The Open Collections website will be unavailable July 27 from 2100-2200 PST ahead of planned usability and performance enhancements on July 28. More information here.

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Elemental characteristics of organic deposits from an area surrounding a lead-zinc smelter : concentration,… Hawke, Michelle Irene 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2004-901947.pdf [ 29MB ]
Metadata
JSON: 831-1.0052634.json
JSON-LD: 831-1.0052634-ld.json
RDF/XML (Pretty): 831-1.0052634-rdf.xml
RDF/JSON: 831-1.0052634-rdf.json
Turtle: 831-1.0052634-turtle.txt
N-Triples: 831-1.0052634-rdf-ntriples.txt
Original Record: 831-1.0052634-source.json
Full Text
831-1.0052634-fulltext.txt
Citation
831-1.0052634.ris

Full Text

ELEMENTAL CHARACTERISTICS OF ORGANIC DEPOSITS FROM AN AREA SURROUNDING A LEAD-ZINC SMELTER: CONCENTRATION, DISTRIBUTION, MODE OF OCCURRENCE AND MOBILITY. By MICHELLE IRENE HAWKE B.Sc., The University of  Guelph, 1994 M.Sc., The University of  Guleph, 1997 A THESIS SUBMITTED IN PARTIAL FULFILMNET OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY In THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF EARTH AND OCEAN SCIENCES We accept this thesis as*conformine  to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 2004 © Michelle I. Hawke, 2004 Statement of  Go-Authorship regarding Chapter 7 of  the PhD thesis entitled "Elemental characteristics of  organic deposits from an area surrounding a Pb-Zn smelter; Concentration, distribution, mode of  occurrence and mobility", fey  Michelle Hawke The chapter in question is entitled "Comparing maceral ratios from tropical peatlands with assumptions from coal studies. Do classic coal petrographic interpretation methods have to be discarded? and was published as WQst, R.AJL, Hawke, MX and Bustin, R.M. (2001) Comparing maceral ratios from tropical peatlands with assumptions from coal studies: do classic coal petrographic interpretation methods have to be discarded? International Journal of  Coal Geology, 48: 115-132. and was included in the PhD thesis of  R.A.J. WUst (2001) This research was conducted jointly between Raphael Wust and Michelle Hawke with contributions by Dr. R. Marc Bustin. Michelle Hawke's contribution to the research included the following: -developing the concept behind the research with her co-authors, -analytical work, in this case preparation of  peat samples for  reflected  light microscopy, and microscopic analysis (point counting), -interpretation of  microscopic analysis Hie resulting paper was written as a collaborative effort,  primarily between Michelle Hawke and Raphael Wiist, with both authors contributing approximately equally to the finished  paper, I agree that the above statement is an accurate description of  Michelle Hawke's involvement with the preparation of  the work in question. Note: Chapter 7, as described in the above statement, is presented in the final version of the thesis as Appendix A. Abstract Monitoring and remediation of  anthropogenic trace elements requires knowledge of  the magnitude of  emissions and their fate  in the environment. This study examines the role of  organic sediments in the partitioning of  Sb, As, Cd, Cu, Ni, Pb and Zn in the environment surrounding a lead-zinc smelter. Trace element concentrations in peat profiles  were highest near the smelter and decreased with distance. Peat concentrations were compared to mineral soil concentrations and to atmospheric deposition rates. Anomalies between peat concentrations and deposition rates identify  locations of  secondary source impact. Concentrations are higher in impacted peats than in corresponding soils, indicating preferential  sequestration. Assigning a geochemical baseline and differentiating  between trace element sources is difficult  in areas with heterogeneous geochemistry. Two geochemical fingerprinting  methods were applied to estimate smelter impact: normalisation to "conservative elements" and rare earth element patterns. Interpretation of  "conservative element" ratios proved difficult,  due to concentration variability. It was concluded that conservative elements may not represent a geochemical baseline. Light rare earth element enrichment was noted in peats and other sampling media closest to the smelter, but not in smelter feedstock  or wastes, suggesting that LREE enrichment is overprinted from background geochemistry. Sources of  trace element-rich particles include stack emissions and fugitive and geogenic dust. Peat ash was examined by SEM-EDX to determine the morphology and elemental composition of  the particulates. Rounded smelter-emitted particles are present in peats sampled close to the smelter. Weathering and mobilisation are indicated by changes in chemistry between fresh  and weathered particles. Angular fugitive  dust sulphide particles also occur. Sequential leaching of  peats and soils illustrates that trace element speciation is controlled by environmental conditions and by source. In smelter-impacted peats, elements are sequestered as exchangeable, carbonate, Fe-oxide or organic species, indicating precipitation from solution. Non-impacted peats contain higher proportions of  residual and sulphide species. Environmental parameters impact peat diagenesis and influence  element behaviour. An organic petrographic evaluation of  smelter-impacted peats aids in determining past and current conditions. A well-humified  profile,  which indicates aerobic conditions, contained abundant Fe-oxide species. Profiles  containing maceral assemblages that indicate anaerobic conditions contained sulphide and organic species. Table of  Contents Abstract  ii List  of  Figures  viii List  of  Tables  xii Preface  xiv Acknowledgements  xv Chapter  1: Introduction  •• 1 1.1 Monitoring of  Anthropogenic Trace Elements 2 1.2 Peatlands as a Sink for  Trace Elements 2 1.3 The Teck-Cominco Trail Smelter 3 1.4 Objectives 3 1.5 Chapter Outline 5 1.6 Other Research-Organic Petrography of  Tasek Bera, Malaysia 7 1.7 References  7 Chapter  2: Trace  elements  in organic  deposits  in the area surrounding  a base metal  smelter  . ' • • 9 2.1 Introduction 10 2.1.1 The Inorganic Geochemistry of  Peatlands ' 10 2.1.2 Atmospheric trace element deposition and the mobility of  trace elements within the peat profile  • 10 2.2 Study Area 11 2.3 Methodology 16 2.3.1 Peat Sampling and Analytical Protocol - 16 2.3.2 Soil Sampling _ _ 17 2.3.3 Moss Monitoring Rates of  Aerial Deposition 18 2.4 Results 18 2.4.1 Trace Element Concentrations in the Peats 18 2.4.1.1 ComincoSite 19 2.4.1.2 Thunder Road Site 26 2.4.1.3 Champion Lakes Site 28 2.4.1.4 Bombi Summit Site 30 2.4.2 Moss Monitoring Results ; 30 2.4.3 The distribution of  trace elements in the environs of  the stationary sources 34 2.5 Discussion 37 2.6 Conclusions 40 2.7 References  41 Chapter  3: Geochemical  finger-printing  in an anthropogenically-impacted  area of heterogenous  geochemistry—Can  rare earth elements  and conservative  element ratios  be used  to identify  source? 44 3.1 Introduction 45 3.2 Study Area ; 45 3.3 Methodology 46 3.3.1 Peat Sampling and Analytical Protocol 46 3.3.2 Soil Sampling 46 3.3.3 Elemental Analysis 49 3.4 Trace Element/Conservative Element Concentrations: A Viable Normalisation Parameter? 49 3.4.1 Introduction 49 3.4.1.1 Selection of  a geochemical background value 50 3.4.1.2 Selection of  conservative elements 51 3.4.2 Results 52 3.4.2.1 Cominco Profile  52 3.5.2.2 Thunder Road Profile  55 3.5.2.3 Champion Lakes Profile  55 3.4.2.4 Bombi Summit Profile  72 3.4.3 Discussion 72 3.5 Rare Earth Element Enrichment Patterns 73 3.5.1 Introduction 73 3.5.1.1 REE behaviour in the environment 74 3.5.1.2 REEs as Geochemical Tracers 75 3.5.2 Results 76 3.5.2.1 REE concentrations of  smelter materials 76 3.5.2.2 Moss Monitoring Station Data 76 3.5.2.3 Peats 76 3.5.2.4 Soil 80 3.5.2.3 Peat Data Normalised to Soil Data 84 3.5.2.5 Stream Sediment Data 84 3.5.3 Discussion 88 3.6 Conclusions _ 8 9 3.7 References  90 Chapter  4: Particulate  matter  in peat from  the environs of  a Pb-Zn  smelter: determining  geogenic  and anthropogenic  input  92 4.1 Introduction 93 4.2 Sampling Locations 94 4.3 Methodology 97 4.4 Results and Discussion 98 4.4.1 Trace element concentration 98 4.4.2 Mineralogy 99 4.4.2.1 Cominco Site 99 4.4.2.2 Thunder Road Site 100 4.4.2.3 Champion Lakes Site 100 4.4.3 SEM-EDX investigation of  peat ash 102 4.4.3.1 Cominco Site 105 4.4.3.2 Thunder Road 105 4.4.3.3 Champion Lakes 106 4.5 Discussion 134 4.5.1 SEM-EDX Investigations of  Smelter-Related Media 134 4.5.1.1 Smelter Wastes 134 4.5.1.2 Moss Monitoring Stations 134 4.5.1.3 Stream Sediments 135 4.5.2 Environmental Significance  135 4.6 Conclusions 137 4.7 References  137 Chapter  5: Determination  of  the mode of  occurrence of  trace elements  in peats and soils  from  the vicinity  of  a Pb-Zn  smelter  using a sequential  extraction  procedure 139 5.1 Introduction 140 5.2 Study Area 141 5.3 Sampling Locations 141 5.3.1 Cominco Site 141 5.3.2 Thunder Road Site 143 5.3.3 Champion Lakes Provincial Park _ ' 143 5.4 Methodology 144 5.5 Results 147 5.5.1 Summary of  Results from Previous Studies 147 5.5.2 Sequential Leaching Results 148 5.5.2.1 Cominco Site 148 5.5.2.2 Thunder Road 151 5.5.2.3 Champion Lakes 162 5.5.2.4 Comparison to results of  sequential leaching from soil and stream sediment samples 168 5.6 Discussion 171 5.6.1 Controls on Speciation _ _ 171 5.6.1.1 Adsorption 171 5.6.1.2 The Carbonate Fraction 172 5.6.1.3 The Reducible Fraction 173 5.6.1.4 The Sulphide Fraction 174 5.6.1.5 Organic complexation 176 5.6.1.6 The Residual Fraction 177 5.6.2 The role of  source material in determining mode of  occurrence 178 5.7 Conclusions 179 5.8 References  180 Chapter  6: The  organic  petrographic  characteristics  ofpeats  from  an area of  high trace element  input  185 6.1 Introduction 186 6.2 Sampling Locations 187 6.3 Methodology 191 6.4 Results and Discussion 191 6.4.1 Trace Element Concentrations 191 6.4.2 Organic Petrography and Mode of  Occurrence 194 6.4.2.1 Cominco 194 6.4.2.2 Thunder Road Profile  200 6.4.2.3 Champion Lakes Profile  204 6.4.2.4.Bombi Summit Profile  205 6.4.3 Inorganic Matter 6.4.4 Organic Sequestration of  Metals-the role of  peat type and humification 6.5 Conclusions 6.6 References Chapter  7:  Conclusions 7.1 Key findings  resulting from this research 7.1.1 Trace element concentrations in organic deposits in the area surrounding a base metal smelter 7.1.2 Geochemical finger-printing,  using element ratios and REE signatures 7.1.3 SEM-EDX study of  peat ash 7.1.4 Sequential extraction 7.1.5 Organic petrography 7.2 Suggestions for  Future Research APPENDIX  A: Comparing  maceral  ratios  from  tropical  peatlands  with assumptions  from  coal studies.  Do classic  coal petrographic  interpretation methods  have to be discarded? 1 A.A.1 Introduction A.A. 1.1 Maceral Ratio Indexes A. A. 1.2 Organic composition of  tropical peats A.A. 2 Tropical Tasek Bera - physiographic settings, peat accumulation and spatial distribution 223 A.A.2.1 Peat Formation A.A.2.2 Climate A.A.2.3 Biological diversity of  the Tasek Bera wetland A.A.3 Methods A.A.4 Results and Discussion A.A.4.1 Pandanus and Lepironia environment (Sites B64 and B102) A.A.4.2 Swamp Forest Environment (Site B83) A.A.4.3 Maceral Ratio Diagrams A.A.4.4 Modern peat depositional environments - a key to the past_ A.A.4.5 Implications for  coal studies A.A.5 Conclusions A.A.6 References List of  Figures Figure. 1-1. Trail, British Columbia, in 1904 (a), 1944 (b) and 2000. Photos a and b were obtained from the City of  Trail Archives (2003). 4 Figure 2-1. Geology of  the Trail area and the location of  the smelter and sampling sites. After  Little, (1982). 13 Figure 2-2. Concentration of  trace elements as a function  of  the radial distance from the smelter. 23 Figure 2-3. Concentration of  selected trace elements within the Cominco profiles.  24 Figure 2-4. Concentration of  selected trace elements in the Thunder Road profiles.  27 Figure 2-5 Concentration of  trace elements in the Champion Lakes profiles.  29 Figure 2-6. Concentration of  trace elements in the Bombi Summit profile.  31 Figure 2-7. Concentration of  trace elements in peats relative to the differential concentrations in peats versus mineral soils. 39 Figure 3-1. Geological map of  the Trail area showing the location of  the smelter and sampling sites. Modified  from Little (1982). 47 Figure 3-2. Concentration of  trace elements in the peat profiles.  BSed = basal sediment. 57 Figure 3-3. Normalised element ratios using soil and rock for  the Cominco profile. Bulk concentration data is plotted on the primary (top) axis, while element ratios are plotted on the secondary (bottom) axis. 63 Figure 3-4. Normalised element ratios using soil and rock for  the Thunder Road profile.  Bulk concentration data is plotted on the primary (top) axis, while element ratios are plotted on the secondary (bottom) axis. 65 Figure 3-5. Normalised element ratios using soil and rock for  the Champion Lakes profile.  Bulk concentration data is plotted on the primary (top) axis, while element ratios are plotted on the secondary (bottom) axis. 67 Figure 3-6. Normalised element ratios using soil and rock for  the Bombi Summit profile.  Bulk concentration data is plotted on the primary (top) axis, while element ratios are plotted on the secondary (bottom) axis. ; 69 Figure 3-7 a. Plot of  the concentrations of  Ti and Zr versus the peat ash contents, b. Whisker-Box diagram showing the spread of  Ti/Zr ratios for  each profile,  where the median value is represented by the dot in the middle of  the box, the upper and lower extent of  the box represent the 75th and 25th percentile, respectively, and the lines extending from the box represent the range of  data. 71 Figure 3-8. Chondrite nomalised REE pattern for  selected smelter feedstock  and waste samples. 77 Figure 3-9. Chondrite normalised REE pattern for  median values of  REE's deposited on moss monitoring stations near the peat sampling sites and the Zn-concentrate unloading site. 78 Figure 3-10. Chondrite normalised REE's for  peat profiles.  82 Figure 3-11. Chondrite normalised REE's for  mineral soils near the peat sampling sites and near the ore concentrate unloading site. 83 Figure 3-12. REE's for  peat normalised to the nearest mineral soil (0-3 cm depth). 85 Figure 3-13. REE's for  peat normalised to the nearest mineral soil (3-10 cm depth). 86 Figure 3-14. Chondrite normalised REE pattern for  stream sediments collected near the peat sampling sites and near the Zn-concentrate unloading site. 87 Figure 4-1. Geological map of  the Trail area (from Little, 1982) showing sampling locations. 96 Figure 4-2. Example of  a geogenic particle (sphalerite) from the Champion Lakes 103 sampling site. The number on the EDX spectra refers  to the frame  number on Table 4-10. 103 Figure 4-3. Examples of  heat-treated spherical or rounded metallic particles (4a and c) emitted from the smelter, observed in the Thunder Road 7-8 cm peat. Also present are a spherical silicate fly  ash particle (4b) and a Zn-containg Fe-precipitate (4d). Numbers correspond to frame  numbers in Table 4-9. 104 Figure 4-4. Examples of  rounded particles containing Pb and Zn present in the Cominco 1-2 cm sample. Numbers on the EDX spectra refer  to the Frame number on Table 4-5. 107 Figure 4-5. Example of  a subangular particle containing Zn from the Cominco 1-2 cm sample. Number on the EDX spectra refers  to the Frame number on Table 4-5. ; 108 Figure 4-6. Example of  a rounded particle of  anthropogenic origin containing Zn and Fe from the Cominco 4-5 cm sample (3a on Table 4-6). The geochemistry of examples of  other particulate matter observed within the frame  is indicated in spectras 3band 3c. 112 Figure 4-7. Example of  a subangular particle of  geogenic origin containing Pb, from the Cominco 4-5 cm sample (4a on Table 4-6). Spectras 4b-d indicate the geochemistry of  other observed particulates within the frame.  113 Figure 4-8. Fe-Zn-Cu precipitate from the Cominco 17-20 cm sample (9a on Table 4-7). Examples of  geogenic particles are also pictured (9b and c). 119 Figure 4-9. Geogenic particle containing Pb, from the Cominco 17-20 cm sample (6a on Table 4-7). The EDX spectra of  other particulate matter observed within the frame  are also presented (6b and c ) _ 120 Figure 4-10. Spherical metal enriched particle emitted from the smelter, from the Thunder Road 0-1 cm sample (4a on Table 4-8). 123 Figure 4-11. Examples of  Zn-containing particles from the Thunder Road 0-1 cm sample, a.: weathered geogenic Zn sulphide mineral, c and d: Fe-Zn precipitates (frame  3 on Table 4-8). An example of  the geochemistry of  other particulate material observed within the frame  is illustrated by spectra b. 124 Figure 4-12. Example of  native Zn particle from the Thunder Road 0-1 cm sample (Frame 5 ,Spot a on Table 4-8), with a quartz particle (b). 125 Figure 4-13. Angular geogenic Zn-rich particle (a) and a spheroidal fly  ash particle from the Thunder Road 7-8 cm sample (Frame 2 on Table 4-9). 128 Figure 4-14. Agglomeration of  Zn rich particles, including a subangular native Zn particle (a), Zn-rich precipitates (b and c) and (d) a diatom with Zn precipitated on the surface  from  the Thunder Road 17-20 cm sample (Frame 3 on Table 4-9). 129 Figure 4-15. Diatom-rich ash material from the Champion Lakes 3-4 cm sample (e.g., b). Non-biogenic materials include an angular Ti-rich geogenic particle (a) and a glassy silicate fly  ash particle with slight Ca enrichment.(c) (Frame 1 on Table 4-10). 132 Figure 4-16. Geogenic particles containing Si, Ca, Fe, Al and S from the Champion Lakes 3-4 cm sample (spots a, c, d and e). Spot (b) is a glassy Si fly  ash particle with slight S, Ca and Fe enrichment. (Frame 6 on Table 4-10). 133 Figure 4-17. Examples of  particulate matter captured by the moss-monitoring stations in the Trail area. a. Geogenic quartz particle, b. Rounded and "fluffy"  PbO particles formed  by thermal alteration in the smelter, and angular ZnS particles from fugitive  emissions, c. Angular ZnS particles derived from fugitive  dust, collected near an ore and slag storage facility.  From Goodarzi et al., 2003. 136 Figure 5-1. Geological map of  the Trail, British Columbia area, with sampling locations for  peats and stream sediments indicated. 142 Figure 5-2. Total element concentrations within the Cominco profile.  BSed = basal mineral sediment. 150 Figure 5-3. Absolute concentrations (in ppm) for  sequentially extracted fractions  in the Cominco profile,  including the basal mineral sediment (BSed). Concentrations for  the Cominco mineral soil (MS) and the Rossland background soil (BG) are included below the results for  the peat profile.  152 Figure 5-4. Relative concentrations (%) for  sequentially extracted fractions  in the Cominco profile,  including the basal mineral sediment (BSed). Relative concentrations for  the Cominco mineral soil (MS) and the Rossland background soil (BG) are included below the results for  the peat profile.  153 Figure 5-5. Total element concentrations within the Thunder Road profile,  including basal sediment (Bsed). 157 Figure 5-6. Absolute concentrations (in ppm) for  sequentially extracted fractions  in the Thunder Road profile,  including the basal mineral sediment (BSed). 158 Figure 5-7. Relative concentrations (%) for  sequentially extracted fractions  in the Thunder Road profile,  including the basal mineral sediment (BSed). 159 Figure 5-8. Total element concentrations within the Champion Lakes profile.  163 Figure 5-9. Absolute concentrations (in ppm) for  sequentially extracted fractions  in the Champion Lakes profile,  and for  a nearby mineral soil (MS). 164 Figure 5-10. Relative concentrations (%) for  sequentially extracted fractions  in the Champion Lakes profile  and a nearby mineral soil (MS). 165 Figure 5-11. Relative concentrations (%) for  sequentially extracted fractions  in selected stream sediments. 170 Figure 6-1. Geological map of  the Trail area, indicating the location of  sampling sites. 189 Figure 6-2. Location photographs a. Teck-Cominco smelter b. Cominco sampling site c. Cominco peat profile  duing sampling d. Thunder Road sampling site e. Champion Lakes sampling site f.  Bombi Summit samplig site 190 Figure 6-3. Organic petrographic composition of  the peat profiles,  expressed as volume % = 193 Figure 6-4 . Graph of  humification  index versus total inertinite content. 195 Figure 6-5. Relative concentrations (%) for  sequentially extracted fractions  in the Cominco profile,  including basal mineral sediment (Bsed). 197 Figure 6-6. Relative concentrations (%) for  sequentially extracted fractions  in the Thunder Road profile,  including basal mineral sediment (Bsed). 201 Figure 6-7. Relative concentrations (%) for  sequentially extracted fractions  in the Champion Lakes profile.  206 Figure 6-8. Examples of  mineral matter found  within the Cominco profile.  A. Sulphide mineral, probably pyrite B. and C. highly reflecting  carbonaceous material (fly  ash) produced from the combustion of  coal in the smelter operation. D. diatom permineralised with sulphides. E. pollen grain permineralised with sulphides F. oxidised framboidal  pyrite and quartz grain. 208 Figure 6-9. Examples of  mineral matter found  within the Champion Lakes profile.  A. woody material permineralised by sulphides. B. fly  ash particle. C. particle of sulphide mineral, probably sphalerite. 209 Figure A-l: Location map of  the Tasek Bera Basin (Tasek = lake) illustrating the dendritic drainage pattern, extent of  peat accumulation (gray shading) and peat swamp tributary names (Paya = swamp, swampy area). Inset map shows the location of  the Tasek Bera Basin in the states of  Pahang and Negeri Sembilan, Peninsular Malaysia. Selected core sites are indicated. Sites B64 and B102 are from a Lepironia articulata and Pandanus helicopus environment, while site B83 is situated in true swamp forest.  225 Figure A-2: Core B64 from Lepironia and Pandanus environment. The site was waterlogged at the time of  core collection. Strati graphic description and chemical characterization of  the profile  and petrographic composition. Petrographic results are reported on a volume percent basis. The huminites (vitrinites) are often  dominant with various amounts of  liptinites. Inertinites, mainly funginite,  dominate the lower part of  this littoral core. 230 Figure A-3: Core B102 from Lepironia and Pandanus environment. The site was waterlogged at the time of  core collection. Stratigraphic description and chemical characterization of  the profile  and petrographic composition. Petrographic results are reported on a volume percent basis. The huminites (vitrinites) are often  dominant with various amounts of  liptinites. Inertinites, mainly funginite,  dominate the lower part of  this littoral core. 233 Figure A-4: Core B83 from the swamp forest  environment of  Paya Belinau (Fig. 1). The site was waterlogged at the time of  core collection. Stratigraphic description and chemical characterization of  the profile  and petrographic composition. Petrographic results are reported on a volume percent basis. The huminites (vitrinites) are often  dominant with various amounts of  liptinites. Inertinites, mainly funginite,  dominate the middle part of  the swamp forest  deposits. 238 Figure A-5: Three cross-plots of  maceral ratios commonly used for  paleodepositional interpretation in coal geology. A) Inertinite versus textinite plus texto-ulminite. B) GI versus TPI. The Diessel (1982) paleodepositional environments are indicated. C) Inertinite versus textinite divided by texto-ulminite. 242 List of  Tables Table 2-1. Location, bedrock geology, vegetation and general notes on the sampling locations. Details of  bedrock geology are from Little, 1982; Hoy and Andrew, 1991. 12 Table 2-2. Compilation of  published geochemistry for  the bedrock underlying the peat sampling sites. Data for  the Nelson Intrusives is presented as published in Sevigny (1990). Data for  the Elise Formation is presented as published in Beddoe-Stephens and Lambert (1981). 15 Table 2-3. Concentrations of  trace elements measured within peats and mineral soils. Environmental soil quality guideline data provided from B.C. Ministry of  Water, Land and Air Protection (2002) lists element concentrations at which soils must be relocated to non-agricultural lands, and at which a permit is required for disposal. Environment Canada Soil Quality Guidelines (2002) list generalized trace element concentration guidelines for  agricultural, residential/parkland and industrial land uses. 22 Table 2-4. Calculated Peat Enrichment Factors (PEFs) for  peats relative to nearby mineral soils. 25 Table 2-5. Concentration of  selected trace elements in the moss used as sampling media for  moss monitoring stations in the Trail area. The concentrations are statistically corrected for  geogenic input, as described in Goodarzi et al., (2001; 2003). Cumulative impact over the two-year sampling period is indicated in the right column. 32 Table 2-6. Relative concentrations of  trace elements in the Thunder Road and Champion Lakes peats and soils, as compared to Cominco peats and soils. Relative rates of  deposition as determined by moss monitoring stations are calculated from data in Goodarzi et al., 2001; 2003. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _  36 Table 3-1 Description of  the sampling sites. • 48 Table 3-2. Mineralogy of  the peat profiles  as determined by X-ray diffraction. Methodology is described in Chapter 4. 53 Table 3-3. Concentration of  selected elements in the peat profiles.  56 Table 3-4. Element concentrations in mineral soils (3-10 cm depth) corresponding to the peat sampling sites, and calculated element ratios. 58 Table 3-5. Element ratios normalized to mineral soil for  Ti and Zr. 59 Table 3-6. Selected published geochemical data and the resulting caluculated conservative element ratios from the bedrock geology of  the Trail Pluton (Sevigny, 1990) and the Elise Formation. (Beddoe-Stephens and Lambert, 1981) 61 Table 3-7. Element ratios normalised to bedrock geochemical data listed in Table 3-6. 62 Table 3-8. Concentration of  REE's and normalized La/Lu ratios from selected smelter feedstock  and waste samples. 77 Table 3-9. Concentration of  REE's and normalized La/Lu ratios for  mineral soils and moss monitoring stations near the peat sampling sites, and near the ore concentrate unloading site. 78 Table 3-10. Concentration of  REE's and normalized La/Lu ratios. 81 Table 4-1. Compilation of  published geochemical data, for  the Nelson Intrusives (Sevigny, 1990) and the Elise Formation (Beddoe-Stephens and Lambert, 1982). ' 95 Table 4-2. Description of  the sampling sites: location, bedrock geology, vegetation and conditions within the peat profile.  97 Table 4-3. Bulk concentrations of  selected trace elements within the peat profiles.  99 Table 4-4. Mineralogy as determined by X-ray diffraction.  101 Table 4-5. Summary of  the observations from the Cominco 1-2 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures.  Elements present in high concentrations are listed in bold face  type. 109 Table 4-6. Summary of  the observations from the Cominco 4-5 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures.  Elements present in high concentrations are listed in bold face  type 114 Table 4-7 Summary of  the observations from the Cominco 17-20 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures. Elements present in high concentrations are listed in bold face  type. 116 Table 4-8 Summary of  the observations from the Thunder Road 0-1 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures. Elements present in high concentrations are listed in bold face  type. 121 Table 4-9 Summary of  the observations from the Thunder Road 7-8 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures. Elements present in high concentrations are listed in bold face  type. 126 Table 4-10. Summary of  the observations from the Champion Lakes 3-4 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures. Elements present in high concentrations are listed in bold face  type. 130 Table 5-1. Organic carbon and total sulphur concentrations of  selected samples. 151 Table 6-1. Description of  sampling sites, bedrock geology and vegetation. 188 Table 6-2. Concentrations of  trace elements in the peat profiles.  192 Table A-l: Data analyses of  the three representative cores of  the Tasek Bera Basin for petrographic investigations. Core number, sample depth in cm, ash content in wt-%, carbon content in wt-% and pH are given. The carbon content of  B83 and B102 was estimated based on the linear relationship between ash and carbon content of  over 130 samples from the Tasek Bera mire deposit. (Wiist, 2001).235 Preface This thesis is a component of  the multidisciplinary Geological Survey of Canada-Teck Cominco project. The thesis is written in the form of  six separate research manuscripts, each of  which is presented as a thesis chapter, and each of which is either intended for  publication as a research paper at a future  date but has yet to be submitted, or has already been published. All chapters pertaining to the Teck-Cominco project (1-7) are original research, conceived and written by Michelle Hawke under the supervision and direction of  Dr. Marc Bustin and Dr. Fariborz Goodarzi. Appendix A is a collaborative effort  between Raphael Wiist and Michelle Hawke under the supervision of  Dr. Marc Bustin. This paper has been published as Wiist et al. (2001). Raphael Wiist (former  PhD student at UBC) performed  the peat sampling. Michelle Hawke performed  all organic petrographic analyses and data compilation. Interpretation and subsequent writing was performed  by both Wiist and Hawke in a collaborative fashion,  with each contributing equally to the final manuscript. References: Wiist, R.A.J., Hawke, M.I., and Bustin, R.M. 2001. Comparing maceral ratios from tropical peatlands with assumptions from coal studies: do classic coal petrographic interpretation methods have to be discarded? International Journal of  Coal Geology, 48:115-132. Acknowledgements I am indebted to numerous people for  their support in the completion of  this thesis. / University of  British Columbia My supervisor, Dr. R. Marc Bustin at the University of  British Columbia has provided both the financial  support and professional  guidance that has made the completion of  this work possible. My collegue and friend,  Dr. Raphael Wiist, currently at James Cook University in Australia, provided both encouragement and advice during the completion of  the thesis and a thorough review of  a draft  of  the completed project. Bert Mueller helped to develop a lab protocol for  the sequential leaching procedure through numerous lengthy discussions and provided training and advice for  the operation of  the ICP-MS. Tammy Cohen, Doctoral Candidate from the Biology Department performed  much of  the ICP-MS analysis for  the second batch of sequential leaching in an efficient  and expert manner. Dr. Kristin Orians generously allowed me to use her laboratory facilities  to complete the first  batch of  sequential leaching work while I was in Vancouver. Calgary I am particularly grateful  to Dr. Fariborz Goodarzi at the Geological Survey of Canada-Calgary for  all the professional,  scientific,  financial  and personal support that he provided during the course of  this project. The GSC-C/Teck-Cominco Trail smelter project was developed under his leadership, and it was he who provided the original suggestion of  a peat project in Trail. This research was conducted in concert with several other GSC-Calgary studies at the Trail smelter, and this body of  research, to which I was allowed access, was invaluable in providing supporting evidence for my own research. Many thanks are due to the other scientific  researchers and technical support personel involved in the Trail smelter project. Hamed Sanei, Doctoral Candidate at the University of  Victoria and member of  the Enviromental Research Group at GSC-C is thanked for  his helpful  discussions and reviews of earlier drafts  of  this thesis. Julito Reyes of  the GSC-C is thanked for  all his help in obtaining data, advice and helpful  discussions. Marcel Labonte is thanked for  his many helpful  suggestions regarding geochemistry and data handling. Sharlene Pollock is gratefully  acknowledged for  years of  scientific  discussions on the topics of peat, coal and elemental analysis, her supportive friendship  throughout years of change, and for  reading an ugly first  draft  of  the thesis. Additionally, I am very thankful  for  the support of  numerous other people at the GSC-C. Dr. Lavern Stasiuk permitted me to use the organic petrography laboratory facilities  and is thanked for  his many helpful  discussions and reviews of early drafts.  Jenny Wong is thanked for  her help with the SEM-EDX and the XRD. Kathleen Bradford  is thanked for  her assistance with the carbon and sulphur analysis, and for  years of  supportive friendship  in the lab and the office. Bill Duncan of  Teck-Cominco provided me with background information regarding the smelter operation and the Trail area, and permitted me to access sites within the smelter property. Don Espahain, Mariah Duncan and Brandee Venne provided their assistance with the peat sampling. I would like to acknowledge my friends  and collegues in the Coalbed Methane group at Suncor Energy for  their support, encouragement and understanding while I completed my thesis. In particular, I would like to acknowledge my former  manager Mike Dawson, currently of  Defiant  Energy, who permitted me to work the flexible schedule that made it possible for  me to finish  my thesis while learning the mysteries of  CBM, and Kim Russell-Symon and Louise Huston, whose words of  encouragment were greatly appreciated. Finally, I would like to thank my partner Ryan Gaetz for  years of  unfailing, encouragement and patience, and for  always having faith  in me. Chapter 1 Introduction 1.1 Monitoring of  Anthropogenic Trace Elements The monitoring of  trace elemental emissions has become an essential aspect of  many industrial operations. Concerns about aerially dispersed materials such as fly  ash and the contaminants contained within them have prompted industries to control emissions and to monitor their impact on the surrounding environment. However, monitoring the impact of anthropogenic activities is a complicated process. Assessments of  trace metal fluxes  based on mass balance equations (e.g. Nriagu and Pacyna, 1988) have come under criticism (e.g. Rasmussen, 1996) because they often  do not account for  geogenic (naturally occurring) trace metals, and because the models are often  based on inaccurate data. Trace element partitioning in the environment following  deposition is poorly understood, and determination of  the dispersal patterns of  elements requires that the methodology employed captures all the phases of  pollution that are of  concern (Rasmussen, 1996). Hence, it is necessary to employ a variety of  monitoring methods and types of  sampling media to accurately characterise the distribution of  trace elements around a point source. 1.2 Peatlands as a Sink for  Trace Elements Peat deposition and early diagenesis are controlled by many factors  that interact to lead to the formation  of  peats which are relatively unique products of  their environment. The geochemical characteristics of  a peat deposit are a function  of  the surrounding country rock, hydrology, climate and the depositional environment (e.g. Goodarzi, 1995). In anthropogenically altered environments, both airborne and waterborne trace elements can become deposited within the peat. Pollutants from the atmosphere are deposited in the peat via wet deposition as ions in precipitation, or dry deposition as particulate matter (Godbeer and Swaine, 1995). Minerotrophic peatlands, which are fed  by groundwater, are subject to the input of  elements carried in solution by groundwater from the surrounding area. Organic matter is capable of  complexing and chelating many trace elements which are of  environmental concern, particularly under low pH reducing conditions (Forstner, 1987), and so may remove much of  the potentially hazardous material from groundwater. For this reason, peatlands may act as a sink for  many elements, and may have elemental concentrations that are elevated relative to the surrounding area. A study examining the concentration and mode of  occurrence of  trace elements within peatlands surrounding a point source of  pollution offers  the opportunity to better understand the role of  peatlands in sequestering trace elements. 1.3 The Teck-Cominco Trail Smelter The location of  this study is Trail, British Columbia, which is situated in the Columbia River valley. A smelter has been in operation in Trail since 1896, originally to process copper ores mined in nearby Rossland. Over the years, the smelter has expanded and is currently one of  the largest facilities  of  its kind in the world (Fig. 1-1). Production totals from the year 2000 are listed by Teck-Cominco as 272 900 tonnes of  Zn, 91 300 tonnes of  Pb, 12 212 000 ounces of  Ag and 56 000 ounces of  Au (Teck-Cominco, 2003). Previously, the main source of  the ore concentrate was the Sulivan Pb-Zn mine in Kimberly, BC, which was in operation for  92 years, from 1909 until its closure in 2001. The main source of  concentrate is currently the Red Dog mine in Alaska, which has been in operation since 1989 and is the largest Zn mine in the world. Emmissions of  Pb from the smelter stack have reportedly been reduced dramatically by the installation of  a KTVCET Pb smelter in 1997 (Teck-Cominco, 2000). Lead emissions from the stack are reported by Teck-Cominco to be 268 kg/day in 1995, but were reduced to 85 kg/day in 1999. The new smelter also resulted in the reduction of  Cd (from 13 to 1.7 kg/day) and As (from 41 to 5.5 kg/day) stack emissions between 1995 and 1999. Zinc stack emissions increased during this time, however, from 383 to 412 kg/day. Particulate matter contents measured in ambient air in the Trail area were not appreciably reduced by the installation of  the new Pb smelter, despite the reduction in stack emissions. Teck-Cominco attributes this observation to the fact  that sources such as fugitive  dust from unprocessed ore concentrates and geogenic dust are also major contributors to the particulate composition of  the air. 1.4 Objectives The objective of  this study is to examine the trace element content of  organic soils and peats in the vicinity of  a Pb-Zn smelter using a variety of  analytical techniques. This program was carried out in order to achieve several goals: Figure. 1-1. Trail, British Columbia, in 1904 (a), 1944 (b) and 2000. Photos a and b were obtained from the City of  Trail Archives (2003). -to assess the extent of  trace element contamination in the area surrounding the smelter -to determine if  trace elements deposited from the smelter are mobile in the environment, thus suggesting greater bioavailability and therefore  a greater threat to health, and further  indicating the need for  environmental trace elemental monitoring programs that account for  potential post-depositional migration -to determine the relative importance of  peats and organic soils as sinks for mobile trace elements relative to mineral soils; -to determine if  it is possible to identify  the source of  the trace elements using some commonly-used geochemical fingerprinting  methods, such as rare earth element patterns and normalisation to conservative elements; -to examine the morphology and chemistry of  particulate material containing trace elements found  within the peat, and to compare the results to those from fresh smelter feedstock,  baghouse wastes and particulates captured by moss monitoring stations, in order to assess the source and the influence  of  weathering on the particulate material; -to determine the mode of  occurrence of  trace elements as a function  of  distance from the smelter, peat depth and trace element source; -to evaluate the organic petrographic characteristics of  the peats in the context of the mode of  trace element occurrence, in order to determine how the early diagenetic conditions within the peat influence  trace elemental behaviour. 1.5 Chapter Outline In order to accomplish these goals, the project was divided into a series of  five phases of  experimental work. Each of  these phases is written up as a separate paper for publication in a refereed  journal, and is presented as an individual chapter in this thesis. A brief  description of  each chapter is as follows: Four peatlands located within a 16-km radius of  the smelter were identified  and sampled. The total concentrations of  trace elements in peats from profiles  at the margin and in the centre of  the peat deposit were compared to trace element concentrations in nearby mineral topsoils and to rates of  atmospheric deposition as determined by a moss monitoring program (Goodarzi et al., 2001; 2003). The results of  this project are discussed in Chapter 2. Two methods of  geochemical fingerprinting  were applied to the peat and soil data in order to determine their usefulness  in distinguishing between sources of  elemental input. The first  method used was the normalisation of  trace element concentrations to concentrations of  a supposed conservative geogenic element, such as Ti or Zr. The data used for  normalisation was derived from numerous sources, including published geochemical data for  the region and local soil data. The second method was to compare REE concentrations and chondrite-normalised REE patterns from peats, mineral soils, moss monitoring stations, stream sediments and smelter feedstocks  and wastes, in order to determine if  the impact of  the smelter can be discerned from REE signatures. The results from  this phase of  the project are presented in Chapter 3. Ash (combustion residue) from selected peat samples was examined by scanning electron microscopy (SEM) and the elemental composition of  individual particles was determined by Energy Dispersive X-Ray spectroscopy. Particle morphology and chemistry are both a function  of  source. Stack emissions are typically rounded as a result of  heating and subsequent rapid cooling, and are most commonly oxides. Fugitive dust particles from unprocessed ore concentrates are angular and are mainly sulphides. Geogenic particles derived from the bedrock are typically angular and have chemistry consistent with the local mineralogy. Using this combined morphology-chemistry approach, the distribution of  trace element-containing particles can be determined as a function  of  peat depth and distance from source. The results of  this investigation are presented in Chapter 4. Selected peat, mineral soil and stream sediment samples were sequentially leached in order to determine the mode of  occurrence of  trace elements. The sequential leaching procedure used in this experiment is selective for  exchangeable, carbonate, reducible Fe-Mn, organically-chelated, organic/sulphide and residual phases. The mode of  occurrence varies as a function  of  the trace element source, and of  the conditions within the sediment. The results of  this study are presented in Chapter 5. The organic petrography of  the peats was examined under reflected  white and blue light. Organic petrographic characteristics such as the extent of  humification  and oxidation, and the presence of  fungal  remains are indicators of  the early diagenetic conditions within the peatland, which in turn are a partial control on the mode of occurrence of  trace elements. Results of  the organic petrographic study are presented in Chapter 6. 1.6 Other Research-Organic Petrography of  Tasek Bera, Malaysia In addition to the Trail smelter study, the author was involved in an investigation of  the organic petrographic characteristics of  a series of  peats from Tasek Bera, Malaysia. The purpose of  this study was two-fold:  1. to characterise the organic petrography of  peat profiles  sampled from known depositional environments in order to construct a history of vegetational and early diagenetic conditions within peatland and 2. to apply the concept of  using ratios of  "index macerals" to determine if  these maceral ratios can differentiate between different  peat depositional environments. This concept has been used extensively for  coals, but little evidence of  its validity exists from peat organic petrography studies. The results of  this study are presented in Appendix A, and were published as Wiist et al., 2001. The organic petrographic study presented in Chapter 6 builds on concepts developed during the course of  this work. 1.7 References City of  Trail, 2003. Museum and Archives. Available: http://www.cityoftrail.com/ourcity/museum.htm Forstner, U., 1987. Metal speciation in solid wastes—Factors affecting  mobility. In: Speciation of  Metals in Water, Sediment and Soil Systems, L. Landner, ed. Lecture Notes in Earth Sciences, Springer Verlag, pp 13-42. Godbeer, W.C. and Swaine, D.J., 1995. The deposition of  trace elements in the env irons of  a power station. In: Environmental Aspects of  Trace Elements in Coal, D J. Swaine and F. Goodarzi, eds., Kluwer Academic Publishers, Dordrecht, pp 178-203. Goodarzi, F., 1995. Geology of  trace elements in coal. In: Environmental Aspects of  Trace Elements in Coal, D.J. Swaine and F. Goodarzi, eds., Kluwer Academic Publishers, Dordrecht, pp 51-75. Goodarzi, F., Sanei, H. and Duncan, W.F., 2001. Monitoring the distribution and deposition of  trace elements associated with a zinc-lead smelter in the Trail area, British Columbia, Canada. Journal of  Environmental Monitoring, 3: 515-525 Goodarzi, F., Sanei, H. and Duncan, W.F., 2003. Deposition of  trace elements in the Trail region, British Columbia; an assessment of  the environmental effect  of  a base metal smelter on land. Geological Survey of Canada Bulletin 573, 50 pp. Nriagu, J.O. and Pacyna, J.M., 1988. Quantitative assessment of  worldwide contamination of  air, water and soils by trace metals. Nature, 333: 134-139. Rasmussen, P.E., 1996. Trace Metals in the Environment: A Geological Perspective., Geological Survey of Canada Bulletin 429, 26 p. Teck-Cominco, 2000. Environmental Performance  Review of  the New KIVCET Lead Smelter and Refineries  Abatement Upgrades: Air Emissions and Ambient Air Quality available: http://www.teckcominco.com/articles/operations/tr-kivcet-envperfrev.htm Teck-Cominco, 2000. Trail Smelter and Refineries  Production available: http://www.teckcominco.com/operations/trail/production.htm Wiist, R.A.J., Hawke, M.I. and Bustin, R.M., 2001. Comparing maceral ratios from tropical peatlands with assumptions from coal studies: do classic coal petrographic interpretation methods have to be discarded? International Journal of  Coal Geology, 48: 115-132. Chapter 2 Trace elements in organic deposits in the area surrounding a base metal smelter 2.1 Introduction The capacity for  organic sediments to sequester and retain metals has long been recognised and a number of  studies have examined peatlands as sinks for  atmospheric and groundwater contaminants in areas of  elevated pollution (e.g. Mathews and Bustin, 1994; Gorres and Frenzel, 1997; West et al., 1997). A study of  the trace elemental distribution in peatlands in the vicinity of  point sources of  trace element emissions, combined with an atmospheric deposition monitoring program, offers  an opportunity to examine the role of  organic deposits in sequestering geogenic and anthropogenic trace elements. This paper discusses the concentration of  selected trace elements measured in peats and soils in the vicinity of  a lead-zinc smelter. 2.1.1 The  Inorganic  Geochemistry  of  Peatlands The geochemical characteristics of  a peat or coal deposit are dependent upon the surrounding geology, hydrology, climate and depositional environment (e.g. Goodarzi, 1995). Minerotrophic peatlands (peatlands which are hydrologically connected to the regional groundwater table) sequester and concentrate elements carried in solution by groundwater from the surrounding bedrock. Organic matter is capable of  complexing and chelating numerous trace elements which are of  environmental concern, particularly under reducing conditions, such as are found  in peatlands (Forstner, 1987). Organic deposits can therefore  act as a geochemical sieve and remove potentially hazardous material from groundwater. Trace elements may also be adsorbed onto clays and iron and manganese oxides, hydroxides and oxy-hydroxides, or form epigenetic minerals such as sulphides (e.g. Allard, et al., 1987; Elder, 1988; Clark et al., 1998; Bendell-Young, 1999). For this reason, peatlands may act as a sink for  many elements, and may have concentrations that are elevated relative to the surrounding area. 2.1.2 Atmospheric  trace element  deposition  and the mobility  of  trace elements  within the peat profile Airborne pollutants may be deposited in peat via wet deposition as ions in precipitation or via dry deposition as particulate matter (Godbeer and Swaine, 1995; Goodarzi et al., 2001). Peat profiles  have been shown to act as repositories of atmospherically deposited trace elements. For example, the trace element chemistry of peat cores from the Jura region of  Switzerland can be correlated with anthropogenic events such as Roman lead-smelting and the use of  leaded gasoline (Shotyk, 1995; 1996; Shotyk et al., 1996; 1997). Similar studies have been conducted by a number of  other researchers (e.g. Gorres and Frenzel, 1997; West et al., 1997, Espi et al., 1997; Schell et al., 1997; Steinnes, 1997; Gilberston et al., 1997; Kempter et al., 1997; Kiister and Rehfuess,  1997; Martinez Cortizas, 1997; Weiss et al., 1997). While Shotyk et al. (1997) indicate that minerotrophic peats are not suitable for  chronostratigraphic trace element studies due to geogenic overprinting, attempts have been made to discern anthropogenically-sourced elements from geogenic inputs through the use of  enrichment ratios, often  normalised to crustal abundance ratios. Once deposited on the surface  and incorporated into the peat, it is likely that some of  the elements will be mobilized and redistributed further.  Elements may be mobilized as a result of  the prevailing conditions within the peatland such as pH, redox potential, the presence of  complexing ligands and adsorption sites, and the presence of  soluble organic colloids with which the metal can form a mobile colloidal species (e.g. Honeyman, 1999; Kersting, 1999). Trace nutrients (e.g. Cu, Zn, Mo), and other elements subject to uptake by plants (e.g. Cd), may be enriched in the surficial  peat as a result of biogeochemical cycling (Trudinger et al., 1979; Rasmussen, 1996; Espi et al., 1997). Due to the large potential for  post-depositional migration, trace element concentrations in the peat profiles  were not assumed to be chronostratigraphic. 2.2 Study Area Trail, British Columbia (Fig. 2-1) is situated in the Columbia River Valley, which bisects the sampling area. The Columbia River is the boundary between the Monashee (west side) and Selkirk (east side) Mountains. The bedrock in the area consists mainly of Upper Jurassic plutonic rocks (the Nelson Intrusives), and Lower Jurassic metavolcanic and metasedimentary rocks (the Elise Formation; Little, 1982; Hoy and Andrew; 1991). The surficial  geology is influenced  by Pleistocene glaciation, and the area is predominantly covered by discontinuous ground moraines (Little, 1982). The Teck-Cominco Trail smelter has been in operation since 1896. Lead-zinc ores are the major product processed, but the plant also processes Ag, Au, Cd, Bi, In, Ge concentrate, Ge02, CUSO4, CuAsO, sodium antimonate and sulphur products (Henderson et al., 1999). Environmental conditions in the area surrounding the smelter have improved in recent decades with the adoption of  better production and containment technologies, and ore handling and storage practices, although Ag, As, Cd, Cu, Hg, Pb and Zn are known to be emitted by the facility  (Henderson et al., 1999). Site Bedrock Geology Vegetation pH and Comments Cominco UTM Zone 11, 0446226, 5440375 Trail Pluton (granodiorite, quartz diorite) Typha  latifolia  (common cattail), Epilobium  ciliatum (purple leafed willow herb) Carex  urticulata  (beaked sedge) Equisetum  palustre,  and Deschampsia  cespitosa (tufted hairgrass) 5.3 (surface) 5.6 (mineral sediment) 5.7 (pore water) -saturated at the time of sampling, although gleying was noted in the uppermost 15 cm-intermittently aerobic. Zones of oxidation were noted around roots. Thunder Road UTM Zone 11, 0445145, 5440974 Trail Pluton (granodiorite, quartz diorite) Typha  latifolia  (common cattail), Pinus  sp. (pine) Betula  sp. (birch) Epilobium  ciliatium Equisetum  palustre,  and Deschampsia  cespitosa (tufted hairgrass) 5.0 (surface) 5.4 (mineral sediment) 5.7 (pore water) -watertable was approximately 10 cm below the surface at the time of sampling Champion Lakes Provincial Park UTM Zone 11, 0453623, 5448726 Elise Formation (flow breccia, lava agglomerate, volcanic breccia, tuff, tuffaceous  conglomerate, andesite, basalt, augite porphyry, metamorphosed to greenschist facies) Limestone xenoliths and calcite amygdules found within mafic flows. Economic deposits of Pb, Zn, Ag, associated with sulphides. Carex  sp. Nuphar  lutea  (cow lily) Assorted grasses 6.5 (surface) 6.5 (15cm depth) 6.4 (pore water) Saturated profile Bombi Summit UTM Zone 11, 0459296, 5454509 Bonnington Pluton (hornblende rich granodiorite, granitic gneiss, amphibolite) Carex  sp. Assorted grasses . Organic rich-sediment, not peat Possible urban contamination, due to proximity to road Table 2-1. Location, bedrock geology, vegetation and general notes on the sampling locations. Details of  bedrock geology are from Little, 1982; Hoy and Andrew, 1991. 117°46' 117*34' ,49*15* QUATERNARY •Unconsolidated sediments; till, sand gravel, silt JURASSIC/CRETACEOUS HNELSON INTRUSIONS: granodiorite; mhor quartz diorite and diorite SROSSLAND MONZONITE: biotite-hombtende-augite monzonitB, mainly medium grained JURASSIC I — E U S E FORMATION: < flow breccia, massive andesites L , I and basalts, agglomerate, tuff, breccia, laminated siltstone MOUNT ROBERTS FORMATION: I - — y *lblack siltstone and argillaceous pHdquartzite, slate, greywacke, chert pebble conglomerate, lava; Imestone AGE UNKNOWN TRAIL GNEISS: amphibolite and grey biotite gneiss, hornblende gneiss, mica schist, aplcte and pegmatite Figure 2-1. Geology of  the Trail area and the location of  the smelter and sampling sites. After  Little, (1982). Sampling locations are described in Table 2-1 and are indicated on Figure 2-1. The sites consist of  small, shallow, high ash minerotrophic peatlands. The climate of  the area is dry and not conducive to the formation  of  large peatlands, and so peat-forming environments are small and isolated. The Cominco site is a small wetland (approximatley 0.5 km 2 ) located in the Columbia River Valley which is subject to the input of  metals from several sources: emissions directly from the smelter approximately 0.5 km away, from fugitive  dust containing Pb-Zn ore concentrates that circulates through the valley in the smelter area (e.g. Goodarzi et al., 2001; 2003) and from secondary sources such as road dust. The Thunder Road, Champion Lakes and Bombi Summit sites are wetlands located 1.6, 13.5 and 16 km from the smelter, respectively. In terms of  topography, these three sites are sheltered to varying degrees from the smelter—the Thunder Road site is located above the river valley and is partially sheltered by a ridge of  rock, while the Champion Lakes and Bombi Summit sites are located in the mountainous terrain east of the valley. Secondary sources may affect  each of  the locations. The Thunder Road site may be subject to input from  the Teck-Cominco fertiliser  factory  (Fig. 2-1), and is also located near an unmanaged automobile dumping site. The Champion Lakes sites are located in a provincial park and may be subject to minor input stemming from recreational activities such as vehicle traffic.  The Bombi Summit site is located near a major roadway. In a geologically diverse environment such as the study area, it is difficult  to establish meaningful  background elemental concentrations. Published bedrock geochemical data for  the region (Table 2-2) indicates a high degree of  heterogeneity between the Elise Formation and the Nelson Intrusives, and between rock types within each formation.  The area hosts numerous ore deposits, which were actively mined until the mid-20th century, and zones of  trace element enrichment are naturally present within the bedrock. Hence, anthropogenic impact can not be calculated based on a comparison to a single set of  established background values. Therefore,  data from the peat was compared to trace element concentrations in mineral soils and to atmospheric trace element deposition rates (Goodarzi et al., 2001; 2002; 2003) in order to better discriminate between anthropogenic and geogenic sources of  trace elements, and to illustrate the role of  organic sediments in the distribution of  trace elements in an area subject to a point source of  anthropogenic impact. Nelson Intrusives Elise Formation % Granodiorite Quartz Gabbro K-Tonalite Granodiorite Ankaramite Basalt K-Andesite Si02 Ti02 Al 203 Fe203 MnO MgO CaO Na20 K 2 0 P2O5 64.2 0.6 16 4.9 0.1 2.0 4.8 3.5 3.1 0.2 54.2 0.9 17.4 9.7 0.2 4.2 7.0 4.0 1.8 0.3 60.5 0.6 16.9 6.4 0.1 2.4 4.8 4.7 2.9 0.3 66.0 0.4 16.5 3.7 0.1 1.1 4.0 4.1 3.4 0.2 48.1 0.8 12.2 12.9 0.2 12.4 9.4 2.0 1.6 0.3 51.3 0.9 14.5 11.6 0.2 6.4 9.4 3.1 1.8 0.3 55.4 0.8 15.1 9.4 0.1 4.1 7.8 3.8 2.8 0.3 ppm V Cr Ni Zn Ga Rb Sr Y Zr Nb Ba Pb Th Cu 100 20 5 69 18 93 654 18 130 16 1039 13 9 223 32 13 126 21 72 744 20 65 8 809 9 5 130 22 7 82 19 96 700 21 122 12 1057 9 6 59 10 5 54 17 102 782 14 107 13 1418 12 6 264.3 664.2 212.2 97.2 39.2 541.5 14.5 61.8 2.8 588.2 118.2 314.5 125.8 39.8 97.2 38.0 755.8 18.5 89.5 3.8 685.0 63.8 222.8 26.5 6.8 90.5 52.2 889.5 17.2 114.8 4.2 939.2 37.8 Table 2-2. Compilation of  published geochemistry for  the bedrock underlying the peat sampling sites. Data for  the Nelson Intrusives is presented as published in Sevigny (1990). Data for  the Elise Formation is presented as published in Beddoe-Stephens and Lambert (1981). 2.3 Methodology 2.3.1 Peat  Sampling  and Analytical  Protocol Sampling sites were selected to encompass a range of  anthropogenic impact levels. For each site, general observations were made about the hydrology, vegetation, topography and proximity to sources of  inorganic inputs. Blocks of  peat were removed from the ground by digging a pit with a steel shovel and cutting block samples from the pit wall with a stainless steel knife.  The samples were bagged in sealable freezer  bags and were refrigerated  at 4°C within 24 hours of  sampling. Locations were sampled twice, in September, 1998, and in September, 2000. The first  time, the peat was sampled from near the margin of  the peatland, while samples were taken from  the centre of  the peatland in the second sampling period, in order to determine lateral variation. There are two separate sampling locations for  the Champion Lakes site. The material from the first sampling session consisted of  a shallow (16 cm) high ash lacustrine peat from First Lake. Since this peatland consisted of  a narrow organic deposit at the lake margin, it was not possible to sample profiles  representative of  the margin and the interior at this specific location. The second sampling site from the interior of  a wider, deeper lacustrine peat (> 1 m), sampled from Second Lake (Fig. 2-1). The Bombi Summit sampling site consists of  a thin margin of  organic soil surrounding a pond. The extent of  the organic soils at this site was too small to obtain samples representative of  the interior of  a peatland, and the results presented here are representative of  a margin environment. The samples were split in the following  manner: Surficial  peat (0-8 cm) was sectioned at 1 cm intervals. Peat below this level was sectioned at 3 cm intervals. Subsampling was performed  wearing vinyl gloves, using a plastic knife,  on a freshly  plastic-covered lab bench in a clean room. The samples were air dried on plastic plates, covered loosely by plastic sheeting. To facilitate  final  air drying, the peat samples were placed in a warm (30°C) oven overnight. A sample for  elemental analysis was removed from the middle of  each peat slice, reducing the possibility of  analysing material that was contaminated during the sampling procedure. All elemental analysis for  the first  sampling period was performed  by Becquerel Laboratories, Mississagua, Ontario, Canada, using Instrumental Neutron Activation Analysis (INAA) (As), and Inductively Coupled Plasma Mass Spectrometry (ICP-MS) (Cd, Cu, Pb, Ni, and Zn). Zinc concentrations were obtained using both methodologies, and the results were consistent between methods, although ICP-MS results are reported here. Analytical precision is variable between elements and is dependent upon concentration and the detection limit. The relative standard deviations for  INAA analysis, as calculated with NIST standard 1633, are as follows:  As: 3.1% and Zn: 9.6%. The peat was not fractionated  in any way prior to analysis, as the desired results were for  the bulk sample. Element concentrations for  the second sampling period were determined by ICP-MS following  microwave-assisted dissolution in 20 ml HNO3-HF-HCI solution (5:4:1), in accordance with the recommended procedure supplied by the manufacturer  (CEM, personal communication, 2001). Due to the highly volatile nature of  the organic-rich samples, it was necessary to perform a pre-digestion step in the acid solution, first  at room temperature in open vessels, followed  by a period of  gentle heating on a hotplate, until a visual inspection revealed that the organic matter had been decomposed. The microwave heating procedure was repeated until the sample was completely dissolved; the number of  repetitions depended on the solubility of  the sample. Average errors, as determined by replicate analyses, were as follows:  Sb: 11.3%, As, 13.3%, Cd: 8.9%; Cu: 11.7%; Ni: 6.7%, Pb, 11.5%; Zn: 7.3%. 2.3.2 Soil  Sampling Mineral topsoils were sampled using a 5 cm diameter probe. Two samples were taken from each site unless otherwise specified:  a surface  sample consisting of  vegetation and the topmost 3 cm of  soil, and a 3-10 cm depth core. The samples were wrapped in separate plastic bags and stored in a freezer  until analysis. Soils were homogenized and analysed for  elemental content using INAA and ICP-MS (Becquerel Laboratories) as described for  the peats. In order to evaluate the extent of  metal partitioning between peats and mineral soils, a ratio of  the trace element concentration in the peat versus the concentration in the nearest sampled mineral soil was calculated. Theoretically, geogenic elements are likely to be concentrated in the soil relative to the peat, while atmospherically introduced elements may be present in equal or greater amounts in the peat, due to organic/anaerobic sequestration of  mobile species. McMartin et al. (1999) conducted a study of  the vertical distribution of  trace metals in soil profiles  surrounding a lead-zinc smelter, and found  that significant  migration occurs, particularly in soils that are very close to the smelter. For the purpose of  comparison, soil from 0-3 cm depth was used to calculate the enrichment factor  for  peats of  the same depth interval, while soil from the 3-10 cm interval was used for  peats greater than 3 cm in depth. Factors of  enrichment in the peat relative to the mineral soil were calculated as follows: Peat Enrichment Factor (PEF) = [ M e ] P e a t / [ M e ] s o i i [Me]peat = the concentration of  the trace element in the peat [ M e ] S O i i = the concentration of  the trace element in the mineral soil 2.3.3 Moss  Monitoring  Rates of  Aerial  Deposition A network of  moss monitoring stations was set up as part of  the Geological Survey of  Canada atmospheric trace element deposition impact study in the vicinity of  the Trail smelter, British Columbia (Goodarzi et al., 2001; 2002; 2003; Fig. 2-1). The moss monitoring project is modeled, with some modification,  after  the methodology developed by Martin and Coughtrey (1982) and Swaine et al., (1983) and is described fully  in Goodarzi et al. (2001; 2003). In brief,  the moss monitoring project involves a network of "moss traps", located throughout the Trail area, which consist of  acid-washed Sphagnum moss inserted into a flat  acid washed nylon mesh bag, which is suspended in a frame  on top of  a 2 m pole. After  three months of  exposure, the moss bags are collected, and the elemental concentrations in the moss are measured by INAA and ICP-MS (Becquerel Laboratories, Mississagua, ON). As the mass of  the moss, the area of  the moss trap and the length of  exposure are known factors,  the rate of  elemental deposition at the moss monitoring station can then be calculated as g/ha/3 months. 2.4 Results 2.4.1 Trace  Element  Concentrations  in the Peats The highest concentrations of  Sb, As, Cu, Pb, Ni, and Zn are found  in the Cominco profile,  0.5 km from the smelter (Table 2-3). Concentrations of  trace elements are typically lower in the Thunder Road profile,  although very high Cd concentrations were measured in the samples from the margin of  the peatland. Trace element concentrations are considerably lower in the Champion Lakes and Bombi Summit profiles.  A plot of  average trace element concentrations versus radial distance from  the smelter illustrates a rapid decrease in concentrations with increasing distance (Fig. 2-2). In order to provide a basis for  comparison, the trace elemental concentrations for  a background site underlain by the Trail Pluton, and soil quality guidelines provided by the British Columbia Ministry of  Water, Land and Air Protection (2002) and Environment Canada (2002) are listed in Table 2-3. 2.4.1.1 Cominco Site The peat from the Cominco site has elevated concentrations of  Sb, As, Cd, Cu, Pb and Zn relative to British Columbia (2002) and Canadian (2002) environmental quality standards for  soils (Table 2-3; Fig. 2-3). Of  particular significance  are concentrations of Pb of  up to 5240 ppm and Zn concentrations of  up to 5910 ppm. Concentrations are variable both with peat depth and with distance from the margin of  the peatland. Concentrations of  Cd, Pb and Zn are higher in the surface  peat from the margin than in the surface  peat from  the centre of  the peatland, while As concentrations are higher in the surface  peat from the peatland centre, and concentrations of  Sb, Cu and Ni are approximately the same at both locations. Concentrations of  elements in the peat from the profile  bases are higher in the peat sampled from the margin for  all elements examined with the exception of  Ni (Fig. 2-3). In the Cominco peat margin profile,  the concentrations of  Sb and Zn are highest at the surface  and decrease with depth, while As and Pb are most concentrated at the base of  the profile  and Cd, Cu and Ni concentrations are consistent throughout (Fig. 2-3). In the profile  from the centre of  the peatland, the concentrations of  elements are consistent in the top 7 cm, decline from 8-11 cm, and are highly enriched in the 14-17 cm sample. The 14-17 cm sample corresponds to the macroscopically-observed transition between the aerobic and anaerobic zone, based on the presence of  iron oxide nodules in the shallow peat and gleying (reduced iron) at depth (Table 2-1). It is hypothesised that this zone of  enrichment occurs as a result of  the transition from oxidising to reducing conditions with depth. Mobile species become sequestered under the anaerobic conditions, possibly as sulphides. Elements adsorbed to or coprecipitated with Fe/Mn oxides may become soluble during periods when the watertable is high and reducing conditions prevail at shallower depths in the profile,  and may be subsequently reprecipitated in a zone of  enrichment when oxidising conditions return. Below this zone of  enrichment, element concentrations decrease with depth to a minimum in the peat above the contact with the mineral sediment. The only element that does not behave in this manner is Ni, for  which concentrations vary little with depth or laterally (8.3 - 13.5 ppm in the peat from the centre of  the peatland, and 10.8-15.4 ppm in the peat from the margin). Nickel concentrations do not exceed the listed soil quality standards. All trace elements, with the exception of  Ni, are enriched in the peat relative to the mineral soil (Table 2-4), with PEFs of  up to 31 times for  Sb, 20 times for  As and 15 times for  Pb in selected samples. Nickel PEFs are approximately lor lower throughout the profiles,  indicating a lack of  preferential  sequestration in the peat. The basal mineral sediment is enriched relative to the mineral soil in Sb, As Cu and Pb, but depleted in Zn, and is approximately at unity for  Cd and Ni. Sample Sb As Cd Cu Pb Ni Zn Depth (cm) ppm ppm ppm ppm ppm ppm ppm CO MINCO MARGIN 0-1 126 78.5 61.7 196 3550 10.8 5910 1-2 141 97.8 65.2 223 3830 11.6 5600 2-3 128 153 53.8 215 3780 15.4 4230 3-4 142 116 60.3 227 3830 13.2 4730 4-5 121 164 47.1 214 3940 14.8 3570 17-20 85.0 247 60.1 226 5240 12.4 3040 COMINCO INTERIOR 2-3 111 152 31.5 160 2609 10.4 3430 3-4 130 148 36.9 184 2698 12.2 3938 7-8 112 159 32.9 166 2758 12.8 3280 8-11 79.5 136 4.8 80.4 1600 12.2 586.7 14-17 450 380 30.3 504 3447 10.8 17 77 20-23 145 265 7.9 122 2097 13.5 1105 32-36 11.3 7.0 1.8 26.0 264.4 8.3 212.9 Basal Sediment 32.1 75.7 4.5 94.0 933.3 8.6 390.3 THUNDER ROAD MARGIN 0-1 34.1 60.0 119 82.0 1670 9.2 2730 1-2 37.7 64.4 105 103 2070 10.6 2270 2-3 33.0 62.0 139 84.0 1750 9.4 2900 3-4 37.6 65.4 134 96.0 1980 11.0 2590 5-6 32.5 67.5 104 92.0 1990 10.8 2330 11-14 34.9 59.0 120 96.0 1940 11.0 2300 THUNDER ROAD INTERIOR 0-1 19.5 50.2 27.5 43.1 580.2 8.0 2204 2-3 22.4 52.4 24.8 74.1 536.9 13.3 1942 3-4 25.3 50.4 20.2 92.0 455.2 12.4 1467 5-6 28.3 42.7 24.5 152 613.5 15.1 1570 14-15 18.0 28.3 9.2 142 428.7 16.4 809.2 19-20 10.7 10.7 3.3 353 193.4 15.0 358.3 Basal Sediment 8.0 11.7 1.0 14.9 144.0 13.3 161.7 CHAMPION LAKES MARGIN 0-1 1.2 1.8 2.4 12.0 77.0 3.0 148 1-2 1.6 3.6 3.7 16.0 102 6.0 194 2-3 5.5 8.3 6.0 14.0 172 9.2 214 3-4 3.8 6.8 4.4 13.0 115 6.0 208 4-5 9.0 15.4 6.8 20.0 246 8.8 268 11-16 15.2 15.8 6.5 24.0 324 4.4 178 CHAMPION LAKES INTERIOR 2-5 5.0 7.0 3.3 10.7 79.5 4.6 211.1 14-15 10.6 9.5 3.5 28.3 110.4 8.8 137.4 22-26 2.9 4.0 3.2 28.5 4.1 8.2 45.9 35-36 1.4 2.6 1.0 30.6 1.5 12.3 24.2 49-50 0.8 2.1 1.3 31.0 0.0 12.6 22.6 63-64 0.9 5.3 0.6 21.4 0.0 10.7 50.2 79-80 0.6 1.8 0.6 18.1 0.2 11.1 37.2 Sample Depth (cm) Sb ppm As ppm Cd ppm Cu ppm Pb ppm Ni ppm Zn ppm BOMBI SU MMIT 0-1 2.2 2.9 2.1 14 79.5 12.8 130 1-2 3.0 7.8 2.1 15 104 10.2 122 2-3 2.8 22.6 2.1 13 109 11.2 116 3-4 5.0 7.8 3.3 18 191 11.4 150 4-5 8.8 9.8 7.9 29 476 10.8 248 6-7.5 3.5 2.4 0.3 7.4 61.7 8.8 66.3 7.5-9 1.3 5.2 0.8 6 35 12.4 76.0 MINERAL SOILS Station and Sb As Cd Cu Pb Ni Zn Depth (cm) ppm ppm ppm ppm ppm ppm ppm Cominco 0-3 42.8 45.9 38.0 107 .1215 13.4 4140 Cominco 3-10 14.3 18.7 6.2 31.2 343 12.8 836 Thunder 0-3 5.2 8.9 1.3 25.4 165 14.6 184 Thunder 3-10 3.4 6.1 0.9 25.2 101 14.0 138 Champion 0-3 2.2 8.5 1.1 31.2 55.5 17.2 218 Champion 3-10 1.9 10.3 0.8 31.8 37.0 17.8 142 Castlegar 0-3 2.6 7.2 1.4 22 68 32 112 Castlegar 3-10 3.2 7.9 0.9 20 61 26 132 Background 0-3 0.9 2.7 0.2 11.6 31.8 25.0 76.0 B.C. REMEDIATION STANDARDS FOR SOILS reloc. non-ag. Disp. auth. req. 20 15 1.5 90 100 100 150 40 15 1.5 90 100 500 150 ENVIRONMENT CANADA SOIL ENVIRONMENTAL QUALITY GUIDELINES Agricultural Residential/Park Industrial 20 12 1.4 63 70 50 200 20 12 10 63 140 50 200 40 12 22 91 600 50 360 Table 2-3. Concentrations of  trace elements measured within peats and mineral soils. Environmental soil quality guideline data provided from B.C. Ministry of  Water, Land and Air Protection (2002) lists element concentrations at which soils must be relocated to non-agricultural lands, and at which a permit is required for  disposal. Environment Canada Soil Quality Guidelines (2002) list generalized trace element concentration guidelines for  agricultural, residential/parkland and industrial land uses. Cd 120 E a ' ^80 1.40 5 10 15 Radial Distance from Smelter (km) 20 250 _200 1150 Cu •5-100 5 " 50 0 "0 5 TO 15 Radial Distance from Smelter (km) 20 5000 Pb • £ 4000 Q. I 3 3000 • S 2000 1 1000 0 1 • • 1) 5 10 " 1 5 20 Radial Distance from Smelter (km) Ni 20 ? 1 6 £ 12 0) • • i . • •  • i 0 5 10 15 ; 20 Radial Distance from Smelter (km) 6000, Zn 4000 S 2000 % 5 10 15 Radial Distance from Smelter (km) 20 Figure 2-2. Concentration of  trace elements as a function  of  the radial distance from the smelter. Interior Concentration (ppm) NJ -U CT> O O O O O O O O O Pb Zn N> O O O 10 15-U 20 25 30 BSed o o o> o o -tr T n, -ui n - - / \\ - m } > ->• N> o o o o o o o CO -tk o o o o o o 10 15 20-25 30-BSed \ 7 Sample Depth Sb As Cd Cu Pb Ni Zn (cm) ppm ppm ppm ppm ppm ppm ppm COMINCO MARGIN 0-1 2.9 1.7 1.6 1.8 2.9 0.8 1.4 1-2 3.3 2.1 1.7 2.1 3.2 0.9 1.4 2-3 3.0 3.3 1.4 2.0 3.1 1.2 1.0 3-4 9.9 6.2 9.7 7.3 11.2 1.0 5.7 4-5 8.5 8.8 7.6 6.9 11.5 1.2 4.3 17-20 5.9 13.2 9.6 7.2 15.3 1.0 3.6 COMINCO INTERIOR 2-3 2.6 3.3 0.8 1.5 2.2 0.8 0.8 3-4 9.1 7.9 5.9 5.9 7.9 1.0 4.7 7-8 7.8 8.5 5.3 5.3 8.0 1.0 3.9 8-11 5.6 7.3 0.8 2.6 4.7 1.0 0.7 14-17 31.4 20.4 4.9 16.2 10.0 0.8 2.1 20-23 10.2 14.2 1.3 3.9 6.1 1.0 1.3 32-36 0.8 0.4 0.3 0.8 0.8 0.6 0.2 Basal Sediment 2.2 4.1 0.7 3.0 2.7 0.7 0.5 THUNDER ROAD MARGIN 0-1 6.5 6.7 94.0 3.2 10.1 0.6 14.8 1-2 7.2 7.2 82.9 4.1 12.6 0.7 12.3 2-3 6.3 7.0 110.3 3.3 10.6 0.6 15.8 3-4 11.2 10.7 145.1 3.8 19.6 0.8 18.8 5-6 9.7 11.1 112.5 3.6 19.7 0.8 16.9 11-14 10.4 9.7 130.4 3.8 19.2 0.8 16.7 THUNDER ROAD INTERIOR 0-1 3.7 5.6 21.8 1.7 3.5 0.5 12.0 2-3 4.3 5.9 19.6 2.9 3.2 0.9 10.6 3-4 7.5 8.3 22.0 3.6 4.5 0.9 10.6 5-6 8.4 7.0 26.7 6.0 6.1 1.1 11.4 14-15 5.4 4.6 10.0 5.6 4.2 1.2 5.9 19-20 3.2 1.8 3.6 14.0 1.9 1.1 2.6 Basal Sediment 2.4 1.9 1.1 0.6 1.4 1.0 1.2 CHAMPION LAKES MARGIN 0-1 0.6 0.2 2.3 0.4 1.4 0.2 0.7 1-2 0.7 0.4 3.5 0.5 1.8 0.4 0.9 2-3 2.4 1.0 5.7 0.4 3.1 0.5 1.0 3-4 2.0 0.7 5.4 0.4 3.1 0.3 1.5 4-5 4.6 1.5 8.3 0.6 6.6 0.5 1.9 11-16 7.8 1.5 7.9 0.8 8.8 0.2 1.2 CHAMPION LAKES INTERIOR 2-5 2.2 0.8 3.1 0.3 1.4 0.3 1.0 14-15 5.4 0.9 4.3 0.9 3.0 0.5 1.0 22-26 1.5 0.4 3.9 0.9 0.1 0.5 0.3 35-36 0.7 0.2 1.2 1.0 — 0.7 0.2 49-50 0.4 0.2 1.5 1.0 — 0.7 0.3 63-64 0.5 0.5 0.8 0.7 — 0.6 0.4 79-80 0.3 0.2 0.7 0.6 - 0.6 0.3 BOMBI SUMMIT 0-1 0.8 0.4 1.5 0.6 1.2 0.4 1.2 1-2 1.1 1.1 1.5 0.7 1.5 0.3 1.1 2-3 1.1 3.2 1.5 0.6 1.6 0.4 1.0 3-4 1.6 1.0 3.7 0.9 3.2 0.4 1.1 4-5 2.8 1.2 9.0 1.4 7.9 0.4 1.9 6-7.5 1.1 0.3 0.4 0.3 1.0 0.3 0.5 7.5-9 0.4 0.7 0.9 0.2 0.6 0.5 0.6 Table 2-4. Calculated Peat Enrichment Factors (PEFs) for  peats relative to nearby mineral soils. 2.4.1.2 Thunder  Road  Site Trace elemental concentrations are relatively consistent with depth throughout the profile  for  the peat sampled near the margin of  the Thunder Road site (Table 2-3; Fig. 2-4). In the case of  the peat from the centre of  the peatland, concentrations decline steadily with depth, with the exception of  Cu, which is most concentrated (353 ppm) in the peat immediately above the basal mineral sediment. Copper exists as a free  ion in solution under acidic oxidising conditions, but may form a sulphide under low Eh conditions (Brookins, 1988) and so it is hypothesised that dissolved Cu migrates from the peat surface  to the zone of  reducing conditions at depth, provided that it is not complexed by other mechanisms. Arsenic tends to exist as a soluble species under acidic oxidising conditions, and as a sulphide under reducing conditions (Brookins, 1988), and is present in higher concentrations in the interior peat profile  than in the margin profile  (Fig. 2-4), similar to the Cominco site (Fig. 2-3). Cd concentrations in the peat from the margin are approximately 4 times higher than the concentrations at the centre of  the peatland (e.g. 119 ppm (margin) versus 27.5 ppm (centre) for  the 0-1 cm sample). Arsenic, Pb, Zn and Cd decrease with depth in the interior profile,  while Sb and Ni concentrations are approximately the same. Unlike the Cominco profile,  concentrations are typically lower in the underlying mineral sediment than in the peat, with the exception of  Sb, Ni and As, which are approximately the same. Calculated PEFs for  Zn and Cd are higher at the Thunder Road site than the Cominco site, despite the lower concentrations of  Zn in both profiles  and Cd from the interior profile.  PEFs for  Cd range from 82-130 for  the margin peat profile  and fall  to 4-27 for  the interior peats. This suggests that Cd is migrating as an aqueous species from the soil, and is progressively removed from solution by the peat. PEFs indicate that the peat is also enriched relative to the soil in As, Sb, Cu and Pb. Nickel shows no enrichment in the peat. era' 3 1-1 CD K> Margin K> -J o o 3 o a> 3 K. o 3 o w SL ciT o a> a. p o o> BS CD 3 3* CO H 3" 3 3 Cu CD >-t o a. >-i o Eh * a> w Interior Concentration (ppm) _>.-». M ro cn o cn o cn _ o o o o o o o o o o o 2.4.1.3 Champion Lakes Site Concentrations of  trace elements generally increase with depth in the margin profile,  particularly Pb, As and Zn (Table 2-3; Fig. 2-5). In the deeper interior profile, concentrations of  As, Pb, Zn, Sb and Cd decrease below 22 cm depth from maximum values near the surface  (79.5 ppm Pb, 211 ppm Zn) to minimum concentrations at 80 cm depth. Zinc concentrations are lower in the middle section of  the profile  (35-50 cm) than at the profile  base, suggesting either downward mobilisation from the surface  to the base of  the profile  or upward mobilisation of  geogenic Zn from the base of  the profile.  Copper occurs in higher concentrations below 15 cm, with the highest concentrations occurring between 22 and 50 cm (31 ppm), similar to the pattern of  distribution within the interior Thunder Road profile.  Nickel concentrations increase below 26 cm to a maximum of 12.6 ppm at 50 cm depth. PEFs for  the margin peat indicate that Pb and Cd are enriched in the peat relative to the mineral soil throughout the profile.  Antimony, As and Zn are relatively enriched at the base of  the peat profile,  but not at the surface.  PEFs of  1 or less are typical for  most elements in the interior peat at depths below 35 cm, with the exception of  Cd, which is slightly enriched in the peat (up to 1.5). PEFs are greater than 1 for  Sb, As and Pb in the surficial  interior peat. However, the actual magnitude of  the enrichment is small, due to the relatively-low elemental concentrations in both the peat and the soil, and the degree of preferential  sequestration by the peats is minimal for  most elements. Arsenic and Pb concentrations occasionally exceed Environment Canada (2002) guidelines for agricultural and residential/parkland use (Table 2-3). 300 200 100 L_ 0+ o o (1) c 40-i 30 20 10 -r- 0 E o a. a. c o (0 k. •*•> c a> o . c o o 400 30a 200 100 0 o 30 20 10 0 sample depth -Q c CL N H •o 0) CO CQ X1 tf>  ~0 =3 ._ CO < O O Z t l t H o T— a) sample depth co Figure 2-5 Concentration of  trace elements in the Champion Lakes profiles. 2.4.1.4 Bombi Summit  Site Antimony, Cd, Cu, Pb and Zn in the Bombi Summit profile  increase in concentration from the surface  to a maximum in the 4-5 cm sample, and then decrease to a minimum at the profile  base (Figure 2-6). Arsenic is most concentrated in the 2-3 cm sample and Ni concentrations are consistent throughout the profile.  Lead, Zn, Cd and As are present at concentrations that exceed the British Columbia (2002) and Environment Canada (2002) soil quality guidelines for  agricultural and residential/parkland use. Peat enrichment factors  for  the site are close to unity for  Zn, Cu, As and Sb for  all but the 4-5 cm depth sample, while Pb and Cd are relatively concentrated in the peat (up to 8 times for  Pb and 9 times for  Cd). Nickel is relatively concentrated in the soil by a factor  of  approximately 2. 2.4.2 Moss  Monitoring  Results The moss monitoring results are discussed in detail in Goodarzi et al., 2001; 2003, but are summarised briefly  here. Arsenic, Cd, Cu, Pb, and Zn deposition was high near the smelter, and decreased with distance (Table 2-5). After  correcting for  geogenic input, approximately 75 944 kg/ha of  Pb and 320 719 kg/ha of  Zn were deposited near the Cominco site in the span of  a two-year period (1997-1999), based on the results in Goodarzi et al. (2003). During the same period, 14 433 kg/ha of  Pb and 64 919 kg/ha of  . Zn were deposited in the area around the Thunder Road site, while 52.5 kg/ha of  Pb and 0 kg/ha of  Zn were measured at the Champion Lakes site. No moss monitoring station is located near the Bombi Summit site. The nearest moss monitoring station is located in the river valley near the town of  Castlegar and recorded 125 .3 kg/ha of  Pb and 246.9 kg/ha of  Zn during the two-year period. 500 400 E Q. Q. C o U—< ca c 0) o c o O sample depth o> CO jd C 0. N + \ .O </) T3 3 ._ co < o o z H i l l Figure 2-6. Concentration of  trace elements in the Bombi Summit profile. Table 2-5. Concentration of  selected trace elements in the moss used as sampling media for  moss monitoring stations in the Trail area. The concentrations are statistically corrected for  geogenic input, as described in Goodarzi et al., (2001; 2003). Cumulative impact over the two-year sampling period is indicated in the right column. Concentration Measured in Moss Sampling Media (ppm) Total Deposition Element Fall '97 Winter '98 Spring '98 Summer '98 Fall '98 Winter '99 Spring '99 Summer '99 (kg/ha/2 years) Cominco As 9.9 14.7 29.6 16.6 25.2 26.9 30.1 32.8 581 Cd 46.5 34.3 78 42.7 57.2 88 56.6 82 1516 Cu 103 73 159 99 83 86 85 137 2574 Pb 2302 3860 2056 3152 3662 4007 3294 1969 75944 Zn 10734 18748 9559 4753 8975 20307 14775 14779 320719 ThunderRoad As 3.1 7 4.6 3.4 2.9 7 2 3 103 Cd 8.27 15.2 23 44.6 44.2 17.6 27.8 52.9 7301 Cu 6 17 24 28 25 19 12 12 447 Pb 302 631.4 613 1010 610 568 422 462 14432 Zn 1592 2409 2806 4142 2913 2162 2426 2324 64919 Champion Lakes As ND ND ND ND ND ND ND ND ND Cd ND ND ND 0.14 ND 0.63 4.06 0.42 16 Cu ND ND ND ND ND 17 9 1 84 Pb ND ND ND 7.8 ND 3 6 ND 52 Zn ND ND ND ND ND ND ND ND ND Castlegar As ND ND ND ND ND ND ND ND ND Cd ND ND 0.66 0.85 0.29 1 1.31 6.34 33 Cu ND ND ND ND 4 1 ND 1 19 Pb ND 1.1 0.2 16.6 9.4 1.2 7.9 3.7 125 Zn ND ND ND 19 25 33 ND 2 247 However, the peat sampling site is located approximately 4 km east of  this station and is elevated above the river valley, and is likely to be subject to less impact from the smelter than the moss monitoring station, as fallout  from the smelter is partially topographically confined  to the valley (Goodarzi et al., 2003). The actual deposition from the smelter at the Bombi Summit site is likely to be similar to or less than that measured at the Champion Lakes site. 2.4.3 The  distribution  of  trace elements  in the environs of  the stationary  sources The distribution patterns of  As, Cd, Cu, Pb and Zn in the peat around Trail indicate that heavy localised deposition of  elements has occurred in the area surrounding the.smelter (Table 2-3). Trace element concentrations in the peat decrease with radial distance from the source, to the point where it is difficult  to distinguish between geogenic and anthropogenic elements in the peats and soils (e.g. Champion Lakes, Bombi Summit). Concentrations of  trace elements in the peats and soils and rates of  atmospheric elemental deposition from the Thunder Road, Champion Lakes and Bombi Summit sites were normalised to the results from the Cominco site in order to compare the concentrations of  elements found  within each of  the sampling media. Results for individual elements are presented in Table 2-6, although the combined average ratio for As, Cd, Cu, Pb and Zn is discussed in the text. As Ni is assumed to be largely geogenic in origin, and moss monitoring results for  Sb were not determined by Goodarzi et al. (2003), these elements were not included. Average and relative concentrations for  the margin and interior peat profiles  were calculated separately. As with the PEF calculations, average concentrations for  the peat profiles  were calculated for  0-3 cm and >3 cm to coincide with the soil data. The average concentrations of  As, Cd, Cu, Pb and Zn in the Thunder Road peat are approximately 47-55% those in the Cominco peats, or 80% when the high concentrations of  Cd obtained from the margin peat at Thunder Road are included. For the Thunder Road soil, the average elemental concentration is 13% of  the Cominco soil concentration for  the top 3 cm. Rates of  deposition recorded by the Thunder Road moss monitoring station average 25% the Cominco station rate when Cd is included, but drop to 19% when it is excluded (the relative deposition rate for  Cd is 48%). This illustrates that Cd is deposited at the Thunder Road site at a rate that is anomalous relative to As, Cu, Pb and Zn, which accounts to some extent for  the high concentration found  within the peat relative to the soil. Assuming that the Cd is deposited as an oxide, it would be readily dissolved in a moderately acidic environment and will become preferentially enriched in the peat, as Cd has a strong organic affinity  (e.g. Sager, 1992) and forms sulphide complexes readily under reducing conditions (Brookins, 1988). Both Cd and Zn behave in a similar fashion  and are likely to exist as free  ions under the conditions present within an acidic soil (Brookins, 1988). Both Cd and Zn show a strong degree of fractionation  in the Thunder Road soil relative to the peat. Although relative deposition rates are 48% and 20% for  Cd and Zn, respectively, relative soil concentrations are 3.3% and 4.4%, while relative peat concentrations are 79-213%) and 50-64%, respectively, indicating a strong preferential  sequestration within the peat. The relatively high deposition ratio for  Thunder Road compared to the lower soil concentration ratio, combined with the peat concentration ratio that is higher than both these parameters, further  indicates that trace elements are preferentially  sequestered in the peat relative to the mineral soil. The anomalous Cd deposition measured by the moss monitoring station also suggests that there is a secondary source of  Cd at this location. The most probable source of  Cd is the Teck-Cominco fertiliser  factory,  which is located near the Thunder Road site and which currently manufactures  nitrogen fertiliser,  but which previously produced phosphate fertilisers  as well. Cd is associated with apatite, the main component of phosphate fertilisers.  It is likely that the relatively high Cd deposition rate measured by the moss monitoring station is derived from historical fugitive  dust from the fertiliser factory  that is circulating in the area. A stream sediment survey (Goodarzi et al., in prep) also detected high concentrations of  Cd in this area. Peat Concentration, Relative to Cominco Peat (%) As Cd Cu Pb Zn Average Average (-Cd) Thunder Road 0-3 (M ) 56.6 200.3 42.4 49.2 50.2 79.8 49.6 Thunder Road 0-3 (1) 33.7 83.0 36.6 21.4 60.4 47.0 Thunder Road >3 (M ) 36.4 213.1 42.6 45.4 63.7 80.2 47.0 Thunder Road >3 (1) 18.1 74.9 102.4 19.7 57.9 54.6 Champion Lake 0-3 (M) 4.2 6.7 6.6 3.1 3.5 4.8 Champion Lake 0-3 (1) 4.6 10.4 6.7 3.0 6.2 6.2 Champion Lake >3 (M) 7.2 10.6 8.5 5.3 5.8 7.5 Champion Lake >3 (1) 2.3 8.8 14.6 0.9 2.9 5.9 Bombi <3 10.1 3.5 6.6 2.6 2.3 5.0 Bombi >3 4.1 9.8 9.4 7.3 3.9 6.9 Soil Concentration, Relative to Cominco Soil (%) As Cd Cu Pb Zn Average Thunder Road 0-3 cm 19.5 3.3 23.6 13.6 4.4 12.9 Thunder Road 3-10 cm 32.6 14.7 80.8 29.4 16.5 34.8 Champion Lakes 0-3 cm 18.6 2.8 29.0 4.6 5.3 12.0 Champion Lakes 3-10 cm 55.1 13.1 101.9 10.8 17.0 40.0 Atmospheric Deposition Rate for 1997-1999, relative to the Cominco site station (%) As Cd Cu Pb Zn Average (- Cd) Thunder Road 17.8 48.2 17.4 19.0 20.2 24.5 18.6 Champion Lakes 0 1.1 3.3 0.07 0 0.9 M and I indicate peats sampled from the margin and interior of the peatland, respectively. Table 2-6. Relative concentrations of  trace elements in the Thunder Road and Champion Lakes peats and soils, as compared to Cominco peats and soils. Relative rates of deposition as determined by moss monitoring stations are calculated from data in Goodarzi et al., 2001; 2003. The average concentrations of  Sb, As, Cd, Cu, Pb and Zn in the Champion Lakes site are 5-7.5% of  the Cominco peat concentrations. Average relative soil concentrations are 12-40%, although this average is skewed by the 102% Cu ratio and 55% As ratio in the 3-10 cm Champion Lakes soil. Soil concentration ratios for  Cd, Pb and Zn range between 3 and 17%, while the average relative rate of  deposition recorded by the moss monitoring station is approximately 1%. The deposition ratio for  Cu was anomalously high relative to the other elements, suggesting a localised source, possibly geogenic dust. Average concentrations for  Cu in the Elise Formation bedrock range between 38 and 118 ppm (Beddoe-Stephens and Lambert, 1981). This discrepancy between the soil and peat concentration ratios and the rate of  deposition strongly suggest a significant  geogenic contribution to the peat trace elemental composition, particularly for  As and Cu. The extent to which peat is enriched in trace elements relative to the mineral soil is relative to the amount of  overall elemental loading at the site (Fig. 2-7). When the average peat trace element concentrations for  the intervals of  0-3 cm and >3 cm depth are plotted against [Me]peat - [Me]soji for  the same depth intervals, a linear relationship (R2>0.85 for  all elements) with a slope of  approximately 1 (slope ranges from 0.79 to 1.33) occurs for  all elements. The closer the slope is to unity without exceeding it, the greater the element's affinity  for  the peat relative to the soil. Arsenic, with a slope of 0.88, has the strongest affinity  for  peat, while the slope of  1.33 for  Ni indicates higher concentrations in the soil. The enrichment of  trace elements in the peats from areas of heavy deposition illustrates the capacity of  organic matter to adsorb and retain metals relative to mineral soils, particularly if  the mineral soils contain little clay (Chapter 4) and little organic carbon (0.2-3.5%; Goodarzi, unpublished data). 2.5 Discussion The results of  this study demonstrate the potential for  natural peatlands to serve as a sink for  trace elements released into the environment and the importance of  maintaining the existing peatlands in areas which are subject to a large input of  trace elements from anthropogenic activities. Experimental constructed wetlands currently in use by Teck-Cominco have been shown to be effective  as a means to remove Cd, Zn and As from smelter waste leachate (W. Duncan, personal communication, 2003). In another example, Holmstrom and Ohlander (1999) discuss the disposal of  high sulphide mine tailings in anaerobic environments in order to prevent acid mine drainage and trace element mobilization. There, the organic material serves to further  immobilize the trace metals. The effective  sequestration of  trace elements leached from the surrounding soils by natural peatlands supports the development of  constructed wetlands for  large-scale soil and water remediation projects. Drainage and subsequent oxidation of  anthropogenically impacted peat (for  urban development, for  example) would likely result in the remobilisation of  the trace el ements into the surrounding soil and water. Incidences of  trace element pollution associated with the oxidation of  black shales (Reichenbach, 1993) and with dredging of  anaerobic estuarine sludges (Forstner, 1987 and references  therein) have been documented. Hence, the release of  metals sequestered in peats from highly polluted areas would be cause for concern. The partitioning of  Sb, As, Cd, Cu, Pb and Zn between the mineral soils and the peats and the evident vertical translocation of  trace elements, particularly Cu, within the profile,  demonstrates the mobile nature of  the deposited trace elements. This indicates that topsoil sampling programs intended to determine environmental impact of  smelters on the surrounding areas may not reflect  the full  impact of  the source, as a significant proportion of  the deposited trace elements are subject to migration rather than in situ accumulation. The evidence for  post depositional translocation suggests that, prior to sequestration in the peat, the trace elements are highly mobile and bioavailable. For this reason, land use guidelines for  areas surrounding Pb-Zn smelters should reflect  the potential for  trace element impact on the local environment, and make recommendations accordingly to limit the impact of  trace elements on the local population. Recommendations might potentially include: restricting agricultural production and home gardening to non-edible plants or plants known not to accumulate metals in the edible parts; prohibiting grazing and feed  production for  livestock on local soils to prevent potential poisoning or possible bioaccumulation of  metals; and limiting the exposure of children to contaminated soils by routinely providing clean surface  fill  in playgrounds. C U y = 0.7997X • O U 0.2155 o co V120 (0 <13 Q. 40 20 40 60 80 100 120 140 160 180 [Me]peat (ppm) _ 250 O ,co 200 CD 5 150 „ 10  CO A 5 0 01 0 " -50 250 o CO 200 cu S 150 i 100 m Cl> i>0 CL cu 0 -50 y = 0.8835X -7.0171 50 100 150 [Mejpeat (ppm) 120 Cd y = 0.5325X +2.163 CO CU s 1 80 Outliers nS 40 • cu n • a) 0 • • 2 3 20 40 60 80 100 120 140 [Me]peat (ppm) Cu y = 0.8013x-20.733 50 100 150 [Mejpeat (ppm) 200 250 O 4000, CO J2 3000 2000 CO CD q. 1000 CD o 4 V) CU 0 ^ 1 -4 M -8 <i) cx cu -12 2 P h y = 0.7943X + r u 27.95 1000 2000 3000 [Mejpeat (ppm) 4000 5000 Ni y = 1.3317x • 18.834 [Mejpeat (ppm) Zn y = 0.8849x -126.32 __ O O) CU 2000 •2 i 1000 • CU a) / Outliers j CL cu 0 I • y --0 1000 2000 3000 4000 5000 6000 [Me]peat (ppm) • Cominco • Thunder Road a Champion Lake Figure 2-7. Concentration of  trace elements in peats relative to the differential concentrations in peats versus mineral soils. Peat enrichment factors  and relative concentration and deposition ratios demonstrate that the underlying geology significantly  influences  the concentration and distribution of  trace elements within the peat, particularly in areas with reduced anthropogenic impact. The concentrations of  As, Cu, Ni and Zn are higher within the Champion Lakes soil than in the peats. The low rates of  atmospheric deposition reported for  this site (Goodarzi, 2001; 2003), and the fact  that these trace elements have maximum concentrations in the basal peat in the margin profile  at this location suggest a significant geogenic contribution for  these elements. Elements that are enriched in the peat, such as Cd and Pb, are likely of  both anthropogenic and geogenic origin, as indicated by both the moss monitoring data and the geochemistry of  the underlying bedrock (Table 2-2). At sites subject to high anthropogenic impact, the influence  of  geogenic sources is difficult to discern by bulk chemical methods alone and further  examination by methods such as sequential leaching, microscopy and geochemical fingerprinting  is necessary. The concurrent use and subsequent comparison of  results from several types of  sampling media and monitoring methods makes a more complete understanding of  the impact of anthropogenic sources of  trace elements on the surrounding environment possible, and may help to identify  input from geogenic sources or secondary anthropogenic sources. 2.6 Conclusions Peatlands offer  a useful  source of  trace metal deposition information  for environmental monitoring for  both highly contaminated and less contaminated areas. However, comparisons with other data, such as soil trace element concentrations and reliable atmospheric deposition rates are important for  determining the overall environmental impact in the region, as peats may contain enriched concentrations of anthropogenic elements, but may be relatively depleted in geogenic elements. Conversely, mineral soils may be relatively depleted in deposited trace elements due to subsequent leaching. The combined use of  peat, mineral soil and atmospheric deposition data makes it possible to develop an understanding of  the role of  anthropogenic point sources in supplying trace elements to the surrounding environment, and the subsequent partitioning that takes place following  deposition. 2.7 References Allard, B., Hakansson, K. and Karlsson, S., 1987. The importance of  sorption phenomena in relation to trace element speciation and mobility. In: Speciation of  Metals in Water, Sediment and Soil Systems, L. Landner, ed. Lecture Notes in Earth Sciences, Springer Verlag, Heidelberg, pp. 99-112. Beddoe-Stephens, B. and Lambert, R.St.J, 1981. Geochemical, mineralogical and isotopic data relating to the origin and tectonic setting of  the Rossland volcanic rocks, southern British Columbia. Canadian Journal of  Earth Sciences, 18: 858. Bendell-Young, L., 1999. Contrasting the sorption of  Zn by oxyhydroxides of  Mn and Fe and organic matter along a mineral-poor to mineral-rich fen  gradient. Applied Geochemistry, 14: 719-734. British Columbia. Ministry of  Water, Land and Air Protection, 2002. Waste Management Act: Contaminated Sites Regulation. Available: http://www.qp.gov.bc.ca/statreg/regAVAVasteMgmtAVasteMgmt375_96/375_96.htm Brookins, D.G., 1988. Eh-pH diagrams for  geochemistry. Springer Verlag, Berlin. 176p. Canada. Environment Canada, 2002. Canadian Soil Quality Guidelines. Available: http://www.ccme.ca/assets/pdf/el_06.pdf Clark, M.W., McConchie, D., Lewis, D.W. and Saenger, P., 1998. Redox stratification  and heavy metal partitioning in Avicennia-AovamaXzA  mangrove sediments: A geochemical model. Chemical Geology, 149: 147-171. Elder, J.F., 1988. Metal biogeochemistry in surface-water  systems—A review of  principles and concepts. U.S. Geological Survey Circular 1013,43p. Elgersma, F., Schinkel, J.N. and Weijnen, M.P.C., 1995. Improving environmental performance  of  a primary lead and zinc smelter. In: Heavy Metals, W. Salomons, U. Forstner, and P. Mavel, eds. Springer Verlag, Heidelberg, pp. 193-208. t Enns, K., 1996. Biomonitoring System Development for  Cominco Ltd.—Synopsis Report. Larkspur Biological Consultants Ltd. 29 p. Espi, E., Boutron, C.F., Hong, S., Pourchet, M., Ferrari, C., Shotyk, W. and Charlet, L., 1997. Changing concentrations of  Cu, Zn, Cd and Pb in a high altitude peat bog from Bolivia during the last three centuries. Water, Air and Soil Pollution, 100: 289-296. Falcone, S. K. and Schobert, H.H., 1986. Mineral transformations  during ashing of  selected low-rank coals. In: Mineral Matter and Ash in Coal, K.S. Vorres, ed. American Chemical Society Symposium Series 301, pp. 70-89. Forstner, U., 1987. Metal speciation in solid wastes—Factors affecting  mobility. In: Speciation of  Metals in Water, Sediment and Soil Systems, L. Landner, ed. Lecture Notes in Earth Sciences, Springer Verlag, pp. 13-42. Gilbertson, D.D., Grattan, J.P., Cressey, M. and Pyatt, F.B., 1997. An air-pollution history of  metallurgical innovation in iron and steel-making: A geochemical archive of  Sheffield.  Water, Air and Soil Pollution, 100: 327-341. Godbeer, W.C. and Swaine, D.J., 1995. The deposition of  trace elements in the environs of  a power station. In: Environmental Aspects of  Trace Elements in Coal, D.J. Swaine and F. Goodarzi, eds., Kluwer Academic Publishers, Dordrecht, pp 178-203. Goodarzi, F., 1995. Geology of  trace elements in coal. In: Environmental Aspects of  Trace Elements in Coal, D.J. Swaine and F. Goodarzi, eds., Kluwer Academic Publishers, Dordrecht, pp. 51-75. Goodarzi, F., Sanei, H. and Duncan, W.F., 2001. Monitoring the distribution and deposition of  trace elements associated with a zinc-lead smelter in the Trail area, British Columbia, Canada. Journal of  Environmental Monitoring, 3: 515-525 Goodarzi, F., Sanei, H., Garrett, R.G. and Duncan, W.F., 2002. Accumulation of  trace elements on the surface  soil around the Trail smelter, British Columbia, Canada. Environmental Geology, 43: 29-38. Goodarzi, F., Sanei, H. and Duncan, W.F., 2003. Deposition of  trace elements in the Trail region, British Columbia; an assessment of  the environmental effect  of  a base metal smelter on land. Geological Survey of Canada Bulletin 573, 50 p. Gorres, M. and Frenzel, B., 1997. Ash and metal concentrations in peat bogs as indicators of  anthropogenic activity. Water, Air and Soil Pollution, 100: 355-365. Holmstrom, H. and Ohlander, B., 1999. Oxygen penetration and subsequent reactions in flooded  sulphidic mine tailings: a study at Stekenjokk, northern Sweden. Applied Geochemistry, 14: 747-759. Honeyman, B.D., 1999. Colloidal culprits in contamination. Nature, 397: 23-24. Hoy, T. and Andrew, K.P.E., 1991. Geology of  the Rossland area, southeastern British Columbia. Geological Fieldwork, 1990, British Columbia Gelogical Survey Branch, pp 21 -31. Kempter, H., Gorres, M. and Frenzel, B., 1997. Ti and Pb concentrations in rainwater fed  bogs in Europe as indicators of  past anthropogenic activities. Water, Air and Soil Pollution, 100: 367-377. Kersting, A.B., Efurd,  D.W., Finnegan, D.L., Rokop, D.J., Smith, D.K. and Thompson, J.L., 1999. Migration of  plutonium in ground water at the Nevada Test Site. Nature, 397: 56-59. Kuster, H. and Rehfiiess,  K.-E., 1997. Pb and Cd concentrations in a southern Bavarian bog profile  and the history of  vegetation as recorded by pollen analysis. Water, Air and Soil Pollution, 100: 379-386. Little, H.W., 1960. Nelson Map-Area, West Half,  British Columbia (82 FW1/2). Geological Survey of Canada Memoir 308. 205p. Little, H.W., 1982. Geology of  the Rossland-Trail Map-Area, British Columbia. Geological Survey of  Canada Paper 79-26, 3 8p. Martin, M.H. and Coughtrey, P. J., 1982. Biological Monioing of  Heavy Metal Pollution. Applied Science Publishers, London and New York, 475p. Martinez-Cortizas, A., Pontevedra-Pombal, X., Novoa Munos, J.C. and Garcia-Rodeja, E., 1997. Four thousand years of  atmospheric Pb, Cd and Zn deposition recorded by the ombrotrophic peat bog of  Penido Velio (Northwestern Spain). Water, Air and Soil Pollution, 100: 387-403. Mathews, W.H. and Bustin, R.M., 1994. Trace metal geochemistry of  peat under a sanitary landfill—a reconnaissance. Environmental Geology, 23: 14-22. McMartin, I., Henderson, P.J. and Nielsen, E., 1999. Impact of  a base metal smelter on the geochemistry of soils of  the Flin Flon region, Manitoba and Saskatchewan. Canadian Journal of  Earth Sciences, 36: 141-160. Rasmussen, P.E., 1996. Trace Metals in the Environment: A Geological Perspective. Geological Survey of  Canada Bulletin 429, 26 p. Reichenbach, I., 1993. Black Shale as an Environmental Hazard: A Review of  Black Shales in Canada. Geological Survey of  Canada Open File Report 2697, 62 p. Sager, M., 1992. Chemical speciation and environmental mobility of  heavy metals in sediments and soils. In: Hazardous Metals in the Environment, M. Stoeppler, ed. Elsevier, Amsterdam, pp. 133-175. Schell, W.R., Tobin, M.J., Novak, M.J.V, Wieder, R.K. and Mitchell, P.I., 1997. Deposition history of trace metals and fallout  radionulides in wetland ecosystems using 210Pb chronology. Water, Air and Soil Pollution, 100: 223-239. Sevigny, J.H., 1990. Geochemistry of  the Jurassic Nelson plutonic suite, southeastern British Columbia. In: Project Lithoprobe; southern Canadian Cordillera transect workshop. Lithoprobe Report 11, pp. 41-52. Shotyk, W., 1995. Peat bog archives of  atmospheric metal deposition: geochemical evaluation of  peat profiles,  natural variations in metal concentrations, and metal enrichment factors.  Environmental Reviews, 4: 149-183. Shotyk, W., 1996. Natural and anthropogenic enrichments of  As, Cu, Pb, Sb, and Zn in rainwater-dominated versus groundwater-dominated peat bog profiles,  Jura Mountains, Switzerland. Water, Air and Soil Pollution, 90: 375-405. Shotyk, W., Cheburkin, A.K., Appleby, P.G., Fankhauser, A. and Kramers, J.D., 1996. Two thousand years of  atmospheric arsenic, antimony and lead deposition recorded in an ombrotrophic peat bog profile,  Jura Mountains, Switzerland. Earth and Planetary Science Letters, 145: E1-E7. Shotyk, W., Cheburkin, A.K., Appleby, P.G., Fankhauser, A. and Kramers,.J.D., 1997. Lead in three peat bog profiles  Jura Mountains, Switzerland: Enrichment factors,  isotopic composition and chronology of atmospheric deposition. Water, Air and Soil Pollution, 100: 297-310. Steinnes, E., 1997. Trace element profiles  in ombrogenous peat cores from Norway: Evidence of  long range atmospheric transport. Water, Air and Soil Pollution, 100: 405-413. Swaine, D.J., Godbeer, W.C. and Morgan, N.C., 1983. Use of  moss to measure the accession of  trace elements to an area around a power station. Proceedings of  the International Conference  of  Heavy Metals in the Environment, Heidelberg, 2: 1053-1056. Trudinger, P.A., Swaine, D.J. and Skyring, G.W., 1979. Biogeochemical cycling of  elements—General considerations. In: Trudinger, P. A. and Swaine, D.J., eds. Biogeochemical Cycling of  Mineral-Forming Elements. Elsevier Scientific  Publishing Company, Amsterdam, pp. 1-27. Weiss, D., Shotyk, W., Cheburkin, A.K., Gloor, M. and Reese, S., 1997. Atmospheric lead deposition from 12,400 to CA 2,000 Yrs BP in a peat bog profile,  Jura Mountains, Switzerland. Water, Air and Soil Pollution, 100: 311-324. West, S., Charman, D.J., Grattan, J.P. and Cherburkin, A.K., 1997. Heavy metals in Holocene peats from south west England: Detecting mining impacts and atmospheric pollution. Water, Air and Soil Pollution, 100: 343-353. Chapter 3 Geochemical finger-printing  in an anthropogenically-impacted area of heterogenous geochemistry—Can rare earth elements and conservative element ratios be used to identify  source? 3.1 Introduction One of  the difficulties  in environmental geochemical studies is distinguishing between different  sources of  elements. In the area surrounding the Teck-Cominco Pb-Zn smelter in Trail, British Columbia, Canada, the heterogeneity in bedrock geochemistry and the long history of  mining, smelting of  ores from different  sources and other human activities have made it difficult  to confidently  assign a geochemical baseline to any site. Hence, it is difficult  to calculate the impact of  anthropogenic activities on the trace element chemistry of  the peats and soils within the area. Trace metal concentrations based on bulk samples provide limited information  from which the influence  of anthropogenic activity can only be inferred.  The contribution of  trace elements from naturally-occurring mineral constituents must be determined in order to estimate the athropogenic impact at a given location. A combination of  several different  methods may provide a clearer picture. Two methods of  geochemical fingerprinting  were applied in order to determine their usefulness  and their shortcomings: normalisation ratios and rare earth element composition. This study is part of  a large project by the Geological Survey of  Canada and Teck Cominco to measure the environmental impact of  the smelter on the surrounding area. Goodarzi et al. (2001; 2002; 2003) present findings  from a moss monitoring program that investigated the magnitude and elemental composition of  atmospheric deposition throughout the region. Total element concentrations in peat profiles  compared with mineral soils and the results of  the atmospheric deposition study are discussed in Chapter 2. The mode of  occurrence of  selected trace elements as a function  of  the peat profile depth is presented in Chapter 6 and the chemistry and morphology of  particulate mineral matter within the peats is discussed in Chapter 4. 3.2 Study Area Trail, British Columbia (Fig. 3-1) is situated in the Columbia River Valley, which is the boundary between the Monashee (west side) and Selkirk (east side) Mountains. The bedrock in the area consists mainly of  Upper Jurassic plutonic rocks (the Nelson Intrusives), and Lower Jurassic metavolcanic and metasedimentary rocks of  the Elise Formation (Little, 1982). The Teck-Cominco Trail smelter has been in operation since 1896. Lead and zinc are the major products produced, but the smelter also produces Ag, Au, Cd, Bi, In, Ge, Cu, Sb, ammonium sulphate fertiliser  and sulphur products. Although environmental conditions in the area surrounding the smelter have improved in recent years with the implementation of  better production and containment technologies and ore handling practices, Pb, Zn, Cd, Cu, and As continue to be emitted by the operation. 3.3 Methodology 3.3.1 Peat  Sampling  and Analytical  Protocol Sampling sites were selected to encompass a range of  anthropogenic impact levels. For each site, general observations were made about the hydrology, vegetation, topography and proximity to sources of  inorganic inputs. All sampling sites are minerotropic peats or organic soils (groundwater influenced)  with vascular vegetation (Table 3-1). Blocks of  peat were removed from the ground by digging a pit with steel shovels and cutting block samples from the pit wall with a stainless steel knife.  The samples were bagged in sealable freezer  bags and were refrigerated  at 4°C within 12 hours of  sampling. Surficial  peat (0-8 cm) was sectioned at 1 cm intervals. Peat below this level was sectioned at 3 cm intervals. Subsampling was performed  wearing non-powdered latex gloves, using an acid-washed plastic knife,  on a plastic-covered lab bench in a clean room. The samples were oven dried at 30°C overnight. Sample material for  elemental analysis was removed from the middle of  each peat slice, reducing the possibility of  analysing material that was contaminated during the sampling procedure. 3.3.2 Soil  Sampling Soils were sampled using a 5 cm diameter probe. Two samples were taken from each site unless otherwise specified:  a surface  sample consisting of  vegetation and the topmost 3 cm of  soil, and a 3-10 cm depth core. The samples were wrapped in separate plastic bags and stored in a freezer  until analysis. Soils from the 3-10 cm sample were homogenized prior to analysis. [49° 15' QUATERNARY •Unconsolidated sediments; till, sand gravel, sift JURASSIC/CRETACEOUS NELSON INTRUSIONS: granodiorite; minor quartz diorite and diorite ROSSLAND MONZONITE: biotite-homblende-augite unite, mainly medium /flrairq ELISE FORMATION: flow breccia, massive a L " t^salts, agglomerate, tuff, ' breccia, laminated siltstone - 3 JUR/ PALEOZOIC MOUNT ROBERTS FORMATION; Is „ bfack fciltstone and argillaceous I v cjaarfate, slate, greywacke, r . ^ctiert, pebble conglomerate, lava; limestone AGE UNKNOWN TRAIL GNEISS: p^flrnpWbdite and grey biotite ITr-jn^i^p, hornblende gneiss, |"y jiica.pdiist, aplite and pegmatite | Peat Sanpling Site Rment Sampling Site 'hi ' 49°03' Kilometres 1 0 D O Figure 3-1. Geological map of  the Trail area showing the location of  the smelter and sampling sites. Modified  from Little (1982). Site Bedrock Geology Vegetation pH and Comments Cominco UTM Zone 11, 0446226, 5440375 Trail Pluton (granodiorite, quartz diorite) Typha  latifolia (common cattail), Epilobium  ciliatum (purple leafed willow herb) Carex  urticulata (beaked sedge), Equisetum palustre, Deschampsia cespitosa  (tufted hairgrass) 5.3 (surface) 5.6 (mineral sediment) 5.7 (pore water) -saturated at the time of sampling, although gleying was noted in the uppermost 15 cm-intermittently aerobic. Zones of oxidation were noted around roots. Thunder Road UTM Zone 11, 0445145, 5440974 Trail Pluton (granodiorite, quartz diorite) Typha  latifolia (common cattail), Pinus  sp. (pine) Betula  sp. (birch) Epilobium  ciliatium Equisetum palustre,  and Deschampsia cespitosa  (tufted hairgrass) 5.0 (surface) 5.4 (mineral sediment) 5.7 (pore water) -watertable was approximately 10 cm below the surface at the time of sampling Champion Lakes Provincial Park UTM Zone 11, 0453623, 5448726 Elise Formation (flow breccia, lava agglomerate, volcanic breccia, tuff, tuffaceous  conglomerate, andesite, basalt, augite porphyry, metamorphosed to greenschist fades). Limestone xenoliths and calcite amygdules found within mafic flows. Economic deposits of Pb, Zn Ag, associated with sulphides. Carex  sp. Nuphar  lutea  (cow my) Assorted grasses 6.5 (surface) 6.5 (15cm depth) 6.4 (pore water) Saturated profile Bombi Summit UTM Zone 11, 0459296, 5454509 Bonnington Piuton (hornblende rich granodiorite, granitic gneiss, amphibolite) Carex  sp. Assorted grasses Organic rich-sediment, not peat Possible urban contamination, due to proximity to road Table 3-1 Description of  the sampling sites. 3.3.3 Elemental  Analysis Elemental analyses were performed  by Becquerel Laboratories, Mississagua, Ontario, Canada, using Instrumental Neutron Activation Analysis (INAA) (As), and Inductively Coupled Plasma Mass Spectrometry (ICP-MS) (Cd, Cu, Pb, Ni, and Zn). Zinc concentrations were obtained using both methods, with consistent results. Analytical precision is variable between elements and is dependent upon concentration and the detection limit. Expected precision for  ICP-MS is estimated to be a minimum of 10% (S. Simpson, pers. com., 2000). The relative standard deviations for  INAA analysis, as calculated with NIST standard 1633, are as follows:  As: 3.1%, Zn: 9.6%. The peat was not fractionated  prior to analysis, as the desired results were for  the bulk sample. Rare earth element concentrations for  peat, soil and the moss monitoring stations were determined by INAA. Reported values from the moss monitoring stations are median values based on eight 3-month sampling periods (May 2000-July 2002). In samples in which the element concentration could not be accurately determined (due to masking by interference  effects  when the element was present only in low concentration), the value is listed as being less than a given value. For graphing purposes, the concentration was assigned a value of lA the stated "less than value". In all cases, chondrite standards were used for  nomalisation. REE concentrations for  the smelter samples were determined by ICP-MS, following  dissolution by microwave digestion with triple acid solution ( H N O 3 , HF and HC1, in a 5:4:1 ratio) and treatment with H B O 3 to dissolve precipitated REE-F complexes (Coedo et al., 1998; Prohaska et al., 1999; Krachler et al., 2002). 3.4 Trace Element/Conservative Element Concentrations: A Viable Normalisation Parameter? 3.4.1 Introduction Many researchers normalise trace element data from peat and other sediments to published global elemental abundance data (e.g. Shotyk et al., 1997, and other articles within this volume) in order to determine the extent to which trace elements are enriched within the sampled sediment relative to background concentrations. Normalisation ratios utilise a "conservative" element, which is assumed to be derived exclusively from background sources and persistent in the environment, and as such, can be used as a proxy for  geogenic input when compared to trace element concentrations. A ratio of  the trace element versus the conservative element is then normalised to a benchmark element ratio, commonly global crustal elemental ratios from geochemical data compilations (e.g. Wedepohl, 1995). Shotyk et al. (1996) employed the following  equation in their study of variations in trace element concentrations in an ombrotropic peat bog: Enrichment  Factor  ={[Trace  Element] pea,/[Sc] peat}/{[Trace  Element]  crust /[S  c] crust} Using this approach in combination with Pb-210 dating, fluctuations  in trace element concentrations were attributed to historical activities, such as Roman Pb smelting (Shotyk et al., 1996). There are many differences  between the conditions in the Shotyk et al. (1996) study and this one, making several modifications  to the proceedings and underlying assumptions necessary. Much of  the material deposited from the smelter is likely to be in a form that is readily soluble in the acidic conditions within the peat, particularly oxides of  Pb and Zn from the roasting process. These ions are likely to be mobilised throughout the peat. Zones of  trace element enrichment exist within the peat as a probable result of redox-controlled precipitation/dissolution reactions, and soils are relatively depleted in trace elements relative to nearby peats, indicating migration and subsequent sequestration within the peat (Chapter 2). The examined peats are composed of  vascular plants which are capable of  taking up and concentrating metals, making a chronostratigraphic interpretation invalid, hence Pb-210 dating was not employed. Furthermore, the sampled sites are clearly not ombrotrophic, and so it must be concluded that the peat chemistry is impacted by both the nature of  the underlying material and the groundwater geochemistry, and by atmospheric deposition. 3.4.1.1 Selection  of  a geochemical background  value Finding a viable geochemical background to determine the extent of  trace element enrichment is problematic. The use of  published global average geochemical values was rejected, as it reflects  neither the geochemistry of  the individual sampling locations, nor the variability in bedrock geochemistry that exists in the sampling area (e.g. Darnley, 1995; Reimann and DeCaritat, 2000). The use of  published geochemical data for  the Nelson Intrusives and the Elise Formation was considered to be more relevant, although even this data may not accurately reflect  the geochemistry of  the bedrock at the sampling locations, as both suites of  rock are heterogeneous. Furthermore, a literature search was unable to locate geochemical data for  all the elements of  interest. Mineral soil data from each of  the sampling sites reflects  the potential geogenic component of  the peat chemistry more closely, i.e. the geochemistry of  the bedrock immediately below the sampling site after  weathering. However, the mineral soil chemistry is also impacted by the smelter activities, although it does not accumulate metals to the same extent as the peats (Chapter 2). Soil from a "background" site, (i.e. a location assumed not to be impacted by the smelter) had elevated concentrations of  As, due either to local anthropogenic activities or to localised natural geochemical enrichment (Goodarzi, unpublished data), and was thus deemed to be inappropriate. Despite the potential drawbacks, published bedrock geochemical data (Beddoe-Stephens and Lambert, 1981; Hoy and Andrew, 1989; Sevigny, 1990) and mineral soil data from  samples 3-10 cm in depth from locations nearby each of  the sampling sites were used for  normalisation. 3.4.1.2 Selection  of  conservative elements Elements commonly used for  normalisation include Ti, Sc, Zr and Al (e.g. Shotyk et al., 1997; Espi et al., 1997). However, Al is subject to biogeochemical cycling in peats composed of  vascular plants (e.g. Wust et al., 2002) and thus cannot be considered to be an appropriate proxy for  background geochemistry. Scandium was found  to correlate poorly with the peat ash content (r = 0.67) and yielded erratic results when used as a conservative element, and so was rejected in favour  of  Ti and Zr. Titanium concentrations in the smelter materials were measured to be less than 0.05% in a variety of  samples taken from concentrates, slags and wastes from the smelter (Goodarzi, unpublished data). Microscopic investigations have indicated that much of  the titanium within the peat is associated with geogenic particles (Chapter 4). The Zr concentrations of  the smelter materials are unknown, but Zr is used as a conservative element in a wide range of  geochemical studies due to its persistence in the environment, and as such may be considered suitable. The following  equations were used: Soil Enrichment Factor = {[Metal]peat /[C.E.]Peat}/{[Metal]SOii]/[C.E.]SOii}; Bedrock Enrichment Factor = {[Metal]peat /[C.E.]peat}/{[Metal]r0ck]/[C.E.]rock}; Where  C.E. = Conservative Element (Ti or Zr) Problems associated with the use of  "Enrichment Factors" for  environmental studies are described in Reimann and DeCaritat (2000). These authors discuss the role of naturally low conservative element concentrations in the sampled material in creating very high EFs, which are subsequently cited as evidence of  severe anthropogenic impact. Both the ash content and the mineralogy (Table 3-2) are known to vary between sampling sites and indeed between samples within the same profile.  For this reason, the magnitude of  EFs alone can not be used to interpret the extent of  anthropogenic impact. Nonetheless, comparing trends in trace element concentrations with EFs may help to distinguish between anthropogenic and geogenically-sourced trace elements. 3.4.2 Results Bulk elemental concentrations are presented in Table 3-3 and Fig. 3-2. Soil element ratios are presented in Table 3-4, and the element ratios of  peat normalised to soil are presented in Table 3-5. Table 3-6 lists elemental concentration data for  the Trail Pluton granodiorite and the Elise Formation K-andesite, from Sevigny (1990) and Beddoe-Stephens and Lambert (1981), respectively. Peat elemental data normalised to the bedrock concentrations are listed in Table 3-7. The bulk concentration data and normalised data are plotted on Figs. 3-3 to 3-6. 3.4.2.1 Cominco Profile The enrichment patterns follow  the overall shape of  the concentration profile  for  most elements in the Cominco profile,  suggesting a non-geogenic source (Fig. 3-3). A geogenic source would theoretically be indicated by a smoothed concentration profile with an EF value of  approximately 1. The trace element concentration would vary as a function  of  the total geogenic input, assuming no preferential  concentration of  geogenic Table 3-2. Mineralogy of  the peat profiles  as determined by X-ray diffraction. Methodology is described in Chapter 4. Sample Depth Ash % Quartz Feldspar Plagioclase Orthoclase Mica Phylosilicates Clay Chlorite Amphi-bole Pyroxene Calcite Anhydrite Cominco Site 1-2cm 51.7 66 16 13 3 0 0 2 0 0 0 2-3cm 63.2 34 10 6 2 0 2 1 46 0 0 3-4cm 62.4 38 35 9 3 0 1 2 12 0 0 4-5cm 66.2 60 21 11 5 0 2 1 0 0 0 5-6cm 64 20 11 3 0 1 1 0 0 0 7-8cm 73.4 40 24 31 2 0 1 1 0 0 0 8-11cm 74.0 48 20 26 2 0 2 0 0 0 11-14cm 81.2 37 26 30 2 0 1 0 0 0 0 14-17cm 79.8 66 25 6 1 0 1 0 0 0 17-20cm 51.4 67 23. 0 4 0 1 6 0 0 0 Thunder Road Site 2-3cm 18.4 62 20 10 2 0 1 6 0 0 0 5-6cm 24.4 53 22 15 5 0 2 3 0 0 0 6-7cm 25.9 59 11 6 6 0 14 2 0 0 2 8-11cm 27.8 50 16 7 7 0 0 2 0 0 19 11-14cm 25.4 53 17 22 5 0 0 4 0 0 0 Champion Lakes Site 3-4cm 38.6 72 14 0 0 2 2 10 0 0 0 6-7cm 21.2 47 34 0 0 2 2 15 0 0 34 8-11cm 16.6 49 33 0 0 5 1 1 0 13 0 11-16cm 12.9 48 28 0 0 6 0 7 0 38 0 Bombi Summit Site 0-1 cm 35.9 50 25 20 1 0 2 2 0 0 0 1-2cm 46.8 47 40 7 1 0 1 3 0 0 0 2-3cm 56.7 43 28 23 1 0 1 4 0 0 0 4-5cm 71.6 51 27 11 3 0 1 7 0 0 0 5-6cm 91.1 38 42 14 1 0 1 5 0 0 0 6-7.5cm 94.9 40 33 20 2 0 1 4 0 0 0 7.5-9cm ' 91.0 47 31 13 2 0 2 5 0 0 0 trace elements within particular fractions  of  the peat had taken place. Arsenic, Sb, Cd, Cu, Pb and Zn show considerable enrichment in the peat relative to the mineral soil, with enrichment factors  ranging from EF-n of  3.7 to 18.1, and EFzr of  2.7 to 11.4 (Table 3-5). Bedrock geochemistry enrichment factors  for  Pb range from EFTj 877-1598 and EFzr 231-327. Zinc bedrock EFTi values range from 152-523, and 35-97 for  EFZr (Table 3-7). Soil normalised enrichment factors  for  Ni range between 1 and 1.5 (EF-n) and 0.7-1.2 (EFZr), indicating that the Ni in the profile  is probably geogenic. When normalised to bedrock geochemistry, the EFs range from 9-13 (Ti), and 36-97 (Zr), indicating either that the published geochemical data does not reflect  the geochemistry of  the sampling site, or that Ni has been concentrated in the peat and soil due to weathering. 3.5.2.2 Thunder  Road  Profile Enrichment factors,  particularly EF-n are much higher in the Thunder Road profile than in the Cominco profile,  despite the lower trace element concentrations and the lower rates of  trace element deposition at the Thunder Road site (Goodarzi et al., 2001; 2003). The higher EFs can be attributed to the lower ash concentrations found  at the Thunder Road site, hence the lower concentrations of  Ti and Zr, and the higher apparent trace element enrichment. This is particularly notable for  the high enrichment relative to Ti measured in the 0-1 cm sample, which reflects  the low concentration of  Ti (0.03%), rather than the extreme concentration of  the trace elements. Even Ni has an apparent enrichment factor  of  up to 5 times in the peat when Ti is used for  normalisation, although that EF is reduced to a maximum of  2 for  Zr normalisation. 3.5.2.3 Champion Lakes Profile Patterns of  enrichment at the surface  and base of  the Champion Lakes profile  can be attributed in part to fluctuations  in the conservative element concentration, rather than to variations in trace element concentration. This profile  is located 13.5 km from the smelter and is sheltered by the physiography of  the region, and is thus subject to deposition from the smelter to only to a small degree, as has been demonstrated by the moss monitoring study of  Goodarzi et al. (2001; 2003). The relatively high concentrations of  Sb, As, Cu, and Pb in the lowermost sample of  the profile  were ascribed to a geogenic source due to the lack of  evidence of  an anthropogenic source. The basal peats have higher EFs than the overlying peats, due to the relatively low concentrations of Zr and Ti. It is possible that these trace elements in the basal peat are from atmospherically-deposited anthropogenic sources, and have migrated to deeper layers of the peat, but it is also possible that these elements are present in the peat due to weathering of  the bedrock. Sample Depth (cm) Sb ppm As ppm Cd ppm Cu ppm Pb ppm Ni ppm Zn ppm Ti % Zr ppm COM NCO 0-1 126 78.5 61.7 196 3550 10.8 5910 0.10 120 1-2 141 97.8 65.2 223 3830 11.6 5600 0.12 130 2-3 128 153 53.8 215 3780 15.4 4230 0.17 120 3-4 142 116 60.3 227 3830 13.2 4730 0.15 120 4-5 121 164 47.1 214 3940 14.8 3570 0.20 170 17-20 85.0 247 60.1 226 5240 12.4 3040 0.16 160 -HUNDER ROAD 0-1 34.1 60.0 119 82 1670 9.2 2730 0.03 83 1-2 37.7 64.4 105 103 2070 10.6 2270 0.05 81 2-3 33.0 62.0 139 84 1750 9.4 2900 0.05 88 3-4 37.6 65.4 134 96 1980 11 2590 0.05 84 5-6 32.5 67.5 104 92 1990 10.8 2330 0.06 73 11-14 34.9 59.0 120 96 1940 11 2300 0.06 77 c HAMPIC IN LAKES 0-1 1.2 1.8 2.4 12 77 3 148 0.01 44 1-2 1.6 3.6 3.7 16 101.5 6 194 0.02 83 2-3 5.5 8.3 6.0 14 172 9.2 214 0.11 97 3-4 3.8 6.8 4.4 13 114.5 6 208 0.04 57 4-5 9.0 15.4 6.8 20 246 8.8 268 0.11 100 11-16 15.2 15.8 6.5 24 324 4.4 178 0.03 53 BOMBI SUMMIT 0-1 2.2 2.9 2.1 14 79.5 12.8 130 0.17 120 1-2 3.0 7.8 2.1 15 104 10.2 122 0.21 130 2-3 2.8 22.6 2.1 13 108.5 11.2 116 0.22 98 3-4 5.0 7.8 3.3 18 191 11.4 150 0.24 130 4-5 8.8 9.8 7.9 29 476 10.8 248 0.18 120 7.5-9 1.3 5.2 0.8 6 35 12.4 76 0.27 150 Table 3-3. Concentration of  selected elements in the peat profiles. Cominco Site Concentration (ppm) K> BSed20-Concentration (ppm) 10 15 BSed 20 r. Thunder Road Site Concentration (ppm) Concentration (ppm) Champion Lakes Site Concentration (ppm) Concentration (ppm) Bombi Summit Site -h-Sb As -EhCd - e - c u —0— Ni Concentration (ppm) IV) CO BSed Pb - • - Z n Concentration (ppm) M CO CJ1 O O O O O o o o o o o BSed t f* > • Figure 3-2. Concentration of  trace elements in the peat profiles.  BSed = basal sediment. N i— s? E Z QJ a I a- s « i i , o a N. CM CO x— CM CM CN o o o o CO oo CM CO CO CO t T— co T— T-(A < JQ E W & Q. o o o o CM C» O CM r t - N t -n rf  co s 1- T" T" CM CO T-^ o CO i- -si-t- in CM oo CO CM CO t-CM <J> 00 00 CD O O O ^ x- « UD a> co •>- co to © ^ T3 •-ro ra E o —' p a: c § Q E C CO E u j . r o o m m o O 0 c -a II J— lt> CO CO LO •! o o o N ° o o o CO t- O P O Q ^ O o o V o LO CO CM o o o d d o o CM o o o o d d d o «4> co CO >1 o o o o o o o o o o d d o d CO •t CO Q o o o o o o d d d d CT> CM ^— o o o o o o o o d o o d to © ^ •a — co ro E O —' p n O u O m .2 o a) i l l i i ._ o SZ SI  O I- O I- O CO N o r^  o c ^ d d ^ NJ •<—•<—'<— CM = 0 0 0 0 t! en w cm in p c\j d d o Q. y CO t- CM CM 3 0 0 0 0 fcj  g o o o T3 • O O O O ° b o d • o o o ^ ° d d o N - S 5 S £ ° d d d w © ^ •a ^ — CD cd E O —' c: o ® -5. CO E § g E O JZ  £Z O OHOm Table 3-4. Element concentrations in mineral soils (3-10 cm depth) corresponding to the peat sampling sites, and calculated element ratios. Table 3-5. Element ratios normalized to mineral soil for  Ti and Zr. Sample Depth (cm) Sb/Ti As/Ti Cd/Ti Cu/Ti Pb/Ti Ni/Ti Zn/Ti COMINCO 0-1 1-2 2-3 3-4 4-5 17-20 15.4 14.6 9.2 11.1 7.3 6.4 7.3 7.7 8.4 6.9 7.6 14.2 17.3 15.4 8.9 10.8 6.5 10.4 11.1 10.6 7.1 8.1 5.9 7.8 18.1 16.5 11.4 12.5 9.9 16.4 1.5 1.3 1.2 1.2 1.0 1.0 12.9 10.0 5.5 6.7 3.7 3.9 THUNDER ROAD 0-1 1-2 2-3 3-4 5-6 11-14 82.7 47.5 48.0 45.6 33.8 37.5 80.3 44.7 49.8 43.8 38.7 34.9 1049.5 480.6 738.6 591.2 392.9 470.4 26.5 17.3 16.3 15.5 12.8 13:7 134.7 86.7 84.7 79.9 68.8 69.3 5.4 3.2 3.3 3.2 2.7 2.8 160.0 70.5 99.6 80.0 56.7 59.6 CHAMPION LAKES 0-1 1-2 2-3 3-4 4-5 11-16 11.6 5.9 3.7 6.1 5.8 32.9 7.4 5.8 2.5 4.9 4.3 15.1 92.4 54.2 16.4 28.7 17.6 57.0 21.5 10.9 1.8 4.0 2.4 9.8 38.3 19.2 6.1 9.7 8.2 36.7 3.5 2.7 0.8 1.2 0.7 1.2 21.4 9.2 4.0 9.4 4.7 10.6 BOMBI SUMMIT 0-1 1-2 2-3 3-4 4-5 7.5-9 0.9 1.0 1.0 1.5 3.6 0.4 0.5 1.2 3.4 1.1 1.8 0.6 3.7 2.9 2.9 4.1 13.0 0.9 1.1 1.0 0.8 1.0 2.2 0.3 1.8 1.9 1.9 3.1 10.1 0.5 0.7 0.4 0.5 0.4 0.5 0.4 1.6 1.2 1.0 1.1 2.8 0.6 Sb/Zr As/Zr Cd/Zr Cu/Zr Pb/Zr Ni/Zr Zn/Zr COMINCO 0-1 1-2 2-3 3-4 4-5 17-20 8.8 9.1 9.0 9.9 6.0 4.5 4.2 4.8 8.2 6.2 6.2 9.9 9.9 9.6 8.6 9.7 5.3 7.2 6.3 6.6 6.9 7.3 4.8 5.4 10.3 '10.3 11.0 11.2 8.1 11.5 0.8 0.8 1.2 1.0 0.8 0.7 7.4 6.3 5.4 6.0 3.1 2.7 THUNDER ROAD 0-1 1-2 2-3 3-4 5-6 11-14 23.2 26.3 21.2 25.3 25.2 25.6 22.6 24.8 22.0 24.3 28.8 23.9 294.8 266.4 326.2 328.2 292.8 321.8 7.4 9.6 7.2 8.6 9.5 9.4 37.8 48.1 37.4 44.3 51.3 47.4 1.5 1.8 1.4 1.8 2.0 1.9 45.0 39.1 44.0 44.4 42.2 40.8 CHAMPION LAKES 0-1 1-2 2-3 3-4 4-5 11-16 1.0 0.7 2.0 2.4 3.2 10.4 0.6 0.7 1.4 1.9 2.4 4.7 8.0 6.5 9.0 11.3 10.0 18.0 1.9 1.3 1.0 1.6 1.4 3.1 3.3 2.3 3.4 3.8 4.6 11.6 0.3 0.3 0.4 0.5 0.4 0.4 1.8 1.1 2.2 3.7 2.7 3.4 BOMBI SUMMIT 0-1 1-2 2-3 3-4 4-5 7.5-9 0.6 0.8 1.0 1.4 2.6 0.3 0.4 1.0 3.7 1.0 1.3 0.6 2.6 2.4 3.1 3.7 9.6 0.8 0.8 0.8 0.9 0.9 1.6 0.3 1.3 1.5 2.1 2.8 7.5 0.4 0.5 0.4 0.5 0.4 0.4 0.4 1.1 0.9 1.1 1.0 2.1 0.5 Trail Pluton Elise Fm Granodiorite K-Andesite Element Concentrations Ti02 % 0.57 0.77 Zr ppm 130.0 114.8 Ni ppm 5.0 6.75 Zn ppm 69.0 90.5 Pb ppm 13.0 N/A Cu ppm N/A 37.75 Ti Ratios Ni 0.001 0.001 Zn 0.01 0.01 Pb 0.002 N/A Cu N/A 0.005 Zr Ratios Ni 0.04 0.06 Zn 0.5 0.8 Pb 0.1 N/A Cu N/A 0.3 Table 3-6. Selected published geochemical data and the resulting caluculated conservative element ratios from the bedrock geology of  the Trail Pluton (Sevigny, 1990) and the Elise Formation. (Beddoe-Stephens and Lambert, 1981) Sample Depth (cm) 0 - 1 1 - 2 2-3 3-4 4-5 17-20 O O O O Cu/Zr N/A N/A N/A N/A N/A N/A Pb/Zr 296 295 315 320 232 328 Ni/Zr 2.3 2.3 3.3 2.9 2.3 2.0 Zn/Zr 97 82 '70 79 40 36 Cu/Ti N/A N/A N/A N/A N/A N/A Pb/Ti 1598 1460 1005 1105 877 1454 Ni/Ti 13 12 11 10 9 9 Zn/Ti 523 407 224 273 152 158 §9 0 - 1 1-2 2-3 3-4 5-6 11-14 N/A N/A N/A N/A N/A N/A 201 256 199 236 273 252 2.9 3.4 2.8 3.4 3.8 3.7 62 54 60 61 58 56 N/A N/A N/A N/A N/A N/A 2712 1745 1705 1608 1385 1394 39 23 24 23 20 21 829 365 516 415 294 309 Q co Q_ LU < 3 X -1 o 0 - 1 1-2 2-3 3-4 4-5 11-16 0.8 0.6 0.4 0.7 0.6 1.4 18 12 18 20 25 61 1.2 1.9 2.5 2.7 2.3 2.2 2.3 1.4 2.7 4.5 3.3 4.1 0.5 0.2 0.04 0.09 0.05 0.2 N/A N/A N/A N/A N/A N/A 45 34 10 15 9 15 88 38 16 39 20 44 CQ O m D cn 0 -1 1-2 2-3 3-4 4-5 7.5-9 N/A N/A N/A N/A N/A N/A 6.6 8.0 11 15 40 2.3 2.8 2.0 3.0 2.3 2.3 2.0 2.0 1.7 2.0 1.9 3.8 0.9 N/A N/A N/A N/A N/A N/A 21 22 22 36 117 5.8 8.7 5.5 5.9 5.5 6.9 5.3 6.3 4.6 4.0 4.5 11.3 2.2 Table 3-7. Element ratios normalised to bedrock geochemical data listed in Table 3-6. Figure 3-3. Normalised element ratios using soil and rock for  the Cominco profile.  Bulk concentration data is plotted on the primary (top) axis, while element ratios are plotted on the secondary (bottom) axis. Sb As Cd Cu Pb Ni Zn Os E o CL 0) Q 0) Q. E CO co 17-20 K> 00 O O O o o o o o cn o Ol Concentrat ion (ppm) N> N3 00 O NJ O O O ro ^ O) o o o o o o o o o J—l . O ui - i - i N) o - ^ r o c o o cn o o o o o o o o cn o cn O M O) o o o o o o Ratio [Me]/[Zr] rock ...A... [Me]/[Zr] soil ~s-~[Me]/[Ti] rock [Me]/[Ti] soil Figure 3-4. Normalised element ratios using soil and rock for  the Thunder Road profile. Bulk concentration data is plotted on the primary (top) axis, while element ratios are plotted on the secondary (bottom) axis. Sb As Cd Cu Concentrat ion (ppm) o o o ° Ratio [Me]/[Zr] rock [Me]/[Zr] soil [Me]/[Ti] rock [Me]/[Ti] soil Pb Ni Zn Figure 3-5. Normalised element ratios using soil and rock for  the Champion Lakes profile.  Bulk concentration data is plotted on the primary (top) axis, while element ratios are plotted on the secondary (bottom) axis. Sb As C\ oo Cd D -t^  oo • • *f< • • i Cu tration i - [Me ] Pb Ni Zn -A M W CO o o o o ° o o o w u o o o O M ^ O I W O O O O O O W O) o o o Ratio [Me]/[Zr] rock ....... [Me]/[Zr] soil [Me]/[Ti] rock - [Me]/[Ti] soil Figure 3-6. Normalised element ratios using soil and rock for  the Bombi Summit profile. Bulk concentration data is plotted on the primary (top) axis, while element ratios are plotted on the secondary (bottom) axis. Sb As Cd Cu Pb Concentration (ppm) —»-[Me] Ni Zn o O -Pi- CO - i N3 CO O O O O O O O Ratio [Me]/[Zr] rock [Me]/[Ti] rock a - [Me]/[Zr] soil - [Me]/[Ti] soil a. 10000 ?1000 Q. Q. I 100 E a) W 10 o Zr R2 = 0.67 • Ti R2 = 0.77 1 0 50 100 Ash (%) b. 24 20 16 rsj 12 i -8 4 0 COMINCO THUNDER CHAMPION BOMBI Figure 3-7 a. Plot of  the concentrations of  Ti and Zr versus the peat ash contents, b. Whisker-Box diagram showing the spread of  Ti/Zr ratios for  each profile,  where the median value is represented by the dot in the middle of  the box, the upper and lower extent of  the box represent the 75th and 25th percentile, respectively, and the lines extending from the box represent the range of  data. 3.4.2.4 Bombi Summit  Profile Only in the 4-5 cm sample is the Bombi Summit profile  significantly  enriched in trace elements relative to the mineral soil. The peat at this depth has high concentrations of  all the examined trace elements relative to the rest of  the profile  with the exception of As, which is enriched in the 2-3 cm sample. The Bombi Summit site is located approximately 16.5 km radial distance from the smelter and is believed to be minimally impacted by the smelter, although it may be subject to anthropogenic impact from secondary sources such as the nearby roadway. At this location, the soil-normalised data closely follows  the bulk trace elemental data for  all elements except Cu, which has a smoothed profile.  This suggests that Cu may have a geogenic source at this site. Lead, Zn and Ni are highly enriched relative to the bedrock (up to 117 times for  Pb) when Ti is used for  normalization, but show enrichments ranging from less than 1 to a maximum of 40 times (for  Pb in the 4-5 cm sample) when Zr is used. This suggests that the published Zr value is closer to reflecting  the true background geochemistry at this location than the Ti value. 3.4.3 Discussion These high enrichment factors  calculated in the Cominco and Thunder Road peats are due in part to the impact of  the smelter at these location, but are also due to the lower Ti concentrations in the peats relative to the bedrock (Tables 3-3 and 3-6) and the tendency for  trace elements to be relatively enriched within weathered materials and dust (e.g. Riemann and DeCaritat, 2000). The higher relative Zr concentration in the peats help to negate this effect.  These discrepancies in the enrichment factors,  depending on the conservative element used for  normalisation, illustrate one of  the potential shortcomings of  this method. The trace elemental and conservative elemental composition of  the geogenic component of  the samples are almost certainly variable, as the mineralogy differs  between samples (Table 3-2). Concentrations of  the conservative elements fluctuate  independently of  the ash content, which can be attributed to both the variable nature of  the mineralogy within the peat profiles  and to the potential for phytogenic mineral material to contribute significantly  to the ash content of  the peat. Figure 3-7 illustrates this point—the Ti/Zr ratio is variable, (standard deviation of  5.2, average 10.4, coefficient  of  variation = 0.503, n=36), rendering questionable the assumption that the conservative elements are a valid proxy for  geogenic input. In studies in which trace elemental input is low, this would not be as readily apparent, but when dealing with high concentrations, the discrepancies become more pronounced. Given that the enrichment patterns are controlled by fluctuations  of  the concentrations in conservative elements relative to the ash content as well as the concentrations in trace elements, enrichment factors  are of  limited use for  determining the extent of  anthropogenic versus geogenic impact in the examined sites. This is particularly true for  the heavily impacted sites, where the signature of  geogenic trace elements is largely masked by the anthropogenic trace elements. Since mineralogy is variable between samples, the ratio of  geogenically-sourced trace elements to conservative elements may be variable as well. The fact  that the ratio of  Ti to Zr is itself variable results in discrepancies in EF, making interpretation difficult.  The environmental significance  of  enrichment factors,  called into question by Riemann and DeCaritat (2000) is further  illustrated here: the so-called trace element enrichment may be observed as a result of  low concentrations of  conservative elements. Since there are so many factors  controlling the concentration and relative enrichment of  trace elements within each of  the peat profiles,  it is difficult  to ascribe an anthropogenic or geogenic source to the trace metals by this method. 3.5 Rare Earth Element Enrichment Patterns 3.5.1 Introduction Rare earth elements have been used to provide a geochemical fingerprint  for sedimentary rocks in order to determine provenance. The REE signature is characteristic of  the parent rock and although it may be modified  by weathering, REE behaviour is considered to be relatively cohesive (e.g. Leybourne et al., 2000). The Lanthanide series of  elements behaves in a fairly  uniform manner owing to the fact  that each element has a +3 charge, although the decreasing ionic radius (from left  to right on the periodic table) causes some disparity in behaviour, resulting in fractionation  between the light (LREE) and heavy (HREE) rare earth elements. The behaviour of  Eu and Ce departs from  that of the rest of  the series. Ce+3 may be rapidly oxidised to Ce+4, which has a higher surface activity than Ce+3 and the other REEs, and is thus preferentially  removed from solution, resulting in a negative Ce anomaly under oxidising conditions. Eu+3 may be reduced to Eu+2, resulting in the formation  of  a negative Eu anomaly. The relative global abundance of  individual elements within the lanthanide series are highly variable between odd and even atomic numbers, and when a plot of  REE concentration versus atomic number is graphed, a sawtooth pattern results. When discussing fractionation  trends in the lanthanide series, REE concentrations are typically normalised to a chondrite or shale standard, resulting in a smoothed line. A high or low concentration of  a single element relative to the rest of  the series results in a positive or negative anomaly when normalised to a standard. 3.5.1.1 REE behaviour in the environment The normalised REE pattern for  plant tissues typically has a negative Ce anomaly (Akagi et al., 2002), but this anomaly was absent from peats sampled in Japan (Akagi et al., 2002), and from other peat REE data observed in the literature (Krachler et al., 2002). The absence of  a Ce anomaly in peats is attributed to the low Eh-pH conditions found within peat-forming  environments (Akagi et al., 2002). A negative Eu anomaly was noted in Carex  and Sphagnum  peats from the surface  and from 40 cm depth (Akagi et al., 2002). Peats sampled from the Holland Marsh in southern Ontario and from the Schopfenwaldmoor,  near Interlaken, Switzerland, were both found  to have slight negative Eu anomalies when normalised to chondrites (Krachler et al., 2002). Chiarnezelli et al. (2001) found  that REE compositions associated with lichens and mosses had little relationship to the underlying mineral substrate, and that the REE composition of  these plants was derived from atmospheric sources. They (Chiarnezelli et al., 2001) found  that the REE composition of  vascular plants was variable between species, and that the REE concentrations were partially dependent on the individual species' capacity to take up and accumulate REEs. A review of  the literature reveals disparities in the observed behaviour of  REE during weathering. In the presence of  humic substances, Takahashi et al. (1999) determined a strong tendency for  REEs to form organic complexes. The potential for REE to sorb to organic matter was determined experimentally by Eskenazy (1999), who observed that REEs sorb to organic matter at pH > 3, with no tendency for  fractionation between HREEs and LREEs. Land et al. (1999), on the other hand, noted fractionation between LREEs and HREEs with depth in a soil profile.  Astrom (2001) observed that concentrations of  La were depleted at the top of  a sulphate-rich soil profile,  highly enriched in the transitional zone between oxic and anoxic conditions, and of  median concentrations in the anoxic zone. Astrom attributed the observed pattern to the mobilisation of  La at the soil surface,  and the subsequent preferential  scavenging by Fe/Mn oxy-hydroxides at depth (compared to HREE). Middle Rare Earth Element (MREE) enrichments in river waters have been attributed to weathering of  phosphate minerals (Hannigan and Shokovitz, 2001), while MREE enrichments in lake waters were attributed to weathering of  Fe-Mn oxides/oxyhydroxides (Johannesson and Zhou, 1999). In a discussion of  REE weathering behaviour, Johanneson and Zhou (1999) state that HREEs become depleted in weathered sediments, due to complexing reactions with the dissolved organic and inorganic ligands in the weathering solution. Iron-oxide precipitates and carbonate ligands are known to sequester LREEs preferentially  to HREEs (Dupre et al., 1996; Leybourne et al., 2000), resulting in negative LREE anomalies in waters and positive LREE anomalies in the Fe and carbonate precipitates. Bau (1999) found  that La and possibly Lu had a lower affinity  for  Fe-oxides than other REEs, while Ce+4 has a higher Fe-oxide affinity. 3.5.1.2 REEs as Geochemical Tracers The REE composition of  lake sediments and lake water may strongly reflect  the underlying geology (Hall et al., 1995). The REE composition of  groundwater may reflect the REE composition of  the host bedrock, if  the groundwater is not subject to oxidation (which results in negative Ce anomalies) or complexing reactions which remove dissolved REEs from solution (Leybourne et al., 2000). Deep groundwater most closely reflects  the host bedrock geology under low pH conditions, as Fe-oxide precipitation reactions occurring under progressively increasing pH conditions subsequently remove REEs from solution (Leybourne et al., 2000). High concentrations of  REEs were found  in stream waters in an area of  Cu-Pb-Zn mining (Protano and Riccobono, 2002). The normalised REE patterns in the stream waters were found  to reflect  the REE chemistry of  leachate from the mine tailings. However, the REE concentrations decreased rapidly downstream due to dilution and precipitation, and the characteristic signature from the leachates was progressively lost (Protano and Riccobono, 2002). The results of  previous studies indicate that, while REEs are potentially useful geochemical tracers, they are subject to fractionation  in the environment, which may result in the characteristic signature being rendered unrecognisable. The concentrations and chondrite normalised patterns of  REEs from smelter wastes and feedstocks  are compared to those of  peats, soils, atmospheric monitoring media and stream sediments from the surrounding area, in order to determine the effectiveness  of  using REEs as tracers to measure the impact of  the smelter. 3.5.2 Results 3.5.2.1 REE concentrations  of  smelter  materials Rare earth element concentrations were determined for  a selection of  samples of smelter feedstock  and wastes (Table 3-8). The La concentration in the Zn concentrate (feedstock)  is 11.8 ppm. Lanthanum concentrations in the slags range from 67.6-73.1 ppm, while the La concentration in the calcine residue from the Indium plant is 17.6 ppm. The high concentrations of  REEs found  within the smelter wastes relative to the feedstock  indicate that REEs are concentrated during processing. La/Lu ranged from 0.48-2.6, and is lowest for  the Zn concentrate sample. There is a positive Eu anomaly in both the feedstock  and the slags and wastes (Fig. 3-8). 3.5.2.2 Moss  Monitoring  Station  Data Moss monitoring stations closest to the smelter typically have the highest concentrations of  REEs (Fig. 3-9; Table 3-9). The patterns from all stations indicate that the REEs collected by the moss are LREE enriched (La/Lu of  up to 11.6, from the concentrate unloading site). However, the Cominco station has a relatively low La/Lu of 4.5. No Ce or Eu anomalies were noted in the moss sampling media data. A positive Nd anomaly is present from the monitoring sites closest to the smelter and the ore concentrate site, although no Nd anomoly was detected within the smelter materials. 3.5.2.3 Peats REE concentrations were highest in the Cominco peats, with La concentrations ranging from 15.2-27.3 ppm (Table 3-10). Concentrations of  La decreased between sampling profiles  in the order Cominco> Bombi Summit> Thunder Road> Champion Lakes. For all profiles,  the concentrations of  REEs generally showed an increase with increasing peat depth, consistent with the findings  of  Ohlander et al. (1996) for  till . La Ce Pr Nd Sm Eu Dy Ho Er Yb Lu La/Lu Zn Concentrate Off  grade slag High Pb slag Indium Plant calcine residue 11.8 13.4 4.9 8.7 4.4 2.2 4.5 3.1 3.4 4.3 2.5 0.5 67.6 114.4 16.9 51.4 12.5 8.2 14.0 4.7 7.6 7.5 3.0 2.4 73.1 128.9 18.0 54.6 13.6 8.8 14.2 4.8 8.2 8.2 3.1 2.4 17.6 29.9 4.2 12.4 2.9 2.0 3.5 1.1 2.1 2.2 0.7 2.6 Table 3-8. Concentration of  REE's and normalized La/Lu ratios from selected smelter feedstock  and waste samples. 1000 100 10 1 Figure 3-8. Chondrite nomalised REE pattern for  selected smelter feedstock  and waste samples. La Ce Pr Nd Sm Eu Dy Ho Er Yb Lu —x— Zn Concentrate —fa— Off  grade slag —a— Hi.qh Pb slag —®— Indium Plant calcine residue La Ce Nd Sm Eu Tb Yb Lu Norm. ppm ppm ppm ppm ppm ppm ppm ppm La/Lu Mineral Soil Ore Concentrate site 0-3 cm 44 80 29 4.9 1.1 0.39 1.7 0.25 18.2 Ore Concentrate site 3-10 cm 41 73 25 4.8 1.1 0.61 1.6 0.25 16.8 Cominco 0-3 cm 34 67 24 4.4 1.0 0.53 1.7 0.26 13.8 Cominco 3-10 cm 37 67 23 4.1 1.0 0.34 1.6 0.23 16.9 Thunder Road 0-3 cm 41 83 28 4.9 1.1 0.56 1.8 0.29 14.4 Thunder Road 3-10 cm 38 76 26 5.1 1.1 0.51 1.8 0.27 14.8 Champion Lakes 0-3 cm 49 88 33 6.3 1.3 0.58 2.1 0.31 16.5 Champion Lakes 3-10 cm 30 56 21 4.2 1.2 0.37 1.8 0.26 12.0 Bombi Summit 0-3 cm 27 51 19 3.8 1.1 0.47 1.7 0.27 10.3 Bombi Summit 3-10 cm 41 74 28 5.5 1.2 0.55 1.7 0.26 16.2 Moss Monitoring Stations Ore Concentrate site 0.83 1.45 2.60 0.10 0.02 0.04 0.10 0.01 11.6 Cominco 1.22 1.60 0.65 0.18 0.07 0.16 0.64 0.03 4.5 Thunder Road 0.66 1.20 0.59 0.10 0.02 0.07 0.09 0.01 12.4 Champion Lakes 0.35 0.72 0.23 0.05 0.01 0.02 0.02 0.004 10.2 Bombi Summit 0.60 0.64 0.32 0.07 0.01 0.02 0.04 0.01 6.7 Table 3-9. Concentration of  REE's and normalized La/Lu ratios for  mineral soils and moss monitoring stations near the peat sampling sites, and near the ore concentrate unloading site. -X-Zn Concentrate Site -B-Cominco ^Thunder Road Bombi Summit -^Champion Lakes Figure 3-9. Chondrite normalised REE pattern for  median values of  REE's deposited on moss monitoring stations near the peat sampling sites and the Zn-concentrate unloading site. profiles.  The high concentrations of  REEs at the Comino site relative to the Thunder Road site suggests that REEs are emitted by the smelter. However, the high REE concentrations found  within the Bombi Summit site suggest a secondary anthropogenic or a geogenic source. The high ash contents in the Cominco and Bombi Summit site relative to the Thunder Road site (Table 3-2) suggest that the REE conentration is partially a function  of  the ash content, which may be assumed to be predominantly of  geogenic origin. The tendency for  concentrations to increase with depth suggests either that the REE's are derived from geogenic sources, or that there is some degree of  post-depositional mobility within the peats. The peats in all profiles  exhibit LREE enrichment when normalised to a chondrite standard (Fig. 3-10; Table 3-10). This trend is most pronounced in the Cominco peat profile,  where La/Lu ranges from 12.6-16.5. The La/Lu ratios decrease in the order Thunder Road> Bombi Summit>Champion Lakes. Although this trend is somewhat consistent with the extent of  input from the smelter, the lower La/Lu ratio found  within the Champion Lakes profile  may be attributable to the more mafic  nature of  the bedrock when compared to the other sites. The La/Lu ratios were higher in the top 3 cm of  the Thunder Road profile  than in the deeper samples; no obvious trends in La/Lu versus depth were noted in the other profiles. There is no Ce anomaly in the Cominco, Thunder Road or Bobmi Summit peat profiles,  consistent with the findings  of  Akagi et al. (2002) who found  negative Ce anomalies in live plants but not in peats. However, there is a pronounced negative Ce anomaly in the top 2 cm of  the Champion Lakes peat profile.  The negative Ce anomaly found  in plants is also typical of  oxidised waters (Leybourne et al., 2000; Protano and Riccobono, 2002) and is probably a reflection  of  the chemistry of  the water taken up by the plants. The negative Ce anomaly found  at the peat surface  suggests that either the peat at the profile  surface  is composed of  relatively unhumified  plant material that has retained the REE signature characteristic of  living plants, and/or that the interstitial water is oxidised. There is a slight negative Eu anomaly in all profiles,  with the exception of  Bombi Summit, consistent with findings  from other peat studies (Akagi et al., 2002, Krachler et al., 2002). This negative anomaly is most pronounced in the uppermost samples of  the peat profiles  in the case of  the Cominco and Thunder Road sites. 3.5.2.4 Soil The concentrations of  REEs in the mineral soils sampled close to the peat sampling locations and near the ore concentrate unloading site do not decrease with radial distance from the smelter as they do with the peats (Table 3-9; Fig. 3-11). The highest La concentration is found  in the Champion Lakes 0-3 cm soil, 13.5 km from the smelter. Lanthanum concentrations are only slightly higher at the ore concentrate unloading site than at the Cominco or Thunder Road sites. The concentrations of  REEs measured in the mineral soils are higher than the concentrations found  within the peats, indicating that REEs are not preferentially  sequestered in the peat. In contrast, the trace elements Sb, As, Cd, Cu, Pb and Zn are concentrated in the peats relative to the mineral soils at the sampling sites close to the smelter (Chapter 2). This suggests that smelter-derived REEs deposited on the mineral soil are less subject to post-depositional mobilisation and sequestration in the peat than the trace elements. Additionally, the relative proportion of geogenic versus smelter-emitted REEs in the mineral soil is likely to be higher than for the aforementioned  trace elements. It is probable that geogenic elements will be concentrated in the mineral soil relative to the peat, and so any characteristic REE geochemical signature from smelter emitted materials is likely to be more "diluted" in the mineral soil. Similar to the peats, the mineral soils are LREE enriched when normalised to chondrites (Fig. 3-11). The Champion Lakes soil has the lowest La/Lu ratio (Table 3-9), consistent with the peat data (Table 3-10). The 0-3 cm soil from the ore concentrate unloading site has the highest La/Lu. No Ce or Eu anomalies were noted for  any of  the soils, unlike the peat, indicating that geochemical fractionation  is taking place within the peats subsequent to REE incorporation. / Sample Depth (cm) La Ce Nd Sm Eu Tb Yb Lu Norm ppm ppm ppm ppm ppm ppm ppm ppm La/Lu Cominco 0-1 15.2 26.4 1.7 1.9 0.4 0.4 0.9 0.1 14.3 1-2 17.4 31.2 3.2 2.2 0.5 0.4 1.1 0.1 12.9 2-3 24.8 47.2 3.5 3.2 0.6 0.3 1.5 0.2 13.6 3-4 20.8 38.4 1.4 2.7 0.5 0.4 1.3 0.1 16.6 4-5 27.3 50.5 2.5 3.4 0.6 0.4 1.4 0.2 15.4 17-20 23.0 41.2 2.4 2.9 0.7 0.5 1.2 0.2 12.6 Thunder Road 0-1 5.1 8.4 1.6 0.6 0.1 0.1 0.3 0.04 14.6 1-2 6.6 11.3 1.3 0.9 0.1 0.1 0.3 0.06 10.9 2-3 4.9 7.7 2.0 0.6 0.1 0.1 0.4 0.03 15.0 3-4 7.0 11.5 1.6 0.9 0.2 0.1 0.5 0.07 10.2 5-6 7.0 12.3 1.3 0.9 0.2 0.2 0.4 0.07 10.2 11-14 8.1 13.3 0.5 1.1 0.2 0.1 0.5 0.08 10.3 Champion Lakes 0-1 0.5 0.3 0.5 0.1 0.03 0.1 0.1 0.01 9.9 1-2 1.0 1.1 5.4 0.2 0.1 0.2 0.1 0.01 10.3 2-3 6.8 12.0 3.4 1.0 0.3 0.1 0.4 0.08 9.2 3-4 3.4 5.9 5.4 0.5 0.1 0.1 0.2 0.05 6.8 4-5 7.8 14.0 2.0 1.1 0.3 0.2 0.5 0.09 8.6 11-16 2.4 4.6 2.2 0.4 0.1 0.1 0.2 0.03 8.8 Bombi Summit 0-1 11.4 21.6 1.1 1.7 0.4 0.3 0.8 0.1 8.9 1-2 14.5 27.0 1.0 2.3 0.7 0.4 0.9 0.2 9.7 2-3 17.1 32.5 1.0 2.6 0.7 0.4 1.1 0.2 11.3 3-4 17.0 31.6 1.3 2.6 0.7 0.4 1.1 0.2 9.7 4-5 12.9 22.8 0.9 1.8 0.4 0.2 0.8 0.1 9.5 7.5-9 22.5 39.5 0.5 3.4 0.8 0.1 1.5 0.3 8.8 Table 3-10. Concentration of  REE's and normalized La/Lu ratios. 00 K> I—T . CTQ c >-1 CD n cr o E3 O. >-t t3 O ft o-s tn 51 i -I TS ft a H o 52 ?T 10 La Ce Nd Sm Eu Tb Yb Lu -0-1 — 1-2 —2-3 —3-4 —4-5 —17-20 100 CHAMPION LAKES La Ce Nd Sm Eu Tb Yb Lu - 0 - 1 - 1 - 2 - 2 - 3 -a-3-4 - 4 - 5 -11 -16 0-1 - 1 - 2 —2-3 - 3 - 4 —5-6 —11-14 -0 -1 - 1 - 2 - 2 - 3 - 3 - 4 - 4 - 5 -7.5-9 100 10 1 La Ce Nd Sm Eu Tb Yb Lu 1000 100 10 1 La Ce Nd Sm Eu Tb Yb Lu -x-Zn Concentrate Site -e-Cominco -a-Thunder Road Bombi Summit -i-Champion Lakes Figure 3-11. Chondrite normalised REE's for  mineral soils near the peat sampling sites and near the ore concentrate unloading site. 3.5.2.3 Peat Data Normalised  to Soil  Data In order to examine how REEs are fractionated  between the peat and mineral soil, the peat REE concentration data was normalised to the 0-3 cm and 3-10 cm mineral soil data (Figs. 3-12 and 3-13). Following normalisation, the LREE element enrichment pattern present in the peat and soil disappears, indicating that LREE enrichment is not controlled by the nature of  the environmental sampling media in this case. The negative Eu anomaly persists in the Cominco peat profile  and to a lesser extent in the Thunder Road profile,  indicating that the Eu anomaly occurs as a result of  the reducing conditions present within the peat. The main feature  that appears as a result of  the normalisation is a positive Nd anomaly, which is pronounced in the Thunder Road profile,  and is present to a lesser extent in the Cominco profile  and to a slight extent in the upper samples of  the Champion Lakes profile.  Protano and Riccobono (2002) found  that Nd is scavenged from solution by Fe-oxides more readily than other REEs. Iron oxides have been shown to be important in the retention of  trace elements in both peat and soils, and the positive Nd anomaly is most pronounced in the surface  peats, where oxidising conditions are most prevalent in the profile.  However, the high affinity  of  Fe-oxides for  Nd would suggest a lower potential for  migration from the soil to the peat than for  the other REEs. The reason for  the positive Nd anomaly in the peat is therefore  uncertain, normalised to soil. The apparent positive Tb anomaly is due to the data handling practice of  assigning a value of  Vi the detection limit when concentrations are below the detection limit. It seems likely from the data pattern that this assigned value is too high. 3.5.2.5 Stream  Sediment  Data Stream sediments have similar REE patterns to soil data (Fig. 3-14). No Ce or Eu anomalies are present in any of  the stream sediment samples. Apparent anomalies in the Dy-Yb-Lu series may be due to decreased analytical precision as a result of  low concentrations, or may reflect  a real anomaly. LREE enrichment is also found  in the stream sediments, with the highest LREE enrichment occurring in the sediment sampled near to the ore concentrates unloading site (Ryan Creek La/Lu-17.1). LREE enrichment decreases with decreasing influence  from the smelter, from Trail Creek, equivalent to the Cominco site (La/Lu of  16.2) to Topping Creek, equivalent to the Thunder Road site (La/Lu of  14.6) to Champion Creek, equivalent to the Champion Lakes site (La/Lu of ^-0-1 —•—1-2 — 2 - 3 ^ 3 - 4 -*-4-5 —•— 17-20 La Ce Nd Sm Eu Tb Yb Lu - - -1 -2 -^—2-3 ^ 3 - 4 — 5-6 —^ 11-14 La Ce Nd Sm Eu Tb Yb Lu - ^ 0 - 1 —•—1-2 2-3 - ^ 3 - 4 —x—4-5 • 11-16 Figure 3-12. REE's for  peat normalised to the nearest mineral soil (0-3 cm depth). La -•-0-1 -*— 3-4 Ce Nd Sm 1-2 •4-5 Eu Tb Yb 2-3 •17-20 Lu 0.01 0.01 0.001 La Ce - • - 0-1 -*— 3-4 Nd Sm Eu La Ce • - 0 - 1 -*—3-4 1 - 2 -*— 5-6 Tb -Ar— 2-3 -•—11-14 Yb Nd Sm - 1 - 2 Eu 4-5 Tb Yb - A — 2 - 3 -•—11-16 Lu Lu Figure 3-13. REE's for  peat normalised to the nearest mineral soil (3-10 cm depth). 1 . , La Ce Nd Sm Eu Tb Dy Yb Lu •s— TRAIL CREEK CHAMPION CREEK*- RYAN CREEK (COMINCO) (CHAMPION LAKES) (Zn CONCENTRATE SITE) TOPPING CREEK TOPPING CREEK (THUNDER ROAD) (THUNDER ROAD) Figure 3-14. Chondrite normalised REE pattern for  stream sediments collected near the peat sampling sites and near the Zn-concentrate unloading site. 9.8). Overall concentrations of  REEs also decrease with increasing distance from the smelter, from 113 ppm La at Ryan Creek, to 46 ppm La at Champion Creek. 3.5.3 Discussion The REE concentrations within the smelter feedstock  samples are lower than the REE concentration in the Cominco and Bombi Summit peats, and lower than the soil REE concentrations at any site. Concentrations of  REEs were elevated in the smelter waste materials (slags) relative to the peats and soils. The pronounced LREE enrichment seen in the peat, soil, stream sediments and most of  the moss monitoring sampling media is absent from all the analysed smelter samples (Table 3-8; Fig. 3-8), despite the high La/Lu observed for  moss monitoring media and stream sediments sampled near the concentrate unloading site. There is a positive Eu anomaly in both the feedstock  and the slags and wastes, that was not noted in any of  the environmental sampling media (Fig. 3-8). These results suggest two possibilities: 1. LREE enrichments observed in the sampled moss, peat, soils and stream sediments are due to weathering and preferential  loss of  HREE in smelter emitted materials. Preferential  weathering of  HREEs in sediments was observed by Ohlander et al. (1996) and Johannesson and Zhou (1999). 2. The pronounced LREE enrichment in the peats, soils and stream sediments is due to the natural background geochemistry of  the sampled materials and the impact of  the smelter is relatively insignificant  to the REE signature. The less pronounced LREE enrichment in the moss monitoring stations is as a result of  the combined input of  smelter-derived particulates and geogenic dust. Given that the media in the moss monitoring stations is changed every three months, which is unlikely to be sufficient  time to degrade the trapped particulates sufficiently  to mobilise the HREEs, it is most likely that the observed LREE enriched signature is due primarily to the natural local geochemistry. The similar La/Lu values for  corresponding peat and mineral soil sampling sites support the geogenic origin of  the LREE enrichment. The low La/Lu at the Cominco moss 'monitoring station reflects  the influence  of  the smelter to a greater extent than the peat and soil at this location, as peat and soil are more impacted by geogenic factors  than the moss. With increasing distance from the smelter, the La/Lu generally increases in the moss monitoring media, indicating a decrease in the smelter impact. 3.6 Conclusions Attempts to apply geochemical signature techniques to differentiate  between anthropogenic and geogenic trace metal inputs in an area impacted by a Pb-Zn smelter were generally unsuccessful.  Element ratios were found  to produce unsatisfactory  results due to the highly variable nature of  the geogenic "conservative" elements within each profile.  Given that these elements do not necessarily represent a stable geochemical baseline in the examined samples, they are not suitable parameters for  normalisation. The factors  controlling the post-depositional mobility of  the trace elements (e.g. redox conditions, hydrogeological flow  paths) are independent of  the factors  that controlled the incorporation of  geogenic materials throughout the depositional history of  the peat (e.g. flooding,  wind) and relative enrichment of  conservative elements and trace elements may occur independently of  each other or of  anthropogenic input. Element ratios have been shown to be effective  in chronostratigraphic trace element studies of  ombrotrophic peats subject to deposition by less mobile species (e.g. Shotyk, 1996; Shotyk et al., 1996). However, the ambiguous results obtained by this study illustrate that the conservative element ratio method is probably unsuitable for  sites subject to a high input of  mobile anthropogenic trace elements. Furthermore, the environmental significance  of enrichment ratios should not be accepted uncritically, as a high degree of  enrichment is as likely to be the result of  the depletion of  the conservative element as the enrichment of the trace element in question. Despite the high rates of  trace element deposition from the smelter to the surrounding area, the relatively flat  REE signature of  smelter materials contrasted with the pronounced LREE enrichment in the sampling media used in this study. LREE enrichments were found  in the peats, soils, stream sediments and, to a lesser extent, moss monitoring media, while the REE signature of  an assortment of  smelter materials was relatively flat,  with a positive Eu anomaly that was not seen in any of  the environmental sampling media. This suggests that the LREE enrichment is characteristic of  the background geochemistry of  the sampled media and of  the geogenic dust captured by the moss monitoring stations, and that the impact of  the smelter on local REE signatures is relatively minor. These findings  indicate that while the REE signatures of  waters have been shown to be useful  geochemical tracers in some cases, REE signatures from emitted smelter materials are subject to masking by the geochemistry of  the surrounding environment, and are of  limited use as tracers in environmental studies. 3.7 References Akagi, T, Feng-Fu, F. and Tabuki, S., 2002. Absence of  Ce anomaly in the REE patterns of  peat moss and peat grass in the Ozegahara peatland. Geochemical Journal, 36: 113-118. Astrom, M., 2001. Abundance and fractionation  patterns of  rare earth elements in streams affected  by acid sulphate soils. Chemical Geology, 175: 249-258. Bau, M., 1999. Scavenging of  dissolved yttrium and rare earths by precipitating iron oxyhydroxide: Experimental evidence of  Ce oxidation, Y-Ho fractionation  and lanthanide tetrad effect.  Geochimica et Cosmochimica Acta, 63: 67-77. Beddoe-Stephens, B. and Lambert, R.StJ., 1981. Geochemical, mineralogical and isotopic data relating to the origin and tectonic setting of  the Rossland volcanic rocks, southern British Columbia. Canadian Journal of  Earth Sciences, 18: 858. Chiarenzelli, J., Aspler, L., Dunn, C., Cousens, B.; Ozarko, D. and Powis, K., 2001. Multi-element and rare-earth element compostion of  lichens, mosses and vascular plants from the Central Barrenlands, Nunavut, Canada. Applied Geochemistry, 16: 245-270. Coedo, A, G., Dorado, M.T., Padilla, I. and Alguacil, J., 1998. Use of  boric acid to improve the microwave-assisted dissolution process to determine fluoride  forming  elements in steels by flow  injection inductively coupled plasma mass spectrometry. Journal of  Analytical Atomic Spectrometry, 13: 1193-1197. Dupre, B., Gaillardet, J., Rousseau, D. and Allegre, C.J., 1996. Major and trace elements of  river-borne material: The Congo Basin. Geochimica et Cosmochimica Acata 60: 1301-1321. Eskenazy, G.M., 1999. Aspects of  the geochemistry of  rare earth elements in coal: an experimental approach. International Journal of  Coal Geology, 38: 285-295. Goodarzi, F., Sanei, H. and Duncan, W.F., 2001. Monitoring the distribution and deposition of  trace elements associated with a zinc-lead smelter in the Trail area, British Columbia, Canada. Journal of  Environmental Monitoring, 3: 515-525 Goodarzi, F., Sanei, H., Garrett, R.G. and Duncan, W.F., 2002. Accumulation of  trace elements on the surface  soil around the Trail smelter, British Columbia, Canada. Environmental Geology, 43: 29-38. Goodarzi, F., Sanei, H. and Duncan, W.F. 2003. Deposition of  trace elements in the Trail region, British Columbia; an assessment of  the environmental effect  of  a base metal smelter on land. Geological Survey of Canada Bulletin 573, 50 pp. Hannigan, R.E. and Sholkovitz, E.R., 2001. The developmnt of  middle rare earth element enrichments in freshwaters:  weathering of  phosphate minerals. Chemical Geology, 175: 495-508. Hall, G.E.M., Vaive, J.E. and McConnell, J.W., 1995. Development and application of  a sensitive and rapid analytical method to determine the rare-earth elements in surface  waters. Chemical Geology, 120: 91-109. Johannesson, K.H. and Zhou, Xi., 1999. Origin of  middle rare earth element enrichments in acid waters of a Canadian High Arctic Lake. Geochimica et Cosmochimica Acta, 63: 153-165. Krachler, M., Mohl, C., Emons, H. and Shotyk, W., 2002. Influence  of  digestion procedures on the determination of  rare earth elements in peat and plant samples by USN-ICP-MS. Journal of  Analytical. Atomic Spectrometry, 17, 844-851. Land, M., Ohlander, B., Ingri, J. and Thunberg, J., 1999. Solid speciation and fractionation  of  rare earth elements in a spodosol profile  from northern.Sweden as revealed by sequential extraction. Chemical Geology, 160: 121-138. Leybourne, M.I., Goodfellow,  W.D., Boyle, D.R. and Hall, G.M., 2000. Rapid development of  negative Ce anomalies in surface  waters and contrasting REE patterns in groundwaters associated with Zn-Pb massive sulphide deposits. Applied Geochemistry, 15: 695-723. Little, H.W., 1982. Geology of  the Rossland-Trail Map-Area, British Columbia. Geological Survey of  Canada Paper 79-26, 38p. Prohaska, T., Hann, S., Latkoczy, C. and Stingeder, G., 1999. Determination of  rare earth elements U and Th in environmental samples by inductively coupled double focusing  sector field  mass spectrometry (ICP-SMS). Journal of  Analytical Atomic Spectrometry, 14: 1-8. Protano, G. and Riccobono, F., 2002. High contents of  rare earth elements (REEs) in steram waters of  a Cu-Pb-Zn mining area. Environmental Pollution, 117: 499-514. Reimann, C. and De Caritat, P., 2000. Intrinsic flaws  of  element enrichment factors  (EFs) in environmental geochemistry. Environmental Science and Technology, 34: 84- 91. Sevigny, J.H., 1990. Geochemistry of  the Jurassic Nelson plutonic suite, southeastern British Columbia. In: Project Lithoprobe; southern Canadian Cordillera transect workshop. Lithoprobe Report 11, pp. 41-52. Shotyk, W., 1996. Natural and anthropogenic enrichments of  As, Cu, Pb, Sb, and Zn in rainwater-dominated versus groundwater-dominated peat bog profiles,  Jura Mountains, Switzerland. Water, Air and Soil Pollution, 90: 375-405. Shotyk, W., Cheburkin, A.K., Appleby, P.G., Fankhauser, A. and Kramers, J.D., 1996. Two thousand years of  atmospheric arsenic, antimony and lead deposition recorded in an ombrotrophic peat bog profile,  Jura Mountains, Switzerland. Earth and Planetary Science Letters, 145: E1-E7. Shotyk, W., Cheburkin, A.K., Appleby, P.G., Fankhauser, A. and Kramers, J.D., 1997. Lead in three peat bog profiles  Jura Mountains, Switzerland: Enrichment factors,  isotopic composition and chronology of atmospheric deposition. Water, Air and Soil Pollution, 100: 297-310. Takahashi, Y., Minai, Y., Ambe, S., Makide, Y. and Ambe, F., 1999. Comparison of  adsorption behavior of  multiple inorganic ions on kaolinte and silica in the presence of  humic acid using the multitracer technique. Geochimica et Cosmochimica Acta, 63: 815-836. Wedepohl, K.H., 1995. The composition of  the continental crust. Geochimica et Cosmochimica Acta, 59: 1217-1232. Wiist, R.A.J., Ward, C.R., Bustin, R.M. and Hawke, M.I., 2002. Characterization and quantification  of inorganic constituents of  tropical peats and organic-rich deposits from Tasek Bera (Peninsular Malaysia): implications for  coals. International Journal of  Coal Geology, 49: 215-249. Chapter 4 Particulate matter in peat from  the environs of  a Pb-Zn smelter determining geogenic and anthropogenic input 4.1 Introduction Most environmental trace element studies utilise a variety of  chemical and statistical methods in order to determine the impact of  a source of  contaminants. A smelter operation emits particulate matter rich in trace metals into the surrounding environment, in the form of  particles emitted during the processing, and fugitive  dust released during the transporting and storage of  ore materials. Particulate matter may also be derived from local soils and sediments as a result of  weathering processes, and may contain high concentrations of  geogenic trace elements. Peat in the vicinity of  a point source of  particulate emissions is a sink for  both atmospheric particle deposition and elements mobilised from the surrounding environment. Ash materials derived from peats sampled from the area surrounding a Pb-Zn smelter in Trail, British Columbia, Canada were observed using scanning electron microscopy (SEM). The primary objective was to observe the size, elemental composition and morphology of  particles containing heavy elements. Natural mineral matter typically exhibits crystalline characteristics and has an angular morphology, while anthropogenic particulate matter emitted from a smelter is typically rounded, as a result of rapid solidification  during cooling in the stack (Goodarzi et al., 2001; 2003). The chemistry of  the anthropogenic and geogenic particles also differs.  Materials emitted from the smelter during processing are typically metallic oxides resulting from the roasting process, and silicate fly  ash from  coal combustion, while geogenic particles will reflect  the local geology. Energy Dispersive Spectroscopy (EDX) is used to determine elemental associations, which help to determine the major modes of  occurrence for elements in the particulate matter found  within the peats. Examination of  the particulate matter using SEM-EDX allows the determination of  the elemental makeup of  the observed particles, and thus makes it possible to relate chemistry and particle morphology in order to determine source. Differentiating  between anthropogenic and geogenic sources of  trace elements in an area of  heavy anthropogenic impact, such as a smelter, is difficult  by bulk chemical methods alone. It is not reasonable to assume that there is no geogenic impact, or that the background geochemistry will be uniform throughout the sampling area, if  indeed a reasonable baseline geochemistry can be established. An SEM-EDX study may help in determining the nature of  the particulate matter, and the elemental composition of geogenic minerals, and may reveal evidence of  industrial impact such as fly  ash particles. The purpose of  this study is to qualitatively observe the morphology and elemental composition of  particulate matter found  within peats in the vicinity of  a Pb-Zn smelter, in order to determine the source. This paper discusses the results of  one phase of a project which examines the role of  peat as a sink for  trace elements in the area surrounding a Pb-Zn smelter. Other phases of  this project have examined the total concentration of  selected trace elements (Sb, As, Cd, Cu, Pb, Ni and Zn) in peat profiles in the vicinity of  the smelter, and discussed the results with reference  to measurements of atmospherically deposited trace elements (Goodarzi et al., 2001; 2003) and trace element concentrations in nearby mineral soils (Chapter 2). A six-step sequential leaching procedure was applied to determine the mode of  occurrence of  selected trace elements in the peat profiles  (Chapter 5). The mode of  trace element occurrence is discussed in the context of  the early diagenetic conditions within the peat, as determined by organic petrography (Chapter 6). 4.2 Sampling Locations The Teck-Cominco smelter is located in Trail, British Columbia, situated in the Columbia River Valley within the Columbia Mountains (Fig. 4-1). The smelter has been in operation since 1896, and produces Pb and Zn as it's primary products, as well as Cd, In, Ge, Ag, Au, Bi, Cu, As, sulphuric acid and ammonium sulphate fertiliser.  The valley is subject to the deposition of  materials containing elevated concentrations of  Pb and Zn, as well as associated elements such as Sb, As, Cu, and Cd. These elements are known to be derived both from material emitted from the smelter during processing, and from fugitive  emissions, such as dust from stored concentrate piles (Goodarzi et al., 2001; 2003). The area is underlain primarily by Jurassic/Cretaceous plutonic rocks, emplaced during the Columbian Orogeny (the Nelson Intrusives), and Jurassic metavolcanic and metasedimentary rocks (the Elise Formation). The geochemistry of  the area is complex and variable. A compilation of  published bedrock geochemical data (Sevingy, 1991; Beddoe-Stephens and Lambert, 1981) shows the variability in element concentrations between rock types in the area (Table 4-1). Small and large scale mining operations have been active within the area over the past century, as a result of  the numerous localised ore deposits. In areas such as this, it is difficult  to confidently  assign baseline geochemistry in order to assess the environmental impact of  the smelter. Three sites were selected in order to obtain a representation of  the lateral distribution of  trace elements in the areas surrounding the smelter. Site selection criteria included distance and direction from the smelter, bedrock geology, and availability of organic sediments. The latter posed a problem, as peatlands in the Trail area are few in number due to the relatively dry climate. A description of  the sampling sites is provided in Table 4-2, and locations are indicated on Figure 4-1. Nelson Intrusives Elise Formation % Granodiorite Quartz Gabbro K-Tonalite Granodiorite Ankaramite Basalt K-Andesite Si02 Ti02 AI2O3 Fe203 MnO MgO CaO Na20 K 2 0 P 2 0 5 64.2 0.6 16.0 4.9 0.1 2.0 4.8 3.5 3.1 0.2 54.2 0.9 17.5 9.7 0.2 4.2 7.0 4.0 1.8 0.3 60.5 0.6 16.9 6.4 0.1 2.4 4.8 4.7 2.9 0.3 66.0 0.4 16.5 3.0 0.1 1.1 4.0 4.1 3.4 0.2 48.1 0.8 12.2 12.9 0.2 12.4 9.4 2.0 1.6 0.3 51.3 0.9 14.5 11.6 0.2 6,4 9.4 3.1 1.8 0.3 55.4 0.8 15.1 9.4 0.1 4.1 7.8 3.8 2.8 0.3 ppm V Cr Ni Zn Ga Rb Sr Y Zr Nb Ba Pb Th Cu 100 20 5 69 18 93 654 18 130 16 1039 13 9 223 32 13 126 21 72 744 20 65 8 809 9 5 130 22 7 82 19 96 700 21 122 12 1057 9 6 59 10 5 54 17 102 782 14 107 13 1418 12 6 264.3 664.2 212.2 97.2 39.2 541.5 14.5 61.8 2.8 588.2 118.2 314.5 125.8 39.8 97.2 38 755.8 18.5 89.5 3.8 685 63.8 222.8 26.5 6.8 90.5 52.2 889.5 17.2 114.8 4.2 939.2 37.8 Table 4-1. Compilation of  published geochemical data, for  the Nelson Intrusives (Sevigny, 1990) and the Elise Formation (Beddoe-Stephens and Lambert, 1982). |49°15' QUATERNARY •UnconsoSdated sediments; till, sand yavel, silt JURASSIC/CRETACEOUS j — — | NELSON INTRUSIONS: A A granodiorite; miner quartz r I diorita and diorite i » -i ROSSLAND MONZONITE: x x ' J biotite-taxnblende-augrte L — J monzonte, manly medium grained 1 ELISE FORMATION: . flow breccia, massive ande sites J and basalts, agglomerate, tuff, breccia, laminated siltstone MOUNT ROBERTS FORMATION: 5black siltstone and argillaceous quartette, slate, greywacke, chert, pebble conglomerate, lava; limestone AGE UNKNOWN TRAIL GNEISS: amphibotite and grey biotits gneiss, hornblende gneiss, rreca schist, aptite and pegmatite Kilometres 1 0 OT~ Figure 4-1. Geological map of  the Trail area (from Little, 1982) showing sampling locations. Site Bedrock Geology Vegetation pH and Comments Cominco UTM Zone 11, 0446226, 5440375 Trail Pluton (granodiorite, quartz diorite) Typha  latifolia (common cattail), Epilobium  ciliatum (purple leafed willow herb) Carex  urticulata (beaked sedge) Equisetum palustre,  and Deschampsia cespitosa  (tufted hairgrass) 5.3 (surface) 5.6 (mineral sediment) 5.7 (pore water) -saturated at the time of sampling, although gleying was noted in the uppermost 15 cm-intermittently aerobic. Zones of oxidation were noted around roots. Thunder Road UTM Zone 11, 0445145, 5440974 Trail Pluton (granodiorite, quartz diorite) Typha  latifolia (common cattail), Pinus  sp. (pine) Betula  sp. (birch) Epilobium  ciliatium Equisetum palustre,  and Deschampsia cespitosa  (tufted hairgrass) 5.0 (surface) 5.4 (mineral sediment) 5.7 (pore water) -watertable was approximately 10 cm below the surface at the time of sampling Champion Lakes Provincial Park UTM Zone 11, 0453623, 5448726 Elise Formation (flow breccia, lava agglomerate, volcanic breccia, tuff, tuffaceous  conglomerate, andesite, basalt, augite porphyry, metamorphosed to greenschist fades). Limestone xenoliths and calcite amygdules found within mafic flows. Economic deposits of Pb, Zn, Ag, associated with sulphides. Carex  sp. Nuphar  lutea  (cow lily) Assorted grasses 6.5 (surface) 6.5 (15cm depth) 6.4 (pore water) Saturated profile Table 4-2. Description of  the sampling sites: location, bedrock geology, vegetation and conditions within the peat profile. 4.3 Methodology Peats were sampled by digging pits with a stainless steel shovel and cutting blocks of  peat from the pit wall. The peat blocks were immediately sealed in plastic freezer  bags and refrigerated  within 24 hours. Peat samples were combusted at 400°C for  24 hours, and the mineralogy of  the resulting ash residue was determined by X-ray diffraction.  Interpretation of  the XRD spectra was aided by the use of  PC-APD (version 3.6), which also provides data for  the normalization of  peak heights, permitting semi-quantitative interpretation. Samples of  peat ash material were carbon, coated for  observation with the SEM in back-scattered electron mode. This procedure was not conducted with the intention of providing quantitative information,  and as such no point counting or image analysis was performed,  rather it was an attempt to characterise the morphology and chemistry of  the materials observed to make up a significant  portion of  the ash. The elemental composition of  selected particles was determined by EDX. Trace element concentrations of  the whole peat samples were determined by Instrumental Neutron Activation and ICP-MS (Becquerel Laboratories, Mississaugua, Canada). 4.4 Results and Discussion 4.4.1 Trace  element  concentration The Cominco site is located 0.5 km from the smelter property and is subject to significant  deposition of  trace elements (Goodarzi et al., 2001; Chapter 2). The Thunder Road site is located 1.6 km from the smelter site, and is partially physically sheltered from deposition from the smelter by its topographic position above the valley, and an intervening ridge of  rock. Deposition from the smelter is a major source of  trace elemental input at this location (Goodarzi et al., 2003; Chapter 2), but to a lesser extent than at the Cominco site. The Champion Lakes sampling site is located approximately 13.5 km from the smelter, and is physiographically sheltered from smelter fall-out  by the intervening mountains. However, there is some suspected smelter impact at this location (Goodarzi et al., 2003; Chapter2). Bulk trace element concentrations for  the peat profiles  are presented in Table 4-3. Concentrations of  all the examined trace elements, with the exception of  Ni, are highest near the smelter and decrease rapidly with radial distance. Concentration differentials between peat and nearby mineral soils indicate that some of  the emitted trace elements are deposited as soluble aqueous species, or as readily soluble solids, which migrate from the mineral soil to the peat following  deposition (Chapter 2). However, sequential leaching indicates that the recalcitrant material of  the residual phase also contributes to the overall trace element concentration (Chapter 5). Sample Depth Sb As Cd Cu Pb Ni Zn (cm) ppm ppm ppm ppm ppm ppm ppm COMINCO' 0-1 126 78.5 61.7 196 3550 10.8 5910 1-2 141 97.8 65.2 223 3830 11.6 5600 2-3 128 153 53.8 215 3780 15.4 4230 3-4 142 116 60.3 227 3830 13.2 4730 4-5 121 164 47.1 214 3940 14.8 3570 17-20 85.0 247 60.1 226 5240 12.4 3040 THUNDER ROAD 0-1 34.1 60.0 119 82 1670 9.2 2730 1-2 37.7 64.4 105 103 2070 10.6 2270 2-3 33.0 62.0 139 84 1750 9.4 2900 3-4 37.6 65.4 134 96 1980 11 2590 5-6 32.5 67.5 104 92 1990 10.8 2330 11-14 34.9 59.0 120 96 1940 11 2300 CHAMPION LAKES 0-1 1.2 1.8 2.4 12 77 3 148 1-2 1.6 3.6 3.7 16 101.5 6 194 2-3 5.5 8.3 6.0 14 172 9.2 214 3-4 3.8 6.8 4.4 13 114.5 6 208 4-5 9.0 15.4 6.8 20 246 8.8 268 11-16 15.2 15.8 6.5 24 324 4.4 178 Table 4-3. Bulk concentrations of  selected trace elements within the peat profiles. 4.4.2 Mineralogy 4.4.2.1 Cominco Site Mineralogy at the Cominco Site consists predominantly of  quartz and sodic plagioclase feldspar,  consistent with the underlying granodiorite bedrock (Table 4-4). The quartz content ranges from 33-63%, with the highest concentration occurring at the surface,  a trend seen in most of  the profiles  examined. Feldspar values range from 10-35% for  plagioclase and 0-31% for  orthoclase. There is little or no clay, indicating a lack of  weathering (supported by the thin to absent soils throughout the area). Phyllosilicates consist of  mica (less than 5%) and chlorite (less than 2%), consistent with the underlying geology. Small amounts of  amphibole are present throughout the profile,  with a maximum of  5% at the profile  base, but usually less than 2%. Additionally, a large amount of  pyroxene is found  in the upper portions of  the profile  (46% at 2-3 cm and 12% at 3-4 cm), but is absent from the remainder of  the profile. 4.4.2.2 Thunder  Road  Site The mineralogy of  the Thunder Road profile  is similar to the Cominco profile, which is anticipated, as both sites overlie the Trail pluton, and are located approximately 1 km from one another. Quartz contents range from 49-62%, plagioclase contents range from 11-22%, and orthoclase ranges from 6-22%. Mica ranges from 2-8%, while chlorite is highly variable, from 0% to 14% in the 6-7 cm sample. No clays were detected in the profile.  Amphiboles range from 2-6%. Anhydrite, which comprises 19% of  the 8-11cm sample, may form from the dehydration of  gypsum during the ashing process. The gypsum may have been present in the sampled peat, as gypsum is produced as a bi-product of  the zinc operation, or it may have formed  as a result of  the oxidation of biogenic sulphur during ashing (Miller et al., 1979). 4.4.2.3 Champion Lakes Site The Champion Lakes mineralogy reflects  the geogenic input from the underlying Elise Formation. The top of  the peat profile  contains amphibole (10-16%) and chlorite (2%), which diminish to 1-7% and less than 1%, respectively, from 8-16 cm. Plagioclase contents range from 14-35%, while no orthoclase was detected. Broad XRD peaks in the phyllosilicate positions suggest that most of  the phyllosilicates are mixed layer clays rather than micas (2-6%, increasing downwards in the profile).  The presence of  calcite (13-38%, found  from 8-16 cm) may be attributed to limestone xenoliths and calcite amygdules found  within the flow  breccias, or to calcite formed  as a product of  green schist facies  metamorphism (Little, 1960). Anhydrite, probably formed  from the dehydration of  gypsum, occurs in the 6-7 cm sample. The gypsum may have formed  from the oxidation of  biogenic sulphur during the ashing process, or may form from gypsum from the smelter present in the original sample. Sample # Ash Quartz Feldspar Phylosilicates Amphi- Pyroxene Calcite Anhydrite % Plagioclase Orthoclase Mica Clay Chlorite bole Cominco Site 1-2cm 51.7 66. 16 13 3 0 0 2 0 0 0 2-3cm 63.2 34 10 6 2 0 2 1 46 0 0 3-4cm 62.4 37 35 9 3 0 1 2 12 0 0 4-5cm 66.2 60 21 11. 5 0 2 1 0 0 0 5-6cm 64 20 11 3 0 1 1 0 0 0 7-8cm 73.4 40 24 31 2 0 1 1 0 0 0 8-11cm 74.0 48 20 26 3 0 2 2 0 0 0 11-14cm 81.2 37 26 30 2 0 1 0 0 0 0 14-17cm 79.8 66 25 6 1 0 1 1 0 0 0 17-20cm 51.4 66 23 0 4 0 1. 6 0 0 0 Thunder Road Site 2-3cm 18.4 62 20 10 2 0 1 6 0 0 0 5-6cm 24.4 53 22 15 5 0 2 3 0 0 0 6-7cm 25.9 59 11 6 6 0 14 2 0 0 2 8-11cm 27.8 50 16 7 7 0 0 2 0 0 19 11-14cm 25.4 53 17 22 5 0 0 4 0 0 0 Champion Lakes Site 3-4cm 38.6 72 14 0 0 2 2 10 0 0 0 6-7cm 21.2 47 34 0 0 2 2 15 0 0 34 8-11cm 16.6 49 33 0 0 5 1 1 0 13 0 11-16cm 12.9 48 28 0 0 6 0 7 0 38 0 4.4.3 SEM-EDX  investigation  of  peat ash The mineralogy as observed by SEM, is in general agreement with the XRD results. The most frequently  observed minerals were quartz, K and Ca feldspar  and Fe-silicates. Lead and Zn are the only trace elements that were frequently  detected by EDX. Lead and Zn occur both in combination and separately and are commonly associated with S, Fe, and Mn, although no Fe-Mn oxides or sulphide phases were detected by XRD. These phases may be amorphous, and thus not detectable by XRD, or may be present in amounts below the detection limits of  the XRD. Zinc is also associated with silicate minerals. Copper occurs only infrequently  in small concentrations, while Sb, As, Ni, and Cd were not observed at all. Ti was frequently  detected in particles from the Champion Lakes profile,  but was seldom detected in the sites close to the smelter, consistent with the low Ti concentration (0.57%) in the Trail Pluton granodiorite (Sevigny, 1990; Table 4-1). Ti concentrations in the Pb and Zn concentrates processed by the smelter are typically in the range of  0.018% (Goodarzi, unpublished data) and so Ti may be assumed to be primarily of  geogenic origin, and is not associated with the smelter emissions. Particles containing heavy metals are typically small and spherical and lack distinctive crystalline morphologies, suggesting a smelter origin. None of  the particles observed were oxides with a distinctive O peak on the EDX spectra, which is characteristic of  freshly  emitted roasted smelter materials. The roasted particulates are more soluble than the ore concentrates, and are therefore  likely to weather rapidly, particularly in an acidic environment such as a peatland. Particles with a morphology typical of  a geogenic origin may be derived from the local rocks and soils, or may be fugitive  dusts from ores transported in open railway cars or stockpiled in uncovered piles on the smelter property (Goodarzi et al., 2003). Geogenic particles and fugitive  dust are commonly silicates and sulphides. Examples of  geogenic and anthropogenic particles are illustrated in Figures 4-2 and 4-3, respectively. s 3 Zn 1 Zn i j l Ua L 0.0 5.0 10.0 kfiV 15.0 20.( Figure 4-2. Example of  a geogenic particle (sphalerite) from the Champion Lakes sampling site. The number on the EDX spectra refers  to the frame  number on Table 4-10. J 4b Si iL z 0.0 5.0 10.0 15.0 20.0 kfiV £ L i C i 4d Fe \ L z.n 0.0 5.0 10.0 15.0 20.0 kfiV Figure 4-3. Examples of  heat-treated spherical or rounded metallic particles (4a and c) emitted from the smelter, observed in the Thunder Road 7-8 cm peat. Also present are a spherical silicate fly  ash particle (4b) and a Zn-containg Fe-precipitate (4d). Numbers correspond to frame  numbers in Table 4-9. 4.4.3.1 Cominco Site Particles commonly observed in the surface  peats at the Cominco site which have a probable smelter origin are small (<2 pm), and rounded to subrounded, with low concentrations of  Pb and Zn associated with Si, Al, Ca, K, S, P, Fe and Mn (Fig. 4-4; Table 4-5). Particles likely to be of  geogenic or fugitive  origin based on morphology are subangular to subrounded, larger than the smelter particles, and contain Zn with Si and Fe (Fig. 4-5). At 4-5 cm depth, particles containing Pb and Zn were identified  as both anthropogenic (Fig. 4-6a) and geogenic (Fig. 4-7a) based on morphology (Table 4-6). With increasing peat depth (17-20 cm), Ti occurs more frequently,  suggesting a geogenic origin for  much of  the examined particulate matter (Table 4-7). Particles containing Pb and Zn are typically more angular at this depth than those observed at the peat surface. Some of  the Pb and Zn is associated with Fe-precipitates that are coating mineral surfaces (e.g. Fig. 4-8a). Small amounts of  Cu were noted infrequently  in association with Pb and Zn (Figs. 4-8, 4-9 a) in the Cominco 17-20 cm sample. 4.4.3.2 Thunder  Road Spherical smelter-emitted particles containing Zn and Pb, with Cu, Fe, and Mn are common in the 0-1 cm sample (Fig. 4-10; Table 4-8). Iron precipitates, which contain elements scavenged from the peat pore water, frequently  have high Zn concentrations (Fig. 4-11 c and d). Angular, geogenic particles of  similar chemistry to the precipitates also contain Zn, possibly as a surficial  coating (Fig. 4-1 la). Zn also occurs as a native element on the surface  of  a quartz particle (Fig. 4-12). Lead was not detected in this sample, other than in the round smelter-emitted particles (e.g. Fig. 4-10). Zinc occurs in a variety of  associations in the 7-8 cm depth sample (Table 4-9): as an angular Zn-S-Si-Ca-Fe particle, probably of  geogenic origin (Fig. 4-13); as an agglomerated particle, including a subangular native Zn particle (Fig. 4-14a); as a precipitate containing Zn, Si, Ca and Fe, (Fig. 4-14b,c); and as an Fe-Zn surface  coating on a diatom (Fig. 4-14d). Spheroidal fly  ash particles composed of  Si-S-Ca-Fe were also observed (Fig. 4-13). The fly  ash particles, identified  based on the composition and morphology, are probably derived from coal burned at the smelter in the Pb-roasting operation, as there are no coal-fired  power plants in the area. 4.4.3.3 Champion Lakes The particles examined in the Champion Lakes peat frequently  contain Zn, with minor occurrences of  Cu and Pb (Table 4-10). Many of  these particles have angular morphologies, typical of  a geogenic origin, such as the sphalerite particle in Fig. 4-2. Few particles with a composition or morphology suggestive of  a smelter origin were detected within the Champion Lakes peat, with the exception of  spherical silicate fly  ash particles (Figs. 4-15 to 4-16). Spherical fly  ash particles typically have a low density and are frequently  hollow, and so are dispersed more widely in the environment than dense metallic particles. Many particles have higher Ti concentrations than were observed in the Cominco and Thunder Road sites (Fig. 4-15), in agreement with the higher Ti concentration associated with the background geochemistry at this site (Beddoe-Stephens and Lambert, 1981; Sevigny, 1990). Diatoms are abundant, in accordance with the lacustrine origin of the peat. 5a Si y « Fe A 0.0 5.0 10.0 keV 15.0 20.0 5c Si J P b ^ s Ca M n F Zn 0.0 5.0 10.0 keV 15.0 20.0 Figure 4-4. Examples of  rounded particles containing Pb and Zn present in the Cominco 1 -2 cm sample. Numbers on the EDX spectra refer  to the Frame number on Table 4-5. Figure 4-5. Example of  a subangular particle containing Zn from the Cominco 1-2 cm sample Number the EDX spectra refers  to the Frame number on Table 4-5. Table 4-5. Summary of  the observations from the Cominco 1-2 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures.  Elements present in high concentrations are listed in bold face  type. Frame Spot Size (Approx) Morphology Major Elements Trace Elements Comments 1 a 2 urn sub-rounded Al, Si, K, Ca, Fe minor Zn, Pb 2 a 1 urn sub-angular Si, K, Ca, Mn, S, Fe minor Zn geogenic? b 0.25 urn rounded Al, Si, K, Ca, P, S, Mn, Fe minor Zn geogenic? c 2 urn sub-angular Si, K, Ca, Mn, Fe minor Cr(7), Zn high amorphous content, interference 3 a 3 (am sub-rounded Si, K, Ca, P, S, Mn, Fe minor Pb, Zn slight O peak-may be altered smelter derived particle b 1 pm sub-rounded Si, K, Ca, S, Mn, Fe minor Pb, Zn c 3 tarn sub-rounded Mg, Si, Ca, S, Fe minor Zn d 1.5 urn sub-angular Mg, Si, Ca, S, Fe geogenic mineral particle 4 a 1.5 urn angular Si, Ca, P, S, Fe Pb, Sn, Cu? Zn? geogenic, based on morphology-fugitive dust? b <1 urn coating? Al, Si, K, Ca, S, Fe minor Zn, Pb(?) precipitate plus underlying mineral 5 a 2 urn rounded Al, Si, Ca, P, S, Fe Pb, minor Zn, slight O peak, probable smelter source b 1 urn rounded Al, Si, K, Ca, P, S, Fe Pb, minor Zn, slight O peak, probable smelter source c 1 |iim rounded Si, K, Ca, S, Mn, Fe minor Zn, Pb slight O peak, probable smelter source d 0.25 urn rounded Al, Si, K, Ca, S, Fe minor Zn, Pb slight O peak, probable smelter source e 0.25 urn rounded Al, Si, K, Ca, S, Fe minor Zn, Pb slight O peak, probable smelter source 6 a 2 fxm sub-rounded Al, Si, P, S, K, Ca, Mn, Fe Pb, minor Zn slight O peak, probable smelter source b 12 pm rounded Si, S, K, Ca, Fe fly ash c 0.5 pm rounded Si, S, K, Ca, Mn, Fe minor Zn, Pb slight O peak, probable smelter source 7 a 1 jam sub-angular Si, S, K, Ca, Fe Zn Possible geogenic source based on morphology b 0.25 urn sub-rounded Al, Si, S,K, Ca, Mn, Fe minor Pb? 8 a 2 urn Sub-angular Si, P, S, K, Ca, Mn, Fe Zn, Pb geogenic morphology b 6 pm Sub-rounded Si quartz particle 9 a 1 (am Rounded Al, Si, S,K, Ca, Mn, Fe Zn, Pb, Smelter source? b 1 urn Rounded Al, Si, S,K, Ca, Mn, Fe Zn, Pb, Smelter source? c matrix Si, S, K, Ca, Fe minor Zn, Pb geogenic mineral particle .10 a 1.5 |xm Sub-angular Si, S, K, Ca Fe minor Zn, Pb angular morphology, fugitive dust? b 0.25 nm Sub-rounded Si, S, Ca, Fe minor Zn, Pb c 0.5 |im Sub-rounded Si, S, Ca, Fe minor Zn, Pb 11 a 1.5 (iim Sub-rounded Si, S, K, Ca, Fe Pb, As(?) probable smelter particle b >15 nm sub-angular, matrix Mg, Al, Si, S, K, Ca, Fe geogenic mineral 12 a 0.5 (im sub-rounded Si, S, K, Ca, Fe minor Pb, Zn fugitive dust? 13 a 1 pm rounded Al, Si, S, K, Ca Fe minor Pb, Zn fugitive dust? b 1 pm rounded Al, Si, S, K, Ca Fe minor Pb fugitive dust? c 1 pm rounded Al, Si, S, K, Ca Fe minor Pb fugitive dust? 14 a 2.5 pm angular Mg, Si, S, K, Ca, Fe Zn fugitive dust b 2.5 pm rounded Al, Si, S, K, Ca, Fe minor Pb, Zn c 2.5 pm rounded Si, P, S, K, Ca Fe minor Pb, Zn 15 a 1 pm rounded Al, Si, S, K, Ca, Mn, Fe Pb, minor Zn slight 0 peak, smelter particle ? b 3 pm sub-angular Al, Si, S, K, Ca, Fe Pb, minor Zn fugitive dust? c 3 pm rounded Al, Si, S, K, Ca, Fe minor Pb, Zn 16 a 5 pm rounded Al, Si, S, K, Ca, Mn, Fe Pb, minor Zn Fe 3a Zn J is Zn U L 0.0 5.0 10.0 kfiV 15.0 20.0 3b Si Fe pSCa i W L M n / L 0.0 5.0 10.0 k f i V 15.0 20.0 Figure 4-6. Example of  a rounded particle of  anthropogenic origin containing Zn and Fe from the Cominco 4-5 cm sample (3a on Table 4-6). The geochemistry of  examples of other particulate matter observed within the frame  is indicated in spectras 3band 3c. Figure 4-7. Example of  a subangular particle of  geogenic origin containing Pb, from the Cominco 4-5 cm sample (4a on Table 4-6). Spectras 4b-d indicate the geochemistry of  other observed particulates within the frame. Table 4-6. Summary of  the observations from the Cominco 4-5 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures.  Elements present in high concentrations are listed in bold face  type Frame Spot Size (Approx) Morphology Major Elements Trace Elements Comments 1 a 4 pm Bulbous Si, K, Ca, Fe Sn b 4 pm Bulbous Si, K, Ca, Fe Sn 2 a 2 pm sub-rounded Al, Si, S, K, Ca, Ti, Mn, Fe Pb, minor Zn b 1.5pm sub-rounded Al, Si, P,S, K, Ca, Ti, Mn, Fe Pb, minor Zn c 5pm Surface coating Al, Si, S, K, Ca, Ti, Mn, Fe Pb, minor Zn precipitate d 10 pm sub-angular Al, Si, K, Ca, Fe Feldspar 3 a 3 pm rounded Al, Si, Ca, Fe Zn smelter-emitted b 0.5 pm rounded Al, Si, P, S, K, Ca, Mn, Fe Pb, Zn smelter-emitted? c 0.5 pm rounded Al, Si, P, S, K, Ca, Fe geogenic mineral matter 4 a 5 pm angular Al, Si, S, K, Ca, Ti, Mn, Fe Zn, Pb geogenic-fugitive dust? b 4.5 pm sub-angular Al, Si, P, S, K, Ca, Ti, Mn, Fe minor Pb geogenic mineral particle c 1.5 pm rounded Al, Si, S, K, Ca, Mn, Fe geogenic mineral particle d 8 pm angular Al, Si, S, K, Ca, Fe geogenic mineral particle 5 a 2 pm Coating? Si, S, K, Ca, Fe Pb Coating of (b)- PbS b 10 pm sub-rounded Si,K, Ca, Fe geogenic mineral particle 6 a 3 pm Sub-rounded Al, Si, S, K, Ca, Mn, Fe Pb, minor Zn smelter-emitted? b 0.5 pm rounded Al, Si, S, K, Ca, Fe minor Pb, Zn smelter-emitted? c 3 pm Sub-angular Al, Si, S, K, Ca, Fe minor Zn geogenic mineral particle 7 a 0.5 pm Sub-rounded Al, Si, K, Ca, Ti, Mn, Fe minor Cu, Zn b 2.5 pm Sub-rounded, Al, Si, K, Ca, Mn, Fe minor Zn, As smelter-emitted? c 2 pm Sub-rounded Al, Si, K, Ca, Mn, Fe Zn d 5 pm Sub-angular Al, Si, K, Ca, Mn Fe, minor Zn geogenic mineral particle Table 4-7 Summary of  the observations from the Cominco 17-20 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures.  Elements present in high concentrations are listed in bold face  type. Frame Spot Size (Approx) Morphology Major Elements Trace Elements Comments 1 a 1 pm Sub-angular Al, Si, S, K, Ca, Ti, Fe Pb fugitive dust? b 5 pm Sub-rounded Al, Si, S, K, Ca, Ti, Fe minor Pb, Zn geogenic mineral particle c 0.25 pm Rounded Al, Si, S, K, Ca, Ti, Fe Pb, minor Zn fugitive dust? or smelter-emitted? d 15 pm Sub-angular Al, Si, S, K, Ca, Ti, Fe Pb, minor Zn geogenic mineral particle e 2 pm Sub-rounded Al, Si, S,K, Ca, Ti, Mn, Fe Pb, minor Zn, Cu geogenic mineral particle? 2 a 4 pm coating? Al, Si, S, K, Ca, Fe Pb, minor Zn, Cu precipitate? b 7 pm Sub-angular Al, Si, S, K, Ca, Fe minor Pb geogenic mineral particle c 9 pm Sub-angular Al, Si, S, K, Ca, Fe minor Pb geogenic mineral particle d 2.5 pm Rounded Si, S, K, Ca, Fe minor Pb? geogenic mineral particle e 4 pm Sub-angular Al, Si, S, K, Ca, Fe minor Pb? geogenic mineral particle 3 a 0.25 pm coating? Al, Si, S, K, Ca, Fe Pb, As precipitate PbS? b 2 pm Sub-rounded Al, Si, S, K, Ca, Ti, Fe Pb, Zn mineral background of (a) c 6 pm Rounded, oval Al, Si, S, K, Ca, Fe geogenic mineral particle 4 a 2.5 pm Rounded Al, Si, S, K, Ca, Ti, Fe Pb, minor Zn, Cu fugitive dust? b 5pm Sub-angular Si, S, Ca, Fe quartz particle c 7 pm Sub-angular Al, Si, S, K, Ca, Ti, Fe Pb, minor Zn geogenic mineral particle 5 a 2 pm Rounded, Al, Si, S, K, Ca, Ti, Fe Pb, minor Zn b 1 pm Rounded Al, Si, S, K, Ca, Ti, Fe Pb, minor Zn c >10 pm Sub-angular Al, Si, K, Ca K-Feldspar particle d 0.5 pm Rounded Al, Si, S, K, Ca, Ti, Fe Pb, minor Zn 6 a 2 pm Sub-angular Al, Si, S, K, Ca, Ti, Fe Cu, Zn, Pb geogenic b 5 pm Sub-rounded Al, Si, S, K, Ca, Ti, Fe minor Zn geogenic mineral particle c 3 pm fluffy Al, Si, S, K, Ca, Fe Zn, Pb precipitate 7 a 1.5 pm Rounded Al, Si, S, K, Ca, Fe Pb, minor Zn geogenic fugitive dust? b 4 pm Rounded Al, Si, S, K, Ca, Fe mineral particle behind (a) c 12 pm Sub-angular Al, Si, S, K, Ca, Fe mineral particle Frame Spot Size (Appro x) Morphology Major Elements Trace Elements Comments 8 a 2 pm Rounded, fluffy Al, Si, S, K, Ca, Fe Pb, minor Zn geogenic fugitive dust? b 3 pm Sub-angular Al, Si, S, K, Ca, Fe minor Pb, Zn mineral particle c 1 pm Sub-angular Al, Si, S, K, Ca, Fe minor Pb, Zn geogenic fugitive dust? 9 a 4 pm Sub-rounded Ca, Fe Cu, Zn precipitate b 20 pm Sub-angular Al, Si, S, K, Ca, Ti, Fe minor Cu, Zn, Pb mineral particle with amorphous coating? c 20 pm Sub-angular Al, Si, S, K, Ca, Ti, Fe minor Cu, Zn, Pb mineral particle with amorphous coating? Figure 4-8. Fe-Zn-Cu precipitate from the Cominco 17-20 cm sample (9a on Table 4-7). Examples of geogenic particles are also pictured (9b and c). Si 6a Pb S b lr Fe J L C u Z n 0.0 5.0 10.0 15.0 keV 20.0 Si 6c J C a BUL 0.0 5.0 10.0 keV 15.0 20.0 Figure 4-9. Geogenic particle containing Pb, from the Cominco 17-20 cm sample (6a on Table 4-7). The EDX spectra of  other particulate matter observed within the frame  are also presented (6b and c) Table 4-8 Summary of  the observations from the Thunder Road 0-1 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures. Elements present in high concentrations are listed in bold face  type. Frame Spot Size (Approx) Morphology Major Elements Trace Elements Comments 1 1b 6 pm spherical Ca, Mn, Fe Cu, Zn, Pb smelter-emitted particle 3 a 12 pm angular Si, S, K, Ca, Fe Zn mineral particle, Fe-Zn coating? b >50 pm angular Si, S,K, Ca, Fe feldspar c 1 pm agglomeration Si, S, K, Ca, Fe Zn precipitate? ZnS d <1 pm agglomeration Si, S, K, Ca, Fe Zn precipitate? ZnS 4 a 2.5 pm Sub-rounded Si, S, K, Ca, Fe Zn native Zn? b >50 pm angular Si, S, K, Ca, Fe mineral particle behind (a) 5 a 2 pm fluffy Si, S,K, Ca, Fe Zn native Zn? from smelter b 30 pm Subangular Si, S, K, Ca quartz particle behind (a) 1 n 1 \ 1 ^ Pb Mm Cu A t\ L ^ P b C a J i J U ^ W l 0.0 5.0 10.0 15.0 20.0 keV Figure 4-10. Spherical metal enriched particle emitted from the smelter, from the Thunder Road 0-1 cm sample (4a on Table 4-8). 1 Fe C Si b 3b Fe L..L ... 0.0 5.0 10.0 15.0 20.0 keV c I > l l 3c Zn i . 0.0 5.0 10.0 kPV 15.0 20.0 Ca 3d Si S i l l Zn — A. — 0.0 5.0 10.0 kfiV 15.0 20.0 Figure 4-11. Examples of  Zn-containing particles from the Thunder Road 0-1 cm sample, a.: weathered geogenic Zn sulphide mineral, c and d: Fe-Zn precipitates (frame  3 on Table 4-8). An example of  the geochemistry of  other particulate material observed within the frame  is illustrated by spectra b. i ^ J / / 2|jm Figure 4-12. Example of  native Zn particle from the Thunder Road 0-1 cm sample (Frame 5, Spot a on Table 4-8), with a quartz particle (b). Table 4-9 Summary of  the observations from the Thunder Road 7-8 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures. Elements present in high concentrations are listed in bold face  type. Frame Spot Size (Approx) Morphology Major Elements Trace Elements Comments 2 a 3 pm sub-angular Si, S, K, Ca, Fe Cu, Zn geogenic fugitive particle b spherical Si, S, Ca, Fe fly ash 3 a <1 pm sub-angular Si, Ca, Fe Cu, Zn native Zn-fugitive? b 5 pm fluffy agglomeration Si, S, K, Ca, Fe Zn precipitate, amorphous c 0.25 pm fluffy agglomeration Si, S, K, Ca, Fe Cu, Zn precipitate, amorphous d 5 pm diatom Si, S, K, Ca, Fe Zn diatom, possible surface ZnS precipitate 4 a 1 pm spheroid Si, S, K, Ca, Fe Zn smelter-emitted particle b 5 pm spheroid Si, S, K, Ca, Fe fly ash c <1 pm sub-angular Si, S, K, Ca, Fe minor  Zn fugitive dust? d <1 pm fluffy Si, S, K, Ca, Fe minor  Zn precipitate y Ca 2 a 0.0 5.0 10.0 15.0 20.0 keV Figure 4-13. Angular geogenic Zn-rich particle (a) and a spheroidal fly  ash particle from the Thunder Road 7-8 cm sample (Frame 2 on Table 4-9). 3a Z n si c.a £Ma ).0 5.0 10.0 keV 15.0 20.0 Figure 4-14. Agglomeration of  Zn rich particles, including a subangular native Zn particle (a), Zn-rich precipitates (b and c) and (d) a diatom with Zn precipitated on the surface  from the Thunder Road 17-20 cm sample (Frame 3 on Table 4-9). Table 4-10. Summary of  the observations from the Champion Lakes 3-4 cm sample. "Frame" and "spot" refer  to the numbers on the EDX spectra for  the figures. Elements present in high concentrations are listed in bold face  type. Frame Spot Size (Approx) Morphology Major Elements Trace Elements Comments 1 a 18 pm Sub-angular Al, Si, S K, Ca, Ti, Fe mineral particle b 3 pm Sub-angular Si, Ca quartz c 15 pm Spherical Si, Ca fly ash 2 a 30 pm Angular Al, Si, Ca plagioclae particle b 40 pm Angular Al, Si, K, Ca, Fe feldspar particle 3 a 50 pm Angular Si, S, Zn geogenic sphalerite 4 a 10 pm Spherical Al, Si, K, Ca fly ash 5 a 5 pm Sub-angular Al, Si, Ca, Ti mineral particle 6 a 7 pm Spherical Si, Ca fly ash b 14 pm Sub-angular Al, Si, P, S, Ca, Fe mineral particle c 2 pm Sub-angular Si, Ca, Fe quartz particle d 4 pm Sub-rounded Al, Si, S, K, Ca, Fe mineral particle e 4 pm Sub-rounded Al, Si, S, Ca, Ti, Fe mineral particle 7 a 4 pm Angular Al, Si, K, Ca, Ti mineral particle 8 a 8 pm Sub-angular Al, Si, K, Ca, Fe quartz particle 9 a 4 pm Sub-angular Si, S, Ca, Fe Zn sphalerite particle Si 1 c u 0.0 5.0 10.0 keV 15.0 20.0 Figure 4-15. Diatom-rich ash material from the Champion Lakes 3-4 cm sample (e.g., b). Non-biogenic materials include an angular Ti-rich geogenic particle (a) and a glassy silicate fly  ash particle with slight Ca enrichment.(c) (Frame 1 on Table 4-10). Si 6 a _ sJl_JLe 0.0 5.0 10.0 keV 15.0 20.0 1 3i 6 d ll  L 0.0 5.0 10.0 15.0 20.0 keV S Ca I 6e J IfJl 0.0 5.0 10.0 keV 15.0 20.0 Figure 4-16. Geogenic particles containing Si, Ca, Fe, Al and S from the Champion Lakes 3-4 cm sample (spots a, c, d and e). Spot (b) is a glassy Si fly  ash particle with slight S, Ca and Fe enrichment. (Frame 6 on Table 4-10). 4.5 Discussion 4.5.1 SEM-EDX  Investigations  of  Smelter-Related  Media The identification  of  particles found  in the peat was aided by comparison to SEM-EDX investigations of  other media associated with the smelter, including smelter wastes and feedstocks,  particles captured by atmospheric deposition monitoring stations and stream sediments. Particulate matter within the peats has potentially undergone weathering and alteration in the environment. Comparisons with "fresh"  smelter materials, with the sampling media from an atmospheric monitoring project (Goodarzi et al., 2003), and with geologic materials such as stream sediments provides insight into the changes that occur in the smelter emitted particles following  their release into the environment. 4.5.1.1 Smelter  Wastes Goodarzi et al. (unpublished report) investigated the chemical and microscopic properties of  materials collected from the zinc baghouse facility  at the Teck-Cominco smelter. The particulate material typically contains Pb and Zn, with the occasional occurrence of  Cr, Ni, Cd, and As. The trace elements had several modes of  occurrence: as silicates or alumino-silicates with or without Fe and/or Mn, as oxides, or as sulphides. Particles were typically small (less than 5 pm), and subangular to subrounded. 4.5.1.2 Moss  Monitoring  Stations A network of  moss monitoring stations, installed to monitor the atmospheric deposition of  elements in the area surrounding the smelter, was set up in 1997 (Goodarzi et al., 2001; 2003). The moss is changed every three months and so particles trapped in the matrix are usually freshly-emitted  particulates that have undergone little weathering. Particles trapped by moss monitoring stations close to the smelter are often  small and rounded or "fluffy"  in morphology (Fig. 4-17). Fugitive dusts are generally angular and similar in chemistry to the ore (sulphides), while the particles emitted from the smelter are typically oxides, native elements or sulphates (Goodarzi et al., 2003). Few of  the particles from the peat profiles  that were identified  as smelter-emitted had prominent oxygen peaks in their corresponding EDX spectrum, unlike the particles from the baghouse and the moss monitoring media. However, the lack of  metal oxide particles within the peat should not be taken as evidence that the high trace element concentrations within the Cominco and Thunder Road profiles  are due exclusively to fugitive  and geogenic sources. The soluble oxide particles are likely to be rapidly altered following  emission to the environment. The pronounced preferential  sequestration of trace elements in the peats relative to the mineral soils that occurs near the smelter (Chapter 2) illustrates that a large proportion of  the trace elements emitted by the smelter are mobilised following  deposition. 4.5.1.3 Stream  Sediments Stream sediments from the area surrounding the smelter were examined (Goodarzi et al., unpublished report). The morphology of  particles depended on source: particles attributed to smelting activities were usually less than 5 pm in diameter and had a rounded or "fluffy"  morphology, while particles ascribed to fugitive  sources (dust from unprocessed ore) were more angular and variable in size. Geogenic particles commonly consisted of  quartz, microcline, albite, amphibole, mica, clinochlore and smectite. 4.5.2 Environmental  Significance The source of  trace elements in mining or smelting areas may have an control the potential for  health impacts on the inhabitants of  the surrounding area. An SEM-EDX examination of  particulate matter may be helpful  in determining the potential bioavailability of  trace elements. Rieuwerts et al. (2000) found  that Pb in garden soil from mining areas (sulphides) is less bioavailable than Pb in garden soil from the area surrounding a smelter (Pb-oxides and sulphates). Significantly  more Pb was found  in the house dust in the area surrounding the smelter, and blood-Pb levels were correspondingly higher. For this reason, geogenic and fugitive  dust sources of  Pb and Zn are of  less concern from a health perspective than the metal oxides and sulphates emitted by a smelter. Furthermore, the size of  the metal-containing particles may have important health and environmental implications. The fine  particles (<10 pm) emitted by the smelter are more likely to be inhaled and become lodged deep within the lungs. Small particles are more likely to "be widely dispersed in the environment by prevailing winds, and are more rapidly weathered than large particles due to a higher relative surface  area. For these reasons, heat treated smelter-emitted particles represent a greater potential threat to the local environment than the fugitive  dust and geogenic particles. ZnS 3.5*2.8 pm 0.8\1 mil \ V 1\1.5 pm (Columbia Gardens) Figure 4-17. Examples of  particulate matter captured by the moss-monitoring stations in the Trail area. a. Geogenic quartz particle, b. Rounded and "fluffy"  PbO particles formed  by thermal alteration in the smelter, and angular ZnS particles from fugitive  emissions, c. Angular ZnS particles derived from fugitive dust, collected near an ore and slag storage facility.  From Goodarzi et al., 2003. 4.6 Conclusions An examination of  the morphology and elemental composition of  particles found in peat ash from the vicinity of  a Pb-Zn smelter is a useful  method for  assessing the sources of  trace element pollution within the environment. The small, rounded Pb and Zn-containing particles found  in the peats near the smelter were likely emitted from the smelter following  roasting, while the more angular particles are likely fugitive  dust from unprocessed stockpiled ores, or natural locally derived geogenic particles. These findings suggest that reducing dust during transport and storage of  ores would be a valuable initiative in reducing the amount of  metals released from the smelter operation, although these particulates may represent a less bioavailable source of  trace elements than the processed particles. No oxides of  Pb or Zn were detected in any of  the peat samples examined, although they were observed in the moss monitoring sampling media. This suggests that these anthropogenic particles are subject to rapid weathering in the environment and are a major source of  mobile and potentially bioavailable trace elements to the surrounding environment. The lack of  anthropogenic particles containing Pb or Zn at the Champion Lakes site suggests that particulate deposition from the smelter is not a major source of  trace elements at this location. Low density Si-rich fly  ash was noted, indicating that deposition of  smelter-derived material does occur in the Champion Lakes area to a small extent, and it is possible that trace elements are deposited in the area in non-particulate forms,  for  example, as dissolved species in precipitation. Moss monitoring studies (Goodarzi et al., 2001; 2003) and a peat trace element study (Chapter 2) indicate that there is possibly minor deposition of  trace elements from the smelter at this location. 4.7 References Beddoe-Stephens, B. and. Lambert, R.St.J. 1981. Geochemical, mineralogical and isotopic data relating to the origin and tectonic setting of  the Rossland volcanic rocks, southern British Columbia. Canadian Journal of  Earth Sciences, 18: 858. Goodarzi, F., Sanei, H., and Duncan, W.F. 2003. Deposition of  trace elements in the Trail region, British Columbia; an assessment of  the environmental effect  of  a base metal smelter on land. Geological Survey of Canada Bulletin 573, 50 p. Goodarzi, R, Reyes, J. and Sanei, H. 2002. Concentrations of  elements in the stream sediments in Trail, BC and the surrounding area and the natural background concentration. Unpublished report. Hall, G.E.M., Vaive, J.E. and MacLaurin, A.I. 1996. Analytical aspects of  the application of  sodium pyrophosphate reagent in the specific  extraction of  the labile organic compounds of  humus and soils. Journal of  Geochemical Exploration, 56: 23-36. Little, H.W. 1960. Nelson Map-Area, West Half,  British Columbia (82 F Wl/2). Geological Survey of Canada Memoir 308. 205p. Little, H.W. 1982. Geology of  the Rossland-Trail Map-Area, British Columbia. Geological Survey of  Canada Paper 79-26, 38p. Miller, R.N., Yarzab, R.F. and Given, P.H. 1979. Determination of  the mineral-matter contents of  coals by low-temperature ashing. Fuel, 58: 4-10. Sevigny, J.H. 1990. Geochemistry of  the Jurassic Nelson plutonic suite, southeastern British Columbia. In  : Project Lithoprobe; southern Canadian Cordillera transect workshop. Lithoprobe Report 11, pp. 41-52. Chapter 5 Determination of  the mode of  occurrence of  trace elements in peats and soils from  the vicinity of  a Pb-Zn smelter using a sequential extraction procedure 5.1 Introduction A common approach to determining the potential environmental impact of  trace elements is by way of  sequential extraction. Sequential extraction procedures differentiate  between operationally-defined  phases in which the trace elements are present within the sediment. By determining the concentration of  trace elements within each phase, the mobility and bioavailability, and thus, potential environmental impact, can be inferred.  The mode of  occurrence can also be used to infer  the source of  the trace elements. The procedure can also be used to differentiate  between anthropogenic and geogenically-sourced elements to some extent. The concept behind the use of  sequential extractions is to treat a sample of  sediment with a series of  increasingly harsh reagents which have been experimentally shown to be selective to a particular phase (e.g., Tessier et al., 1979). Phases typically extracted may include: exchangeable (adsorbed to clays and organic matter); associated with carbonates; associated with reducible iron and manganese oxides, hydroxides and oxy-hydroxides; associated with organic matter; associated with sulphides; and residual elements within the crystal lattice of  mineral matter (e.g. Tessier et al., 1979). This study examines the mode of  occurrence of  selected trace elements (Sb, As, Cd, Cu, Pb, Ni and Zn) in profiles  of  peat and organic sediments from the area surrounding a lead-zinc smelter. The capacity for  peats to retain trace elements is of interest in studies in which the objective is to determine the fate  of  trace elements in the environment. This study builds on earlier work in which: the total trace elemental content of  peat profiles  in the vicinity of  the smelter was examined and discussed with reference  to measurements of  atmospherically deposited trace elements and trace element concentrations in nearby mineral soils (Chapter 2); geochemical tracer methods were applied to the peats in order to determine if  the relative impact of  the smelter on the trace elemental composition of  the peats could be deteremined by geochemistry (Chapter 3); peat ash materials were examined by SEM-EDX and compared to fresh  smelter wastes and feedstocks  (Chapter 4): and the peats were characterised using organic petrographic methods (Chapter 6). The results of  previous investigations are summarised briefly  in the results section. 5.2 Study Area The Teck-Cominco smelter in Trail, British Columbia has been in operation since 1896. The smelter's primary products are Pb and Zn, but Ag, Au, Cd, Bi, In, Ge concentrate, GeC>2, Q 1 S O 4 , CuAsO and sulphur products are also produced as by-products. Ammonium nitrate fertilizers  are also produced by Cominco in a separate plant located near the smelter. Improvements in technology and operating practices have reduced trace element emissions from the smelter in recent years, however, high concentrations of  trace elements measured in peats from the surrounding area record the cumulative impact of  a century of  operation (Chapter 2). 5.3 Sampling Locations Sampling sites were selected in order to obtain a representation of  the lateral distribution of  trace elements in the area surrounding the point sources. Considerations for  site selection included the distance and direction from the smelter, bedrock geology, and availability of  organic sediments. The latter posed a particular problem in the Trail area, as peatlands in that region are few in number. Trail, British Columbia (Fig. 5-1) is situated in the Columbia River Valley, which bisects the sampling area. The bedrock in the area consists mainly of  Upper Jurassic plutonic rocks (the Nelson Intrusives), and Lower Jurassic metavolcanic and metasedimentary rocks (the Elise Formation; Little, 1982; Hoy and Andrew, 1991). 5.3.1 Cominco  Site The site consists of  a series of  small fens.  The underlying bedrock is the Trail Pluton, consisting predominantly of  granodiorite, with quartz diorite and diorite at the margins (Little, 1982; Simony, 1979; Fig. 5-1). Vegetation consists of  Typha  latifolia (common cattail), with Epilobium ciliatum  (purple leafed  willow herb), Carex  urticulata (beaked sedge), Equisetum palustre  (marsh horsetail), and Deschampsia cespitosa (tufted hairgrass). The pH of  the peat, measured in the field,  ranged from 5.3 at the surface  to 5.6 in the underlying mineral sediment, and the pore water had a pH of  5.7. At the time of  sampling, the entire peat profile  was water saturated. It was noted in the sampling pit that features  associated with reducing conditions, such as gleying, were not present in the uppermost 15 cm of  peat, hence it is assumed that the top 15 cm are at least intermittently aerobic. This site is located approximately 0.5 km from the smelter and is known to be 117*46* 117°34' 49°15* QUATERNARY UnconsoBdated sediments; till, sand gravel, silt JURASSIC/CRETACEOUS NELSON INTRUSIONS: granodiorite; minor quartz diorite and diorite H ROSSLAND MONZONITE: biotite-hornblende-augite monzonite, mainly medium grained JURASSIC EUSE FORMATION: flow breccia, massive andesites and basalts, agglomerate, tuff, breccia, laminated siltstone MOUNT ROBERTS FORMATION: black siltstone and argillaceous - j quartette, slate, yeywacke, chert, pebble conglomerate, lava; Kmestone TRAIL GNEISS: amphibolite and grey biotite gneiss, hornblende gneiss, mica schist, aplite and Kilometres 1 El Figure 5-1. Geological map of  the Trail, British Columbia area, with sampling locations for  peats and stream sediments indicated. subject to heavy impact from the smelter (Goodarzi et al., 2003; Chapter 2). 5.3.2 Thunder  Road Site The site consists of  a group of  small fens  parallel to a small gravel road, with a high ridge of  granodiorite behind them. This site is also located on the Trail Pluton (Little, 1982; Fig. 5-1), and has granodiorite bedrock similar to that described for  the Cominco site. Vegetation is also similar to the Cominco site, with Typha  latifolia,  Pinus sp. (pine), Betula sp. (birch), Epilobium ciliatium, Equisetum palustre,  Carex  utriculata and Deschampsia cespitosa. The watertable was approximately 10 cm below the peat surface  at the time of  sampling. The pH ranged between 5.0 (surface)  to 5.4 (mineral sediment), and the pore water pH was 5.7. This site is located approximately 1.5 km from the smelter and is subject to significant  trace element deposition from the smelter, although to a lesser extent than the Cominco site, due to the larger intervening radial distance and partial topographic sheltering (Goodarzi et al., 2003; Chapter 2). The site is also subject to input from a fertiliser  factory,  which currently produces N-fertiliser,  but previously produced phosphate and potassium fertilisers  as well (W. Duncan, personal communication). 5.3.3 Champion  Lakes Provincial  Park The bedrock in the Champion Lakes area is the Lower Jurassic Elise Formation (Fig. 5-1), which is characterized by flow  breccia, massive lava, agglomerate, volcanic breccia, tuff,  tuffaceous  conglomerate, andesite, basalt and augite porphyry, metamorphosed to green schist facies.  Limestone xenoliths and calcite amygdules are found  within the mafic  flows.  Clasts from the tuffaceous  conglomerates are commonly limestones derived from the underlying Mount Roberts Formation (Little, 1982; Hoy and Andrew, 1991). The Elise Formation hosts most of  the gold and copper (pyrrhotite with chalcopyrite) vein deposits of  the Rossland area (Hoy and Andrew, 1991). Lead, zinc and silver vein deposits are also associated with the contact of  the Elise and the overlying Hall Formation. Associated minerals include sphalerite, galena, arsenopyrite, pyrite, chalcopyrite and boulangerite. This site is located in a provincial park, approximately 13.5 km from the smelter. The site is sheltered from deposition from the smelter due to the intervening mountainous topography, and due to the fact  that the prevailing winds are generally from the northwest (Goodarzi et al., 2003). Peat was sampled from a marsh surrounding the lake. The peat-forming  vegetation consists mainly of  Carex  sp., with Nuphar  lutea  (cow lily) and several species of  grasses. The water table was approximately 8 cm below the peat surface  at the time of  sampling. The peat pH was 6.5 at the surface  and at 15 cm depth, while the pore water had a pH of  6.4. 5.4 Methodology Peats were sampled by means of  digging pits with a stainless steel shovel, and cutting blocks of  peat from the pit wall. The peat blocks were immediately sealed in acid-washed plastic freezer  bags and refrigerated  within 24 hours. Measurements of  pH were taken in the field  using a portable pH meter. Peats were subsampled in a clean room, using an acid-washed plastic knife. Samples for  analysis were removed from the interior of  the peat block, so as to use material that was uncontaminated by the sampling procedure. Samples were selected to include peats from a range of  depths within each profile.  The peats were subsequently freeze-dried,  and ground to a fine  powder with an agate mill that was thoroughly acid-rinsed and washed with deionized water between each sample. Organic carbon and sulphur were determined using a LECO-SC444 carbon and sulphur analyser. Samples were treated with 10% HC1 for  24 hours to remove carbonate, and were subsequently rinsed with distilled water and dried at 80°C overnight. The sequential extraction procedure used in this research was developed from methods described in Tessier et al. (1979), Hall et al. (1996a,b), Land et al. (1999), Clark et al. (1997), and Kersten and Forstner (1987). After  reviewing the literature and considering the type of  sediment involved and the objectives of  the study, the following procedure was developed: 1 g of  sample was added to acid washed Nalgene ® centrifuge  tubes. Water soluble/exchangeable phase extraction: Due to very low recoveries in an initial batch of  sequential leaching, the water soluble fraction  was pooled with the exchangeable phase. The water soluble fraction  was extracted by adding 25 ml of  distilled deionised water to each sample. The samples were shaken for  1 hour, and were then centrifuged  for 15 minutes at 10000 rpm. The samples were then rinsed with an additional 10 ml of deionized distilled water, shaken briefly,  centrifuged,  and decanted. The decanted supernatant was then centrifuged  at 15000 rpm for  30 minutes to separate any suspended solids that were accidentally decanted earlier.; The supernatant was decanted into an acid washed Nalgene® HDPE sample bottle. The exchangeable fraction  was extracted by adding 25 ml of  1 M MgCl2 solution to each sample. The samples were shaken for  one hour and were subsequently centrifuged,  rinsed and decanted in the manner described for the water soluble fraction. Carbonate phase extraction: 25 ml of  1M Na-CH3COOH, adjusted to pH 5 with glacial acetic acid, was added to each sample. The samples were shaken for  two hours, and centrifuged  and rinsed with water as described above. Iron and Manganese oxide, hydroxide and oxy-hydroxides (reducible Fe and Mn) phase extraction: 25 ml of  0.1 M NH2OH HC1 was added to each sample. The samples were heated in a 80°C water bath for  2 hours, then were shaken for  an additional 2 hours. Subsequently, the samples were centrifuged  and rinsed in the manner described above. Organically-chelated ions: The samples were transferred  to 250 ml beakers. 100 ml of 1M NaP207 solution was added to each sample. The sample was shaken for  one hour, centrifuged  and decanted. The procedure was repeated. The samples were subsequently rinsed with deionized distilled water, as with the other extraction steps. Sulphide phase extraction: 50 ml of  30% H2O2 was added to each sample. The reaction is quite vigorous (effervescent)  when the samples contain a large amount of  organic matter, resulting in almost instantaneous "bubbling over" when the procedure developed for  mineral soils and sediments was followed  (heating to 80°C for  two hours). For this reason, this extraction had to be performed  in a number of  steps and over a longer period of  time than the peroxide extractions described in the literature for  non-organic sediments. The samples were gently swirled by hand to mix, and were then allowed to settle. When bubbling had ceased, the samples were shaken more vigorously, then allowed to settle. The samples were subsequently heated in a 40°C water bath. The samples were removed as soon as the bubbling became vigorous, and were allowed to cool. This procedure continued, until the reaction rate became observably slower. Throughout this procedure, the hydrogen peroxide evaporates, and thus further  additions of  H2O2 are necessary. When the reaction rate at 40°C slowed, the samples were heated to 80°C. The samples were heated continuously, with periodic cooling to prevent boiling over as necessary, with further  additions of  H2O2 until all bubbling had ceased (this procedure may take several days, depending on the organic content of  the samples). The samples were then centrifuged  and rinsed as described previously. Residual fraction: The remaining residue from each sample was treated with 10 ml of H F / H N O 3 / H C I solution (5:3:1 ratio), in a Teflon  microwave digestion vessel. Microwave digestion followed  the manufacturer's  recommendations for  soils (CEM Technologies, personal communication, 2001). Samples were digested until all residue was observed to be dissolved, or until no further  dissolution appeared to be occurring. Total concentrations were determined using microwave digestion of  homogenised whole samples in triple acid solution similar to the digestion of  the residual fraction, following  a pre-digestion stage in open vessels to allow volatile gasses evolved from the decomposition of  organic matter to be released prior to sealing in the microwave vessels. Extractions were performed  in sets of  7 with one blank per extraction step per set. Following extraction, all extracts were heated to dryness in a deionised-distilled water bath and redissolved in 20 ml of  2% high purity nitric acid (250 ml in the case of  the organic chelate fraction). ICP-MS analysis was performed  with a Ru internal standard and a high purity Seastar ® nitric acid matrix. Instrument calibration was performed  with a multi-element standard solution. Replicate analyses, at both equal and variable dilutions, analysis of two isotopes of  Ni, Cu, Zn, Sb and Cd, and blanks were used as a check on data reliability. The exchangeable, carbonate and organic chelate fractions  were analysed using standard additions to a sample of  the blank in order to account for  matrix interference  effects. When the sum of  the concentrations from the sequential leaching procedure is compared with the total digestion concentration, the percent error is variable. Percent error for  the three profiles  averages +/-35% but ranges between +/-0.7% to 100% for individual samples. This may be due to sample heterogeneity or to cumulative error associated with each leaching step. However, correlation coefficients  for  the summed concentration versus the total concentration for  each data set are high (0.71 for  Thunder Road, 0.91 for  Cominco and 0.96 for  Champion Lakes), indicating that the overall trends in the total concentration are reflected  in the summed concentrations. While sequential extraction methods have been shown to be helpful,  they are by no means infallible  (Kersten and Forstner, 1987; Martin et al., 1987). The results of  each extraction step can be considered to be operationally-defined,  rather than absolute, since extraction may be incomplete due to incomplete solublisation or to readsorption, or may inadvertently release metals associated with another fraction.  Individual elements may respond differently  to the extracting solution (e.g. Dominik et al., 1983; Luksiene and Shvedov, 1985). A major problem encountered with the use of  sequential extractions with anaerobic sediments such as peats is that changes in phase association may occur when the samples are oxidised during sampling, storage or drying. Forstner (1987) and Kersten and Forstner (1987) examined this problem experimentally, and found  that the majority of  the oxidative transfer  of  trace elements occurs between the organic and the reducible iron phases. For this reason, it should be assumed that at least some of  the trace elements measured in the reducible iron phase originated in the organic matter phase. However, knowledge of  the potential errors associated with the use of  these procedures enables data to be discussed in a useful  way. 5.5 Results 5.5.1 Summary of  Results  from  Previous  Studies Chapter 2 examines the total concentration of  selected trace elements of  the peat profiles  in the vicinity of  the smelter, and discusses these concentrations with reference  to measurements of  atmospherically deposited trace metals and trace metal concentrations in mineral soils near the peats and from background locations. Peats closest to the smelter have elevated concentrations of  Pb, Zn, Sb, As, Cd, and Cu, which decrease with increasing radial distance from the smelter. The peats are enriched in the aforementioned elements relative to the mineral soils, and the extent of  enrichment is relative to the extent of  element loading at the site. This finding  indicates that the elements deposited on the surface  of  the soil from atmospheric fallout  are subsequently solubilised and transported from the soil to the peat, where they are sequestered as a result of  the geochemical conditions found  within the peat. Nickel concentrations, however, did not decrease with increasing distance from the smelter, nor was Ni relatively enriched in the peat relative to the mineral soils. Based on these findings,  it was concluded that Ni was largely of geogenic origin. Selected samples of  the peat ash material examined by SEM-EDX were found  to have trace elements present in several elemental associations (Chapter 4). Heat-treated rounded particulate matter from the smelter was noted to contain Pb and/or Zn in association with Si, Al, and Fe. Angular particles of  sulphide minerals with high concentrations of  Pb or Zn were determined to be either fugitive  emissions from the smelter operation or detrital minerals. Both particle types occurred in the peats from the Cominco and Thunder Road sites, but were not detected in the Champion Lakes site. The only evidence of  anthropogenic impact in the Champion Lakes peat was the presence of spherical fly  ash particles composed of  Si-rich glassy material. Zn was also noted in geogenic particles. The organic petrographic composition of  the peats was determined by point counting in reflected  white and blue light (Chapter 6). While the practice of  using the maceral composition to classify  the depositional environment (e.g. swamp, bog, marsh or fen)  has come under criticism, the organic petrographic characteristics are useful  in determining the past and present conditions of  early diagenesis within the peat (e.g. Wiist et al., 2001). The peat from the Cominco profile  was well humified  with abundant fungi, indicating that conditions within this peat are frequently  aerobic. The Thunder Road peat was better preserved, with higher contents of  textinite (intact cell structure) and texto-ulminite (anaerobic gelification)  indicating less oxidizing conditions. The Champion Lakes peat maceral composition is indicative of  a relatively well preserved peat subject to some oxidation. 5.5.2 Sequential  Leaching Results The mode of  occurrence of  elements is largely a function  of  the Eh and pH conditions within the peat, and the source of  the trace elements. While there is a great deal of  variability in behaviour between the elements, generally, the reducible fraction  is more significant  as a sink for  elements in the uppermost, more aerobic portion of  the profile,  while the organic and residual fractions  are of  greater importance with increasing depth. 5.5.2.1 Cominco Site Total concentration trends follow  a similar pattern for  Sb, As, Cd, Cu, Pb and Zn (Fig. 5-2). The concentrations at the peat surface  are relatively high, followed  by a zone of  lower concentrations in the 8-11 cm sample. Concentrations for  the aforementioned elements increase sharply in the 14-17 cm sample, and decrease steadily in the deeper samples. Concentrations in the mineral basal sediment underlying the peat profile increase sharply relative to the overlying peat, they are low relative to the concentrations measured in the upper portion of  the peat profile.  Concentrations of  Ni are relatively consistent throughout the profile.  Organic carbon ranges from 20.9% at the surface  to 2.9% in the basal mineral sediment, and total sulphur ranges from 0.04-0.32% (Table 5-1). The exchangeable fraction  is an important sink for  Ni, As, Zn, Cd and Pb (Figs. 5-3 and 5-4). The relative importance of  this sink increases with profile  depth for  Ni, As and Pb, while the Zn exchangeable fraction  is consistent throughout the profile  depth (3.6-10.6%), and the Cd exchangeable fraction  fluctuates  throughout the profile  (7.5-44.6%). The carbonate fraction  is a significant  sink for  Pb (8.6-19.5%) and to a lesser extent, Zn (1-7%) and Cd (2-7%), while generally contributing less than 4.5 % of  the total element concentration for  Ni, Cu, As and Sb. The relative proportion of  carbonate-associated elements decreases with increasing peat depth for  Zn and Cd and to a lesser extent for  Pb, although the absolute concentration follows  overall concentration trends, with a spike in the 14-17 cm sample. The reducible Fe and Mn fraction  is the major sink for  Pb, Zn, As and Cd (19-72%) throughout the profile  depth, and a major sink for  Ni, Cu and Sb in selected samples. There is a relative and absolute concentration spike of reducible-fraction  associated elements in the 14-17 cm sample, coinciding with the increase in total concentrations. Typically <20% of  the total Pb, Sb, Cu and As occur within the organic chelate fraction,  and the highest relative and absolute concentrations coincide with the total concentration spike at 14-17 cm depth. The sulphide fraction  represents approximately 2-4% of  the As, Pb, Zn and Cu, and 7-10% of  the Sb and Cu at the surface  of  the peat. The relative percentage of elements extracted from the sulphide phase decreases with increasing depth to the basal mineral sediment. The relative percentage of  sulphide-associated elements increases abruptly in the basal mineral sediment for  all the examined elements. The sulphide fraction  represents 8% (Cd) to 48% (Cu) of  the total element concentration in the basal mineral sediment, while the peat immediately above (32-36 cm) contains little or no Concentration (ppm) M O) CO O O O o O O O O O Concentration (ppm) Pb Zn Figure 5-2. Total element concentrations within the Cominco profile.  BSed = basal mineral sediment. Sample % Sulfur %Carbon Cominco 2-3 0.29 20.88 Cominco 3-4 0.27 18.28 Cominco 8-11 0.21 13.37 Cominco 14-17 0.32 12.31 Cominco 32-36 0.01 0.41 Cominco Basal Sediment 0.04 2.91 Cominco Mineral Topsoil 0.05 6.44 Rossland Background Soil 0.00 0.05 Thunder Road 0-1 0.33 31.03 Thunder Road 3-4 0.36 28.62 Thunder 2-3 0.33 30.27 Thunder 5-6 0.35 26.74 Thunder Road 17-20 0.07 4.45 Thunder Road Basal 0.01 0.52 Sediment Champion Lakes 2-3 0.97 31.96 Champion Lakes 3-4 0.85 29.67 Champion Lakes 4-5 0.86 34.24 Champion Lakes 35-36 1.52 39.78 Champion Mineral soil 0.06 13.50 Table 5-1. Organic carbon and total sulphur concentrations of  selected samples. sulphide-associated trace elements. The residual phase is the major mode of  occurrence for  Ni, Cu and Sb. The relative proportion of  Zn, Cd and Pb associated with the residual phase is variable throughout the depth of  the profile,  but is negligible in the basal mineral sediment, (<3%). Trace elements associated with the silicate particulates emitted by the smelter are most likely extracted in the residual fraction,  and so the residual fraction  is not necessarily exclusively representative of  geogenic materials. 5.5.2.2 Thunder  Road Total concentrations of  trace elements are lower at the Thunder Road site than at the Cominco site (Fig. 5-5). Concentrations of  Pb and Zn are highest at the surface  of  the peat and decrease with depth, as do concentrations of  As, Sb and Cd, although to a lesser relative extent. Concentrations of  Cu and Ni increase with increasing depth in the peat. All elements are present in lower concentrations in the basal mineral sediment than in the JQ D. O •o o O IO o 10 r- r CN N o J2 cn ro CO <*> a) W to S CD CO in o 10 •«- CM CM 8 / \ / \ / X • .,•« : i . • t - ,. '^rr^t I  . 10 o w ^ CM CM s J2 "5 c/J (5 « n ^ m CD in < w o m o m o (£-ornco T-T-CMCM tocoojig^ </) ^ CO CQ A CO <D <D jn'sz ja a> Q-0 .2 a)Oc/D_ c5™-- ro to o •C-Q.^  as ca-2 O J^ rj) rn W X (0 O c ' c '© UJOQTOOQi Mil sample depth (cm) Figure 5-3. Absolute concentrations (in ppm) for  sequentially extracted fractions  in the Cominco profile,  including the basal mineral sediment (BSed). Concentrations for  the Cominco mineral soil (MS) and the Rossland background soil (BG) are included below the results for  the peat profile. Figure 5-4. Relative concentrations (%) for  sequentially extracted fractions  in the Cominco profile,  including the basal mineral sediment (BSed). Relative concentrations for  the Cominco mineral soil (MS) and the Rossland background soil (BG) are included below the results for  the peat profile. W l Depth (cm) Depth (cm) W M -»• M O Ji i i i 0) w s m w w ffl  10 o ( D « I 0 Q. 0) W CO M i i rv* N) o rv* o cv* 0) o (V-00 o 0s o o o o cn o oo o o o SSI Depth (cm) Depth (cm) 0 3 , I -vi i 00 73 (D </> CL C 0) 0 0 CQ 0) CO c "D 3" d CD CO 0) 5 o' O zr CD_ 0) 0 70 CD a c o cr CD W M i i k W 1 0 R 4- i i I " I 1 o 0) o 3 0) <-t-CD m X o •y 0) 3 (Q CD 0) d ; CD Ni O A o 0s 00 o O o 03 0 CO ° D 0) 0-03 w N) (/) N) O ® u ro a 0) w m i 00 I >1 ^ r r i 00 W N) •l W 1 1 N) O A O 0) o oo o 0s o o U 0 a overlying peat. Organic carbon ranges from 31 weight % at the peat surface  to 0.5% in the basal mineral sediment, and total sulphur ranges from 0.33 to 0.01 % (Table 5-1). The exchangeable fraction  is a significant  sink for  Ni, Zn, As, Cd and Pb, (Figs. 5-6 and 5-7). Patterns of  distribution with profile  depth vary between elements: exchangeable Ni is particularly enriched in the 2-3 cm sample, while exchangeable Zn, As and Cd distributions are variable throughout the profile  and the relative proportion of exchangeable Pb increases with increasing profile  depth. The carbonate fraction  is a significant  sink for  Pb, Zn and Cd (2-25%), and a minor sink for  As, Ni and Cu (up to 4%). The relative percentage of  element sequestration associated with the carbonate fraction  decreases with increasing profile  depth for  Zn and Cd, but increases with depth for  Pb at the expense of  the organic chelate fraction.  The reducible fraction  is a major sink for  Ni near the surface  (up to 27%),  and for  Zn, As, Cd and Pb throughout the profile (17-76%o). The organically chelated fraction  is a major sink for  Pb and As, and a minor sink for  Sb. The percentage of  As and Pb sequestered in the organic fraction  decreases with increasing profile  depth. The sulphide fraction  is significant  for  Ni, Cu, Zn, As, Sb, and Pb at the surface and in the basal mineral sediment. The relative percentage of  Ni, Cu, Zn, As, Cd and Pb associated with the sulphide fraction  decreases with increasing profile  depth, to a minimum in the 19-20 cm sample, and then sharply increases in the underlying mineral sediment, similar to the Cominco profile.  The absolute concentrations of  trace elements in the sulphide fraction  are also significantly  higher for  Ni, Cu, Zn and Pb, Concentration (ppm) - i N) W ^ o o o o o o o o o Concentration (ppm) • -k N3 K> cn o cn o cn o o o o o o o o o o o -A-Sb — Cd -•-Cu -4-Ni Pb Zn Figure 5-5. Total element concentrations within the Thunder Road profile,  including basal sediment (Bsed). c N -Q CL 3 o 73 o (A < A W © jD 1o jc xi <D .9-® m -£- 3 ® ® ©Oc/D_ j=?SS:9 o o eg ro o " " E c - g -C.O.2 co co~ O ^ ^ OJOJC x CO 0) " tl Q) u j o c e o o o ; HIHt Concentration (ppm) Figure 5-6. Absolute concentrations (in ppm) for  sequentially extracted fractions  in the Thunder Road profile,  including the basal mineral sediment (BSed). Figure 5-7. Relative concentrations (%) for  sequentially extracted fractions  in the Thunder Road profile,  including the basal mineral sediment (BSed). J^- CD LO O - o 6 cvi co to Y $ ? 2 m (LUO) L|;d9Q CO Tt CD LO CM CO 10 T T— (LUO) mdaa CT\ Sb 0% 20% 40% 60% 80% 100% E o CL CD Q 0-1 2-3 3-4 5-6 14-15 19-20 BSed E o Q. CD O 0-1 2-3 3-4 5-6 14-15 19-20 BSed -+- I I I 4 xxwmmm 1 A l i i l M m l l l W i M i l l l l l l l l l -j vmmmmmmmmmMMmimimmwM Pb 0% 20% 40% 60% 80% 100% i i i i i mmrnrnmmmmmmm -W///////////M -J mm/mm J YmmMtTtTtrtfTxim 1 ----wmm wmrnm 'mmk^kmm 1 J mmrnrn WMtmmm 1 0% 20% 0-1 2-3 3-4 5-6 14-15 19-20 Cd 40% 60% 80% 100% ___ \-------wmmmmm --WZ/M/y/y/MAVA — - i ~ WMyyyyyyyyyyyyyyyyyi\ l"~'T i I — W ^ m m M m m m m m W M W M M M f m E x c h a n g e a b l e Carbonate Reducib le Organic Che la te Organic /Sulphide Residual and slightly higher for  As and Cd in the basal sediment than in the overlying peat, despite the overall lower concentrations of  both trace elements and sulphur in the basal sediment. The residual fraction  is the main association for  Ni, Cu, and Sb, and a major mode of  occurrence for  Zn and Pb. The relative contribution of  the residual phase for  Ni, Cu, Pb and Zn increases with increasing profile  depth, but decreases in the basal sediment with a subsequent increase in the sulphide fraction.  The Sb residual phase is relatively constant throughout the profile. 5.5.2.3 Champion Lakes Total trace elemental concentrations are much lower in the Champion Lakes profile  than in either the Cominco or Thunder Road sites. Both Pb and Zn concentrations are relatively high in the 2-5 and 14-15 cm samples (Figure 5-8). All of  the examined elements, with the exception of  Ni and Cu, decrease in concentration with increasing profile  depth (Fig. 5-8). Organic carbon ranges from 29.7 to 39.8 weight %, and total sulphur ranges from 0.85 to 1.5 weight% (Table 1). To better compare with the more impacted sites which have higher total trace elemental concentrations, extraction fractions  with elemental concentrations of  less than 1 ppm are not discussed as "significant",  although the relative percent contribution of  the fraction  to the total concentration may be large (Figs. 5-9 and 5-10). The exchangeable fraction  is a significant  sink for  Ni throughout the profile  (8-39%). A minor carbonate and organic chelate association is noted for  Zn and Pb at the profile  surface  (1.5-8.5%). The reducible fraction  is a sink for  Zn, As and Cd at the peat surface  (16-21%) and for  Zn throughout the profile  depth (up to 40 %). Sulphides are relatively more important in the Champion Lakes profile  than in the Cominco and Thunder Road profiles,  although overall concentrations are much lower. Sulphur concentrations are higher in the Champion Lakes peat than in the Cominco or Thunder Road peats and thus there is a greater pool of  sulphur for  the trace elements to react with. The major proportions of  Ni, Cu, and As, Sb and Pb at depth are found  within the sulphide fraction. The residual fraction  is a significant  mode of  occurrence for  Ni, Cu, As, Sb, Cd and Pb, indicating a probable significant  geogenic contribution to the geochemical makeup of  the peat. Concentration (ppm) (D g. 40; <D T3 r+ IT Concentration (ppm) N> CO Sb -As -Cd Cu -Ni Pb Zn Figure 5-8. Total element concentrations within the Champion Lakes profile. 1?-JE R Z B 4 o -o 1?0|-n ? 80-40-10 8 6 c/>£ 4 2 0 S w oo ca XI CO ni a> iB © 2>(0 5 c c ~ (OO " -5-e -a x to a) HI o an © 0 TJ JO 1c a) .a. -C 3 O CO _ o "Q co ' c "c -5 CO CO s O O Q f l i n n Sample Depth (cm) Figure 5-9. Absolute concentrations (in ppm) for  sequentially extracted fractions  in the Champion Lakes profile,  and for  a nearby mineral soil (MS). Figure 5-10. Relative concentrations (%) for  sequentially extracted fractions  in the Champion Lakes profile  and a nearby mineral soil (MS). Depth (cm) Depth (cm) --J CD 00 ro i. CD CO CD cn ro ro i 00 CD cln clo ro i X O o CD CD cn cn ! I I i CD •si O) ^ U M - i _ (D CO (D Ul N) ^ ^ i i i i i i ^ 00 O) Ol W N) - 1 1 / 1 O ^ O O) ffl  Ul IV ) o -fv CD N 3 CD CD OO O O CD u ro CD cn NJ i ro cn co to 0 0 ) 0 ) 0 1 0 1 CD -t^  CO CO CO CO cn CD CO (1) UI IJO -P» ^ 00 CD (Jl w M A C/D q o © O) Oi N3 CD -H* CD o C CD CD OO CD C D C D ON "-J Sb 40% 60% E o Q. a) Q 100% 0% 20% Cd 40% 60% 80% 100% 2-5 14-15 22-26 35-36 49-50 63-64 79-80 MS -I mwmm& - u wniwwiai i m m - T i a v . - . v rmzFTXxx; ' i - - v//""//'///////////// r WV S/////////. c w i i i i i 0% 20% Pb 40% 60% 80% 100% E o a. a> Q 2-5 22-26 49-50 MS E i h 1 1 i 1 1 mmmmmtm -1 1 vmmmtmtm^ WMwmmsm \----mmmmmmmmmi: Residual 5.5.2.4 Comparison to results  of  sequential  leaching from  soil and  stream sediment samples Samples of  mineral topsoil from the Cominco and Champion Lakes sites and from a background site located near the town of  Rossland were sequentially extracted following  the procedure used for  the peats in order to determine if  differences  in mode of occurrence exist between mineral and organic soils. The extraction data for  the Cominco and background soils is presented in Figures 5-3 and 5-4, and the Champion Lakes data is presented in Figures 5-9 and 5-10. Total concentrations for  trace elements in the Cominco site soil were typically lower than those measured in the surface  samples from the peat profile,  with the exception of  Ni and Sb, which were similar. Trace element concentrations in the Rossland soil were much lower than the Cominco site soil, with the exception of  Ni, which was higher in the background soil (Fig. 5-3). Organic carbon and sulphur contents are 6.4 weight % C and 0.05 weight % S for  the Cominco site and 0.05 weight% C and 0 weight% S for  the background soil, respectively. The total concentrations of  Cd, As and Sb in the background soil are less than 2 ppm each, and extraction recoveries of  these elements do not exceed 1 ppm for  any fraction.  Concentrations of  trace elements in the Champion Lakes soil are similar to those found  in the surface  peats for  Cd, Cu, As, and Sb, higher for  Ni, and lower for  Pb and Zn. The organic carbon content of  the Champion Lakes soil is 13.5%, and the sulphur content is 0.06%. The mode of  occurrence of  trace elements in the mineral soils differs  somewhat from that seen in the peats. Nickel, Cu, Zn, As, Cd and Pb are associated with the sulphide fraction  in both the Cominco and the background soil to a greater extent than in the peats. Sulphides are also a significant  mode of  occurrence for  the aforementioned elements in the Champion Lakes soils, and absolute concentrations of  trace elements in the sulphide fraction  of  the soil are equivalent to those in the surface  peats. The sulphide component of  the background soil is surprising, given that sulphur was not detected by the sulphur analyser. This result may be due to sample heterogeneity, or the inability of the sulphur analyser to decompose the sulphur species present within the soil. Alternatively, this result may indicate that the peroxide extraction is not specific  to sulphides. Carbonates are an important sink for  Pb, Zn and Cd in the Cominco and Rossland soils, and a minor sink for  Pb in the Champion Lakes soil, similar to the peats. Pb and As were measured within the reducible fraction  of  the Cominco soil at similar percentages to the peat, while the reducible Cd and Zn fractions  represent a smaller percentage of  the total element concentration than in the peats. The reducible fraction  is a major sink for Pb, Zn and Cd in the Champion Lakes soil, and a minor sink for  As. The relative importance of  the reducible fraction  is greater in the Champion Lakes mineral soil than in the peat (e.g. 53% of  the Pb in the soil is found  in the reducible fraction,  versus 6% in the surface  peat sample). The reducible fraction  is a relatively minor sink in the background soil from Rossland. The background soil contains little organic matter and so has little capacity to sequester metals in the organic fraction.  The Cominco site soil has higher organic C, and the organic fraction  is a sink for  As, Cu and Sb (10-14%). The organic fraction  of  the Champion Lakes mineral soil retains Pb, As, Sb and Cu (8-11%). The residual fraction  is a major contributor to the total Ni, Zn, Sb, Cd and Pb in the background soil from Rossland, while the residual fraction  from the Cominco site soil is a major mode of  occurrence only for  Ni, Zn, and Sb. The Champion Lakes mineral soil residual fraction  is the largest sink for  Sb, and an important fraction  for  Ni, Cu, Zn, As, and Pb (11-28%). Surprisingly, the residual fraction  of  the surface  peat is a larger contributor of  all the examined trace elements than the residual fraction  in the mineral soil, in terms of  both relative and absolute concentrations. The reason for  this finding  is uncertain. In summary, trace elements within both the Cominco soil and the background soil are concentrated in the sulphide and residual fractions,  and are not sequestered by the reducible, carbonate exchangeable or organic fraction  to the same extent as in the peats. Both of  these soils are sandy and contain little organic matter, and so have little capacity to retain trace elements mobilised by weathering from particulates or deposited as ionic species in precipitation. The Champion Lakes mineral soil is an organic-rich topsoil from a forested  area, and is capable of  retaining mobile species through a variety of mechanisms. 0% -t—• Hanna Creek 20% 40% 60% 80% 100% Sb As Cd Cu Pb Zn +-S b b b mmwam _ T - . W ^ K —vs/m 3 W/Zmv-A mmm o% 20% Ryan Creek 40% 60% 80% 100% Topping Creek 20% 40% 60% 80% Blueberry Creek 40% 60% 80% 100% t = j Exchangeable g g Carbonate HH Reducible Organic Chelate Organic/Sulphide Residual Figure 5-11. Relative concentrations (%) for  sequentially extracted fractions  in selected stream sediments. Stream sediments were sequentially leached following  the same procedure as was used for  the peats. Sediments from four  localities were examined: Upper Trail Creek, located near the Thunder Road site; Lower Ryan Creek, located near a zinc concentrate unloading site, which has been shown to be subject to high trace element inputs from fugitive  emissions (Goodarzi et al., 2G03); Upper Blueberry Creek, located north west of the smelter; and Middle Hanna Creek, located between Trail and Blueberry Creeks (Fig. 5-1). The Ryan Creek sample has higher concentrations of  sulphide-associated Cd and Zn relative to the other streams, due to the impact of  the ore concentrate unloading site (6% Cd, 13% for  Zn; Fig. 5-11). However, the relative contribution of  other elements in the sulphide fraction  is minor at Ryan Creek, and negligible at the other sites. Antimony, As, and Cu are found  mainly within the residual fraction  at all sites, reflecting  the geogenic nature of  these elements, or at least their non-bioavailability in stream sediments. Cadmium exists primarily in the exchangeable fraction  (>41%). Cadmium, Zn and Pb are also found  in the carbonate fraction,  particularly in the Hanna, Ryan and Trail Creek sites (9-17% for  Pb). Reducible Pb, Zn, Cd and As are important modes of occurrence at all locations. 5.6 Discussion 5.6.1 Controls  on Speciation 5.6.1.1 Adsorption The adsorption (ionic bonding) of  elements to surfaces  is a function  of  pH, the charge concentration of  the adsorbing substrate and the concentration of  the charged ionic species (e.g. Elder, 1988). Adsorption behaviour of  a given ion to a given surface  is strongly influenced  by pH, with the relative percentage of  adsorbed species versus pH being described by a steep "adsorption edge" Langmuir-type isotherm over a short range of  pH. The pH at which this isotherm occurs is variable, depending on the properties and concentration of  the adsorbed element and the availability of  sites for  adsorption. It has also been shown that organic substances and oxyhydroxides of  Fe and Mn may coat mineral particles, thus altering the absorptive capacity of  the minerals (Davis, 1984). The expected adsorption behaviour may also be disrupted by competing reactions in heterogeneous systems (Jacquier et al., 2001). For these reasons, it is difficult  to predict the adsorption behaviour of  elements in natural systems. Nickel, Cd, Zn, and As were found  in the exchangeable fraction  in all profiles, while exchangeable Pb and Sb were only measured in selected samples. Exchangeable Cu is not an important fraction  in any peat, soil or sediment. Forstner and Haase (1998) point out that the metalloid As, unlike the heavy metals, commonly exists as an uncharged species in natural waters through a range of  Eh and pH conditions, and so is not as susceptible to immobilization by sorption. However, As was removed in the exchangeable fraction  in most samples, indicating that the As was present either as an adsorbed or water soluble species. Adsorbed elements are held to surfaces  via ionic bonding, and are highly susceptible to desorption due to changes in pH or displacement by other charged species. For this reason, the exchangeable fraction  can be considered to have a high potential mobility. 5.6.1.2 The  Carbonate  Fraction Carbonate ligands maybe produced as a result of  dissilimilatory sulphate reduction, according to the equation 9CH20 + 4S042" + 4Fe(OH)3 + 4Me2+ - » 4FeS + 4MeC03 + 15H20 + 5C02 (Berner, 1984; Clark, 1998). Carbonate ligands may also be present in solution as a result of  inorganic processes, such as the dissolution of  carbonate minerals, and from dissolved atmospheric CO2. Carbonates are an important sink for  Pb, Zn and Cd in the peat profiles  and soil samples, indicating reprecipitation of  mobile ions of  these elements. Based on Eh-pH diagrams, the weakly acidic to neutral conditions in the peat favour  the formation  of PbC03 relative to Cd and Zn carbonates (Brookins, 1988), in agreement with the observations. Gee et al. (2001) determined that PbCC>3 is soluble below a pH of  5, releasing the Pb to solution, suggesting that the carbonate fraction  would not act as an effective  sink for  Pb in more acidic peats. The importance of  the carbonate fraction  as a sink for  Pb, and to a lesser extent, Zn, and Cd, in a variety of  environments has been noted by numerous authors. Lead carbonate was noted as a weathering product on the surface  of  Pb shot from a shooting range (Rooney et al., 1999). Carbonate species were determined to be an important mode of  occurrence for  Cd, Zn and Pb in soils around a modern Pb-Zn smelter (Verner et al., 1996), and in soils contaminated by mining wastes (Song et al., 1999). Cerussite was determined to be the major mineral association for  Pb at historical smelter sites, and an important control on Pb mobility (Maskall et al., 1996; Gee et al., 1997; 2001; Lee and Thornton, 2001). Carbonate-complex formation  has been shown to be an important reaction in anaerobic constructed wetland environments (Barton and Karathanasis, 1998) and in anaerobic soils (Charlatchaka and Cambier, 2000). 5.6.1.3 The  Reducible  Fraction In agreement with the findings  of  the current study, the reducible Fe/Mn fraction has been shown to be an important sink for  trace elements in peats (Bendell-Young, 1999; Bendell-Young et al., 2002), mineral soils (Karczewska, 1996; Verner et al., 1996; Astrom, 1998; Rieuwerts et al., 2000; Lombi et al., 2000; Palumbo et al., 2001; Lee and Thornton, 2001; Davranche et al., 2003) and sediments (Clark et al., 1998; El Bilaili et al., 2002; Lee et al., 2002; Mucci et al., 2003). The reducible fraction  is a significant  sink for  As, Zn, Cd and Pb in the heavily-impacted Cominco and Thunder Road peats, and for As, Zn and Cd in the Champion Lakes peat. The reducible fraction  is also an important sink for  these elements in mineral soils and stream sediments. The relative significance of  the reducible fraction  is typically highest in aerobic environments, such as the mineral soils and the Cominco profile.  However, it should be noted that the peats were dried prior to treatment, and so some of  the operationally-defined  reducible fraction  may have been attributable to the organic or sulphide fraction  within the in situ peat (Forstner, 1987; Kersten and Forstner, 1987), particularly in the case of  peats that are predominantly waterlogged, such as Champion Lakes. The precipitation of  reducible Fe and Mn compounds is controlled by both biotic and abiotic factors,  including pH, Eh, (Brookins, 1988; Kirby et al., 1999; Lee et al., 2002) and the presence of  iron-oxidising bacteria (Kirby et al., 1999). Metals are retained by Fe-Mn oxide complexes both by adsorption to negatively charged sites on Fe-Mn oxide surfaces  and by isomorphous substitution during precipitation (Herbert, 1996). Palumbo et al. (2001) noted significant  enrichment of  Pb, Cd, and Ni in Fe-Mn nodules found  in alfisolic  soils, and found  that Cd, Cu, and Ni show a strong affinity  for  Mn, while Pb shows a strong affinity  for  Fe. This is in accordance to the findings,  of  Lee et al. (2002), who found  that the adsorption of  Pb coincided with the precipitation of  Fe-oxide species at pH s4, while Cu, Zn, Ni and Cd were adsorbed at higher pH's, coinciding with the precipitation of  Al-oxide and Mn-oxide species. Bendell-Young (1999) found  that sorption of  Zn was associated with Fe-oxides, rather than Mn oxides, in peats from a range of  environments. 5.6.1.4 The  Sulphide  Fraction The formation  of  sygenetic sulphides occurs under anaerobic conditions, as a by-product of  the dissimilatory metabolism of  sulphate-reducing bacteria. These bacteria use organic oxysulphur compounds as terminal electron acceptors in their respiration processes, generating HS~, which reacts with ferrous  iron to form framboidal  pyrite (Altschulter et al., 1983), according to the following  reactions (Tuttle et al., 1969; Altschulter et al., 1983; Southam et al., 2001) CH20 + S042" H2S + 2HC03" H2S + Fe2+ FeS + 2H+ FeS + S° FeS2 Iron (III) is also reduced to Fe (II) by bacteria as a by-product of  dissimilatory metabolism. This process is enhanced in the presence of  organic matter, which acts as an "electron shuttle" between the microbes and the surface  of  the Fe-bearing mineral (Fredrickson et al., 1998). Iron (III) reduction has been shown to result in the mobilisation of  associated trace elements (e.g. Charlatchka and Cambier, 2000; Davranche et al., 2003) in open systems where leaching is extensive and other complexing ligands are not present. However, in a "semi-closed" hydrological system such as a peatland, it is likely that trace elements liberated by the reduction of  iron oxides will be retained as sulphides or organic complexes, and hence be retained within the peatland. Arsenic associated with Fe complexes in estuarine sediments was found  to be rapidly scavenged by newly-formed  Fe complexes following  changes in redox conditions. This occurred both in the case of  anaerobic Fe (II) monosulphides that were rapidly oxidized to form Fe oxyhydroxides (Saulnier and Mucci, 2000), and in the case of  aerobic Fe oxides that were rapidly buried, which retained As as Fe-sulphides (Mucci et al., 2003). It is probable that some redox-induced leaching occurs in the peat profiles.  The enrichment of  reducible fraction-associated  elements (Sb, As, Cd, Cu, Pb) in the Cominco profile  at 14-15 cm depth may have formed  due to leaching from the overlying sediments accompanying the dissolution of  Fe-oxides during periods of  flooding,  with reprecipitation as suphides under the reducing conditions found  at depth, followed  by conversion to Fe-oxides upon the re-establishment of  oxidising conditions. Such concentration "spikes" at redox transition zones were noted by Astrom (1998) and Clark etal. (1998). It has long been known that many trace elements found  in anaerobic sediments are associated with syngenetic pyrite and other sulphides (e.g. Finkelman, 1980; Casagrande, 1987; Miiller, 2002). In a study of  anoxic sediments from a Norwegian fjord,  Muller (2002) found  rates of  trace element coprecipitation with pyrite to be greater than 98% of the total "reactive concentration" of  Cu, Ni, Zn and Pb. Factors controlling the rate of formation  of  ferrous  sulphides include the availability of  sulphide, the availability of ferrous  iron, and the availability of  organic substrates for  dissimilatory microbial respiration. As Fe and S are quite ubiquitous in the environment around the smelter, it is likely that the main control on the formation  of  pyrite is the presence of  reducing conditions, and the main control on the degree of  coprecipitation is the extent of competing reactions. Elements may also be precipitated as non-ferrous  sulphides (e.g. ZnS). Sulphide-associated Zn, Cd, Pb, Sb, Cu and As make up a significant  portion of the total element concentration in soils and the surface  peats at the Thunder Road site and in the Cominco mineral soil. The relative contribution of  the sulphide fraction  to the total element concentration decreases with increasing depth in the Cominco and Thunder Road profiles,  particularly for  Zn, Pb and Cd. This suggests that some of  the sulphide-associated elements in the surface  peats and the mineral soils may be derived from sulphide-rich fugitive  dust from the smelter, rather than from syngenetic sulphides (Goodarzi et al., 2003; Chapter 4). Sulphides are unlikely to precipitate in aerobic mineral topsoils, and so sulphide precipitation is an improbable mechanism for  trapping mobile ions from solution in the mineral soil environment. Syngenetic sulphide formation  would be expected to increase with the increasingly reducing conditions found in the deeper peats. However, sygenetic pyrites are often  observed within plant cell structures, where reducing micro-environments are created as a result of  the decomposition of  plant materials (e.g. Altschuler et al., 1983; Cohen et al., 1983; Chague-Goff  et al., 1996). This suggests that trace elements may be coprecipitated with sulphides in sediments that are predominantly aerobic. Alternatively, it is possible that not all the organically-associated elements were removed by Na-pyrophosphate in the previous extraction step. Hall et al. (1996a) determined that two 1-hour extraction periods were sufficient  to extract the majority of  the Na-pyrophosphate extractable metals, although it is possible that some of  the organically-associated elements are not extractable by this method. The pattern of  increasing sulphide association with depth in the Champion Lakes profile  is contrary to the observations in the Cominco and Thunder Road profiles.  It is likely that the majority of  trace elements associated with sulphides precipitated from solution under the reducing conditions within the peat. The relative contribution of  the sulphide fraction  is minimal in the peat directly above the basal sediment in both the Cominco and Thunder Road sites. This suggests that the mobile trace elements are sequestered by another mechanism before  percolating to greater depth. The increase in relative and absolute concentrations of  sulphide-associated elements in the basal mineral sediments suggests either syngenetic coprecipitation of  anthropogenic and/or geogenic trace elements from groundwater which preferentially  flows  through the mineral sediment, or the presence of  geogenic sulphide minerals, or possibly a combination of  the two factors. 5.6.1.5 Organic complexation The retention of  metals by organic matter via complexation with negatively charged functional  groups, typically carboxyls, phenols, thiols or amines, as unidentate complexes or bidentate complexes (chelates) is an important factor  in the fate  of  trace elements in the environment. The potential for  trace metal complexation is a function  of the solution pH, the metal ion in question (Kerndorff  and Schnitzer, 1980; Stack et al., 1994; Cohen et al., 1995), the functional  group chemistry of  the organic matter, and the physical structure of  the organic matter (Sohn and Rajski, 1990; Wu and Tanoue, 2001; Leenheer et al., 2003). The fraction  of  organically associated elements is relatively small when compared to the reducible and carbonate fractions.  This may be partially due to the transformation of  organically-complexed elements to reducible Fe or Mn species as a result of  the oxidation that occurred during sampling and storage (Kersten and Forstner, 1987). The relatively minor contribution of  the organic fraction  to the total elemental concentration may also reflect  a greater tendency for  the examined trace elements to form Fe-oxide or carbonate complexes. Nonetheless, the organic fraction  is a major sink for  Pb in all profiles  and is a significant  sink for  Zn, Cd, Cu and Sb. The Cominco and Champion Lakes soils also contained organically-associated Pb, Sb, As and Cu, while the organic C-poor background soil from Rossland did not. Organic carbon is thus an important parameter to be considered when conducting trace element surveys of  mineral soils. 5.6.1.6 The  Residual  Fraction Trace elements extracted from the residual fraction  are believed to be predominantly geogenically derived, and are frequently  associated with silicate minerals. The residual fraction  also contains all the elements associated with other phases that were incompletely extracted previously (e.g. some crystalline sulphides and Fe-oxides may remain within the sediment). Silicate slags from the smelter may also be resistant to all previous extracting solutions. The residual fraction  is the major source of  Sb and Ni, and is a significant  mode of  occurrence for  Cu, Zn, Cd and Pb in the Cominco and Thunder Road profiles.  The relative contribution of  this fraction  typically increases with increasing profile  depth, indicating a probable partial geogenic source for  these elements, or at least increasingly limited mobility for  the trace elements in the deeper peats. The relative significance  of  the residual fraction  decreases to zero for  all elements other than Zn and Sb with increasing peat depth in the Champion Lakes peat. The relatively high concentrations of  Pb, Zn, Sb, and Cd at the surface  of  the profile  suggest deposition from the smelter, although a large proportion of  these elements are in the residual fraction,  which suggests a geogenic origin. The proposed geogenic origin for  the trace elements is supported by the SEM examination of  the peat, in which no trace element enriched particulate material from the smelter was observed (Chapter 4), and by the atmospheric deposition monitoring program, which measured little smelter impact at this location (Goodarzi et al., 2003). The increase in residual trace elements at the peat surface  may be due to increased input of  inorganic particulate matter due to enhanced erosion as a result of  anthropogenic activities. The lack of  residual-associated elements below 50 cm may be due to the gradual dissolution of  detrital minerals within the peat at depth (Bennett et al., 1991), or due to a lack of  deposition of  detrital materials at the time of  peat formation. 5.6.2 The  role  of  source material  in determining  mode of  occurrence There are several potential sources of  trace elements to the peats, which influence the mode of  occurrence. Heat-treated particulate material is released from the smelter during the course of  ore processing (Goodarzi et al., 2003; Chapter 4). These particles are typically oxides, sulphates and silicates enriched in Pb and Zn. Freshly deposited particles from the atmospheric monitoring stations observed by SEM frequently  had strong O peaks on the EDX spectra (Goodarzi et al., 2003). A similar SEM investigation of  the peat ash materials did not reveal this strong O signal in similar particles, suggesting that these heat-treated materials are subject to weathering and solublisation in the environment, in accordance with the findings  of  Davranche et al. (2003). These mobile elements are leached from the site of  deposition and migrate as soluble species until they become sequestered in the peats, in the exchangeable, carbonate, reducible, organic or sulphide fractions.  Carbonates and the reducible fraction  are also a major sink for elements within the soils, however, overall trace elements concentrations tend to be higher in the peats, particularly for  elements that do not readily form carbonate species (Cu, As). Elements which are not weathered from these silicate particles are likely to be extracted in the residual fraction. A second source of  trace elements in the area surrounding the smelter is fugitive dust released from ore feedstock  material during transportation and storage. This dust is composed primarily of  sulphides and the associated trace elements. Goodarzi et al. (2003) cite fugitive  dust from smelter feedstock  as the most significant  source of  trace elements in the area immediately surrounding the smelter, although stack-emitted materials make up a greater relative contribution with increasing distance. The significance  of  this fugitive  material in the peats from the Cominco and Thunder Road sites was noted in Chapter 4 in an SEM-EDX study of  peat ash materials. This suggests that a portion of  the sulphide fraction  from the Cominco and Thunder Road sites is derived from fugitive  dust, rather than sulphides precipitated in situ. Secondary sources may also contribute to the trace element concentrations and influence  their mode of  occurrence. Cadmium at the Thunder Road site is believed to be partially derived from past emissions from the fertiliser  plant, which previously produced phosphate fertilisers,  although only N-fertilisers  have been produced for  the past decade (W. Duncan, personal communication). The Cominco site is located in close proximity to a major roadway, and so may be subject to significant  input from automobiles (e.g. brake dust, and previously, Pb from gasoline), road dust and salt. There may be additional secondary sources that were not identified  (e.g. old mining spoils, undocumented landfills,  agricultural chemicals) and so it is difficult  to accurately determine the impact that these diffuse  sources may have on the concentration and mode of  occurrence of  the trace elements. A fourth  source of  trace elements is local geogenic input. The geology of  the Trail area is diverse and natural background geochemistry is known to be variable, with areas of  high grade sulphide-hosted ores (Beddoe-Stephens and Lambert, 1982; Little, 1982; Sevigny, 1990; Hoy and Andrew, 1989,1991). For this reason, it is not reasonable to assume a single value will represent the geogenic background for  the study area. Crystalline sphalerite was observed in the Champion Lakes peat by SEM-EDX, and several geogenic particles were observed to contain Pb and Zn, although these may have been surface  coatings, but no minerals associated with high trace elemental concentrations were detected by XRD (Chapter 4). Elements derived from local geogenic sources may be associated with the residual or sulphide fraction,  or may be found  within any of  the other fractions  as a result of  mobilisation during weathering and subsequent sequestration. 5.7 Conclusions Sequential leaching of  peat and soil samples subject to variable input from a Pb-Zn smelter illustrates that mode of  occurrence is a function  of  the geochemical conditions within the peat and the sources from which the trace elements are derived. In the heavily impacted peat profiles,  As, Cd, Zn and Pb are typically sequestered as exchangeable species, carbonates, Fe-oxides or organic complexes, indicating that these elements were most likely precipitated within the peat from solution, and may be subject to subsequent mobilisation, while Ni and Sb are commonly found  in more stable residual forms. Mineral soils, which are relatively depleted in mobile trace elements compared to the nearby peats, generally contain higher proportions of  trace elements associated with the recalcitrant residual and sulphide fractions,  due to leaching of  more labile species. Peats from a profile  subject to less deposition from the smelter contained a greater proportion of  residual and sulphide associated elements, due to the higher relative contribution of geogenically-derived materials. 5.8 References Altschuler, Z.S., Schnepfe,  M.M., Silber, C.C. and Simon, F.O., 1983. Sulphur diagenesis in everglades peat and origin of  pyrite in coal. Science, 221: 221-227. Astrom. M., 1998. Partitioning of  transition metals in oxidised and reduced zones of  sulphide-bearing fine-grained  sediments. Applied Geochemistry, 13:607-617. Barton, C.D., and Karathanasis, A.D., 1998. Aerobic and anaerobic metal attenuation processes in a constructed wetland treating acid mine drainage. Environmental Geosciences, 5: 43-56. Beddoe-Stephens, B. and Lambert, R.St.J., 1981. Geochemical, mineralogical and isotopic data relating to the origin and tectonic setting of  the Rossland volcanic rocks, southern British Columbia. Canadian Journal of  Earth Sciences, 18: 858. Bendell-Young, L., 1999. Contrasting the sorption of  Zn by oxyhydroxides of  Mn and Fe, and organic mater along a mineral-poor to mineral-rich fen  gradient. Applied Geochemistry, 14: 719-734. Bendell-Young, L., Thomas, C.A., and Stecko, J.R.P., 2002. Contrasting the geochemistry of  oxic sediments across ecosystems: a synthesis. Applied Geochemistry, 17: 1563-1582. Bennett, P.C., Siegel, K.I., Hill, B.M., and Glaser, P.H., 1991. Fate of  silicate minerals in a peat bog. Geology, 19: 328-331. Berner, R.A., 1984. Sedimentary pyrite formation:  an update. Geochimica et Cosmochimica Acta, 48: 605-615. Brookins, D.G., 1988. Eh-pH diagrams for  geochemistry. Springer Verlag, Berlin. 176p. Casagrande, D.J., 1987. Sulphur in peat and coal. In: A.C. Scott ,ed. Coal and Coal-bearing Strata: Recent Advances. Geological Society Special Publication No. 32., 787-105. Chague-Goff,  C., Goodarzi, F. and Fyfe,  W.S., 1996. Elemental distribution and pyrite occurrence in a freshwater  peatland, Alberta. The Journal of  Geology, 104: 649-663. Charlatchka, R. and Cambier, P., 2000. Influence  of  reducing conditions on solubility of  trace metals in contaminated soils. Water, Air and Soil Pollution, 118: 143-167. Clark, M.W., McConchie, D., Lewis, D.W. and Saenger, P., 1998. Redox stratification  and heavy metal partitioning in Avicennia-dominated mangrove sediments: a geochemical model. Chemical Geology, 149: 147-171. Cohen, A.D. and Stack, E.M., 1995. Peat: Environmental and Energy Uses. Encyclopedia of  Energy Technology and the Environment. John Wiley and Sons, New York, pp. 2205-2218. Cohen, A.D., Spackman, W. and Dolsen, P. 1983. Occurrence and distribution of  sulfur  of  peat-forming environments in southern Florida. In: R. Raymond, Jr., and M.J. Andrejko, eds. Mineral matter in peat: Its occurrence, form and distribution. Los Alamos, NM, p 87-112. Cohen, A.D., Stack, E.M., Eltayeb, S. and Durig, J.R., 1995. Applications of  peat-based sorbents for  removal of  metals from water. Abstract from I & EC Special Symposium, American Chemical Society, Atlanta, Ga, September 17-20,1995. Davis, J. A., 1984. Complexation of  trace metals by adsorbed natural organic matter. Geochimica et Cosmochimica Acta 48: 679-691. Davranche, M., Bollinger, J-C. and Bril, H., 2003. Effect  of  reductive conditions on metal mobility from wasteland solids: an example from the Mortagne-du-Nord site (France). Applied Geochemistry, 18: 383-394. Dominik, J., Rapin, F. and Vernet, J.P., 1983. Sequential extraction of  Pb-210, Po-210 stable Pb and other metals from marine and lacustrine sediments. Proceedings of  the International Conference  on Heavy Metals in the Environment, Heidelberg. CEP Consultants, eds. Edinburgh, pp 1078-1081. El Bilali, L., Rasmussen, P.E., Hall, G.E.M. and Fortin, D., 2002. Role of  sediment composition in trace metal distribution in Lake Sediments. Applied Geochemistry, 17: 1171-1181. Elder, J.F., 1988. Metal biogeochemistry in surface-water  systems—A review of  principles and concepts. U.S. Geological Survey Circular 1013, 43pp. Forstner, U., 1987. Metal speciation in solid wastes—Factors affecting  mobility, in: Speciation of  Metals in Water, Sediment and Soil Systems, L. Landner ,ed. Lecture Notes in Earth Sciences, Springer Verlag, pp 13-42. Finkelman, R.B., 1995. Modes of  occurrence of  environmentally-sensitive trace elements in coal, in: Environmental Aspects of  Trace Elements in Coal, D.J. Swaine and F. Goodarzi, eds., Kluwer Academic Publishers, Dordrecht, pp. 24-50. Forstner, U., and Haase, I., 1998. Geochemical demobilization of  metallic pollutants in solid waste— implications for  arsenic in waterworks sludges. Journal of  Geochemical Exploration. 62: 29-36. Fredrickson, J.K., Zachara, J.M., Kennedy, D.W., Dong, H., Onstott, T., Hinman, N.W. and Li, S.-M., 1998. Biogenic iron mineralization accompanying the dissimilatory reduction of  hydrous ferric  oxide by a groundwater bacterium. Geochimica et Cosmochimica Acta, 62: 3239-3257. Gee, C., Ramsey, M.H., Maskall, J. and Thornton, I., 1997. Mineralogy and weathering processes in historical smelting slags and their effect  on the mobilisation of  Pb. Journal of  Geochemical Exploration, 58: 249-257. Gee, C., Ramsey, M.H. and Thornton, I., 2001. Buffering  from secondary minerals as a migration limiting factor  in lead polluted soils at historical smelting sites. Applied Geochemistry, 16: 1193-1199. Goodarzi, F., Sanei, H. and Duncan, W.F., 2003. Deposition of  trace elements in the Trail region, British Columbia; an assessment of  the environmental effect  of  a base metal smelter on land. Geological Survey of Canada Bulletin 573, 50 pp. Hall, G.E.M., Vaive, J.E., and MacLaurin, A.I., 1996a. Analytical aspects of  the application of  sodium pyrophosphate reagent in the specific  extraction of  the labile organic compounds of  humus and soils. Journal of  Geochemical Exploration, 56: 23-36. Hall, G.E.M., Vaive, J.E., Beer, R. and Hoashi, M., 1996b. Selective leaches revisited, with emphasis on the amorphous Fe oxyhydroxide phase extraction. Journal of  Geochemical Exploration, 56: 59-78. Herbert, R.B., 1996. Metal retention by iron oxide precipitation from acidic ground water in Dalarna, Sweden. Applied Geochemistry, 11: 229-235. Hoy, T. and Andrew, K.P.E., 1989. Tectonics and mineralization in southeastern British Columbia: Structure, stratigraphy and mineral deposits of  the Rossland Group, Nelson Area. In:Geologic guidebook for  Washington and adjacent areas: Washington Division of  Geology and Earth Resources Information Circular 86, p 63-67. Hoy, T. and Andrew, K.P.E., 1991. Geology of  the Rossland area, southeastern British Columbia. Geological Fieldwork, 1990, British Columbia Geological Survey Branch, pp21-31. Jacquier, P., Meirer, P. and Ly, J., 2001. Adsorption of  radioelements on mixtures of  minerals—experimental study. Applied Geochemistry, 16: 85-93. Karczewska, A., 1996. Metal species distribution in top- and sub-soil in an area affected  by copper smelter emissions. Applied Geochemistry, 11: 35-42. Kerndorff,  H. and Schnitzer, M., 1980. Sorption of  metals on humic acid. Geochimica et Cosmochimica Acta, 44: 1701-1708. Kersten, M. and Forstner, U., 1987. Effect  of  sample pretreatment on the reliability of  solid speciation data of  heavy meals—Implications for  the study of  early diagenetic processes. Marine Chemistry, 22: 299-312. Kirby, C.S., Thomas, H.M., Southam, G. and Donald, R., 1999. Relative contributions of  abiotic and biological factors  in Fe(II) oxidation in mine drainage. Applied Geochemistry, 14: 511-530. Land, M., Ohlander, B., Ingri, J. and Thunberg, J., 1999. Solid speciation and fractionation  of  rare earth elements in a spodosol profile  from northern Sweden as revealed by sequential extraction. Chemical Geology, 160:121-138. Lee. G., Bingham, J.M. and Faure, G., 2002. Removal of  trace metals by copreciptation with Fe, Al and Mn frm natural waters contaminated with acid mne drainage in the Ducktown Mining District, Tennessee. Applied Geochemistry, 2002: 569-581. Lee, X. and Thornton, I., 2001. Chemical partitioning of  trace and major elements in soils contaminated by mining and smelting activities. Applied Geochemistry, 16: 1693-1706. Leenheer, J.A., Wershaw, R.L., Brown, G.K. and Reddy, M.M., 2003. Characterization and diagenesis of strong-acid carboxyl groups in humic substances. Applied Geochemistry, 18: 471-482. Little, H.W., 1982. Geology of  the Rossland-Trail Map-Area, British Columbia. Geological Survey of Canada Paper 79-26, 38pp. 1 Lombi, E., Sletten, R.S. and Wenzel, W.W., 2000. Sequentially extracted arsenic from different  size fractions of  contaminated soils. Water, Air and Soil Pollution, 124: 319-332. Luksiene, B.J. and Shvedov, V.P., 1985. Presence and forms  of  210-Pb, 210-Bi and 210-Po in ground level aerosols. Journal of  Radioanalytical Chemistry, 91: 215-219. Martin, J.M., Nirel, P. and Thomas, A.J., 1987. Sequential extraction techniques: Promises and problems. Marine Chemistry, 22: 313-341. Maskall, J., Whitehead, K., Gee, C. and Thornton, I., 1996. Long-term migration of  metals at historical smelting sites. Applied Geochemistry, 11:43-51. Mucci, A., Boudreau, B. and Guignard, G., 2003. Diagenetic mobility of  race elements in sediments covered by a flash  flood  deposit: Mn, Fe and As. Applied Geochemistry 18: 1011-1026. Miiller, A., 2002. Pyritization of  iron and trace metals in anoxic fjord  sediments (Nordlsvannet fjord, western Norway). Applied Geochemistry, 17:923-933. Nriagu, J.O. and Pacyna, J.M., 1988. Quantitative assessment of  worldwide contamination of  air, water and soils by trace metals. Nature, 333: 134-139. Palumbo, B., Bellanca, A., Neri, R. and Roe, M.J., 2001. Trace metal partitioning in Fe-Mn nodules from Sicilian soils, Italy. Chemical Geology, 173: 257-269. Rieuwerts, J.S., Farago, M.E., Cikrt, M. and Bencko, V., 2000. Differences  in lead bioavailabilty between a smelting and a mining area. Water, Air and Soil Pollution, 122: 203-229. Rooney, C.P., McLaren, R.G. and Cresswell, R.J., 1999. Distribution and phytoavailability of  lead in a soil contaminated with lead shot. Water, Air and Soil Pollution, 116: 535-548. Saulnier, I. and Mucci, A., 2000. Trace metal remobilization following  the resuspension of  estuarine sediments: Saguenay Fjord, Canada. Applied Geochemistry 15: 191-210. Sevigny, J.H., 1990. Geochemistry of  the Jurassic Nelson plutonic suite, southeastern British Columbia. In: Project Lithoprobe; southern Canadian Cordillera transect workshop. Lithoprobe Report 11, pp. 41-52. Simony, P.S., 1979. Pre-Carboniferous  basement near Trail, British Columbia. Canadian Journal of  Earth Sciences, 16: 1-11. Sohn, M. and Rajski, S., 1990. The adsorption of  Cd (II) from seawater by humic acids of  various sources of  origin. Organic Geochemistry, 15:439-447. Song, Y., Wilson, M.J., Moon, H.-S., Bacon, J.R. and Bain, D.C., 1999. Chemical and mineralogical forms of  lead, zinc and cadmium in particle size fractions  of  some wastes, sediments and soils in Korea. Applied Geochemistry, 14: 621-633. Southam, G., Donald, R., Rostad, A. and Brock, C., 2001. Pyrite discs in coal: Evidence for  fossilized bacterial colonies. Geology, 29: 47-50. Stack, E.M., Eltayeb, S., Liu, J., Cohen, A.D. and During, J.R., 1994. The use of  characterized peats as sorption media for  heavy metals. Extended Abstract, I and EC Special Symposium, American Chemical Society, Atlanta, GA, September 19-21, 1994. Tessier, A., Campbell, P.G.C. and Bisson, M., 1979. Sequential extraction procedure for  the speciation of particulate trace metals. Analytical Chemistry, 51: 844-851. Tuttle J.H., Dugan, C.B. and Randies, C.I., 1969. Microbial sulpfate  reduction and its potential utility as an acid mine water pollution abatement procedure. Applied Microbiology, 17: 297-302. Verner, J.F., Ramsey, M.H., Helios-Rybicka, E. and Jedrzejczyk, B., 1996. Heavy metal contamination of soils around a Pb-Zn smelter in Bukowno, Poland. Appied Geochemistry, 11: 11-16. Wu, F. and Tanoue, E., 2001. Geochemical characterization of  organic ligands for  copper (II) in different molecular size fractions  in Lake Biwa, Japan. Organic Geochemistry, 32: 11-20. Wiist, R.A.J., Hawke, M.I. and Bustin, R.M., 2001. Comparing maceral ratios from tropical peatlands with assumptions from coal studies: do classic coal petrographic interpretation methods have to be discarded? International Journal of  Coal Geology, 48: 115-132. Chapter 6 The organic petrographic characteristics of  peats from  an area of  high trace element input 6.1 Introduction While chemical analyses are critical in evaluating sediment samples from areas subject to anthropogenic environmental impact, microscopic methods which characterise the enclosing sediments can contribute valuable complimentary information.  In the case of  peat sediments, organic petrographic methods are particularly useful.  Organic petrography can be used to quantitatively characterise the phyteral composition of  a peat, and to reconstruct a general depositional history of  the peat in terms of  the redox conditions, microbial populations and the nature of  the vegetation. The conditions within the peatland have important implications for  the capacity to sequester metals, as metal mobility is a function  of  numerous interrelated factors  such as the redox conditions, the pH, and the potential for  metal retention by sorption, complexation, coprecipitation and chelation by various components found  within the peat. These environmental parameters are imprinted on the peat in terms of  the extent of  humification,  gelification  and oxidation, as indicated by the content of  huminitic and inertinitic macerals. This study examines the organic petrographic characteristics of  peats from the area surrounding the Teck-Cominco Pb-Zn smelter in Trail, British Columbia, Canada, and is part of  a Geological Survey of  Canada-Teck Cominco project to determine the trace elemental impact of  the smelter on the surrounding area. The capacity for  peats to retain selected trace elements (Sb, As, Cd, Cu, Pb, Ni and Zn) relative to the mineral topsoils was discussed in Chapter 2. It was observed that the small shallow peatlands in the region are significant  sinks for  mobile metal species deposited from the smelter. Both the total concentration and the mode of  occurrence of  selected trace elements as determined by sequential leaching vary as a function  of  profile  depth (Chapter 5). Thus, characterising the maceral composition of  the peat enables discussion of  the behaviour of the trace elements as a function  of  the prevailing conditions within the profile. A reflected  light microscopic investigation was conducted to achieve two objectives: 1. To characterise the peat in terms of  the nature of  the constituents and the degree of diagenesis due to chemical and biochemical humification.  Stack et al. (1994) and Cohen et al. (1995) have demonstrated that metal sorption by peats varies both as a function  of the individual metal and of  the peat characteristics. The extent of  humification  may have implications for  the sorption capacity of  the organic fraction  for  metals. Although there is no single accepted model for  the early diagenesis of  organic matter, it is generally proposed that as humification  proceeds, cellulose is mineralised and the more recalcitrant lignin compounds become increasingly molecularly condensed, leading to a decrease in functional  groups available to chelate and sequester metal ions. Due to overprinting by factors  such as differing  amounts of  anthropogenic deposition and variable bedrock geochemistry between the sites, it is not reasonable to expect metal concentrations to be meaningfully  correlated with the degree of  peat humification  on a site-to site basis. However, since the organic fraction,  as defined  by sequential leaching, has been shown to be an important repository for  metals (Chapter 5) the degree of  humification  is a parameter worthy of  further  observation. Organic petrographic methods may also be used to construct a history of  diagenetic conditions within the peat profile.  Characteristics such as the Eh and pH conditions may be inferred  from the degree and type of  diagenesis. Gelification  (loss of  ligno-cellulosic cell wall integrity) occurs under anaerobic conditions due to bacterial activity, while humification  (mineralisation of  organic matter) and fungal  degradation of  lignin require aerobic conditions, and occur at a much faster rate than gelification.  The presence of  funginite  (fungal  remains) may also be used as an indicator of  past or present aerobic conditions, as fungi  are obligate aerobes. Other inertinite macerals such as semi-fusintite  and inertodetrinite may indicate in situ oxidising conditions, or may be allochthonous (e.g. ash material transported by wind from forest fires). 2. To qualitatively characterise the mineral matter within the peat samples. Mineral matter within peats may be of  both detrital (e.g. quartz particles) and neoformed  origins (e.g. pyrite). Detrital mineral matter may be a source of  geogenic trace elements, while neoformed  sulphide minerals may sequester aqueous metal species. 6.2 Sampling Locations The Teck-Cominco smelter is located in Trail, British Columbia, situated in the Columbia River Valley within the Columbia Mountains. The smelter has been in operation since 1896, and produces Pb and Zn as its primary products, as well as Cd, In, Ge, Ag, Au, Bi, Cu, As, sulphuric acid and ammonium sulphate fertiliser. The sampling locations are underlain by Jurassic and Cretaceous plutonic rocks emplaced during the Columbian Orogeny (the Nelson Intrusives), and Jurassic metavolcanic and metasedimentary rocks (the Elise Formation). Due to the relatively dry climate, peatlands in the region are few in number, and of  limited size and depth. A description of  the sampling sites is provided in Table 6-1, and their locations are indicated in Figure 6-1. Photographs of  the sampling sites are provided in Figure 6-2. Site Bedrock Geology Vegetation pH and Comments Cominco UTM Zone 11,0446226, 5440375 Trail Pluton (granodiorite, quartz diorite) Typha  latifolia (common cattail), Epilobium  ciliatum (purple leafed willow herb) Carex  urticulata (beaked sedge) Equisetum  palustre, and Deschampsia cespitosa  (tufted hairgrass) 5.3 (surface) 5.6 (mineral sediment) 5.7 (pore water) -saturated at the time of sampling, although gleying was noted in the uppermost 15 cm-intermittently aerobic. Zones of oxidation were noted around roots. Thunder Road UTM Zone 11,0445145, 5440974 Trail Pluton (granodiorite, quartz diorite) Typha  latifolia (common cattail), Pinus  sp. (pine) Betula  sp. (birch) Epilobium  ciliatium Equisetum  palustre, and Deschampsia cespitosa  (tufted hairgrass) 5.0 (surface) 5.4 (mineral sediment) 5.7 (pore water) -watertable was approximately 10 cm below the surface at the time of sampling Champion Lakes Provincial Park UTM Zone 11,0453623, 5448726 Elise Formation (flow breccia, lava agglomerate, volcanic breccia, tuff,  tuffaceous conglomerate, andesite, basalt, augite porphyry, metamorphosed to greenschist facies). Limestone xenoliths and calcite amygdules found within mafic flows. Economic deposits of Pb, Zn Ag, associated with sulphides. Carex  sp. Nuphar  lutea  (cow lily) Assorted grasses 6.5 (surface) 6.5 (15cm depth) 6.4 (pore water) Saturated profile Bombi Summit UTM Zone 11, 0459296, 5454509 Bonnington Pluton (hornblende rich granodiorite, granitic gneiss, amphibolite) Carex Assorted grasses Organic rich-sediment, not peat Possible urban contamination, due to proximity to road Table 6-1. Description of  sampling sites, bedrock geology and vegetation. 117°46 117°34 "149°15* | QUATERNARY •Unconsolidated sediments; till, sand gravel, silt JU^STC/CRETACEOUS 1 NEUSON INTRUSIONS: "granodiorite; minor quartz diorite and diorite X ROSSLAND MONZONITE: -ibJdJite-bornblBnde-augite monzonite, mainly medium grained C iELj^E FORMATION: da, massive andesites aiu Lwsdlts, agglomerate, tuff, breccia, laminated siltstone PALEOZOIC [TH.UO|JNT ROBERTS FORMATION: L^jblagksiltstone and argillaceous quartzitB, slate, greywacke, chert pebble conglomerate, lava; Kmestone J J^OWN L GNEISS: libolite and grey biotite gneiss, hcmblende gneiss, mica schist, aplite and Kilometres 1 0 1 i i r Figure 6-1. Geological map of  the Trail area, indicating the location of  sampling sites. Figure 6-2. Location photographs a. Teck-Cominco smelter b. Cominco sampling site c. Cominco peat profile  during sampling d. Thunder Road sampling site e. Champion Lakes sampling site. f.  Bombi Summit sampling site. 6.3 Methodology Sites were selected in order to encompass a range of  suspected impact from the smelter. Each site was sampled by digging a small pit and removing blocks of  peat from the pit wall. The peat was sealed in zip-top freezer  bags and was refrigerated  within 12 hours of  sampling. Whole peat samples were sectioned into 5 cm oriented blocks in order to obtain a representative continuous profile  for  analysis. The blocks were freeze  dried until desiccated. The samples were placed in rubber moulds greased with petroleum jelly, impregnated with resin under a vacuum and allowed to cure overnight. The samples were then removed from the moulds and polished with progressively finer  grit paper on a polishing turntable. The final  mirror surface  was achieved through the use of  Al-flour slurry and a silk polishing cloth. The point counts were carried out using an oil immersion objective on a Leitz reflected  light microscope, using white light to observe mineral matter and blue light for organic matter. Four hundred counts per block were used to calculate volume percentages for  maceral composition. Maceral terminology followed  ICCP guidelines (ICCP, 1971; 1997) adapted for  peats (e.g. Esterle et al., 1991). Sequential leaching was performed,  and is discussed in detail in Chapter 5. The procedure followed  is modified  from Tessier et al. (1979) and Hall et al. (1996). Elemental concentrations were determined by Instrumental Neutron Activation analysis (INAA) and inductively coupled plasma mass spectrometry (ICP-MS), performed  by Becquerel Laboratories in Mississaugua, Canada, and ICP-MS performed  at the University of  British Columbia (Refer  to Chapters 2 and 5 for  further  details). 6.4 Results and Discussion 6.4.1 Trace  Element  Concentrations The highest concentrations of  Sb, As, Cu, Pb, Ni, and Zn are found  in the Cominco profile  (Table 6-2). Cadmium concentrations are highest in the Thunder Road profile.  Trace element concentrations are considerably lower in the Champion Lakes and Bombi Summit profiles.  A complete discussion of  total trace element concentrations in the peats and the results of  a sequential leaching process is presented in Chapter 5. Sample Depth (cm) Sb ppm As ppm Cd ppm Cu ppm Pb ppm Ni ppm Zn ppm COMINCO 0-1 126 78.5 61.7 196 3550 10.8 5910 1-2 141 97.8 65.2 223 3830 11.6 5600 2-3 128 153 53.8 215 3780 15.4 4230 3-4 142 116 60.3 227 3830 13.2 4730 4-5 121 164 47.1 214 3940 14.8 3570 17-20 85.0 247 60.1 226 5240 12.4 3040 THUNDER ROAD 0-1 34.1 60.0 119 82 1670 9.2 2730 1-2 37.7 64.4 105 103 2070 10.6 2270 2-3 33.0 62.0 139 84 1750 9.4 2900 3-4 37.6 65.4 134 96 1980 11 2590 5-6 32.5 67.5 104 92 1990 10.8 2330 11-14 34.9 59.0 120 96 1940 11 2300 CHAMPION LAKES 0-1 1.2 1.8 2.4 12 77 3 148 1-2 1.6 3.6 3.7 16 101.5 6 194 2-3 5.5 8.3 6.0 14 172 9.2 214 3-4 3.8 6.8 4.4 13 114.5 6 208 4-5 9.0 15.4 6.8 20 246 8.8 268 11-16 15.2 15.8 6.5 24 324 4.4 178 BOMBI SUMMIT 0-1 2.2 2.9 2.1 14 79.5 12.8 130 1-2 3.0 7.8 2.1 15 104 10.2 122 2-3 2.8 22.6 2.1 13 108.5 11.2 116 3-4 5.0 7.8 3.3 18 191 11.4 150 4-5 8.8 9.8 7.9 29 476 10.8 248 6-7.5 2.5 4.1 1.1 8.2 82.1 9.8 85.3 7.5-9 1.3 5.2 0.8 6 35 12.4 76 Table 6-2. Concentrations of  trace elements in the peat profiles. (1) CT) O E x a5 -g 2 P fc  w 3 <D X cn <D .-*=: c .g t= -b <u TJ o -f—» CL Q) _C <u -C O EE 0 » I ^ 0) D j j -F ' — i -O C QJ D) _Q c 8 3 3 3 0 —> C 'E 0 [c X 0 -4—' c 0 •4—' ' c >> •C Q. O O ._ CL u-h ° O CL _ O (/) Q py Japuniij. jqiuog OOUjlUOQ >n uoidiueqo Figure 6-3. Organic petrographic composition of  the peat profiles,  expressed as volume % Results of  the point count are summarised in Figure 6-3. A cross plot of structured/unstructured huminite macerals versus inertinite macerals (Fig. 6-4) provides an illustration of  the extent of  humification  within each of  the samples. 6.4.2 Organic  Petrography  and Mode  of  Occurrence 6.4.2.1 Cominco The peat from the Cominco site is sapric, with small (approximately 5 mm) fragments  from reeds and rootlets. At 17 cm depth, large woody roots, approximately 1 cm in diameter, are present. At the time of  sampling, the water table was very close to the surface,  and the peat profile  was completely saturated. The pH at this site ranges between 5.3 (surface  peat) to 5.6 (mineral sediment), while the pore water had a pH of 5.7. This site is located approximately 0.5 km from the smelter and is known to be subject to heavy impact from the smelter (Goodarzi et al., 2003; Chapter 2). The organic petrographic makeup indicates that the Cominco profile  consists of well humified  peats (22-33 % textinite, 3-18% texto-ulminite and 5-7% humodetrinite/gelinite). The peat at the surface  contains less textinite than the peat at depth, indicating that the peat surface  is currently subject to more intense humification than it was in the past, and that the capacity of  peatland to accumulate organic matter is decreasing, due to drainage and subsequent oxidation. The most abundant liptinite maceral is cutinite, ranging from 2-7%, with lesser amounts of  resinite, liptodetrinite and sporinite. Funginite is prevalent throughout the profile,  particularly in the uppermost peat (almost 40 %). The role of  metals in inhibiting humification  by restricting the growth of microbial populations is somewhat uncertain: Brookes and McGrath, (1984) determined that the application of  sewage sludge containing heavy metals resulted in decreased microbial mass and respiration, while Filcheva et al. (1996) found  no significant  decrease in either parameter. Metals may be toxic to some microbes, but the impact is variable between species and between metals (Ledlin, 2000). Although copper sulphate is used as an agricultural fungicide,  and the Cu concentration in the Cominco profile  is high (up to 226 ppm), the abundant funginite  suggests that fungal  growth has not been inhibited, and the well-humified  state of  the 50 45 40 35 .E 30 •e 0) Z 25 <3 P H 20 15 10 5 0 Well Humified Well Oxidised Well Humified Unoxidised Well Preserved Unoxidised 10 15 Humification Index Textinite/(Texto-ulminite+Humodetrinite) 20 Cominco • 4 - 9 cm m 9-15 cm Champion Lake • 5-9 cm O 9-16 cm 25 Thunder Road ' A 0-5 cm A 5-10 cm ®10-15cm Bombi Summit • 0-5 cm • 5-10 cm Figure 6-4 . Graph of  humification  index versus total inertinite content. peat suggests an active community of  plant-decomposing microflora.  The presence of relatively large amounts of  semifusinite  and inertodetrinite (4.7-5.8%) indicate that the site is frequently  subject to intense oxidising conditions or input of  combusted materials from forest  fires. The peat from  the Cominco site contains elevated concentrations of  Sb, As, Cd, Cu, Pb and Zn relative to British Columbia (2002) and Environment Canada (2002) environmental quality standards for  soils, including Pb concentrations of  up to 5240 ppm and Zn concentrations of  up to 6170 ppm. Nickel concentrations are below the listed soil quality standards. Concentrations of  Sb and Zn are greatest at the surface  and increase with depth, while As, and Pb are most concentrated at the base of  the profile,  and Cd, Cu and Ni concentrations are consistent throughout (Table 6-2). The oxidising, weakly acidic depositional environment indicated by the organic petrographic makeup is confirmed  by the mode of  occurrence for  the smelter-derived trace elements. Both the carbonate and the reducible fractions  are important sinks for  Pb, Zn and Cd (Fig. 6-5). The carbonate fraction  is a significant  sink for  Pb (9-20%) and to a lesser extent, Zn (1-7%) and Cd (2-7%). The prevalence of  the carbonate fraction  in this profile  corresponds with the well-humified  nature of  the peat: highly acidic conditions that prevent the precipitation of  carbonates also inhibit humification.  The formation  of Pb carbonate species may occur within the pH range of  the peat (Brookins, 1988), although carbonate complexes for  Zn, Cd, Ni and Cu are more stable at higher pH's and are thus less prevalent. The reducible Fe and Mn fraction  is the major sink for  Pb, Zn, As and Cd (19-72%) throughout the profile  depth, and a major sink for  Ni, Cu and Sb in selected samples. The organic chelate fraction  is retains Pb, Sb, Cu and As (generally less than 20%). The sulphide fraction  represents approximately 2-4% of  the As, Pb, Zn and Cu, and 7-10% of  the Sb and Cu at the surface  of  the peat. The relative importance of  the sulphide fraction  decreases with increasing depth until the basal mineral sediment Figure 6-5. Relative concentrations (%) for  sequentially extracted fractions  in the Cominco profile,  including basal mineral sediment (Bsed). CD Ca) N ) C/> h O | O £ 1 0 Ca) N ) 1 Q 0 ) Ca) • I 71 (D w d c Q) o 0 CQ Q) 2 Si U) c T3 :x CL (D CO 0) D O O IT (D r+ (D O 0 s N) O ox ^ O 0 s "0 o) 0" 0 s CO o ox O O x° 0 s o co -i cr o D Q) (D m X o D" £D 3 (Q (D 0) g; (D CD Ca) N ) - A ( / ) K ) O ^ o o ® i) N) ' •M C O IV) n ^ iv — —x i i i CD 0) l\) J ro o a 0) u s j co u _ __ I I I I i P is reached. This suggests that the oxidising conditions within the peat do not favour  the formation  of  sulphides. 6.4.2.2 Thunder  Road  Profile This site is located approximately 1.5 km from the smelter and is subject to significant  trace element deposition from the smelter, although to a lesser extent than the Cominco site, due to the larger intervening radial distance and partial topographic sheltering (Goodarzi et al., 2003, Chapter 2). The site is also subject to input from a fertiliser  factory,  which currently produces N-fertiliser,  but previously produced phosphate and potassium fertilisers  as well (W. Duncan, personal communication). Macroscopically, the Thunder Road peat is uniform in appearance throughout the profile.  Fine rootlets (<lmm in diameter) and larger reed fragments  in varying states of decomposition/preservation make up the majority of  the material. The reed fragments  are variable in size, ranging from approximately 3 mm to 1 cm in diameter. The identifiable plant fragments  are set in an amorphous medium brown matrix. The uppermost surface has abundant reeds. The pH ranges between 5.0 (surface)  to 5.4 in the basal sediment, and is 5.7 in the pore water. The Thunder Road profile  is less humified  (60-75% textinite, 0.25% texto-ulminite and 2-3.5% humodetrinite), and contains less funginite  (11-11.5 %) than the Cominco profile.  The lower abundance of  other inertinite materials (max. 2%) and low level of  humification  suggests that this site is typically anaerobic throughout most of  the profile.  The textinite content decreases slightly with increasing depth, to a minimum of 60% in the 10-15 cm sample, and there is a corresponding increase in resinite, liptodetrinite, sporinite and chlorophyllinite, while the texto-ulminite and humodetrinite contents remain constant. Funginite is present throughout the profile,  albeit in lesser amounts than in the Cominco profile,  indicating past or intermittent aerobic conditions (11%). Liptinites are present only in small amounts, the most abundant of  which is cutinite (up to 7%). Total concentrations of  trace elements are lower at the Thunder Road site than at the Cominco site, with the exception of  Cd (Table 6-2). Trace elemental concentrations are consistent throughout the depth of  the Thunder Road profile.  The carbonate fraction is a significant  sink for  Pb, Zn and Cd (2-25%; Fig. 6-6). Figure 6-6. Relative concentrations (%) for  sequentially extracted fractions  in the Thunder Road profile,  including basal mineral sediment (Bsed). zoz % ? 2 . g ^ cn co ho o U1 O) i CO A CD (J) CD Q. O o N ro o O CD O 0 s 00 o O O ^ i cn co k> • • i CD -t^ CO 0 1 K> O OJ Sb 0% 20% 40% 60% 80% 100% 0-1 _ 2-3 1 3-4 CL 0) •U E o 5-6 14-15 19-20 BSed 0-1 2-3 3-4 •5. 5-6 0) "°14-15 19-20 BSed -J 1 1 1 1 I K m 1 ife1-mam mm* i -I I I mmm WMBM 1 1 \ M M wMmfflwmmmm j i — I \ m m m AViTAWA' 1 WMM&MW^^ -] mmmw i i i i i i i Pb 20% 40% 60% 80% 100% mm Cd 0% 20% 40% 60% 80% 100% 0-1 2-3 3-4 5-6 14-15 19-20 BSed m. m ^ ^ Exchangeable Carbonate Reducible Organic Chela te Organic/Sulphide Residual m The relative percentage of  element sequestration associated with the carbonate fraction decreases with increasing profile  depth for  Zn and Cd, but increases with increasing depth for  Pb at the expense of  the organic chelate fraction.  The reducible fraction  is a major sink for  Ni near the surface  (up to 27%), and for  Zn, As, Cd and Pb throughout the profile (17-76%). The organically chelated fraction  is a major sink for  Pb and As, and a minor sink for  Sb. The percentage of  As and Pb sequestered in the organic fraction  decreases with increasing profile  depth. The sulphide fraction  is significant  for  Ni, Cu, Zn, As, Sb, and Pb at the surface,  but decreases with increasing profile  depth. 6.4.2.3 Champion Lakes Profile This site is located in a provincial park, approximately 13.5 km from the smelter. The site is somewhat sheltered from deposition from the smelter due to the intervening mountainous topography, and the fact  that the prevailing winds are generally from  the northwest (Goodarzi et al., 2003). Peat was sampled from a marsh surrounding First Lake. The peat is fibric  at the surface  (the top 3 cm), and is yellowish-brown in colour. It gradually grades from fibric  to dark brown hemic peat (approximately 7 cm) to black sapric peat (approximately 12 cm), with intact rootlets, and partially humified  reeds set in an amorphous matrix. The peat pH was 6.4 at the surface  and at depth, while the pore water had a pH of  6.5. The Champion Lakes site peat displays a humification  pattern typical of  an active peat-forming  environment, in that the surficial  peat is predominantly minimally-altered textinite-rich peat (55% textinite, 6% texto-ulminite, 2% humodetrinite), while the peat at depth is more gelified  (37% textinite, 17% texto-ulminite, 4% humodetrinite). This indicates that the peat below 9 cm is waterlogged, permitting bacterial gelification.  The presence of  funginite  and other inertinite (12-16%) is evidence of  intermittent aerobic conditions within the profile,  likely due to fluctuations  in the lake water levels. The near-neutral pH conditions illustrate the moderating influence  of  the lake water, which acts to dilute the impact of  organic acids generated during humification. Total trace elemental concentrations are much lower in the Champion Lakes profile  than at the Cominco and Thunder Road sites. Concentrations of  trace elements generally increase with depth (Table 6-2). A minor carbonate and organic chelate association is noted for  Zn and Pb at the profile  surface  (2-9%; Fig. 6-7). The reducible fraction  is a sink for  Zn, As and Cd at the peat surface  (16-21%) and for  Zn throughout the profile  depth (up to 40%). Sulphides are relatively more important in this profile  than for  the Cominco and Thunder Road profiles,  although overall concentrations are much lower. Sulphur concentrations are higher in the Champion Lakes peat than in the Cominco or Thunder Road peats and thus there is a greater pool of  sulphur for  the trace elements to react with. The major proportions of  Ni, Cu, and As, Sb and Pb at depth are found  within the sulphide fraction. 6.4.2.4.Bombi Summit  Profile The Bombi Summit organic deposit consists of  7 cm of  peat overlying sandy sediment, located approximately 16 km from the smelter. There is a thin (<1 cm) layer of fibrous  peat at the surface,  while the underlying material is a dark brown to black amorphous matrix with fine  rootlets and reed fragments  in variable states of decomposition, and coarse sand grains. Large wood fragments  (approximately 3.5 cm diameter) are also present throughout the profile. The Bombi Summit peat is moderately humified  at the surface  (45% textinite), but highly detrital and oxidised in the 5-10 cm sample (6% humodetrinite, 26% liptodetrinite, 19% funginite,  8% other inertinite). The high funginite  content (25-28%) indicates that oxidising conditions prevail at this site. Antimony, Cd, Cu, Pb, Zn in the Bombi Summit profile  increase in concentration from the surface  to a maximum in the 4-5 cm sample, and then decrease to a minimum in the 7.5-9 cm sample. Arsenic is most concentrated in the 2-3 cm sample and Ni concentrations are consistent throughout the profile  (Table 6-2). Trace elements are associated primarily with the reducible, sulphide and residual phases (M. Hawke, unpublished data). Ni 40% 60% Cu 60% 100% Zn As _ 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% « 2-5 m m m - ^ i - — — n 2-5 1 — — £ 1 , | ^14-15 14-15 w/m T3 Sb 40% 60% Cd 40% 60% 80% 100% CL d) •o 0% 2-5] 14-15 fe 20% i M W W Pb 40% 60% + • ! 80% 100% Exchangeable Carbonate j | j Reducible Organic Chelate ^ Organic/Sulphide I Residual Figure 6-7. Relative concentrations (%) for  sequentially extracted fractions  in the Champion Lakes profile. 6.4.3 Inorganic  Matter Micrographs (Fig. 6-8) illustrate some examples of  the types of  mineral matter observed in the Cominco profile.  Of  particular interest is the occurrence of permineralised diatoms, pollen grains and plant tissues (Fig. 6-8 d and e), which act as nuclei for  precipitation reactions. Finely dispersed sulphide minerals (Fig. 6-8a) are common in all profiles,  but are typical epigenetic minerals found  in peatlands, and as such cannot be used to infer  anthropogenic impact. Examples of  high temperature carbonaceous neoformed  particulate material (fly  ash) occur in several samples (Fig. 6-8b and c). Coal is used to fire  the slag furnace,  and coke is used as a flux  in Pb production, and so the smelter is the most likely source of  this material. The peat is also rich in detrital minerals such as quartz (Fig. 6-8f).  Figure 6-9 illustrates examples of  mineral matter from the Champion Lakes peat, which are similar to the Cominco peats, including permineralised plant materials, fly  ash from the smelter and sulphide minerals such as pyrite. Other than the fly  ash, no smelter-derived particulate material was observed using light microscopy, although an SEM-EDX study of  peat ash material (Chapter 4) identified Pb and Zn containing particles, originating both from the smelter stack and from fugitive feedstock  dust, based on the particle morphology. These particles were typically less than 2 pm in diameter, and it is likely that their defining  characteristics (rounded or "fluffy" morphology) would not be recognisable at the level of  magnification  used for  the point count. Since the particles emitted from the smelter had varying chemical compositions, as determined by EDX (Fe, S, and Si associations), their optical properties under reflected white light may also be variable. 6.4.4 Organic  Sequestration  of  Metals-the  role  of  peat type and humification Trace element fractionation  and mobilization is subject to a number of  controls, including Eh and pH (e.g. Forstner, 1987). Since redox and pH conditions may fluctuate within peatlands due to both day-to-day and seasonal climatic variation, the concentration and mode of  occurrence of  the trace elements are likely in a state of  flux  (e.g. Van Den Berg et al., 1999). Eh-pH diagrams, such as those presented in Brookins (1988), are helpful  in determining potential speciation, but are not entirely A m Figure 6-8. Examples of  mineral matter found  within the Cominco profile.  A. Sulphide mineral, probably pyrite B. and C. highly reflecting  carbonaceous material (fly  ash) produced from the combustion of  coal in the smelter operation. D. diatom permineralised with sulphides. E. pollen grain permineralised with sulphides F. oxidised framboidal  pyrite and quartz grain. c Figure 6-9. Examples of  mineral matter found  within the Champion Lakes profile.  A. woody material permineralised by sulphides. B. fly  ash particle. C. particle of  sulphide mineral, probably sphalerite. applicable to situations where mode of  occurrence may be influenced  by reactions with organic matter or microbial activity. Co-precipitation of  trace elements with Fe and Mn oxides is an important mode of  sequestration within the aerobic sediments, notably the upper portions of  the Cominco and Thunder Road profiles  and the Bombi Summit profile. Neo-formed  sulphides, similar to the examples illustrated in Figures 6-8 and 6-9, are important trace element sinks in the anaerobic portions of  the profiles,  such as the Thunder Road and Champion Lakes peats. The role of  the microbial community in trace element behaviour is frequently overlooked. In all profiles,  there is evidence of  an active microbial community, as indicated by the direct evidence of  funginite,  and the indirect evidence of  humified  and gelified  plant tissues. Microflora  have the potential to accumulate significant  amounts of trace metals through a variety of  mechanisms, including ionic and covalent adsorption to the cell surface,  passive and active uptake, and complexation within extracellular polymers such as capsid layers or slime (Ledin, 2000). Amine groups found  within humic substances are also attributed in part to microbial and algal sources (e.g. Hatcher and Spiker, 1988). While the population of  the microbial community was not determined in this organic petrography study, it is recognised that some of  the organic substances responsible for  the retention of  trace elements in the peat are likely derived from microflora. Rates of  peat growth and humification  will also affect  the concentration of  non-mobile atmospherically-deposited elements, as slowly-accumulating peat will be exposed to the atmosphere for  longer periods of  time (Kempter et al., 1997). The well humified peat at the Cominco site likely accumulates at a slower rate than the well preserved peat in the Thunder Road profile  due to higher net rates of  organic matter mineralisation, which may partially account for  the higher concentrations of  trace elements found  within the Cominco peat. Dating of  the peat deposits would provide information  regarding the age of  the peats and the accumulation rates, however, this is beyond the scope of  the current project. However, the higher rates of  trace elemental deposition at the Cominco site relative to the Thunder Road site (Goodarzi et al., 2003) are the most significant reason for  the higher concentrations found  in the Cominco peat. The ability of  organic matter to sequester metals has long been recognised. Metals are held by the organic matter via both adsorption (outer sphere complexes, i.e., exchangeable fraction)  and by complexation by organic acid functional  groups (inner sphere complexes, i.e., the organic fraction  as defined  by sequential leaching). Reactions with organic substances may serve both to retain and to mobilise metals, depending upon whether the organic complex is soluble or insoluble. Studies which have noted that the fine  fraction  of  peats typically have greater metal retention capacity than more fibrous  peats have tentatively attributed this finding  to the greater surface  area available for  exchangeable ions (Cohen et al., 1995). However, the contribution of  the exchangeable fraction  may consist of  only a small percentage to the total metal concentration, particularly for  Sb and Cu (Chapter 5). A more significant mechanism for  the retention of  metals by organic matter is complexation with negatively charged functional  groups, typically carboxyls, phenols, thiols or amines as unidentate complexes or bidentate complexes (chelates). Numerous studies have examined the nature of  metal-organic interactions, and have demonstrated that the potential for  metal complexation is a function  of  solution pH, the metal ions in solution, the functional  group chemistry of  the organic matter, and the physical structure of  the organic matter. Kerndorff  and Schnitzer (1980) illustrated that metal ions compete for  complexation sites, and the order of  sorption from solution at a pH of  5.8 was as follows: Hg=Fe=Pb=Al=Cr=Cu>Cd-Zn>Co>Mn. Organic coatings on alumina particles and dissolved organic complexes were shown to enhance adsorption of  Cu (Davis, 1984). Amine groups combined with carboxyl groups are thought to be more efficient  chelators of  metals than carboxyl groups alone (Sohn and Rajski, 1990; Wu and Tanoue, 2001). The structure of  humic substances is also cited as an important parameter in controlling the potential for  complexation, i.e. structurally complex molecules with abundant negatively charged ligands are more capable of  forming  bidentate complexes (Wu and Tanoue, 2001; Leenheer et al., 2003). With increasing humification,  quinone groups with free  radicals are thought to take on a more important role in sequestering metals (Jerzykiewicz et al., 2002), although these complexes are less stable than chelates. As humification  proceeds in the early stages of  peatification,  oxidation of  lignin side chains results in an increase in carboxylic acid groups, thereby increasing the potential for  metal retention. These side chains are increasingly lost as humification continues to more advanced stages of  diagenesis. The peats examined in this study were derived from a variety of  vascular plants with rooting systems that penetrate to varying depths, as was observed in the macroscopic observations and by the organic petrographic study. Under these circumstances, fresh  materials are constantly input throughout the profile  to the maximum depth of  rooting, and so the availability of  potential complexing organic ligands is unlikely to be a limiting factor.  It is difficult  to accurately model the organic complexation of  metals in an active peat environment, as each organic substance has a different  equilibrium constant for  each metal (see, for  example, Prapaipong et al. (1999) for  a summary of  thermodynamic properties of  metal-dicarboxylate ligand complexes) and the organic composition of  the peat is constantly changing. However, it would be worthwhile to characterise the functional  group chemistry of  peat materials from different  vegetation types at different  stages of  humification,  in order to assess their effectiveness  as environmental remediation sorbants for  metals. 6.5 Conclusions Microscopic investigations of  contaminated sediments aid in characterising both the contaminants and the enclosing sediments. An assessment of  the organic petrographic characteristics of  the peat is a valuable tool in determining the depositional history of  the peatland and thus, past and present environmental conditions within the sediment. The environmental conditions within the profile  determine the mode of  occurrence and thus the potential for  remobilisation, should the prevailing conditions change. Well humified peats indicative of  near-neutral pH and oxidising conditions are conducive to the formation  of  reducible Fe-oxide complexes, which may be subject to remobilisation and leaching under reducing conditions. Carbonate complexes are likely to be rapidly solubilised, should the pH decrease. Predominantly reducing conditions are indicated by gelified  textinite macerals. Elements precipitated as sulphides under prevailing reducing conditions may be mobilised under oxidising conditions. Light microscopy also allows one to observe neoformed  precipitated mineral matter and particulate matter from external sources. 6.6 References British Columbia. Ministry of  Water, Land and Air Protection, 2002. Waste Management Act: Contaminated Sites Regulation. Available: http://www.qp.gov.bc.ca/statreg/regAVAVasteMgmtAVasteMgmt375_96/375_96.htm Brookes, P.C. and McGrath S.P., 1984. Effects  of  metal toxicity on the size of  the microbial biomass. Journal of  Soil Science, 35, 341-346. Canada. Environment Canada, 2002. Canadian Soil Quality Guidelines. Available: http://www.ccme.ca/assets/pdf/el_06.pdf Cohen, A.D., Stack, E.M., Eltayeb, S. and Durig, J.R., 1995. Applications of  peat-based sorbents for removal of  metals from water. Extended abstract, I&EC Special Symposium, American Chemical Society, September, 1995. 4p. Davis, J.A., 1984. Complexation of  trace metals by adsorbed natural organic matter. Geochimica et Cosmochimica Acta, 48: 679-691. Esterle, J.S. and Ferm, J.C., 1994. Spatial variability in modern tropical peat deposits from Sarawak, Malaysia and Sumatra, Indonesia: analogues for  coal. International Journal of  Coal Geology, 26: 1-41. Forstner, U., 1987. Metal speciation in solid wastes—Factors affecting  mobility. In:  Speciation of  Metals in Water, Sediment and Soil Systems, L. Landner, ed. Lecture Notes in Earth Sciences, Springer Verlag, pp 13-42. Filcheva, E., Cheshire, M.V., Campbell, C.D. and McPhail, D.B., 1996. Effect  of  heavy metal contamination on the rate of  decomposition of  sewage sludge and microbial activity. Applied Geochemistry, 11: 331-333. Goodarzi, F., Sanei, H., and Duncan, W.F., 2003. Deposition of  trace elements in the Trail region, British Columbia; an assessment of  the environmental effect  of  a base metal smelter on land. Geological Survey of Canada Bulletin 573, 50 pp. Goodarzi, F., Reyes, J., and Sanei, H., 2002. Concentrations of  elements in the stream sediments in Trail, BC and the surrounding area and the natural background concentration. Unpublished report. Hall, G.E.M., Vaive, J.E., and MacLaurin, A.I., 1996. Analytical aspects of  the application of  sodium pyrophosphate reagent in the specific  extraction of  the labile organic compounds of  humus and soils. Journal of  Geochemical Exploration, 56: 23-36. International Commission of  Coal Petrology, 1971. International Handbook of  Coal Petrography. Centre National de la Recherche Scientifique  (reprinted 1985, the University of  Newcastle-upon-Tyne). International Commission of  Coal Petrology, 1997. Inertinite Classification-ICCP System 1997. International committee for  Coal and Organic Petrology. Wellington, New Zealand, 1997. Jerzykiewicz, M., Jezierski, A., Czechowski, F., and Drozd, J. 2002. Influence  of  metal ions binding on free  radical concentration in humic acids. A quantitative electron paramagnetic resonance study. Organic Geochemistry, 33: 265-268. Kempter, H., Gorres, M. and Frenzel, B., 1997. Ti and Pb concentrations in rainwater fed  bogs in Europe as indicators of  past anthropogenic activities. Water, Air and Soil Pollution, 100: 367-377. Kerndorff,  H. and Schnitzer, M., 1980. Sorption of  metals on humic acid. Geochimica et Cosmochimica Acta, 44: 1701-1708. Leenheer, J.A., Wershaw, R.L., Brown, G.K. and Reddy, M.M., 2003. Characterization and diagenesis of  strong-acid carboxyl groups in humic substances. Applied Geochemistry, 18: 471-482. Little, H.W., 1960. Nelson.Map-Area, West Half,  British Columbia (82 FW1/2). Geological Survey of  Canada Memoir 308. 205pp. Little, H.W., 1982. Geology of  the Rossland-Trail Map-Area, British Columbia. Geological Survey of  Canada Paper 79-26, 38pp. Prapaipong, P., Shock, E.L. and Koretsky, C.M., 1999. Metal-organic complexes in geochemical processes: Temperature dependence of  the standard thermodynamic properties of  aqueous complexes between metal cations and dicarboxylate ligands. Geochimica et Cosmochimica Acta, 63: 2547-2577. Sevigny, J.H., 1990. Geochemistry of  the Jurassic Nelson plutonic suite, southeastern British Columbia. In: Project Lithoprobe; southern Canadian Cordillera transect workshop. Lithoprobe Report 11, pp. 41-52. Sohn, M. and Rajski, S., 1990. The adsorption of  Cd (II) from seawater by humic acids of  various sources of  origin. Organic Geochemistry, 15:439-447. Stevenson, F.J., 1994. Humus Chemistry: Genesis, Composition, Reactions. John Wiley and Sons, New York. 496 p. Tessier, A., Campbell, P.G.C. and Bisson, M., 1979. Sequential extraction procedure for  the speciation of particulate trace metals. Analytical Chemistry, 51: 844-851. Van den Berg, G. A., Loch, J.P.G., Van der Heijdt, L.M. and Zwolsman, J.J.G., 1999. Mobilisation of  heavy metals in contaminated sediment in the River Meuse, The Netherlands. Water, Air and Soil Pollution, 116: 567-586. Wu, F. and Tanoue, E., 2001. Geochemical characterization of  organic ligands for  copper (II) in different molecular size fractions  in Lake Biwa, Japan. Organic Geochemistry, 32: 11-20. Wiist, R.A.J., Hawke, M.I. and Bustin, R.M., 2001. Comparing maceral ratios from tropical peatlands with assumptions from coal studies: do classic coal petrographic interpretation methods have to be discarded? International Journal of  Coal Geology, 48: 115-132. Chapter 7 Conclusions 7.1 Key findings  resulting from this research The overall objective of  the research was to investigate the role of  peat deposits in the partitioning of  deposited trace elements in the region surrounding a Pb-Zn smelter. The researcher was pursued through a series investigations, some of  which were more successful  than others. The principal conclusions and recommendations for  further research are summarised below. 7.1.1  Trace  element  concentrations  in organic  deposits  in the area surrounding  a base metal  smelter Trace elements deposited in the region immediately surrounding the smelter were shown to be highly subject to weathering and mobilisation in the environment, as indicated by concentration differentials  noted between peats/organic soils and nearby mineral soils. The preferential  sequestration of  the mobile trace elements in peats is due to the capacity of  organic matter to sequester ionic species, the enhanced potential for precipitation of  trace elements under reducing conditions, and possibly to the net inflow of  groundwater to peatlands. The finding  that the deposited trace elements are subject to mobilisation following  deposition indicates that the trace elements should be considered to be potentially bioavailable, which has implications for  both remediation (i.e. phytoremediation methods may be effective  for  removing trace elements from soils) and for  environmental policy (i.e. guidelines based on an awareness of  the potential for  the trace elements to enter the food  chain via locally grown produce or through the grazing of livestock on contaminated soils). The concurrent examination of  trace element concentration data from peats and mineral soils and the depositional rate data obtained from the moss monitoring program made it possible to identify  locations subject to impact from secondary sources (Cd from the fertiliser  plant at Thunder Road) and geogenic input (Cu at Champion Lakes). These findings  underline the importance of  using several types of  sampling media when conducting environmental impact studies. 7.1.2  Geochemical  finger-printing,  using element  ratios  and REE signatures It was found  that the use of  conservative element ratios is generally ineffective  in the case of  this study, both as a means of  subtracting the background geochemical input from the peatlands, and as a method for  calculating a meaningful  "enrichment factor"  to express the extent of  trace element contamination relative to a benchmark factor.  It was found  that the conservative elements do not necessarily represent a stable geochemical baseline to which it is possible to normalise trace element concentrations. Indeed it was found  that the concentrations of  the two conservative elements used in this study (Ti and Zr) fluctuated  with respect to each other, and relative to the ash content, indicating that so-called conservative elemental concentrations do not actually represent a reasonable proxy for  geogenic input. Furthermore, the calculated enrichment factors  varied widely, depending on the choice of  the media used for  the benchmark, and on the conservative element used, rendering their significance  as environmental indicators questionable. These findings  indicate that although conservative element ratios have been shown to be effective  indicators of  anthropogenic input in remote ombrotrophic peatlands, in the case of  this study they are ineffective. Rare earth element signatures were also found  to be relatively ineffective  tools for indicating the extent of  smelter impact on the surrounding environment. Chondrite normalised REE signatures from smelter feedstocks  and wastes were relatively flat,  while the REE signatures from peats were highly LREE enriched, similar to soils and stream sediments sampled in the region. This finding  indicates that the peat REE signature is most influenced  by local geochemistry, and is not extensively impacted by the smelter. REE signatures from moss monitoring media, however, were more characteristic of  the smelter materials. 7.1.3  SEM-EDX  study  of  peat ash Scanning electron microscopy was found  to be a useful  method to identify  the source of  particulate matter found  within the peat ash from the sampling sites. Stack emitted particles, identifiable  by a rounded morphology due to rapid cooling, were commonly found  in the peats sampled in close proximity to the smelter. Energy dispersive X-ray spectroscopy indicates that these particles contain Zn and Pb, commonly in association with Si, Fe and S. Precipitates containing Zn and Pb in association with Fe were also detected in the samples from the area immediately surrounding the smelter. Angular particles, particularly Zn in association with S, are indicative of  either fugitive dust particles or geogenic materials. 7.1.4  Sequential  extraction Peats and soils from the region surrounding the smelter were subjected to a sequential extraction technique to determine the mode of  occurrence as a function  of source and deposition environmental conditions. Trace elements from sites which are known to be heavily impacted by the smelter are typically found  to have a mode of occurrence indicative of  solubilisation and subsequent reprecipitation or sequestration (i.e. exchangeable, carbonate, reducible Fe/Mn, organic chelate). This finding  supports the conclusion from Chapter 2 that trace elements that are emitted from the smelter are subject to extensive mobilisation within the environment following  deposition. The predominant modes of  occurrence found  at the less impacted sites were indicative of geogenic origins (i.e. the residual and, to some extent, the sulphide fractions).  The relative distribution of  trace element concentrations in each fraction  was found  to be a function  of  the redox conditions within the peatland, and the trace element source, and was found  to vary between individual elements. 7.1.5  Organic  petrography An organic petrographic investigation of  the peats indicates that each site is subject to early diagenesis to varying degrees, and that the predominant early diagenetic processes differ  from  site to site. The Cominco site contained a relatively low volume of textinite, and had abundant inertinite macerals, particularly funginite.  This organic petrographic makeup is indicative of  a peatland subject to oxidising conditions, supported by field  observations and the high relative significance  of  the Fe-oxide fraction,  as determined by sequential extraction. The Thunder Road site, on the other hand, had a higher relative proportion of  textinite, with low volumes of  inertinite macerals, indicative of  a peatland with predominantly reducing conditions. Correspondingly, the sequential leaching results from this site indicated a lower relative percentage of  trace elements associated with Fe-oxides, with a correspondingly higher proportion of  organically-chelated elements. The organic petrographic composition of  the Champion Lakes peat indicates a peatland with moderate preservation potential, but as a significant  proportion of  the trace elements at this location are within the residual phase, the redox characteristics of  the peat have less impact on mode of  occurrence. 7.2 Suggestions for  Future Research Numerous questions were raised during the course of  this project that could be addressed in future  research. It would be worthwhile to pursue the following  topics further: 1. The peats sampled in this study were from small, shallow, minerotrophic peatlands in an area of  complex bedrock geology. A study of  ombrotrophic peatlands, subject only to atmospheric input (except at the peatland margins) from an area of  high trace elemental impact would provide an opportunity to better discriminate between geogenic and anthropogenic trace elements within the peatland. Detailed transects of  a larger ombrotrophic peatland would be useful  in determining the extent to which trace elements are mobilised from surrounding mineral soils and sequestered within the peat subject to groundwater influence,  while the ombrotrophic portion of  the peatland could potentially provide a chronostratigraphic record of  trace element deposition. 2. Sampling and monitoring programs employing similar methodologies could be conducted at other smelters in order to determine the extent to which the sandy, undeveloped nature of  the mineral soils in the Trail region influences  the potential for trace element mobility. Soils with higher clay and organic matter contents could potentially retain trace elements more effectively  than the soils surrounding the Trail smelter. 3. The peat was sampled in the field  and subsampled in the lab in such a way that oxidation occurred, which most likely resulted in subsequent changes to the mode of  trace element occurrence. As a result, the results from the sequential extraction study may not be truly representative of  conditions within the in situ peat. While this problem was judged not to be critical in this case, a sampling and analytical program designed to address this problem would be helpful  in determining the true nature of  the in situ peat geochemistry. Such a program would involve the immediate storage of  all samples in an inert gas, and performing  all lab work within a glove box under an inert atmosphere. 4. An investigation of  pore water chemistry would provide further  insight into the processes occuring within the peatlands. APPENDIX A Comparing maceral ratios from  tropical peatlands with assumptions from  coal studies. Do classic coal petrographic interpretation methods have to be discarded?1 1. A version of  this chapter has been published. Wttst, R.A.J., Hawke, M.I., and Bustin, R.M. (2001) Comparing maceral ratios from tropical peatlands with assumptions from coal studies. Do classic coal petrographic interpretation methods have to be discarded? International Journal of  Coal Geology, 48: 115-132. A.A.I Introduction A.A.1.1 Maceral  Ratio  Indexes Depositional environments play a major role in determining the composition and preservation of  peat and coals. The geology, climate, vegetation, and the hydrological regime, (e.g. level and flow of  groundwater) of  the peat deposits influence  the distribution and quality of  coal. Coal maceral compositions, determined through petrographic analyses, are used to determine the paleoenvironment and contribute to the understanding of  composition, compaction and degree of  preservation of  plant material. Macerals are frequently  considered to be plant- and/or environment-specific,  and so the assessment of vegetation type, occurrences of  droughts and fires,  and intensity of  humification  and microbial activity is possible by studying coal macerals (e.g. Diessel,1982; Styan and Bustin, 1983; Goodarzi and Hill, 1989; Lamberson et al., 1991; Markic and Sachsenhofer,  1997), and many others). Organic petrographic studies have also been utilised to interpret the degree of  preservation of  peat deposits, and to examine the spatial distribution of  inorganic material (e.g. pyrite, clays, quartz, diatoms) as a function  of environment. Hence, paleoecological interpretations, such as main floral  composition, can be inferred  based on petrographic studies of  peat deposits (Cohen and Spackman, 1972; Styan and Bustin, 1983; Cohen et al., 1987; Grady et al., 1993; Hawke et al., 1996; Shearer and Clarkson, 1998; Hawke et al., 1999). Ratios of  selected "diagnostic" macerals, such as the gelification  index (GI) or the tissue preservation index (TPI) of Diessel (1992) have been used in coals as indicators of  paleo-peat preservational environment. The use of  maceral ratios was proposed as a technique to illustrate the interrelationship of  two parameters (such as humification  versus gelification).  The TPI is a ratio of  macerals that have undergone little humification  (in the case of  this study, textinite, texto-ulminite, and semi-fusinte  are used) versus macerals which are indicative of  advanced humification  (humodetrinite). The GI is a ratio of  gelified  macerals (texto-ulminite, eu-ulminite, macrinite) versus ungelified  macerals (textinite). Cross plots of TPI and GI are then subdivided into interpreted "peat environments of  deposition", based on the supposed characteristics of  peats from specific  environments. This approach was pioneered for  paleoenvironmental interpretations of  Permian coals by Diessel, (1986) and has since been used and abused extensively (e.g. Calder et al., 1991; Kalkreuth et al., 1991; Lamberson et al, 1991; Markic and Sachsenhofer,  1997; Vessey and Bustin, 2000). Modified  maceral indexes, such as the groundwater index (GWI) or the vegetation index were also applied to non-Permian coals (Calder et al., 1991). The GWI is the ratio of strongly gelified  to weakly gelified  tissues and includes the amount of  mineral matter as an indicator of  flooding  activity. The VI contrasts macerals of  forest  affinity  with those of  herbaceous and marginal aquatic affinity.  Although coal maceral indexes have been • used extensively, it has been shown that temperate peats from known depositional environments are not readily classified  by maceral ratios (Hawke et al., 1999) and that plant composition is probably a more important factor  in governing preservation than has previously been thought (Shearer and Moore, 1994). Moreover, no correlation between peat type and depositional environment, climate or tectonic setting occurred in four  bogs from New Zealand (Shearer and Moore, 2003). Maceral ratios have also been shown to give incorrect interpretations of  Tertiary coals (Collinson and Scott, 1987; Crosdale, 1993). The purpose of  this study is to examine the distribution of  organic constituents in polished blocks of  modern peat deposits from tropical Malaysia, and to determine whether the method of  petrographic analysis of  peat (as used in coal studies) can be utilised in reconstruction of  past environments. In order to examine the potential usefulness  of  representing maceral ratio indices of  modern peat deposits, three different cross plots of  maceral contents and maceral ratios were constructed. This organic petrographic study is part of  a larger project, examining the peats from Tasek Bera both in the context of  biologically and ecologically dynamic sediments, and in the context of  coal precursors (Wtist and Bustin, 1999; Wust and Bustin, 2001). A.A.1.2 Organic  composition  of  tropical  peats Climate, depositional environment, mire type, and vegetation type dictate peat composition. Tropical lowland peat deposits are often  dominated by trees and shrubs (e.g. Polak, 1933; Merton, 1962; Polak, 1975; Anderson, 1976; Morley, 1982; Anderson, 1983; FRIM, 1997; Giesen, 1998; Phillips and Bustin, 1998; Wust and Bustin, 1999). material (branches, roots, and stumps with bark) and variable quantities of  sapric, Peats Peats from  Malaysia, Indonesia, Irian Jaya or Thailand have large amounts of  woody from Malaysia, Indonesia, Irian Jaya or Thailand have large amounts of  woody amorphous matrix. The matrix contains residues and fragments  of  fungi,  bacteria, plankton, sponges, fibres,  leaves, roots, cuticles, epidermis, trichomes, spores and pollen. The predominantly forested  tropical peat-forming  environments result in wood-rich peat deposits, unlike most temperate peat deposits, which are often  dominated by shrubs, grasses and bryophytes (Barber, 1993). Tropical peats are therefore  often  rich in lignin with only small amounts of  hemicellulose, cellulose, protein, and water-soluble compounds (Polak, 1975; Orem et al., 1996; Kuder et al., 1998), which are lost due to vigorous microbial activity. It is commonly thought that woody peats generally result in bright, vitrain-rich coal (Taylor et al., 1998, p. 37) and peat derived from bryophytes or Cyperaceae produce dull coal. Well-humified,  inertinite-rich or gelified  organic deposits are thought to result in dull coals (Taylor et al., 1998, p. 292). A.A. 2 Tropical Tasek Bera - physiographic settings, peat accumulation and spatial distribution A.A.2.1 Peat  Formation The wetland basin of  Tasek Bera is located between the eastern and the western mountain ranges in Peninsular Malaysia and covers an area of  552 km2 at an altitude of 30 m above sea level (Fig. A-l). The dendritic, lowland basin occupies the largest natural freshwater  lake system in Malaysia. The basin is covered by 300 km2 of  rubber and oil palm plantations and 252 km of  wetland and lowland rain forest.  The basin measures 27 km from north to south and 13 km at its widest point. Peat accumulation is concentrated along the tributaries but is widespread in the central and northern part of  the basin, where the water level is high year round (Fig. A-l). In the central and northern part, the water level is usually 20 to 100 cm above the soil surface.  During El-Nino events (low precipitation, <100 mm/month), water levels may drop up to 50 cm below the peat surface,  and water flow  is restricted to the channels and lakes. Similar water level changes in mire systems have been documented from Kalimantan (Moore et al., 1996). Because of  the high water retention of  the peat, the organic-rich sediments are saturated during low water level periods and the water table away from  the lake/channel system drops to a maximum depth of  15 cm below the sediment surface.  The peaty topsoils in the southern tributaries may be exposed to air for several weeks during months of  low precipitation. The basal deposits of  the Holocene sedimentary basin that form the floor  of  the peat are composed of  lateritic, kaolinite-rich massive clay, clay-rich quartz sand, and lesser amounts of  clay-rich silt or sand. Radiocarbon data suggests that peat accumulation started some 5000 years BP in the tributaries of  the central and southern Tasek Bera Basin (Wust and Bustin, 1999). Rising water levels, due to various factors,  led to paludification  of  the northern part of  the basin, and the vegetation community changed from a forested  to a palm- (Pandanus  spp.) and sedge-dominated environment (Lepironia spp., Eleocharis  spp.). In the northern part of the basin, peat began accumulating approximately 4000 years BP. Since then, climatic and vegetational changes have led to terrestrialization of  the lakes and channels within the wetland basin and to paludification  of  the deeply weathered, riparian lowland forest  zone (Wust, 2001). Several micro-climatic changes led to vegetation changes and punctuated peat accumulation. Peat accumulation rates increased during times of  abundant precipitation to maximum 2 mm/a, but decreased (to 0.6 mm/a) or ceased altogether during times of  low precipitation (Wiist and Bustin, 1999). The undulating rate of  peat accumulation caused changes in drainage (water level) and thus runoff,  which has a major impact on peat preservation. In the Tasek Bera area, organic-rich deposits from the central and northern areas are often  less than 3 m thick and have high ash contents (>20 wt-%), independent of  the type of  vegetation (swamp forest  and Lepironia - Pandanus  environment). In some southern swamp forest  tributaries, such as in Paya Kuang or Paya Belinau, ash contents are low (<10 wt-%) and peat deposits are thick (2-6 m). The latter deposits are ombrotrophic with low nutrient levels (Wiist and Bustin, 1999). However, none of  the deposits are ombrogenous. to Temertoh 57 km 5? 10.064 -02° 30.681 N 3° 08.49: E 102° 32.( XelatungX, ty^t PAHANG DARULMAKMUR r23 km Paya 'Tasek ; 'Damparj Burung j Bankung Paya Belinau; NEGERI SEMStU DARUL KHUSUS Sungai Ayer lighway 11 Kuning Sungai Palong 12 km Eastern Watershed X Sungai to Temertoh 67 km Western Watershed Sungai Bera |~| raMkB*raS««m lo Cempaka N 3" 00.176 £102° 46.627 The Tasek Bera Basin 12.5 km 17.8 km to Keratong N 2° 54.184 E 102° 45.626 10 km . 6.2 miles ' Figure A-l: Location map of  the Tasek Bera Basin (Tasek = lake) illustrating the dendritic drainage pattern, extent of  peat accumulation (gray shading) and peat swamp tributary names (Paya = swamp, swampy area). Inset map shows the location of  the Tasek Bera Basin in the states of  Pahang and Negeri Sembilan, Peninsular Malaysia. Selected core sites are indicated. Sites B64 and B102 are from a Lepironia articulata and Pandanus helicopus environment, while site B83 is situated in true swamp forest. A.A.2.2 Climate Peninsular Malaysia is situated in the tropical zone and is characterised by high temperatures throughout the year, with an average of  30°C in the lowlands. Precipitation in the vicinity of  Tasek Bera is up to 2600 mm/a, with the average precipitation exceeding 2000 mm/a. However, fluctuations  of  more than 1000 mm can occur from year to year. During the 1992 and 1997 El-Nino years, Felda Bera (4 km southeast of  TB Basin) and Felda Triang (10 km west of  TB Basin) recorded seven and four  months, and seven and six months, respectively, of  low (below 100 mm/month) precipitation (Wiist and Bustin, 1999). The amount of  rainfall  peaks during the two monsoon seasons (Wiist and Bustin, 1999). Rainfall  studies during field  campaigns in 1997 and 1998 revealed that major wind changes (e.g. El-Nino, La Nina) influence  the amount of  precipitation drastically. From June to November 1997, the water level in the Tasek Bera lake fell below the top of  the peat soils, causing widespread death of  phytoplankton and compaction of  the top organic soil layer. In the lowland forest  and swamp forests, temperatures are generally lower, because the dense canopy provides microclimatic regulation, and net water loss due to evapotranspiration is lower than in open reed swamp areas, and so soil and understory vegetation are protected from drying out. A.A.2.3 Biological  diversity  of  the Tasek  Bera wetland Tasek Bera Basin is characterised by three main natural vegetation communities: dry lowland dipterocarp forest  (trees dominated by Dipterocarpaceae), freshwater  swamp forest,  and freshwater  lake system. The dry lowland dipterocarp forest  can be subdivided into two groups: a dipterocarp-dominated area and a mixed-forest  area. The dipterocarp forest,  which covers the lowland hills bordering the swamp area, is dominated by various species such as Shorea  spp., Dipterocarpus  spp., and Koompassia  spp. The mixed-forest  is a transition between the high-ash freshwater  swamp and the dipterocarp forest,  and hosts species of  both forest  types. The freshwater  lake system hosts three different  sub-environments: 1) open water, limnic environment, with patches of  open water (average depth of  2 m) connected by a complex of  channels. The plant assemblages are mainly: Utricularia  flexuosa,  Nymphea sp., Cryptocorynepurpurea  (in channels of  forested  areas), Batrachospermum  spp. 2) A littoral or sedge marsh environment with an average water depth of  0.8 m, with a floral composition dominated by sedges and grasses including Lepironia articulata,  Eleocharis ochrostachys, Zachinellia  spp., Utricularia  spp., and Limnanthemum spp. 3) A pandan-dominated environment along open water bodies and in areas less prone to desiccation during months of  low precipitation dominated by Pandanus  helicopus with subordinate Thoracostachyum  sumatranum. The freshwater  swamp forest  environment is divisible into a high-ash and a low-ash freshwater  peat forest  and has a seasonally fluctuating  water depth (Wiist and Bustin, 1999). High ash swamp forest  is found  mainly in the northern lowland drainage tributaries, such as Paya Kelatung, Paya Kemiyan, or Paya Tebuk (Fig. A-l). The vegetation is dominated by Eugenia  spp., Tristania  spp., Koompassia  spp., Macaranga spp., Ficus  spp., Pandanus  atrocarpus,  and Thoracostachyum  sumatranum. High-ash peat deposits also occur in some tributaries in the southern part of  the basin in the areas of confluence  (e.g. northern part of  Paya Tasek Dampar, Fig. A-l). The high-ash swamp forest  has shorter (rarely taller than 25 m), slender trees with small, sparse open canopies compared to the dense and high canopy lowland forest  (<50 m). The vegetation of  the northern swamp forests  differs  from  the low-ash freshwater  swamp forest  areas in the south such as Paya Tasek Dampar, Paya Belinau or Paya Kuang (Fig. A-l). The dominant species of  the high-ash swamp forest,  Eugenia  spp., are subordinate and sometimes absent for  several hundred metres in the southern low-ash peat swamps. Paya Belinau's dominant tree species are Pandanus  atrocarpus,  Cratoxylon  arborescens, Cyrtostachys  renda,  and Macaranga  puncticulata,  with Thoracostachyum  sumatranum and Ficus  spp. as understory shrubs (Wiist and Bustin, 1999). A.A.3 Methods Three representative samples (core B64, B104 and B83) were collected for  this study using a Macaulay corer. Stratigraphy was described in the field  following  the classification  systems described in Esterle and Ferm (1990) and Wust (2001). Soil pH (Cardy® Model C-l) and Munsell colours were determined. All samples were oven dried at 85° to 105°C for  four  to seven days for  determination of  moisture content. Organic and ash contents were determined by weight loss on ignition after  ashing at 750°C for  2 to 4 hours. The mineralogical analyses of  the peats of  Tasek Bera illustrated that biogenic inorganic material contributes up to >75% of  the total ash content (Wiist et al., 2002), which makes it impossible to quantify  the detrital mineral matter of  the peats. Organic carbon analyses were done on a Carlo Erba NA-1500 analyzer according to the analytical method of  Verardo et al. (1990). The carbon content of  B83 and B102 (Table A-l) was estimated according to the linear correlation of  ash and carbon content as demonstrated in Wiist (2001). Polished blocks were prepared for  petrographic analysis in reflected  light. Representative samples from each of  the macroscopically observable peat facies  were selected for  petrographic analysis. Bulk core sections were freeze-dried  and impregnated with epoxy resin under vacuum (Epo-Thin®, Buehler Technologies), and were subsequently polished using standard coal petrographic methods adapted for  peats (e.g. Stach et al., 1982; Esterle et al., 1991). Fluorescent light was used to conduct organic petrographic point counts, while observations of  mineral matter were made under white light. Results are reported as volume percentages (mineral matter free),  and are based on a count of  300 points with a 0.2 mm step interval. A.A.4 Results and Discussion Three representative cores were selected from the sedge (site B64 and B102) and forest  (site B83) mires. These peat deposits are autochthonous, as is evident from the peat stratigraphy. Most deposits have high macroscopic organic fibre  contents that show little humification,  which is typical for  tropical peat deposits. Peat deposits from the tropics are also known to have higher lignin contents (e.g. textinite) in the basal peat than in the deposits towards the topsoil (Esterle, 1990). A.A.4.1 Pandanus  and Lepironia  environment  (Sites  B64 and B102) The site of  B64 is located in a swamp dominated by Lepironia, Pandanus  and sedges (Fig. A-l). The profile  was water saturated at the time of  sampling. Macroscopically, this profile  (Fig. A-2) consists of  a sapric to fine  hemic surficial  layer with abundant Lepironia roots and root fragments  (0-60 cm). With increasing depth (60-100 cm), Pandanus  roots become predominant. Below 100 cm, the presence of  abundant woody fragments  and leaves within a hemic matrix indicates that the locality was at one time a forested  swamp. The profile  grades into a lowland Dipterocarp  forest (approximately 225 cm depth) with wood and woody fragments,  and subsequently grades into an increasingly inorganic sediment consisting of  sand and silt. The ash yield fluctuates  throughout the profile  and ranges between 30 and 60 wt-% in the peat deposits (Table A-l). The ash content is lowest during the time of  deposition of  peat in the swamp forest  and increases towards the topsoil. Abundant textinite (35%) in the upper peat layer (37-45 cm) is composed mainly of  live and minimally-humified  root tissues, plus relatively unhumified  material deposited at the surface  (Fig. A-3). Textinite increases in the middle peat layer (102-110 cm) to 53%, most of  which is comprised of  large, intact woody fragments  found  in this horizon, and is still prevalent (33%) in the lowermost peat layer (220-228 cm). The humodetrinite content increases from 15% at the surface  to 21% in the swamp forest  peat facies,  and declines to 4% in the dipterocarp forest  facies.  The high humodetrinite content at the surface  is attributed to the high susceptibility of  the surface  vegetation to humification processes (i.e. the cellulose-rich sedges), while the humodetrinite within the swamp forest facies  is consistent with an environment in which aerobic humification  processes are the most significant  diagenetic pathway. Texto-ulminite is a relatively minor constituent (6%), indicating that relatively little gelification  (anaerobic diagenesis) has occurred. Corpohuminite is present in significant  amounts (12%), indicating the presence of  tannins within the peat. The relatively low carbon content (15%) found  at the surface  attests to the unaltered condition of  much of  the material making up the peat. The increase in carbon content with depth correlates with the increasing levels of  humification. The majority of  the liptinites consist of  cutinite, resinite, liptodetrinite and suberinite. Due to the thermally immature nature of  the peats, volatile oils within the plant structures are highly susceptible to reaction with the impregnating resin and with fluorescent  light used in sample preparation, resulting in the formation  of  artifact "resinites" and "exsudatinites". These artifacts  may, in part, account for  the high "resinite" content within the 37-45 cm sample. Inertinite, consisting almost entirely of funginite,  increases with increasing depth (2-21%). The presence of  funginite  indicates that aerobic conditions existed within the peat, as fungi  are obligate aerobes (Brady and Weil, 1996). Figure A-2: Core B64 from Lepironia and Pandanus environment. The site was waterlogged at the time of  core collection. Stratigraphic description and chemical characterization of  the profile  and petrographic composition. Petrographic results are reported on a volume percent basis. The huminites (vitrinites) are often  dominant with various amounts of  liptinites. Inertinites, mainly funginite,  dominate the lower part of this littoral core. B64 Stratigraphy Munsell Characterization color N> UJ 0 -> 50 100 S. © 150 Q 200 -250 J « H I 3/2 4/1 3/2 5/2 7/1 abd roots ot Lepironia small, short root fragments abd roots ot Pandanus long fibers and roots ot Pandanus abd roots abd woody fragments, highly humified abd woody fragments, twigs, leaves woody fragments woody fragments and mud with sand and silt, organic fragments hemic fine hemic sapric organic rich mud a clay wood leaves UTTORAL PANDANUS AND SEDGE SWAMP SWAMP FOREST 50 -100 -150 -(abd = abundant) 2 0 0 -LOWLAND DIPTEROCARP FOREST 250 -3.5 roots Total moisture in wt-% C in wt-% 0 20 40 60 80 100 10152025303540 0 ' I L :/N ratii Carbon 7 0 20 40 60 80 100 10152025303540 0 HT-ash in wt-% C in wt-% r Petrographic analysis in vol-% 10 20 30 40 50 60 70 80 90 100 I I I I I I I I I I I ' M : — 0 20 30 Textinite ] Texto-Ulminite [HI Humodetrinite 80 90 100 Corpohuminite Liptinites Inertinites (Funginite) Site B102 is situated within a sedge marsh, similar to site B64 (Fig. A-l). The peat profile  (Fig. A-3) is approximately 250 cm in depth, and records a vegetation history similar to that of  B64. The sediment grades from white clay (7/1, 10YR) with few organic fragments  at the base to an organic-rich mud with abundant short plant fragments and wood (216-219 cm). The mud is overlain by sapric and fine  hemic peat (145-200 cm) with abundant plant fragments  such as leaves, bark, charcoal, and wood (189-196 cm), as well as pandan fragments.  Two fine  hemic peat units with 100% pandan fragments  (128-145 cm and 100-109 cm) are interbedded with a fine  hemic peat (109-128 cm) with woody fragments  and leaves. This interval is interpreted to have been deposited in a forest  swamp environment. Above 30 cm, the peat has a clay-rich sapric matrix with abundant live Pandanus and Lepironia roots. The ash content of  the peat deposits ranges from less than 20% to over 60% (Table A-l). Hence the carbon content fluctuates  as well, with about 35% throughout most of  the Textinite is less abundant in the upper portion of  the profile  (25%), increases to 67% in the 55-63 cm sample, and subsequently decreases to 16% in the 180-188 cm sample (Fig. A-4). Texto-ulminite increases consistently with depth, from 14-28%. Humodetrinite is consistent within the three uppermost samples (approximately 5%) but increases to 10% in the lowest sample. These trends indicate that in general, alteration of  the peat increases with age, however the peat at the surface  is more susceptible to alteration than the woody peat found  at depth, and that conditions (water level, vegetation, climate) at the surface  are not favourable  to preservation. The majority of  the liptinites present throughout the profile  are liptodetrinites (up to 9%) and sporinites (up to 5%). Unlike the other profiles,  semifusinite  is present in significant  amounts (up to 5% at site B102), indicating that fires  have occurred in the past. Funginite is prevalent (4-20%), especially at the surface  and in the lowermost sample, indicating that intermittent aerobic conditions prevailed. Figure A-3: Core B102 from Lepironia and Pandanus environment. The site was waterlogged at the time of  core collection. Stratigraphic description and chemical characterization of  the profile  and petrographic composition. Petrographic results are reported on a volume percent basis. The huminites (vitrinites) are often  dominant with various amounts of  liptinites. Inertinites, mainly funginite,  dominate the lower part of this littoral core. to B102 o 5 0 -§ loo-ts. u 150-200-Stratigraphy Munsell color 3/1 3/1 2/2 2b0-> V V fibric s. s coarse hemic hemic fine hemic Characterization abd. roots of Pandanus,  Lepironia abd. charcoal fragments abd. roots of Pandanus,  Lepironia abd roots of Pandanus,  Lepironia UTTORAL PANDANUS AND REED SWAMP WITH FOREST FIRES abd roots of Pandanus,  Lepironia abd long roots of Pandanus,  Lepironia abd roots of Pandanus  (100%) no Lepironia  — abd. roots of Pandanus  and woody fragments FOREST (leaves, bark) LITTORAL? abd. roots of Pandanus  and woody fragments (leaves, bark, charcoal, roots, braches) abd. short, small woody fragments (roots, bark, leaves) abd organic fragments few, short organic fragments sapric ^ leaves organic rich mud roots clay (abd = abundant) SWAMP FOREST LOWLAND DIPTEROCARP FOREST wood Total moisture in wt-% Petrographic analysis in vol-% 0 20 40 60 80100 0 10 20 30 40 50 60 70 80 90 100 | Humodetrinite Inertinites (Funginite) Table A-l: Data analyses of  the three representative cores of  the Tasek Bera Basin for petrographic investigations. Core number, sample depth in cm, ash content in wt-%, carbon content in wt-% and pH are given. The carbon content of  B83 and B102 was estimated based on the linear relationship between ash and carbon content of  over 130 samples from the Tasek Bera mire (Wiist, 2001). deposit. r-N , 7 .c 102- 220-Depth in cm 37^45 1 1 Q 2 2 g 125- 220- 310- 435-J U 1 3 0 133 228 318 443 12-20 55-63 Vegetation type Sedges/pandan True Swamp Forest Sedges/pandan Ash content in wt.% Carbon content in wt.% (approx.) PH (H20) 48 44 35 23 25 30 4.7 4.6 4.5 12 5 8 8 8 >40 >45 >45 >45 >45 3.9 4 4.1 4.2 4.5 60 25 20 17 15 33 35 37 4.8 4.5 4.8 4.9 Core # B64 B83 B102 Petrographic analysis in vol.% Textinite Texto-Ulminite Humodetrinite/ Humogelinite Corpohuminite Cutinite Resinite Exsudatinite Suberinite Lipto-detrinite Sporinite Chlorophyllinite Fluorinite Funginite Other Inertinite 35.3 53 33.7 49.3 47.3 38.3 41.7 64.3 24.7 66.3 51 16 6 2.3 0.7 2.7 9.3 3 7.7 4 13.7 13.3 26.7 28.3 15.3 20.7 4.3 17 9.7 14.7 19 12.3 5 5.7 5.3 9.7 12.3 2.3 22.7 0.7 12 2.7 2.3 0.7 4.3 0 2.7 9 4.3 6 3.7 3.7 7.7 3 2.7 0.3 6 3 1.3 4.7 18 0 0.3 1 1.7 2.3 3.3 2.3 0 0 0.7 1.7 0 0 1.7 0 0 0.7 4.7 0 4.3 0 0.7 0.3 1.7 0.7 1.3 0.7 2.7 4 0.3 0.7 1.3 1.3 0.7 1.3 2.3 1.3 6 5 3.3 6 7 5.3 9 4.3 2.7 4 1.3 1 2 7 0 2.7 2.3 1.3 5 1 0.3 2.7 0 0 0.3 0.7 0 0 0.3 0 0.3 0 0 0 1.7 2.3 1.7 0.7 0 0 0 0.7 0.7 0.7 0 0.3 1.7 10.3 21.3 11.3 5.7 23.3 8.7 8 21 4 6.3 21 0 0 0.3 0.3 0.7 0 0 0 4.7 0.3 1.7 1 A.A.4.2 Swamp Forest  Environment  (Site  B83) Sampling site B83 is located within a low-ash, ombrotrophic swamp forest  (Fig. A-l). The profile  (Fig. A-4) consists of  hemic to fine  hemic peat in the uppermost 150 cm, underlain by hemic to sapric peat to a depth of  approximately 520 cm. The contact with the mineral sediment grades from organic-rich mud to clay at 560 cm. Abundant small woody fragments,  leaves and roots in varying stages of  preservation are embedded within the peat matrix throughout the profile.  Ash content is often  less than 10% and carbon content exceeds 40 to 45% (Table A-l). Textinite is the most abundant maceral throughout the profile,  ranging from 38-68% (Fig. A-5). In the surficial  horizon (30-38 cm), much of  this textinite is derived from roots, while at depth the abundant woody fragments  are the major source. Texto-ulminite is present in much smaller amounts, ranging from 3-9%, indicating a lack of  anaerobic bacterial alteration processes that have affected  the peat. Humodetrinite ranges from 10-23%, and is usually found  agglomerated with the clay minerals, diatoms and sponge spicules that make up much of  the hemic to sapric matrix. Corpohuminite is present only in limited amounts in most of  the samples (less than 1-12%). Liptinites consist mainly of  cutinite, liptodetrinite, suberinite, and sporinite, consistent with a hemic to sapric peat derived from wood and leaves. Inertinite is almost exclusively funginite,  and ranges from 5-23%. The high funginite  content is indicative of intermittent aerobic conditions throughout the depositional history of  the peat. Figure A-4: Core B83 from the swamp forest  environment of  Paya Belinau (Fig. 1). The site was waterlogged at the time of  core collection. Stratigraphic description and chemical characterization of  the profile  and petrographic composition. Petrographic results are reported on a volume percent basis. The huminites (vitrinites) are often dominant with various amounts of  liptinites. Inertinites, mainly funginite,  dominate the middle part of  the swamp forest  deposits. K> OJ Stratigraphy Munsell Characterization color 2/2 pH Total moisture in wt-% 3.5 4.0 4.5 5.0 0 20 40 60 80100 0 abd roots ot Lepironia small, short root fragments Petrographic analysis in vol-% 10 20 30 40 50 60 70 80 90 100 I I I I I I abd long roots of Pandanus long fibers and roots, woody long fibers, woody fragments leaves, bark 2.5/2 woody fragments, leaves, bark (5YR) small woody fragments, fibers, wood highly degraded, abd small woody fragments abd fibers highly degraded, abd small woody fragments abd woody fragments dense, abd. woody fragments highly degraded, roots, leaves, bark dense, abd. woody fragments highly degraded, roots, leaves, bark organic-rich mud abd charcoal fragments with sand 100 SWAMP FOREST 300 fine hemic ^ coarse hemic [JJJ sapric E/J5 hemic • organic rich mud LOWLAND 5 5 0 DIPTEROCARP FOREST 600 clay 3.75 4.25 4.75 wood roots leaves (abd = abundant) 0 10 20 30 40 50 60 70 80 90 100 Textinite g Corpohuminite Liptinites • Inertinites (Funginite) 0 20 40 60 80100 HT-ash in wt-% A.A.4.3 Maceral  Ratio  Diagrams Maceral ratios for  interpretation of  peat depositional environments are based on the concept of  diagnostic macerals, which are either plant- or environment-specific  (e.g. Diessel, 1982). The textinite plus texto-ulminite content versus the inertinite content is commonly assumed to delineate peat deposited under prevailing anaerobic versus aerobic conditions because the texto-ulminite content requires anaerobic or low-aerobic conditions, while the inertinite often  indicates fires,  drought or microbial activities. For site B64, inertinite increases with depth (Fig. A-5a). This suggests that anaerobic conditions prevailed during deposition of  the uppermost peat, but that aerobic conditions were more common during the formation  of  the peat that is found  at lower positions within the profile.  Dissolved oxygen measurements (Wiist and Bustin, 1999) show that perennial high water levels prohibit aeration of  the uppermost peat deposits. Changing water levels in the past on an annual or decade scale (water level fluctuations  during monsoon and inter-monsoon time) could have resulted in periodic exposure of  the peat surface  at site B64 (e.g. Wiist and Bustin, 1999). In contrast to the perennial flooded  site B64, the uppermost 7 to 10 cm of  the peat deposits at core site B102 were above the water level during the El-Nino in 1998. For site B102, the textinite plus texto-ulminite content is relatively constant, whereas the inertinite (mainly funginite)  content is highest in the surface  horizon, lowest in the 55-63 cm sample and then gradually increases with depth in the remainder of  the profile.  This suggests that aerobic conditions have recently developed within the upper portion of  the profile,  while the peats below have been subject to intermittent oxidation. The swamp forest  peat (B83, Fig. A-4) shows no pattern of  variability with depth, indicating that conditions within the profile  fluctuated substantially over time and that no simple conclusions can be drawn from using maceral ratios in this case.The results from this study plotted in maceral ratio diagrams produce mixed results (Fig. A-5b). Although sites B102 and B64 represent similar environments in terms of  vegetation and hydrology, the plots suggests very dissimilar environments. TPI's for  site B102 are generally much higher than those obtained for  site B64, while GI's are similar, with the exception of  the highly gelified  basal peat, and to a lesser extent, surface  peat, of  profile  B102 (Fig. A-5b). A similar pattern (as TPI and GI) emerges when texto-ulminite is cross-plotted with inertinite (Fig. A-5c). The B102 profile  shows that both parameters increase concurrently with depth, with the exception of  the surface  sample, which is very high in inertinite, whereas in the B64 profile,  the texto-ulminite content decreases with increasing inertinite content and increasing depth. The B83 profile  does not show any clear pattern in maceral content variation with depth, similar to the other cross plots. A comparison of  the results from these cross plots illustrates some of  the problems inherent to the use of  maceral ratio diagrams to determine paleodepositional environments. Although B64 (Fig. A-2) and B102 (Fig. A-3) are similar vegetational environments at the present time and have similar macroscopic peat compositions (Fig. A-2,7-3), the results are quite different  (Table A-l), suggesting that the two sites have been subjected to differing  postdepositional factors  that control early diagenetic processes. For example, periodic drying of  the profile  and subsequent colonisation by fungi,  or fires  in the nearby dipterocarp forest,  swamp forest  or Pandanus  helicopus, result in increased inertinite contents. During peatification  and coalification,  most of  the readily hydrolysable substances of  the plants are lost, such as cellulose and hemicellulose, while the humic substances become concentrated (Taylor et al., 1998). As a result of  the differing  early diagenetic pathways, original organic petrographic features  of  the profiles, derived, for  example, from the vegetation type, are overprinted. The organic petrographic characteristics become increasingly divergent as time and diagenesis progress, resulting in coals that are entirely dissimilar. In addition, secondary alteration processes occur after burial and further  alter the maceral composition of  peat deposits, rendering maceral ratios almost useless. .•ti c <D 30 25- j 20 15 10 ^ • 220-228 O 12-20 * B64 : 4 B83 O B102 : A O 180-188 220-228 30-38 £ 102- ^ ^310-318 o 110 • : - 435-443 4125-133 o 7-45 5 5 " 6 3 115-122 •30 - 2 5 20 -15 10 S: 0 10 15 20 Textinite + Texto-Ulminite (%) 25 30 0 <3. 2.5-2 1.5 1 0.5-0 :8; 10 1:2 14 16 s la j °180-188 m B64 4 : B83 B fc a).' O B102 • r. o i . . . . . E •P-i Wet I-j Swa orest mp 6 c E: ] 310-318 Q12-20  115-122 ' 3 7 " 4 5 * 2 2 0 - 2 2 8 125-133 D L V F o r e s t O !l 02-110 3 0 " ? 8 ^ 1 435-443 220-228 ° 55-63 4 6: 2.5 K2, 1.5 1 |-0.5 0: 8 T P I 10 12 14 16 0 5 10 15 20 25 30 35 40 45 50 55 Textinite (%)/Texto-Uiminite (%) Figure A-5: Three cross-plots of  maceral ratios commonly used for  paleodepositional interpretation in coal geology. A) Inertinite versus textinite plus texto-ulminite. B) GI versus TPI. The Diessel (1982) paleodepositional environments are indicated. C) Inertinite versus textinite divided by texto-ulminite. Coals formed  in forested  depositional environments are interpreted to have, in general, higher amounts of  structured macerals (textinite) than coals deposited in herbaceous/sedge environments, which often  have high contents of  unstructured macerals. During coalification,  the lignin-rich wood remains structurally better preserved than, for  example, the cellulose-rich tissues of  herbaceous plants (Taylor et al., 1998). The inertinite macerals reflect  the intensity of  microbial activity and thus the degree of humification  or gelification  (Moore et al., 1996). The petrographic analyses indicate that funginite  dominates the inertinite maceral group in the Tasek Bera peat deposits. The inertinite group includes also the macerals fusinite  and semifusinite,  which represent charred humic constituents from fires.  In modern peat environments, such as the Tasek Bera peat deposits where flooding  of  the mire systems is common, charcoal fragments  are abundant and usually allochthonous, transported by water or wind. Floating charcoal fragments  originating from modern fires  have been observed during the field  investigations. Charcoal fragments  can also result from  in situ fires  of  trees and shrubs or subsurface  material. A radiocarbon date, for  example, from a charcoal fragment at a depth of  150 cm suggests that the fragment  belonged to a tree that was burnt in situ about 280 yrs BP (Wiist and Bustin, 2001). Clearly, the original cross plots developed by Diessel cannot be used, as they were intended for  use in Permian coals, not peats; however, if  the TPI/GI has merit for  coals, as it is anticipated, peats from similar peat mire types should have similar TPI/GI characteristics. The maceral indexes devised by Calder et al. (1991) were also considered, but were determined to be inappropriate for  the delineation of  depositional environments within the Tasek Bera peats. The Ground Water Index devised by Calder uses the ash content as a proxy for  flooding,  on the assumption that most of  the mineral matter found  within peat is detrital. However, it is known that a large part of  the ash content of  the peats under study is of  biogenic origin (Wiist, 2001) and not indicative of flooding  but of  vegetation type. The biogenic inorganic component is assumed to represent less than 2% of  the total inorganic fraction  in coals (Diessel, 1992) and thus must likely removed or transformed  during diagenesis (Wiist and Bustin, in rev.). Further, the ash content of  the peats exhibits no detectable pattern of  distribution between environments, and so is not a viable parameter to distinguish between depositional environments. A.A.4.4 Modem  peat depositional  environments  - a key to the past Petrographic analyses show that tropical peats, which are regarded as analogues to the precursors of  most non-Permian coal deposits, have a high textinite content. Tropical peat deposits, such as the ones from Malaysia or Indonesia, have a more woody nature than the widespread, modern boreal bryophytic peat deposits and are thus believed to be precursors of  many thick coal deposits. The present investigation shows that textinite-rich peats also form in sedge and pandan environments where shrubs and trees are absent (e.g. Table A-l, core B64 and B102). The high textinite content in the sedge and pandan environment is most probably due to a high root-input oi Pandanus  helicopus, which undergoes little aerobic humification. Humification  in tropical peats is commonly believed to be very high because of high temperatures and high biomass input (Taylor et al., 1998). However, maceral analysis shows that anaerobic bacterial alteration is limited in the thick and low-ash peat deposits, although intermittent aerobic conditions prevailed during short periods of  time, as indicated by the high (5-23%) inertinite contents, consisting exclusively of  funginite. In both open area deposits (B64, B102), the inertinite content is highest at the base of  the deposit, decreases towards the top, and reaches high values in the top 20 cm. Similar observations were reported from East-Malaysia (Esterle, 1990) and central Kalimantan (Moore etal., 1996). Petrographic analyses also indicate that hydrologic conditions have varied throughout the depositional history of  the peat deposit, and likely correspond to climatic events, such as El-Nino. Although the results indicate that dissimilar environments are likely to have different  organic petrographic characteristics, it is also possible that environments with similar vegetational characteristics will contain different  maceral ratios. Thus, paleoenvironmental interpretations based on maceral ratios must remain tentative. After  burial and diagenesis, coals of  the same paleoenvironment may therefore show dissimilar maceral compositions. For these reasons, paleoenvironmental interpretations of  coals based on maceral ratios should be viewed with caution. A.A.4.5 Implications  for  coal studies Peat studies have been used as templates for  understanding coal composition and depositional environment. Organic petrographic methods of  peat and coal offer  valuable information  about the paleoenvironment. Previous studies have shown that petrographic analysis of  peat deposits can provide information  about bacterial and fungal  activity, droughts, fires,  or depositional conditions (Teichmuller, 1968; Cohen, 1974; Yaekel, 1981; Styan and Bustin, 1983; Esterle, 1990; Chague-Goff  and Fyfe,  1996; Moore et al., 1996; Hawke et al., 1999). The petrographic analyses of  the Tasek Bera peat deposits record valuable information  about vegetational changes as well as depositional and burial conditions. Similar observations during comparisons with palynological investigations were reported from Southern Florida (Cohen and Spackman, 1972). Despite the usefulness  of  the maceral analyses, maceral ratios to interpret modern and ancient peat deposition environments have little value. Maceral ratios for  peat paleoenvironmental analysis do not provide paleo-environmental interpretative tools. Hence it follows  that maceral ratios of  coals at best provide little useful  information  about paleodepositional environment. Similar conclusions were reported by Hawke et al. (1999), who compared maceral ratios and interpretation possibilities of  cold temperate peat deposits. The failure  of  maceral ratios for  peat studies is due to some erroneous assumptions for  interpreting mire types (Crosdale, 1993): These assumptions, discussed in the context of  the current study, are: 1) Structured macerals imply woody vegetation. The Tasek Bera peat also has high textinite contents derived from littoral vegetation, in particular pandan species and to a lesser extent sedges. 2) Oxidised macerals are related to the height of  the mire surface  above the groundwater level. For example, the domed deposits of  Kalimantan (Moore and Hilbert, 1992) showed little inertinite contents whilst the less domed deposits had abundant inertinite (Moore et al., 1996). In the Tasek Bera area, the inertinite content has no trend, and the water table changes are frequent  and ephemeral. Field observations show that the water level may drop below the soil surface  during extreme climatic conditions (e.g. El-Nino) for  short periods of  time regardless of  the depositional environment. In addition, detrital charcoal fragments  (eolian or water-borne) may also contribute to the total inertinite content. 3) Gelified  macerals result from groundwater fluctuations  and thus depends on mire type. In the Tasek Bera area, gelification  is often  enhanced in the central and northern peat deposits where major water level fluctuations  occur. Comparing GI and TPI values of core B102 (Fig. A-3) with core B64 (Fig. A-2) suggests a dissimilar depositional environment although both sample sites are from similar depositional environments. In addition, the basal deposits are highly humified  (sapric or fine  hemic) and thus more humified  than the middle organic deposits. Hence, unstructured macerals are more abundant than in the less humified  peat deposits. Similar decomposition stages are found in the top peat deposits. The swamp forest  peat deposits, on the other hand, show little variations in TPI or GI (Fig. A-5b). Moreover, the maceral ratios from the swamp forest environment have similar TPI and GI values as the sedge peat deposits. Hence, differentiating  the swamp forest  from the sedge mire depositional environment is impossible using maceral ratios such as the GI or TPI. The importance of  the water level on the composition of  peat has also been documented from New Zealand peat bogs (Shearer and Moore, 2003). 4) Detrital macerals imply transportation. The Tasek Bera peat deposits are autochthonous and show fragmentation  of  plant material in situ with little "detrital" material. Detrital macerals are most common in the swamp forest  peat where little to no transportation occurs due to the fact  that the deposits are not flooded  (e.g. Wiist and Bustin, 1999). The detrital macerals have been shown previously to form in-situ from lignin-poor or cellulose-rich particulate matter, such as leaves, or stalks from sedges (Crosdale, 1993). A.A.5 Conclusions The application of  maceral ratios to modern tropical peat-forming  environments failed  to fully  characterize the present and past environment. In addition, maceral ratios were not able to clearly delineate swamp forest  from  Pandanus  or Lepironia environments, although peat composition and humification  (i.e. pH, CEC, Eh, DOC, etc., see Figs. 7-2 to 7-4) are remarkably different  (Wust and Bustin, 1999). Previous investigations on temperate peats (Hawke et al., 1999) and Miocene coals (Crosdale, 1993) have also questioned the usefulness  of  maceral ratios for  delineating paleo-vegetational environments. During diagenesis, e.g. peatification  and coalification, biochemical and geochemical alterations will change the maceral composition, concentrating the more resistant lignin-rich macerals, which may eventually result in similar maceral ratios. Our study shows that maceral ratios of  modern peat-forming environments do not allow interpretations about depositional environment. In coal geology, maceral ratios rely on some assumptions that are not supported by analyses of modern peat deposits. Therefore,  paleoenvironmental interpretations of  coals based on maceral ratios rely on assumptions that are not supported by the analysis of  modern peat deposits. A.A.6 References Anderson, J.A.R., 1976. Observations on the ecology of  five  peat swamp forests  in Sumatra and Kalimantan, Peat and podzolic soils and their potential for  agriculture in Indonesia, Tugu, pp. 45-55. Anderson, J.A.R., 1983. The tropical peat swamps of  Western Malesia. In: A.J.P. Gore (Editor), Mires: Swamp, Bog, Fen  and Moor,  Regional Studies. Ecosystems of  the World.  Elsevier Scientific  Publishing Co., Amsterdam, pp. 181-199. Barber, K.E., 1993. Peatlands as scientific  archives of  past biodiversity. Biodiversity and Conservation 2, 474-489. Brady, N.C. and Weil, R.R., 1996. The nature and properties of  soils. Prentice Hall, Upper Saddle River, NY, 740 pp. Calder, J.H., Gibling, M.R. and Mukhopadhyay, P.K., 1991. Peat formation  in a Westphalian B piedmont setting, Cumberland basin, Nova Scotia: implications for  the maceral-based interpretation of  rheotrophic and raised paleomires. Bull. Soc. geol. France  162(2), 283-298. Chague-Goff,  C. and Fyfe,  W.S., 1996. Geochemical and petrographical characteristics of  a domed bog, Nova Scotia: a modern analogue for  temperate coal deposits. Organic Geochemistry 24(2), 141-158. Cohen, A.D., 1974. Petrography and paleoecology of  Holocene peats from the Okefenokee  swamp-marsh complex of  Georgia. Journal  of  Sedimentary Petrology  44(3), 716-726. Cohen, A.D. and Spackman, W., 1972. Methods in peat petrology and their application to reconstruction of paleoenvironments. Geological Society of  America Bulletin 83, 129-142. Cohen, A.D., Spackman, W. and Raymond, R., 1987. Interpreting the characteristics of  coal seams from chemical, physical and petrographic studies of  peat deposits. In: A.C. Scott (Editor), Coal and Coal-bearing Strata: Recent Advances. Blackwell Scientific  Publications, Oxford,  pp. 107-126. Collinson, M.E. and Scott, A.C., 1987. Implications of  vegetational change through the geological record on models for  coal-forming  environments. In: A.C. Scott (Editor), Coal and coal-bearing strata; recent advances. Geological Society Special Publications. Geological Society of  London, London, United Kingdom, pp. 67-85. Crosdale, P.J., 1993. Coal maceral ratios as indicators of  environment of  deposition: do they work for ombrogenous mires? An example from the Miocene of  New Zealand. Organic Geochemistry 20(6), 797-809. Diessel, C.F.K., 1982. An appraisal of  coal facies  based on maceral characteristics. Australian Coal Geology 4,474-483. Diessel, C.F.K., 1986. The correlation between coal facies  and depositional environments. Advances in the study of  the Sydney Basin, 20th Symposium, Univ. of  Newcastle, pp. 11-22. Diessel, C.F.K., 1992. Coal-bearing depositional systems. Springer-Verlag, Berlin, 721 pp. Esterle, J.S., 1990. Trends in petrographic and chemical characteristics of  tropical domed peats in Indonesia and Malaysia as analogues for  coal formation.  Ph.D. Thesis, University of  Kentucky, Lexington, KY, USA, 270 pp. Esterle, J.S. and Ferm, J.C., 1990. On the use of  modern tropical domed peats as analogues for  petrographic variation in Carboniferous  coal beds. International  Journal  of  Coal Geology 16, 131-136. Esterle, J.S., Moore, T.A. and Hower, J.C., 1991. A reflected-light  petrographic technique for  peats. Journal  of  Sedimentary Petrology  61(4), 614-615. FRIM, 1997. A consultancy report on the extent and types of  vegetation cover at Tasek Bera, Forest Research Institute of  Malaysia for  Wetlands International-Asia Pacific,  Kuala Lumpur, Malaysia. Giesen, W., 1998. The habitats and flora  of  Tasik Bera, Malaysia: An evaluation of  their conservation value and management requirements, Wetlands International-Asia Pacific,  Kuala Lumpur, Malaysia. Goodarzi, F. and Hill, P. A., 1989. Intermontane coal at Hat Creek, British Columbia: Review of  macerals, minerals, tonsteins, elements, International Symposium on Intermontane Basins: Geology & Resources, Chiang Mai, Thailand, pp. 89-98. Grady, W.C., Eble, C.F. and Neuzil, S.G, 1993. Brown coal maceral distributions in a modern domed tropical Indonesian peat and a comparison with maceral distributions in Middle Pennsylvanian-age Appalachian bituminous coal beds, pp. 63-82. Hawke, M.I., Martini, LP. and Stasiuk, L.D., 1996. Petrographic characteristics of  selected Ontario peats: possible modern analogues for  coals, 13 th annual meeting TSOP, abstracts and program, Southern Illinois University, Carbondale, Illinois, pp. 22-23. Hawke, M.I., Martini, I.P. and Stasiuk, L.D., 1999. A comparison of  temperate and boreal peats from Ontario, Canada: possible modern analogues for  Permian coals. International  Journal  of  Coal Geology 41, 213-238. Kalkreuth, W.D., Marchioni, D.L., Calder, J.H, Lamberson, M.N, Naylor, R.D. and Paul, J, 1991. The relationship between coal petrography and depositional environments from selected coal basins in Canada. International  Journal  of  Coal Geology 19, 21-76. Kuder, T, Kruge, M.A, Shearer, J.C. and Miller, S.L, 1998. Environmental and botanical controls on peatification-a  comparative study of  two New Zealand restiad bogs using Py-GC/MS, petrography and fungal  analysis. International  Journal  of  Coal Geology 37, 3-27. Lamberson, M.N, Bustin, R.M. and Kalkreuth, W, 1991. Lithotype (maceral) composition and variation as correlated with paleo-wetland environments, Gates Formation, northeastern British Columbia, Canada. International  Journal  of  Coal Geology 18, 87-124. Markic, M. and Sachsenhofer,  R.F, 1997. Petrographic composition and depositional environments of  the Pliocene Velenje lignite seam (Slovenia). International  Journal  of  Coal Geology 33, 229-254. Merton, F, 1962. A visit to Tasek Bera. Malayan  Nature  Journal  16, 103 - 110. Moore, T.A. and Hilbert, R.E, 1992. Petrographic and anatomical characteristics of  plant material from two deposits of  Holocene and Miocene age, Kalimantan, Indonesia. Review of  Paleobotany and Palynology  72, 199-227. Moore, T.A. and Shearer, J, 2003. Peat/coal type and depositional environment—are they related?, International  Journal  of  Coal Geology 56, 233-252 Moore, T.A, Shearer, J. and Miller, S, 1996. Fungal origin of  oxidised plant material in the Palankaraya peat deposit, Kalimantan Tengah, Indonesia: Implications for  "internite" formation  in coal. International Journal  of  Coal Geology 30, 1-23. Morley, R.J, 1982. Origin and history of  Tasek Bera. In: J.I. Furtado and S. Mori (Editors), TasikBera: The  Ecology of  a Freshwater  Swamp. Monographiae  Biologicae. Dr. W. Junk Publishers, The Hague, NL, pp. 12-45. Orem, W.H, Neuzil, S.G, Lerch, H.E. and Cecil, C.B, 1996. Experimental early-stage coalification  of  a peat sample and a peatified  wood sample from Indonesia. Organic Geochemistry 24(2), 111-125. Phillips, S. and Bustin, R.M, 1998. Accumulation of  organic rich sediments in a dendritic fluvial/lacustrine mire system at Tasik Bera, Malaysia: Implications for  coal formation.  International  Journal  of  Coal Geology 36, 31-61. Polak, B, 1975. Character and occurrence of  peat deposits in the Malaysian tropics. Modern  Quaternary Research Southeast Asia 2,71-81. Polak, E, 1933: Ueber Torf  und Moor in Niederlandisch Indien. Verhandelingen der koninklijke Akademie van Wetenschappen te Amsterdam, XXX, No. 3. Van de N.V. Noord-Hollandische Uitgevers-Maatschappi, Amsterdam, 84 pp. Shearer, J.C. and Clarkson, B.R, 1998. Whangamarino wetland: effects  of  lowered river levels on peat and vegetation. International  Peat Journal  8, 52-65. Shearer, J.C. and Moore, T.A, 1994. Botanical control on banding character in two New Zealand coal beds. Palaeogeography,  Palaeoclimatology,  Palaeoecology  110, 11-27. Shearer, J.C. and Moore, T.A, in rev. Coal type and depositional environment - are they related? International  Journal  of  Coal Geology. Stach, E, Mackowsky, M.T, Teichmuller, M, Taylor, G.H, Chandra, D. and Teichmiiller, R, 1982. Stach's Textbook of  Coal Petrology. Gebruder Borntrager, Berlin, 535 pp. Styan, W.B. and Bustin, R.M, 1983. Petrography of  some Fraser Delta peat deposits: Coal maceral and microlithotype precursors in temperate-climate peats. International  Journal  of  Coal Geology 2, 321-370. Taylor, G.H, Teichmuller, M, Davis, A, Diessel, C.F.K, Littke, R. and Robert, P, 1998. Organic Petrology. Gebruder Borntrager, Berlin, 704 pp. Teichmuller, M, 1968. Zur Petrographie und Diagenese eines fast  200 m machtigen Torfprofils  (mit Ubergangen zur Weichbraunkohle?) im Quartar von Philippi (Mazedonien). Geologische Mitteilungen  8(1), 65-100. Verardo, D.J, Froelich, P.N. and Mclntyre, A, 1990. Determination of  organic carbon and nitrogen in marine sediments using the Carlo Erba NA-1500 Analyzer. Deep-Sea Research 37(1), 157-165. Vessey, S.J. and Bustin, R.M, 2000. Sedimentology of  the coal-bearing Mist Mountain Formation, Line Creek, Southern Canadian Cordillera: relationship to coal quality. International  Journal  of  Coal Geology 42, 129-158. WUst, R.A.J, 2001. Holocene evolution of  the intermontane Tasek Bera peat deposit, Peninsular. Malaysia: Controls on composition and accumulation of  a tropical freshwater  peat deposit. Ph.D. Thesis, University of British Columbia, Vancouver, B.C., Canada. Wiist, R.A.J. and Bustin, R.M, 1999. Geological and ecological evolution of  the Tasek Bera (Peninsular-Malaysia) wetland basin since the Holocene: evidences of  a dynamic system from siliciclastic and organic sediments, Wetlands International-Asia Pacific,  Kuala Lumpur, Malaysia. Wiist, R.A.J, and Bustin, R.M, 2001. Low ash peat deposits from a dendritic, intermontane basin in the tropics: A new model for  good quality coals. International  Journal  of  Coal Geology 46(3-4), 179-206. Wiist, R.A.J. and Bustin, R.M, 2003. Opaline and Al-Si-bioliths from a tropical topogenous mire system of West-Malaysia: Their abundance, habit, elemental composition, preservation and significance  for geochemical peat analysis and coal studies. Chemical Geology Wust, R.A.J, Ward, C.R, Bustin, R.M. and Hawke, M.I, 2002. Characterization and quantification  of inorganic constituents of  tropical peats and organic-rich deposits from Tasek Bera (Peninsular Malaysia): implications for  coals. International Journal of  Coal Geology, 49: 215-249. Yaekel, J.D, 1981. Geochemistry and petrography of  peats from the Okefenokee  Swamp, Georgia and the Everglades and coastal regions of  southern Florida. Ph.D. Thesis, The Pennsylvania State LJniversity, Pennsylvania, PA, USA. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0052634/manifest

Comment

Related Items