatmustbeclearedtosustainthesamenumberofminersover

time.Pitsaredugwithhandtoolsandwaterisremovedwithgaspoweredpumps


26

Kafiawemamineralprocessing

TheminersofKafiawemauseadrymillwith steelballs.Themill andballs arevisiblyworndown.

Thereisnocontroloverthemillspeed,thoughtheycontrolthesizeofmilloutputbyusinga~1mm

sieve.Thisisaroughlyappropriatediscriminatingsizeformostprimaryconcentrationapplications.

Nevertheless,theprecisecontrolsizecanonlybeassessedbyformalgranulometricanalysis.

This is a relatively simple analysis that would normally be conducted during the baseline and

planning stageof aproject. Insteadof returningoversizematerial to themill for further grinding,

miners insteadhand itover to themanual stampmillers.Thesemillersuseametal variantof the

traditionalmortarandpestlemethodforhandpoundingmaniocrootflour.

In Kafiawema, the drymill discharge ismixedwithwater to create a slurry that is fed to a sluice

throughawashbasinwith1-2 cmholes cut into thebottom.This creates a somewhat controlled

flowofmaterialontothesluice.Fortunately,theydonotfirstimpregnatethecarpetswithmercury,

sothisfirststepisnotwholeoreamalgamation.

Photo17:Thisolddieselmotormustregularlybefedfreshoiltokeepitrunning.Thecapacityofminerstoimprovisethis

roughbutfunctionalmilldrivesystem,demonstratestheircapacitytolearntooperatebestpracticesequipment

Next,theresultingamalgamisburnedoveracharcoalfirewithgreenleavescoveringtheamalgam

totrapsomeoftheevaporatedmercury.Mostofthemercuryis lost,butthesmallportioncaught

on the leaves iswashed into thewashbasinandmoremercury is added toabsorb the recovered

mercurymoreeasily.Suboptimalconcentrationtechniqueslikelyresultinverylowgoldrecoveries.

This conclusion is supported by the fact that they can productively rework tailingsmultiple times

(while also, of course, serially reapplying mercury). Miners have no other way to assess their

productivity per unit ore because they do not study the bulk content of primary material (also

knownasthe‘headgrade’).


27

Theassessmentteamwasunabletoobservethegoldproduction fromsluiceconcentrationof the

orethatisgroundintheballmillinKafiawema,norinventorythemercuryusedintheamalgamation

ofthesluiceconcentrateofthatmaterial.Thisisunfortunate,asitisundoubtedlywheremostofthe

gold is produced, yet the expected amounts were not being sold at the PAC project’s legal gold

buyingshop6.Thereseemedtobemistrustandamistakensuspicion that the realpurposeof this

projectwastobuythegoldandprofitfromexport.Thatopinionwasexpressedexplicitlybyminers

atthecommunityconsultationinButuzi,andprojectstaffindicatedthiswasaperceptionsometimes

foundinSomealso.Ononeoccasionwhenthesluicewasnotbeingused,itwasobservedthatitwas

linedwithamoreoptimalsyntheticsluicecarpetforconcentrationoftheprimaryore,insteadofthe

towels andwoollenmats usedbyothers.When asked about it, the PDG said that they got those

carpetsfromaformalminenearby.Itwouldbeveryusefultoinvestigatethisasapossiblesourceof

bettersluicecarpetslocally,orpotentiallyalocaldistributor.

The“owner”oftheore(inKafiawema,thisisthepersonthatdiscoveredthequartzvein)keepsthe

tailingsaspaymentfortherighttomineandalsofordistributingmercurytotheminersforfree.The

mining tails of all the aforementioned processes are reprocessed 6 or more times using more

mercury each time. All theminers are required to sell their gold to the PDG, who probably also

collectsafractionofitasroyaltiesforthe“owner.”

Photo18:Continuousreprocessingofminetailings.Notethelackoforganizedtailingsstorage,oreffluentcontrol

6PAC’sJustGoldModelTradingHouse(Maisond’achatmodèle)actsasthelegalconduitforartisanallyproducedgoldbypurchasingitatcompetitiveprices,packagingit,andsellinggoldtoapartneredlegalexporter(comptoir).


28

Photos19and20:This(left)isprobablyanappropriatemeshsizefortheKafiawemaore(approximately1mm),atleastfor

the primary grind. Granulometric analyses would better indicate the optimal size.Mortar and pestle secondarymilling

(right)

Photo21:InKafiawematheslopeofthissluicecouldbeoptimized,andthetwosluicesectionsshouldbearrangedinLorZ

configurationtoforcetheflowtostopandrestart(mostgoldiscaughtinthefirstmetreofthesluice,makingaturngives

moreeffectivesluicelength


29

Oreextraction

Mortarandpestlegrinding

BallmillGrinding

Amalgamation

Sluice

ManualCrushing

Ore

+1mm

-1mm

Sluiceconcentrate

Burning

Tailings

Amalgam

MercuryVapour

Gold

Orepowder

+0.15mm

-0.15mm

Diagram2:FlowdiagramformineralprocessinginKafiawema.Sieves(symbolizedbydasheddiagonal

lines)areusedtocontrolthesizeofmaterialthatpassestothesluiceandmanualgrinding.Tailingsare

oftenreprocessed5timesormoretoextractfurthergold.


30

2. MercuryassessmentmethodsEvaluationof theamountsand impactsofmercurycontaminationwas throughsite investigations,

communitymeetings,andenvironmentalsampling(sedimentsanddust).Siteinvestigationsincluded

direct measurement of mercury usage and recovery in each stage of mineral processing, which

provides an estimate of the mercury emissions to air and land/water for each gram of gold

recoveredatthesite.

Theassessmentmethodologyincluded:

i) discussions and interviewswith artisanal ‘mining cooperatives’ andminers, gold traders,mining

technical administration and local authorities concerning priority environmental and occupational

safetychallenges;

ii)sitewalkoversurveytoscopeenvironmentalproblems;

iii) inventoryquantifyingmercury inputsand releases,andestimate the ratioofmercury losses to

goldproduced;

iv)collectionofsoil/sedimentanddustsamples.Intotal57sampleswerecollectedfrombothsites

comprising of 29 soil/sediment samples; 26 swab samples and 2 geological samples. All samples

wereshippedtoSpiezLaboratoryinSwitzerlandforanalysis.

In addition in both Butuzi and Some, consultationmeetingswere organizedwith artisanalmining

cooperatives,miners,goldtraders,theMiningMinistryandlocalauthoritiesexplainingthepurpose

of UNEP's environmental assessment, initial findings from the field visits and potential

recommendationsforprogressiveremedialsolutions.Thesemeetingsalsoprovidedanopportunity

to raise awareness and share informationmaterials about the health and environmental risks of

mercuryuseinAGM.

SedimentsamplingSamplingfollowedascloselyaspossibletheProtocolsforEnvironmentalandHealthAssessmentof

Mercury Released by Artisanal and Small-Scale Gold Miners (Veiga and Baker 2004). In general,

samples were collected from least to most contaminated but it was not always feasible to

successfullydefine those zones.Thedisorderlyarrangementofoperations, legacymining impacts,

and upstream operationsmake it difficult to find “pristine” control points. Siteswere chosen for

their potential to accumulatemercury. This included tranquil areas along river banks where fine

sediments accumulate (e.g. on the inside of a meander), amalgamation ponds, and areas of

depositionofminetailingsorsedimentsthathadbeensluicedandamalgamated.

Most samples were of river or amalgamation muds and sands collected from beneath standing

water. Bottle samples were obtained by taking a composite sample of several trowels full of

submergedcreekbedsediment,mostlynearthecreekedgeswherewatereddiedoutanddropped

itsfineload.


31

Photo22:Swabsamplingonmarketstallsnexttowhichopenairamalgamburningtakesplace

Eachsampletooksedimentfrom4-6locationsoftheupper10-20cmofsediment,mixinginroughly

equalamountsofdeeperandsurfacesediment.Eachsamplewasmixedinaplasticbag,andmixing

bagswereonlyusedonce,and the trowelwasgivenawayaftereachmine sitevisited toprevent

fieldsamplingcontamination.Oncevisiblyhomogenizedintoauniformcolour,themixwaspoured

into the sampling jar, taking care to take somematerial from the top, middle, and dregs of the

sample mixing bag. Each sample was labelled with the geographic coordinates, sampling time,

sample number, description of the context in relation to the mine site, river or sediment

characteristics and other aspects of the sampling site. The bottles were then sealed tightly and

packagedintwoplasticbagstightlyclosedbytwistingandtying.

DustsamplingDustsamplingwasdonewearingrubberglovestoavoidcontaminationduringhandling,andmarked

withgeographiccoordinates,samplingtime,andadescriptionofthesitebeingsampled.Sampling

focused on finding sites near amalgam burning locations such as in home residences including

kitchens,goldbuyerandsurroundingmarketsstalls,foodkiosksandtreesnearopenburningspots.

Dust collection siteswere always taken from shady locationswhere solar exposurewould not be

able to evaporate the mercury (such as roof beams, under shelves, under eaves, or walls in

permanentshadeorartificiallighting).


32

Photo23:Sedimentsamplingfromtheamalgamationponds

MercuryinventoryTheButuzisiteisachallengingplacetodoamercuryinventory.Theminersareoftenusingminute

amounts of mercury (<1g) to recover tiny amounts of gold. The masses involved are near the

effective detection limit of the electronic balance used, which is higher than the manufacturers

statedprecisionof0.01gramsduetodifficultiesinfindinglevelgroundandtheeffectsofwindgusts

(whichproducepressuresinthe0.02-5grange).Turbulentwindeddiespeaklaterinthedaywhen

minersareamalgamatingtheirore.

Tocounterthisinstability,aplasticbagfilledwithsandwaspressedtothegroundandthescalewas

thendrivenintothebaginawaythatitwaslevel.Inthefuture,asmallcircularlevelshouldbeused

tomake sure that the scale is evenlybalanced.A clearplastic coverdomeor a glassbowl should

thenbeplacedoverthescaletoblockthewindwhilerecordingthemass.

Theminers also all tend to amalgamateat around the same time, so following theamalgamation

process of oneminer often precludes finding that of others. Tomakemattersmore complicated,

there is typicallyonemanwhosuppliesallof theminerswithmercury,so thatonecannotsimply

trackthechangingweightofthemercurybottletomeasuretheamountofmercuryrecovered.Once

thismangivesmercurytooneminerheproceedstogivemoretoanother,thentogetthemercury

recoveredfromyetanotherminerandperhapstogivemoremercurytoathirdminer.Thismakesit

difficulttotracktherecoveredmercury.Furthermore,dustanddirtgrainssometimesadheretothe

tinymercurydroplet,whichmay add a relatively significantmass to the recoveredmercury given

howsmalltherecoveredmercurydropletcanbe.


33

Photo24and25:Mercuryinventoryusinganelectronicbalancetomeasureratioofmercurylosstogoldproduced

As a result, it is sometimes necessary to use a piece of paper to hold the recoveredmercury or

amalgam. Theminer,however,oftenwetsordropsparticlesoforeormudontothispaperwhen

transferringthemercurytothepaper.Thesolutiontothisistotarethebalancewiththepaperonit,

andthentomeasure theemptypaperafterwardtoseethechange inmassduetocontamination

duringmercury/amalgamtransfer.

Unfortunately,ahigherprecisionbalancewouldhaveevenhigherrequirementsforlevellingandis

likelynotpracticalinthiskindoffieldsetting.Also,mercuryisusedandstoredunderwater,soitis

importanttoseparatethewaterbeforemeasuringthemercury.Asoftplasticspongecanbepressed

tothesurfaceofthemercury,whichwilldrawupthewaterbutnotmercury.


34

3. AnalysisresultsMercurywasfoundineverycontextinwhichitwasexpected,thoughconcentrationsarerelativelylowcomparedwithartisanalminingcommunitiesinothercountries.Thisreflectsthefactthatlocalminersdonotburnamalgamindoors,norusemercuryfarinexcessoftheamountofgoldtheycanrecover. Artisanal mining in eastern Congo creates a vast, dispersed, low grade web of mercurycontamination and transformationhot spots. This can impact the long termhealth of ecosystemsand people across broad swathes of poor, rural areas.Mercury vapour exposure is amore acuteproblemforminersthemselvesandthoselivingandworkingnearminingrelatedmercuryemissionsin mining camps, villages, and cities. All values are measured in mg/kg (equivalent to ppm). Fullanalytical results are given in the appendices. Normal levels in soil range from 0.05 to 0.08 µg/g(Veiga et al. 2004). As the DRC has not established standards formercury in soil and sediments,samplesareevaluatedagainstCanadianandSwissguidelinevalues.

Table2:SoilandsedimentqualityguidelinevaluesinCanadaandSwitzerland(mg/kg)

Canada Switzerland

Mercury Arsenic Mercury Arsenic

Soil 6.6 12 0.5 Novalueinlegislation

forsoilprotection

Freshwater

sediment

0.17 5.9 0.5 Novalueinlegislation

forsoilprotection

Map5:Kafiawemaminecampdetailshowingmercurysamplingsites.GoogleEarthimagery


35

Map6:Butuzimercurysamplinglocations.GoogleEarthimagery

Map7and8:Bukavucitycentre,andKatudumarket(Kaziba)samplinglocations.GoogleEarthimagery


36

Map9:OverviewmapofMambasa/Kafiawema/Somearea.GoogleEarthimagery


37

Map10:Mbembesemercurysamplelocationmap,showingoldunfilledalluvialminingpitsontheright.GoogleEarth

imagery


38

SedimentanalysisBackgroundsoilconcentrations(0.018–0.34mg/kg)intheKivuregionareelevatedbutwellbelow

international standards. Sediments in the area of the Luzinzi river (0.064 – 0.091 mg/kg) where

Butuziandotherminersdischargeandhundredsofartisanalalluvialminerspracticeamalgamation,

exceed guideline levels downstream of Kaziba town. Natural arsenic released by oxidation of

sulphideminerals inminetailingshas increasedsubstantially.Arsenicconcentrations10or20 fold

above recommended standards in Luzinzi River sedimentswere detected, thus posing a potential

risktopeopledownstream.TheamalgamationpondsattheButuziminesitearerichinmercuryup

to3.6mg/kg)andarsenic (14000mg/kg),whichwillperiodicallyoverflowanddischargedownthe

slopeintotheLuzinzi.

Background concentrations ofmercury in soil, sediment, and rock are low in the Some area and

upstreamof theKafiawemacamp (~0.1mg/kg).Mercury levelsare significantlyenriched inwaste

dumps(upto6mg/kg)andsedimentsimmediatelydownstreamofmines(upto5mg/kg).Orefrom

Kafiawemacontainssignificantamountsofarsenic(upto310mg/kg),whichisliberatedovertimeby

oxidation in waste piles. Each mine site emits tailings and practices amalgamation in or near

freshwater bodies that are slow or stagnant and rich in dissolved organic compounds, thus

promotingthegenerationofreactive,mobile,andbioavailablemercurycomplexes.

DustanalysisSwab sampling shows that mercury concentrations in Katudu market (near the town of Kaziba,

wheremuchoftheButuzigoldissold)increase(from0.2to1.7mg/kg)asoneapproachesthegold

buyerscornerwhereamalgamisopenlyburned,despitethefactthatminersburntheiramalgamat

themine site first. Evidence of dust contamination from open air burning suggests that airborne

mercury concentrations in the market are frequently far in excess of WHO health guidelines (1

microgrampercubicmetre,WHO2003),and is thereforeasignificantrisk forpeopleworkingand

shoppingthere.Nearbystreamsedimentalsocontainmercury (0.3mg/kg), thoughconcentrations

do not exceed guidelines. In Bukavu, swabs of walls near gold shop chimneys (1.2 – 7 mg/kg)

indicatethatspongegoldsoldtherecontainsignificantamountsofmercurywhichcouldbeputting

thousandsofurbanresidentsatrisk.Dustswabstakeninhutswheremercuryisstoredorburned,

andontreesnearamalgamburnsites,werealsofoundtocontainsignificantmercurycontent(2–

1200mg/kg). In hopes of recoveringmercury,miners place green leaves on top of the amalgam

duringburning,andasampleoftheseleavescontainedmorethan2%mercurybydryweight.Tens

ofgramsof leavesascontaminatedasthisareproducedanddiscardedhaphazardly inandaround

miningsitesonadailybasisacrosseasternDRC.

MercuryinventoryAnadequatenumber(5)ofmercurymassbalanceswereundertakenatbothsitestogiveageneral

ideaofmercuryuseandreleaserates,andlocalPACstaffweretrainedtorepeatthemregularly.This

will build up a longer term time series of mercury usage that offers amore reliable estimate of

mercuryreleasesandalsotrackreductionsduetochangesinpractices.Onaverage,minersatboth

sites released more than 4 grams of mercury to the environment for each gram of gold they

produced.One thirdof thatmercurywas emitted as vapour, and the rest is released to landand

water.


39

Table3:Butuzimercuryinventory(ingrams)

Butuzi Average

Mercuryapplied 0.31 0.72 0.84 0.12 0.50

Amalgam 0.06 0.32 0.14 0.20 0.47 0.31 0.25

Recoveredmercury 0.10 0.10 0.68 0.29

Goldsponge 0.03 0.08 0.11 0.25 0.05 0.10

Mercurylost 0.24 0.94 0.22 0.09 0.22 0.38 0.35

Mercury:Gold 8.00 2.75 0.82 0.88 7.60 4.01

Table4:Somemercuryinventory(ingrams)

Kafiawema Somesite2 Average

Mercuryapplied 6.01 5.88 5.3 12.13 5.47 5.06

Amalgam 0.13 1.22 0.68 0.75 1.2 0.58

Recoveredmercury 5.62 4.04 4.76 11.17 2.5 4.73

goldsponge 0.08 0.56 0.29 0.36 0.46 0.25

Mercurylost 0.44 2.5 0.93 1.35 3.71 1.11

Mercury:Gold 5.50 4.46 3.21 3.75 8.07 4.25


40

4. MercuryreductionThefollowingrecommendationsapplytobothAGMsites.Mills7andretortsshouldbeplacedatleast

500metresdownwindofhabitation,foodpreparation,andallnon-miningactivities;andasfaraway

aspossiblefromnaturalbodiesofwater.Themillsiteshouldhaveasluicingpondintowhichsluice

tailingsarecollected,whichdrains intoaseriesofsedimentationpondsthatprogressivelyremove

fine sediments before discharging. A pond approximately ~5 m2 in area should be sufficient, as

tailingsshouldbeperiodicallydugoutandreprocessedorstored.

Themill siteshouldalsohavea longtermtailingsstoragesite that ishigher thanandat least500

metresfromwaterbodies,abovethewatertable,dugoutofironrichclaysoil.Thelocationshould

also be marked with warning signs indicating the presence of mercury and toxic hazard. This

excavation should be lined by iron rich (red) fine clay or a geomembrane to reduce percolation.

Decommissionedmine pits can also be used for tailings storage. Themill site also needs a small

concrete lined amalgamation pond, perhaps with low brick walls to increase the convenience of

panninginit.Onlytheverylaststepofamalgamatingthepannedconcentrateshouldbedoneinthis

pond,thusisolatingallofthepotentialmercuryreleases.

There should be a proper retort and heat source on site, preferably a gas torch rather than the

traditionalcharcoalmethod(thelatterdoesnotproducesufficientheat).Eventuallythereshouldbe

asufficientlyhightemperaturetorchtodirectlysmeltthegolddirectlyinsteadofusingmercury.The

retortshouldbestoredfar fromhousesandrestaurants,andnever inornearpeoples’houses.As

themill, tailings and amalgamation pond are all located right at the edge of themine camp, the

retortiscurrentlykeptrightoutsideminers’residenceswhichisdangerouspractice.

DustexposurereductionThedrymillingmethodsusedatKafiawemapresentasignificantdustcontaminationproblemthat

canproduceshorttermandlongtermhealthproblems.Thereforethemillshouldberelocatedtoa

site far downwind of the mine camp. It would also be beneficial to determine typical prevailing

windsfortheareatoensurethatthemercuryanddustcontaminationisplaceddownwindfromthe

village.

ImprovingmineralprocessingtoeliminatemercuryuseEliminatingmercurycontaminationandincreasinggoldproductionarecompatiblegoals;infactthe

latterisessentialifminersaretomaketheswitchtobetterpractices.Mercuryisthefastest,easiest,

andmostreliablemeanstoseparategoldfromtheore.Bycontrast,gravityconcentration ismore

time consuming, fastidious, and inevitably loses gold in direct proportion to the degree of

concentrationachieved.Tomakeahighgradeconcentrate(gradeistheconcentrationofgoldinthe

concentrate) incurs greater gold losses than ifonewere satisfiedwitha lowergrade concentrate.

Therefore, it is important to reform the entire production chain to improve gold recovery by

leveraging various inefficiencies that are a product of the rudimentary methods used by miners.

Increasing gold production by closing efficiency leaks can offset the losses incurred by generating

higher grade concentrates. This consists of improving milling efficiency (in the case of hard rock

7NotemillingisnotpracticedattheButuzisite,butthismayberelevantinthefutureasminingactivitiesadvance.


41

mining)toensurethatgoldparticlesareoptimallyliberatedfromtheore(mechanicallybrokenaway

fromquartzandotherminerals)whilenotgrindingtheseparticlesintoshapesandsizesthatarenot

amenabletogravityconcentration.Furthermore,lossesfromgravityconcentrationcanbereduced

by optimizingmethods according to the nature of the gold itself (size distributions and shapes of

goldgrains)andbydesigningachainofconcentrationtoolsthatupgradeconcentratessequentially

andcapturegoldlossesbyrecyclingandupgradingtailingsofearlierlinksinthechain.

This impliesmorecomplex systemsshouldbe laidout soas to flowconcentratesand tailings into

subsequentstepsattheappropriateandconsistentfeedrates.Onceestablished,thesesystemsalso

require less labour, which can free miners to spend more time actually extracting ore from the

ground.Asaconsequenceitalsoestablishesmoretechnicaltasksthatmustbemanagedbypeople

with a higher degree of training, thereby diversifying the labour market within camps and

encouraging participation in training and skills upgrading. This reformedmineral processing chain

canbebuiltlinkbylink,beginningwiththesimplestmethodsrequiringtheleastcapitalexpenditure

and training. Ideally, the gains in each step canhelp offset the cost of adding further links in the

chain.

Analysis of gold concentrations in their primary ore should be a priority for this project, as it can

show that gold is being lost and that alternatives aremore efficient. It is important to note that

precisedeterminationsofefficiencyaredifficulttoobtainexceptovermanyrepeatsthroughalong

operational period. This averages out variability due to changes in ore grades and statistical

difficulties such as the ‘nugget effect’ that plagues gold ores in particular. Nevertheless, initial

mineralogicalanalysescanstillprovidesufficientevidenceofpotentialefficiencygainstomakeand

justifyfinancialmodelsofprocessingplantsthatsuitagivenore.

Inparticular,thetargetmilledgrainsizeshouldbedeterminedbylaboratoryanalysis,andoversize

materialsscreenedoutandreintroducedtothemillormilledinasecondarymill.Thiswillproduce

the optimal gold liberation without over grinding. Instead of first grinding the oversize material

enoughtopasstheirprimary(~1mm)sieve,Kafiawemaminersgrindtheoversizematerialtoamuch

finermeshsize(~<0.1mm;probablytoofine).Thisresultsingoldlossesduetothefactthatthegold

ishammeredtooflatforefficientsluiceconcentrationandtoofineforamalgamation.Theyusethe

samemortarandpestlestampmillingtechniqueusedtopreparefood,andsievetheresultwitha

very finemesh to separate the coarse grains for further grinding. Itwould be best tomill all ore

particlesto~1mm(ortotheoptimalsizeoncethissizehasbeendetermined),andthenreprocess

tailingsofthatprocesstoafinersizeoncethegoldhasbeenremoved.Theover-milledmaterial is

also treated using whole ore amalgamation leading to maximum mercury contamination for

minimumgains.Optimizationshouldbedonethroughaseriesofcontrolledexperimentswithlarge

homogenous split ore samples that will enable miners/experimenters to directly compare gold

recoveryofdifferenttrials.Theremayalsobewaystousetheartisanal“assay”thatartisanalminers

are familiar with (in which they hand grind with a hammer and concentrate in a small pan to

evaluate the gold content of the sample) as third party evaluation of the grade of resulting

concentratesormilloutput.

Minersdonotcurrentlyusemagnetstoseparateoutthemagneticcomponentoftheirconcentrate,

andsometimesgrainsofblacksandcouldbeobservedadheringtotheiramalgam.Stirringastrong

magnetcovered inclotharoundinthepanningconcentratecaneliminatea lotoftheblacksands,

thusupgradingtheconcentrateandimprovingamalgamation.Furthermore,thisisanimportantstep


42

inpreparingtheconcentratefordirectsmelting(meltingtheconcentratetoseparatethegoldinthe

formofingots).

In summary,mineral processing should be improved by the following actions in each step of the

processingchain:

Crushing: strive for auniformcrush size so thatmillingproceedsuniformly.Amechanical crusher

wouldfreeuplabourforotheractivities.

Milling:Optimizemillspeed,loading(includingratioofgrindingmedium,ore,andwater),useawet

mill,replacewornoutgrindingmedium(steelballs),andoptimizegrindtime.

Concentration: Use proper sluice carpets instead of towels and blankets, use shaker tables, bluebowl concentrators, and panning to upgrade sluice concentrates to the point where they can bedirectlysmelted.

ActionplanformercuryreductionBasedontheabovediscussiononmercuryreduction,astructuredplanofactionisproposedforthe

twoAGMsites.Theplanstartswithchangesrequiringtheleastinvestmentandtraining,followedby

progressivelycostlierandmorecomplexchangesthatcanbeaddedwhentheminersarereadyto

take thenext step.Consultationswithminersand local communitieswill need tobeorganized to

obtaintheirviewsandbuy-inontheproposedplan.(Seealsoflowdiagramformineralprocessingof

thetwosites).

Immediateaction:

1. Promote use of retorts or fume hoods to recover mercury and reduce exposure and






Mediumtermobjectives:










Longtermgoals:





43



Photo26:SedimentsamplecollectionfromtheLuzinziriverbedwherealluvialminingistakingplace


44

5. Butuzi:SlopestabilizationandriskmitigationSlopestabilityisacomplexproblemandimminentthreat.Largecracksarevisibleinadjacentslopes

to theeastof themainButuzi site indicating that a largedestructive landslide is inevitable. Slope

failurewilldestroyseveralamalgamationponds,mineentrances,andshacksbelowit,andendanger

the lives ofminersworking in the river below. Thiswill likely occur during a heavy and sustained

periodofrainfall,oraftershortburstsofconcentratedrainfall;ahighfrequencyeventinButuziwith

a relatively lowchanceof triggering foranygivenevent.The landslidemayalsobe initiatedbyan

earthquake; near certainty in the unpredictable and low frequency event of large earthquakes

(magnitude4andabove).

Map 11: Massive slope failure induced by artisanal mining at the Butuzi site underscores the risk of a

‘catastrophic’landslidedammingtheLuzinziRiverandcausingdevastatingdownstreamflooding.GoogleEarth

imagery

Miners tend already to avoidworking the site in heavy rains and this should be encouraged and

reinforced. In the event of a massive landslide, it could dam the river below, potentially to a

significantheightbecausethevalleyissteepsidedandtheriverhasonlyanarrowspaceinwhichto

flow.Thetemporarydamthuscreatedwillbuildalargereservoirbehindit,whichmaynecessitate

evacuationof low-lyingareasalongtheLuzinziRiverwherea largeportionofagriculture inKaziba

Chefferietakesplaceandwheresomepeoplehavebuiltpermanenthomes.Itmaytakemonthsfor

thedamtobreach,andthecatastrophicoutwashfloodcouldkillhundredsandeventhousandsof

people if they are not evacuated. A significant area of agricultural fieldsmay bewashed away or

coveredwithdebris,renderingcultivationimpracticalwithoutsignificanteffortstoremoveit.

LuzinziRiver


45

Photo27:Cracksvisible inmountainslopes intheupper leftandrightofthispictureareanalarmingsignalof imminent

collapse

Actualmeaningfulreinforcementoftheminesiteusingconventionalengineeringstructuresislikely

prohibitively costly with low probability of success. The rocks are shattered and altered both by

tropicalweatheringandbythegeothermalalterationprocessesthatemplacedthegold inthefirst

place.Insidetheminingtunnels,therockdoesnotappeartobecomeanymoresolid,butratheris

poorly consolidated and deeplyweathered. Not only does this present an extreme risk of tunnel

collapseevenwithoutaseismicevent,butalsoitmeansthattheslopeinstabilityrunstoodeepand

is toopervasive touse rockboltingorotherheavy industrial techniques. In any case, itwouldbe

impracticaltogettheequipmentforsuchengineeringtothisremotesite.

Under these circumstances, disasters are inevitable. Indeed, a landslide last yearwas reported to

have caused several deaths. In fact, injuries or deaths are likely simply from rockfall fromminers

operatingaboveoneanother,asnoonewearshelmetsandthere isnomineplanning.Apparently

theminersaredisposedtoleavingsomeareasalonetodecreaseriskstootherminers,ortoworkin

rotating shifts. In practice this may be hard to agree on or enforce, but should nevertheless be

pursuedaspartofmineplanning.

Bioengineering techniques provide a relatively easy and low-cost option for slope stabilization.

Vetivergrass(Chrysopogonzizanioides)isagoodcandidateforhelpingstabilizethesteepslopesof

the Butuzi mine site. Although classified as a grass, vetiver plants used for land stabilization

purposesbehavemore likefast-growingtreesorshrubs.Theirextensiveandfinelystructuredroot

system can extend 2-3 meters in the first year reaching up to eight meters. While some of the

exposedslopeisbarerockorlooseshaleslopes,therocklayersaregenerallydeeplyweatheredwith

many weak spots through which the vetiver’s fast-growing and deep roots can penetrate. By

plantingthevetiverplantscloselytogetheratcritical locations(e.g.alongthetopofthemineface


46

and the footpath), dense hedges can be created to structurally reinforce the slope and also slow

down and spread thewater flow to help control erosion. Finally, it should be noted that there is

goodexpertise in theDRCon vetiver use for slope stabilization anderosion controlwhich canbe

utilizedbytheproject.

Photo28:WatercollectionpitslocatedabovetheButuziminesiteincreaseinfiltrationintolandslideinitiationzones

Controllingdrainage is theothermainmethodwhichcanbeusedtomitigateslope instabilityand

landslidesattheminingsite.Buildingalargeconcretelinedcisternandraincollectionsystemonthe

hill above themine site couldprovide theminerswith a larger andmore reliable sourceof clean

water that would not impact stability and could be fed to the miners using gravity. Ideally, the

cistern shouldbe locatedon theeast faceof theButuzimine summit, at theelevationwhere the

easternapproachtraildescendsintotheminesite,butfarthersouthsothatifitfailsitdrainsaway

fromtheminesite.Fixedtubesshouldconnectthecisterntothetopoftheminesitewherewater

managerscanconnecttoanozzlemanifoldtowhichtheycanconnecttheirhoses.Hand-dugditches

tochannelwatertosluicesshouldbestoppedandreplacedwithlargehoses.

Existingwaterpitsshouldbefilledin,perhapswithmaterialdugouttocreatethenewcistern,and

planted with vetiver and ideally some trees also. If miners, however, are reluctant to stop using

existing water pits, than these should be lined and reinforced with sand bags and terraces and

plantedwithvetiver.Ineithercase,vetiverhedgesandtreesshouldbeplantedalongthetopofthe

entire mine face to soak up and armour against high surface flows of water, and reduce rapid

infiltrationduringstorms.


47

Photo29:Hosesaresometimesusedtodeliverwaterforon-siteprocessoperations;amajorchallengeandcostforminers

Apart fromthepresentwatersupply, themainButuzi siteappears tohavenomajorcracks in the

earthwhereunstablelandundertensionispullingawayfromthemountainside.Tarpsmaybelaid

over thecracks in locationswhere this is thecase,with the topmarginof the tarpdugdiagonally

downintothesurfacesoiltocaptureallsurfacerunoffandephemeralstreams.Thebottommargin

ofthetarpsshouldbelinedwithplasticwaterchannelsthatfeedrainwaterstoragecisternslocated

asfarfromthedisturbedslopeasispractical.

Theabovementionedtensioncrackrunoffdiversionwatersystemsandcisternscouldfeedsluicing

and amalgamation stations far away from the disturbed site, thereby reducing the exposure of

miners to slope failurewithoutmaking themhaul their ore uphill. Care should be taken tomake

properly armoured,well drained trails to these safeareas toavoid creatingnew stabilityhazards.

Vetiver and trees should be planted along the trails and around the processing areas to reduce

erosion.


48

Photo30and31:Arecentlandslidefollowingheavyrainfalldestroyedseveralminingtunnels


49

Another significant geotechnical risk is that the unplanned, unmapped, and haphazard tunnelling

further weakens the slope overall, and presents serious risk of tunnel collapses from miners

burrowing under each other without knowing it. Even without such collapses, the inadequate

buttressingofundergroundworkings (madeworseby thepoorly consolidatedandheterogeneous

hostrock)hasalreadycausedseveralcollapses.Minerspointedoutseveralmineopeningsthathad

collapsed.At leasttwotunnelcollapseshaveclosedoffundergroundaccessessincere-initiationof

operations in 2010. Relatively straightforward measurement and mapping of the existing mine

tunnelscanbemadeusingahipchainandcompass.Bytrackingexcavationprogress,minerscanbe

alerted when they are in danger of compromising each other’s workings.

Photo32:Handdugchannelsareanimportantcauseofslopeinstability


50

6. Some:Forestdegradationandbiodiversityloss

BiodiversitylossIn Some, deforestation and biodiversity loss is potentially even a greater environmental concern

thanmercurycontamination.Sinceartisanalmining is takingplace inaprimarytropical rainforest,

with its wealth of biotopes and numerous species, it is important to be mindful of its wider

environmentalfootprint.TheIturiForestisrenownedforitsoutstandinglydiversefloraandfauna.It

is the prime habitat of theOkapi, an endemic forest giraffe that is a national DRC symbol.Other

spectacular and endemic forest species include the bongo, dwarf antelope, forest buffalo, forest

elephant,thegiantforesthog,theaquaticfishinggenet,andchimpanzees.Itisalsoanexceptional

areaforbirds,includingtheendemicCongoPeafowl.

Asnotedearlier,thelocationoftheSomesitewithinthedesignated50-kmbufferzoneoftheOkapi

Wildlife Reserve poses a specific risk. In effect, the Reserve is potentially at the receiving end of

mercurypollution loadsas it liesdownstreamoftheSomesitealongthe IturiRiver.Thiscreatesa

risk of mercury bioaccumulation in the tropical food chain from bacteria to carnivorous fish to

animalsfeedingonfishandfurthertohumans.

Primaryimpactsofartisanalminingonwildlifestemfromhabitatlosscausedbyvegetationremoval

and landdegradation.The resultinghabitat fragmentation is likely todisturb thehome rangeand

migrationpathsofwildlife.Theriskofhuman-wildlifeconflictandthespreadofzoonoticdiseases

increases as animals are displaced and search for new habitats. Miners may also become

opportunistically or deliberately involved in wildlife poaching. In addition, mining camps, which

overtimemaygrowintovillages,arelikelytobecomecentresforbushmeatconsumptionandtrade.

Ivorytrade,whichhasexplodedacrossnorth-easternDRCinthelastseveralyears,isanothermajor

issueofconcern.

The high conservation value of the Ituri rainforest creates a challenging context for the roaming

approachtomininginSome.ThegeneraltrendofencroachingartisanalminingbothwithintheORW

anditsbufferzonefurthercomplicatesthesituation.Moreover,thispressureislikelytocontinueto

increasewiththesteadyimmigrationofpeoplefromthedenselypopulatedhighlands.

DeforestationIn Kafiawema,miners found a gold vein under their mining camp and so they destroyed several

hectares of forest to establish a new camp.Miners in effect clear-cut the forest,whereby all the

treesinthecampareaarecut-downwithonlyasmallnumberoftreesleftstanding.Environmental

impacts of clear-cutting range from soil erosion and increased riskof flooding todepletionof the

carbon sink andmodification of themicro-climate. Remote sensing could be used tomonitor the

evolutionandextentofdeforestationcausedbyartisanalmining.


51

Map12and13:SatelliteimagerytakeninApril2013(above)andAugust2015(below)revealexpansionofdeforestation

patchesandalluvialminingpitsintheMbembeseareaoftheSomeminingsite.GoogleEarthimagery

UndisturbedForest

Mbembeseminecamp

Alluvialminingpits

UndisturbedForest

Alluvialminingpits

DeforestationMbembeseminecamp


52

Photo33:Ontheedgeoftherapidlyshrinkingclimaxequatorialrainforest,minersslashandburnvaluabletimber

MitigationoflanddegradationSerious legacyproblems frompreviousworkingswere alsoobserved. In the Somearea, extensive

alluvial pits arrayed in a disorderly manner and never refilled dot the landscape. This creates

stagnant ponds that prevent proper forest regeneration and vastly increase the potential for

methylation (and therefore mobilization of more toxic and bioavailable mercury compounds) of

legacy mercury contamination contained within the tailings and pond bottoms left by previous

mining activities and augmented by present mining activities that discharge into these old mine

holes.Whileaddressingthesecomplexenvironmentalissuesisbeyondtheimmediatescopeofthe

‘JustGold’project,asustainableminingmodelneedstoprogressivelydesignawin-winsolutionthat

balances the interestsofminerswith thatof theReserveand its surrounding landscape. It should

also be underlined that in order for remedial measures to gain the buy-in of miners, and be

disseminatedandreplicatedtoothersites,technicalsolutionsproposedmustbeaffordable.

Mitigating measures to address the above mentioned impacts may be promoted through the

project’s agreement with miners on provision of technical assistance. This may include specific

provisionspromoting:

1. Systematic reclamation and backfilling of mined-out pits. Holes containing mercury

contaminated tailings should be covered with one meter of clay or laterite, compacted,

coveredwithsoil,andre-vegetated.

2. Restrictions on wildlife hunting and trade including respecting national regulations on

endangeredspeciesandthehuntingseason.

3. Selectiveapproachtofellingtreesinformedbytechnicalassistanceonmining exploration

andplanning.


53

To incentivize miners to gradually implement environmental improvements, the project may

considerofferingan‘environmentalpremium’and/oradditionaltechnicalassistanceforminerswho

demonstrateenvironmentalcompliance.Market-basedinitiativesfor‘greengold’(e.g.Fairtradeand

Fairminedgold)maybeleveragedtohelptheprojectinincentivizingbetterenvironmentalpractice.

Such initiatives should be undertaken in a gender inclusive perspective. Targeted outreach and

follow-upwillalsobenecessarytodriveminerstowardenvironmentalimprovementsovertheshort

andmediumterm.

Photo34:HavingpaidanillegalloggertofellhundredsofcubicmetresofAfricanMahoganyandothertropicalhardwood

onhisdayoff,minerspurchasesawnwoodandcarryitintocamp


54

7. MiningexplorationForeverymillion-ouncescaledeposit thatwould interest industrialminers thereare thousandsof

surfacedepositsthatcouldfeedgenerationsofartisanalminers.Artisanalminersgenerallyexplore

atthemarginsofexistingdepositsbymakingtestexcavationsandpanningsmallsamplestocheck

forvisiblegold.Manyjustworkforotherswhohavesomeownershipormanagementofthemines.

Mappingouttherealsocialstructuresandorganizationalarrangement isbeyondthescopeofthis

environmentalassessment,but isan important topic thatneeds tobestudied,especiallyas ithas

importantimplicationsontheadoptionofimprovedenvironmentalpracticesbyminers.

Minerswill expand favourable excavations into amine and continue as long as they canmeet or

exceed theirdailyneeds. Theydo thiseitherbyopeningupmore surfacedeposits laterally,orby

chasingdepositsdeeperintothegroundbydiggingtunnels.Theirapproachshouldbeexpandedto

includeexploration,mineclosure,healthandsafety,andmineplanning.Tunnelsarecreatedwithout

a thought to ventilation, stabilization, or the location and extent of other tunnels. This leads to

collapsesoftunnelsthatundermineeachother,inadditiontonaturalones.Speleologicalsurveysof

tunnelsareneededfortrackingandavoidanceofman-madecave-ins.

Photo35:Openingnewgroundforexploitation.Technicalguidanceisneededtohelpminerslocategolddepositsandplan

extractionaccordingly

Minersalreadyhave limitedsystemstoorganize forwardthinkingworkthatdoesnot immediately

produce gold. For example, a large number of workers in Butuzi were observed excavating

overburden in search of new gold ore at the margins of existing workings. Furthermore, in

Kafiawemamanyminerswereworkingtoopennewexcavationsandbuttressdeeperpartsoftheir

mine pit. This system for organizing “unproductive” but necessary roles within themining group


55

should be studied for possible support of expanded “unproductive” roles such as exploration and

waste management. Not only are these essential roles in a more orderly, formal, and

environmentallyresponsibleminingoperationbuttheycanalsocounterbalancetheunemployment

that can be generated when replacing inefficient methods with best practice mineral processing

methods.

Photo36:SteepslopemininginButuziposesseriousrisksoflandslideswhichcoulddamtheLuzinziRiverbelow


56

8. OccupationalsafetyWorkingintheButuziandSomeminesitesisahighriskanddangerousactivity.Thisisduetoboth

the site terrain (steepmountain slopes, inadequately buttressed tunnels, deepopenpits, hazards

fromworkinginareacoveredwithfelledtrees),riskofmercurypoisoning,andtheverylimiteduse

ofpersonalprotectiveequipment(PPE)byminers.

Photos37:Steep-slopeminingisdangerouswork

Somebasicmeasuresandpracticescanbeintroducedtoimproveworkingconditionswhichshould

beaccompaniedwithprovisionofappropriatetrainingandinformation.Thisincludes:

1. Usingwheelbarrowsorhandcartstohaultheorebags.

2. Wearingstronggloves(preferablynon-latex)whenhandlingmercury,particularlyduringtailing

reprocessing and in whole ore amalgamation. Gloves should normally be used during all wet

work.

3. Allmercuryshouldbestored indurablecontainers thatareclearly labelledasmercuryandas

toxic,andcoveredwithalayerofwatertopreventthemercuryfromevaporating.4. Puttingondisposabledustmasksandeyeprotectionwhendrymillingrockore.Notedustmasks

needtoberegularlyreplaced.

5. Safetyhelmetsshouldnormallybewornatalltimeinthesite,particularlyinButuzi.

6. Providesimplefirstaidequipmentattheminesite.

7. Encourageminerstostorecleandrinkingwaterandconstructasanitarypit latrineatthemine

site.


57

Photo38:Steep-slopeminingisdangerouswork.Safetyhelmetsanduseofprotectivepersonalequipmentarelacking


58

9. Barrierstoformalityandtheartisanalminingpovertytrap

Artisanalminerstendnottoexploreforgolddeposits,butinsteadcongregateinsiteswheregoldis

already being produced. Globally speaking,most known gold deposits are usually orphanedmine

sites, many of which were established in colonial times and are largely exhausted, though the

remnant deposits are still suitable to those seeking a way out of poverty. Some deposits were

singledoutfor internationalminingexplorationprogramswhosedevelopmentdidnotproceedfor

lackofeconomic justificationorsocial license,andthatwerethentakenoverbyminersthatwere

alerted to the site by the presence of these prospectors. Perhaps more commonly, however,

industrialminingprospectorsidentifypotentialmineraltargetsbyseekingoutartisanalminingsites

andusingtheirexcavationsandtunnelstoassesspotentialorebodies.Prospectorsthenproceedto

purchasethemineralrightsfromthegovernmentorthelegaltitleholder,withouttheknowledgeof

artisanalminers operating on site. They then pressure the government to use police andmilitary

forcestoenforcetheirtitlerightsbyremovingthe‘invaders.’

Artisanalminersareoftenseekingentirelydifferentdepositsfromthoseofinteresttolargemining

companies,ordifferentpartsofthesamedeposits.Artisanalminerschaseshalloworeinsparseand

well definedhigh grade veins (in hard rockmining) or strata (in alluvialmining) that require little

investment in excavation and mechanization to operate. Large companies use those deposits to

identify and constrain larger, deeper, lower average grade deposits whose mine life production

would justify massive capital expenditures required to exploit the ore for least cost and highest

returns. Often industrial mines remove the “overburden” (overlying ‘waste’ rock that artisanal

minerswouldexploit) inorder toaccess thedeeper industrial scaledepositwitha largeopenpit.

Theytakeafulllifecycleapproachtominingoperations,inwhichthecostofexploration,excavation,

processing, waste storage andmanagement are compared against the net value of the exported

lifetimeproduction.

Government regulations, particularly mineral title licensing rules, are often structured to favour

largeindustrialminingoperations.Obtainingmineraltitleisoftenthefirstobstacleencounteredby

any artisanalmining project aiming for transparency, formalization, or sustainability, and globally

remains a vexatious and difficult barrier. Governments in Peru, Colombia,Mongolia, Guyana, and

Suriname (among others), as well as in the DRC, are actively wrestling with this challenge with

limitedsuccessduetothemisinformationandcompetingintereststhatattendnationallevelpolicy

changeinitiatives.

In well-functioning property rights regimes, mineral titles expire if the title holder makes no

investment inexplorationor exploitation in the site.Where corruptionand loopholes allow, titles

areoftenrenewedorheldindefinitely,andalloftheknownandmostpromisingmineralconcessions

arealreadyclaimed.Thereforeartisanalminersareoften immediately labelled“illegalminers,”or,

where theirprofitsare funnelled toorganizedcrimeandconflictgroups,“criminalminers.”This is

usuallyagrossoversimplification(artisanalandsmallscaleminersinColombiafundcriminalgroups

involuntarilythroughextortion),buttheperceptiondrivesgovernmentassembliestosetunrealistic

legislative barriers to formalization and development of the artisanal mining sector. As a result,


59

minerscannotentertheformaleconomybecausetheycannotobtainlegaltitletothedepositsthey

arealreadyworkingandareunabletonegotiateequitableagreementswithexistingconcessionaires.

Where miners have entered formal contracts with legal concessionaires, their lack of legal

knowledgeorcounselcanproduceabusiveagreementsthatstronglyfavourtheconcessionaireand

establishfeudalarrangementsthatperpetuatedangerous, inefficient,andpollutingpracticeswhile

concentratingexcessivewealthinthehandsofdistanttitleholders.Inthesecases,theyalsotypically

control the mercury supply and therefore may see its reduction as a diminished income.

Furthermore,artisanalminerstypicallylacktheskillsandresourcestoexploreavailableconcessions,

whichareusually so remoteor insecureas tomakeaccessdifficult (and thereforeprovide secure

fundingsourcesforillegalorconflictgroups).

AnartisanalandsmallscalemininglegalloopholeAtransparentgoldsupplyis,bynecessity,alegalone.Legalgoldbydefinitioncomesfromoperators

oflegalconcessions,thereforeformalizationandmineraltitlearethehighestorderpriorityforany

gold transparency program. Congolese provincial governments are currently managing artisanal

mineral titles under a legal loophole: an ingenious reinterpretation of provincial jurisdiction

concerningaggregatemining.

Aggregatesarenon-metallicrockdepositsthatareminedforroadfill,concreteproductionandother

constructionactivities.Thenon-metallicdesignation isusuallyclearlyspecified inaggregatemining

legislations, so the present work-around may not be an intentional regulation. In other ways,

artisanaloperationsarequite similar toaggregatemining,mainly in thesense that theyareoften

mining poorly or un-consolidated materials that are close to the surface with a limited surface

footprint. Therefore aggregate concessions are by definition small, shallow, cheap, and available.

Where they overlap with centrally administered precious metals concessions, there are no real

conflictsofinterestsbecausethelargescaleminersarenotinterestedinthesurfacedepositsother

than as indicators of deeper potential strikes. This differentiation is less clear when considering

alluvialandmediumscaleminingoperators.

Kafiawemaisagoodexamplewhereaggregateconcessionsareusedforsemi-legalpreciousmetals

exploitation.Onememberofthecommunitywasdiggingapitlatrineandfoundagoldbearingvein.

Rather than obtaining a national gold mining concession which would have required travel to

Kinshasawithnocertaintyofbeingable toafford the licensing fee (inpartbecause theminimum

concessionsizemaybeprohibitive,andinpartbecausegoldismorevaluablethanaggregate),the

minersinKafiawemaacquiredanaggregatelicense.Theprojectneedstofurtherinvestigatenational

and provincial mineral concession legislations and para-legal functional agreements under which

minersoperate.

Loopholes that misuse provincial concession rules to circumvent federal ones may be a useful

starting point from which to identify and harmonize the concession rules and establish a clear,

equitable, achievable, and fully legal concession system that is accessible to artisanal miners.

Jurisdictionalconflictcouldberesolvedbydefiningclearrulesforindustrialsectorprospectorsand

developers wishing to operate concessions where small artisanal concessions overlap centrally

administeredconcessions.Oneofthegoalsofartisanalminingdevelopmentorganizationsistofind

a way to establish formal agreements among industrial and artisanal mining groups where their


60

interestsareinpotentialconflict(eventhoughtheyareinterestedindifferentpartsofthedeposit).

Thiskindofharmonizedlegislationcouldbeafirststeptowardthatgoal,thoughitmightalsoreduce

potential international investment because the industrial sector is afraid of the liability issues

concerningpartnershipswiththeinformalsector.Gran-ColombiaGoldinSegovia,Colombia, isone

example of a successful relationship between sectors, and Barrick Gold’s operations in Tanzania

offer a useful case study in the formal sector’s worst scenarios. The latter case is more

representativeofrelationsamongartisanalandindustrialminersglobally.

The insecurity and disorganization of informality preventsminers from investing in exploration or

bestpractices for fearofevictionand seizurebyauthorities. It forcesanopportunistic,migratory,

andmarginal lifestyle onminers and traps them in poverty. This project needs to investigate and

ensurethetruelegalityoftheminingclaimsinwhichtheyareoperating,andusesuccesscasesto

designandpromote the reviewand reformofnationalminingandprovincial regulation so that it

betterandmorerealisticallyservesbothartisanalminersandthestate.


61

10.ConclusionsandrecommendationsThis studyshows thatmercurycanbesignificantly reducedandultimatelyeliminatedby replacing

the need for mercury by artisanal miners processing the ore. The strategy should be based on

prioritizing pre-concentration before amalgamation. Efficient pre-concentration incurs gold losses,

soprocessinginallpartsofthechainneedstobeimprovedincludingfrom:explorationandplanning

ofdepositextractionandmineworkings,choiceandpre-processingofore,efficientgrindingforgold

liberation, aswell as concentration andmonitoring of gold in ore andwaste.Overall, higher gold

recovery can be achieved without mercury. Technology transfer and training is an important

pathwaytowardsthatend,assumingnotonlyformalitybutaccesstoinvestmentandservices.

SummaryofrecommendationsIn addressing themultiple environmental challenges at the twoAGM sites, the recommendations

are sequenced by interventions requiring the least investment and training, followed by

progressivelycostlierandmorecomplexchangesthatcanbeaddedwhentheminersarereadyto

take thenext step.Consultationswithminersand local communitieswill need tobeorganized to

obtaintheirviewsandbuy-inontheproposedinterventions.

Mercuryreduction1. Promote use of retorts or fume hoods to recover mercury and reduce exposure and





















62

Slopestabilization(Butuzi)1. PlantVetivergrass(Chrysopogonzizanioides)tohelpstabilizethesteepslopesoftheButuzi

minesite.

2. Controldrainagetominimizegroundsaturationanderosion.

3. Buildaconcretelinedcisternandraincollectionsystem.

4. Establishsluicingandamalgamationstationsfarawayfromthedisturbedsitetoreducethe

exposureofminerstoslopefailure.

5. Mapandplantunnellingactivities.

Forestandlandprotection(Some/Kafiawema)1. Systematically reclaimand-backfillmined-outpits.Holescontainingmercurycontaminated

tailings,shouldbecoveredwithonemeterofclayorlaterite,compacted,coveredwithsoil,

andre-vegetated.

2. Restrictwildlifehuntingandtrade.

3. Employ a selective approach to felling trees informed by technical assistance on mining

explorationandplanning.

Occupationalsafety1. Usewheelbarrowsorhandcartstohaultheorebags.

2. Wearstronggloves(preferablynon-latex)whenhandlingmercury,particularlyduringtailing

reprocessingandinwholeoreamalgamation.Glovesshouldnormallybeusedduringallwet

work.

3. Allmercuryshouldbestoredindurablecontainersthatareclearlylabelledasmercuryand

astoxic,andcoveredwithalayerofwatertopreventthemercuryfromevaporating.

4. Usedisposabledustmasksandeyeprotectionwhendrymillingrockore.Notedustmasks

needtoberegularlyreplaced.

5. Safetyhelmetsshouldnormallybewornatalltimeinthesite,particularlyinButuzi.

6. Providesimplefirstaidequipmentattheminesite.

7. Encourageminers to store clean drinkingwater and construct a sanitary pit latrine at the

minesite.


63

Appendix1:LaboratoryanalysesinterpretationBottlesamplesAllvaluesmeasuredinmg/kg(equivalenttoppm).CanadasoilqualityguidelineHg=6.6mg/kg;InterimfreshwatersedimentqualityguidelineHg=0.17mg/kg

Samplecode

Description Discussionofchemicalanalysis Hg As Pb Au

B1B Riverwestofmarketwheregold

buyersburnamalgam.1-2mwide,

highbanks,resistantoldchannel

basin.Sampleconsistsoffineblack

sandwithmicaflakes.Sample1-10

cmdepthsedimentunderwater.

MugeraRiver.Brownturbid,

surroundedbyagriculturalfields.

Evidenceofmercury

contamination,asexpectedfrom

beingdownstreamofthegold

buyersinthemarketthatburnHg.

Noothermetalsshowelevated

concentrations.

0.03 6 6.5

B2B BLANKsamplebottledatLuzinzi

riversite(b3b)

Thisblankmaterialwassandfrom

aconstructionsiteinBukavu.No

significantmetalconcentrations

present.

0.008

5

1.

8

7.4

B3B LuzinziRivercompositesample,

bimodalimmaturesediment

braidedriverheavilyreworkedby

AGMwithHg.Compositesampleof

quietwatersedimentsupstream

nearfasterflowchannelinslower

channelinthemiddleof

amalgamationarea.Panningis

doneinfasterflownearby,butalso

lotsupstream.

Hglevelsreflectthefactthatthis

entireriverisanartisanalalluvial

mineatthissiteandforseveral

kilometresupstream.Arseniclevels

are100xguidelines.

0.064 71

0

15

B4B LulinjaRivercompositesample

along3mlengthofquietbendin

river;eddybank.Riverisbright

brownturbidfromupstream

mining.Largeblackshalerocks

makesmallrapidsinthefast

flowingrivermiddle.Sedimentis

finesandandmud,brownwith

blackstrata.

LesscontaminatedthanLuzinzi

River,stillevidenceof

amalgamationupstream.

0.034 8.

9

7.5

B5B Heavilyworkedminerbuildriver

braidchannelling,moderatewater

flow,siltybottom.Composite

sample.

LessHgcontaminationthan

expected,andillustrative

differenceinmercury

concentrationascomparedtothe

duplicate.Mercuryvariabilityfar

exceedsthevariabilityinother

elementsbecauseitisdeposited

anddisturbedinpatchy,haphazard

waysbyhumans,whereasthe

othermetalconcentrationsare

0.068 95

0

19


64

Samplecode


morerelatedtobackgroundand

tailingsdischargewhichproduce

moreuniformsignals.

B6B DuplicateofB5B consistentmetalconcentrationsin

bothsamples,Hgvariesmore0.091 86

0

20

B7B Butuziminesiteamalgamation

pond.Muddyyellowwithstanding

water.Severalminers

amalgamatedinthispondatlast

visit2dayspreviously,suggesting

thisisadailyoccurrence.

Ascontaminatedasanamalgam

pondwouldbe.Arsenicis

surprisinglyhigh.

3.6 44

00

61

B8B Backgroundsample.50cmbelow

surface10cmintoscarpface.

Brownmoistsubsoil(Bhorizon)

withsharprocksandsomeroots.

TopofButuziminesite.

Surprisingamountsofmercuryin

theoreitself.Perhapsminershad

beenburningnearby.Thenext

sample,B9B,isabettermercury

backgroundsample.

0.34 19

00

28

B9B GPSmarksmineentrance.Ore

sampletakenintunnel~40mdeep

accordingtominerwhotookthe

sample,butshaftiscircuitous.

Abetterbackgroundsamplefor

mercurybecauseitcamefrom40m

depth.Arseniclevelsareveryhigh.

0.018 35

00

47 <8

B10B Anotheramalgamationpond

sample

TypicallyhighconcentrationsofHg

andAsforamalgamationpondsat

thissite.Interestingvariability

amongtheponds.

0.95 12

00

0

82

B11B Amalgamationpondsample

(compositemixedinthesamplejar

notplasticbag)

Arsenicishighinallpondsbecause

itisconcentratedwiththesulphide

mineralslikepyrite,thenis

discardedwiththeconcentrate

afteramalgamation.

0.74 14

000

28

B12B Luzinziriver5kmnorthand

downstreamofKaziba

Alarminglyhighermercury

concentrationsdownstreamof

mostoftheminingactivity.Isthere

someverynear?Alsointeresting

howtheHgisconsistentwiththe

concentration2kmfurther

downstream.

0.2 43

0

12

B13B Luzinziriver7kmnorthand

downstreamofKaziba

Curiouslyhighmercurycontent 0.26 39

0

14

BConc Concentratesamplefrombanana

barksluice

VeryinterestingthattheHgfrom

thisconcentrate,whichwasnot

exposedtomercury,isstill15x

higherthanthebulkoreinsample

B9B.ItindicatesthatperhapsB8B

reflectspossiblebackground

concentrationsinpartsofthis

depositafterall.

0.27 57

00

110 10-

40


65

Samplecode


mleaf Leavesthatwereusedtocapture

mercuryduringamalgamburn

Tensofgramsofleavesas

contaminatedasthisareproduced

anddiscardeddailyateachmine

siteinCongo.

2200

0

M-Ore Quartzoresample Quitelowlevelsofallmetalsother

thangold.Thisgoldvaluematches

thelowestvaluesofthegeological

oresampleanalyzedinCanada.

0.008

2

62 8.1 13

M1B Upstreamofminesite.Clearbrown

quiet2mwidecreekveryslow

almoststill,forestedbanks,muddy.

Handsampledcompositebylocal

personundersupervision,samples

from2mlengthofriver,various

waterdepths,surface1-10cm.

PerhapshighHgindicatesprevious

mininginorupstreamofthissite

hascontaminatedthesediment

already,orthebackgroundlevels

arehigh.Theblanksampleseems

tosuggesthighbackground

regionally.

0.16 3.

1

5.4

M2B Compositesampleincreek

immediatelyupstreamofintakefor

minewatersupply.Teacoloured

waterwithmuddybottom4mwide

withlowtomodestflowandold

forestedbanks.Mbelariver.

PerhapshighHgindicatesprevious

mininginorupstreamofthissite

hascontaminatedthesediment

already,orthebackgroundlevels

arehigh.Theblanksampleseems

tosuggesthighbackground

regionally.

0.11 7.

3

21

M3B Compositesampleattheoutflow

oftheminediversionchannel,4-5

mwide,veryslowflowwithmuddy

bottomandshadingforested

banks.Translucentwithmoderate

turbidity.Sabasabacreek.

Hgis10xthebackgroundand

upstreamvalues1.1 37 12

M4B BlanksampleatsiteM3B,material

wasoriginallyfromasunbaked

termitemound1hourawayfrom

MambasaontheroadfromBunia,

farfrommajorsettlements.

Regionalbackgroundmercury

levelsaremodest0.12 1.

8

6.2

M5B Sluicewasteandamalgamation

pondcompositesample

Hgcontaminationasexpected 6.6 13

0

23

M6B CompositesampleatM5Bsite;5-6

timesreworkedtailings(newHg

addedeachtime)

Asaduplicateoftheprevious

sample,thisagainindicateshow

variablemercuryconcentrationsin

hotspotscanbe;allothermetals

arenearlyidenticalin

concentration.

3.3 11

0

16

M7B Upstreamcompositesample

downstreamofclotheswashing

spot,smallmuddycreek.

Welaterfoundoutthatthiswas

originallywheretheydidtheir

sluicingandamalgamation.

5 13

0

20


66

Samplecode


M8B DuplicateofM7B. Asaduplicateoftheprevious

sample,thisagainindicateshow

variablemercuryconcentrationsin

hotspotscanbe.

0.94 54 7.7

M9B Backgroundsampleofmudfrom

3mdepthinminepitwallin

afternoonshade.Mottledgrey

/brown/redmottledtexturewith

quartzpebblesinit.

LocalbackgroundHgdoesseemto

behigh,consistentwithregional

background.

0.16 31

0

17

M10B Amalgamationpondcomposite

sampleMbembeseminesite.

HighHgconsistentwithlongterm

amalgamationuseinthispond.26 18

0

52

M11B Oldalluvialholesfilledinwithdark

stagnantwater,lookslikemarshes

andlakeswithteacolouredwater.

Compositesampleofrecently

reworkedtailings.

Perfectconditionsformethylation:

contaminatedtailingsinstagnant,

anoxicpondswithhighdissolved

organiccontent.

3.6 10

0

20

M12B "Upstream"intheMbutelecreek.

NB.Thisisallonebigun-reclaimed

alluvialminedisasterthatended

about3yearsagoandispartly

revegetated.Thisspothadatrickle

ofcurrent,clearwaterthatistea

colouredwhendeeper.

Clearlythereisgeneralizedlegacy

Hgcontaminationinthisarea,as

indicatedbythedisturbed

landscape.

0.65 7.

7

5.9

M13B Opensandy/gravellypitcomplexin

whichitwashardtofindfines.

Unreclaimedalluvialdisaster.

Clearlythereisgeneralizedlegacy

Hgcontaminationinthisarea.1.1 17 4.1

M14B Roadcutsample~5cmintocutface

and~1mbelowgroundsurface.

Hardpebblysaprolite,red.

AnotherevenmoreoddHg

enrichmentinthesoil.Local

backgroundisquitehigh.

0.41 4.

1

18

M15B Illegiblesampleinformation.

ProbablyafieldduplicateofM13B

Exactlyequalmercurycontent

comparedwithm13b1.1 48 9.5


67

SwabsamplesAllvaluesmeasuredinmicrograms(µg)perwipe.

SampleCode

Description Discussionofchemicalanalysis Hg As Pb

S1B NorthwestcornerofKatudumarket

Kabiza,goldbuyerhut1.5metres

fromburnsite.Swabbedwallunder

roof.

Relativelyhighmercurycontamination,as

expected1.7 0.77 0.54

S2B KatudumarketKabiza,goldbuyer

hut2.4metresfromburnsite.

ModestHgatamodestdistance 0.42 <0.2 13

S3B KatudumarketKabiza,goldbuyer

hut3.7metresfromburnsite.

Lessmercuryfartheraway 0.22 <0.2 3.7

S4B EastendofKatudumarket,FIELD

BLANK

Mercuryfreeblanksuggeststhatcross

contaminationisnotanissue<0.2 <0.2 1.9

S5B EastendofKatudumarketroofpillar

brick

Non-significantresult <0.2 <0.2 1.5

S6B Katudumarket,EastandSouthend Thisvalueishighforbeingontheopposite

sideofthemarket.Possiblythereareother

burnsites?

0.4 <0.2 3.1

S7B KatudumarketEastandNorthend Non-significantresult <0.2 <0.2 <0.2

S8B Katudumarket,middleNorthend

nearrestaurants.

Non-significantresult <0.2 <0.2 <0.2

S9B KatudumarketWestandNorthend

neargoldmarket

Nearertotheburnareatheconcentrations

ofHgrise0.29 <0.2 <0.2

S10B MarketLuakabiri,roofpillarof

market

Non-significantresult <0.2 0.23 0.73

S11B InsideKabizaelementaryschool

classroom,wallnearestmarket

wheregoldbuyerssetup


S12B Duplicateof11 Non-significantresult <0.2 <0.2 3

S13B Wallandtopofroofbeamofschool

classroombesidegoldbuyerhutsin

market.

Non-significantresult <0.2 1.1 0.87

S14B courtyardofgoldbuyerhouse,wall

aboveburnsite

OddnottofindHghere.Perhapsasminers

usuallyburnattheminesiteandthis

person'scharcoalfireisnotanyhotter,it

doesnotemitsignificantamountsofHg?

<0.2 <0.2 0.47

S15B Insidewallofgoldbuyerhouse Non-significantresult <0.2 <0.2 0.45

S16B Insideburnroombehindbuyers

house


S1BUK NeargoldshopchimneyatHotelMt.

Kauzi,Bukavu

Clearlyspongegoldwithresidualmercuryis

foundhere7 2.6 0.4


68

SampleCode

Description Discussionofchemicalanalysis Hg As Pb

S2BUK Shadowywallspacebetweengold

shopandhotelnearburnarea.

Hgdecreasesquicklyintheshortdistance

fromthelastsample.1.2 0.29 0.81

M1S Amalgamburnsitewall Definitiveevidenceofhealthhazard.Hgis

relativelylowforbeingimmediatelyabove

theburnsite.

2.1 0.81 <0.2

M2S 5metreswestofM1S,above

amalgamburnareaundertheroof

Hgdecreasesrapidlyasonemovesaway

fromtheburnsite0.52 0.92 <0.2

M3S Swaboftheinsidemercurystorage

roomofPDG'shouse,swabbedroof

beams,undershelves,countertop

andunderside.

Mercurystorageroomiscontaminated 2.9 0.46 0.41

M4S BlanksampleatM3Ssite. Non-significantresult <0.2 <0.2 <0.2

M5S DuplicateofM3S Again,duplicatesshowthatHgvariesmuch

morethanothermetals,butduplicates

generallyagreewithin30%error.

1.9 0.36 0.52

M6S Wallsandroofbeamsabovethe

PDG'sburnsiteatcoveredporchof

hishouse.

ByfarthegreatestHgcontentindustis

abovethecampburnsiteatthePDG's

house.

1200 4.9 0.83

M7S Roofbeamsofkitchenhutwhere

theyburnamalgam

MuchlowerHgthanintheamalgamburn

siteadjacent,consistentwithminershabitof

burningonthefrontporchinsteadofintheir

hut.

69 0.89 0.28

M8S Locationoffirepitwhereamalgam

wasburnedinopenair,thisisdust

sampleof3surroundingtreeswithin

4metresoftheburnsite.

Modestelevatedmercuryindust 3.3 0.34 1.5


69

Appendix2:LaboratoryanalysesresultsAllsoil,sediment,ore,leavesandwipetestssampleswereanalyzedformercuryandothermetalsbySpiezLaboratoryinSwitzerland.

Thefollowingtableshowssampleswiththemostsignificantamountsofmercury,othermetalsandthegold

ores.

Sample Analysis/Matrix Massfraction Swissthreshold(TVA1)

M6S Mercury/wipetest 1200µg/wipe -

M10B Mercury/sediment 26mg/kg 0.5mg/kg

MLeaf Mercury/leaves 2.2% -

B11B Arsenic/sediment 14000mg/kg 15mg/kg

BConc Copper/concentrate 1100mg/kg 40mg/kg

B9B Gold/ore <8mg/kg -

BConc Gold/concentrate 10–40mg/kg -

M-Ore Gold/quartzore 13mg/kg -

1Swissregulationaboutwastes(TechnischeVerordnungüberAbfälle)


70

Laboratoryanalysesresultsforwipetests

UNEPCode

UA-2016-12

Crµg/wipe

Mnµg/wipe

Coµg/wipe

Niµg/wipe

Cuµg/wipe

Znµg/wipe

Asµg/wipe

Cdµg/wipe

Pbµg/wipe

Hgµg/wipe

S1B -15 2.7 18 0.57 1.2 1.7 4.8 0.77 <0.2 0.54 1.7

S2B -16 <0.2 <0.2 <0.2 <0.2 0.25 0.87 <0.2 <0.2 13 0.42

S3B -17 <0.2 1.1 <0.2 <0.2 <0.2 0.24 <0.2 <0.2 3.7 0.22

S4B -18 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 1.9 <0.2

S5B -19 0.8 5.2 <0.2 <0.2 0.22 0.72 <0.2 <0.2 1.5 <0.2

S6B -20 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 3.1 0.4

S7B -21 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2

S8B -22 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2

S9B -23 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 0.29

S10B -24 0.41 9.9 <0.2 <0.2 0.28 1.8 0.23 <0.2 0.73 <0.2

S11B -25 <0.2 4.1 <0.2 <0.2 <0.2 1.2 <0.2 <0.2 0.9 <0.2

S12B -26 0.34 8.5 <0.2 <0.2 0.21 3.1 <0.2 <0.2 3 <0.2

S13B -27 4.3 96 1 2 3 19 1.1 <0.2 0.87 <0.2

S14B -28 <0.2 1.5 <0.2 <0.2 <0.2 88 <0.2 <0.2 0.47 <0.2

S15B -29 <0.2 3.4 0.43 <0.2 <0.2 0.6 <0.2 <0.2 0.45 <0.2

S16B -30 <0.2 4.8 <0.2 <0.2 0.27 510 <0.2 <0.2 4.2 <0.2

S1

BUK

-31 0.88 6.2 <0.2 0.24 0.42 4.7 2.6 <0.2 0.4 7

S2

BUK

-32 2.2 11 0.21 0.42 0.49 1.5 0.29 <0.2 0.81 1.2

M1S -33 <0.2 2.3 <0.2 0.24 <0.2 1.1 0.81 <0.2 <0.2 2.1

M2S -34 <0.2 2.1 <0.2 0.25 0.26 0.41 0.92 <0.2 <0.2 0.52

M3S -35 0.22 1.3 <0.2 0.42 0.5 2 0.46 <0.2 0.41 2.9

M4S -36 <0.2 <0.2 <0.2 <0.2 <0.2 0.55 <0.2 <0.2 <0.2 <0.2

M5S -37 <0.2 0.88 <0.2 0.28 0.22 0.63 0.36 <0.2 0.52 1.9

M6S -38 0.66 6.6 <0.2 0.85 1 2.2 4.9 <0.2 0.83 1200

M7S -39 0.48 28 <0.2 0.68 1.2 4.4 0.89 <0.2 0.28 69

M8S -40 0.22 2.2 <0.2 0.29 0.25 0.9 0.34 <0.2 1.5 3.3


71

Laboratoryanalysesofmercuryinsoil/sediment/ore/leafsamples:

UNEPCode UA-2016-12 Hgµg/kgdrymatter

B1B -1 30

B2B -2 8.5

B3B -3 64

B4B -4 34

B5B -5 68

B6B -6 91

B7B -7 3600

B8B -8 340

B9B -9 18

B10B -10 950

B11B -11 740

B12B -12 200

B13B -13 260

BConc -14 270

M-Leaf -41 22000000(2.2%)

M-Ore -42 8.2

M1B -43 160

M2B -44 110

M3B -45 1100

M4B -46 120

M5B -47 6600

M6B -48 3300

M7B -49 5000

M8B -50 940

M9B -51 160

M10B -52 26000

M11B -53 3600

M12B -54 650

M13B -55 1100

M15B -56 1100

M14B -57 410


72

Laboratoryanalysesofothermetalsinsoil/sediment/oremg/kgdrymatter/l=leached;t=total/*goldonlyinformationalvalues,notaccredited

UNEPCode

UA-2016-12

Cr–lmg/kg

Mn-lmg/kg

Co–lmg/kg

Ni-lmg/kg

Cu–lmg/kg

Zn-lmg/kg

As–lmg/kg

Cd–lmg/kg

Pb–lmg/kg

U–lmg/kg

Au–t*

mg/kg

B1B -1 65 360 14 33 40 84 6 <0.5 6.5 1.2

B2B -2 7.2 54 1.3 4.4 3 18 1.8 <0.5 7.4 3.7

B3B -3 20 250 6.6 12 190 26 710 <0.5 15 6

B4B -4 36 410 7.9 17 22 28 8.9 <0.5 7.5 0.87

B5B -5 20 160 5.7 13 240 16 950 <0.5 19 9.3

B6B -6 19 190 6 13 220 14 860 <0.5 20 9

B7B -7 39 15 1 1.9 500 7.3 4400 <0.5 61 60

B8B -8 69 42 1.1 2.5 170 7.3 1900 <0.5 28 9

B9B -9 25 14 0.71 4.5 710 5.1 3500 <0.5 47 72 <8

B10B -10 58 22 3.2 3.2 920 5.1 12000

<0.5 82 120

B11B -11 47 21 3.2 3.1 770 8.4 14000

<0.5 28 57

B12B -12 23 270 6.7 13 200 77 430 <0.5 12 4.1

B13B -13 30 290 7.9 18 120 41 390 <0.5 14 4.8

BConc

-14 36 40 5.6 11 1100 6.9 5700 <0.5 110 44 Oct-40

M-Ore

-42 12 30 3.5 16 31 12 62 <0.5 8.1 <0.5 13

M1B -43 31 130 8 12 7.4 25 3.1 <0.5 5.4 0.56

M2B -44 100 480 27 37 21 75 7.3 <0.5 21 1.1

M3B -45 24 140 7.6 13 11 25 37 <0.5 12 <0.5

M4B -46 46 400 13 17 16 25 1.8 <0.5 6.2 0.59


73

UNEPCode

UA-2016-12

Cr–lmg/kg

Mn-lmg/kg

Co–lmg/kg

Ni-lmg/kg

Cu–lmg/kg

Zn-lmg/kg

As–lmg/kg

Cd–lmg/kg

Pb–lmg/kg

U–lmg/kg

Au–t*

mg/kg

M5B -47 33 50 4 22 18 12 130 <0.5 23 <0.5

M6B -48 30 40 3.2 19 15 16 110 <0.5 16 <0.5

M7B -49 19 56 4.6 16 16 15 130 <0.5 20 <0.5

M8B -50 19 120 3.9 5.9 8.5 63 54 <0.5 7.7 <0.5

M9B -51 66 38 4.5 18 42 32 310 <0.5 17 2.3

M10B

-52 20 34 4.1 23 25 19 180 <0.5 52 1

M11B

-53 35 130 4.6 12 14 31 100 <0.5 20 <0.5

M12B

-54 33 280 8.9 11 9.9 25 7.7 <0.5 5.9 <0.5

M13B

-55 30 120 5 8.4 9.7 19 17 <0.5 4.1 <0.5

M15B

-56 15 160 4.8 6.5 57 150 48 <0.5 9.5 <0.5

M14B

-57 21 97 3.1 11 3.4 35 4.1 <0.5 18 6.2

Sinuous patterns from extensive reworking of the Luzinizi riverbed by artisanal miners

environmental assessment of mercury pollution in two artisanal gold

Documents