Environmental Impact

Soulh Pociflc Reglonol Envlronmenl Progromme

SPREF Reports ond Studies Sedes no.90

Environmental imPact of large-scale mining in Papua New Guinea: Sedimentolory and potential mobilization of trace metals from tnine-derived material deposited in the FIy River Floodplain

by Jiirg llettler and Bernd Lehmaaa oryricbt @ South Paeifi e Re.gional :Enryirou,rneni Progtamlngn l-g96

The South Farific Regional En'rd.ronuent Pmgraeue authotises the reprodnctisr of thie material whole or iU part, in an1'form pmvid€il apfiropoiate acknooulofuenaat ie g:i-tren.

Original Text English

Prurblshed in Sepl:ember l9g5 by: Soulh Poeifie Reglonol Eh viro'nm e n,f F ro;gro rnrrn e 'F",O. Eox 240 Apio. Westerrn Somoo

Commerclsl Ptinters Apio. \{esfer, rt Son loq

F 39/e5 - 4C

Fdn:ted with finsneislossisfonc,e fr,om fhe U nlled N otlons Environm gnf Progrorn m e ( lJ NIEPI

loyout by Weslery Wsrd, SP,REP,

SPREP Gataloguingtnfublieation Data

Hattler, J6ng Eqviro-nmental impaet of large-seale mining in Fap*a Nerv Gurnea ; gedi.urentology and p'ortchtial ulobill'aansn of tiace metals fiou. m ne.derivEd mateiial d.ep.osited itr tle Fl6, River F.'losdptain / by Jiitg Hettler and Bernd I;eht'rann. - Apia, Westetn Sargoa : SP.EEP, 1995.

vi, 7lp, I 29rcm - (SPflEF Reports addstrrfi€s aelles,; n'g. 9O).

L$IEN: 982-O4.Org5-9

1. tsnvirsaueqtal iurpaot analysis = Papua Ne:rtr Guinga. 2. Mines,and r.rtioer.al resour'ces, Environrneotal aspe,€ts ' PaBua N,ew Guirirea. [. Lehmann. Bernd. I. South P.acific R,egional Env,ir,onmeut Frogranls.e. III. Title IV, Series.

$68;709549 South Pocific Regionol Environment Progromme

SPREP Reports ond Studies Series no.90

Environmental Impact of Large-scale Mining in Papua New Guinea:

Sedimentology and Potential Mobilization of Trace Metals from Mine-Derived Material Deposited in the Fly River FloodPlain

by Jdrg Hettler and Bernd Lehmann

Te cht i sche U n iu er sitcit C lau sthal Freie (hiuersitrit, Berlin,, G erman'y

Published in August 1995 in Apia, Westem Samoa Preface

The Ok Tedi mine in PNG ranks amongst the This is particularly tlue of the road. l)ower.. world's largest copper mines. It is a major health and water systems, transport and contributor to the PNG economy, accounting for communication facilities. These services are up to 45o/o of PNG's total annual export meeting the basic needs of local villagers by increasing life expectancv and provicling access revenues. The PNG Government 3O% is a to education, employment (OTML). and business shareholder in Ok Tedi Mining Ltd. opportunities. Mining for gold commenced in 1984, and for copper in 1986. Mining for copper is expected to Ok Tedi Mining Ltd. has taken a numbel of continue for a further 10-15 years. iniatives to encourage as nruch participation as possible from people rn the regron and so enstrle It was initially intended that a tailings dam human developrnent goes hand rn hand with should be constructed to contain the mine's mineral development. waste. However for economic reasons and given Sustainable development also requires good that the area is structurally unstable, no tailings environmental stewardship. During the dam was constructed. So, since 1986, 100,000 to devlopment of the Ok Tedi ploject there have 150,000 tonnes of waste per day have been been criticisms of its environmental impacts, discharged into the Ok Tedi and Fly River many of thern based on inaccurate infornration. system. Ok Tedi Mining Ltd. is a responsible resources development company which recognises that all In 1989 the PNG Government set compliance environmental effects must be carefully standards covering suspended particulate load, considered. It is the extent of these impacts and dissolved and particulate copper levels, whether they are reversible or not, which must biological parameters like fish catch, and others. be considered in relation to the social and It is argued by some that the environmental economic benefits to the communities involved and the nation management of the mine is based on monitoring as a whole." for compliance with standards that have been Whereas at the same seminar Mr Alex Maun. a established "to facilitate the realization of prominent landowner stated: OTML's mining schedule" rather than to protect "With the mining operation our life has changed. the environment. The loss of our rainforest and degradation of the Ok Tedi Mining Ltd. has an extensive environment cannot be calculated in terms of environmental monitoring programme. Compl- short-ternr money. We are rural village subsistence farmers who depend the iance monitoring and interpretation are on envlronnrent for survival. assessed routinely by the company (and the Government), but no detailed data are available Before the Ok Tedi Mine operation our life was for public scrutiny (if at all) for at least a year paradise, we en;oyed both the aquatic and after data are collected. This lack of public ten'estrial lesoul.ces. We used the river for fishiug, washing, dnnking and transportation. accountability for environmental p e rform ance is We nrade gardens near the rrver banks which unacceptable. lasted3toSyears. Inevitably the Ok Tedi mrne evokes different Now we rrver people can no longer drink from emotions and perceptions by different groups in the river nor can swrm. bath or wash clothes in PNG. For example, Mr Kipling Uian. the former the nver'. We lack the protein rn orlr diet that Deputy General Manager of OTML and now !!'as fot'rnerly provided by the aquatrc and BHP PNG General Manager, siated rn lgg3 at telrestlral resources. Overflow of the Ok Tedi the 20th Waigani Semrnar on "Environment and Rrver has caused the wild life near the river. banks and Development" at, the Unrversrty of PNG: floodplarns to disappear. Some game anrnrals were drowned by sudden floods. "Importantl-v, the mrne has brought positrve changes to the quality of life of neighbourrng communities in the previously trnderdeveloped Westeln Province. lnve.stment in proJect infrastructure has amounted to about li300m and many of the basrc services now established are available to all citizens in the area, 'lrt'es iu tht' tlootlplaius at'e dving corupletely The Floodplains lblcrrrg thc rvild life to nrigrate to other aleas. river Ncrs' galdt'us al'e llo longer tnade near the The Ok Tedi mine adds about 58 million tonnes lla nks. of sediment to the Ok Tedi River per year. It is I rvoull hkc to stress that Ok Tedi Miuing is estimated that 30% is deposited along the Ok t:arrsing ll'l'eversible destntction along the Fly Tedi, much of the rest reaches the Fly Delta. An Ilirrrl An Ok Te'di River'. Wlratever rt is unknown amount is deposited along the Fly rlestlovrng u'rll ner.el coure back to not'tual. e.g. River and in the floodplains in the middle Fly. crrstrlrui.tt'v land, sago swallll)s, etc' OTML is Studies on the coastal and maline ecosystems in blrnging ttttsrtslatttable tleveloptnent. We Papuan Gulf and Torres Straits afTcctetl rtvct'peoltle thitrk rt is ttcrusense talking the Fly Delta, ltbont srrstainallle developnrent whetr the ruess are being done by both Ok Tedi and Australian rlonc bv the Ok Tedi lUirring rs not cleaned trp." scientists funded by Ok Tedi. Studies on the floodplains of the Middle Fly are being done by "Ok In 1994 Oh Tedi again hit the' headlines: Ok Tedi scientists. and are also the focus of this Zetpl", so says 'ledi has to build tailings dam ' UNEP study. Post Courit'r' headline on 18 lUarch 1994. The report gocs 0n: OTML. in its booklel Oh Tedi, the Erwirotunerfi al.d, You' states that: "Ok Tedi Illining Ltd (OTML) will be tokl to build a series of danrs to dispose of mine wastes "coppel in the sedinrent in the Fly River is also or' "ship ottt" il it leftrses. Mr Zeipi has always being transported during floods onto the insisted a tailings darn be built because, he savs. floodplain where it settles into the lakes, the river rlunrping has darnagirlg effects on the streams aud the flooded forest' At t'hie time ecosvsteltts." there is very little scientific information to tell trs what effect thrs copper on the floodplain (\VIr. orr.d Zeipi i.s lllliltister for Enuironrnettl mrght have on hsh life, feed and breed in this Cott,sert'al iotl). atea". again says "[i2b case looms on mine damage" Consequently OTML has commissioned studies 4 May 1994. Post. Cou.rier headline. this time on of the arnounts in and effects of copper on the (; Then in ltosl Oorurie,'on I\Iay a leport states: floodplain ecosystems. Unfortunately, the PNG "'l'he leatlel of a Papua New Guinean clan Government and independent researchers lodged a writ in the Melbottrne Srtprenre Coult generally lack the financial and human .vesterdav itgainst Attstralia's biggest company' resources to do independent monitoring and BHP, seeking trnspecified datttages for allegedly research of depth and detail. Hence most studies his poisonine the Ok Tedi River aud destroyrng are done by or commissioned by OTML- people's sttbsistence way of' life. It also alleges thc PNL} Gover:tuent. a 30 percent shareholder' This study by Jtirg Hettler and Bernd l,ehmann in Ok Terli Mining Ltd.. had "failed. neglected is valuable since it is done by overseas scientists and t'eftrsetl" to enfot'ce e nvtronrnental funded by UNEP and SPREP through the aHreelnents and convenants'" universities of Berlin and Clausthal and the These rnattels are still being dealt wrth in both tiniversity of PNG. It is apparent from their PNG and Attstralizrn courts and havc not been work that copper is being deposited in the resolved. Furthermore present debate continLles floodplain at higher than expected levels- between Govet'ntnent ministers and leaders and Hopefully future independent seientists can Mr Zeipi on the 11rpe o{'courpensirtiol"r paYnrents assist in deciding if this copper contamination is for envirotrtucrntttl dautage :llld rvhethcr a liliely to have biological effects, for example, if tailings dam should and cottld he' btrrlt. fish populations may be affected. The present study is a valuable contribution to out present state of knowledge of the Ok Tedi/ Flv River system. In fact, more environmental research has been done on this tropical river systen) over the last l5 years than probably any other troptcal rlver system in the world. Yet tht're. rs strll uruch unceltainty! Back in, 1989 in hls 'book The Pst of Gold,. Bichard Jae.kson made a statement whieh (in slishtly nodified forr) stiil is applicable 13 years later: tln our present state of knowledge of rhe wgrLinca of natural sy€tems' it is only in the aarest of oecasions that honest snvironmental ercpetts agrec as a body and confidenlly prediat a se(Iueace sf futrrre events 4ndoutcomes. The Ok Tedi Brotsct is not one of those occasisns. Decisions ha,ve bepn mada ib the prqiect hoping that environnental rieks will be worti it, that is, hopiaC thart in the light of the fants available the maggitude of the enrir.onmerrtal impact wlll be min:irhal. But tbe facts available are, still. too fen'. In the OkTedi case, the expertl, ifthey are, honest, will b€ kesping their fiagere,srpssed and environmental monitoring syst€fts under consta[t scrutiny ..." and hoprug that the ennironnental da'r,ages remain accciptable,! Q^h€kW

Dr Doutd,lWowbray, 'Prgfussar of E- nuirontnerfial $cienre at the Utduercity of Papua New Guinea, and, Gouerntnetfi Ad,ui.ser tn the De-parfinentaf En airotutneitt,an d' Canservatistr; ItlE" The opinions expressed in the Frefaee are thoee ofit€ arrthor and not necessarily those of the SPREP.

iv Gontents

Surnrnary-. 1.lntroductio[....,'...... '...... ].r.'..idri.|t...r.t 2.sa.nrplinganditnalyticalh4ethodology,'..-.-....j.'...... l..... 2.1 Sampling...... -..-i'.--..- 2.2 Analytical Methods 3 3. The'Ehvironment of the Ok Tedifly River Re-gion --.i-!.--r- 6 3,1 Geolsgy of the Mount Fubilan Ore Deposit..'*--..-."....-....j'r....r.r'..i..i 6

8 3. Geonoorpholory and Vegetation...... '.. 8.3 Hydrslory and Climate ..n.i...'...... 3,.4 Aquatic Biology...... r....-...... !,..; 3.5 Populatior,l,,..*.,..... 4. Input of the Ok Tedi Mine iu-to the Ok TedilFly Biver Systcn....-.'...--...---!.r-q-....,'. 11 4.1 DescriBtion of the Mining Project .r.+.!.r!'...r.:.'.-r..'..-.;....ir..?- 11 +.2 Discharge of Tailiags and tri[aste Rock.-'..-...-. ..,'...!.*..--...... r!'.r"."r 11 5. Sedimentolog,y of the Fly River and its Floodplain- i'.rji'"' 1g 6,1 Natural Sedirnentary Pfocesses i.n the Floodplain,-;..i.!.i.-'-.r..r. 1B 6.9 DeposiGion of Mine-Detived Material.. ..-.-i-ri...."r.--.!,.-'r..:r...r--.i.'..'- l5 6. Foteatial Mobilizat-ion of heavy metals and hydroc'hemiehy..'..-..,--'--...a:i;.,-....-'.." 2l 6,1 tleaw MetaLe in Sedirnents of the Ok TcdilPly Biver System."...'...'..."."""" 2L 6-9 Hyilrochemietry of tbe Ok Tedi/Fly River Syetern .--.--..i.--.--r--...;-..-...-dr-...... -- n

7 . Biological Impaets...... -i....p..i.,,.i.. 8. eonclusioRs ---1...r....,-....! 43 9. AcknowledBeuleata Summary

The Ok Tedi copper-gold mine, located at the Due to the flat terrain, turbid Fly River water eastern end of the central mountain range of intrudes regularly upstream of the floodplain New Guinea, discharges approximately 80,000 tributaries (measured intrusions up to 25 km). tons of ore processing residues daily, and a A thin layer of 1-5 cm of copper-rich material similar volume of waste rock and overburden (400-900 mg/kg Cu) was usually found on the into the headwaters of the Ok Tedi River. bottom of drowned (tributary) valley lakes. Copper in sediment deposited pre-mining The Mount Fubilan orebody, which is the source in the period gave a median value 44 of heavy metal-rich sediments deposited along of eO = 25-63) mg&g. the Ok Tedi and on the Fly River floodplain, contains a suite of base metals, of which copper Riverine particulate rnatter also settles down on is the primary environmental concern. The Ok the flooctplain at times of overbank flow, which Tedi River flows into the Fly River 200 km leads to extensive copper contamination in low- downstream of the discharge point, where the lying swamp sites close to the r.iver. Natural mining wastes carried as suspended load are deposition rates in the floodplain were diluted. determined by C14 age dating to range between 0.1-1 mm/year, wrth exception This study investigated the deposition of rnine- the of oxbow lakes, where natural sedimentation derived sediments in the lower part of the is much higher. l,eaching copper material Middle Fly River flood-plain, and the from deposited on swampy, vegetated floodplain hydrochemistry and potential mobilization of sites was detected sediments trace metals, particularly copper, in this alluvial in which were also strongly depleted in calcite and sulfide, which plain. To this end, a total of 156 sediment cores are the components most and surface sediment samples and 117 water easily mobilized of mine-derived material. samples were taken frorn the upper Ok Tedi and the Middle and Lower Fly River floodplain. The variable water table. oxidizing environment and acidic conditions generated The suspended matter content Middle by decomposing in Fly vegetation. River water today is about 5-10 times higher typical features of low-lying floodplain swamps, facilitate the mobilization of than the natural background of about 60 mg/I. copper from the solid into dissolved phase. Near-surface sediments deposited along the the Water taken from swampy floodplain locations river channel contain up to 1100 mg/kg copper, showed copper values of up pg/I with the mean value is 530 (*O 240-820) to 50 copper in - the (membrane mg/kg. filtered sample filter 0.4b pm). Average copper content in mixed waters of the Of the floodplain water bodies, cut-off meanders inner floodplain is around 9 (+0 = 5-14) pg/l; the receive the largest quantities of mine-derived Fly River water has around l7 (+0 = 13-19) pg/l sediment. Deposits of up to 70 cm in thickness coppef. Copper concentrations in unpolluted of copper-rich material (with 800-1000 mg/kg floodplain waters were measured at below 2 FCll. copper) were detected in oxbow lakes, which Nearly all dissolved copper have accumulated srnce the nine started in the Flv River system rs complexed discharging residues in Very hieh by dissolved orgaruc carbon 1984. compounds, however, deposition rates (around 4 cm/year) of mine- a fraction of these complexes appears derived sediment were determined in locations to be labile and reactive. Comparrson of dissolved copper levels close to the creeks and channels whrch link the measured during the present Fly River with the outer floodplain. study in the Middle Fly River floodplain with literature data on copper toxicity and interuational water quality guidelines shows that chronic toxicity of the metal to the aquatic community is to be expected. Sign- ificant negative ecological effects, particularly on the local fish population, may develop with a considerable trme lag, since aquatic organisms at the base of the food web are the biota most sensitive to copper.

1. Introduction

The environmental problems of mining activities The present study, funded by the United have been receiving increasing attention Nations Envirorunent Programme and the worldwide in the last decade. The shift in the South Pacific Regional Environment search for raw materials outside of the classical Programme, is the first independent research producer countries to the world's marginal project undertaken in the area. It investigates regions like tropical rainforests has led to a one of the most important environmental number of environmental and social problems. aspects of the mining project. Based on discussions with the Environnent Department natural In many cases, the extraction of of Ok Tedi Mining Ltd., it focusses on the fate of resources has been the first industrial activity mine-derived sedrments deposited in the FIy being developed in peripheral regions. River floodplain. Requirements of infrastructure, logistics and workforce may completely change the The project's coordinator was Dr- Bernd environmental and social patterns in regions Lehmann, Professor of Applied Geology at which have previously been nearly untouched by Technical University Clausthal. The respons- man. ible research officer was Jcirg Hettler, M-Sc. (Geology), of the Department of Environmental examples of such a One of the best d.ocumented and Resource Geology at the Free University of development is the Ok Tedi gold-copper mining Berlin. The sedimentological part of the study project in Papua New Guinea. Before the was co-ordinated by Dr. Georg Irion of economic value of the Mount Fubilan orebody Senckenberg Research Institute at was discovered in the late 1960's, the mountain Wilhelmshaven. The University of Papua New was a sacred place to the sparse population of Guinea provided comprehensive logistical were the local Wopkaimin Papua tribe, who support during the field and laboratory work. inhabiting one of the world's most isolated regions. A draft of this report was presented by the study team in a series of discussions held in The environmental impact of the mrning September 1994 in Port Moresby and Tabubil, operation, located in a difficult physical Papua New Guinea, with Ok Tedi Mining environment, has been subject of controversy Limited, the Departments of Mining and since the project's early planning stages. The Petroleum (DMP) and Environment and fact that the Ok Tedi mine works without waste Conservation (DEC) of the Papua New Guinea of the retention facilities has been in the focus Government, and the University of Papua New debate in the last few years. Guinea. The mining company is engaged in an extensive environmental monitoring program and also has commissioned a number of studies on individual problems which have been executed by international research institutions and consultants. 2. Sampling and Analytical Methodology

The sediment cores recovered were usually in 2.1 Sampling the same range of lengths as the cores from water bodies.

Sampling locations were determined with a handheld Global Positioning System (GPS) 2.1.2 Water receiver and available topographic maps of scales 1:100,000 and 1:250,000. Water depth A total of ll7 water samples in the Ok Tedi/Fly and bottom relief were measured with a portable River system was taken. Water temperarure, electronic depth sounder with LCD. pH, conductivity and oxygen content were measured in the field using portable electronic equipment. The reduced species HS- 2.1.1 Sediments NHI*, and NO2- were also determined in the field using "Merck Aquaquant" reagent kits for rapid water Thirty two (32) surface sediments from the river analysis, based on a colorimetric method. bank of the Ok Tedi at Tabubil and from a few locations in the Middle Fly region were sampled All water samples were gulp samples taken by with a plastic spoon and collected in geochemical hand 20-50 cm below the water surface, the sampling paper bags. volumes ranging between 250 and 1500 ml. The wide-mouth polyethylene bottles were soaked in The large majority of sediment samples (124) dilute nitric acid for two days and washed with taken along the Fly River was recovered with a demineralized/distilled water before use in the Sravity corer. field. The containers were rinsed twice with Because of the difficult access to swampy, water from the sampling site before the sample vegetated floodplain sites, the large majority of was taken. Upon return to the field laboratory sediment cores (and water samples) were taken withrn a few hours after collection, alkalinity from channels, lakes and small water bodies was determined by titration with 0.02 mol HCI which were accessible by boat or dugout canoe. to pH 4.5 (APHA Standard Method No. 408). The gravity corer consists of a massive The water was filtered through 0.4b pm aluminium tube with a backslash (floating) cellulose nitrate membrane filters (Sartorius, valve mounted on the upper end. Fins are Germany) with a portable "ANTLIA" pressure welded to the aluminium body to stabilize the filtration system (Schleicher and Schuell, corer during its free fall through the water Germany) which consists of a pneumatic pump, column. Coring tubes of transparent acrylic a 50 mI syringe cylinder and a filter holder of 50 glass ("Plexiglas") of I m or 1.5 m length were mm diameter. Sometimes a pre-filter (Schleicher fixed inside of the alumimum shaft. The corer and Schuell BIue Ribbon ashless) had to be used was brought in an upright position above the for highly turbid water. After sufficient rinsing water and then released for free fall. of the pump and filtration system, a 120-ml water sample was taken and was immediately To increase penetration depth in the bottom acidified with 3-4 ml of concentrated nitric acid sediment, a ring-shaped lead weight of l0 kg (Merck Suprapur) per litre of sample. was fitted to the alumimium shaft. The corer is hoisted from the bottom of the sampled water Filters were retained for gravimetric body with a rope fixed to the backslash valve, determination of suspended solids. The which remains closed. suspended matter content in some samples was measured using a 1.S-litre "Imhoff' funnel- The sediment cores recovered have a diameter of shaped sedimentation cylinder. Few water 36 mm and a length of between 20 and 70 cm samples were split and a subsample was depending on the nature of the bottom sediment. acidified unfiltered to determine the trace metal The cores, which showed no or minimal content associated with the particulate matter. disturbance, where pushed out of the plexiglas Water samples selected for anion analysis were tubes with a rubber stopper and were packed in filtered but not acidified, DOC samples were clean polyethylene bags. On floodplain sites, the stabilized with 2 mlAitre of 50% H,SO4. sampling tubes were driven by hand into the sediment. After closing the open end with a rubber stopper, the tube was pulled out. Great care was taken to avoid cross- Insoluble carbon and sulfur were determined by contamination of collected material. AtI Leco furnace, in which CO2 and SO2 gaees are equipment was acid-washed and rinsed with released during combustion and then detected. distilled and deionised water between use for A l5o/o HCI leach prior to combustion was different samples in the field laboratory. Blanks performed to remove soluble eulfate and were included and subrnitted to the same carbonates. This procedure ensures that only procedule as the sarnples. organic carbon and sulftrr bound as metal sulfide is detected. Two reference sediment samples, NBS 2704 2.2 Analytical Methods Buffalo River Sediment and EEC/BCR RM 280 Lake Sedim.ent, were submitted to the commercial laboratory for quality control- 2.2.1 Sediments Results are shown in Table 1. was performed on five Sediment cores were cut in halves in the Radiocarbon age dating peaty sediment from drowned valley laboratory with a stainless stell knife to allow samples of at the C-14 laboratory of Kiel University, visual inspection and taking of photographs. lakes Germany. Sub-sections of sediment cores were wet sieved with distilled water through nylon sieves of 20, 2.2.2 Y,later 60, 100 and 200 pm mesh size. The resulting size fractions <20 pm, 20'6O prn, 60-100 pm' samples were analyzed at a commercial and >200 pm were washed from the Water 100-200 um (XRAL, at the laboratory of sieves into plastic containers and dried to laboratory Canada), of Environmental and Resource weight at 80o in a drying oven' Grain the Department constant Universitit Berlin (FUB), and was determined after weighing. Geology at Freie size distribution at the laboratory of the Environment The fraction less than 20 pm in samples selected Department of Ok Tedi Mining Limited at for clay rnineral analysrs was submitted to Tabubil (OTML). gravity in "Atterberg" sedimentation separation for metals was by multi- cylinders. Of the resulting three fractions (<2 Commercial analysis ICP-S, of which measurements for 13 pm, 2-6.3 prn, 6.3-20 prn), the finest was used to element elements are reported in this study. The . trace prepare smeat" slides which were submitted to copper, lead, zinc and cadmium, which X-ray diffraction analYsis. metals were present in some samples in concentrations Samples selected for chemical analysis (grain close to the detection limit of ICP-S' were size <20 pm was the standard fraction used) additionally determined by graphite furnace were shipped to commercial laboratories (ACME atomic absorption spectrometry (GFAA) at FUB and XRAL. Canada), where the sediment was and OTML laboratories. and pulverized in an agate mill. homogenized (sulfate, chloride, nitrate) were ln the "total digestion" procedure, 250 mg of Major anions by suppressed ion chromatogtaphy on sample is digested with 10 ml HCI0a-HNO"- analyzed samples, Dissolved organic carbon HCI-HF at 200"C to fumrng and diluted to 10 ml non-acidified (DOC) was determined with a "Technicon" auto with dihrted aqua regra. analyzer. vigorotrs digestion procedure leads to The water analysis was more potential loss of As, Sb and Cr due to Quality control for compared to sediments, as no during HClO,r fuming' The leach, diffictrlt when volatilization was available' partial magnetite, clu'omite, standard reference material however, is for values, which were of Zr and Mn. 35 elements Calcium and bicarbonate barrte. oxides Al. methods, gave a by inductively cotrpled plasma determined by two different were determined 99% in the data set of all spectrometry (ICP-S), of which 24 are reported correlation of in the present study. Arsenic and antimony measured values. were additionally rletermined by hydride Values obtained for blanks and measurements generation with aqua regia digestion and ICP'S of the same sample by different methods and analysis. laboratones are given in Table 2. The analyses (demineralized / distilled laboratory together with some other elements were of blanks Gold no minimal contamination, in selected satnples by neutron water) showed or analyzed of zinc. Reported values for activatron analysts (NAA) in a commercral with the exception metal have been corrected by subtraction of laboratory (Bondar'Clegg, Canada). this zinc content in blank samPles- Toble 1: Quality control with stondard. reference ,naterials. Volues with na wlcertairlty range giuen are noncertified,. As ualue with * is by hydride generatiar./ICP-9.

N€+ -+t o6 B 6

k qH o

6 6 d E E €€

e t'- rt€

! H h F g=eo F F

fiN e

ed F+t€

o6 N

-€A+I F +€ g 60 HN E 6+t -

G.6 . O -+td NE €+td-i G6 6F od € 6

N 6+t € NN-+

66 o€ 6i 6 Q+t 6 HH

d€ta oa Bg + d- -+t -+ oo eo6 o aa d€ 6 r64 G+t € FF 8o a €e€ 66RA €-F €6 {+t N d eooo ga N-oo €€ a -# d Eoooa A BA o-o €o6 E N+' N )< d eo o BA d E 5= 6 E.FR - R 4+l = : A ge a 5E ts E B @ e -E o it ? B a e s* E-5 d 18-E E '.i.s s E'= :H@O 'i&> e- -

Sampler ad Lrbor.todet fin ppb)

w1n8"t0 &l Cfl Pb Gd

XBALICP€ ld282<1

FUB6fM 25 n 3. <.1

w1.t30:3. F' tn Zn Gu Pb cd

,XRAL lGF.$ f26 1:61 I 4 td td

FUB.GFAA nd nd '10 5 <1 <.1

OTMLGF.AA 10? 76 I 5

w31tr.1, Fp lln ZA Cu Pb cd

XRALICF€ t1l 21 28 <2 nd nd

FUBGFM nd nd 2U 1.5 <'l <.1

OTMTGFAA w u 19 1.3 <1 <.0f1

'wtt/|.1, Fe Itln Zn Crt P! Gd

xfiALrcPs 319 ll3 ll 53 nd nd

<1 <.1 'FUtsr,GF,M nd nd 20 50

OT'II/IL GFAA 134 t6 t5 31 ,4 0.1

ui3t8.1. F. fn tul Cu Fb cd

XRATICP€ 50 <5 <5 14 nd rxt

FUE 6FAA nd hd <5 {6

QTML GFIAA 484 1.3 16

Blank(DDWtabwatal [e iln Zn GU Pb cd

XRAL]ICP.s <10 <5 lt 4 nd nd

FUg GFAA rrd nd 20

OTMLGFM 27 0.5 13 0.2 <1

$tatietical analysis of eed,ixqent arrd i+ater v,ahres was perforned ua-ing the U-S- EnvironrnEntal Prstection Ageney "Geostatistical Environnental l\sseesmeqt" s.oftware (dEO-EAS, USEPA t$SE). Geoehennieal iuodelling and speciation determrnation wase done with aR updated version sf the PHREEQE eoftware (Plurmer. Parkhurst & Thorsteaso r, 1980),, 3. The Environment of the Ok Tedi/Fly River Region

Table 3: Oh Tedi ore contpositiott 3.1 Geology of the Mount Fubilan Ore Deposit 1. Ore minerals of the oxide zone: cupriferous goethite Cu.FeOOH, cuprite Cu20, native Cu, malachite Trace metal contamination of the Ok Tedi/Fly Cu2CO3(OH)2, azurite Cu3(COj2(OH)2, copper River system through mine discharges is closely carbonates geochemical composition of the Ok related to the 2. Ore minerals of the sulfide zone: chalcocite Cu2S, Tedi ore. di genite, covellite Cu S ( supergen e f ormati on sl

The orebody at Mount Fubilan consists largely ? chalcopyrite CuFeS2, bornite Cu5FeS4 (very minor), of an intruded and altered monzonite porphyry stock of Lower Pleistocene age, which hosts a 4. pyrite/marcasite FeS2, molybdenite Mo52 mesothermal. stockwork and disseminated ( (p rotore m in e ral izati onl copper-gold mineralization (Rush and Seegers f990). Intense weathering at Mt. Fubilan led to o. Sulfides of Se and Ag are accessory minerals in the the formation of a copper-depleted, but gold- main orebody (Fubilan Monzonite Porphyry) enriched cap (gossan) overlying the primary Pb, Zn, Cd, As, Ag and Se minerals are found mainly in mineralization. The deposit is similar to other skarn ores big copper porphyry mines in the Pacific Rim region. Pyrite and chalcopyrite are the dominant sulfide minerals in the protore below the 3.2. Geomorphology and Vegetation supergene enrichment zone. Average metal concentrations in the orebody are O.75o/o of The minesite at Mount Fubilan. which had an copper and 0.67 grams per ton of gold (OTML elevation of 2094 m above mean sea level (MSL) 1993). before mining started, is located in the upper At the contact between the intrusive and catchment of the Ok Tedi (see Map l). The site adjacent sediments, calc-silicate, sulfide and receives rainfall of 10,000 mm per annum, magnetite skarns have formed, which make up which is amongst the world's highest. approximately 10% by volume of the orebody Dense tropical rainforest blankets the ridge and (Jones and Maconochie 1990). ravine topography. The area is seismically Porphyry base-metal deposits of hydrothermal active and prone to landslides. North of the Ok origin like Ok Tedi typically contain a Tedi catchment stretches the massive Hinden- paragenetic sequence of the metals copper, burg Range with maximum heights of 3325 m. molybdenum, silver, gold, lead, zinc, iron, The range is a highly unstable, cliff-like struct- manganese, selenium, arsenic. cadmium, ure btrild up of Tertiary Darai Limestone, which tungsten and others, which are not evenly is responsible for the high natural calcium and distributed within the orebody. but are located bicarbonate content in Ok Tedi water. quartz in distinct zones around the low-grade Tributary slreams in the area are very steep core of the intrusion. Very high values for lead and form narrow, gorge-like valleys, with (1000-3000 ppm) have been found in overburden boulder-size alluvral debris. Upstream of sampled in October 1991. Ningerum (60 m above MSL), the Ok Tedi is a The skarn mineral paragenesis has an elevated braided nver, up to 100 m wide. At Ningertrm, and more variable content of trace metals as 70 km to the south of the mine. the Ok Tedi compared to the coppef porphvry ore, and is River leaves the mountaineous region. Between more critical fronr an environmental pornt of Ningerum and l{onkonda. the river passes view. through a transrtion phase and enters the Fly Platfor:m. a vast alluvial plateau wrth very little However, arsenic and mercury, whrch ilre relief and meandering streams. 130 km frequently associated with gold mineralisatton, downstlearn of Ningerum, the Ok Tedi flows show no enrichment above bacl

!f tut. runit nl OK T EDI MINE iqt?0"0" PORGE i \"*, \x

ilE | _': Ningerum / / f/ .\_r{f rJt KonkondaIt \ on x",*i5ffrlun D* rJ iba d

Lake Murray

Obo to

I o Lervada i I I i Daru i 0 25 50kn L-l-l

142 " South of the junction, the river channel is flat Their bottom relief clearly reflects their origin and meandering, flanked by thick jungle, as tributaries to the Fly River. During the sea extensive backswamps, lakes and lagoons. The level rise which ended about 5,000 years ago, reach between D'Albertis Junction and the the river ends of the larger floodplain creeks confluence with the Strickland River south of were filled with sediment derived from the Fly Obo (Everill Junction) is called Middle Fly River. Water can flow freely in both directions, River, with a length of 410 river kilometers. The depending on the water levels in lakes and the sinuosity factor is about 2.4. The channel width Fly River. varies between 200 and 300 m on average. Due to the low relief. the edges of lakes or Downstream of Kai and Agu Rivers/Lakes, the "lagoons" are not well defined and tend to merge floodplain vegetation pattern is dominated by with neighbouring grasslands at high water tall, dense reed (Phragrnites, Saccharutn) and level. Higher banks. on which the villages are smaller grasses. Jungle is restricted to a few usually located, are remnants of eroded plateaus isolated areas of higher elevation. Trees grow of Pleistocene age. The water level in the lakes only in rarely inundated areas (with the is highly variable. Following consecutive dry exception of Sago palms and thin-stemmed months, lakes may dry out completely which can Melaleuca) because of their susceptibiiity to result in massive fish kills. prolonged flooding. The fact that trees are largely missing in the lower part of the Middle Fly points to frequent natural flooding. 3.3 Hydrology and Climate Lakes and shallow water bodies and channels of the floodplain often have a dense vegetation of The climate in the investigated area is wet aquatic grasses, floating Azolla (Equisetum), tropical. The minesite receives well distributed Hstio (Araceae), water lillies, and the rainfall on 325 days throughout the year. submerged Ceratophy Llurn. Average temperatures show a nocturnal minimum of 20'C and a mean daily maximum of Below Everill Junction, the lower Fly River is 27"C. Mean annual relative humidity is about tidally in-fluenced, reaching the delta 400 km 85% in the Fly River lowland. Temperatures downstream. Due to the increased flow rate and are constant throughout the region with an a lower gradient, the channel width and annual mean of 29oC. meander amplitude are greater than in the Middle Fly. The Fly River has a total length of 1120 km. Its most important tributaries are the Ok Tedi and The Fly River floodplain is made up of an inner the Strickland Rivers. The catchment comprises part, the channel migration zone comprising the more than 76,000 kmz. The mean annual runoff present river channel, cut-off meanders and per unit of catchment area is about 2500 mm, meander scroll complexes, and the outer part which is higher than for any other river system with drowned vallev lakes and extensive in the world (Maunsell and Partners 1982). As backswamps. the river catchment is relaiively small, The floodplain widens from 4 km width near streamflow is characterized by short-term water Kiunga to over' 14 km close to Obo. level fluctuations with rapid changes between Discontinuous low natural levees, about 1-2 m flood peaks and drought periods. above mean river level, flank the existing river Rainfall is highest in the upper Ok Tedi regton, channel. with recorded annual values up to 14,000 mm, Drowned valley lakes are typically connected to 7,800 mm at Tabubil, and decreasing in the the river by a narrow, slnuous channel with a lowland. At,I{iunga on the Upper Fly River, the length of l-2 km, 3-10 m width and a depth of 2 annual mean is 4.700 mm. at Lake Bosset in the to 5 m, becoming more shallow towards the lake. Middle Fly area 2,670 mm, and at coastal Daru, The channel itself is being kept open by the 2,100 mm. Rainfall and river flow show limited scouring action of sediment particles in the seasonality. June to October usually are the water. The lakes have a depth of about 1.5 to 2.5 dryest months in the Middle Fly region. m, cut by slightly deeper channels, decreasing to The long-term mean discharge rate of the Ok less than I m towards the end awav from the Tedi at the lunction with the Upper FIy River river. (flow 1178 m3/s at lhunga) was measured at 923 m3/s. The combined flow is 2161 m3/s in the upper Middle Fly. Thcrc exists it n'ateL exchtlnge between the The fish population in the FIy is remarkable for f'loodplarn altd rts rviltet' bodies and the Middle bhe large size of sone species (e'g. black bass Flv River'. The net supply of seclitlent-poor' but and barramundi) and the abundance of endemic DO(l-r'ich f'lor-'clplain rvater to ther Nliddle Fly' species like catfish. The most bargetted species however'. apperlls to be small. At Obo' the long- by commercial fisheries is the barramundi Lotes telln mcirn dischalge is 22'14 trr:t/s. The calcarifer, which roams the entire length of the Stricltlnnd Itivel dischalges llll0 m;lis into the Fly and resides in floodplain water bodies Fll' Riverl rtt, Evelill ,Ittnctiotr (OTI\'IL 199't). during most of its lifecycle. The barramundi N'lurn daill' florv data for the last I'ew years ale miglates annually to the coastal areas for sumllarlzerd rn Trrble {. spawning (Eagle 1993). Watet' lervel lluctttations in the Middle Fly r:rre Fish ecology is characterized by an overlap of smallel thirn in ths ITppel FI-'- River. where species types resident in the two main habitats, rvirter' level chrtnges of up to 15 m have been the river channel and the floodplain with its lccorded at liiungl. The discharge of the Fly waterbodies. Due to the high biologrcal Rivel at its tnouth of 6.000 nrlt/s makes it productivity of the Fly floodplain, the majority comparable tn size rvith the Niger and Zambesi of the food for the Fly River fishes originates in i\fricir ol tltc l)auube in Eulope' from off-river sources (Kare 1992). Overlap in diet and habitat requirements is an important periods Tubl.a .l; llleon d,oi|.y flou' dot.o collected by mechanism for survrval since prolonged drying out of the )'ll\IL lor Lh.e Ok Tedi,, FLY atd of lorv water level may result in (Eagle St ri"chl.a t r.d llr.r'crs o t tl,ifl'erenl floodplain and its shallow water bodies locttl,iorLs itt the coI'clturcril. 1993). Under these conditions, the fish take refuge to the stream channel and oxbow lakes, which may represent the only standing water on 1990 1991/92 199493 Site 1988 1989 the floodplain.

Flow rate (m3/s) The fish populations decline substantially, i however. the recovery usually is rapid with 27 24 27 Bukrumdaing 25 recolonisation of the newly flooded habitat by i Tabubil 175 110 the surviving stocks (Smith & Bakowa f994). The most important food items for fish are Ningerum 305 222 183 193 290 a quatic and terrestrial invertebrates (freshwater Konkonda 965 850 1122 456 805 prawns, mayfly larvae and other insects, worms etc.), algae, plant and organic detritus' 1435 1197 1044 807 941 Kiunga Predatory fish like barramundi and catfieh feed Kuambit 2400 2124 2094 1407 2061 on srnaller fish like Netnatalosa herrings. Obo 2515 2613 2424 2A57 1978 Strickland 3785 3869 3769 2870 3141 3.5 Population Ogwa 6300 5792 5461 In the entire Ok 'ledi/Fly River drainage area Iived about ?3,500 people, according to the 1980 ' no data avatlable nattrnal census. They speak 28 different languages and ftlrm five different language lanrilies: the Ok and Awin people of the Ok Tedi 3.4 Aquatic BiologY ancl Lrpper Fly River. the Marind of the Middle Flv and Lower Strickland (Lake Murray) region, A st.rrdy couducted prior to OTML's opelation the Strki-Gogodala people of the l,ower Fly. and (Robelts I978) carne to the ctlnclusion that t'he thc Tre.rns-Fly peoples of the southern coastal Fly River systetn supports thc rnost diverse {'rsh plarn. regloll, with at least' f'auna rn the Australasian The rnountain people are hunter-hortr'- 105 freshwater species from lJiJ ftrmilies. 'l'l're culturalists living a subsistence lifestyle. Fish is Seprk Itiver rn Northern New (iuinea' the nver ir mote tmportant part of the diet for the lowland clnsest to the size of the Fly on the rsland. has a nverine people. relatrvely low overall fish densrtv rvtth a total o{' 5? freshwatel specres (Allen & (irates 1990)- In the Middle Fh-Lakc Murray regi.on, The rlevelopment of the Ok Tedi Mine ha*s habitable leqd end ground suitable fsr food resulbed in massive cul'tural and eseioeconogrig eul-hvatiqn is s-carcg, TliE irth*bitarrte (sbs'{rt changes in the a{fected region, A f,or:uer 4,500 in 1980) are mainly hunter'gatherers who government adrniser errrphasized the ",XLsycho' eollect wild sago. cooklng-bananas and sugar logical trauma that Ok Tedi developnrent wiltr caqe, and rige t-he fieb reeorree of barranundi, bring to the simple life stlrliee of the bountain black baes and eatfish (Busse 1991). The sale of people" (Pinta 1584 . arocodile eirins is an important source of casl,r. The ndning comllany has established the "L,ower The largest Fopixlations in the Middle Fly arca Ok Tedi/Fly River Developrnent Truet'' whieh are at take Beeset, Lake Pangpa and Lake sffers basic village inftastructufe in the fieltl of Davia,nobu. S-nall villages and homesteads are trsnspef,t, health, educafion and busiqegg ffiically located along lakes aud rivers, developr,re.ut. Ii supports an estimated 30,000 V,iitually all rravel. ing is fu dugout canoe. people in 10[ villages living along the riuer Dcdng extrem-eli dry periods' irr wlliah la-kes syutem, who rnay be negatively affeqted by' slio may drrr o,ut co'mpletelJ" p;eople ab-andon their disehalg""""""""e of mine tes,idugs ,and assoqated viUages and move to the main rirr,ers wh,ich probleme. rer-tai[ the only source of water.

10 4. Input of the Ok Tedi Mine into the Ok Tedi/Fly River System

Following a recent restructuring of ownership, the shareholders of Ok Tedi Mining Limited are 4.1 Description of the Mining Project helcl by Broken Hill Proprietary (51%), a leading Australian mining company, which is also the I\lining of thc r\'lotrnt li'ubilan orebody as ern open project's operator; the Government of Papua pit stzrrted irr 198'1, sixteen 5rsi1I's zrfter the New Guinea has increased its stake from 20 to discover."- ol' thc deposit bv liennecott 30%: and 19% now held by Metall Mining, a explolation gcologists in 1968' 'fhe particular German-Canadian mining company (Milt'irr,g f'c:rtures of'the ot'e deposit were responsible lbr Jourttol. May 6, 1994). the developnrcnt ol' the tnine in mainly two stages: from Nlirl' l()tt4 to late 1988. the gold- cnriched cap was rrrirred and processed. This 4.2 Discharge of Tailings and Waste rnvolved the use' ol' sodiutu cyanide and other Rock process cheurrcals ftl' gold extractiort. Following this stage, the phase of extractttlg copper ore The tailings from the copper extraction in the comurenced. The golcl is now recovered in the ruill, approximately 98% of the original feed to sulfide fl otation concentrate. the processing plant, is piped without further Cr,rrrently, thc' pt'oduction is about 80.000 tons of tleatment to the Ok Mani, a tributary of the Ok ore and a similar volttnte of waste rock per day. Tedi. Waste rock is hauled to erodible dumps The latter is uraterial in which lhe metal adjacent to Mt. Fubilan, from where the content falls below the cub-olT grade of 0.2% fot material is washed into the headwaters of the copper or 0.8 g/tons of gold. The copper cotrtent Ok Tedi. the ore cttllenLly being mined is 0.84% on of The thickened, alkaline tailings slurry consists average and overbttrden contains about 0.11% of approximately 55% solids of which 78o/o have (OTML 199;l). Asstrming a urean recovery rate of a grain size less than 100 pm. The material has 85%, the copper content of tailings is arotrnd about l0-2Oo/o of its original copper content, 0. I 3'.1/o - varying amounts of trace metals ingluding zinc, The ore is crushed and subsequently f'ed ro a Iead and cadmium which occur naturally in the series of ball rnills where the material is ground porphyry ore, and small quantities of organic to fines. A copper-rich concentrale. containrng flotation chemicals. valtrable metals, is recovered by means of a all Waste rock and overburden is coarser, but has a conventional sulfide froth-flotation process. For high percentage of soft material (siltstone and naxunurn sulfide recovery and pyrite depress- limestone) which breaks down easily, containing ion. the pH of the flotation mixture is adjusted copper and significant quantities of other heavy 10.5- I I and organic process with lime to ,o"tul". Although some of the waste rock chemicals ale added (England et al- 1991)' remains temporarily in the dumps and adjacent The final conccrltrate is transported by a 160 km creek valleys (depending on the rainfall slurry prpehnc' to bl"re rlver port of Iiiunga on the activity), most of it is washed down rapidly into Upper Fly ltiver. r\t the l(runga whari the the Ok Tedi River. sltrlry is lilteled. dried and stored to await Since the beginning of mining at Mt. Fubilan, all b1' bulk czrt'riers down the Fly River' shiprrrent and overburden (with the exception facility in the Fly River Delta. waste rock Frorl a storrrge period during the gold stage) has been ls sold urrder long-term of a short the coltccntrilte disposed off in the river system. The to surelters rn Japan' Gerlnany, South contracts construction of a tailings dam was halted in Phrlipprnes- Iiolea. I,-rnlancl and the early 1984 when a land slide forced the In 1992, OTI\'{L pr"oduced 193,400 tons of copper abandonment of the construction site, which was rn concentratcs. coutatmng also l0. l tons of located in a geotechnically highly unstable zone' gold. about Zl-) tons clf silver, and other metzrls. The rnrne' rs the world's fifth brggest copper producer. Olt Tedi's nrrneable resel'vcs at the end of 1992 were estrmated at 4jJl Mt of ore' sufTicient fol another l5 yeerrs ol' operation (Mitirtg ,Jort.t'rutl , Octotrer l. l9$3).

11 The larger part of the mine-derived nraterial The presence of iron and base metal sulfides in entering the upper Ok Tedi has a particle size of the waste rock and tailings indicates the less than 100 pm and can be transported as possibility of developing acid mine drainage suspended load throughout the entire length of (AMD), which arises from the oxidation of the Ok Tedi/Fly River system, given sufficiently sulfide mrnerals and subsequent sulfurrc acrd high hydraulic transport capacity. production. AMD generatron rn the Ok Tedi waste rock dumps was considered a potentially The massive input of mine-derived sediments serious problem in the Ok Tedi Environmental into the Ok Tedi exceeds the sediment transport Study by Maunsell and Partners (1982). Low capacity of the river system, which has led to pH in waste rock drainage may result in severe aggradation of the river and a rise of the enhanced solubility of trace metals like lead, Ok Tedi channel bed by ten meters and more in silver, zinc, cadmium and copper (Ferguson and the upper reaches. Riverbank food gardens and Erickson, 1988). plantations have been flooded and the natural ecology of the river and subsistence fishing has However, acid formation may be buffered by been seriously disrupted. alkalinity released from carbonate minerals in the mme waste, such as calcite (CaCO$. Given About 60 million tons of material are delivered the relatively high calcium content in the waste to the Ok Tedi/Fly River system per year, material, development of AMD in the waste rock containing approximately 69,000 tons of copper. dunrps appears unlikely. Despite the fact that The daily discharge rates are 160,000 tons of sulfide oxidation is occurring in the mine rock material with 190 tons of copper. discharges. no pH values below 7 have been recorded in water of the upper Ok Tedi, neither by OTML nor in measurements during the present study.

12 5. Sedimentology of the Fly River and its Floodplain

Maunsell and Partners (1982) recorded a perrod of very high flow in the Fly River in JuIy 1981, 5.1 Natural Sedimentary Processes in which was the highest since 1977. The water the Floodplain level was about I m above the natural levees. According to their observation, grassland and The Fly River has a highly vztt'iable flow regine forests adjacent to the river channel in the reach where extensive flooding and overbank between Lake Bosset and Obo were flooded to a deposition on its wide floodplain alternates with width of about 16 km on either side of the extremely low water levels in drought periods. channel. Although this situation is highly The sedimentary processes in the Fly River anomalous, it is evident from aerial photographs floodplain are dominated by the river itself and that at high water level in the Fly River, turbid the suspended load it carries. Sediment delivery flood water may flow several kilometers across by floodplain creeks and chaunels to the systern the floodplain. catchment in the outer is negligable. The background deposition rate (i.e. before the adjacent piedmont plain The floodplain and the mine started discharging material) in drowned forested areas with some swarup comprises valley lakes is very low. Sediment fractions less savannah. The content of particulate matter in than 2 were analysed from small islands (0-1 creek waters gencrally is very low lrm floodplain m above mean lake level) in Lake Bosset and (below 15 rng/l), consisting of organic material, Bai Lagoon as seen in Map 2. These indicated a plant and strongly weathered mainly debris, -soil mineral composition dominated by kaolinite, material. alurninium chlorite, quartz and gibbsite. All At mean flow in the Fly River and low rvater four minerals are typical of highly weathered table in the floodplain, intt'usion of t'ivet'iue tropical sediments. Sediments probably from suspended sediment occurs along pre-existing the Pleistocene age were found on the surface of channels linking the liver with the floodplain' these shallow islands, showing that no Fly River Sedimentation is highest in oxborv lakes aud tie material has been deposited there in the last channels, whele medium to coarse silt from the 120,000 years, when the sea level began to drop river suspended lozrd settles dorvn. and mnch and Pleistocene sedimentation in the alluvial lower in drowtred vztllel' laltes and floodplain plain ended. where clay ilnd fine silt tlray become depressrons Table 5 shows the results of Cla age dating deposited. which was performed to determine Drainage direction throughotrt most of the Fly sedimentation rates in drowned valley lakes. River floodplain i.s towards the Fly, although'the The overall deposition rate seems to be well inner floodplain, the erctive rrleauder belt' tends below I rnn/year and may be as low as 0.1 to be highcr in elevation Lhan the ottter mm/year. The sediment deposition at a floodplain due to sedimentation along and in the particular site within a lake depends largely on main river channel. No broad continttotts levee bottom reliefand shape ofthe lake. is developed along the river and where there rs a Sarnple Core 3/30.3. was taken from the distal dam. it rs cttt b1' smzrll drainage channels, end of Lake Daviambu. The upper 13 cm consrst During penods cll' hrgh watel level rn the Fly entirely of peat-like organc debris. River, whrch are assocrated wrth higher Accumulation of this material, derived froru the sediment loads because of tncreased transpot't catchrnent of the lake rtself, is much slowel than energy, rn'er bank overllow and innndatron of at those sites which receive suspended sediment the alluvral plarn occuls. Overbank floodrng flom the Fly River. floods is locahzecl close to the during rnoderate With the other three cores, the sediment layer channel rn flirnhing srvaml:s. 'l'he Fly River ebove the peat was made up of Fly River' to ther rivcr rvhctr the flood water returns matertal, as detet'mined from its characteristic recedes. nlthough urost of the snspcrncled load cltrl ruineral assemblage with dominantly (low t.he floodplain lvill be deposrted carried into charged) smectite. A thin layer of copper-rich there due to vegetative lilterurg in the gt'assland mlrteual on the top of the cores was observecl' bordering the uratn river, ilud iu seclttucnt flitps such as lakes.

13 Map 2: Tlte lower port of llrc Middle FIy Riter (stu,dl oreo).

n\u\ , o Lake Agu o -4,

lake Hurray

\ Bosset

Kemea Lagoon

f..t',4 r.'. ,', t dkc K ibuz

*tr{

14 Toble 5: Resllrs of racliocarbotr age dating of sedinterfi cares front' drown'ed ualley lo,hes-

Core Number 3'30.3. 3111.1. 115.4. 1'9.4.

Locality Units L.DaviambuW L. EossetNW BaiL.C KaiL.SW

28 Depth of peaty layer cm 13 26 27 below sediment surface 4380 C14 age years before present 5265 2950 4030 ye?rs t215 t70 t110 *125

Sedimentation rate mm/1000 years 249567M

Water depth at site mean water level 1.10 2.40 1.70 2.60

Distance from Fly air km 6539

Due to the lack of dateable organic matter in Flow inversion in channele linking the sediment cores frorn tie channels and their floodplain with the Fly River is an inportant banks, where sedimentation certainly was much process because it is responeible for suspended higher, no data are available to establish sediment transport to off-river sites during deposition rates at these sites. The simple fact mean flow conditions. Reverse flow upetream that the shallow lakes still exist after about the tributary channels was frequently obeerved 5,000 years of possible sedimentation from the durrng the three field trips undertaken' FIy River points to very low deposition rates. On April 8, 1993 slightly turbid water (sample channel depth of Assuming that the maximum W1/8.4.) rvith elevated calcium, copper (9 FSil) channel 5.4 meters, measured in Lake Bosset and cadmium (0.12 pgll) levels was encountered close to the Fly River, reflects the original in the Agu River/Lake system about 25 km channel depth of the drowned tributary, then upstream of the junction with the Fly River, the deposition rate was about I mm/year in the which had about mean flow. The Agu River/ last 5,000 years at sites close to the Fly. Lake in this reach runs roughly parallel to the At times when discharge from the Strickland Fly. As a result of a recent Fly River intrusion River is greater than the flow of the Fly' a upstream the Agu River, turbid water was backwater effect develops, which results in visible in the densely vegetated parts of Agu decreasing river velocity in the Mrddle Fly nnd Lake. whereas in the Agu main channel, Fly subsequent settling of suspended material on rvater was slowly pushed out in southerly the river bed. The sarne occurs during low-tlow direction (downstream) due to heavy rainfall in periods. The deposited material rnay be the days before the site was visited. resuspended at hish flows. A sirnilar observation was made in the Kai Pickup et al. (19?9) estimated the natural Itiver/Lake -system rvhich was sampled on April suspended load of the Middle Fly at 7-10 niilbon 9. 1993. nlthough in this case Fly River water tons and bed lorrd at l-2 rnillion tons per 1'car'. was still flowing ttpstreatn. The upstream cllrl'el1t observed close to the Iiai River mouth w'as rt,uralkablv strong and tnade the use of the 5.2 Deposition of Mine-Derived outborrrd notol: nccessilry for passage- Material Water wrth hrgh condrtctrvtty, neutral pH and elcvirtcd copper valttes (39 pg/l in unfiltered The strspended sediment conccnlraLtons rn tl're saruplc W4/9.,1.) was encountered 19 channel krn Middle Fly Rrver today are 5 - 10 tttttes above upstr(,an1 of the Iiai ltiver confluence with the the natural load of about 60 rng/I. Srgnificant Fl-\, Il.rver'. lVater ntovenlent at the sampling quantitities of' mine-derived sedimcnts al'e Iocutrtrn was in northcrl,y direction (upstream)- cleposrted and trapped in creeks, [3lis-s rind ( )ne c

15 However, satellite images and aer-ial Sediruentation of mrne-derived material was photographs show no such connecting channel highest in oxbow lakes with up to Z0 cm and give no indication of flow across the measured at sites close to the river end of Lake floodplain, away from the river. The same Pangua rrnd Lake Ilibuz. Twenty-nine oxbow observation was made for the Agu River system, lakes at different infilling stages exist along the although maps show a northerly channel Middle Fly River. connection of the Agu with the Fly River. Two sediment According to local villagers, this flow-through cores were taken from the Fly River (see mechanism is only active at very high water channel bed Figure 2a-c), one at Lake level in the Fly. Pangua (water depth 10.5 m) and the other downslream of Obo (depth l4.b m), both in the The frequent intrusions of highly turbid Fly deepest part in the channel cross section. The River water should result in widespread entire bed core from the Lake Pangua site (b6 sedimentation of mine-derived material in the cm) consisted of fine gr.ained sediment with bb- Kai and Agu Lakes. Sediment sample analysis ?5% finer than 20 pm (clay, fine and medium revealed moderate copper pollution. Only the silt fraction), and displayed a constantly high uppermost 3-5 cm of sediment cores showed copper content of about 850 ppm. strongly elevated trace metal contents, with Cu The upper 30-35 around 300-600 ppm. In the Kai River samples, cm of the core taken close to Obo were very the contaminated sections consisted of deposited similar in composition, however the bottom (4I.48 riverine suspended matter- In the Agu River section cm) showed much coarser system, the uppermost, copper-rich sediments material with 7b% fine sand (69-200 were mainly made up of humic material in Fm). The copper content dropped to 48 ppm in this section. From both cores, which the high copper content may be it is evident that suspended load has secondary, due to adsorption on flocs of organic settled down on the channel matter. floor. Only the Deposition of rnine-derived sediments generally section with mainly fine sand, from the sample taken downstream is highest at locations close to the channels of Obo, shows the typical grain size which connect floodplain water bodies with the of bed load. It is known from earlier sedimentological Fly River. Comparatively coarse material studies (Pickup et al. 1979) (medium to coarse silt) is deposited there, that fine material does temporarily deposit on whereas the remaining fractions of clay and line the river bed upstream of Everill Junction. Due to silt of the riverine suspended load is carried into the cohesive forces among fine grained particles, high current speeds are the floodplain water courses shown in Figure 1 necessary Trace metals are generally enriched in the finest to erode these deposits. The particle fractions. resuspension of mine-derived clay- and silt-size material from the channel floor today may be In a sediment core from the Kai River channel limited because much more sediment settles bank (U10.4.), tahen approximately B km down, and will be flushed downstream at high upstream of the Fly River confluence, the upper flows only at hydraulically preferred sites like 38 cm showed strongly elevated copper values meander bends and immediately upstream of (400-700 ppm). This matenal must have been the Strickland j unction. deposited within the last ten years, since the Ok In connection, Tedi mine is operating. The sedimentation rate this it is interesting to note that rs about 4 cm/year. two islands within the Fly River channel, one immediately downstream of Lake Bosset and the Similar observations were made at lhe other other at the lunctron of Tarnu Creek, have been drowned valley lakes investigated (Bosset, observed during the field trips undertaken. Kongun, Kemea, Bar, Daviambu). Where the Sediment samples taken showed relatively well defined channels extend into the lake. coarse, copper-nch material (see Figure 2d). No substantial deposition of 2O-40 cm of copper- such island,s are visible rn aenal photographs contaminated sediment was detected several from the lg60's and lg?0's. The formation of kilometers away from the Fly River, whereas rn rslands rn the river channel also indicates the proper lake copper-rich sedrmentation did insufficrerrt transport capacity of the Fly River not exceed 5 cm. to carrJ- all mine-imposed waste material. OTML (1993) reports Sedirnent core 4/11.4. showed elevated trace an rncrease in bed level at ntetal levels in the upper Kuambrl. immedrately downstream of the Ok 23 cm with about 200 'fedi/t-ly ppm copper at the bottom and 800 ppm at the lliver confluence. of more than one metel'above 1984 pre-mine top section This core was taken in Lake Bosset the baseline. at the location where channel and Iake water mc-rge. about 2 krn rnside the lake (see Map 2).

16 ('ore Figtp'a I : ()ottr Jtrtrt.so,t s al sed i.rrrtrtt groi,rr. siza for f'ort.r' satn.ples t'ront uarious sites ht the I I i d d la FIt Rircr.s.t'.src,,r.

t c o

U o Iry c I 6- A E E c) a6-:= n OE() --eD- 'E i( 'n oo t HRi (oc ! lr c') aP: (/)+3S 4.e J c 5o ; N o = a\mv \/ '"H A og -- o€ S ^s :i m> (!a.! f gr : (I:q c.rK E EO ::.! i J- .!t'6 E =>g$r e a '= d- o ja 5 Fc; -.=:,1U E ct -.F E YOt (o 96 F (!r Nu ,== .9 c!s N9 v? d o .? I6 o F-@!noooooooo ra (I) rf) rf co (\t E o/o + or'o ,4 .t

n tJ 5 0 q 6 E ; 6 ob otEo! AIN\ o3 5iV oE t* oo =\v o-L o :L- E :E :- 3E G)-€ ,(J 6- E .\(J L Ee S € Y=e9 'd -or- C\,l 6^; c - )(C\l ovo ory d';: d 96 "R€ I E H€ ' ^ tor E ctN o, N I5 :'>L-: 9NO (trq @'rn € (D$ d'6 E :<3 -=-D E E .>cci .t(U ,g cE@ ,o-b' IS E olF ! NE:f E u v, ':E-4 '\. org c v'5 ! 6 .EE i no a rf(rrNooooo d l\(Onl'c')C! o E al o/ tO ci -i rA 'a

17 Figure 2: Comparisotts of sedinent grailt size for fotr corut soltples frorn uari.ous sr,les ht the Middle Fly Riuer systetn.

'a i 9v *.= X.EO=r* +E niP Od E() aE o f, tr -Os \:+vRE:r --=F (o_ o6 9= s Ee _'= t OY-(f) E b FO aY s (l); o.9 e o.o G -\t N:'i c .C) tf, 9E I c00 Ho r-) P.E ? E= (Yar= 9il !, E: eor..! t? i=P qF^H CE E .e-- o'E .9 :.(f, \rY:4r- U) .= n\ rdFi = tL N :;E .- OF? fl K.ET oG VE ;tL I o t;- [J o oc *; rnt- ta

6 o, t G o o I J o 6 (u oc! { N 0 f d E o) o c o E 6r ()Eo- o E !Oxtr q{1 d cltr C\ h qc o-6 o Oro- 9.- ar (0 E -N I _v9 ,c tr oc.{ 4 q rn- o o g ^o (J E.o- F= cO (U : sro -J-ro o 6 r-(\l @ol q E> o E 9= .N I .N .= >- a .! a I ;+N o- E Ee (oo E *tc; (o -E r-J '6 t A(U G (D t).r cr) N (', Xtst o ='d lI c') 6 .>N -\ tr E. o -> N r o, LL o I i ! t cal

\o vi t v)

18 Trvo ef'l'ects probably are responsible for the and Obo shows a strong decrease in all sampling in strorrgl5' inclellsed deposition rates as compared periods. This effect can not be seen clearly to earlier sedintentation: filstly, because of the the depth-integrated measurements in which increase in the (mine-derived) suspended the suspended matter content is determined in sedirnent load of the Flv River. which today also vertical sections through the water column. contains rnore silt than the natural particulate Typically, the suspended sediment concentr' load. Secondlv, wrth redr'tced channel capacity ation of near-surface rlver water (gulp samples) due to bed aggradation, overbank flow frequency is ?0-90% of the depth-integrated euspended incleases. In the lower part of the Middle Fly. content (Higgins 1990). No such junction, matter close to the Stricliland River relationship can be established in the data of aggladation of the tnain river channel is higher Table 6. Differences between suspended sedi- dne to the bacliwater effect fi'our the Stricklarld. ment content determined by gulp and depth- in Sedimentaiiot't studies which were ttndertaken integrated sampling were up to 9Ao/o date before the Oli'Iedi mine was developed (Pickup measurements from the same site and et al. l9?9) assumed that frequent landslides, (data in OTML 1993). particularly in the upper Ok Tecli catchment, It is difficult to explain this discrepancy, which have lead to a high trattual suspended sedin'rent is particularly evident at Obo' There obviously load in the FIv River. Had this been the case' exists a pronounced stratification within the this should have resulted in much higher flowing water body, with particulate-rich water Fly natural deposrtion rates in the Middle travelling close to the bottom. Due to the shift of floodplain. Tl-re courparison of sedimentrrtion the main current in river bends, which leads to a rates before and after the urining at Mt. Fubilan more turbulent (spiral) flow, the bott'om strata started points to the fact that the backgrotrnd with a higher suspended solids content rise to suspended sediment concentration in the Fly the surface. This effect was observed in the River was very low. probably in the range of'40' depth-integrated measurements at Obo where 60 rng/l. suspended sediment concentrations are usually Attenpts were made to establish a suspended highest close to the outer bend. sediment balance using data collected by OTML' The stratification in suspended sediment A proper balalnce would show sediment losses concentration within the water body points to (or gains) in the systeur' A brief examination of the settling down of suspended load on the large data in Table 6 shows that there is a channel floor. River bed sedimentation is largely scatter in the measured concentrations over dependent on flow velocity, which at Obo is several sampling periods. significantly lower than at KuambiUNukumba' gathered in Of main interest to the present sttrdy is the A detailed comparison of flow data OTML gave a deposition of suspended riverine material in the February and April of 1992 by Kuambit and 0'66 Midcile Fly region. The measured suspencled mean value of 1'12 m/s for natter content in gulp samples of near-surface m/s flow velocity at Obo. water in the FIy River reach between I(uambit (surface) an'd depth' To,bl.e 6. Auerage d.o.ta otv sttspend.ed. seclJ,truent corrcen trotiorts irv gulp ;rlegraticl (DI) samptis. Med.ians shown and rtttrnber of nteasurem'ents in' brackets' Dut.a fron OTML (1988-92)'

1988 9'88-8'90 10,91'9'92 ss/gulP (mg/l)

64 Bukrumdaing . 151(3) 44(5) 57 10626 Tabubil 4709 7825(251 66e6(12) 8095 5882 Ningerum 1802 524(124) 3477(121 5846 nM Konkonda 6e8 1420124\ 949(12) 2072 . 60(24) 26(12) 186 149 Kiunga utl Kuambit 305 520(24) 316(12) ut: Bosset . 171(23]t 1 18(12) Obo 65 e2(25) 103(12) 414(24\ 239(12) Stnckland ' * Ogwa 326 2S3(23)

* nodata available

19 Ok Tedi Mining Ltd. has tried to establish an In the data set for the following year, the annual suspended sediment mass balance based discrepancy between observed and predicted on its regular measurements taken in the Ok Ioad is rninor. Given that sedirnent input by the Tedi and the Fly River (see Table 7). The Ok Tedi mine into the system is fairly constant calculations illustrate the complexicity of the over the years. the massive adjustment of the system and the difficulties in predicting predictive model between the two years (at Obo sediment transport in the two rivers. The 38 instead of 55 Mt/a) is hard to understand. measured load values (calculated from a small The flow weighted load data for 1992/93 suggest number of measurements) in the 1991/92 data heavy erosion of material in the Middle Fly indicate a loss of sediment frorn Ningerum to between Iiuambit and Obo (increase of Obo (46 to 35 Mt/a), whilst the model predicts a suspended load by 4 Mt/a), which is hard to steady increase in the same reach (43 to 5l-: explain, too. This may be attributed to the Mt/a). limited number of data collected (9 depth- rntegrated sarnples).

Table 7. Suspen"ded sedirnent tnass balances for 1991/92 (aboue) and 1992/93 (below) co,lutloted by OTML (1993, t994).

Station Flow Weighted Flow Weighted Predicted Load Predicted Conc. Load (MUa) Conc. (mg/l) (MUa) (mg/t)

Ningerum 46 5750 43 5991

Konkonda 40 2001 43 1862

Kuambit 37 646 M 724

Obo 35' 450' 55 687

Strickland R. g4* 900. B5

Ogwa 1 19' 323- 140' 807

Station Flow Weighted Predicted Load Load (MUa) (MUa)

Ningerum 40

Konkonda 31 35

Nukumba/ 34 38

Kumabit

Obo 38 38

- insufficient records: estimates based on available data

20 6. Potential Mobilization of heavy metals and hydrochemistry

populations, i.e. sediments which were deposited before the Ok Tedi mine started discharging 6.1 Heavy Metals in Sediments of the residues into the Ok Tedi/Fly River system Ok Tedi/Fly River SYstem (Table 8), and sedimente controlled by mine- derived material (Table 10). This classification is Analytical resnlts for sediments from the Fly easily practicable due to the distinct River floodplain are shown in Tables 8 and 10 geochemical signature in the mine-derived (At and A3 in annex) and from the upper Ok material, which shows for some elements a Tedi in Table 9 (A2 in annex). The samples from significant enrichment above the natural the FIy River section are grouped into two background. 27 Ta.ble 8. Meatt, stand.ard deuiatiott, med.ian ualues and 25% - 75% confidence iruteruals for (rt eLetnerr.ts in Mid.d.Ie Fly Riuer backgrowtd sedimerfi sanples = 128).

Element Units Mean SD Median Percentiles 25 75

I cd ms/Ks 0.26 0.13 0.2 0.2 0.2 ] 8.5 Au pg/kg 6.7 6.5 3.0 1.0 0.3 Ag mg/kg 0.24 0.26 0.2 0.1 5.0 As mg/kg 4.6 3.6 4.0 2.0 70 Cr mg/kg 61 16 ol 55 2.0 Mo mg/kg 1.7 1.4 1.0 1.0

Co mg/kg 13 6 12 916 41 Ni mg/kg 33 13 29 24 196 V mg/kg 165 49 171 143 4400 Ti mg/kg 3765 1269 41 00 3100 83 Zr mg/kg 67 25 67 49 28 La mg/kg 247 25 20 361 Ba mg/kg 309 119 309 255 162 Sr mg/kg 137 57 127 101 Qn mg/kg 't7 { 18 14 20 ok 6.7 9.4 2.6 1.3 6.0 q mg/kg 1447 1445 800 300 2525

21 Gold gives the highest enrichment factor of 53, The comparison of mine-derived sediments from followed by molybdenum (factor 23), copper the Fly River floodplain (Table f0) with material (factor 12), lead (factor 4.4), calcium (factor 4.1), from the upper Ok Tedi, which consists nearly silver (factor 3.5), sulfur (factor 2.6) and exclusively of tailings and waste rock (Table 9), strontium (factor 2.5).Zinc shows an enrichment gives the following results: Calcium is found in factor of 1.6. upper Ok Tedi material 4 times and sulfur 3.8 times higher as compared mine- The elements Al>V>Co>Cr>Sc>Ni>Ti>Zr (in to lowland controlled sediments. decreasing order) are present in mine-derived material in lower concentrations than in Factors for other important elements which are background sediments. It is evident that the found in the Mount Fubilan orebody are: copper elements associated with the copper-gold ore (2.2), silver and cadmium (both 2), zinc and from Mount Fubilan are also found in the strontium (1.8), gold (1.7), arsenic and lowland depositional sites. Those metals which molybdenum (1.6) and lead (1.5). The elements are typically enriched in soils during tropical K>Ba>Mg>La>V>Ti>Ni>Sc>Zr (in decreasing weathering are found in higher concentrations order) are present in tailings and waste rock in in unpolluted lowland sedi-ments as in material lower concentrations than in mine-derived from the mine, deposited in the floodplain. sediments of the Middle FIy region.

Table 9: Mean, standard deuiation, mediant ualues and 25% - 75% confiderrce itderuals for 27 elements in sedimer*s from the upper Oh Ted,i, mairtly tailings antd waste roch (n : 24).

Element Unit Mean SD Median Percentiles 25 75

Na mg/kg 11921 4159 10950 8100 15000 K mg/kg 30350 7395 27950 24400 37200 Mg mg/kg 6733 1463 6350 5700 7700

Ca mg/tg 86117 33796 98450 54800 1 15900 AI mg/kg 61462 11766 62750 54800 65800 Fe mg/tg 47512 29854 40900 29300 54200 Mn mg/kg 700 323 775 436 965 Zn mg/kg 541 450 378 182 779 Cu mg/kg 1523 976 1158 805 1791 Pb mg/kg 463 744 123 63 488 cd mg/kg 1.6 1.3 1.0 0.6 2.2 Au pg/kg 426 381 266 107 609 Ag mg/kg 2.4 2.5 1.4 0.8 3.3 As mg/kg 14.9 11.5 11 519 Cr mg/kg 30 12 26 21 33 Mo mg/kg 38 14 36 27 45 Co mg/kg 13 10 11 5 16 Ni mg/kg 179 14 10 23 V mg/kg 111 21 105 99 120 Ti mg/kg 1788 305 1750 1500 1900 Zr mg/kg 22 11 19 13 29 La mglkg zJo 22 19 26 Ba mg/kg 410 202 368 275 599 Sr mg/kg s76 86 548 510 620 Sc mg/kg 7.3 2.4 I 59 ?/o 1.8 0.5 1.9 1,3 2.1 S mg/kg 15211 20643 7900 4425 10625

22 Evaluation of relationshrps between elements in the case of sulfur, the positive relationehip is the data obtained for the upper Ok Tedi lost or becomes weaker in the Fly River sediments give the result that sulfur versus Fe, floodplain data for the same metals. It is Mn, Zn, Cu, Pb, Cd, AS and Mo is strongly interesting to note that gold showe no positively correlated (r = 0.80-0.99, signifrcance relationship with any of the trace metals level p <0.05). This can be explained by the fact mentioned above in the upper Ok Tedi that most metals are dicharged in sulfidic form sediments. [n the mine-affected lowland into the Ok Tedi River. sediments exists a positive correlation with lead (r 0.68) and molybdenum (r = 0.66). In the Fly River floodplain sediments, the = positive relationship between trace metals and There are mainly two processes operating in the sulfur is much weaker (in the range of r = 0.31- river system which are responsible for the 0.?l for the metals mentioned above) with the differences in the element concentrations in exception of iron, which shows no correlation upper Ok Tedi and lowland eediments: with sulfur in lowland mine sediments. Sediment admixture/erosion and mobilization of metals from the solid phase. The Ok Tedi River upper Ok Tedi, the In sediments from the on its 200 km long way to the FIy River junction Cu, Cd, Ag and Mo are metals Fe, Mn, Zn, Pb, receives water and suspendgfl ssdimsnts &om (r 0.65-0.95). As in all highly intercorrelated = several tributaries.

Table l0:Meant, standard deuiation, medio,tL ualues and 25% - 75% con'fid'en'ce hfieruals far 27 elernents in, mine-corttrolled sedinents from the Middle Fly Riuer floodploin (tr' = 197).

Element Units Mean SD Median Percentiles 2s 75

Na mg/kg 8010 3218 8200 6025 9800 K mg/kg 28807 11610 30300 18850 36950 Mg mg/kg 7242 1831 7400 6200 8575 Ca mg/kg 30139 23842 24800 8200 50350 AI mg/kg 80994 14915 81 100 72325 91025 Fe mg/kg 40004 11012 39500 32200 46850 Mn mg/kg 528 254 510 310 710 Zn mg/kg 211 68 204 163 245 Cu mg/kg 530 289 529 278 732 Pb mg/kg 79 36 80 48 104 f-d mg/kg 0.5 0.3 0.5 0.2 0.7 Au pg/kg 166 81 160 120 205 Ag mg/kg 0.7 0.5 0.7 0.3 1.0 AS mg/kg 8.3 5.8 7.0 4.0 12 ?t mg/kg 48 14 46 39 56 Mo mg/kg 24 13 23 13 32 'l mgftg 105 10 I 12 Ni mg/kg 21 I 19 15 24 V mg/kg 145 35 139 123 163 Ti mg/kg 2753 801 2600 2200 3200 Zr mg/kg 44 16 41 32 52 LA mg/kg 267 27' 22 30 Ba mg/kg 415 100 420 363 466 Sr mgftg 330 139 3'13 226 438 Sc mg/kg 13 4 12 10 15 o/o 1.6 1.5 0.9 0.6 2.0 b mg/kg 2464 2160 2100 675 3425

23 Further dilution with uncontaminated riverine Table 11 shows analytical data for two sediment particulate matter occurs at the Ok Tedi/Upper cores sampled from locations close to the Fly Fly River confluence. Due to the fact that River channel (Kai River channel. l/10.4.. and unpolluted tributaries have a much lower Fly River bank at Obo, 5127.8.), whereas Table suspended load than the Ok Tedi, dilution 12 shows data from swampy sites (locations in effects, however, are small. Lateral erosion in Table A3). the highly sinuous channel of the Middle Fly Calcium and sulfur values appears to be a more important process. are much higher than in the cores shown in Admixture of weathered floodplain sediment in Table 12. Very little calcite dissolution and sulfide the river course is responsible for increasing oxidation has taken place in the material concentrations of elements like zirconium, from the river channel. The declining copper values scandium, titanium and chromium (which are towards the base of the sediment core 1/10.4. highly persistent in chemical weathering) in the reflect the development mine-derived sediments stages of the mine- [n the early mining phase of deposited downstream gold in the Middle Fly floodplain. extraction only, tailings and waste rock consisted mainly of oxidized gossan rnaterial A sediment particle which is discharged into the with a relatively low copper content and almost upper Ok Tedi travels about 7 days until it no sulfide minerals, which can also be seen in reaches Everill Junction 610 km downstream of the base section of core 1i 10.4. It was frequently the discharge point. During this time, mineral observed that copper values increased in steps dissolution occurs. It is evident that particulate, from the base to the top of a sediment core mine-derived calcite is being dissolved in Ok whereas gold showed the opposite behaviour Tedi and FIy River water. Particulate sulfide (Table A3 in annex). minerals, mainly pyrite and chalcopyrite, are Sediments of Table 12 display also unstable in the oxygenated river waters and very low calcium and sulfur values, are oxidized to sulfates. The trace metals although high lead and gold contents indicate associated to these minerals either go in solution that it is largely mine-derived material. Gold very or become adsorbed to particulate matter, is a resistant element in mainly to unsoluble tropical weathering, and lead also shows very iron oxyhydrates or organic mobility matter. They are partitioned between the solid Iittle in the Ok Tedi/Fly River environment (see and dissolved phase according to their following section), Copper, together with calcium geochemical mobility in the aquatic and sulfur, is clearly depleted in the sediments en'rironment of the Ok Tedi/Flv River svstem. in Table 12. See also Figures 3 and 4.

Table 11. Typical uertical profiks of rnine-controlled seclirnerils (fractiott <20 pnt) ilr, the Mid.d.Ie Fly Riuer floodplafu depositcd o.t sites close to the riuer at higtt deposition rates.

Sediment core 1,10.4. CaCuPbS sec{ion ppm ppm ppm ppm

0-2 cm 35400 663 51 2300 24cm 44300 668 54 1700 '18-20 cm 50000 664 98 nd 28-30 cm 56800 618 78 2700 3&32.4 cm 50200 574 94 nd 32.4-34.5cm 35200 483 83 <100

Sediment core 5/27.3. CaCuPbAuS section ppm ppm ppm ppb ppm

19.5-22cm 48700 977 76 205 5300 30-32 cm 56000 929 79 nd 3000 4648 cm 47800 891 71 nd 4600

24 Ca and S remain in the sediment body when the To the contrary, mine-derived material deposition rate is high and each layer of riverine deposited sporadically on swampy floodplain suspended matter is rapidly covered by fresh sites is subject to intense leaching. The sediment. Sulfide minerals are stable under undulating water table permits penetration of conditions of oxygen deficiency, which are atmospheric oxygen into the ss,limsnf fedy, generated by the decay of riverine organic which is faciltated by the roots of floodplain rnatter. The pore water in the fine-grained vegetation. Both factore result in sulfide sediment will soon be saturated with calcium oxidation. Rainwater leaches calcite from the which prevents further calcite dissolution. deposited material, and the rotting of swamp Mobilization of trace metals will be minimal vegetation generates acidic pore watere which under these conditions. may bring trace metals in solution.

Toble I2:Vertical profiLes of rruhrc-corttrolled seditnerfis (fraction.40 pm) deposited at swatnpy sites ort th,e Middle Fly River floodplain.

Sediment corc 417.4. Sediment corc427.3. CaCuPbAuS CaCuPbAuS section ppm ppm ppm ppb ppm section ppm ppm ppm ppb ppm

0-2.5 cm 7800 583 92 1s0 600 3.5-5.5 cm 6100 430 110 240 <100 2.5-5 cm 6600 576 122 160 300 5.5-8 cm 6900 320 81 nd 100 7-9 cm 4200 557 140 nd nd 30-32.5 cm 5900 40 19 nd 100

1 9-21.5 cm 6400 37 16 <2 <100 Sediment core 1/30.3. Sediment core1n.4. Ca Cu Pb s CaCuPbAuS section ppm ppm ppm ppm section ppm ppm ppm ppb ppm 2-4cm 7300 409 81 100 7.5-10 cm 8700 40't 102 310 200 26-28cm 5800 35 14 400 27-29 cm 14000 356 79 291 500 33-35 cm 7500 69 29 nd 200

Figure 3: Copper distribution of alhnio,l, seditnents itt. th,e Midd,le FIy Riuer floodpain. The probability graph of th,e composite population, af 385 data poirtts (opett. squares) seporates i.rr.to approrinately log-norrnal su,bpopu,Iations (closed squares) with a, natura,l bockgroutd' of a,bou,t 50 ppttt Cu, attd, a, second popula.tiott of mhrc-d,eriued material ot 610 (geometric ttttttt.tts). l l 99.50 99.00 .]- 98.00 '''-q -.1 G- gs.oo - Pop. 2: o\ : f = 510 ppm so,oo I -l t 1d= 42G880PPm () + 5 Eo.oo J e 7o.oo (l) 60.00 :j ri so.oo = O t10.00 j= .= 3o.oo (g l Pop.1: 20.00 f=60ppm =tr j+ 5 lo.oo I 1cr= 3G120 ppm cJ I: s.oo -l

2.00 --..1 -.l L00 0.50 =

100 Cu (ppm)

25 tr'ittwre *Gold dictrf,butian of alluaioll, sedirnents iu, th,e Middle FIy Riuer ftoadpain. \he probabitity graph of the composite population of 2f datu poi,nts (open squpres) seporates inta appraxhnateX! rc-g,rtartnal euhpopalatians (clased. squares) witlt, a n:e;tural backgraund of oboat 7 ppb Au, ond, a second populntion of ninn-d,eriued, materi,at at ppb.

9E.00 Fly River 95.00 Fop.2: 90.00 f = 17Q ppb s t 1d= 12S250 ppb 80.00 (J oC 70.00 60.00 ET=, E 50.00 ]L o f0.00 .= 30"00 $ E E Pop.'1,: 5 (J x= Tppb t 1cr= 2.21 pgb 5.00

a00 1.m

Au (ppb)

26 6.2.1.3 Trace metals 6.2 Hydrochemistry of the Ok Tedi/Fly Copper, zinc, molybdenum, cadmium and lead River System were analysed in waters. Between pH 6.5 and 8.5, lead speciation in the aquatic environment is dominated by carbonates 6.2.1 General and hydroxides (Hem and Durum, 1973), which are almost insoluble. Pb may be complexed with organic ligands, yielding soluble, colloidal and 6.2.1.1 Earth alkaline and alkaline metals particulate compounds. It can easily be adsorbed to particulate organic and inorganic matter, These elements have a high solubility of their which the dominant mechanism controlling is independent of redox is salts in comnron. which the distribution of lead in the aquatic changes and partly independent of pH (Na, It). environment above pH 6 Garrah and Pickering Because of the reaction: r977). + <+ + 2 CaCO" HzCO" Caz+ HCO,r' Copper generally shows a higher geochemical calcite dissolution re.quires the presence of mobility than lead. The distribution of copper carbon dioxide, which forms the carbonic acid species in the aquatic environment depends on consumed in the reaction. The carbon dioxide pH and on the presence ofinorganic and organic partial pressure in water decreases with ligands. Carbonates and hydroxy-anions are the increasing ternperatures. Dissolved Ca, Mg and favoured inorganic complexing ligands in Sr compounds are stable in water in the oxidized freshwaters with pH above ?. In the plesence of carbon dioxide. presence of soluble organic matter, sorption of copper to particulates may be relatively Calcite is abundant in the mine waste ineffective because complexation with humic Its discharged into the Ok TedilFly River. acids is the dominant process (Jackson and exerts a buffering effect on pH which dissolution Skippen 1978). will remain in t,he range of about 6.4 to 8.0, in spite of concoruitant dissolution of sulfides from Cadmium and zinc are characterized by a mine discharge. Calcium forms complexes with lower affinity to carbonate and hydroxy anions humic acids^ Mg compounds are, in general, at near neutral pH values. The influence of pH more solnble than their Ca counterparts. Na is and alkalinity on the solubility of Cd2+ and Zn2+ the most mobile metal in the aquatic is much lower as compared to lead and copper. environment. I( is usually present in lower Zinc also forms complexes with humic concentrations than Na. The natttral leaching of substances, a process which is favoured by I( from minerals does not result in significantly increasing pH. increased solubilization because K is usually Molvbdenum occurs in oxidized waters as incorporated into new mineral readily molybdat" 6[o042-) and bimolybdate (HMoO4-) structures like clavs. anions. The anions are readily adsorbed by iron and aluminium oxyhydroxides at pH values 6.2.1.2 lron and manganese below 5. Above this value, molybdenum in natural waters is essentially dissolved. Both rnetals show a similar hydrochemical behaviour. Manganese tends to be more mobile Alurninium in the aquatic environment is as compared to iron. mainly present in the form of dissolved and colloidal aluminium hydroxide, AI(OH)e' Its Compounds of I'erric iron 1Fe3*) and Mn4+, minimum solubility is at pH 5.5-6. Above pH which are the stable oxidation states in aerobic 6.5. soluble aluminium exists primarily as waters, waters, are nearly tnsoluble. In reducing is capable of forming complex ions (ferrous/manganous) al(OH)+- It the divalent reduced forms with inorganic and organic substances. can persist in the absence of sulfide and carbonate anions. Iron and manganese may also occur in both oxidation states in inorganic or 6.2-1.4 PH alkaline organic complexes. In oxygenated, The pH value in the Ok TediiTly River system present colloidal waters both metals are as is not only controlled by the dissolved carbon suspensions of ferric resp. manganic hydroxide dioxide concentration in waters and calcite particles which pass through a 0.45 Fm dissolution, but also by the biochemical the membrane filter- For this reason, "dissolved" processes of photosynthesis and respiration. concentrations reported for iron and manganese rn neutral or alkaline waters in most cases do not correspond to electrolyte solutions.

27 Photosynthesis of subaquatic plants is Only the elements C, N, O, S, Fe and Mn are accompanied by the assimilation of ions such as important participants in redox processes in the NOJ-, NH4+ and HPO.2-. Charge balance is aquatic environnent. However, since the maintained by the uptake or release of H+ or mobility of trace metals is influenced by the OH', which lead to alkalinity changes. adsorption to Fe and Mn oxides and fixation in reduced species (e.g. Alkalinity increases in oxygenated waters as a formation of metal sulfides), redox result of photosynthetic nitrate assimilation conditions are of importance to predict trace metal behaviour. according to the bulk reaction (Stumm and Morgan 1981): 6.2.1.6 Disso/ved organic (DOC) 106 COz + 16 NO3- + HpO42- + 122H20 + 18FI+ carbon The term DOC is a sum parameter <+ Clo6H26ootroNtaPr (mean algae for a number of polymeric organic substances which contain a composition) + 138 Oz sufficient number of hydrophile functional (-COO, -NHz,R2NH, -RS-, NH4+ is the dominant nitrogen species in sroups ROH, RO-) to reducing waters. NH*+ assimilation remain in solution despite their molecular size. during Polypeptides, photosynthesis causes a decrease in pH: amino acids, certain lipids, polysaccharides, humic and fulvic acids and 106 COz + 16 NHn+ + HPO42- + 108 HzO Gelbstoffe belong to this group. Humic substances in general are result c+ cl06H2caolroNrapr + 107 02 + 14 H+ a of the transformation of biogenic material. DOC levels Respiration of plant material leads to reactions in interstitial waters of deposited organic debris in opposite direction. When photosynthesis and may be much higher than in the overlying respiration are in overall equilibrium, no change water. in pH will be observed, although there exists a Humic acid is extracted from pH variabfity over the day/night cycle. humic matter in When alkaline solution. Fulvic acid the rate of production of organic matter is the humic fraction that remains in acidified solution (assimilation of NH4l is larger than the rate of and is soluble over the entire pH range. decomposition, alkalinity will decrease. peat Humin is the fraction which cannot be extracted formation, which was frequently observed in the by either acids or bases. Structurally, FIy River floodplain, leads to low pH values the three groups in are similar; differences the overlying water. are in molecular weight and functional group content. The analytical determination of soluble chelates in natural 6.2.1.5 Reducedspecies waters is very difficult, particularly with the minute quantities of metal NH4*, HS- and NO2- were ions that are usually determined in waters present. as indicators of reducing conditions. All three compounds are unstable rn oxygenated water. Humic substances have a strong tendency to Nitrite is metastable in both reducing become adsorbed on inorganic surfaces like (conversion to N2 and N2O) and oxidizing waters hydrous oxides and clays through a mechanism (nitrate formation). The stability of NH4+ and involving ligand exchange of humic anionic HS- is pH-dependent. HS- is converied to groups with HzO and OH- of mineral surfaces volatile H2S at pH values below ?. To the (Tipping 1981). These negatively charged contrary, NHa* forms volatile NHg at pH above coatings may themselves become active 9 (at 30'C). absorbers of trace metals. Colloidal iron and aluminium oxides are stabilized by Within the pH range observed in the humates. In highly dispersed form, these pass investigated waters with oxygen deficiency, only colloids through an 0.45 pm membrane fi.lter. NHa* waa a reliable indicator of reducing conditions. Many redox reactions are slow and The most important feature of humic and fulvic depend on biological mediation, hence the acids is there tendency to form complexes with concentrations of species encountered in natural metal ions. As polycarboxylic acids, humates waters may be far from those predicted precipitate in the presence of dissolved Ca and thermodynamically (Stumm and Morgan f g81). Mg due to coagulation. Humic substances display colloid-chemical behaviour. In the present study, three types of water were distinguished for their different chemistry and main constituents: Fly River water, unpolluted floodplain waters, and mrxed waters.

28 6.2.2 Fly River Water Within the temperature range observed in Fly River water, pco:l does not change to an extent \\'atel fi'om thc Nliddle lrl.v Iliver (Table 13 and which would markedly affect calcite dissolution. Tablc A.t in tlre Anrrex) is charactelized by a The plot of pH versus calcium (Fie. 5a) shows a nrodcralelv high content of' earth alkaline and significantly negative correlation G <0.05). The allialine nrctlls u'hich is dtte to the active plot of suspended solids versus calcium (FiS. 5b) nrinelal clissoltttion ol' 1i'eshlv eroded tock displays a significantly positive relationship. It nrat.erial rvhrch is caltted in sttspension from thc can be concluded that at low pH' more calcite Ntount l'-ubrlln lnrnestte and the Ok Tedi from the suspended particulate fraction is dissolved cirtch ur e tr t. dissolved, which means that the calcium concentration is controlled by pH the 'fht urirjol trniot't is lricarbonate. followed by (assurning constant atmospheric pcot. rrrinor aniotrs sr.rl{irtc, chloride and nitrate. The cloruinirting drssolvcd electrolytes are calcium The fact that the highest suspended solid ;rnd bicarbonittu. rvhich accottnl lbr approx- concentrations are associated with the lowest the imatelv 90'% of'conductivity (calculated fi'om nrol pH values may indicate that high flows in equivnlents). itncl rvhich also control the Fly River are associated with low pH. Since high moderat,eh' allialirte pH of 7.7. The content of flows are a result of heavy rainfall. and because dissolved organic carbon (DOC) is fairly high at rainwater is saturated with atmospheric CO2, aborrt 6 mg/l (low trutnber of samples, not the increased input of carbonic acrd may explain repolted in Tables). near neutral pH values in the river water at high flows. Oxygen satttration ltteastll'enrents of Fly Rrver rvater gave a tneirn valute of only 66% which is There exists a positive relationship between obviously inflrtenced by the oxyElen-consumitrg conductivity and suspended solids (p <0-05' Fig' dcrcay of dissolved and particulate riverine 5c). A high concentration of suspended matter is organic matter'. Reduced species (NH.r+' HS-' associated with high values for earth alkaline NO2) were plesent at or below detection limit and alkaline metals (Ca, MS, Sr, Na and K)- rvhich indicates ox-vgenated conditions.

25% - 75oto con"fidence iltteruala Fly To b Ie I i] : II eq r r,. st.ot ul.ar d de u ia t io t r,, nrc d iol t rnlues uad, for Ri.vttr uto.l,er- so trtples (n' = I I ).

Parameter Unit Mean SD Median Percentiles 25 75

Temperature 28.2 1.6 28.9 26.4 29.4 pH 7.7 02 7.7 7.7 7.8 Conductivity pS/cm 138 15 136 128 146 ot 68 Oxygen Saturation TO 66 0.2 64 60 Suspended Solids mg/l 199 149 | 1s5 92 207 I 299 1350 1660 Na ug/l 1564 | 14zs 87 615 536 667 K ug/l 622 | 24750 31300 wd pg/l 29380 6654 I zoooo pg/l 170 11eo 1075 1345 Mg 1237 | pg/l 172 36 | 158 144 181 pg/l below dtitection ttmit of AI 501tg/l I Fe tjg/l 72 102 lso875 25 Mn PVrl 18.4 13.2 I rz,s 8.5 10.5 Zn pg/l 9.2 6.2 I uo 4.0 below dr ttection limit of 0.1 pg/l 1tg/l I 3u pg/l 196 11.8 rz o 13.0 1e.3 I'detection below timit of 1 Pb ug/l 1tg/t I Mo pg/l 7.9 5.0 | 70 3.0 12.0 unn I rvv3 mg/l 77 4,0 1767378 uvdcn 2- mg/l 5.0 2.5 I s.s o.o 6.0

29 Ftgure 6;Ecahur pletl for setrected: Wrarnaters ilt FIy Biner watcr (Fi,g. fia,e) on'd, floadptailt' 111:6fisrs (fiA, 5d), fr.g.5a Fie.5b

8,? 600

8.1 s00 I

7;:9 400 7.A .o, T- z.t' E soo U1 7.6 an 200 7.5

7..4 100 7.3

7.2 222{.26283039 77 ?4 26 28, S0 52 54 5:6 5,8 42 Cqleium' Ccleium mgll W6e w.5d

000 t&0

160

t+0

I 2t0

\ \ roo o| of E lso I (rl 'v1 a80

200 e0

40 t00 r l' 2D

0, r r0 120 110 140 150 160 170 180 lD t5 .20 25 30 55 .40 45 Gonductivit'y pS/crn eondtrctivity pS/cm

30 No correlation was found for suspended solids The content of suspended solids is very low as versus F-e, Ivln and Mo. A negative correlation compared to the Fly River. The material orange beLween sttspended nratter and dissolved copper retained on the membrane filter is (about and zinc was observed, however statistically not brown, consisting of iron oxides 10-15% significant. Na, Ca, Mg and Sr ale highly Fe) and particulate organic matter. The high of intercorrelated. content of iron. most probably in the form oxyhydrate colloids, is particularly evident in Sample Wl/21.2.. which has the highest Fe the non-filtered water samples, but also in the values value of 339 ug/I, also gave the highest filtered water in which the iron values are much water for dissolved Zn and Cu. In the acidified higher than in the Fly River water. Sodium, an sample, the colloidal iron oxide particles are ubiquitous element, is present in floodplain are dissolved and adsorbed trace metals watei-s in a similar concentration range as in the released. This explains the high Zn and Cu Fly River. Aluminium values were highest at vellues associated with iron- This correlation Iow pH and conductivity (Fig. 5d)' The overall Al was not obsen'ed in other samples. content was higher than in the FIy River, where (50 ps/l) Dissolved trace metal levels, with the exception the metal was below the detection limit of copper (rnedian value: 17 pg/l) are generally in the four samples measured. low due to alkaline pH and high bicarbonate An influence from the mine discharges was content. Solubility of inorganic coppet species in detectable in some water samples (slightly the pH range neasured in the river water ts elevated trace metal, calcium and sulfate levels)' very low. The fact that, in spite of the high However, since the main features of floodplain (which concentratiou of suspended matter offers drainage such as low pH and conductivity and adsorption sites), dissolved copper values are high DOC levels were observed in these relatively high. points to the plesence of soluble samples, they were included in this category' organic copper cornplexes' Metal levels for Cu, Pb, Cd and Mo in floodplain Analytical data from the pre-mining period waters unaffected by mining are at or below (e.g. Table 14)' (Maunsell 1982, Iiyle 1988) are of poor quality detection limit tZ Pgll for Cu, in make cornparison with present data Zinc is the only heavy metal present and floodplain Dissolved calcium concentrations are measurable concentrations. Since difficult. intensely reported to have been in the range of 13 to 16 waters drain lowland areas of extremely low mg/l, which corresponds to about half of the weathered. Pleistocene sediments, present values. Fly River water has a high trace metal are to be exPected. natural content of earth alkaline and alkaline Low alkalinity is a result of active accumulation metals because of the limestone formations of organic material. Peaty sediments were mainly rn the Ok Tedi catchment. repeatedly encountered during lake bottom Fly River water chemistry probably was not Oxygen deficiency in stagnant or slowly "o"i.rg. allow complete much different f'rom present day conditions, rnoving, warm waters does not matter' There although Maunsell (1982) and l{vle (1988) respiration (oxidation) of organic tenperature report slightly acrdrc pH values rn the range of rs a positrve correlatron between (Fig' 6a), which is 5.5 to 6.7 which appear erroneous. Trace metal and oxygen saturation However, the levels measured by Maunsell (1982) were at or contradictory at first sight- result of photoeynthetic below the detection limit of 1 ptg/l with the oxygen measured was a intense sunshine, exception of Fe. Mn and Zn. Pickup et al. (f979) activrty. which is highest at high water report mean suspended solid concenbrations in which in turn is responsible for the range of 60 to 80 mg/l for the Middle Fly and temperatures. note that "the Fly and the lower Ok Tedi are It is obvious that the waters are not in redox very clean rivers by Papua New Guinea equilibrium. The photosynthetic- oxygen standards". pr:oduction masks the overall oxygen deficiency of the system, which is evident from the species 6.2.3 FloodPlain Waters relatively hrgh concentrations of reduced like NHj* and HS-. Of both compounds, NH4* reducing eonditions The term floodplain waters is used for courses of is the better indicator of water which clrain the outer Fly River because hydrogen sulfide is volatile at the low temperatures measured' floodplarn. Therr catchment comprises forested pH-I.lHn" and high water positive correlation areas and low-lying swamps of high biological tho*s a significantly (DOC) (Fie' 6b)' productivity with dense subaquatic and swamp with dissolved organic carbon water vegetation. The main features of floodplain which is also positively linked with waters are low pH and conductivity (Table 14 !en1perature. and Table A5 in annex) and their yellow-brown colout' ("blackwatcr"). 31 Table | 4: Meq,tt, st a nd,trd, deuiatiort, meanl ua,hxes and, 95'"/, - V5% coufid.ertcc irtte:n:r.is for outer fl:aodplain, wate:r sanBl,es (n = r5).

Fammeter Unit Mean SD Median Percentiles 25 75 .cc Temperafure 30.6 2.2 30.0 29,2 31.4 prl 6,0 0,5 6.0 5,7 6.4 OonduetiMty pSrcm- 2&.5 10 27 2A 34 .Solids Suspended mg/l v1 11 19 11 Z4 Na pg/J 1,23X 409 1 140 1020 t2€0 K ugfl 188 10? 193 89 244 ea ugll 4295 1992 41 10 3060 5890 ttrlg pg/t 477 141 480 3V2 499 Sr ugil 27 10 u 21 34 AI pgn 6V 48 64 25 80 Fe ug/t values highly,variable Mn pgn 154 15,1 7.5 2.5 27 Zn ugil ;hse to detecton limitof 5 pg/l cd ugfl nlourdeteofon lirnitof 0,1 pg/l

Cu ug/,1 :lo.m ts de-tection lirnitoI2 pg/l Pb pg,l belor detection lirnit of 1 pg/l Mo ug/l belorr detecton llmlt of 5 pgll HC03' rngll 13 6:5 13 $: 1'l NHr* mgll 0.3 a.2 0_20 0,14 0.38 so41 mgn 0.37 0.26 0.27 0.18 0.45 DOe. mg/l 9,4 3"0 8:0 7.0 't1.0

Ffuure 6;Sca0lerplods for sel,ected petra,t*tzrs i,rr,l,:lnad,plah,t wsi,ets (Fi6 6o'6b),

Fig,.6a Fis.6b

I60 rg

't 40 \ro ot E 6\\o il20 r, tr 5 o n 100 qt- U L C)t2 .g q 80 c v'1 8,'t c lg o 60 O O) x Eg a o 40 r{ 6o .96 20 o lt I 0 2V 28 29 50 31 32 33 34 55 3.7 0.1 0 2 0:5 0.4 0i.5 0,6 av Ternperoture oC ]irll-{4 mgrrl

32 NH+* and DOC ale both a result of At high flow conditions in the Fly River and decomposition of' organic ttratter. The DOC previously little rainfall in the lowland, river ploduced in ofT-river sites ts ihe nratn source of water intrudes several kilometers upstream of organic ligands which are responsible for trace the channels and lakes which drain into the ruetal courplexation rn polltrted waters. main river under reverse conditions' The intrusion of FIy River s'ater is associated with transport of mainly mine'derived suspended 6.2.4 Mixed Waters matter, which is deposited in the waters of the inner floodplain. Hence, it is not possible to ge in Mixed waters are nerally intermediate distinguish between dissolved metals in mixed compositiou between Fly River and floodplain waters which are directly derived from an River waters, although the influence of the Fly intrusion of Fly River water and those that may in is clearly dominant (Table l5 and Table AG be rnobilized secondarily from the sediments annex). already deposited. generall.v- the locatlon of Due to the {iat terrarin. Because of the different composition of Fly River and FIy the mrxrng zone of {loodplarn drarnage waters and those draining the outer floodplain, River water depends marnly on rainfall in the the physical and chemical interactions occuring upper catchtnent of the Fly River/Ok Tedi and in are of environmental interest. Particular high the Fly River lowland. When there is attention has to be paid to the behaviour oftrace mixing rainfall in both catchrnent areas, the metals, of which copper is the most relevant. front between Fly water rich in suspended solids and blackwatet li'om bhe floodplain will be located close to the mouths of creeks and tie channels.

Toble 15. Mean, stondarrl d.euiotiort, rnedion ualues and.25% - 75% con'fidertce interuals for mixed u)a.ter so,,,tples (n' = 65).

Parameter Unit Mean SD Median Percentiles 25 75

Temperature 30.2 2.0 30.1 28.9 31.7 pH 0.8 7.1 6.6 7.7 Conductivity pS/cm 130 on 119 80 151 Oxygen Saturation % 77 42 77 41 107 '15 Suspended Solids mg/l 36 72 728 1755 Na ug/l 1 587 541 1480 1185 K pg/l 539 657 385 272 603 Ca pg/l 21 390 8372 21 900 13425 28050 1345 Mg pg/l 1226 725 1140 881 Qr pg/l 150 111 138 82 176 rqo 101 AI pg/l 102 25 25 50n 457 Fe trg/l 357 145 32 24.5 Mn trg/l 82 308 2.5 2.5 Zn pg/l 15.4 15.5 12 619 pgil close to detection limit of 0.1 pg/l pgil 11.8 11.3 14 Cu | Pb pg/l close to detection limit of 1 Pg/l Mo psl /.6 9.9 5.0 2.5 9.0 62 HCO3' mg/l 53 28 47 39 NHa' mg/l 0.21 0.16 0.15 0.10 0.25 Sooz' mg/l 2.5 2.0 2.1 0.7 3.4 DOC mg/l 9.0 2.5 9.0 7.3 11.0

33 As mentioned above, the pH of floodplain waters and fulvic substances. In the absence of calcium is much lower than in the Fly River. Decreasing bicarbonate buffering, the organic acids control alkalinity may lead to increased heavy metal water pH. mobility due to desorption from particulate Reducing conditions, matter and formation of free dissolved metal indicated by an elevated NHo+ species. content, also seem to have little influence on dissolved trace metal concentrations. Only The plot of pH versus Ca shows a weakly zinc gave a moderately positive correlation with positive correlation (Fig. 7a). Mg, Sr, Na, K and NHa*' HCO"- give similar plots. Calcite in particulate Ca shorvs a weakly positive form carried into the floodplain waters is rapidly correlation with Mo (Fig. 9a). A positive dissolved and exerts a buffering effect on the relationship of the alkaline and earth alkaline metals (calcium local waters. Opposite to the main trend in the with sulfate versus sulfate, Fig. 9b) was pH/Ca plot, there are samples showing a high observed. Elevated concentrations calcium content at comparatively low pH. These of Ca, Mg, Na, K, and Sr are clear indicators data are from small ponds on the swampy of the influence of mine discharges. High sulfate levels floodplain and slightly acidic floodplain seepage, are a product of active dissolution i.e. extremely iron-rich water trickling from of mine-derived sulfide minerals. Although exposed channel or river banks into the river the metals iron, zinc, copper during low flow conditions. and molybdenum are discharged into the river system primarily in the form of sulfides and No clear influence of pH on dissolved iron and undergo oxidation to sulfates, no positive manganese concentrations was observed, correlation between sulfate and any of the although the highest Fe values tend to be metals was observed. Iron and zinc even associated with low pH. displayed a weakly negative relationship with sulfate. This None of the trace metals showed a significant illustrates the complexity of solution chemistry correlation with pH. The plots of pH versus zinc in the investigated waters, (Tig. 7b) and cadmium display slightly and the fact that aqueous metal transport is not a controlled negative trend. No correlation between pH by sulfate complexation. There was and also significant Cu (Fig. 8a) and pH and Mo was detected. DOC no correlation of dissolved organic carbon (DOC) and pH are weakly negatively correlated (Fig. and trace metals found. OnIy the plot 8b) which is due to the acidic nature of humic of zinc versus DOC shows a weakly positive correlation.

Figure 7:Scatter plots for selected, parameters in, ntixed uaters (FiS. Ta-7b).

Fig.7a Fis. 7b

90 90

80 80

70 70

- 60 60 o, Eso (- 1 c I .l ao .E oo N E II O:O t tl t tt, I r I , i I r ,l t ,r rtr - I 20 20 r t t r r I tr ,t t . I rr r ri 'l rtl -f - I I. 10 t I 10 rl II r, | | , , r t artt!*-- I ' r lra tttt att 0 0 7.3 8 85 9.5 55 6 b.) 7 7.5 8.5 ql pH pH

34 Figure 8: scatter pLots for selected pararneters hr, mixed waters (Fig. 8a-8b).

Fig. 8o Fig.8b 60

o)12 rI Ltr c. rtl o I I 40 -ot0L U rD rT l- Lln II o-* nc o. g) II o r) rt 6 -2o tl -n rl I o) f I I I o tttt { l0 I I' I I tttl r I .U4 I I trl I o trl at II r I L 0 t 5.5 6 6.5 7 7.5 I 8.5 I 9.5 5.5 6.5 7.5 d.) pH pH Intercorrelations between elements in mixed Both factors interact in a complicated manner waters are very high in the group Ca, Mg, Sr, which cannot be predicted from thermodynamic Na, K and HCO3-, with the exception of Na equilibrium calculations. positive versus K. Zn and Cd display a Despite the fact that lead in mine wastes is correlation, too. A significantly positive clearly enriched above background values, the in correlation was found for Al and Cu dissolved metal contents were in the great unfiltered water samples (Fig. l0). Aluminium majority of samples below detection limit (< I hydroxide may complex or adsorb dissolved pgn;. El"rr"ted concentrations were found only copper. Al/Zn displayed a similar correlation. Fe in unfiltered samples and in a few samplee clear versus Cu in unfiltered water shows no whrch also showed a high content of presumably trend. Fe is weakly positively correlated with colloidal iron and manganese, which offer DOC and Mn. adsorption sites for lead. The heavy metal concentrations in waters of the Speciation modelling of inorganic lead with Fly River inner floodplain, the zone of mixed pUnpnQp suggests that at near neutral pH water, are of particular interest because of the values and oxic conditions in waters, more than play in prominent role which off-river waters 7\o/o of soluble lead exiets in the form of the the the ecology and biological productivity of PbCO3o complex which easily precipitates out' entire rrver system. BetwJen 0-30% of lead may be present in the environmental Dissolved trace metal levels in mixed waters are free ionic form Pbz+ under the controlled by a number of abiotic and biotic conditions in the floodPlain. factors. The most important inorganic factor is Cadrnium and zinc Erre among the geo- of FIy the moderately high bicarbonate content chemically most mobile elements. Their mobility metal River water and the high earth alkaline is less pH dependent as compared to lead' Cd concentrations, domrnated by calcium in was found in concentrations above detection dissolved and particulate form, which are limit (> 0.1 pg/l) only in waters with a pH below responsible for the neutral to alkaline pH values ?. Because of the low Cd values in mining biotrc in mrxed waters. The most prominent residues, the metal is not considered as being of factor ate the dissolved organic carbon environmental concern. The same holds true for substances which play a very important role as zinc, which showed a similar concentration complexing agents. range in all waters investigated.

35 Figure 9: Scatter plots for selected paranrcters ilt rttixecl u:eters (Fig. ga-gb). Fig. 9a ,o Fi.g. 9b ,

q, aJo E 3zs o _o 20 c4 o > 15 J

n 0 r0 20 l0 40 50 60 70 F,') 9,) D l0 l0 i0 4r) 50 60 7a Colcium nrg/i Colciunr mg/l Figure 10: Scatter plot for Al uersus Cu in mixed wa.ters.

Molybdenum. because of its anionic form in water, showed a different behaviour than the other trace metals. Highest concentrations were usually found in alkaline waters, or linked with high calciurn values. Mo as well as Zn is considered an essential element to biota. The EO concentrations \ observed are much below the c', toxrc threshold. f,_ Copper is an essential element to plants b50 and n animals, too. o_ r)o Because of its environmental importance, the 1A fate of copper in the investigated waters will be discussed in some detail in the following chapter.

6.2.5 The Behaviour of Copper

The statistical data analysrs showed that copper o roo :oc roo ooo too ,Jo tJolrlilToo concentrations rn water do not appear to be si gnificantly Atuminium i.rg/l controlled by rnorganic factors. Alkalinity and pH. adsorption to oxide surfaces (with the exceptron of alumrnium; and reducrng conditions (e.g. formatron of soluble copper amlne complexes) do not seem to have a srgnificant effect, on drssolved copper levels. The abundance of organic complexing agents in the investigated Middle Fly floodplain waters may be responsible for the observed behaviour of dissolved copper. No coruelation was observed between DOC and copper levels in the floodplain waters.

36 It is evident that dissolved organic ligands, even This behaviour is contrary to that predicted by at comparatively low concentrations in waters of inorganic solubility considerations. Complex- the Fly River system, are always present in ation of the trace metals by dissolved organic excess of trace metals which may become matter is the moet reasonable explanation. complexed (although some complexing capacity Organic copper complexes are known to be may be occupied by major cations like Ca and particularly stable becauee of their favourable Mg). In the Lower Fly River, downstream of the electron configuration (Stumm and Morgan Strickland junction, measured dissolved copper 1981). high at 6 pg/l (Table values are still moderately Fe and Mn in floodplain waters, despite their 4,6, annex) despite the massive admixture of chemical similarity, showed no significant uncontaminated suspended matter from the correlation in any data set. Of the elements River. appears that the lowering of Strickland It investigated by Sholkovitz and Copland (1981), Lower is dissolved copper levels in the Fly Fe>Cu>Ni>Cd (in decreasing order) showed the that mainly due to dilution effects and strongest affinity for the dissolved humic riverine particulate matter is of adsorption to substances, Mn and Co the least. The differing secondary importance. tendency of Fe and Mn to form complexes with Davis and Leckie (1978) tested the influence of dissolved organic carbon may explain the dissolved organic substances on copper missing correlation in data from the Middle Fly adsorption. Certain organic ligands which form River region. particles coatings on suspended mineral In a copper adsorption experiment (spiking to adsorption, enhance the extent of trace metal yield 20 pgA Cu) with natural waters containing opposite behaviour. Some while others show dilferent concentrations of suspended solids and organic acids inhibit copper adsorption by dissolved organic carbon, Sholkovitz and complexes which keep forming soluble, stable Copland (1981) obtained results which suggest When complexing the netal in solution. the that solubilization of Cu by dissolved organic dissolved carbon substances capacity of organic ligands, forming ultrafine colloids (< 0.01 pm), is the rate of adsorption to inorganic is larger than a more important process than the adsorption copper remain solution. surfaces, will in onto riverine particulate matter. Dven in water Sholkovitz and Copland (1981) also investigated containing 100 mgil mostly inorganic suspended the solubility and adsorption properties of a matter and 3 mg/l DOC, most of the copper was number of trace metals in the presence of humic kept in solution as the pH increased from 4 to 9. studies which were acids. Contrary to similar CSIRO of Australia (1989) performed mixing performed Iaboratory waters, with synthetic experiments commissioned by Ok Tedi Mining Copland used natural water Sholkovitz and Ltd. with water from the Fly River and a from a small stream in Scotland which drains floodplain tributary. Mine derived sediment peaty soils. The properties of this water (DOC 7 with high copper values was added to different pH 6.5, low conductivity) are similar to ng/I, admixtures. Total dissolved copper concent- floodplain. The authors those of the Fly River rations increased with sediment admixture, the humic acids and detected a coagulating effect on pH remained fairly constant. Dissolved copper trace metals, particularly iron and copper, upon species in the resulting solutions were analyzed mmolA Ca, which is adding of only 0.5 of by Anodic Stripping Voltametry 6SV). The equivalent 20 Ca. to mg/l method is used to determine labile trace metal The mean Ca concentration in mrxed waters of species supposed to be present in the ionic, most the Fly River inner floodplain is about 25 mgll. bioavailable form. Coagulation of humic substances with adsorbed lnterestingly, between 30-50% of dissolved trace metals may be an important process when copper in the final test solutions was measured water mixes with calcium-rich Fly River as Iabile or "ionic" copper. Assuning that the floodplain blackwaters. This is consistent with copper was present as dissolved organic species, the detection of elevated copper levels on the top the results point to great reactivity of humic of black, peat-like sediments sampled from lake copper complexes in the Fly River system. This bottoms far from direct Fly River in{luence. The is consistent with more recent laboratory work experiments carried out by Sholkovitz and undertaken by CSIRO on behalf of OTML (1981) gave Copland also the result that there (OTML 1994). Electrochemical measurements no precipitation of trace metals and humic was of bioavailable copper in the Ok Tedi/FIy River (starting point pH was acids when pH 6.5) system gave the result that the fraction of changed the range of 9.5 to 3, below which in potentially bioavailable copper is up to 5oo/o of humic acids, Fe, Mn,'Cu, and Cd began to Ni the total dissolved copper concentration. precipitate.

37 In the presumably unpolluted waters of the Where floodplain swamp waters wrth leached outer floodplain (it is difficult to establish the copper drain into lakes and the main river, they maximum intrusion range of Fly River water) will increase copper concentrations. This may copper was at or below the detection limit of 2 explain why OTML (1991, 1998, 1994) reporrs pg[. In the Fly River, copper values were about higher dissolved copper levels at Obo as tenfold above this "background", which may compared to the upstream site at Nukumba, actually be much lower than 2 pg/l, i.e. 0.2 pgn. below the Ok TedilUpper Fly River con{luence. Mixing of both waters should result in dilution. A different type of water in which high traee This is the case in most mlxed water samples, metal concentrations could be expected are the however in some waters copper concentrations samples called "floodplain seepage". These were much higher. Sample WLlT-4., which had a moderately acidic waters percolate through the dissolved copper concentration of 52 pgll in the floodplain sediments and drain spring-Iike into filtered and 106 pg/l in the unfi-ltered sample, the channels and rivers. The waters are highly was taken from a depression in flat swampy reducing and have a very high Fez+ and Mn2+ terrain about 150 m behind the low dam content which is immediately oxidized to orange paralleling the Fly River channel. coloured gels and flocs upon exposure to atmospheric Water sampled from small pools and shallow oxygen. Although the floodplain seepage samples showed pb water courses in the periodically flooded high Cd and levels, this was not grassland sites usually gave high copper values. the case for copper, which probably remains The water samples taken generally had iow pH in sulfidic binding in the sediment body. and/or alkalinity. It appears that active leaching of copper from deposited mine-derived Table l5 shows predictions based on computer material is responsible for the high dissolved modelling by OTML (1990) of various values observed. environmental parameters in the FIy River system response In the floodplain swamps, redox conditions in to the impact of mine discharges. change easily depending on the undulating These additional environmental monitoring water table which may result in trace metal conditions were established to ensure mobilization. Because of the dense vegetation that the Acceptable Particluate Leuel of 940 mg/l at Nuhurnba does and the low rate of water exchange, soluble not result ht actual etr,uiruunerttal dantage organic chelates are abundant may to tlrc Fly Riuer Systent and beyond th,e facilitate copper mobilization. The buffering leuel achnlly specified ilt the precLiction s (OTML I990). effect of calcium disappears as the easily soluble element is leached from the sediments. Because The predictions resulting from the of the favourable conditions for trace metal Supplementary Environmental Investigations leaching, even a thrn Iayer of deposited mine- (undertaken in 1986-89 by OTML) have been derived material will be an important source of established for testing whether or not the State's copper. Due to difficult access, only a few conditions for environmental management of the samples from the swamp sites were taken Fly River System and off-shore are met. Due to (water was sampled within a maximum distance a number of reasons, there has been a non- of one kils6sls1 from the main river channel). It compliance with monitoring conditions seems that the copper levels in floodplain waters begrnning in 1990- Particularly the dissolved investigated in the present study are biased copper levels were much higher than expected. towards low values (in samples from large lakes Measured values at the APL sites and tie channels) and are not representative for Kuambit/Nukumba, Obo and Ogwa have been in the entire floodplain. a steady increase between lgg0 and lggg (OTML 1994) contrary to the trend predicted by the models. Revrsed prediction values for the key environmental parameters were developed by OTML in July 1992 and again in December 1993.

38 Toble 16: Pt-ed.ictiotts nto.de by OTML fit 1989 for hey enuironmental paraneters in' the Fly Riuer system, and. reuised predictiotts for dissolued copper hr, 1992 and 1993 (OTML 1990, 1993' 1994).

a.. Po.rticu.late and. d.issolued copper and fish catch for the Kuambit/Nukunba, Obo and Ogwa (OTML 1990).

1990 1,36 11 15 nvz 17 35 524 3 95 1991 905 6?,8 n9 966 401 2 118 LWZ 715 338 693 594 321 tt 35 1993 n5 429 850 670 388 t 120 t994 581 349 5g 4 119 n0 L t47 1995 581 349 5ffi 4 119 NOL M7 1996 562 349 544 4 119 E3 L L47 ro 2008

Source:pC\r-OCu -supplementarylnvestigatiou$YolFSppcndixB' -Supplementary Investigations, Vol.I' Figure 3' Fisb catch -supptementary Investigatioos, vol 4 Appendixll' Figures tOa, 114 &.\?aandTable4A

b. Dissolued. copper predictiort's (pe/l), prouided to State ht' June 1992.

Year Nukumba Obo Oclra

l99l r l (10.5) l5 (r4.2) 6 (s.4) t99Z l3 (13.r) r7 (r5.0) 6 (6.6) 1993 l4 t7 7 1994 ll t5 6 1995 ll l5 6 t996 t2 l8 I

Error Estimate +-78Yo +- SOVo +-85Yo

( ) atrnui (calcodar ycar) mcaas of obscnred '{-n

c. Reuised. predictiorts for dCu in Fly Riuer, and meon' da.ta for 1993 (OTML 1993).

Year I Nukumba Obo Oglva (7..1) 1993 16 (4 - 28)(13.E) 7 (3 - irl (17.6) 6.1 {l - 12} 1994 18 {4 - 12) 8(4-i2) 7.i (l - l4l 199i r.r {3 - 25) 4 {i - 25). 6.7 lt - tzl 1996 r{ {i - 25} 4(3-2r) 6.6{l-12} 1991 B {i -2i) i{i-25} 6.r (r - ll) r998 tl (j-251 I {i - 25'l 6.6{l-l1l { ) Uppcr rnd lorvcr error estimatcs () Anrruel (srlcndrr -vc:tr) rnsrns of obscn'cd datit

39 7. Biological lmpacts

The discharge of ore processing residues and Both total fish catches and the diversity of the mining wastes has significantly changed the aquatic fauna are affected. No clear negative aquatic environment of the Ok Tedi/Fly River impact of mine discharges on the abundance of system. From the discussion in previous sections fish in the Middle Fly River downstream of it can be concluded that there are three factors D'Albertis Junction can be established from the of particular environmental concern: the data collected so far because of their high increase in particulate copper, in dissolved variability. Fish biomass monitoring in the last copper, and in suspended solids concentrations. three years has been strongly influenced by low The negative impact of each effect on the fluvial river levels concentrating fish into the river system is spatially di-fferent. channel as the floodplain dried, hence leading to high catches at the river sites. The massive increase in the suspended load of Clearly, the exact determination populations (including the Ok Tedi is mainly responsible for the decline of fish migratory species) in aquatic life upstream of the conJluence with in a dynamic fluvial ecosystem is a difficult undertaking the Fly River, where fish populations and since fish catches are inJluenced by species diversity have been dramatically a number of factors which cannot be attributed reduced (Smith 1991). The persistently high to the mine waste, Iike periods of droughts and commercial concentration of suspended sediments in the and subsistence fishing pressure. water interferes with gill respiration of fish and In addition, OTML (1994) admits that its database aquatic organisms, modifies the movement and on fish biomass is limited with respect migration of fish and prevents successful to baselne data and comea to the conclusion development of eggs and larvae. Heavy metals that "changes between before and after mine start-up in particulate and dissolved form, and toxic conditions cannot be quantified for any site (in process chemicals like xanthates probably play a the Fly River system)". minor role in the adverse impact on the aquatic life of the Ok Tedi River. Considering the homogeneity of the Middle Fly River system and the fact In the Fly River itself, a combined detrimental that the key water parameters influenced effect of suspended sediment, particulate and by the mine'e operation (dissolved and particulate copper and suspended dissolved copper on the aquatic ecology is to be solids) show little changes expected. Monitoring undertaken by OTML in the Middle Fly River reach between Kuambit/Nukumba (Smith 1991, OTML 1993, 1994) shows that fish and Obo, negative effects on populations populations in the Middle Fly River at Kuambit, the fish in the lower part of the Middle FIy region be immediately downstream of D'Albertis J unction, are to expected. are continuously declining, in the range of 30- 5oo/o (M. Eagle, pers. comm.). Statistical OTML (1993) report elevated copper levels in analyses gave the result that particulate copper body tissue of the catfish Arius benrcyi caught in (pCu) was the best negative correlate of fish Lake Pangua, a deep oxbow lake. Arlus is a catches, but dissolved copper and suspended bottom feeding omnivore. Aquatic invertebrates solids also have a negative effect. OTML (f994) form the most important part of its diet, but speculates that the relationship between pCu detritus/mud is also an important food item and fish catch may either be attributed to an (I{are 1992). It is not clear whether the copper avoidance of pCu by fish or by a correlation of detected in the fish was taken up via the bottom pCu with a toxic fraction of dissolved copper. dwelling invertebrates, or directly from the copper-rich sediment, the metal being released in the rntestinal tract of the catfish. In both cases, it is evident that copper in particulate form on the lake bottom is bioavailable.

40 larva In the off-river floodplain sites of the Middle FIy The most sensitive species, a chironomid region, elevated concentrations of dissoived (midges, Diptera), was eliminated at 13 pg/l (1989) copper may negatively affect aquatic life. The copper after l0 days. Moore and Winner floodplain plays an irnportant role in primary also found out that benthic chironomid and to productivity to support fish stocks in the river mayfly larvae show a high sensitivity for the recruitment of fish- OTML dissolved copper. Burrowing mayfly larvae of system, and with (199f) states that the floodplain supports a the genus Plethogenesia (together prawns) were the greater stock of fish than do oxbow and drowned Macrobrachiurrr. freshwater valley lakes. The Fly River channel is to a large dominant constituents of the macroinvertebrate extent a fish tnigration route and a refuge fauna in terms of biomass in the Fly River food items for cluring clry periods. Subsistence and commercial (OTML 198?). Both are important particulate copper in fisheries target mainly the off-river water bodies the local fishes. Toxicity of been proven by because of their high productivity. Although the mrne wastes on mayflys has 1988 (OTML vegetated flooclplain sites may not be inundated bioassays undertaken by OTML in over the whole year, fish invade rapidly into the 1989). alluvial plain at higher water levels because of Williams et al. (1991) report a 50% mortality the abundance of food sources. after three days of exposure of the extremely Carid'irt'o Animals at lower trophic levels, such as aquatic sensitive tropical freshwater shrimp studies on copper invertebrates, and juvenile stages of fish are sp. to 4 FSfi copper. Most as the known to have a high sensitivity to dissolved ttxicity consider the free ionic speciee copper complexed copper. According to Moore and Ramamoorthy most bioavailable form, and organic carbon (1b84), sensitivity to copper is inversely related with naturally derived dissolved toxic or even non'toxic (e'g' to the age or size of an animal. A decrease in the (DOO as much less 1989; Meador, 1991)' abundance of macroinvertebrates, which are at Flemming and Trevors (Winner and Owen, the base of the food web, may lead to a decrease Recent scientific work provided further evidence in the overall fish population. Copper is a strong 1991), however, has toxicant to aqualic organisms but does not that dissolved organic carbon also may enhance in the food chain. End consumers copper toxicity. The authors used the alga biomagnify to test the like carnivorotrs fish show a much lower copper Ciia,tnyd.ontonas rein'hard,tii organically body burden than benthic invertebrates like bioavailability and toxcity of from a freshwater burrowing mayfly larvae which feed directly complexed copper in water frorn the contaminated substrate (Karbe 1988)' pond. alga The average copper concentration in mixed Toxic effects of dissolved copper on population growth were waters of the inner floodplain is around 10 pg/l; de{lagellation and 12.2 pgl in natural the Fly River water has about 17 pgil of copper' observed at values above DOC concentrations varying from 5 These are concentrations which may be harmful waters with to biota. The applicable United States to 14 mg/l. (USEPA) Environmental Protection Agency They found a positive correlation between DOC of "Water Quality Criteria for the Protection and copper on alga toxicity and concluded that Aquatic Organisms and Llses" (1986) labile organic copper complexes may be pgn recommends a maximum value of 6'5 responsible for this effect, demonstrating that a (hardness 50 mg/l as CaCO'3, unfiltered water) simple relationship between DOC concentration for the "chronic" 4-day average concentration, an& copper toxicity cannot be established' and 9.2 pg/l for the "acute" l-hour average OTML (1994) has commissioned toxicity tests concentration. The Canadian Water Quality with the freshwater green alga Chlorella Guidelines (198?) are even more stringent, as protothecoides. seen in Table 16. No reduction in algal growth rate occutred at mixed Dissolved copper levels detected in dissolved copper concentrations of L2 pgll, which the floodplain waters of the present study are in was the highest concentration used in the tests' the concentration range which is reported in Since copplr values measured both during the literature to be harmful to sensitive freshwater present study and by OTML in waters of the Fly aquatic organisms. Clements et al' (i989) found iUver and its floodplain were much higher than total aquatic insect tests a significant decrease in 12 $gn, it appears justified to repeat the abundance after 4 days of exposure to 6 pg/l of with copper concentrations of up to 50 pgll in copper in an outdoclr experimental stream' order to obtain more information on the species' sensittvity.

41 Toble 16:Cantad,ian Guidelilws for the protectint, of freshwater aquatic life (CCREM tgST).

Plnmccr Gurr.0 mg.L-r; DOC=].o E|8.L-l

.Antimony IDT

Arscnic 0.05 ng'L- t

Bcr-vllium ID

CNdEiurn 0.! pg.l-t Hsrdncss 0-6O mg.L-r fCrCOrt 0.E FS.L- | Hatdness 6&120 mg.L-t (C!CO1) 1.3 pg'l- t Hardness ll&lE0 me'L-t (GCO3) l.t pg.L-l Hardncss >ltO m-g.L-r (CeCOr)

Chlorim (.otrl 2.0 pg'L- anpcromertc r€sidud chlonnel ' lvtcasured bs or cquivalent mcthod

Chromium 0.01 mg'L-t To proreo fish 2.0 Pgl- t To proreet aquatic lifa iacluding rooplar*ron ard phyroplrntron

Copper 3 Pg'L- t Hardness G60 mg'L- t (CeCOr) 3 Pg'l- t Hardness 6&130 mg-L-t (CaCO31 3 Pg'L- t Haldncss l3GlE0 m-s.L-r {CaCOr) 4 pg'L-t Hardncss >lE0 ms.L-r (CeCOr)

C:aridc !.0 Pg'1-- I Ftec c-ranide as CN

Dirsolvrd oxygen 6.0 mg'L- t bioa - carly lifc sages 5.0 mg'l-t "t'arm-*acr - othcr lih soges 9.5 mc'L-l Cold.watcr biota - carly life sogcs 6.5 mg'L- t - othcr lifc ruqes

Lron 0-3 mg'L- t

be I Pg'!- t Hardacss G60 mg.L- I (CaCOr) 3 pt'L-, H6r{nc!s 6G130 ng.L-t (CaCO5) .l pg.L-l Hardnass l2Sl80 mg.L-t (C.CO3) 7 ;rg.!- t Hardness >lt0 mg.L-r (CaCO1)

Mcrcury 0.1 Pg'l- I Nickcl 25 Ft'L-r Hardacss S6O mg'L-t (CaCO3) 65 pg'L-l Hadncss 6{}.120 mg.L-t (CICO3) I l0 pg'L- t Hardncss 12tr180 mg.L-t (CaCOr) 150 pg'L- t Hardness >lt0 a1g.L-r (CaCO3)

Nitogen Aruroaia (total) !.2 mg'L-r pH 6.5: anpcrrtuc lOt (stc Tablc 3-12) 1.3? mg.L- t pH t.0: ampcrerurt lfC Nidte 0.06 utg'L- t Nitrerc Cosccaczrroos that s.imuhc ptoliic ud goflh should bc rwidcd Nimreraiaes ID

pH 6.5-9.0

Sclcuium I Fg'L- t

Silvtr 0.1 Pg'l-t

Thelliurn ID

7i,l.c' 0.03 mg'L- I

Physical paramacr's

Temperarurc Thcrmal addiuons should not oltcr lhcnflal stntificuion or rurnorsr dacs. :rcEd maltmum $'cekll,.lleB_qe tempcrlluEs. and ancced marimum shon_ :efin rcmpenilurss rscc Sccrion .3-:.3.1-l)

Total surpcndcd solids inccasc of 10.0 Background tuspendcd solids =100.0 me.L-t t!g'L - | rncrrrsc rrf l09t abovc Backgtouno susprnded solids >100.0 mc.L-t backcround I Coocenratroas of hcavy mctals reporad rs ronl menl in rn untilcnd samoic. : lD = insutficienr dua to recomrncnd a cuidcline, I Tcntarive guidclinc.

42 8. Conclusions

The following conclusions can be drawn from the The most sensitive biota to dissolved Cu are informations gathered in the present study: juvenile stages of fish and aquatic invertebrates which form the base of the 1. Of the trace metals contained in the Ok trophic web. Populations of larger frsh main Tedi mine waste, copper is of species will only show a response after a environmental concern because of its strong full reproduction cycle has been completed. enrichment above background (about Thus, measurable negative elfects may high twentyfold) and its relatively develop with a considerable time lag. geochemical mobilitY. 5. The copper pollution in the Fly River 2. The discharge of mining residues into the floodplain is of persistent nature. Even Ok TedilFly River sYstem leade to a after the cease of deposition of copper-rich substantial deposition of copper-rich suspended load, the environrnentally detri- material in the Middle Fly River floodplain, mental effects will continue to exigt. There although most of the mining wastes carried is no mechanism to "de'toxifu" the alluvial as suspended load finally reach the delta. plain. Erosion processes will only remove Areas of standing water and vegetated deposits in the main river channel and on parts of the floodplain play an important the banks immediately adjacent to it, and, role in the recruitment of fish and primary to a minor degree, along the channels productivity. Hence, this part of the fluvial connecting drowned valley lakes with the ecosystem is particularly sensitive to mine- main river. The mine'derived material induced changes in the local environment. deposited in oxbow and drowned valley will 3. The pollution by copper-rich material is not lakes, and in the floodplain swamps, only a problem of quantity, but also of remain and may be covered bY lesg spatial distribution and the potential contaminated sediments carried by the Fly mobilization of the trace metal' Deposition River in the post-mining period' Howevet, of mine-derived sediments in the FIy River when sedimentation rates in the floodplain highest in oxbow lakes. return to pre-mining values, it may take floodplain is gediment Because of the naturally high organic centuriee until copper'rich carbon content in sediments, which is deposits are sealed by an unpolluted responsible for reducing conditions, only sediment cover. (few copper in the uPPermost laYer The possibilities to mitigate the detrimental may centimeters) of bottom sediments effects of the Ok Tedi mine waetes on the become dissolved. When copper-rich Ok Tedi/Tly River system are limited when sediment is permanently covered by water, the construction of tailings and waete mobilization and bioavailability of the retention facilities is deemed uneconomical metal, except to bottom-dwelling inverte- by the mining company' In order to reduce brates, is low. However, when mine- the amount of metal discharged into the derived sediments settle on extensive areas environment, the recovery of copper in the of the vegetated floodplain, the conditions mill, currently at around 85%, could be are quite different. Chemical and biological increased. This may be Possible bY factors facilitate the mobilization of copper. installing more flotation cells, re'flotation of Even a thin layer of copper-rich material tailings, increase of the residence time of may have a negative ecologrcal impact- ore in the cells, improved control of grain of 4. The way in which copper affects the aquatic size in the flotation process, sulfidization community is not clear. Dissolved organic non-sulfide ores, and other optimisation carbon substances, which are present in strategies. Thus, a recovery of 90% may be high concentrations in the Fly River achievable. The cut-off gpade for waste rock system, complex the metal and keep it in could be lowered, which would decrease the solution. Little is known about the bio- copper reaching the fluvial environment' availablity and toxicity of these organic These measures are most probably not copper species. Recent research has shown economically viable; however, for environ- that these compounds may exert toxic mental protective action, they should be effects comparable to the free ionic copper investigated thourougily by the mining species. company.

43 L Acknowledgements

The authors of the study wish to thank the Very special thanks go to Mr. Peter Gelau of United Nations Enyironment Programme for Obo Station on the Middle Fly River, who helped financial support of the research project. The us in accomodation and transport problems study team is particularly indebted to Mr. during the various sampling campaigns, and Mr. Gerhart Schneider, Programme Offrcer with the Ben Kapi Mekeo, our reliable field guide and Water and Lithosphere Unit at UNEP boat pilot. We also wish to thank the people of headquarters in Nairobi, who offered prompt Aiambak, Bosset and Manda villages for the assistance in solving all emerging problems. hospitality they have extended to us. We also may David Mowbray of the Department of express our gratitude to Mr. Joseph Gabut. then with Environmental Science of the University of the Department of Foreign Affairs of Papua New Guinea (UPNG) provided invaluable the Papua New Guinea Government, for continued support of support througout the various stages of the our research project. We wish to thank reeearch project. The authors of the present Ok Tedi Mining Ltd. for analyzing a batch of water study wish to thank him and Dr. Ian Burrows, samples and further logistical assistance. Head of the Biolory Department of UPNG, who provided working space and laboratory facilities during the field work in Papua New Guinea.

44 10. References

Allen, G.R. and Coates, D. (1990). An England, J.K., Kilgour, I. & Kanau, J.L. ichthyological sttrvey of the Sepik River, (f991). Processing copper-gold ore at Ok Papua New Guinea. Rec. West.. Atlst. Mtt's. Tedi. in: Rogerson @d.) Proceedin'gs of the Su,ppl.34,3t-116. PNG Geology, Exploration anrd, Min'ing Conference 1991, P. 183-190' The American Public Health Association (1985). Australasian Institute of Mining and Standald Methods for the Examination of Metallurgy, Melbourne. Water and Wastewater. 16th edition, (198E). Washington, D.C. Englund, E. & SParks, A. Geostatistical Envrronmental Assessment Applied Geology Associates Limited (f989). Software. Environmental Monitoring Reuiew of the fitmL draft report bv Ok Tedi Systems Laboratory, United States Miling Linited - Etr,uirotunen'taL Stu'dy, Environmental Protection Agency, Las the Nouenr.ber 1988. Unpublished report to Vegas. Department of Minerals and Energy, Port (LS77l' Moreshy, Papua New Guinea' Farrah, H. & Pickering, W.F. Influence of clay-solute interactions on human Busse, M. (1991). Environment and aqueous heavy metal ion levels- Water Air Fly ecology in the Lake Murray'Middle SoiI Pollut. 8, 189-197. Area. in; D. Lawrence & T. Cansfield-Smith (eds.) Su'stailmble Deuelopmen't fo, Ferguson, K.D. & Erickson, P.M. (1988). Pre- Troditiona.l hr.habitants of th'e Torres Strait Mine Prediction of Acid Mine Drainage. irt': Regiott, pp. 261-282. Great Barrier Reef Enuironnrctr,tal Man'agement of Solid Waste pp. Marine Park Authority, Townsville. flil. Salomons and U. Filrstner, Eds.), 24 - 43, Springer-Verlag, Berlin- Canadian Council of Resource and Environment Ministers, CCREM (f 987)' Flemrning, C.A. & Trevors, J.T. (1989)' Cqtmdiilt. Water Quolity Guidelhrcs. Copper toxicity and chemistry in the Environment Canada, Ottawa (Ontario)- environment: A review. Water Air Soil Pollut. 44, 143-158. Clements, H.W., Farris, J.L., Cherry, D.S. and Cairns, J. (1989). The influence of Hem, J.D. & Durum, W.H. (f973). Solubility water quality on macroinvertebrate and ocurrence of lead in surface waters. J. courmunity responses to copper in outdoor Atn. Water Works Assoc. 65, 562-568' experimental streams. Aquot. Toxicol. 14, Higgins, R.J. (1990). Off'river storages as 249-262. sources and sinks for environmental Davis, J.A. & Leckie. J.O' (f978). Effects of contaminants. Regu'lated, Riuers: Research adsorbed complexing ligands on trace metal & Manngemen t 5, 401-412- uptake by hydrous oxrdes. Enuirorr,- Sci. Jackson, K.S. & Skippen, G.B. (1978)' Teclt,rrcl. 12, 1309-1315. Geochemical dispersion of heavy metale via Department of Mining and Petroleum / organic complexing: a laboratory study of DMP (1994'1. Milr.es, Projects art'd Prospects. copper. lead, zinc and nickel behaviour at a boundary. J. QuarLerl1 Report April ta Jurrc 199'1. DMP, simulated sediment-water Port Moresby, PaPua New Guinea. Geochem. Explor. 10, 1 17-138. Eagle, A.M. & Higgins, R.J. (1991). Jones, T.R.P. & Maconochie, A.P. (1990)' Environmental Investigations of the effect Twenty five million tonnes of ore and ten of the Ok Tedi copper mine in the Fly River metres of rain. Proceedin'gs of Mhrc system, in: Rogerson (Ed') Proceedilrys of Geologists' Conference 1990, p. 159-165. The and tlr"e PNG Geology, Exploro.tiort otr'd Mhtin'g Attstralasian Institute of Mining Conference 1991, P. 232-238. The Metallurgy, Melbourne. Australasian Institute of Mining and Karbe, L. (1988). Interactions with biota. rrt"' Metallurgy, Melbourne. Metals and ntetalloids ht' the lrydrosphere: Eagle, A.M. (1993). Copper Mining and the Impact through ntittirtg and in'dustry, an'd' Environment in Papua New Guinea: The preuentiort tech.n'ology, pp. 165-186' Ok Tedi Case Study. Ok Tedi Mining Ltd., UNESCO, Paris. Tabubil.

45 Kare, B.D. (1992). Di.ets of seuen freshwater fish Ok Tedi Mining Lirnited (1994). ApL species irt the Fly Riuer system, Papua New Complian ce an d Addition aI En uironnrctilal Guhrca. Unpublished report to Ok Tedi Monitorhtg Progron: lgg7/ 93 Arur.ual Mining Ltd. Report. An annual report to the Independent KyIe, J. (1988). Water quality and sediment State of Papua New Guinea as a part of OTML's chemistry of Lakes Bosset, Pangua and environmental compliance monitorin g requirements. Daviambu. pp. 9-18 in: J.C. Pernetta (ed.) Poterutial irnpacts of ninhr.g on the Fly Parkhurst, D.L., Thorstenson, D.C. & Riuer. United Nations Environment Plummer, L.N. (1980). PHREEQE - a Programme, Nairobi. computer program for geochemical calculations. Liiffler, E. (f977). Geomorphology of Papua U.S. GeoL Suru. Water Resou.r. hr.uest.80-96, 210 p., New Guinea. CSIRO and ANU Press. Washington, D.C. Melbourne, 196 p. Pickup, G., Higgins, R.J. and Warner, R.F. (1979). Impact Maunsell & Partners (1982). Tedi of waste roch disposo.I front Oh proposed Enuironmental Stud,y, Main report the Oh Tedi nthre on sedirnentatiorl processes Environmental Impact Statement, and in the Fly Riuer arr.d its tributaries. supporting studies in seven volumes. Unpublished report to the Department Unpublished report to Ok Tedi Mining of Minerals and Energy Linaited, Papua New Guinea. and Offrce of Environment and Conservation, Port Moresby, Papua New Meador, J.P. (199f). The interaction of pH, Guinea. dissolved organic carbon, and total copper Pintz, W.S. (f984). in the determination of ionic copper and Oh Tedi - Euolution of a toxicity. Aquat. Toxicol. 19, 13-32. Third Warld niniltg project. Mining Journal Books, London, 206 p. Moore, J.W. & Ramamoorthy, S. (f984). Rernacle, J., Houba, C. (1982). Heauy metals hr, twtural waters. Springer & Ninane, J. Verlag, New York. Cadmium fate in bacterial microcosms. Water Air SoiI Pollut.18, 4b5-46b. Moore, M.V. & Winner, R.W. (f989). Relative Roberts. T.R. (f978). sensitivity of Ceriod,apfuua dubia laboratory An ichthyological survey of the FIy River tests and pond communities of zooplankton in Papua New Guinea with descriptions of new species. and benthos to chronic copper stress. Aquat. Smith. Contrib. ZooL Toxicol. 15, 311-330. 28l,l-72. Rush, P.M. Seegers, Ok Tedi Mining Limited (1987). Studies of & H.J. (f990). Ok Tedi the betr.thic faur.a of lowland (potamon) Copper-Gold Deposits. iru..F.8. Hughes (ed.) Geology of the Iocalities of th.e Ok Tedi and. Fly Riuer, with Mineral Deposits of Australia atr.d Papua New Guinea, pp.I'147-12b4. The reference to nfitirtg actiuities. ENV 87-1f . Australasian Institute of Mining and Ok Tedi Mining Limited (1989). Metallurgy, Melbourne. Supplenentary Enuironntental Salomons, ht uestigatiort s, Volurnes I-III, W. & Eagle, A.M. (1990). unpublished. Hydrology, sedimentology and the fate of copper in mine-related discharges in the Fly Ok Tedi Mining Lirnited (1990). APL River system, Papua New Guinea. Scj. Compliarrce and Additional Monitoring Total Enuiron., 97 198, 31b-334. Prograrn, unpublished. Sholkovitz, E.R. & Copland, D. (1981). The Ok Tedi Mining Limited (1991). APL coagulation, solubility and adsorption Conpliartce and Additional Monitorhry: properties of Fe, Mn, Cu, Ni, Cd, Co and Artrtual Report 1990. Report to the humic acids in a river water. Geochin. Independent State ofPapua New Guinea. Cosnrcchim. Acta 45. 181-189. Ok Tedi Mining Limited (1993). APL Srnith, R.E.W. (f991). Biological investigations Compliarrce an d Additianal En virorr.nten ta.l into the impact of the Ok Tedi copper mine. Monitoring Prograrn: 1992 Anrtual Report. iru: D. Lawrence & T. Cansfield-Smith (eds.) An annual report to the Independent State Sustailmble Deuelopmerr.t for Traditional of Papua New Guinea as a part of OTML's Ittha.bitants of the Torres Strait Regiort, pp. environmental compliance monitoring 261-282. Great Barrier Reef Marine Park requirements. Authority, Townsville.

46 Smith, R.E.W. & Bakowa, K.A'. (1994). Williams, N.J., Kool, ILM. and Simpson' Utilization of floodplain water bodies by the R.D. (f99f). Copper toxicity to fiehes and fishes of the Fly River, Papua New Guinea. an extremely sensitive shrimp in relation to Mitt,. Intentat. Vereilu Lintn'ol. 24, 187-196. a potential Australian tropical pining waste seep. hdent'. J, En'uironmental (f98f). Aquatic Stumm, W. & Morgan, J.J. Studies 38, 165-180. Chendstry. Wiley and Sons, New York. R.W. & Owen, H-A. (1991). Toxicity (198f). Winner, Tipping, E. The adsorption of aquatic of copper to Chlanyd'o,nonfl's reinhordtii humic substances by iron oxides. Geoehim. (Chlorophyceae) and Ceriodapbtia d,ubia Costnochiut. Acta 45, l9f-199. (Crustaceae) in relation to changes in water United States Environtnental Protection chemistry of a freshwater pond. Aquat. Agency (1986). Quality Criteria for Wo,ter. Toxicol. 21, 157-170. Office of Water, Regulations and Standards, Washington, D.C.

47 Annex:

Table A1: Arwlytical results for sedintents front flood.plaht sites in the Midd,le and, Lower Fly Riuer region, where no d,epositiort, of mine-deriued, material was d.etected., and. of sedimetLts deposited ilr, the pre-mhirtg period. Sediments with. no grain size a.re <20 n. Statistics were calculated for the fraction <100 m. tl Ill" ee€ EEEE l-l* e=a Fcece E cceeFe EEEE EEEE eeI fecee E EEEE-E EEE= :t- 9n- =:-6-:* _;__ :l FFh €==FP q heSPA= e--- :l NdNda;; ;ix.l=R :l gH= qF5=E A ESFXFg Efi== -FA n =ulal--- F;__ :l ssxSG g *;€egF =FF =s,,€F :l es8 -l =E= =EEqR EF€= I gRF gEEFd = xFFF=g=EE€FF gssg = Ntui_ -l s ;ee*pE ;=E= ;l = F==s=R ns== :l gsga =ea F€s 3 FFRS;= ;F=: ;l =.-=-. ---* G:."fN dc;c; Jeidc; ": *l^ ;;; cje;dic; e jcje;e;i==.:-rq] ;;;; EIE EEE EEEE EEE E=EFE E EEEEEE cec* ^:.:..:r:.:..:.:.tN- .:.:.:.:.:.:-:^rdNN a oodd :l -,-<;c; ooo€oo c;c;e;c, =-- R== -c6Q6 s ==-=== *E3 €;l AF:gR FgF E AgRRSF e;4=* ol tSBFe -l ==s =85;- FSEg= F €3S€*8 R-83 s ==€ Eq== ;5= =l -h6 apsEgg F€H= aSaBe DI E=F FEF= E=F = gEFFEE -l ;=.j;j ;1== + ;G=ti'=X =€gRE;fr= I FAE EgEE ACB EEaga = e F=F 6=€e ==F GS==F F FFFEgE FF=F :l EFE REEg aaaaa E F=EEEF AFEB = ==F=5= =S=i :t EFE EEEE EgF==E o6000 E gF=€ BAAAe a a3e€aB SEAE I E=E EEgg E;E F==FF =EEEEE AE== EEC aeEB a-aa BAEAA ======h ;:1" s=seF :xF Esg=EEaa€aaa aSas 'l- Fr F".? =*B=6 -- ,+ E "E ;E -=*J .L..:= I j #E=o.ae N.EEE>= E >g qe E .E E d^ - I :e_= € a^= ;c_3q Ea_= =O -^ t I ? 6-ia^3=3e iqa-e FFI EFF EFFE 5EF '-6566 TRg -Rs-: €6J3-- E .'i^i^.': =F'+ -^*== .til ;;; ; =:E=** gil i= a- F€ F=EEs

48 Table Al corr,tirt'ued

E €E EE E E= Eg E = E EE EE= EE ;-= E o 3e E I fe E EE ecE EE E 3e =3 = da o 6o t ;\t= - i- EE E E g= s s* s; == = =I = == €s;- a 3q a 5 Fq E Ai E= EE g E=

€r oG F:{6F a r ======3E F €g FE E= A EE Z A= E =E H E= E €= ER Rs a- r E= E =E == = F !4'- cf F F= X= e N Nd = =R == = == o 5+ n o ======5 = € I 99 u -- 9V g €g 6 ;F F= p sa o EH ; h8 = =g *a * 3- eA 6 V -3- - -F 2 =* := :: : 5 T: T := : : =: = =T =: - 9e E EE E E E* E EE EE EE * EE r:- f,f, ovwovo - e: : : : := ;f, l: : E =3 n 66 a R€ 6 -r! - 9= = =n = =s == g gs g t gs Hg F= d= a 6S s =g = a F= E 3E ss € EE FF ======E gF T EE FR E E= E EE = =F = == o FE 3 E F E= EE EF EE E EE EE R EF =E EF-G E= EFEE a FE EE = ==EF g =F HH F F= = =E=8 = gE E€ g E= d A FE E F FF R EE a FE g gE EE F EE E E E EE AF A =F == E= EE E E F= E= E= E Fg EF =F FF =E ==EE E Es E=E EF 3 * q = =F 6+ i F F ; r:B F , E i.: s ;* -; I : 5R:: ,= .A .,!. A: t iI f": i:!o- =s 3; ?.s ":'i^3q 2+; E. E E ?- Z ;: eE €F7a I E€;> ;n Z €i E ; €E E $: E*=3d 'EE r B E; € Ef : - €i E E= 3 :: @ i= f ; i ;E iFE l; 6 E =; =; = = == a ===:i=!ji===;;ji=**33 3 33 S 3 33 = s* * * ss s's-'E

49 Table AI cottthtued,

ooo E. €€€t €€ a8eR EE ceeeee s€ Ro ssRBoeele

€€ ro*?.-1s u €€EE i €€€€e--eee €€ EE6eegN-= =€ -":8

o da-o oo-a6 ed € -<

6ts€6dd 6666F H €6Ndd6- 6

o€o- 666AN-O € 600a<

6F-{G6 €€€{H

N--€Od odsd6r6Fr ts O-FFF@O

000006 oo6600000 oooooo QQOOOGOOO rs 6

6-dF 6€-666- o--606 {N di-NN

66tsrotsd - *-Nd

o d666€6 66€-6O@a- o AN-diNN

o d- d-d6F-Nd+ E v

aN€oo-€d6 -666-O €€€€€F@

E €N Fd AgA--

6iN€ g oo606€ -ooovvvoo

a EE€ &o

6- 6? €o oooooo o600voooo

5 6-€ONN6 -oo6 ---d?

t €€:604 o-do6-F o

E *o6660 €_ €r€66tsOi€

65-6 NNON6 6-6 @€oN E d€ 6-d-a6- ooo oooooo ga a qe--N+E€- aa aaaaa aa - €6€ o€@d o€oooo €ooooo660--6€€€€6F oooooo eoooGoo 66€:- aaa 6A ea €- 666-6 ooo oooooo 5F€ a 8A oo aa aa aaaaa oa 6F- €€€- 66666 oeo e€oooo ooooooooo g ooo ooooooo €€€FF oooooe oooooeo tsFOaa3 aa a dNN6dO ooooo60 H--dd 6€OOOO ooo a a oooooo eooo E AE. :1-!B -666+E6 !E a -EE dro BB 96O EE oo- i E6 B €6 rl otF ':E A B I o a NOdO-- -ie -NE q 4g e6 6u .E-il F =u -6 bl d>6F.E - €F=E € a E": 9= Eg d g": 'rg-t J E EI,- EFC9 F.A g.A EE!BA= ==qr o .E E E *a o i.Et-== EE A_dN -r - €_ j*; JJ F €Hj j E39? a { ddN- HH> E FRsE H> =;eo oo60000 E ooe .-: i 6 <-dd- a d\\\ \\ \\ \\\\\

50 Table AI cotttitr.ued

R EE E EE EF EE E A- E E F EEEE .: EE E ; E E EEEE E o EE E EE = = == 6 6 I e e92 6- R ======g h E 3=== EE E EE ======RE6+ E dF a:8 E= E 3 €EFE = = = = o€ Fq: u-f E 2 FF!== j i.; --r = = g t GgRg s Bg s s3 Fs= S€ a = 3 gEEg 66 E E F E EE E EE EF €F = = s aE E tra aFe E= =g E R a g==E = s : 3 3 R =R 3 FE ======R3= 6 6 0 6 66 5 I I ======F=- 6 N-dd o FN 6- g E $ vv - gss= k gg s a s R 3 a E B EA F3 =6 F-e E iL -* V VV VV --V V - -: n-: -:33 -:F: -- 3 5 vdevvvvvodo = =:3 ==== € E E* C EE EE EE E E E E E EEEE s:: n n sl t':nr: -: tr: -- : :v- ci Q o ooaa E v oe - -<:vo \tJ oo o € rF-rr2-6F===Fa==-== e a I !B s astcs a s e6 e s= == E€ a - E E= 5 gE g= E F E *=EF =a aa s =€aB EE 3E E B a= = a a aasg o E; F= + €F++ E= E E E = = = =*E= H 5EHH ;E XeE= FRHE ==FE FB 3 FE € Fs E6 EFEEd=gd = = = tra eee; a FEaB aE eEEa E F a F a Es E EFEE EE = = =Ea* EE=F - E FF A EF F F E E F c aa a =F€s EE ==B.q g a E 3 3F C E Asaasses - E EE E= =5 == = E E 33 =a 8a i;H3= eE a a e B g ; EF F= = =?a==-== z =s BE-Eg-8 = = = = = 6s;i 3"9 E Ii*:$ i + s .E .E= g E € : ; q x E== E g BuEu - z fi: ; F: E5 + a : F F s4 E g : ts: s:ai E* EE n q: ; ; :E6 ;Ii! r --: = -=:- i F HHP> . ?= '-:-; =E..rd d= 4 ": ''! ? ai'iia 6E = = t **3s4s$3s =ss33F33*

51 Table AI continu.ed

g E EEg E F 3 € €- E E F E E EEE EE_ o : E ; 5e e E E E EEE =e3 = = ==ii o A g * = ==R =x = 3 = = = =-- sR o 3=I F3 rn E E s B== F= = = = = a F=g s xd 6. g H gx =AJ; =; E = =e=

d d 3F= R R 3 F= R gR = x = F =-- h agB p a g =3 tr a a R a sb =n9 EEE E +EE -.i g € E F g€g Eg = E -; E s s g e A^6= e 3 5 rFs =6= -= N a ==dN RFS E R x= s - = F R 5 = =-- == go ==- =

+NN -f d N s vV- 'j k o *3a h a a B= € € a B s E== =R 5 V *V d 6€ =V.i ;: V V .:F:.]-:.:'1 5 vev o e d .rl ; f, 3 : : :: =3: = EEE E E E EE E E S E E EEE EE .: a --::vvv o: o -' ---- J -q f, : ::: :: = a R R= S E =F= = = = = F-- =R RSF 3 A e ex a s g a = a R-- =B E EEg E sE E E E 3-======ji 3E= R g E a ss======q o i. EF FEF--^t E ;E E -..F= E;-i F E 5R RF Si{tuE=F 3==EEE aF *F F EE E = Fg € E Heg =R E = H E== =H a3E E a F == E3 =a E= € a a= 8Ea aa=E o se= R F F= F E H E-- = = = sE - FgF ; E € F € E F E== eF 8B E=E E= F EE=F F E GE GF i;-_EEE EeF F= aa E= = a a aaB == ==s s F= = = E E X E;<;< 8aFE = BJ = =- = = = El;;e5:E=:efTi=6-E:s.Eg Eh f a a F - i B'T g o J= "r ..* g A *;+ gin a t +( : 6; T ".i4l- = E- ;I 6= ,= ; ;' : : E $ E ^-.li F1 E : d E F : d ; il 7= He d F E E**=" f ff;**= : E Ef € Lw 5 EB F 4 r ;S* I > Ee 5 :,-:*.:.i =F.! ': ? a q= I:: 5 :l: d <; =:=- J= J : ': ': '':^: a SSSSSSSS=SSSSSSS= - = =: = S*

52 Table AI cotltitLued

ee EEEE= EEE e 3==€ EE CE ;e =

GFsr.t R======Es ssss R E=s5 gg 3 =s= = Fs =F6= E E=E R €=E=

d RRS 6 SBSF ad =S =g*E = EE ERE= F FFF RB R = =EEE > a3 €EFq E REF F EE=E

-e6q 5==F = = =sg = o Node t:-a e 9PoF o Ndd AO -o --gV V -U- --U? g6 G g€AC F6€ U FgEE = €B€ * €'- F;=vi3 -:i *- F E =v:+ v 66 .-".? -1"?-:.: -: ':"?": : 3nF:n ovv b ovvvvovaooovovv + a cEd = E EE EEE+ E EEE E EEEE EE a+ u:- ---<: - '-"-: I ---- vvv 9 eovvvo<;cioovvvv @-6 a 6-a.t E=RS -- = >== = 6< 4 AAg't g F=g fi 3=3= o bH s qFE G 6€g= ER R e !8= =aF; =5 ;R B E s= e=3d 6=q E - =- =^- =ssE a aa aaEa a E8E A 8388 aa }sSaa S E E=F RS N- o EF E=F= Fi;td -E ESSE ooo = R N6a8 e e€e= FFE FEEE o€ EE=E HEEA s xbE F? h=s= =a - = aa a8 A Eg EEER E EgF ? €FF3 g3 G ooo gg R g RE F EFF E EE=3 @ BB ssAA a aE8 E E=aa ag BA a =EEE E*a gBs= RF -d - Bs =hl:= = e sEE = EsFq BB a8 B a3E= aaaatr==F T .e R=s= F === E Els 5E E Fl e eEi 6 E 6 ,.tE! € o o ; Ei-a'- H og ="-tag <; : I.rusi ; 4 F ": Es -l$ ,A J.-?A= .;- +' - f J: -- E= 7^*E^ 3 seq€ EE @ : a: re! ;E =73i_= dB;;t ; E6= 6: +r ts+r g H= AUEEg E E":s 4 fi_vfr a6 JETF8 .E a,; .iJ si EiEEH g s_eqs .a d'- F ii;;i i :ii e i-Ji E'= i: :;: : atr HE 'Er HE;== H tr -*E> H> 5 =E :JJJr ..! ==E-!d^r - -a-: H ;;;;; J JJ+ a6 q € =={{-JJ a

53 TahlcAI cotltitund

E=E E EEE flE 9€

ef,c E EEE dE €EGE xTF = ssR.F rh gg= Fi ca' Ne' ==E HH€, F HSF

s6qRNd N @ G6,d -*-

EgH 53 BF?g 4* €€ E',€g € Ege qa H5F E gg5

*ErNNE E -6+

6Fd 6 H ooo a- a€

e QrEYVVVVW ddN, i,i j

dFh FEF lC- --6(dCF

vvvvvvw ++d

,.-: *? .€ ..: cr -:, Fl 6:n-: 'v€vvvvv- n= e F.€

tj .d-; EEE E EEE €€q4 €€6e +-* 6-- vvvvFtvv -

FO- @GE NdN dN* -d s,Eg R 5R!9 F6 aE E R= g tE= A9 == ;Si 'EE E$FtEi RR *R ,-1 4l a gaa 3E! gss B:88E€e SIPE eg gg3A g= iF- d- FF EFENo.n d Nh6 FE Eg H66 6F ai€ '--n =gE -- -F aEE Belt AE f:e!8a8a a8 de 888E' EFE FE= FE FE E€a a= g=aa8a ta 8a :i:l= Sl == e= esg a aB8 feFF3EFE E 3H= .a8St= cF8a EOEBE rrF T BB E6 6 e1' n oi tii{ rD 66.d €fl 6 o- g:!3 = .As: Fi si 1s A :- E--* ; Ei8 -A -E &Ife . i6!! E€ B !g .5:_= : :'3q s= -ds. F E=j E5 s: Ee 3 5s:H>r >b a F:-- E--H ---!.,!!:- - -- 4 € --:- -- -- e.! -E*g 1E q-R. q. €.= q=. q. <-q. <-<- Slot HA -+6.6 -q- the Toblc il!]: Atml.y1;col resrrlfs lor riuer banlt sed.inten'ts and suspen'd,ed solid,s from (Ippe,r Ok Tpct;ftiurr.. Sedi.nrcrr.ts utith. no grain size are <20 m. Statistics were calculated l'or the |roctiort' <100 trt.

€ EE F aaaa g EE= EEEEE :=E,'E= ECE=E ;E E=: EEE:? :-E=E EEE=E -rj-:n a = i :-* 66*-6

R EEE F*NE€ *E€C ====d€€6€6 F E==s* =sE =H=S dao6F = sR= Rgxg> ==== 6O6€€ -FGFr E EEE EE== 8R3RBaeooo =EE=E E3=3F -6-F@ = =Es ===s rdo€@o-o -o606 F

N66o €F-45 o iNaN-@ts6-H o =- FaFi; L gcts== 6O6r+ - F ag= =5== o6-6- gF==

oQoooo60 H66: n 9EEr 3-eto -N< EE FEE -h6-+a-e o diaH A== €d -!:-: -EE=E N,eee e 6F€g q 6-€ aF E€E€ rjFgsj

OFE6 90aN d€6- -.6 t6 6d6O b 6S66 6-+ FF-F

€il66 6d:F Fd d

a aG r- -FiFdHNH -6hdiH NNSd

Fk tsaFdi P6€ Eaql

e6€at €oe oQ eoQc B' d-HdTRRA REE A3

q66 F6 F =abB -q€€6 *-ed6 R€ - =.-- gt@ 6G6a *?+, {t+

a 9F9A FH€ NO -, FF6E NNd QGr Ald 5 6{d- 6<- €d {.{ 6ra'a - *AT E€

4'Q€ry .:tf €,d-o?.:a- 900-Fts -6 -a oded at EEE EE EEEE e*-d 1= *FOd .e-B EI doHo €a:-9- cJ g€ €a€€

€-F a€d- 5 E5d}E fi- d€dE -*a F

6€6. €o6-d FFNH 6€€d6 -N E E F:A=E

Fd d€r-- p N'JF d€ N: 6 6 +

OFEe e E6 €e 6ldFdt A6N d6 E 6dN6d

ataE 6-At8a .td ag aaE'g o 53=g dOH'N€ Ndd FI 6--- ggs= €eq oo e9ee NN€aaa aa a9sa FFts€ts 9€€€ eo€o €eo oo EgaB F@-+€ABE Rg FtsF- od 6F

oe66 6e,o 8e @Ooe eoo +6 F€€O Fdo €_o g F9= 4€ F€56 'eB: AESB €6d6-BAB BA Elg 6S€ F- E N+

aaaa 50+tBB €dEB h66€6 GqF9_ G H<- -d 6 E Ei 4vd- o ea P aaag FH 6 '!i o !i. E u. E F - -ER*o I o - 6eER=-e a CBPf E oo=' €:o 6f i'& 9.,9 9r Egga 8EE EE JiiN€n EO g - .eq -e € G! t +'J €g I N' SolE =e -b-

56 antd,I'ower F-Iy Riuer To.ble AJ: Atw.lytical results for flood.plain sites in' the middle n'o groin siz'e giuen Regiotr., where iepositiott of mhrc matLrial occured. Sed,iments with u,ie <20 r,.- Sto.iists were calculated for the fractiorts <100 m otily.

E EEEE E EEEE EE E E EEE'lcE j EEEEz gF C EEEE E T-:=EF- E: E = d6€a€

-aaa :9e=B * F F =n EEEF E= ==EEF F € * EEHEs =x=g EESH SE = 6 * ===R == = =::=s

J6€F€ + 6==a =g = - g SEEE E EEFF F€ F E EE==E e 3 g= EEsSE ==s5 ==E= = = 3:-=5 E - -6F6 = € = =9=S s = =-*== o €de- P--= s s;--d - == g;RE g 3€gga * €nF3 s ;= = € !be d-e€ R= qn._*.: ee .: - \e!:n':- : 6000 o€vvo =::: = Ps 2 E EEEE E EEEE EE E E EEEEE u: o-' :.---- q -':': a - vvvov -1^:':':':vee6€a ooee s g 955=g I e S-t3= ; E==8 A= gaBgsS gR€s g J a s e gEgE E5 = e ti G==€ B FF *88:€ E3E= = = SBg=E € F *=FE F EnaE 3 = =s BA EEE= EE E E EEEEF E 5 a8 ESEE =EFEEFFE EE F EEF=F F BA = =E EEEE= E FEEF E EgEE EE F AR =F E=CFF g d GFF€ FEEF EE = =€gF = BA R E E=EE E FFEE= =EEE E E=EF =E E =F EEEE= F FEEF EF Ee 6g -U .q-a 8-:6 8:E ="N i'a€ "tT.ta 2 E E 6+ .; EE E E g8!i Jex e=E-a :; 6 E €-8 EE^e+ ; a,ai tI E E E*.-"5"= g gFF= g- a =H sE=s ;: H_ f -s f ?; : lseE .j .: .-i -i a -..i.64i:;;<. ^r ^: =EEg= 5a SE 333 SS S =S 3

57 Table A3 contirtued. ll oo 9 9 EFcEc EEE F= F: E E eFcE E E l- l* = .; 33=5e EEE i= ie E e5cl E E E = = l;f 56e-66€a+oq g==ER dBg s= Hg a 3 l-l <<- = ==Rg G E sF=qE 8=s g ada €E R!8 6a s +_6FsBex ;x R -h-*+ -- l;l o-666 g6 6 q € N o --Nd6 -N

q G€6 N 6d66 S = l:l FE€EE FFE F= EE E F E EEEF F 5s= F= R g= =E=FrddA ta H e l:l - =3 = =F*S 6-F€F €6 F6 6 Ei g! €6N€ HHiid ; l=l -Fa _dr< o€66H -d- 6oF6 € == = = l;l ooo-56F__666 60Nd- € c! ts+6N E;d] ;- l'l - l=l *a-3s i-= *- v_v v=vv i 6-Fr€dN6 : j-i i .-< qo:?g: .1 oooed 6oo oo o - - -t *l_ -o oo€o o t€ =lE €eeee s== EE ftE E E E EEEE * E ;I :i:;: ::: :: := : i :=:: : *l = = o *ES5= EE* EE E E T F "l =5 ==SH = FsF=s sRF Fs F3 € E .FRF h F "l =B NNNN Es€Ea RFR F6 € s R RAss d; - 6€66 s =F == = = "l FaEaa gF€ gE EE E a E aaaa a E ol a=e=E E=a E EFFH R -N+-6d =E =E -l aaaea aaa = = = aFEe= 3a 3a a E E EFs ;!l FE qF LE 5 o€6o 6 =l ; esFF=FF= = Eiaaa€ s=H Eaa Ea a3 € E F aaaa a 3 -l Ea€=a r3rdci R 4]., * FF== E -66-- == = ;l EEFSE F= EF E F FEFF E F 3s33= 8aa aa SeG=E= RRBa s3 E F= FER= g =eE€€ aJ ;F. €otsN -l =ii - = aEss a -le d6F;He= HREa 3A Ee A e E R F A==F 53 -l^ =g€Eg =H==Fg = E = E .F F-q o-sE - qE =o 5 5- 3 g ST T" O? J 6 B= E ;' s+ - i I ? I ]4 -5 ,=\ ts d €E !i ;dE= ;

58 Table AJ cotttirr'ued.

FEc E=c= - E E EEEE eeF:-:-E F=Fe= E E ue 3ee=8 EE EEEE eciEEe 3c3=e E : 3fe

N66No6 -=-= -6Na --======-- dxRg E=REF F gHE FEEH =€ =s€==;= = 3R S-;:3 H€SER= E===E * H g=g 33E*

d@oaeF: 3 s ==:*3 ==9 ===3 : F=sR 3= l;=gt ==Rgg = =s3 F E EEER ? EE EEFF =E=EF= =EE=A =EE 3= GssF €Rg;gRl =B======3= =3=E 6Fa€6O g s=-= z oo6r ==3=s = === e 6€@@ d66F rF5 I o -€iaN =-9 == gRggBR SSR gg== -o s= 5EA3 =S=-: = g==cB= g gSSg o gs asE= =R3=d = =3= €d€d v ivvv -!l --* vvii* = =--

a-:.?.1.1.-. '1 -'1 -:-.: oooooa":::: : ::: 3==: 5 ooFd-H €o-ooo = t E E EEEE EEEEEE EEE:-:-t.EEEEY-:.==E or-<6 cu-ts: ---- a-ar: d'9 J- -;.;.; eoee € n.'t oeoo aaa =-":':=- ==;';J ADi==R€ e F a=a sge= a =g =ggg ==3eH EFE= - sa;E EE==gE gEE=5 F E F€E =n I F*E= - *= E=R= B=enH= EE=s= = =gF 3 S FEE EF€= 5e €= SEgH EF=EEE *=E3E g FEF= o FREEF E A== =q66 ==== A=A=EE Eg aBsE Sg NEFE EEFEF F F EEE 3A aaaa Baaaaa =AE=EE=E 5€ ESFH =AE=FF FE==F E EFE =5=FFs = aF= eeEa 6@6 F g FE FFEE =FAgaF =E==E = EE EFA=EF E E E=EA a a a =CEEFE=E =g= ==E=F e F =F= E==F - E= =ESFs= =EF=F6FEl =EE 6 - €-oa i si E 3 1!g E E E E e -l;a E H6 E3- E i * SE Efr'E 6eti=eT .=-T=s=.E EL*l;" g"i"f FqE a "i, _ Z 4aeLeFu F:8 F iT jf vf eF.-:=EE=- E EE;El E E Ei= E=E= -i -RzRq oo9 7 a 7 ; 'i d '--'-: - E 77 A -El i ;== € SSSS {- \\\ SssS** 33s

59 Table A3 contiltued.

oo o6 h€€F gF== €qe- E E EE EE E EEEE =5=E ; .G€ €. o oeeo =3e3 E : EE I3== EE E EEEE I

o F66? O-16-€6tsA@a\

+o66ots-6 !t9srg B S g5 ag3= =g a Fs;=Fj

o€6€ n€6N dd€N-666 -i-6;-.i G_

666e 99f G 6 €6<6 6--N NNY1 Nda- @- e€-€ - -d - -_;_

arO_ 6+N€ 66-F No €st€;ia i;s F=Bg ooe€ oooo ts -aoN FFF= E E F==F FF F €A== s-a- =E aa

€ -q€€ Ndry4 N6N- iiru - --dN o --iqqo € reu'! nna6 €o o de€6 o -5€d4 r-

o 9ddP d6 - -€€€Fed- F6Fdi

FFN€ €€Fl 64N654€6eao o 6 6- 6i6-6 i.i6 -6A6 3-*= s&-a o 6F li-*- ta __a!i -vvH v-v- a - q q .:.-:-:": .: t -=e-odHo€oovvv-voedoo ?: e-: -:-:a.'l + = EEEE A= =EEE E E EE EEEE E EEEE d €Oo6--d-6F+N€-@6 o oooovoovvv6eoeo E E €= € €aaa ==F= Ag = Hg=g

€6€+€6--

e sRRb $ R qE sEEa

60 Table AS corutirt'ued.

o oo oe EL seSs E€ EE €€€3s-oo2 AeedeeeEq CCCEEFE e 3 e e R eR e e e E -€ oo.!e €€ e€ dd€Y-oo-t? o E CE G E E G E eeeoeee-a =f=cE=3==e e3

iFdaed o€E-€o6r@ dFd-da5Ee+ a ;i dada i

ON€6 od6 FN+OO@-O NoHFO€O6€ Rrt=aaeRBF:g 5d id ei6 -=6666--N6

€FFT+F e6€N60 sR==E=B8Ss a ----66

N6€--€ 66FAFF e F 6 6 N O -@ .NdNNNd-N -----ddN-6 --NNdA

oF-n €€ o6F6 66FNOOr6 -o6€€o --NF-6+-r -d665-< 60000000 oooooooeo ooo€eooo9o € ooooeooo €-d6€F C F n=-tu-- d -Nd-Fd =====ga=a =E=RFe3a3= N-O- 60 c=s=5R=€= b-6---e9--

o-e€o-9 6 ;\i;.Fi-.:-dQdd - -oal€d+€dNdd -dN F€FONOO oo 60F€d6 6€n66Nded

o do€o€-5- oooN-F-F6 -*-6-d-+a - -d-*

H N6-A5€ 6N-Fo66e+ g==g==EB=d I 5€F6F€ * -€<--n

6 66hV----v 3-e-;i'--!i*o -d€-NNNd v- vv-v- -

6ldFaa66 -:f:.:o:F:-.:--:g'! ': ??":--:.:-:-:e: daO O .g aoaaeood odeoeoooe -OOO € E tsos Eo €E - 6tGeeee= EEEEEE E E E EeEEEEr-EEE E

4666 dA 6d .:-a.:'':-F:':- -: -:s:nr?F:":u:-9F: e €o eo600eoo ooeooaeev do-6€ooeo6

6d g a agss==sieR :s;gss==aEgd

r€ a 6-FO aggS=sas= gesEsssEFs E dN €6@€-

G oH6050 ;=R=Sgg=RA S -66F6+N-HdAA =Fs3Etgs33gNdd --tuN:dda-d

_O E _o E €€--€66€ =ara=s5g==N6< =g==r=*==g65A6€6€6F5 - saaaaaaaaE F 8A3A 3A H6A8 aaaaaaaSa F€€€ *etr=ag$Raa E o dN b3!34=gasE aaaa aa aa aasSgaaat saaaaSaaSS-N-6--666? a gF tr[sSesaalQgR E6 dBaS S5 6€Fa66sa3EB56sga €€66 =€Feoi66r6 8s8A A3 Ag aagaaaBaS saaaaaaEaa a tu-66 6€ SdpEEsFAC a l-6 =€EeRsEgdaa/a = oooeoooo 8A88AA3Aaa a ;T€OFFa6o€6OOe E€a=Er;dF6S E g dd FEFFEFE=E oo ooee SAAsAAAAt aa=aSaaaaa a -6-€ s-*=Rat€€ra s b+-< =3€=d5=>38 --: oooooooo asaaaaaaaE E o600 3R==hs=Rfrsaasgaaac g E eRge =E sls ==s=B=€aets - E B- o a I oEl di oT €t? E! A8 E!l=

A -rl Z; -e 6=: =r*-'d-A E== -"2-E a--B o $=€e Es. -B:?.83i3Eo - EaaStusrEI E' gs - =;=J,= F €A-+ .5s *ii:8 EF+€AF?3; iF:sfEt?Eg E =e++ -==--Rg=E 3s=ER==gR; i -;<; -; -; -! cj.S^i ..i.'i "3 cS E5 = oi di-idi aioi d d € €€ o = === F E <\\\ F \\\===

61 Table A3 continued..

oooooo oo o €EE€E€d6€

o NNFON6@6A66 a -6N-NOd-oNNo

16€6€-O o-Qo6dr o<€-€€Noo r6f.6- <-66+6 5666*665

6€- 9EENhd+r€€€a E NO66F€F6--d<

€F604€d6€€N rG€oo

e -F66< -6N eooo€ooo€€e60 ooooeooooooo oo€oo600000eo oooeo 6-€6-O66+€€ F N-NdN -dd

6€€€F -6€606d FN€65 i--dd

NO€<6€-e€€56 N6€NOF€O-6dF - NdH--Hd NH_NdiHNHdd_

o oots€o€6 o d

o frN€FOO 6€d€6€6€O - a6--N- 66666N--6

d€604?6€ €€6F666FA€6O q +-N

6 6d---;:. g3* A €::do*€€<-+- vddv

66rNAdO6- ".:.:.:-:".:.: 5 oeo60i/o o d4td-ooo- -=-a.: + E€€g€EEEE€N€E g€€q€EE ?-::: = eeeeeeeeeeNee €e eEe=eleoeeee

€F@-F 6d€E6F€d6F6 o ooeoooovevvvw e-eeeooEdooo .€

g €O6NN6TF€ 6NN€dO€FOOq€ Hi

o t F606 €o6d+-tsts6 €-FFA- 6<€€6€F6F-6NtsOrd€4

6d66G€6F6dN €-6+6€N E ONts€HN6€€€6J r66€ts€6€€€FN N€N 6FNNNNN-NNNN

H€OF6O 660FF€<€6 e6 6€6€d 5i €d66€€+dd6N @oaFa€r€€ oeooo oooooeooo600 ooooooooo ooooooeooooo 664-O*NFO€66 €d€-€ 3N--6 oooo oo6ooo€o€ ooooooooeooo ooooeoooo600 ooo€o6eo66€ oro€- 6 €-N 6r66€ €6€€€rFFO eooooo€ooo€o oeeooooooooo a 6-+6 eeooo o€€r I -F€-€O-€6 6€r-4 €6-r6r606€6€ eoooooooooooe ooooooooooo6 oooeooooo€o qN6s4r6nNN€€ - €F€ oo600€o6000 ooooo€o€ oeoo d€o@-6 €a€:60- -6* oooooo oooooooeoooo NO666 €6-$E--ON€+O = 6€N @HdFF6€6N6€6 EEE EB ' Ei-uh-E El" t O ;e EI ?5 ;t,A N t9 E <6 I a -E q e. -d o EE -F- - e a aE - I e E E I=- d'F a o g Eie-E E r66 -I g--9-- 6-E F.3- F ; *FE=->5 *.

62 Table A3 con'thtued.

o R=9 EE=e eFe E E e 3ee E --: F=F EE= ,E ? g ; 3 g:P -6 E e E 3ee c =-:EJ =-;Ee;;ie 3fe= E=E 2 o 9-- =-= ==-* ======: ==: A B gs? 6 EAE E*E= 853 ==-/ = =€= =FE !e 3Rg E E== =F3 =;.i --d €=F H5e =83 =FeE = psR sFS R 5 RFR E=9 F=R =R== = = * g s dgg a HE= 55F ==R =3== ==R E=F € € E E H FF= EEF EEE =FFF =FF8Sa 3 5=R icE a€g = = R E==!; =EE F = = R:= =R: ==x ======@66F o g-- €eG I s=s = o -=- ==- = = F:P o s=R g==- - === =Rs ==g = = g A A FSg F o 3FS 5g5 gEE 3s3= 333

6 8-Yv : @ *s* -=g *v= =*i === rP ':': ': ': q9- J c; o'e eae F ---oo€ odo =a= =5i-: -c;t; -oo- =:= CEEEEEE d Hni EEi *== E€e= eE= 6: ':-^: ": - o oee ; - €0 :if, :=: ::f, :=:'= ::: g F EIB= I n geR sEE EsF HS== 8== = E E5= € E5g = =

OJ Toble A3 continued,.

e9 o oo aE. = 9tr Br E=ce E==EEE E Fee FEeE=F dFEE e o jcec o.'so eE A E==EEE : iee 5le3=3 oEcde

o 4 €+--Fgr NF6 o_€ =reF.P - ai- -o6

a F=F5 H==qEE F€ = 5=F E5E3=H SEESEE E EiiaB s _=ira3sF RssE=R -o- -

6Fts@ d o-€ 6O€ o- -.q6 ; _i\i= -€ k r>-ne 6d€€-!<9 o 66-€+€ + f;i\i- 166 96:

999e o€oooo o ooo eea€oo oooeo AFEE o€ooo ts AAEREg GiF E=E F====H €N -dd -;_<= dNH tseEq E5s=;sE p a== si=s=5a d6 - .:a ;. ;i €€

sRFS R:=s=s dd - = *== ==R==- o clnio eF66€O €6_ 6d o -o

o R9s= g===s= g== €h69d - = *FE=g= 66o€--!--6+6d€ o EGF€F-r-N- -6- --N6F

€ +6-- - - ! :j;.-=*-* F666d =-- ===:-6 NN-N o6F--!n 6 ooh 56ts €F6EF €;.;-;-; ic;-j''-i-' ooooo E = ==J ==-_:=== a E EEEE eBeeEe E EEE EEEEEi 8oB dE== '::l! u?-::-- F? a.?- \?-g?are9 a €oeoovweovooooovoeoo veovo € E=3Et B 6;rFHs =RRsBao =nG 6CF = FEE"€H€- RBFpEEd€rN6d E a!sE==s oa a^ ilio€=Rs iese €€<€€ e <606- xeHH FFeFHqe4aadd---_n E it;5ssa N-r ==e NNd e 6€Nfl S== R=5€Bs *6€NN - ==Eg ==i;eEF = 6F; l_n__N o6000ooe 4 RaFa EeaEE€ a aaa aaaasa 650 HEE= FE=g=* E 5FF o6 eooeo g gEFE€F eoooo ERqAEE=E 358=s=EgEEEF r EFF Fe=5s;====F= =E= ooooeoooeo q EFFF ts60 EFEsE=dN- 8h €6- 6+666- qEEa aaF€Ee E E3e=FF =ggEEFaaasaa ooooo - FBEx aEB€aF s F€F :tFEF=F saaaEE a ooooo Faag EE8 BEaAsa aN6 F== ooo6d EEEfi FRGeEe 3 e-F e+qa==FEE 6N qasa aaB ooooo oo E. 8888 3=ESRxE€aE€.a Aa s*[3 asaasa : ==EF=F -6r6+ -qi --Ei=-a E E 5+En 9nB9'e I g E E_ F?-B":F : -68=i gElE :?;^Fl Fq- E E6RJ=i o g!e ":":!h.i-\F E -T.Liag _ oE-^o?-*J,-A i 4 J tA,e a 6 E *r E-:ratE: E i;;=;; bsa e E_":I EgSEEA ,i li"5 €'E'=EE= .= g'64 Fs F d*Ji g {6TFE= r-H-1;€;jjs4 ji jg.!1eE-=.c.a--:aE oE*-o€E* 'd9E E--HH - H;<;=E* HdG *=E>FF EA EV 5 { -3-.i iJi*JJ -S *N= € .j iJ i;;:-- -i -:JJ:J: .:-t E : a s 33 s=<

64 Table A3 cott'lrin'ued',

I e, l" EEE R? e 's:EE -":a "tr- 5e E. EE **-Fd?r han G5;-: :ggF g dd''Bd =gj3 :l - "l ERB -E g =l F=* Fsg s ,sl -*gF s'e S' -q FgE EH A €a *id h se8d FF :l -a SRs F.E S :l .aF= Fa d ryr qi El -dgA A-,

doid 6e 4 tri-€ +r

"l s +6' w- 3-

"l .: .: !? !g r-1 {s-!- -, :9 5l o y v €,9 Y ,r-i -d6 €da EEE Ei€ €€ae

,= fE a 'a€o a- ==-r.vv€6 -ul =? R rq.Et HeiB =g Hl l:l €E ss* FF= = E3 E =Ee =f;H g'FS R .5!A= =F E=B l:l E E,Egr 'EE de ==a .aa edE l:l EHE Eg A F=d g3 s! f3 cl FEF Efi 'E 8A €e f=g F',8 'H 'B8 AB I'l E'F' g E E=El o€ eB ee IA'=E FE B BFXts=FA oE 6a6 a T EIEG: R ! I'l- gfi : 3 a ra Ahaq d -- d E -E8g OQ Ll FIl E=€ E,u € d.E 4e ?!E FF He eF3 *:tF: 5 3R 'a"+ -, + EE lgl =-!.q-q R3g+* E,€

65 Ta.ble A4: Attalyses of the waters of the Fly and, Oh Tedi Riuers. Medians, meatls and stardard deuiatiort. only giuetl where appropriate.

I I r50n o fr rri ql otf):6tI] g €E€ . .=€€E E-€€6-- [email protected] oNo 8l E= ol I o at.('E€N€rI)€s\.OE F! co

e6e.sn66ndch€ @ratF e o rF€ 66 o d-Hd\zVvVFE <-Nvv := -o |r||tr) g]!nnOt'l 6)r) 6 -cl ' .E= -o A vvEEvvvvv vvv A. vvvvvE -€

oC:oOo-

E dtsF6cl\q6For-€ cn!oq' E ro n F66 N HiVNVE N vvvd OrCD

E N€6r

ooForr)9rfi€6- NNO d) miidNm\oa-tcod.e FO- q) r'o€Q < Ir. ivv-e t- €!.o- Orn'io

dmdotc|€r,.clHdN cr ro co d) cO EOF\I) COiJ) t{ q'!=€ovrfrr9\.olt)€ Fmr) l{ nnco---g tt) NN:Nd-HdHHE c/) N--O-9H

ooeooae9€e F9Q aooQooe 6NclrOF€scocnc'\ FIFOI \c rtl(ts E'r €tl1dse-oec!dE, rU.)s--N -- - dHdddi----E N NNN Nd ooooaoeoeo o:o QOOQOEQ ooooaaoaoo or}o a e eoe Qa NOeO\rsio-o|r o\5a NrFf)

>r lrr

IJ]eFnc'rcOOFrHF !-l Fc{F dcoro-cr\ o\F A. FOOFFFFFFF@FF-F A. q; .:qi* - N-g60c'rr.ooNrc\, q. ;E N(^ocn E @< o) !o\ororo@o\oe \oct\ @d6 o,) CO NNNCINNN--NNNN NEEF E€E

G o c)E 6 gr o, 9r CJ k-y - r -oo\ €J c) C) oeo .o i=

oo ottd, T,ab1e Ai, A6olryses af th,e waters af tlve ftlt River f,l,oodpl,a'in. fuIeiliww-rneanw sam9fi4 stard,ard, ttreui;a-tiip, otily gi,;nen uhere ap,prop-rioate. NF rtearus wtfulteted' uatar

- aC :ct 6iddd H ga==F,=oFEtsF€-*EE hch^t rn-r :Erv FIOF F]-':''::o n=B *l ffi I ]] ll "e g ooeoo€=EiE=i=-:q -iide a:e e c6 aB6 h ul s'! s'! :=a' ,:EdciiEee3-== 33ee gsERe:A SF 9 =3 33ee €oeeeeon(lq.G Qe tsQaa ======e33e=o--666,-H-re E :: d: r r;;;;; -q-;;Ez E6-. o. ,oRE9*-,*=t ii.O a4 u E =X= =F== rtl -o 9999e9u9,e9e9€ee9 € i?vd"tiuvv"itE?c * .t

EU ;;3i5:;;1 ;5:;;=;? F rn ifi h Ei F r'l 6 to q Fi iiv'Hr-vvvwtvvv-F n ctr !'! n $ 3{ ce16. rat St F -e evRgFSR:aSSs'6=IFp HSsSSsBASE.==EFEs ag -..-e a":ERF --SS-"= li4 caE I a SEFtr;s3HuB 6,8aEEE < N N + 6 I !q F < fi .1 19 fi !9 riO€ t .j ;'i.d m: d- N- N fit if 60 6l nt fr .\ F htd6l 4t Ff@ B 3H,gHEBH.H,*B*SEHRE ,- q@e N6d66d al' iEEFF, (J F,E$$$Fg$EFEE d) a\ dt @ectt ,;ra ESSESSEqHHFRFHee ai oi i5 NqHa,rQ!r e B EEEEEE=SFEEEEEES ta g dro<(tr @ =EEsEgHgEEsgEggN -H x* SS,r'ECFB--BS hBEF c, llte.e € E: GA €, a-- EI s ;;-Rs=']**== RsEFx a$rdN 6F gN'€s€ O€,ltls I s rfi CF a5 liot d !€A +u; d.j t; 6i.d d gi-d ut dd G G; .r;r E; s; r o6N<€€N.€a\€!{d'lrn- €6i€ ci sl (a .E sisRdBiRRs: s sRE= fre 9:i *... aE E EEEcFFasgns€f;c;Ee$s$a;ii:q; tr#Fg .:+icint ,o J; ',1.: o;:::-:l::al;-.r:;'iE + 6 6 6 F F F @ q| >n n,N a{ c E :i-i\<\\--{<<------8aE 0'F S S * B S S g a = =- ======

67 Ta'ble 46: Arwlyses rnixed waters. for Medianrs, tneons antd, stord,ar-d d.euiatiott ortly for filtered waters of the Middle Fly region.

..:

EJ € EEg oe€EEE a E-EE-.*E EC= E deceE a r- o €6-66 o6. 66N q J.JEE*;J jrqE-E E E =gE .; JEEE;E q o 6 NN ET eSeEEEE o -;EE<;E E 3E=EEE € oo 606 o o oo.-qE.n E3=EEEE o=eo- E 3E=EE= 6 5 E €N 666 z eIeeS=e o -;EE-;E € E 33=EE=

ON a NhEE€66 E qE EE E S€roc$ =EEAE EC Eeee o 9V*9999 606N o66iN66 ;E € 1'vwa4

.,q tdad r v.r''r' € dN s44rE4

€N€T-6NN O= 6N@N 66*o-*N v viN sv

dd4Adaa do-N 5 € E odHA-d O vv.-,v\,!'v veo-Ev o€.-,.-,vv\.

s666C-A NE €66: @€6F_ts6 -dvv!.< NdNd NN!, €oh66 R=3-=?- NF-6t- N€666€ E 6N@N s=--v!.N - N N6:r-OffN o---N*+ N60606 o:F-60N6 o @€- Ad6d<:6 - = 'Jd

< p E BS=eaB =SecFie SFEEFE c = vdeeev F-OO€hF-- .1 k :oo6F_€- l--€F-€ 5€6 O-6€Nd- a €dd 9-ar-€r-@ -Ndi-N €oooeoo r-o=oeo r €-dNNNFr o cooooo6 -dNd<6€ FEHH== F g*=-sE - -NNNNN o dH-- o e 6-:r-€+NRAA=AAa 3A=AAA e r-F-9@F-F-|5 a aaaEat dNNNNN AESREA 5 6N NiNNNdASS€[39

h ei=e€5= ;=EE=EE €e E 6NeeeN83sssS ooooeeo oe6000Q o oooooo€ s;sF3=gsg 46

E 6<6€r-r-G 6€€ .-: €€F-r-r-a-F. r:==--e r €@6€ @€@F_r-r-@ Gi C.F-N€6Nh €.oo€e 6@N6*N 5i €6€@66a- N€sE-.€ NNNNNNN E c;;s;u;6-; NEN ENNNNiT- o?

odH EOO Eee l! €ee E F F F.g €.8.8 6 H H ditaSg'EE- € -ooF. qe-=.==E Ezz.E-E€ . E E oE*= .E= U-g.E 6udd5E-s=iq- €d: -E E E L-! E--t E's'= i E €okk*o E €' g g e 3 i-oEd e.E.E.E 6 o= 'F"ESStst#E6 I E b.9-i F EO€'E€E€A E5E Ee e;:r gr-E b Jo .:= sE::s.5 N.NN I g..- .i 5-66..6 I €=I{;QdR'=i;F;.F; '': t|..: B-; g.- . a- 6Es {;E={..: F q F 4F 8-E 8--_: 8-R r R ER .==E. b;"-;-:- -:-:- E:- E :-:-:"-;"-:-:--=.jR=p= ----t-- =eFgE-g=-gsiH=-€=-E*- €-aF-----

68 Toble AG cotr'l.ilut'ed.

e uo E € p E E IJ E E EEE E E F E EEEE EEE o =EE - - E E € 6 € E o Je=EEEei=€EE € e e E EEEE =3e AA € € E 4 EEEE EEE E E Sce33eEEEEEE e E NON oo € E E o 3 E E E E E e E T-EEE EEE - 3 e c = =-:.:- rn € E € € E E i=ej!.eEEEEEE E E icee E:=E

4 E Ee EEE E Ea sEE==EEEEEEE F E E EEEE 6o€.1 o 6@F-e vd o ir-i=-'"?-?--?i66O56Or-Q-NOO v :E aa- 1 daA4 vv & -v'v!tYvEev!'='-'\' jiEg--:nn v!'\'!'v

€6€= F-€i€ N:66*6h6{-€6N o= v = -n- - -!"r'vv!' -"1-1- vvvv €a 4--:-:------,r'v\-\.!.!.vov'r'vv

F- 6€€€6 €o€@ E 6d-66€:66r--6 da € 2.jaa'-'/s,-N'/ o 66656 v !'N \' 9V9 -6S6666n!6666:-i-v -V t'r'w' v v v = -e FP E--- Nsa= @ ======gFiGS=xjr;:.a!'- ===R F=eFFeEE:=EEE EEEEE 5EEE EEE E

se=:? R=Gada H shg=F=3=E==e ==g5;a;h ---- = @ '- = = a6o-FOoooec- H==E r E===ssFRsR== --- - ===a=oe? =EE = ==aa=aEaaEEE F O .:i-i=;;-N-N ==Eeq =R==N=sRsFaq =;(==;s =R=E =EE 3 c F * e e:-E E E E E.EEEE Sece :=EE E * = ooo9 E 6 !ie--iriFmsr'r-9:€€ N-N .==Na EE=E FFE -;r==3=A=F3s3E=F ==s€= :6GN@6 +N660-d€N N=K== €N E= EE- q FB= a =sR=A-aa iia =E=s.3I:383=I ==E=s <; s*=e atd 6 seEF-E==E=EE --======o =€F3B+=-N .1 -:"9*l E q=q:'':==es:-='-' ;r-r-€ -=q= =r:a-t-r-ae@r_r = F. .rieor-@€F-r_@6€6 @o 6ON<6F_O .i e o As@ j E iesE====R=R=-'-tl-- E=T=JR *=s= === \g Hq E H*-:Hi^'de-@-- E*E E sl ad-e t > E E*-- .=i--=;;; Ed, @^ =?=? EEEEEE€ i € EEE C-aC. 'i: EEEi gFE.F EEE EdE BBe-Eo ; EEEEEEE*EEEE. --9 E--6 Bc5€€ *i-EE-rrrrri- Cg-;q;Z *aoi *;EIe ffE:E= J-;-; .--;-;<;c;e;e;d ;F;F;s-iEl-; E=E .liim -;-- . E - E - F- F- T:iE* iii :d*s:xxxx*-NF TE RE E.E {.4 < F=i-E-=-=-=s-=--=-F gSFs=iEi= E=3S3 SSs; F:

69 Table 46 corttiruted.

-.: u- o TJ EEE €E €€ .=E a Eec EE E EEEE-€EE 3

=o--6 o €€E €€ EE = o ee 33e E EEEE=EEE o6 1")s E EdE=-i.qgq E 1€€ €d EE 60ee,EEeeoEeE : =E 60

99= 6 6nnr-FhoF- -o -0 NN v!"JV!'-

dSaN€Ai-d-d- 5 €-E

d6 6€dr-€€ o= i a o E=- E '1 .:-'1 € ve.€ ---.,. I vvv -e'J-'r'vv

€E F66 € -69-€F-6N6:6@r ddNd-'/-e 6r6666666e6 66 '-' 'J '-a F- € 'r'v 6€OO6e@6--r_d ho o :== o s ==sk3--SR

g€ f;Gr E €EE=3ee'q <' €€€ €€GE !,'re=Eeee'-'EE=

t-o o6€o€F- gH€=$=58 qH Ndd ==H H = ooo 6€O66@CO-OOO @6€ gG= g3==F-'E=3 r N = - o€ oode oo eoo oe 8€A A AeAA=a€a 6Ct- S X=93=g=!i a€ -N- =Fg N

E€€ €g EE E3: € €€EsE:ss * e= €N-eeeec-EeE oooooQo o:ooo€6NoN€: eoo N:+€66-OOsr-6 €€€ oN dr-< : 66

o N6@:€O€

o 6r- R== - g=s35E:r Fo =r€r-e rcE .d oo€ N*F.F.NO6 E F. r- ;<-6:-6@Od€ o --e

E dF-€ ?"] e'e= a -F:=?"?F:'lr: 9. F-€-- € r-€--6r-6€l-€F-F-F-

ci F. r-€N€-do60@ A6dE€ EI €oo ! Q so -N15CF eo do.n o ee stE g EE Eev- E= os) €€6 6 ^E-=uu EEA a -a\ 90ee ;-;-E €tuu !!!FH 6E oe ssoo gA-E -E= 8AA EBS-- "?'Eh 6roo isa qq 8 8o -ts88AA ASBS 5?.9 €F€ tB E6@--88 sjs /\i €d qs * .9 0b gg I fle e :J,tjaeerg-E-9-E-q-* . .E ts€-=k o€ ssa.i8-j9s55,5.5jj 6 O'J .-€ arxN/. rl .: .: "-{ 1": t': : - i'':h€-d:66 a- "': €€.6 jd :aN,NONNN= = N N \-o t or-o o-- HN_ s- fB -@-s -- -- -F-s-scE------

70 To,bla A6 cort't i"lt*ln:d,.

r5 ei 6o; rtr':a= a\if qlr E E1. O r1t t! |.1.l 9 id'Ea rr'} ll) O r{ € .D rFa lli . "ia ,.E g F mddlHF eHeie 6a (\t nr fft lti Ni ot Ft 4- rFi rl qt t= e9. .== o@l EO€ E!ut C'EEE d d @cta p€ EEgrfr|.l Gt-F cl 9-' '- ee ooo .6' s eo .ic-'(' 6 q ' d€Gi a e.o E = dE d.; ft dtrn Fl tA lg' tn '6:i- 0 lat €€a= (i| di i{ + . r'g ! a e 90Ee dE eSEi ddc; o:FI H GNOGA dFirdto FI@ts C' Ei" s tit tlt € il f F riQ

Fi. (\! '8-no-*

lfl rcOOlc)t]ots€eo€15 C= -t .@l L{ FrJi;qcdFrn+3!F ul d c"l vt _FF at]1 rn (2 iiilgi (iit At € g al EEEEF 6EF-+ oooe€9Qe9Q €ftrii o€ =fiHF=ornoe€eoeQq gHB 6 .\r=rFg)q qt et a\t Nerofiio-!tsdf,-FNN- Fa G, EJ li-rura.HctN(\lH 6rN I osr..t 6 lai tfl 6 il1 aGl rro ,Fl 14 Exe=E sHsEg F.d tt a{ .s qoe€oee9?l: 6sOtt €6ts

-l tC, ct, 6 jt Ei .E ESgr=- EE =EE== x* gff F€ ;ei * BEFS E € E) +H€eg'l!l.tP(\l e Ctr Ctl .gB Ll|FFo\F(\INF a-tad o 9,ij+ € H d fi@d 6i rg{alelA