UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The geochemistry of Fe, Mn, Ni, Cu, Zn and As in the water column, sediments and porewaters in a seasonally.. 1996

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_1997-0073.pdf
ubc_1997-0073.pdf [ 11.51MB ]
Metadata
JSON: 1.0052322.json
JSON-LD: 1.0052322+ld.json
RDF/XML (Pretty): 1.0052322.xml
RDF/JSON: 1.0052322+rdf.json
Turtle: 1.0052322+rdf-turtle.txt
N-Triples: 1.0052322+rdf-ntriples.txt
Citation
1.0052322.ris

Full Text

THE GEOCHEMISTRY OF Fe, Mn, Ni, Cu, Zn AND As IN THE WATER COLUMN, SEDIMENTS AND POREWATERS IN A SEASONALLY ANOXIC LAKE by A L A N MARTIN B.Sc, The University of British Columbia, 1991 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER V OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Earth and Ocean Sciences) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1996 © Alan Martin, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or . her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my _ written permission. Department of £^ r r#7 CH^/ Oc&2tn ?>0<£<s>rr*< The University of British Columbia Vancouver, Canada Date XfM/ DE-6 (2/88) A B S T R A C T The distributions of Fe, Mn, Ni, Cu, Zn and As in the water column, interstitial waters and associated solid phases in Balmer Lake, Ontario, were determined from samples collected in July and October of 1993, and March and May of 1994, in order to assess the seasonal biogeochemical controls governing trace metal behaviour and mobility. The basin has served as a repository for tailings pond effluents since 1967, and as a result, hosts elevated levels of contaminants in the sediments and lake waters. During the ice-free periods of summer, fall and spring, the water column is characterized by fully oxygenated bottom waters and homogeneous distributions of all measured parameters. However, reducing conditions develop in deeper areas during the period of ice cover in response to the high biological demand of the organic-rich sediments coupled with restricted atmospheric exchange. Trace metal profiles collected during winter exhibit considerable lake- wide variation and appear to reflect variability in the duration and extent of bottom water anoxia, and the relative influence of metal-rich lateral inputs from the tailings circuits. Solid-phase profiles indicate that the top decimetre of the sediment column has received variable contributions of both organic matter- and feldspar- rich natural detritus, and carbonate-, chlorite- and metal-rich tailings inputs. High resolution profiles of porewater N O 3 " , Fe, Mn and SO4 2 " illustrate that the sediments in Balmer Lake become anoxic within a few centimetres of the sediment-water interface during well-mixed periods. Evidence for the precipitation of a Cu-bearing sulphide phase in the lowermost winter bottom waters suggests that the sulphate redox-cline migrates above the sediment-water interface at some point during ice cover. The seasonal porewater distributions of ii dissolved Ni, Cu and Zn exhibit pronounced consumption profiles at depths consistent with sulphate reduction zones, suggesting they are sequestered as metal-sulphide phases. Enrichments of dissolved Ni and Zn in the surficial sediments at most sites reflect the dissolution of labile particulates at the sediment-water interface. Diffusive influxes calculated for Zn, Ni and Cu suggest that diffusion mechanisms contribute insignificantly to the accumulation rates of these metals. Arsenic is remobilized at deeper sediment depths and appears to be largely governed by the redox geochemistry of Mn. The data collectively demonstrate that trace metal mobility in the Balmer Lake water column/sediment system varies significantly over the course of the four seasons. With the exception of arsenic, however, dissolved metal fluxes indicate that the underlying contaminated sediments are providing a significant and permanent sink for dissolved metals. iii TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF FIGURES .7. viii LIST OF TABLES xiv ACKNOWLEDGEMENTS xvi I. INTRODUCTION 1 1.1 Trace Metal Geochemistry 1 1.2 Trace Metal Cycling in Lakes 3 1.2.1 Inputs of Trace Elements to Lacustrine Systems 4 1.2.2 Major Processes Governing Trace Metal Behaviour and Mobility 5 1.2.3 Transformations Across the Sediment-Water Interface 10 1.3 Contamination of Aquatic Systems From Metal Mining 13 1.3.1 Bioavailability and Toxicity 14 1.4 Research Objectives 15 II. STUDY SITE 18 2.1 Physiography, Physical and Chemical Limnology 18 2.2 History of Mine-Related Inputs 20 2.2.1 Campbell Mine 21 iv 2.2.2 Arthur White Mine 22 2.3 Sampling Periods 23 2.4 Sampling Sites '. 24 III. SAMPLING AND METHODS 26 3.1 Sampling 26 3.1.1 Water Column 26 3.1.2 Physical Profiling 27 3.1.3 Porewaters 27 3.1.3.1 Peeper Description and Preparation 29 3.1.3.2 Peeper Deployment and Subsampling. 32 3.1.4 Sediment Sampling '. 34 3.2 Instrumentation 35 3.2.1 ICP/MS 35 3.2.2 GFAAS 36 3.2.3 Ion Chromatography 36 3.2.4 X-ray Fluorescence 36 3.2.5 Oculometry 37 3.2.6 Carbon, Nitrogen and Sulphur 37 3.2.7 Spectrophotometry 37 3.2.8 Digestion of Polycarbonate Filters 38 IV. RESULTS 39 4.1 Water Column 39 4.1.1 Physical Profiling 39 4.1.2 Nitrate, Ammonium and Sulphate 42 4.1.3 Suspended Organics 53 v 4.1.4 Trace Metals 57 4.1.4.1 Dissolved Fraction 57 4.1.4.2 Particulate Fraction 64 4.2 Interstitial Waters 69 4.2.1 Peepers 69 4.2.1.1 Nutrients and Sulphate 69 4.2.1.2 Trace Metals 79 4.2.1.3 pH 95 4.2.2 Core Porewaters 97 4.2.2.1 Nutrients and Sulphate 97 4.2.2.2 Trace Metals 99 4.3 Sediments 101 V. DISCUSSION 109 5.1 Water Column 109 5.1.1 Physical Limnology I l l 5.1.2 Sulphate I l l 5.1.3 Ammonium and Nitrate 113 5.1.4 Suspended Particulates 117 5.1.5 Trace Metals 121 5.1.5.1 Iron 122 5.1.5.2 Manganese 124 5.1.5.3 Arsenic 126 5.1.5.4 Nickel 127 5.1.5.5 Copper 129' 5.1.5.6 Zinc 130 5.1.6 Water Column Model 131 vi 5.1.7 Comparison of Four Lake Zones 135 5.2 Sediment Geochemistry 142 5.3 Porewater Chemistry 147 5.3.1 Ammonium and Nitrate 149 5.3.2 Sulphate 153 5.3.3 Iron and Manganese 157 5.3.4 Nickel 169 5.3.5 Copper 174 5.3.6 Zinc 179 5.4 Factors Controlling the Diagenetic Behaviour of As 186 5.4.1 Experimental and Field Observations 186 5.4.2 The Behaviour of As in Balmer Lake Sediments 194 VI. SUMMARY A N D CONCLUSIONS 202 VII. BIBLIOGRAPHY 205 VIII. APPENDICES 220 Appendix A. Typical ICP/MS operating conditions 220 Appendix B. Quality Assessment/Quality Control 221 Appendix C. Porewater Data 225 Appendix D. Sediment Data 235 Appendix E. Core Logs 237 Appendix F. Diffusive Fluxes 239 Appendix G. Accumulation Rates 243 vii LIST OF FIGURES Fig. 2.1 Location map showing the sampling stations, lake bathymetry and tailings circuits of the adjacent mines. p<r' ^ Figure 3.1. Schematic diagram of membrane dialysis sampler (peeper). Pfl- - ^ ° Fig. 4.1 Summer water column profiles of temperature, dissolved oxygen and pH for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, June, 1993. The hatched line represents the sediment surface. f>f • 4*° Fig. 4.2 Fall water column profiles of temperature and dissolved oxygen for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, October, 1993. The hatched line represents the sediment surface. Fig. 4.3 Winter water column profiles of temperature and dissolved oxygen for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, March, 1994. The hatched line represents the sediment surface. ^ ' Fig. 4.4 Spring water column profiles of temperature, dissolved oxygen and pH for stations 1, 2 and 4 (A-C, respectively), Balmer Lake, May, 1994. The hatched line represents the sediment surface. Pf • ¥ ^ Fig. 4.5. Winter water column profiles of nitrate for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. p<j .sv Fig. 4.6. Winter water column profiles of ammonium for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. 57 Fig. 4.7. Winter water column profiles of sulphate for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface.^. £2 viii Fig. 4.8. Winter water column profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 1, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. ^ Fig. 4.9. Winter water column profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 2, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. Pft' & ° Fig. 4.10. Winter water column profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 3, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. pt' <*1 Fig. 4.11. Winter water column profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 4, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. P%- ̂ Fig. 4.12. Duplicate summer peeper profiles of dissolved NH4+, NO3" and SCH^- for station 1 (A-C) and station 2 (D-F), Balmer Lake, June, 1993. Replicate samples are represented by double symbols at specific single depths. - ^ ° Fig. 4.13. Fall peeper profiles of dissolved NH4+, NO3" and SO42- for stations 1 (A-C) and 5 (D-F), Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. pft • T'2- Fig. 4.14. Duplicate fall peeper profiles of dissolved NH4+, NO3" and SO42" (A-C, respectively) for station 2, Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. pft- ^ 3 Fig. 4.15 Winter peeper profiles of dissolved NH4+, N03~ and SO42- for stations 1 (A- C) and 2 (D-F), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. ix Fig. 4.16. Winter peeper profiles of dissolved NH4 + , NO3" and S04^_ for stations 4 (A- C) and 6 (D-F), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. ^ ' Fig. 4.17. Duplicate spring peeper profiles of dissolved NH4+, NO3" and S04^- (A-C, respectively) for station 1, Balmer Lake, May, 1994. Replicate samples are represented by double symbols at specific single depths. ^ ' Fig. 4.18. Duplicate summer peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A- F, respectively) for station 1, Balmer Lake, June, 1993. Replicate samples are represented by double symbols at specific single depths. . b° Fig. 4.19. Duplicate summer peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A- F, respectively) for station 2, Balmer Lake, June, 1993. Replicate samples are represented by double symbols at specific single depths. Ff-6^ Fig. 4.20. Fall peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 1, Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. pq- 5 5 Fig. 4.21. Duplicate fall peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 2, Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. - 6 ^ Fig. 4.22. Fall peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 5, Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. f*§ - ^ Fig. 4.23. Winter peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 1, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. Fig. 4.24. Winter peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 2, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. p%. x Fig. 4.25. Winter peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 4, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. PQ'**° Fig. 4.26. Winter peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 6, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. P^- Fig. 4.27. Duplicate spring peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 1, Balmer Lake, May, 1994. Replicate samples are represented by double symbols at specific single depths. ' ^ Fig. 4.28. Spring profile of porewater and bottom water pH measured from peeper 2, station 1, Balmer Lake, May, 1994. f% • f 6 Fig. 4.29. Distributions of dissolved NO3- and SO42" obtained from peeper (open circles) and core porewaters (diamonds) at station 1 (A-B) and station 6 (C-D), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. Pff-9* Fig. 4.30. Dissolved distributions of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) obtained from peeper (open circles) and core porewaters (diamonds) at station 1, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. Pf- / o € > Fig. 4.31. Dissolved distributions of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) obtained from peeper (open circles) and core porewaters (diamonds) at station 6, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. Pft • / o 2 - Fig. 4.32. Sedimentary weight ratio profiles of Si/Al, Ti/Al, Mg/Al, K/Al, Na/Al and P/Al (A-F, respectively) for station 1 (duplicate cores la and lb, open symbols) and station 6 (core 2, closed symbols), Balmer Lake, March, 1994. f^. /"V xi Fig. 4.33. Sedimentary weight ratio profiles of Zr/Al, Sr/Al and Rb/Al (A-C, respectively) for station 1 (duplicate cores la and lb, open symbols) and station 6 (core 2, closed symbols), Balmer Lake, March, 1994. Pf- '°5 Fig. 4.34. Sedimentary concentrations of organic carbon, calcium carbonate, total nitrogen, values for organic carbon/nitrogen, total sulphur and iron/aluminum weight ratios (A-F, respectively), for station 1 (duplicate cores la and lb, open symbols) and station 6 (core 2, closed symbols), Balmer Lake, March, 1994. p^- / 0 < a Fig. 4.35. Sedimentary concentrations of manganese, nickel, zinc, copper, arsenic, and lead (A-F, respectively) for station 1 (duplicate cores la and lb, open symbols) and station 6 (core 2, closed symbols), Balmer Lake, March, 1994. Fig. 5.1. Time series of progressive trace metal water column profiles (A-E) over the fall-winter transition in Balmer Lake: A) well-mixed water column, B) influence of remobilized bottom-water source, C) influence of interfacial sulphide sink, D) greater extent of sulphide removal, and E) influence of metal-rich lateral advective flow. Depth and concentration axes are in arbitrary units. rf • Fig. 5.2. Seasonal peeper profiles of ammonium for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. P% • >s2- Fig. 5.3. Seasonal peeper profiles of nitrate for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. Pf- 1 5 * Fig. 5.4. Seasonal peeper profiles of sulphate for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. n- / r 6 Fig. 5.5. Seasonal peeper profiles of dissolved Fe for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. pft • ^° xii Fig. 5.6. Winter peeper profiles of dissolved Fe for stations 1, 2, 4 and 6 in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. p̂ . Ibt Fig. 5.7. Seasonal peeper profiles of dissolved Mn for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. f^.i^ Fig. 5.8. Time series of progressive bottom water and porewater profiles (A-D) over the fall-winter transition in Balmer Lake: A) fall profile, B) instantaneous input of Mn to bottom waters, C) profile after some time of equilibration, and D) late winter profile. Depth and concentration axes are in arbitrary units. Pf/6>^ Fig. 5.9. Seasonal peeper profiles of dissolved Ni for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. Pfi"' Fig. 5.10. Seasonal peeper profiles of dissolved Cu for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. P%' Fig. 5.11. Seasonal peeper profiles of dissolved Zn for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. Prj- / & l Fig. 5.12. Summer peeper profiles of dissolved Fe, Mn and As for station 2 in Balmer Lake. Replicate samples are represented by double symbols at specific single depths, pg. fif Fig. 5.13. X-Y scatter plot illustrating the dependence of dissolved As on the distribution of dissolved Mn. The values (molar concentraions) represent a depth interval spanning from ~3 cm above the sediment surface to ~20 cm below. Coefficients of determination (r2) from linear regression analyses for the respective stations are included. p<g / < ? ^ Fig. 5.14. Seasonal peeper profiles of dissolved As for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. ill x i i i LIST OF TABLES Table 2.1. Estimated average yearly ranges of mine-related inputs from the Campbell Mine tailings system to Balmer Lake, Ontario, before and after chemical treatment. All values expressed in mg/L. Pft-22- Table 2.2. Estimated average yearly concentrations of mine-related inputs from the Arthur White tailings system to Balmer Lake, Ontario. All values expressed in mg/L. ft- 3 3 > Table 2.3. Summer, fall, winter and spring sampling periods at Balmer Lake, Ontario, indicating stations of peeper deployment, coring and water column sampling. P?" z Table 4.1. Summer water column distributions of dissolved Mn, Fe, Ni, Cu, Zn, As, Pb, NH4+, N03", S042" and P O 4 3 - for stations 1, 2, 3, and 4 in Balmer Lake, Ontario, June 1993. All values are expressed in u\g/L unless specified otherwise (BDL denotes below detection limit). F%• ^b Table 4.2. Fall water column distributions of dissolved Mn, Fe, Ni, Cu, Zn, As, Pb, NH4+, NO3- , S O 4 2 - and P O 4 3 - for stations 1, 2, 3, 4 and 5 in Balmer Lake, Ontario, October, 1993. All values are expressed in | ig /L unless specified otherwise (BDL denotes below detection limits). Pf- ̂  ̂ Table 4.3. Winter water column distributions of dissolved Mn, Fe, Ni, Cu, Zn, As, Pb, NH4+, NO3- , SO4 2 - and P O 4 3 - for stations 1, 2, 3 and 4 in Balmer Lake, Ontario, March, 1993. All values are expressed in ug/L unless specified otherwise (ND denotes not determined). P(f" ¥ ® Table 4.4. Spring water column distributions of dissolved Mn, Fe, Ni, Cu, Zn, As, Pb, NH4+, N03', S O 4 2 - and P 0 4 3 ' for stations 1, 2 and 4 in Balmer Lake, Ontario, May, 1993. All values are expressed in ug/L unless specified otherwise (ND denotes not determined). • s * xiv Table 4.5. Concentrations of total suspended solids (TSS), particulate organic carbon (SPOC), particulate organic nitrogen (SPON), Corg^itrogen ratio (C/N) and total particulate organic matter (POM) in the water column of Balmer Lake, Ontario, for the summer, fall, winter and spring field sessions. TSS values were determined from the particulate mass on pre-weighed 0.45 um Nuclepore filters, while all organics were determined from glass fibre filters. P%' ss~ Table 4.6. Particulate trace metal concentrations of Fe, Mn, Ni, Cu, Zn, As and Pb for the summer, fall, winter and spring water columns in Balmer Lake, Ontario. All values are expressed in ug-L~l. Pf - Table 4.7. Concentrations of trace metals (mg-kg"l), organic carbon (%) and organic nitrogen (%) in suspended particulates of the summer, fall, winter and spring water columns in Balmer Lake, Ontario. Trace metal values were determined from the particulate mass on pre-weighed 0.45 um Nuclepore filters, while all SPOC and SPON values were determined from glass fibre filters, p*. (af- xv A C K N O W L E D G E M E N T S This project was jointly jointly funded by Placer Dome Inc. and Gold Goldcorp Inc. Special thanks goes to Richard Janicki, Dave Hiller and Elias Dibb of Campbell Mine, Red Lake, who helped ensure the success of the fieldwork. We also thank Jim Robertson (PDI, Vancouver) for his interest and continued support. I wish to thank Drs. Tom Pedersen, Steve Calvert, George Poling and Bill Cullen of my supervisory committee for their various degrees of input into my thesis. In particular, I deeply express my thanks to Tom Pedersen who is not the uncultured, useless, dirty, stinking heathen everyone says he is. I will always have fond memories of sitting in his office, Tom reclining back in his chair, hands clasped and brought to his mouth, his eyebrows progressively lowering over his piercing stare as I presented him with my latest views. "You have done a good job" he would say, as he passed you your crimsoned document, the pages weighted and thickened by the rigorous onslaught. I also extend my gratitude to Steve Calvert, whose insight and depth would intimidate the likes of Carl Sagan and Mr. Spock. As for Bill Cullen, I appreciate the fact that he left the country as I heard he can be quite intimidating at oral examinations. Additional thanks goes to my fellow research group members, Remy Chretien, Darcy McDonald, Jay McNee, Raja Ganeshram, Robert Mugo, Brad Mckelvey, Maureen Soon and Bert "high sticking" Mueller, for making my graduate years (all four of them) a lot of fun. I will also never forget John "Leatherman" Ridley, whose mighty power aided in both the peeper deployments and his ability in snow wrestling. The boys (you know who you are) should also be praised for their efforts in getting me out of the house. I dedicate my thesis to my partner in crime, Debbie, who not only proofread this document, but ensured I received regular doses of mountain air throughout my degree. xvi I. INTRODUCTION 1.1 Trace Metal Geochemistry The study of trace metal geochemistry is driven partly by curiosity, and as a result of the inherent toxicity of many metal species, partly by an obligation to the maintenance of human health and the well-being of the environment. The last decades have seen a large increase in the burden of trace metals in the environment, and as a consequence, enrichments now commonly exist in terrestrial and aquatic reservoirs (Ferguson and Gavis, 1972; Nriagu, 1989). Considerable attention has been given to the fates and effects of toxic metals. In recent times, the major research objectives on metal-polluted systems have progressed from initial surveys of sources and pathways toward more detailed investigations of the mechanisms controlling the distribution, mobility and bioavailability of different metal species. Knowledge of the distribution and behaviour of trace elements in aquatic and marine systems has grown significantly over the past two decades, largely in part to advancements in sampling techniques and analytical methods. The development of clean collection techniques (Boyle and Edmond, 1977; Bruland et al, 1979; Schaule and Patterson, 1981; Nriagu et al, 1993) and sensitive analytical instrumentation has facilitated the reliable measurement of trace constituents; the use of clean methods has also exposed inaccuracies in previously reported data. As a result, well-defined vertical and horizontal variations in concentrations of dissolved components for the world's oceans now exist for the majority of trace elements. Prior to the implementation of clean techniques, the state of trace metal biogeochemistry of lacustrine systems suffered similar misrepresentation. Contamination artifacts introduced during the collection, 1 handling and analyses of samples severely compromised published data acquired in the absence of rigorous methods. The assumption that stringent, ultraclean laboratory methods are not required in "polluted" waters warrants close scrutiny following recent studies of the Great Lakes which report trace element concentrations comparable to, or even lower than, the levels commonly observed in the pelagic ocean (Nriagu, 1996; Flegal, 1988; Coale, 1989). In the open ocean, dissolved trace element distributions are generally considered to be in steady state (Murray, 1987). Integrated, systematic studies have been used to determine the biogeochemical and physical processes governing the vertical and horizontal distributions of many trace elements (Bruland, 1980; McKelvey, 1994). Application of one-dimensional, vertical advection diffusion models fitted to oceanic profiles allow the extraction of chemical parameters such as production and scavenging rates (Morel and Hudson, 1985). In addition, oceanographic investigations have revealed the usefulness of many trace elements as effective markers of various marine processes. The science of limnology has evolved greatly since the original pioneering works of Forel (1892, 1895, 1904), Mortimer (1941, 1942, 1951) and Hutchinson (1957). In spite of significant progress, chemical limnology has not yet reached the same state of sophistication as its marine counterpart. Essentially the same mechanisms act to control trace element distributions in lacustrine settings; however, trace metal distributions are driven by time-dependent processes which vary from days to seasons (Balistrieri and Murray, 1992). The variability induced by the transient nature of lake processes can make steady state assumptions invalid, and as a result, may inhibit the isolation of the variables controlling trace metal cycles (Murray, 1987). Furthermore, the dynamic nature of lakes hinders comparisons of different systems due to variations in factors 2 such as the surrounding geology and trophic status. Defined patterns of trace metal distributions in lakes are further hindered by the greater availability of competing scavenging phases (e.g., detrital particles, phytoplankton and authigenic precipitates) and compressed depth scales. Although the transient nature of lake processes adds extra complexities, time-dependent variations (e.g., oxic/anoxic cycles) can be taken advantage of to establish the dynamics of biogeochemical cycles (Balistrieri and Murray, 1992). 1.2 Trace Metal Cycling in Lakes The intrinsic properties of a particular metal (e.g. crystal field stabilization energy, electronegativity, ionic radius, etc.) and the environmental conditions prevalent (e.g., types and concentrations of ligands present, available adsorptive surface areas, pH, pe) are ultimately responsible for the distribution of particular species and their corresponding redox and coordination chemistries (Stumm and Morgan, 1981). Systematic studies of trace elements in lakes provide information on the physical, chemical and biological processes controlling their respective mobility and distribution. These processes include diffusion, advective transport, nutrient-like biological cycling, sorption by sediments or suspended particles, precipitation, atmospheric deposition, redox cycling, and fluxes across the sediment-water or air-water interface (Murray, 1987). The relative importance of these mechanisms depends on the prevalent abiotic and biotic environment as well as the element under consideration. Spatial and temporal changes in the chemical environment of lacustrine systems can strongly influence biogeochemical pathways of various metal species. For example, lakes may possess chemical gradients with depth (e.g., 3 dissolved O y S0 4 , etc.) and exhibit significant seasonal differences in such major variables as p H and pe (Jenne, 1986). Furthermore, seasonal variations in productivity can foster corresponding shifts in pe and concentrations of organic products. Increases in autochthonous particles for instance, may provide additional complexation sites for trace metals, resulting in lower concentrations of dissolved metal species (Francois, 1990). These multiple influences are briefly elaborated upon in the general review of trace element cycling in lake systems that is presented in the following sections. A brief introduction to the sources of trace metal inputs will be followed by more detailed descriptions delineating the processes governing metal behaviour and mobility in both the water column and underlying sediments and porewaters. 1.2.1 Inputs of Trace Elements to Lacustrine Systems Trace metals are incorporated into lake systems primarily via tributary inflow, ground water inputs, coastal runoff, industrial and domestic effluent discharges and atmospheric fallout. Natural sources are primarily water or wind-borne soil particles, seasalt aerosols, volcanism, forest fires and biogenic particulate matter (Nriagu, 1989) . Nriagu (1989) estimated that aerosols of biogenic origin dominate the natural fraction and can account for 30-50% of the global baseline emissions of trace metals in non-urban areas. Hydrothermal activity has also been shown to provide significant quantities of trace metals to specific fresh waters (Agett and O'Brien, 1985). Background metal levels in inland waters due to weathering processes are more difficult to establish than in the marine environment as a result of varying geology and fluctuations in water discharge (Baccini, 1984). In contrast, anthropogenically enriched levels of 4 metals are derived from domestic wastewater effluents, mining-related inputs, gasoline combustion, coal burning power plants, non-ferrous metal smelters, iron and steel plants and the dumping and incineration of sewage sludge (Nriagu, 1989). 1.2.2 Major Processes Governing Trace Metal Behaviour and Mobility The distribution of trace metals between solution and particulate forms has widely been recognized as a major factor controlling the geochemical behaviour, transport and biological effects of metals in natural waters. Many trace metals (e.g. Cu, Pb, Hg, Cd, Zn, Ag) have been shown to be primarily associated with suspended and sedimentary particulates as opposed to dissolved phases (Davis and Leckie, 1978b; Nriagu et al, 1981; Laxen, 1985; Tessier et al., 1985). Particulates in natural waters consist predominantly of detrital particulate and colloidal organic matter, inorganic solids such as metal oxides and hydroxides (e.g., Si02, MnC>2, FeOOH, A I 2 O 3 ) , algal skeletal remains, carbonates and detrital aluminosilicates (e.g., clay minerals, feldspars) (Tessier et al, 1985). In general , particulate organic matter (POM) often presents the major component of suspended material, accounting for up to 70% of the total particulate fraction (Nriagu et al, 1982). Settling particles, especially organic aggregates, can play a dominant role in the binding and transfer of heavy metals to lake sediments, thereby regulating the concentrations of dissolved species (Nriagu et al, 1981; Sigg, 1985; Jackson and Bistricki, 1995). In general, inverse relationships between the residence times of particulate metals and the standing crops of suspended matter in the water column are evident for most lakes (Sigg, 1985). Variations in the residence times of various metals primarily reflect differences in their leachability from source 5 particulates and in the rates of biotal assimilation (Nriagu et al., 1982). Sigg (1985) and more recently Morel and Hudson (1985) have extended the Redfield ratio concept (Redfield, 1958) to include trace elements, suggesting a stoichiometric formula of C106N16P1 (Fe, Zn, Mn)n.oi (Cu, Cd, Ni , Co) o.ooi- Sigg (1985) suggested that algal-metal associations could result in metal uptake with constant stoichiometric proportions. Although the concept of Redfield ratios may be applicable to the metal content of algae, the dynamic nature of lakes limits its usefulness because only under severely restricted conditions will the stoichiometry be reflected in dissolved metal concentrations (Reynolds and Hamilton-Taylor, 1992). Cyclic variability in the composition of the total suspended matter is evident in many temperate lakes. In winter and early spring, isothermal conditions typically predominate and storm events can resuspend quantities of fine sedimentary matter (Baccini, 1984). This particulate fraction may have undergone considerable diagenetic alteration and be characterized by relatively low organic carbon and high mineral contents (e.g., clays). With the onset of large-scale vernal primary production, phytoplankton, zooplankton and the associated detrital material may dominate the particle pool. During periods of thermal stratification, the epilimnion is largely decoupled from the underlying colder waters of the hypolimnion resulting in unidirectional, downward transport of particulate matter. During the latter stages of stratification in the late summer and fall, the p H may be driven high enough by the production of organic matter to promote the precipitation of CaC03 in surface waters. Inorganically precipitated calcite has been shown to account for 20-90% of the particle pool in the Great Lakes during such periods (Eadie and Robbins, 1987). The influence of biologically- and chemically-mediated redox reactions on the behaviour of trace elements has received considerable attention owing to the 6 pronounced effect such transformations have on the spatial and temporal distributions and on the speciation of such elements in solution, suspended particulates and sediments. The predominant mechanisms driving the redox cycling of elements in lakes are plankton growth in surface waters and the bacterially-mediated degradation of organic matter in sub-surface waters and sediments (Hamilton-Taylor and Davison, 1994). Redox-driven biogeochemical pathways of elements may involve direct redox transformations (e.g., As(V) -» As(III)) in which the metal species itself undergoes a redox reaction (De Vitre et al, 1991). Alternatively, metal species may be subject to indirect involvement, in which the distribution of the element is affected by redox-driven cycling, but apparently does not itself (e.g., Zn) undergo a redox transformation (Reynolds and Hamilton-Taylor, 1992). The redox behaviour of iron and manganese in lakes has received considerable attention as a result of the central role that these abundant metals play in the geochemical cycling of trace elements (Sholkovitz, 1985; Leckie, 1986; Davison, 1993; Hamilton-Taylor and Davison, 1994). Both elements are common to natural waters in discrete mineral phases and in surface coatings on particulate material, usually occurring in oxic conditions as insoluble oxides at their higher oxidation states. Their reduced states, Fe2+ and Mn2+, are soluble and relatively uncomplexed (Davison et al, 1982). Sedimentary iron and manganese oxyhydroxides exhibit high affinities for metal ions (e.g., As, Ni, Cu, Cd, Zn, and Pb) and have been repeatedly inferred to be sinks for trace metals (Crecelius, 1975; Nriagu et al, 1981; Davison and Woof, 1984; Tessier et al, 1985; Tessier et al, 1996). The sorptive affinity of trace metals for oxide surfaces is strongly influenced by pH (Vuceta and Morgan, 1978), ionic strength (Stumm and Sulzberger, 1992), adsorbed ligands (Davis and Leckie, 1978a; Vuceta and Morgan, 1978) and humic substances (Tipping and Cooke, 1982). Theoretical and 7 experimental works on solid-solution interactions of trace metals with amorphous iron oxyhydroxide demonstrate the existence of a multiplicity of discrete binding sites with wide-ranging adsorptive characteristics (Davis and Leckie, 1978b; Benjamin and Leckie, 1981). The preferred adsorption sites for such trace elements as Cd, Zn, Cu, and Pb are apparently distinct from one another, which explains why competitive effects observed are weaker than one might otherwise expect. This heterogeneity of complexation sites has been explained in terms of the number and identity of nearest neighbours, proximity to various crystal defects, microcrystallinity, local electrical field strength, and physical constraints such as the size of pores leading to internal sites (Benjamin and Leckie, 1981). Seasonal variations in the redox chemistry of iron and manganese in lakes can result in pronounced inter-annual variability in the distribution, behaviour and mobility of many trace elements. Non-steady state redox environments induced by productivity pulses and changes in local hydrography can significantly affect the distribution of their respective soluble and particulate phases (Tipping et al, 1981). Meromictic lakes offer a range of redox conditions which vary spatially and temporally, and consequently, such basins are particular well suited to studying nuances of redox processes (Davison and Woof, 1984). Temperate lakes tend to exhibit seasonal stratification of the water column in the summer months which is induced by surface heating of a usually shallow mixed layer (Baccini, 1984). The redox boundary in a seasonally stratified lake may migrate from a horizon within the sediments under well- mixed conditions upwards into the water column during episodes of maximum stratification (Laxen, 1985). During unstratified conditions, oxygen may be present at the sediment-water interface, favouring the formation of Mn and Fe oxyhydroxides in the surficial sediments (Davison et al., 1982). Upon burial of 8 the sediment, diagenetic respiration processes deplete oxygen and promote a reducing environment. The reductive dissolution of oxyhydroxides in reducing sediments promotes the mobilization of Fe(II) and Mn(II) and the concomitant mobilization of sorbed metals. In the absence of further chemical reactions, Fe(II), Mn(II) and the associated trace metals may diffuse upward along concentration gradients towards the sediment-water interface, where the predominant reactions are reprecipitation under oxic conditions, and release into the hypolimnion when anoxic conditions prevail (Agett and O'Brien, 1985; Agett and Kriegman, 1988). In general, as the oxic zone of the sediments is compressed, retention of metals in the sediments primarily associated with metal oxides is decreased. Those elements with lower ionization potentials (i.e. decreased capability to take up electrons and thus a reduced tendency to form complexes) are more readily mobilized and thus enter solution at higher pe values (Stumm and Morgan, 1981). During the later stages of stratification, the hypolimnion may become sufficiently devoid of oxygen to enable the dissolved species to diffuse across the sediment-water interface where they may accumulate until overturn (Davison and Woof, 1984; Agett and Kriegman, 1988). The duration of stratification and the volume of the hypolimnion strongly influence the fluctuations in the concentrations of dissolved trace metal species (Davison 1982). The misinterpretation of diagenetic metal enrichments as anthropogenically related features is, however, a potential problem in mildly reducing, Mn- and Fe-rich lake sediments. 9 1.2.3 Transformations Across the Sediment-Water Interface Bottom sediments of lakes represent complex mixtures of diverse components that include: residues of weathering and erosion (e.g., aluminosilicates), inorganic and organic products of biological activity, and diagenetic products such as iron and manganese oxyhydroxides and sulphides (Tessier and Campbell, 1988). Trace metals may be scavenged from the water column and incorporated into the bottom sediments via several mechanisms including: sorption onto hydrous metal oxides, ion exchange with clay minerals, sorption on organically coated particulates, adsorption to or incorporation with detrital organic matter, and via diffusive influxes along concentration gradients. The particular physical and biogeochemical characters that dictate sediment geochemistry ultimately determine the nature of the various diagenetic (i.e., post depositional) reactions and resulting authigenic fractions. Authigenic phases encompass a wide spectrum of inorganically-derived precipitates which form in the water column or in the sediments after deposition. Organic matter constitutes a very important sedimentary component, as its diagenesis during decomposition provides the primary impetus behind the post-depositional reactions responsible for the formation of numerous hydrogenous components (Suess, 1979). Organic matter degradation has additional ramifications with respect to nutrient regeneration and benthic community structure (Emerson, 1976; Froelich et al, 1979). Studies of the microbially-mediated decomposition of organic matter have led to the development of a well-established model in which the distributions of oxidants, reductants, and metabolites with depth follow a predictable and universal sequence, given an elastic depth scale. Detailed accounts have been described by several authors (Sholkovitz, 1973; Froelich et al, 1979; 10 Klinkhammer, 1980; Van der Weijden, 1992). Briefly, this degradation follows a thermodynamic order of sequential oxidation reactions which yield progressively lower free energy changes per mole of organic carbon oxidized. In practice, the reaction sequence is subject to oxidant availability and kinetic constraints, including the dominance of particular microbial populations. As previously stated, thermodynamics dictates that relative pore water distributions of metabolites will be consistent from site to site with the only variable being the depth scale over which these reactions occur. Organic-rich sediments, characterized by high diagenetic intensity and thus steep thermodynamic gradients, may, however, exhibit significant overlap in the sequence of oxidation reactions (Froelich et al, 1979). A recent evaluation of the redox zonation during the degradation of organic matter assessed the controls responsible for observations of the simultaneous reduction of Fe(III) and sulphate in several field studies (Postma and Jakobsen, 1996). They suggest that the sequential occurrence of these processes can be better explained by a partial equilibrium approach where initial fermentation determines the overall rate (i.e., the reactivity and amount of organic matter) while equilibrium is approached by the terminal electron accepting processes. They concluded that depending on the stability of the iron oxide and porewater pH, sulphate may be reduced simultaneously with, or even prior to Fe(III). The flux of organic matter to the sediments exerts the primary control on diagenetic intensity and has a corresponding influence on metabolite distribution. In environments of rapid sediment accumulation and those with high input rates of organic matter, interstitial metabolite concentrations (e.g., H C O 3 - , H P O 4 2 " , HS") may exceed the solubilities of certain inorganic equilibria 11 resulting in the formation of authigenic minerals and thus preferential removal of one or several of the dissolved constituents (Suess, 1979; Berner, 1984). Thermodynamic calculations on the bulk solution of most freshwater and seawater typically show undersaturation with respect to known metal solid phases (Tessier et al. 1985). Even for interstitial waters of oxic sediments, in which trace metals have longer residence times than in the bulk solution and are thus more apt to approach a state of equilibrium with solid phases in the sediments, metals are usually undersaturated with respect to their least soluble compounds (Schindler, 1981). Trace metal adsorption to solid surfaces has been invoked to account for observed distributions. The trace metal-particulate association may include: sorption of metal ions at oxide surface sites, ion exchange within clay minerals, binding by organically-coated particulate matter or organic colloidal material, or adsorption of a metal-ligand complex. In particular, the important roles played by iron and manganese oxyhydroxides and humic materials in the adsorption of trace metals in soils and sediments have been strongly emphasized due to their high sorptive properties and ubiquitous presence in natural systems . Thermodynamic modeling of interstitial waters is a useful technique for identifying possible controls on authigenic mineral formation. Experimentally derived ionic strengths, activity coefficients, and activities of various interstitial metabolites are often used to calculate the controlling mechanisms responsible for the distribution of various aqueous and solid fractions. Production and consumption rates of interstitial metabolites can be used to constrain further thermodynamic gradients and redox pathways. The availability of powerful mathematical tools which incorporate variables such as saturation indices, distribution coefficients and diffusive fluxes, aid in the development of conceptual models of sediment geochemistry. In more complex natural settings, 12 however, calculated values often deviate from the observed as complications arise from mixed mineral phases, adsorption and/or ion exchange equilibria (Matisoffef al, 1980). 1.3 Contairrination of Aquatic Systems From Metal Mining While estimates of trace element emission rates to the environment differ (Lantzy and Mackenzie, 1979; Jaworowski et al, 1981; Zoller, 1984; Nriagu, 1989), it is agreed that for most toxic metals emissions from industrial activities are larger than natural fluxes. Human operations have thus become the key agent in the global atmospheric cycle of trace metals and metalloids. For example, a comparison of world-wide median emission values from natural and anthropogenic sources suggests that industrial emissions of Pb, Cd and Zn exceed the flux from natural sources by factors of 18, 5 and 3, respectively (Nriagu 1989). Further, contributions from anthropogenic sources of As, Hg, Ni, Sb and V may exceed those from natural sources by 100-200% (Nriagu 1989). The metal mining industry contributes significantly to the anthropogenic input of many trace elements to the Earth's atmospheric, terrestrial and aquatic systems (Nriagu, 1989). Impacts resulting from such activities can be highly localized, such as an acid-generating waste-rock pile, or can extend over hundreds of kilometres via emissions from the smelting of ore concentrates. In the case of sulphide-bearing ore deposits, the potential for loadings of acidity and metals to the environment presents the primary concern. In particular, smelter emissions, acidic drainages and the erosion of waste rock deposits can foster extensive and lasting environmental impact (Aimer et al, 1978; Salomons, 1995). 13 1.3.1 Bioavailability and Toxicity The uptake of trace elements by biological systems is a speciation-sensitive process for which chemical specificity varies from one element to another (Turner, 1984; Campbell et al., 1988). Bioavailable forms include free metal ions, organometallic species including organic complexes, and inorganic complexes including hydrolysis products (Turner 1984). Though much remains to be learned about the mechanisms of trace element uptake and nutrition in phytoplankton, the assimilation of trace metals appears to be in most cases a two stage process involving the binding of a relatively large pool of metal to the cell surface and transport through the cell membrane, presumably via metal-porter molecules (Morel and Hudson, 1985). The exact nature of either the surface binding groups or the porter molecules remains questionable. Seemingly, a significant fraction of the surface ligands must be complexed by a particular metal for that species to be taken up, the extent of binding being related to the intracellular assimilation (Turner 1984). Indiscriminate uptake of non-essential metal species appears to be related to the relatively low specificity of metal transport mechanisms and thus presents an important mode of toxicity (Harrison and Morel, 1983). In addition to the major nutrient elements, many trace metals including Fe, Mn, Zn, Cu and Ni have been determined to be essential for life, and it has been argued that many major and trace elements, both essential and toxic, may be simultaneously controlling biological production in the oceans (Morel and Hudson 1985). These trace metals exhibit fairly narrow "concentration windows" that separate essential and toxic levels (Florence 1982). In the open ocean where concentrations are low, toxic effects are negligible; however, in coastal regions and lakes, metal inputs from weathering and anthropogenic processes result in 14 substantially higher metal concentrations that may exceed toxic thresholds (Elinder, 1984). 1.4 Research Objectives Many trace elements exhibit toxic effects towards plant and animal species when present in sufficiently high concentrations. Particular attention has been given to the environmental significance of several heavy metals and metalloids (e.g., As, Hg, Pb, Ni , Zn, Cd, Cr) in view of their inherent toxicity and common occurrence (Elinder, 1984; Cullen and Reimer, 1989). Lake sediments represent the ultimate site of deposition for many trace elements introduced into the environment. Since an important fraction of the trace metals present in the aquatic environment is reversibly associated with the surficial sediments (Campbell et al., 1988), the study of trace metal transformations and dynamics across the sediment-water interface is prerequisite to the understanding of their behaviour in whole aquatic systems. The primary impetus behind this research was to investigate the poorly constrained biogeochemical cycling of metals within the water-sediment system in Balmer Lake. It is anticipated that the project will provide basic insight into the cycling of metals in shallow, boreal contaminated lakes as well as data which can be used to assess both the overall impact of the final discharge from Balmer Lake on the local water system, and prospects for mitigation of the contamination problem over the long term. More specifically, the research comprises three principal objectives: 1) To determine the biogeochemical processes governing the behaviour and mobility of trace metals in Balmer Lake sediments and assess the extent to which these metals are mobile; 2) To determine the direction and significance of benthic fluxes of trace metals; and 3) 15 To assess how the chemistry of the water column, underlying sediments and associated porewaters, change over the course of a full four-season cycle. A seasonal study of Balmer Lake was essential in order to resolve the nature and extent of geochemical variations in response to seasonal changes in hydrography, oxygenation and plankton productivity. In particular, it was desired to link periods of seasonal anoxia and the influence of redox-related reactions on metal mobility and diagenetic chemistry. To meet these objectives, a two component research program was implemented: first, a seasonal assessment of the water column chemistry and second, a detailed seasonal investigation of the chemistry of the sediments and associated porewaters. The latter involved high resolution sampling of the interstitial waters and bottom waters. Balmer Lake provides an ideal in situ laboratory to address the nature of metal transfer mechanisms and partitioning across the sediment-water interface for two reasons: first, there is a large signal of metal contaminants; and second, the large benthic area to volume ratio makes the chemistry of the lake suceptible to influences from interfacial fluxes. To determine accurately the nature of post-depositional reactivity, detailed analyses of both solid and dissolved phases are essential (Colley et al, 1984; Carignan and Nriagu, 1985). Undetectable changes in the sediment fraction can significantly alter the chemical composition of the associated porewaters which can thus provide a sensitive indicator of the diagenetic reactions and equilibria between solid phases and dissolved species (Matisoff et al. 1980). In addition, the residence times of trace elements in the interstitial fraction are longer than in the overlying waters; porewaters are therefore more apt to approach a state of equilibrium with solid sediment phases (Schindler, 1981). Consequently, the composition of interstitial waters can be used to identify and quantify mineral equilibria and thus infer chemical mass transport (Matisoff et al. 16 1980). Porewater studies done in conjunction with investigations of the water column can provide a link between water column transport processes and sedimentary accumulation by showing evidence for the release of trace elements associated with the degradation of biogenic particulate matter (Shaw et al, 1990). 17 II. STUDY SITE 2.1 Physiography, Physical and Chemical Limnology Balmer Lake is a small, shallow, boreal lake situated in the Canadian Shield of northwestern Ontario (Fig. 2.1). The waterbody is 260 hectares in area, has an estimated mean depth of 2.4 m and a maximum basin depth of 4.0 m. Typical of shallow Precambrian shield lakes, Balmer Lake is productive and hosts organic-rich sediments (Conroy and Keller, 1976). The shallow water column maintains fairly well-mixed conditions during the ice-free periods of late spring, summer and fall. Periods of bottom water anoxia develop in the winter months in deeper areas (>2.5 m) of the ice-covered lake. The waterbody has served as a repository of mining-related wastes for two underground gold mines since 1965 (Fig. 2.1). Both mines process high grade ores rich in Ni, Cu, Zn and As, and as a result, concentrations of these metals are significantly enriched in the lake waters and sediments. In addition, the lake hosts high inventories of other mining-related byproducts, including cyanide, ammonia, nitrate and sulphate. The water retention time in the lake is approximately 230 days, which aids the natural degradation and settling of cyanide and heavy metals before final discharge to the local water system. In the design of the tailings circuit, the lake was intended to receive only dissolved and colloidal loads. However, historical dam breaches of the surrounding tailings ponds have resulted in significant and episodic particulate loadings to areas of the lake. Over the period of mining activity, the lake has supported few fish and is not used recreationally. Since January of 1994, only the Arthur White Mine (Fig. 2.1) has been discharging effluents that have exceeded the recommended 18 Tailings Discharge Balmer Creek Fig, 2.1 Location map of Balmer Lake showing the sampling stations, lake bathymetry and tailings circuits of the adjacent mines. 19 C O A (Certificate of Approval) limits with respect to total arsenic and heavy metals. The general topography of the Balmer Lake region has been described as flat to rolling, with lowland coniferous and upland mixed forest cover. Examinations of the Quaternary geology show that the soil types are dominated by glacial deposits of bouldery to silty till one to several metres thick. The area is also characterized by scattered bedrock outcrops and glaciolacustrine deposits of clay and silt deposited in Lake Agassiz (Gulson et al., 1993). Balmer Lake receives inputs from three main drainage basins, encompassing a total watershed area of 3,290 hectares (Masala, 1995 ). The natural drainage area comprises 2,550 ha and has been influenced to a moderate degree by road construction, timber harvesting and mining exploration, contributing to breaks in the natural forest cover. The remaining two drainage basins are represented by the Campbell and Dickenson mine sites which comprise active and revegetated tailings, clear water (polishing) ponds, and parts of the mines and townsite. Inputs from these drainages are predominantly mine-related with a small fraction being natural runoff (Masala, 1995a). 2.2 History of Mine-Related Inputs The Archean greenstone belts in the Red Lake area of northwestern Ontario have been mined for their gold deposits since the gold rush to the district in 1926. Ore-grade gold mineralization in the area generally occurs in three specific associations: (i) massive ferroan-dolomite veins that contain gold + quartz + arsenopyrite in crosscutting veins; (ii) silicified patches with disseminated gold; and (iii) sulphide disseminations (arsenopyrite, pyrite, and lesser pyrrhotite) (Gulson et al., 1993). More detailed summaries of the geology 20 of the Red Lake area can be found in Colvine et al. (1984), Corfu and Wallace (1986) and Corfu and Andrews (1987). Production mining in the area of the Balmer Lake began in 1948 upon operation of the Arthur White Mine. Balmer Lake, however, did not receive mining-related inputs until 1965, at which time, tailings were deposited along the basin of the south shore of the lake. A rock berm was subsequently constructed in 1967 to contain particulate mill effluents. Since this time, the waterbody has served as a tertiary polishing pond for mining wastes from both the Campbell (Placer Dome Inc.) and Arthur White (Goldcorp Inc.) mines. 2.2.1 Campbell Mine The Campbell Mine is an underground gold mine which uses sub-level longhole and cut-and-fill ore extraction methods. Ore processing proceeds via cyanidatioh with carbon-in-pulp and Merrill-Crowe Gold Collection Processes. The coarse tails fraction (~ 460 tons per day (TPD)) is combined with flyash and cement for underground backfilling while the fines (« 470 TPD) are sent to the tailings impoundment and settling ponds. Prior to January, 1993, decanted excess water from the main tailings pond proceeded through of a series of polishing ponds (including Balmer Lake) before final discharge to the environment at the Balmer Lake outlet (Fig. 2.1). However, this system was insufficient in reducing contaminant concentrations of Ni, Cu, Zn, As and CN" to within C O A limits on a consistent basis. Subsequently, Placer Dome Inc. commissioned the installation of a wastewater treatment plant which commenced operations in 1993. The chemical process utilizes INCO SO2 technology (Devuyst et ah, 1982) and brought effluent parameters to within guideline specifications, thus effectively removing Balmer Lake from the 21 Campbell tailings circuit. A comparison of the effluent quality to Balmer Lake before and after chemical treatment is presented in Table 2.1. Table 2.1. Estimated average yearly ranges of mine-related inputs from the Campbell Mine tailings system to Balmer Lake, Ontario, before and after chemical treatment. All values expressed in mg/L. Year Total Cyanide Total Ni Total Cu Total Zn Total As Prior to treatment 1992 0.5-30 2-6 1-7 0.3-1.5 0.2-0.5 1993 0.5-25 0.4-2.0 0.1-8 0.05-1.5 0.01-0.5 After treatment 1994 0.4-1.5 0.05-0.3 0.05-0.2 0.02-0.06 0.01-0.03 1995 0.1-0.6 0.04-0.07 0.02-0.08 0.02-0.04 0.01-0.03 2.2.2 Arthur White Mine The Arthur White Mine began production mining in 1948 and shares a similar mining protocol to its PDI counterpart. Ore production, rated at approximately 800 TPD, is achieved by shrinkage, horizontal hydraulic cut and fill, and long-hole stoping mining methods. Arthur White also possesses its own internal primary tailings disposal system, which consists of series of inter- connected ponds (Fig. 2.1). Clarified effluent from the primary tailings area overflows the primary dam and drains into a secondary polishing pond. The drainage then proceeds through two rockfill seepage-dams before final discharge to the tertiary polishing pond (Balmer Lake). At the present time, no effluent 22 treatment facilities exist at the mine, and as a result, inputs from this system present the most significant contributor of heavy metals, sulphate and cyanide to Balmer Lake. Typical effluent concentrations of various parameters are presented in table 2.2. Table 2.2. Estimated average yearly concentrations of mine-related inputs from the Arthur White tailings system to Balmer Lake, Ontario. All values expressed in mg/L. Year Total Cyanide Total Ni Total Cu Total Zn Total As 1983 9 0.9 1.0 0.7 0.4 1984 15 1.1 3.2 1.1 1.3 1985 7 0.2 0.4 0.1 0.5 1986 3 0.4 1.1 0.3 0.8 1987 5 0.6 1.0 0.3 0.4 1988 2 0.6 0.9 0.1 0.8 1989 4 0.9 2.1 0.2 1.3 1990 15 1.1 3.3 1.1 1.3 1991 8 0.9 1.1 0.7 1.1 2.3 Sampling Periods Four sampling periods encompassing a summer-fall-winter-spring transition were carried out in order to represent accurately the full seasonal spectrum of the chemical environment in Balmer Lake. The summer sampling 23 period (Jun-Jul, 1993) was chosen to coincide with a potentially stratified water column, the condition being common to many temperate lakes in the summer months. The fall session (Oct-Nov, 1993) was planned to follow complete breakdown of stratification developed in the summer; sampling was performed just prior to freeze-up and after convective overturn of the shallow water column. In order to observe the maximal extent of sub-ice stratification established during the winter months, sampling was carried out just prior to ice- off in March-April of 1994. Gravity coring was also performed during this period to take advantage of the stable surface offered by the ice; this permitted slow and precise corer insertion. The spring sampling period was carried out two weeks subsequent to ice-off (May 1994) and was designed to capture a post- thaw episode of elevated heavy metal and cyanide levels previously observed during this time. 2.4 Sampling Sites A survey of Balmer Lake in 1990 (PDI, 1990) established a suite of 45 permanent sampling sites, five of which were chosen as sampling stations for this project (Fig. 2.1) During the winter period, the ice extended to the bottom sediments at station 5 (~1 m deep), and consequently, sampling in that region of the lake required a move into deeper waters to station 6. It was originally anticipated to sample at stations 1 and 2 during every season. However, as a result of an improper insertion, peeper profiling in the winter session was restricted to station 1 only (see below). To evaluate the precision of the porewater sampling method, two sets of tandem peepers (~ 10 m apart) were first deployed at stations 1 and 2 during the summer session. Subsequent results indicated good agreement among the 24 porewater profiles yielded by duplicate samplers. The fall period included deployments at stations 1 and 2, including a tandem arrangement at the latter. A fourth peeper was deployed at station 5 in order to assess the porewater chemistry in an area of the lake that has received substantial tailings inputs. This area, coined "Tailings Bay", is bordered by a tailings beach and occupies a significant portion of Balmer Lake (Fig. 2.1); the recent sediments are predominantly mine-related. Deployments at stations 1 and 2 were repeated in the winter session. In addition, peeper sampling was performed in a shallow area of the lake near the out-flow (station 3) and at another deeper site in Tailings Bay (station 6., Fig 2.1). Only three peepers were available for the spring field session, two of which were deployed in tandem at station 1. Upon retrieval of the third sampler positioned at station 2, it was concluded from sediment smears on the aluminum lander and stains on the polyethylene frit that the peeper had fallen on its side at or shortly after deployment. As a result, no data for station 2 were acquired. A summary of the peeper and water column sampling stations is presented in Table 2.3. Table 2.3. Summer, fall, winter and spring sampling periods at Balmer Lake, Ontario, indicating stations of peeper deployment and water column sampling. Sampling Period Peeper Deployment Water Column Sampling Summer Tandem peepers at stations 1 & 2 Stations 1,2,3,4 Fall Stations 1,5; Tandem peepers at station 2 Stations 1,2,3,4,5 Winter Stations 1, 2, 3, 6 Cores at station 1 and 6 Stations 1,2,3,4 Spring Tandem peepers at station 1 Stations 1,2,4,6 25 III. SAMPLING A N D METHODS 3.1 Sampling 3.1.1 Water Column During ice-free conditions, all sampling of the lake water column and sediments, as well as peeper deployment, was carried out from a free-floating surface platform powered by a 9.9 Hp outboard. The deck provided approximately 12 m^ working area, a sufficiently large area to accommodate a gas generator, winch, a peristaltic pump, a nitrogen gas cylinder, and permanent attachment of a nitrogen glove bag containment structure. Gravity cores, surface grabs and water column samples were drawn through a 3 n\2 hatch in the platform centre. Four 20 kg anchors maintained position at any given site. Winter sampling was carried out through 30 cm holes augered through the ice. Water column samples were drawn from the lake using a "Masterflex" peristaltic pump (Cole-Palmer Instrument Co., Chicago, IL) and sampled directly in a nitrogen-filled glove-bag. Three litres were taken at each sampled depth, placed in a cooler, and taken immediately to the lab for filtration. One litre, to be analyzed for dissolved and particulate metal fractions, was filtered through an acid-cleaned, pre-weighted Nuclepore® 0.45 um polycarbonate membrane filter using nitrogen overpressure, acidified to p H 2 with nitric acid (Seastar Chemicals), and stored in acid-cleaned polyethylene bottles. The remaining two litres were filtered through a 0.4 um precombusted Glass Fibre filter using nitrogen overpressure for determinations of suspended particulate organic carbon (SPOC) and nitrogen (SPON). Ammonium concentrations were 26 determined from the filtrate spectrophotometrically on site; the remainder was frozen for subsequent analyses of nitrate, phosphate and sulphate. 3.1.2 Physical Profiling Hydrographic profiling of temperature, dissolved oxygen and p H was performed using a Cole Palmer water analyser (model 5150). Profiles were collected by slowly lowering the probe to desired depths and allowing the electrodes and thermistor to equilibrate before proceeding to the next depth. During periods of stratification, higher resolution profiling was attained across the depth of the pycnocline. Profiles were obtained commensurate with water column sampling in order to be temporally consistent. For both ice-free and ice- covered sampling periods, all hydrographic profiling was performed at least 5 m away from the point of water column sampling to minimize potential contamination and water column disturbance in the vicinity of the sampler. The oxygen electrode was calibrated in the field on a daily basis while calibration of the p H sensor was performed in the lab. 3.1.3 Porewaters The diagenetic cycling of metals in the sediments, and their exchange with the overlying water column was assessed by high resolution sampling of the interstitial and bottom waters using dialysis arrays. Diffusion sampling of sediment porewaters has become widely used since their initial application by Hesslein (1976) and Mayer (1976). Due to lower requirements of processing time and equipment, dialysis methods offer a suitable alternative to squeezing and 27 centrifuging techniques. Samplers typically consist of plastic sheets machined with compartments down their length which are filled with DDW and covered with a semi-permeable membrane; upon insertion into the sediments, the samplers are allowed to approach diffusive equilbrium with the adjacent porewaters. In situ dialysis has been shown to be particularly well suited to the study of trace constituents in sedimentary porewaters under field conditions (Carignan et al., 1985) and has been successfully implemented in studies of Canadian Shield Lakes (Carignan and Nriagu, 1985; Carignan and Tessier, 1985a; Gaillard et al, 1986). Potential artifacts are inherent to the spectrum of interstitial water sampling techniques. Methods such as squeezing and centrifugation followed by filtration, for example, are prone to effects resulting from sample oxidation and temperature variations (Bischoff et al, 1970). Diffusion techniques share a similar succeptibility to potential sources of error including the type of membrane used, construction material, dialyzer design, equilibration time and sampler preparation (Carignan, 1984; Carignan et al, 1985; Brandl and Hanselmann, 1991; Carignan etfal, 1994). Knowledge of the permeability characteristics of each membrane type, its sorptive properties for chemical compounds and microorganisms, and its resistance to microbial degradation are essential to the application of diffusion samplers and determination of optimal incubation times. A review by Brandl and Hanselmann (1991) determined from a study of 13 different membranes that cellulose-based dialysis membranes were the best choice in lake sediments at low temperatures. The cellulose membranes tested, however, exhibited marked weight losses due to microbial degradation when incubated at elevated temperatures (~28 °C). Martens and Klump (1980) also observed some physical breakdown of cellulose-based membranes. A digestible membrane polymer 28 could easily produce false porewater compositions due to a weakening of membrane stability, changes in membrane permeability or blockage of membrane pores. In consideration of warm water conditions (>20 °C) characteristic of Balmer Lake in the summer and fall, such membranes were excluded from this study. In its place, a Gelman (HT-450) polysulfone membrane was chosen for its mechanical stability and resistance to microbial attack. The suitability of such a dialysis system has been validated in several examinations of lake porewaters (Carignan, 1984; Carignan et al, 1985; Gaillard et al, 1986; Carignan et al, 1994). In some peeper preparations, a thin Nuclepore polycarbonate membrane (0.2 urn )was used in conjunction with the Gelman membrane to provide a better seal. Most samplers, however, were prepared solely with a 0.45 u.m Gelman polysulfone sheet. For measurements of porewater constituents, Carignan et al (1985) showed that profiles obtained with a dual 0.45 urn (polysulfone) + 0.03 urn (polycarbonate) membrane system were not significantly different from those obtained with a 0.45 um membrane alone. This result suggests that the Brownian diffusion of colloidal metal species through relatively large pores of 0.45 urn is unlikely 3.1.3.1 Peeper Description and Preparation The dialyzers implemented in this study are similar in design to that of Hesslein (1976) and consist of 120 x 20 x 2.3 cm Plexiglas plates with two rows of chambers, each 6.5 x 0.7 x 0.7 cm, machined in at 1.25 cm intervals, centre to centre (Fig. 3-1). This scheme is interrupted by a 20 cm section containing 31 chambers, each 3.5 x 0.7 x 0.7, spaced at 0.625 cm intervals. This zone is intended to coincide with the sediment-water interface upon deployment and provides 29 Peeper base plate Machined sample wells, Acrylic face plate machined to match base plate High resolution section (5mm interval) r- \ 'A' * I 1 i i i m i 70 fxm polyethylene frit 0.45 fim filter" DETAIL 'A* Figure 3.1. Schematic diagram of membrane dialysis sampler (peeper). 30 greater resolution in the surficial sediments and bottom waters. The peeper design affords high resolution pore water and bottom water profiles of the dissolved constituents from 40 cm above the sediment-water interface to a sediment depth of 50 cm. A membrane filter arrangement positioned between the chamber and face plates was implemented to facilitate dialysis and preclude the entrance of particulates (Fig. 3.1). In addition, the exterior surface of the face plate was fitted with a polyethylene frit to provide a coarse filtering screen, facilitate easier insertion and thus minimize sediment disturbance. Three rows of nylon screws positioned at 4.5 cm spacing down the peeper securely fastened the frit, face plate and filters to the chamber plate. Peeper preparation and assembly was performed at U.B.C. prior to shipment to the study site. The peepers and their Plexiglas cases were first washed in a mild detergent (Isoclean, ISOLAB Inc, Akron Ohio) followed by several rinses in DDW. The components were then soaked in dilute ultrapure H N O 3 (Seastar Chemicals) followed by several 24 h soaks in DDW to remove any residual contaminants and acid. Other than serving as nitrogen flushing sleeves, the plexiglass cases served as sturdy transport containers and provided a sterile transport environment. The Nuclepore and Gelman membranes received a dilute soak in ultrapure H N O 3 and several rinses in DDW. Peeper assembly was carried out in a laminar flow hood (class 100) in a shallow plexiglass tank of deoxygenated, distilled, deionized water (DDDW). The chamber plate was first submersed, and then using a clean pipette tip, all bubbles on the acrylic surface were dislodged. Two peeper membrane configurations were utilized. The first consisted of a twin membrane arrangement of a Nuclepore filter (polycarbonate, 0.2 u,m poresize) overlain by a Gelman HT-450 filter (polysulfone, 0.45 urn poresize). The second arrangement consisted of only a 0.45 |xm Gelman filter membrane. Upon placement of the 31 filter membrane(s) on the chamber plate, the acrylic face plate and polyethylene frit (70 um pore size) were placed appropriately and securely fastened with 51 nylon screws. The assembled peepers were stored in sealed acrylic cases containing DDDW bubbled with nitrogen until deployment. All peeper flushing cases were purged with nitrogen for approximately 48 hours before transport and for a further 24 hours prior to deployment. Carignan et al. (1994) have recently brought attention to potential effects of free oxygen liberated from acrylic samplers. The presence of such oxygen traces could introduce an artifact by reacting with ferrous iron which may diffuse into peeper cells after deployment and perhaps precipitate as Fe(OH)3. Carignan observed that acrylic (Plexiglas) in equilibrium with the atmosphere can absorb 1.6% (vol/vol) O2 that is lost slowly (half-time -5.7 d) once the material is exposed to anoxic conditions. This translated into significant differences in the observed porewater distributions of total Fe, Fe2+, dissolved reactive P, and SO42" between those samplers conditioned by a 30-d exposure to N 2 before deployment and those that received an 18-h bubbling in N 2 (Carignan et al., 1994). This phenomenon was originally described by Carignan (1984), who observed an orange discolouration on peepers deployed in anoxic sediments. The peeper preparation in this study included flushing periods greater than 72 h and suggests the potential for post-deployment releases of oxygen is minimal. 3.1.3.2 Peeper Deployment and Subsampling Each sampling session involved the deployment of four peepers at various sites within the lake. Peeper deployment for the first sampling period was achieved via diver insertion; peepers were pushed vertically into the sediments to a specified depth marked on the acrylic surface. Each sampler was marked via 32 a 6 m slack line from the peeper to a 20 kg anchor and a line from the anchor to a surface float. Subsequent peeper deployments and retrievals during ice-free periods were performed using weighted aluminum "landers". These cube- shaped support structures (1 m x 1 m x 0.8 m) remained in situ during the equilibration period with the peeper attached and ensured vertical insertion to a specific sediment depth. During the winter sampling period, peepers were fixed to 5 m long teflon-coated pipes and pushed vertically into the sediments through augered holes in the ice surface. A i m long peg attached horizontally to the pipe and placed flush with the ice ensured a correct insertion depth and maintained the position of the pole over the deployment period. All peepers were allowed to equilibrate for at least 14 days. The most important factors determining the equilibration time are the diffusion coefficient of the constituent, its sorptive properties, and the temperature and porosity of the sediments (Carignan, 1984). Carignan determined that for the porosities observed in recent lake sediments, equilibration periods of 15 days for warm (20°-25° C) and 20 days for cold (4° -6° C) appear to provide sufficient incubation. Peeper removal and subsampling occurred on a daily basis, with one peeper being removed per day. Upon retrieval, each peeper was brought to just below the lake surface, briefly agitated to dislodge adhering particulates, transferred to the boat and immediately placed in a nitrogen-purged sleeve for transport to the lab. The former procedures took less than 2 minutes. Peepers were subsequently processed in a nitrogen-filled glove bag and involved the careful removal of nylon screws, frit and face-plate, followed by the removal of the 0.45 urn Gelman filter (when used in tandem with the 0.2 um polycarbonate membrane). Each peeper was then subsampled by direct puncturing of the filter membrane with an Eppendorf pipettor and acid-washed pipette tips. The dual 33 row chamber arrangement of the peepers allowed one row to be subsampled for metals while the other for nutrients and sulphate. Direct probe insertion into chambers at selected depths allowed for measurements of porewater pH. Samples to be analyzed for dissolved metals, nutrients and sulphate were treated as described above for the water column protocol. Sample extraction and partitioning was accomplished within 2 hours of peeper retrieval. 3.1.4 Sediment Sampling Coring was performed during the winter survey as it was hoped that the stable ice platform would facilitate the acquisition of high quality cores. Three such cores with well-defined sediment-water interfaces were collected through augered holes in the ice by lowering a light-weight gravity corer (Pedersen et al, 1985) suspended from a metre-block and tripod. Cores were carried by hand to the lab and immediately processed. Cores were first logged and then firmly attached to the underside of an extrusion table fitted with a nitrogen-filled glove-bag. The core was extruded through a hole in the center of the table and carefully sectioned; core slices were placed in acid-washed 250 mL high-density polyethlyne (HDPE) centrifuge bottles and spun at ~1,200 gravities for 20 minutes at which point they were transferred to an additional nitrogen-filled glove bag for decantation of the interstitial waters and filtration. Supernatant water from the centrifuged sample was decanted into an acid-washed 50 mL disposal polypropylene syring fitted with an acid-washed Syfril 0.45 um mixed cellulose acetate disposable filter. An acid-washed polyethylene piston was utilized to express the interstitial solution through the filter and into an acid-washed HDPE sample bottle. The remaining 34 filtrate was passed through a 0.1 um disposable filter and acidified to p H 2 with ultra-pure nitric acid (Seastar Chemicals). Sediment samples were bagged and frozen until analysis. 3.2 Instrumentation 3.2.1 ICP/MS A V G "PQ Turbo Plus" inductively-coupled plasma mass spectrophotometer (ICP/MS) (VG Elemental, Surrey, UK) was used for the determination of the concentrations of Mn, Ni, Cu, Zn, As and Pb. Data analyses were handled by a Dell 486 computer equiped with PQ vision 4.1.1 software. Pure (99.998 %) ICP-grade argon (Medigas, Vancouver, B.C.) was used as the plasma and carrier gas. A Rheodyne 6 port flow injection system (Mandel Scientific, Guelph, Ont.), positioned between the peristaltic pump and the nebulizer, was used to facilitate small sample volumes. Operating conditions of the ICP/MS were optimized daily to obtain maximum sensitivity. Resolution and peak shape were optimized using a solution containing 10 ppb Co, In, Pb and Bi. Sample analyses were performed using pneumatic nebulization (de Gallen v-grove glass nebulizer). The instrument was run in a multichannel peak jump mode with a 10 milli-second dwell time, one point per peak and a 20 second aquisition time. The concentrations of the respective metals were calculated using a linear calibration curve derived from certified metal standards. A 10 ppb In internal standard was applied to all samples, standards and blanks to correct for sensivity variations during analyses. Typical operating conditions are outlined in Appendix A. 35 3.2.2 GFAAS Graphite furnace atomic adsorption spectroscopy (GFAAS) was performed using a Varian Spectra AA300 with a Zeeman background correction to determine Fe concentrations in the water column and porewaters. All determinations were conducted using L'Vov platforms. Concentrations were caculated from linear calibrations derived from certified metal standards using instumental parameters suggested by the manufacturer. 3.2.3 Ion Chromatography Determinations of porewater and water column concentrations of nitrate and sulphate were performed using a DIONEX DX-100 ion chromatograph. The system provided efficient isocratic separation of the respective inorganic anions using a 1.8mM Na2C03/ 1.7 mM NaHC03 eluent in conjunction with an IonPac® AS12A anion exchange column and a ASRS-1 anion self-regenerating suppressor. 3.2.4 X-ray Fluorescence An aliquot of each freeze-dried sediment sample was ground in a tungsten carbide disc mill prior to preparation for X-ray fluorescence, C-N-S and coulometric analyses. Metal concentrations were determined in the solid phase by X-ray fluorescence spectrometry using pressed powders and glass discs for minor and major elements, respectively. Calibration procedures and precision estimates for most elements were very similar to those reported by Calvert et al. (1985) with the exception that the calibration range for As was extended by 36 making artificial standards spiked with known weights of AS2O3. A linear calibration was obtained for As up to a maximum concentration of 5000 ppm. Accuracy of the elemental analyses was monitored by including aliquots of the standard sediment powders MESS-1, BCSS-1, PACS-1 with each batch of samples. 3.2.5 Coulometry Carbonate carbon was determined photometrically in a Coulometrics Inc. 5010 coulometer. Precision, as the relative standard deviation, was ± 3.7 %. Reagent-grade CaC03 was used to assess routinely the accuracy of the coulometer. 3.2.6 Carbon, Nitrogen and Sulphur Total C, N and S of sediments and suspended particulates were determined by combustion /gas chromatography using a Carlo-Erba (NA-1500) elemental analyser. Precisions were ± 0.6, 3.5 and 3.5 % (1 a rsd), respectively. Organic carbon was obtained from the difference between the total and carbonate carbon values (combined 1 a rsd = 1.9 %). 3.2.7 Spectrophotometry Water column and porewater ammonium concentrations were determined spectrophotometrically similar to the alternative method outlined by Parsons et al. (1984). Ammonia reacts with phenol in an alkaline citrate medium with 37 sodium nirtoprusside added as a catalyst and forms blue indophenol. The absorbance is then measured at 640 run. 3.2.8 Digestion of Polycarbonate Filters Polycarbonate membranes were placed in pre-weighed 5 mL teflon vials and oxidized at 80°C with 1 mL of ammonia (Seastar Chemicals). After being dried down (80°C) / 0.5 mL each of HC1, HF and H N O 3 (Seastar Chemicals) were added and microwave-digested for 45 minutes at 60 PSI. After being dried down once again, the digest was brought up in 250 uL H N 0 3 and diluted to approximately 4 mL. ICP-MS was used to determine the concentrations of Mn, Ni, Cu, Zn, As and Pb. Total Fe concentrations were determined via GFAAS. 38 IV. RESULTS Results for the water column, interstitial waters and sediment components of the Balmer Lake survey are presented in this chapter. The quality assurance and quality control (QA/QC) for all measured parameters was excellent (Appendix B). Values obtained for field blanks, field replicates, laboratory blanks, laboratory replicates and reference materials all met acceptable criteria. 4.1 Water Column 4.1.1 Physical Profiling Hydrographic data of temperature and dissolved oxygen suggest that the water column of Balmer Lake maintains fairly well-mixed conditions during ice- free periods. The first sampling period in the summer of 1993 was chosen to coincide with a stratified water column, since stratification is common to many temperate lakes in the summer months. Basic hydrographic data, however, revealed fully oxygenated bottom waters and invariant temperature and p H profiles (Fig. 4.1). Temperatures ranged from 16-18 °C while dissolved oxygen concentrations ranged from 7-9 mg/L. Profiling of pH revealed a slightly acidic to slightly basic water column with values ranging from 6.2 to 7.8 p H units. The overall homogeneity was probably a result of the common windy conditions and the resultant mixing of the shallow water column (~4 m deep). Fall sampling took place subsequent to convective overturn and was also characterized by a well-mixed water column. Cooler water temperatures (~2-3°C) and higher dissolved oxygen concentrations (10-12 mg/L) were evident (Fig. 4.2). A p H 39 Summer Station 1 Temperature, deg. C 14 15 16 17 18 19 20 5 6 7 8 9 Dissolved O ppm; pH 10 B Summer Station 2 Temperature, deg. C 14 15 16 17 18 19 CL &: <D To 5 20 " " 1 " ' ' 1 " " 1 • ' o f : : i / / - " ? / 1 * 1 : * 1 1 * \ - 6 \ i i ; pH-*- \ 1 A - 1 i<—Temp. 1 1 o 1 i - . . . . 1 . . . . 1 . . . . 1 . . . . 1 . . . . 1 . . . . 10 Dissolved O z > ppm; pH Summer Station 3 Temperature, deg. C 14 15 16 17 18 19 20 D CL 5 " ' " ; ;Temp. i t - i 4 1 - : P H — » » ; , / r - ° - . 1 1 • \ f : / _ - ' [ +\ 9 - Vs/s/s/s/s/s/s/s/s/sSs/s/s/sM . . . . i . . . . i . . . . i . . . . i . . . . i... .' 10 14 CL <D 3 CO 5 Summer Station 4 Temperature, deg. C 15 16 17 18 19 20 : f y * T M I I \ A • 1 ? A - I \ V I I I [ P H — H ,  I ?\ I •_ Tjjmp.-*^ I 1 I 1 ? f i I i j 6^*^ 0 . . . . 1 • • • • 1 • • • • 1 10 Dissolved 0 2 , ppm; pH Dissolved 0 2 , ppm; pH Fig. 4.1 Summer water column profiles of temperature, dissolved oxygen and pH for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, June, 1993. The hatched line represents the sediment surface. 40 Fall Station 1 Temperature, deg. C 0 1 2 3 4 5 6 7 0 1 1 1 1 1 1 1 1 1 • 11 B Fall Station 2 Temperature, deg. C 1 2 3 4 5 6 7 Q _ 5. 5 o • 1 1 , 1 1 1 1 1 1 1 1 1 1 1 1 1 r • H : M M O o 1 o '• Temp.—»>', :  %  j - ' — i — ' — ' — i — i — ' 6 8 10 12 14 Dissolved O ppm 8 9 10 11 12 13 14 15 Dissolved O ppm Fall Station 3 Temperature, deg. C 0 1 2 3 4 5 6 7 D Q _ <D Q I 2 I IT 1 | 1 1 1 1 | 1 1 I 1 | 1 1 1 1 | O i i i | i i i i | i i i i ! \ : Temp.—•! O l 6 8 10 12 14 Dissolved O z ppm CL a T5 2 5 Fall Station 4 Temperature, deg. C ) 1 2 3 4 5 6 7 • 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 " ' 1 1 \ 1 1 1 1 1 1 1 1 1 1 1 1 1 - > 0 I - ' Temp.—» j - - o • - 'f s s / / s ' • ssssssss. 1 . . . 1 . 8 10 12 14 Dissolved 0 2 ppm Fig. 4.2 Fall water column profiles of temperature and dissolved oxygen for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, October, 1993. The hatched line represents the sediment surface. 41 meter was unavailable for the fall and winter field sessions, and as a result, such data are absent. Winter profiling near the end of the ice-over period indicated a moderately stratified water column at the stations sampled (Fig. 4.3). Temperatures increased gradually from 0 °C immediately below the ice (~1 m thick) to 3-4 °C in the bottom waters. Conversely, dissolved oxygen levels decreased quickly in the lowermost half metre from approximately 5-7 mg/L to below detection limits 10-30 cm above the sediment surface (Fig. 4.3). However, at the shallowest station (station 3, 2.6 m depth) only a slight basal oxygen depletion was observed during this period (Fig. 4.3). The early-spring sampling, conducted less than two weeks after ice-off, revealed the reoccurrence of a well- mixed water column (Fig. 4.4). Constant profiles of temperature (9.5-10.5 °C) and dissolved oxygen (8-9 mg/L) were evident. Profiling of p H revealed a neutral to slightly basic water column with values ranging from 7.0-7.7 pH units. 4.1.2 Nitrate, Ammonium and Sulphate Variations in lake-wide average concentrations of nitrate, ammonium and sulphate closely parallel one another throughout the seasonal sampling periods. For all three ions, the seasonal hierarchy of mean concentrations follows the order winter > spring > summer > fall. The well-mixed water column and well- oxygenated bottom waters characteristic of the summer and fall field sessions, are consistent with the uniform profiles of dissolved constituents measured during these periods. Stations sampled during the summer period exhibited constant depth profiles of nitrate and sulphate with lake-wide values averaging 42 Winter Station 1 Temperature, deg. C 0 1 2 3 4 5 6 7 B 1111 . i i i | < . 1 i Ice .A S J J J J J J J J J J J J J J ' ' I . . . . 0 1 2 3 4 5 6 7 [O ], ppm Winter Station 2 Temperature, deg. C 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 [O ], ppm Winter Station 3 Temperature, deg. C 0 1 2 3 4 5 6 7 a. CD Q to 5 !o ( ;: Ice 1 ' " • ! » • • • \ ' } : \ / q f Temp »S f>—<v : V Winter Station 4 Temperature, deg. C 0 1 2 3 4 5 6 7 SI a. o> Q r • I ' H • ! >< i i | , < > >| t i p< | " I " ' - Ice \ i \ L . i . • ! hr* Temp. i 6 '(3 c f " S * - — ° - — 1 ' • • • 1 1 1 1 1 0 1 2 3 4 5 6 7 [OJ, ppm 0 1 2 3 4 5 6 7 [O ], ppm Fig. 4.3 Winter water column profiles of temperature and dissolved oxygen for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, March, 1994. The hatched line represents the sediment surface. 43 Spring Station 1 Temperature, deg.C 5 6 7 8 9 10 11 E f •1 11 11 1 1 1 1 1 11 i I|I ifi|I I I I[QiI I '- f f '• '- t 9 : j j • pH->J • f • '- i Y°> 1 - '- i • 6 - 1 i JTemp.W? - • 6 SSSSss/ssssssss. • I I I 1 t t • I 1 I I 1 5 6 7 8 9 10 11 Dissolved O , ppm; pH B Spring Station 2 Temperature, deg.C 5 6 7 8 9 10 11 E f CO Q 2 CO i I 111 I I 11 I pH- 111 I I <E I I I I 1 6 1 9 -i * 6 iemp.->i * P vs///s//sssssss 5 6 7 8 9 10 11 Dissolved 02, ppm; pH C Spring Station 4 Temperature, deg.C 5 6 7 8 9 10 11 0 i i i i I i i i i I i i r i | 11 i y | i 111 [ Q1 rr* • i f -. 1 t f :- 2 p H ^ 4 9 ~ 3 i i Temper • ? - • 6 .////////// / /./ / s. 1 1 1 1 1 1 1 1 1 1 t t 5 6 7 8 9 10 11 Dissolved O z , ppm; pH Fig. 4.4 Spring water column profiles of temperature, dissolved oxygen and pH for stations 1, 2 and 4 (A-C, respectively), Balmer Lake, May, 1994. The hatched line represents the sediment surface. 44 330 umol/L (274-342) and 3.30 mmol/L (2.69-3.45), respectively (Table 4.1). Ammonium was fairly uniformly distributed in the surface waters, with values ranging from 86-123 umol/L (Table 4.1). Depletions of ammonium with depth were evident for stations 2 and 3, where bottom water concentrations were approximately 60 % of their respective surface values (Table 4.1). At all sites and depths, phosphate concentrations remained below detection limits (~1 umol/L). Significant decreases in average ammonium concentrations were evident over the summer-fall transition with average lake values falling from 94 to 50 umol/L (Table 4.2). Water column nitrate and sulphate contents during the fall period remained consistent with the summer season and averaged 312 umol/L (292-346) and 3.41 mmol/L (3.32-3.69), respectively (Table 4.2). As seen for the summer period, phosphate concentrations were not detectable throughout the water column. The major ion distributions observed during the winter period show distinct relationships to the stratification indicated by the temperature and oxygen data (Table 4.3; Figs. 4.5 -4.7). Nitrate removal in the suboxic to anoxic bottom waters, and/or in shallow sediments, results in pronounced depletions at all stations 40-60 cm above the sediment-water interface (Table 4.3; Fig. 4.5). Sub- ice nitrate values averaging 415 umol/L (372-435) decline to minima of 16 to 80 umol/L in the bottom waters. Less significant reductions in nitrate concentrations are observed in the 'deep' waters at shallow station 3, where values decrease to -175 umol/L (Fig. 4.5C). A large increase in the inventory of ammonium was evident during the winter sampling period (Fig. 4.6); contents in bottom waters at all locales except station 4, ranged from 260 to over 500 umol/L. Ammonium values were not obtained below 2.6 m at station 4, and as a result, the presence or absence of a bottom water increase was not determined. 45 Table 4.1. Summer water column distributions of dissolved Mn, Fe, Ni, Cu, Zn, As, Pb, N H 4 + , NO3*, SO42- and PO4 3 " for stations 1, 2, 3, and 4 in Balmer Lake, Ontario, June 1993. All values are expressed in ug/L unless specified otherwise (BDL denotes below detection limits). Station Depth, m Mn Fe Ni Cu Zn As Pb NH4 NO3 SO4 PO4 mM mM mM mM 0.5 271 35 387 164 46 282 0.4 128 327 3.32 BDL 2 287 41 405 170 49 288 0.4 128 329 3.19 BDL 2 285 43 409 168 50 286 0.3 130 326 3.28 BDL 3.5 282 39 399 171 59 286 0.3 128 327 3.26 BDL 0.1 281 38 406 172 45 276 0.4 86 362 3.38 BDL 1 277 43 403 173 48 277 0.6 70 331 3.39 BDL 2 274 46 409 172 48 286 0.6 77 274 2.69 BDL 2 275 51 411 175 50 283 0.4 75 294 2.87 BDL 3 300 50 437 181 56 293 0.7 77 337 3.44 BDL 3.6 289 39 420 179 60 290 0.4 55 313 3.15 BDL 0.5 282 47 423 172 46 285 0.4 99 335 3.35 BDL 2 280 35 400 175 48 287 0.3 68 341 3.38 BDL 2 282 35 405 179 48 289 0.4 70 342 3.42 BDL 2.8 292 58 411 169 52 288 0.4 59 337 3.25 BDL 0.5 287 35 407 175 45 290 0.5 103 338 3.45 BDL 0.5 289 32 410 175 46 288 0.5 105 342 3.43 BDL 2 289 39 411 169 48 292 1 103 332 3.32 BDL 3.5 286 44 418 173 52 289 0.4 91 337 3.34 BDL 3.5 284 39 421 175 54 290 0.4 91 332 3.31 BDL 46 Table 4.2. Fall water column distributions of dissolved Mn, Fe, Ni, Cu, Zn, As, Pb, NH4+ N03 -, SC»42_ and PO43- for stations 1, 2, 3,4 and 5 in Balmer Lake, Ontario, October, 1993. All values are expressed in ug/L unless specified otherwise (BDL denotes below detection limits). Station Depth, m Mn Fe Ni Cu Zn As P b N H 4 N O 3 SO4 PO4 uM uM mM uM 0.5 161 242 114 63 203 0.4 45 302 3.33 BDL 2 165 95 262 112 59 207 0.5 51 309 3.38 BDL 2 170 92 260 110 60 208 0.4 299 3.32 BDL 3 171 88 265 114 55 210 0.3 56 307 3.42 BDL 05 171 89 278 121 48 214 0.3 51 346 3.44 BDL 1.5 158 93 273 121 63 208 0.3 49 310 3.41 BDL 3 166 101 269 119 47 211 0.4 48 311 3.36 BDL 05 186 79 281 121 63 203 0.5 55 317 3.47 BDL 0.5 180 90 283 124 58 203 05 306 3.46 BDL 15 185 84 283 123 59 207 0.6 55 313 3.42 BDL 1.8 196 99 288 126 55 210 0.5 56 292 3.42 BDL 0.5 181 94 279 124 40 214 0.3 46 307 3.38 BDL 0.5 182 89 282 118 45 219 0.2 48 316 3.52 BDL 1.5 183 97 284 122 65 217 0.4 45 316 3.45 BDL 1.5 182 90 280 125 62 215 0.4 338 3.40 BDL 3 180 88 275 112 56 210 0.4 307 BDL 0.1 227 98 312 130 124 242 0.3 300 3.62 BDL 0.1 203 88 300 101 59 235 0.3 304 3.69 BDL 47 Table 4.3. Winter water column distributions of dissolved Mn, Fe, Ni, Cu, Zn, As, Pb, NH4+ NO3-, SO42- and PO43- for stations 1, 2,3 and 4 in Balmer Lake, Ontario, March, 1993. All values are expressed in \ig/L unless specified otherwise (ND denotes not determined). Note that replicate samples were collected at a number of depths at all stations. Station Depth, m Mn F e N i Cu Z n A s Ft! NH4 NO3 SO4 PO4 uM uM m M uM 1.0 404 i'A) 455 294 83 337 0.6 372 4.94 NIL) 1.0 396 438 285 84 343 0.6 2.0 285 158 432 274 68 384 0.5 121 409 5.19 ND 2.3 299 147 473 331 63 358 0.2 133 415 5.17 ND 2.3 151 2.5 306 141 474 301 67 338 0.1 125 418 5.31 ND 2.8 245 2.8 504 73 502 177 130 199 0.3 251 288 8.09 ND 3.0 926 76 532 52 149 260 0.1 446 174 13.79 ND 3.0 173 3.2 972 48 554 46 158 283 0.2 502 16 12.04 ND 3.2 1043 52 578 46 161 289 0.4 1.0 312 131 523 450 68 386 0.1 211 395 5.02 ND 1.8 311 139 512 444 63 365 0.1 208 390 4.97 ND 1.8 133 211 2.3 307 129 503 438 77 392 0.4 213 388 5.00 ND 2.5 275 134 425 289 64 346 0.3 190 379 5.09 ND 2.5 300 467 314 65 345 0.3 2.7 312 115 407 216 63 270 0.2 214 282 4.92 ND 2.8 662 73 493 100 121 209 0.2 248 185 8.66 ND 2.9 978 119 503 58 178 282 0.6 406 82 11.63 ND 2.9 1015 133 532 59 180 298 0.4 1.0 347 177 468 314 67 353 1.1 160 411 5.16 ND 1.5 325 170 445 302 66 337 1.1 162 414 5.26 ND 1.8 348 175 471 307 74 354 0.3 158 420 5.32 ND 1.8 334 173 457 287 61 324 0.9 2.0 358 167 487 310 67 349 0.9 160 385 5.20 ND 2.0 162 2.2 362 165 525 328 65 349 0.4 170 407 5.18 ND 2.2 369 539 320 64 345 0.3 2.4 342 160 524 388 67 381 0.3 198 394 5.13 ND 2.6 667 152 451 382 135 410 0.4 263 175 4.93 ND 48 Station Depth, m Mn Fe Ni Cu Zn As Pb NH4 N 0 3 SO4 PO4 4 1.0 322 154 466 235 73 331 0.1 94 379 4.95 ND 1.0 291 427 221 71 319 0.3 ND 1.5 295 150 416 208 75 317 0.3 15 158 2.0 89 2.0 89 2.5 315 140 461 240 73 324 0.2 91 425 5.27 ND 2.6 326 148 494 320 87 357 0.4 118 436 5.38 ND 2.7 362 156 528 346 87 367 0.2 2.7 341 500 310 79 333 0.2 2.8 674 129 429 149 132 310 0.4 48 6.55 ND 49 Winter Station 1 0 100 200 300 400 500 Dissolved N O ', pmol L" 1 B Q . "O 2 i_ To 5 Winter Station 2 • • I : c > !> « • « « 1 • • • • 1 • • 1 1 1 • • ' 1 1 • * • • 0 100 200 300 400 500 Dissolved N O " , pmol L"1 1 t Q . "D 2 cs 3 r Winter Station 3 Ice \ \ \ \ \ \ \ \ \ \ \ \ \ I I • 1 1 I I 1 I I I I • • « • 1 I 0 100 200 300 400 500 Dissolved N O ' pmo lL " 1 D Winter Station 4 3 h i i i i 1 i i i i 1 i • • • I • • • • I < 0 100 200 300 400 500 Dissolved N O ", u.mo.L" 1 Fig. 4.5. Winter water column profiles of nitrate for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. 50 Winter Station 1 1 t Q . CD 9 X) k. CD CO 5 • I I I • • • I ' ' ' • 1 ' • • • I • • • ' 0 100 200 300 400 500 Dissolved N H 4 + , umo lL " 1 B Winter Station 2 0 100 200 300 400 500 Dissolved N H 4 + , ixmolL" 1 cL CD T3 Winter Station 3 Ice 3 I i • • • ' 1 ' ' • ' I ' ' ' ' 1 ' ' ' ' 1 ' • • • 0 100 200 300 400 500 Dissolved N H 4 + , nmol L ' 1 Winter Station 4 0 100 200 300 400 500 Dissolved N H 4 + , |xmolL" 1 F i g . 4.6. Win te r water column profiles o f a m m o n i u m for stations 1, 2, 3 and 4 (A-D, respectively), B a l m e r L a k e , M a r c h , 1994. Replicate samples are represented by double symbols at specific single depths. T h e hatched l ine represents the sediment surface. 51 A Q . Q) o •o CO Winter Station 1 Ice > k \ \ V* V ' 0 5 10 15 Dissolved S0 4 2", mmolL" 1 B Winter Station 2 o •O 2 * CD CO <: " " " " I ' ' * 0 5 10 15 2" m m r \ l - l "1 Dissolved S0 4 , mmo lL Winter Station 3 D Winter Station 4 0 5 10 15 Dissolved S042", mmol L"1 0 5 10 15 Dissolved S042", mmolL" 1 Fig. 4.7. Winter water column profiles of sulphate for stations 1, 2, 3 and 4 (A-D, respectively), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. 52 The winter water column was also characterized by a significantly greater burden of sulphate, with values averaging 5.2 mmol/L immediately below the ice surface (Table 4.3; Fig. 4.7) . Pronounced bottom water enrichments of sulphate were observed at or below depths of 2.8 m throughout the lake (Fig. 4.7). In the deeper lake areas (stations 1 and 2), sulphate values increase to near 14 mmol/L at water depths greater than 2.9 m (Fig 4.7A-B). During spring sampling, the concentrations of the major ions exhibited little lake-wide variation throughout the water column with values averaging 145 umol/L for NH4+, 268 umol/L for N O 3 - and 3.47 mmol/L for S O 4 2 " (Table 4.4). Due to analytical difficulties, phosphate concentrations were not determined for the winter and fall periods. 4.1.3 Suspended Organics The seasonal water column concentrations of total suspended material (TSM), organic-C (SPOC), organic-N (SPON) and total particulate organic matter (POM) are shown in Table 4.5. A crude estimate of the particulate organic matter (POM) content was obtained from the relationship, [POM] = 1.8 x [POC] (Nriagu et al, 1981). The TSM content of the water column exhibited negligible variability over the summer-fall-winter transition with values averaging 3.2-4.4 mg/L (Table 4.5). However, the winter water column was characterized by bottom water enrichments of total suspended solids at stations 1 and 4. Significant increases in the concentrations of suspended material were observed in the spring, during which values averaged 11 mg/L (Table 4.5). The fraction of suspended particulates represented by POM over the summer, fall, winter and spring varied considerably, with contributions averaging 47, 24, 26 and 14 %, respectively (Table 4.5). During the well-mixed periods of the summer, fall and 53 Table 4.4. Spring water column distributions of dissolved Mn, Fe, Ni, Cu, Zn, As, Pb, N H 4 + , NO3-, SO42- and P043" for stations 1,2 and 4 in Balmer Lake, Ontario, May, 1993. All values are expressed in u,g/L unless specified otherwise (ND denotes not determined). Station Depth, Mn Fe Ni Cu Zn A"s PS NTT4 NO3 3O4 PO4 m uM uM mM uM 1 1 301 86 428 220 59 224 1.1 136 252 3.25 ND 2 284 86 403 210 87 223 0.9 141 268 3.47 ND 3 299 85 428 223 57 225 1.3 142 275 3.46 ND 2 1 250 82 350 190 57 216 1.1 139 259 3.34 ND 2 257 91 380 202 57 217 1.7 144 265 3.36 ND 2 257 81 384 201 55 219 1.7 144 270 3.42 ND 3 244 92 358 191 60 229 1.4 144 273 3.45 ND 3 60 226 1.6 ND 4 1 267 80 377 184 65 230 0.9 148 269 3.63 ND 1 262 78 361 180 65 244 1.3 152 ND 2 265 76 363 180 64 250 1.6 152 269 3.62 ND 3.2 251 75 343 171 59 230 1.7 156 275 3.71 ND 54 Table 4.5. Concentrations of total suspended solids (TSS), particulate organic carbon (SPOC), particulate organic nitrogen (SPON), C0rg:N weight ratio (C/N) and total particulate organic matter (POM) in the water column of Balmer Lake, Ontario, for the summer, fall, winter and spring field sessions. TSS values were determined from the particulate mass on pre-weighed 0.45 urn Nuclepore filters, while all organics were determined from glass fibre filters. ND deontes not determined. Season Station Depth, m TSS mg-LT1 SPOC Hg-L"1 SPON ug-Lf1 Wt. Ratio POM ug-L-1 Summer 1 0.5 3.7 988 135 7.32 1778 2.0 3.3 876 119 7.36 1577 2.0 4.0 3.5 3.3 920 122 7.54 1656 Summer 2 0.1 3.7 772 103 7.50 1390 1.0 4.2 883 116 7.61 1589 1.0 4.2 2.0 3.9 893 118 7.57 1607 3.0 4.7 762 no 6.93 1372 3.6 3.6 1191 171 6.96 2144 Summer 3 0.5 2.0 3.2 2.1 1017 128 7.95 1831 2.8 4.0 1308 165 7.93 2354 Summer 4 0.5 3.8 923 114 8.10 1661 2.0 2.4 856 107 8.00 1541 3.5 5.5 1033 130 7.95 1859 3.5 4.7 ND ND ND ND Fall 1 0.5 3.9 540 81 6.67 972 2.0 4.4 521 77 6.77 938 3.0 4.2 552 82 6.73 994 Fall 2 0.5 4.5 ND ND ND ND 1.5 5.5 ND ND ND ND 1.5 4.4 ND ND ND ND 3.0 4.6 ND ND ND ND Fall 3 0.5 4.0 ND ND ND ND 1.5 4.0 ND ND ND ND 1.8 4.1 ND ND ND ND Fall 4 0.5 3.3 515 76 6.78 927 1.5 4.0 508 74 6.86 914 2.8 6.2 ND ND ND ND 55 Season Station Depth, m TSS SPOC SPON C / N POM Winter 1 1.0 3.0 376 82 4.59 677 2.0 2.4 336 79 4.25 605 2.25 2.6 ND ND ND ND 2.5 2.3 ND ND ND ND 2.75 2.6 332 77 4.31 598 3.0 4.1 416 69 6.03 749 3.15 6.3 754 138 5.46 1357 Winter 2 1.0 4.0 577 137 4.21 1039 1.75 3.1 498 131 3.80 896 1.75 2.7 498 131 3.80 8% 2.25 3.2 ND ND ND ND 2.5 3.1 530 149 3.56 954 2.65 3.5 ND ND ND ND 2.8 2.7 376 112 3.36 677 2.9 4.4 ND ND ND ND Winter 3 1.0 360 80 4.50 648 1.5 3.0 ND ND ND ND 1.75 3.2 ND ND ND ND 2.0 2.2 372 84 4.43 670 2.2 2.7 ND ND ND ND 2.4 3.1 ND ND ND ND 2.55 ND 945 157 6.02 1701 Winter 4 1.0 2.6 300 63 4.76 540 1.5 3.2 ND ND ND ND 2.0 2.5 333 67 4.97 599 2.5 2.6 310 65 4.77 558 2.6 2.0 404 83 4.87 727 2.7 3.5 ND ND ND ND 2.8 5.6 722 110 6.56 1300 Spring 1 1.0 8.2 798 142 5.62 1436 2.0 7.9 731 135 5.41 1316 3.0 9.3 743 139 5.35 1337 Spring 2 1.0 ND 749 137 5.47 ND 2.0 ND 706 135 5.23 ND 2.0 ND 700 134 5.22 ND 3.0 ND 739 137 5.39 ND Spring 4 1.0 10.8 816 138 5.91 1469 1.0 11.9 ND ND ND ND 2.0 12.1 834 144 5.79 1501 3.2 12.7 823 142 5.80 1481 Spring 5 0.5 11.2 913 151 6.05 1643 0.5 10.9 867 148 5.86 1561 56 spring, concentrations of suspended POC and PON varied only slightly with depth. Values for SPOC over these seasons averaged 960 ugC/L (762-1308), 530 HgC/L (508-552) and 815 ugC/L (731-913), respectively (Table 4.5). Furthermore, concentrations of SPON over the summer, fall and spring periods averaged respectively 125 ugN/L (103-171), 78 UgN/L (74-82) and 142 ugN/L (135-151) (Table 4.5). The discrepancy in the ranking of magnitudes in the SPOC and SPON values over these periods can be explained by seasonal differences in their respective particulate C / N ratios (Table 4.5). With the exception of station 2, all sampling sites during the winter period exhibited bottom water enrichments of SPOC and SPON (Table 4.5). The bottom waters of stations 1, 3, and 4 were characterized by higher C / N ratios and SPOC/SPON concentrations over two times greater than their respective near-surface values. 4.1.4 Trace Metals 4.1.4.1 Dissolved Fraction The well-mixed conditions of the summer, fall and spring water columns, revealed by physical profiling of temperature and dissolved oxygen, are in good agreement with the constant vertical distributions of dissolved trace metals observed during such periods. Concentrations of dissolved species during the summer period averaged 286 | ig/L for Mn, 40 ug/L for Fe, 411 ug/L for Ni, 173 ug/L for Cu, 51 ug/L for Zn and 287 ug/L for As (Table 4.1). During all seasons and at all depths, concentrations of dissolved Pb hovered at or below its detection limit of approximately 0.5 Ug/L. The fall sampling period was characterized by generally lower dissolved metal inventories with values 57 averaging 175 ug/L for Mn, 91 ug/L for Fe, 275 ug/L for Ni, 119 ug/L for Cu, 56 ug/L for Zn and 210 ug/L for As (Table 4.2). The winter period exhibits marked differences in water column structure with respect to dissolved trace metals. In general, concentration profiles relate to the observed stratification (Table 4.3, Figs. 4.8-4.11). At depths shallower than the oxycline (approximately 2.5 m), dissolved metal concentrations are significantly greater than those observed during other seasons with values averaging 323 ug/L for Mn, 152 ug/L for Fe, 468 ug/L for Ni, 307 ug/L for Cu, 70 ug/L for Zn and 348 ug/L for As (Table 4.3; Figs. 4.8-4.11). Dissolved trace metal profiles in the sub- or anoxic bottom waters of Balmer Lake contrast greatly with those seen during other periods. At all stations, dissolved Mn and Zn are characterized by bottom water enrichments up to 3.2 and 2.7 times their respective near-surface concentrations (Table 4.3; Figs. 4.8-4.11). Conversely, dissolved Cu profiles suggest removal for this element in the bottom waters at all locales except at the shallowest site, station 3 (Table 4.3). Depletions of dissolved copper were more pronounced in the deeper sites (stations 1 and 2; Figs. 4-8 and 4-9), where bottom concentrations represent 16 and 13 % of their respective near-surface levels. The bottom water gradients evident for Mn and Zn occur across the corresponding interval for dissolved oxygen depletion (see Fig. 4.3); similarly, zones of decreasing Cu concentrations are consistent with that of dissolved oxygen. The water column distributions for As and Ni are more complex. At the two deepest stations (stations 1 and 2, Figs. 4.8 and 4.9, respectively) concentrations of dissolved As begin to decrease between 2-2.5 m, reaching minima at approximately 2.7-2.8 m depth. At slightly greater depths, arsenic levels increase sharply to values in bottom waters approximately 100 ug/L greater than their respective minima (Figs 4.8 and 4.9). At the shallowest site 58 Winter Station 1 CD T3 k_ CD OS 200 400 60( Dissolved [Ni], ppb Winter Station 1 » » • • I ' • • • 1 • • ' ' 1 • 1 ' ' I 1 1 0 100 200 300 400 Dissolved [Cu], ppb c . a •c i_ a 5 Winter Stat ion 1 3 L 0 100 200 Dissolved [Zn], ppb Fig. 4.8. Winter water column profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 1, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. 59 ->—» CL 0) o Winter Stat ion 2 ICE B Winter Station 2 i i i I i 1 V CD T J k_ CD CO 5 2 f 0 100 200 Disso lved [Fe], ppb 200 400 600 800 1000 Dissolved [Mn], ppb Q . CD o •o ' CD I Winter Station 2 0 100 200 300 400 Dissolved [As], ppb Winter Station 2 Winter Station 2 Winter Station 2 E < Q . CD o •o CD CO 0 200 400 60) Dissolved [Ni], ppb 0 100 200 300 400 Dissolved [Cu], ppb a CD "O >_ CD « ICE . , . i . . . . i 1 1 1 . . 0 100 200 Dissolved [Zn], ppb F i g . 4 .9 . Win te r water co lumn prof i les o f d isso lved F e , M n , A s , N i , C u and Z n ( A - F , respectively) fo r station 2 , Ba lme r L a k e , M a r c h , 1994. Repl icate samples are represented by double symbo ls at speci f ic single depths. T h e hatched l ine represents the sediment surface. 6 0 CD •S 2 r 52 CO 5 Winter Stat ion 3 Winter Station 3 Winter Station 3 0 100 200 Disso lved [Fe], ppb 0 200 400 600 8001000 Dissolved [Mn], ppb 0 100 200 300 400 Dissolved [As], ppb Winter c Winter c Winter Station 3 fc Station 3 r Station 3 0 200 400 60i' 0 100 200 300 400 0 100 200 Dissolved [Ni], ppb Dissolved [Cu], ppb Dissolved [Zn], ppb Fig. 4.10. Winter water column profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 3, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. 61 Winter Station 4 B Q. CD , i i CD Winter Station 4 Ice 3 L A W W W W W W • 1 • • 1 1 1 * 1 1 1 1 • ' Winter Station 4 0 100 200 Dissolved [Fe], ppb 0 200 400 600 8001000 Dissolved [Mn], ppb 0 100 200 300 400 Dissolved [As], ppb Winter p Winter c Winter Station 4 ^ Station 4 ' Station 4 0 200 400 60) 0 100 200 300 400 0 100 200 Dissolved [Ni], ppb Dissolved [Cu], ppb Dissolved [Zn], ppb Fig. 4.11. Winter water column profiles of dissolved Fe, Mn, As, N i , Cu and Zn (A-F, respectively) for station 4, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. The hatched line represents the sediment surface. 62 (station 3, Fig. 4.10), profiles of dissolved As lack a defined minimum, although values increase slightly below 2.2 m. Conversely, the water column at station 4 is characterized by a near-bottom maximum of dissolved As at ~2.7 m, below which concentrations decrease (Fig. 4.11). Bottom water maxima are also evident for dissolved Fe, Ni, Cu and to a lesser degree Zn, at the same horizon at this site (Fig. 4.11). Dissolved Ni distributions during the winter period differ markedly between sampling sites. Stations 1, 3 and 4 all exhibit some increase with depth (Figs. 4.8, 4.10 and 4.11); values at stations 3 and 4, however, fall back to near- surface levels below deep maxima (Figs. 4.10 and 4.11). Concentrations of dissolved Ni remain fairly constant at station 2, except for a minimum evident between 2.5 and 2.8 m (Fig. 4.9). Distributions of Fe share similar inconsistencies to those seen for dissolved nickel. The steady drop in dissolved Fe below 2.5 m at station 1 occurs across the same interval observed for decreases in dissolved As and Cu (Fig. 4.8). The iron profile is characterized by a bottom water minimum at station 2 coincident with the As minimum observed at this site (Fig. 4.9). In contrast, stations 3 and 4 exhibit fairly uniform profiles of dissolved Fe (Figs. 4.10 and 4.11). During the well-mixed conditions of the spring sampling period, concentrations of dissolved metal species varied little throughout the water column with values averaging 265 ug/L for Mn, 80 ug/L for Fe, 372 ug/L for Ni , . 196 ug/L for Cu, 61 ug/L for Zn and 224 ug/L for As (Table 4.4). In general, the magnitudes fall within the ranges observed for the summer session. 63 4.1.4.2 Particulate Fraction The seasonal concentrations of particulate trace metals in the water column of Balmer Lake have been expressed in terms of both mass of metal per unit volume (Table 4.6) and mass of metal per unit mass of total particulate matter (Table 4.7). The well-mixed periods of summer, fall and spring, are characterized by considerably more intra-site variability in comparison to associated dissolved trace metal distributions. In general, the particulate fraction for most metals (except Fe) contributes a small percentage of the total pool of metals in the water column during all seasons. The water column concentrations of particulate Fe, Mn, Ni, Cu, Zn and As during the summer period averaged 107, 3, 5, 13, 27 and 7 ug-L - 1 , respectively, and represented 71,1,1, 7, 30 and 2% of the respective totals (Table 4.6). These lake-wide averages translate into mass ratios of 29,600 mg-kg_ 1 for Fe, 850 mg-kg" 1 for Mn, 1,330 mg-kg-1 for Ni, 3,250 mg-kg-1 for Cu, 7900 mg-kg"1 for Zn and 1760 mg-kg - 1 for Pb (Table 4.7). The autumn particulate data contrast little from the summer values; concentrations of particulate Fe, Mn, Ni , Cu, Zn and As averaged 205, 4, 5, 15, 16 and 9 ug-L"1, respectively (Table 4.6). The fall proportions of the dissolved and solid-phase fractions show additional inter- season similarity with the particulates averaging 69, 2, 2, 11, 21 and 4 % of the respective totals of Fe, Mn, Ni, Cu, Zn and As. The influence of ice cover on the chemical and physical environment during the winter months can be further realized upon observation of the vertical distributions of particulate metal species (Tables 4.6 and 4.7). Slight decreases in dissolved Fe with depth are mirrored by significant increases in concentrations of Fe-bearing particulates which surpass 400 ug-L"1 at some sites (Table 4.6). Conversely, Mn, N i , Cu and Zn particulates exhibit minimal variation with 64 Table 4.6. Particulate trace metal concentrations of Fe, Mn, Ni, Cu, Zn, As and Pb for the summer, fall, winter and spring water columns in Balmer Lake, Ontario. All values are expressed in ug-L~l. Season Station Water Fe Mn Ni Cu Zn As Pb depth, m Summer 1 0.5 105.2 1.5 3.1 6.8 56.4 3.8 0.1 2.0 108.1 2.3 3.8 9.1 26.3 3.9 0.2 2.0 126.6 2.8 4.5 11.9 3.0 5.1 0.3 3.5 102.2 2.0 3.2 8.9 28.1 4.0 0.3 Summer 2 0.1 99.3 2.5 5.3 11.6 93.9 6.3 0.2 1.0 118.0 1.0 95.9 2.8 4.7 11.1 9.0 6.1 0.1 2.0 138.0 3.8 6.2 15.8 5.1 8.0 0.2 3.0 16.3 22.2 50.1 43.9 35.8 1.8 3.6 122.9 3.5 4.9 11.8 4.5 6.6 0.2 Summer 3 0.5 88.7 1.5 2.5 7.1 28.2 3.2 0.2 2.0 68.1 1.4 2.3 6.0 25.9 2.5 0.2 Summer 4 0.5 101.5 2.0 3.7 9.9 33.0 4.1 0.4 2.0 70.5 1.6 2.6 6.7 17.1 2.8 0.2 3.5 147.0 2.9 4.3 11.3 7.3 4.5 0.3 Fall 1 0.5 182.0 3.2 4.2 12.5 14.6 7.3 0.9 2.0 212.6 3.6 4.8 14.5 43.1 8.8 1.1 3.0 203.7 3.4 4.5 12.8 20.8 8.0 0.9 Fall 2 0.5 182.3 3.3 4 -7 14.8 6.9 7.6 0.9 1.5 214.3 3.9 5.5 17.7 10.8 9.5 1.3 1.5 195.2 3.4 4.9 15.5 8.7 8.1 1.0 3.0 219.8 3.7 5.3 16.9 7.4 9.2 1.1 Fall 3 0.5 213.6 4.1 5.7 17.5 9.0 10.2 1.2 1.5 193.4 3.4 4.8 14.6 7.3 8.4 1.1 1.8 192.0 3.4 4.8 14.5 6.6 8.4 1.0 Fall 4 0.5 171.0 2.8 3.8 11.7 16.5 7.2 0.9 1.5 189.1 3.3 ' 4.5 13.2 14.9 8.3 1.0 2.8 283.2 4.5 5.3 14.8 38.4 9.4 1.1 65 Season Station Depth Fe Mn Ni Cu Zn As Pb Winter 1 1.0 130.8 1.6 5.6 32.1 6.3 11.6 0.8 2.0 156.3 2.0 7.7 36.0 13.5 9.5 0.9 2.25 157.0 2.6 9.6 50.8 8.6 11.4 0.7 2.5 160.5 2.6 8.9 42.6 7.0 11.7 0.7 2.75 149.4 2.4 6.5 51.2 5.6 11.7 0.5 3.0 388.3 3.0 6.4 22.6 6.9 121.4 0.6 3.15 366.2 6.2 14.8 38.0 16.3 42.3 0.7 Winter 2 1.0 205.5 2.0 11.7 109.6 15.5 6.4 0.2 1.75 148.7 2.25 193.6 2.5 13.4 115.2 14.9 16.0 1.3 2.5 223.9 2.6 15.4 97.4 18.2 14.0 1.3 2.65 222.3 2.7 15.6 81.6 18.8 14.0 1.3 2.8 175.3 2.4 12.6 59.5 19.1 10.7 0.8 2.9 424.4 3.4 13.1 53.3 19.4 110.9 1.4 Winter 3 1.5 161.5 2.2 7.8 45.5 31.7 12.3 0.9 1.75 153.3 2.1 7.7 45.4 50.1 12.9 0.8 Winter 4 1.0 163.9 2.6 7.2 29.1 8.7 11.4 0.8 1.5 176.0 2.8 7.6 31.3 11.8 12.2 1.0 2.0 151.1 2.2 6.1 25.2 3.9 10.0 0.7 2.5 143.3 2.0 6.0 27.6 16.3 8.8 0.8 2.6 130.1 1.8 5.5 32.8 11.3 8.4 0.7 2.7 190.2 2.5 8.6 54.5 16.7 13.5 1.1 2.8 479.9 4.5 11.2 49.7 73.9 116.1 2.3 Spring 1 1.0 445.5 5.7 7.5 70.7 22.6 49.2 0.8 2.0 566.0 7.2 9.4 88.5 29.9 70.8 1.2 3.0 569.6 7.5 10.2 90.8 27.5 69.0 1.0 Spring 4 1.0 733.0 10.2 11.2 101.1 30.6 73.4 1.4 1.0 739.3 11.3 12.5 105.9 31.3 75.8 1.4 2.0 744.5 11.1 12.1 105.2 33.7 78.3 4.7 3.2 742.3 11.4 11.6 96.8 27.3 69.1 1.3 66 Table 4.7. Concentrations of trace metals (mgkg'l), organic carbon (%) and organic nitrogen (%) in suspended particulates of the summer, fall, winter and spring water columns in Balmer Lake, Ontario. Trace metal values were determined from the particulate mass on pre-weighed 0.45 um Nuclepore filters, while all SPOC and SPON values were determined from glass fibre filters. Season Station Water depth, m Fe Mn Ni Cu Zn As Pb S P O C S P O N Summer 1 0.5 28,510 410 840 1850 15380 1030 27 26.8 3.7 2.0 32,560 690 1130 2730 8010 1160 70 26.4 3.6 2.0 31,320 700 1100 2940 830 1270 65 3.5 31,290 620 970 2730 8700 1230 77 28.2 3.7 Summer 2 0.1 1.0 26,770 28,190 690 1420 3120 25410 1690 60 20.8 21.1 2.8 2.8 1.0 22,590 670 1100 2610 2180 1440 26 2.0 35,790 980 1610 4100 1390 2070 50 23.2 3.1 3.0 3480 4750 10710 9450 7670 380 16.3 2.4 3.6 34,020 970 1350 3270 1320 1840 52 33.0 4.7 Summer 3 0.5 27,780 460 790 2230 8930 1020 66 31.8 4.0 2.0 33,060 680 1140 2900 12720 1220 110 32.9 4.1 Summer 4 0.5 26,940 520 990 2620 8840 1100 95 24.5 3.0 2.0 29,300 670 1100 2790 7210 1150 67 35.6 4.4 3.5 26,600 520 780 2050 1380 810 54 18.7 2.4 Fall 1 0.5 46,590 810 1060 3190 3870 1870 220 13.8 2.1 2.0 47,940 810 1080 3270 9830 1970 260 11.7 1.7 3.0 48,040 790 1060 3010 5030 1890 210 13.0 1.9 Fall 2 0.5 40,760 740 1060 3310 1650 1710 210 1.5 38,750 710 990 3200 2040 1710 230 1.5 44,110 770 1100 3510 2080 1830 230 3.0 47,540 810 1140 3660 1710 1980 240 Fall 3 0.5 53.940 1030 1440 4410 2400 2590 310 1.5 48,730 850 1210 3680 1970 2120 270 1.8 46,650 810 1180 3530 1720 2040 240 Fall 4 0.5 52,450 870 1170 3580 5220 2220 260 15.8 2.3 1.5 47,790 840 1130 3350 3890 2100 260 12.8 1.9 2.8 45,620 720 860 2380 6260 1520 170 67 Season Station Depth Fe Mn Ni Cu Zn As Pb SPOC SPOIS Winter 1 1.0 42,960 520 1850 10540 3340 3800 260 12.4 2.7 2.0 64,660 810 3170 14880 7200 3950 360 13.9 3.3 2.25 59,280 970 3620 19190 4730 4300 260 2.5 71,220 1140 3970 18920 4840 5190 300 2.75 56,490 900 2480 19370 3600 4410 190 12.6 2.9 3.0 95,010 740 1580 5540 2650 29700 140 10.2 1.7 3.15 58,520 990 2360 6080 3220 6760 120 12.0 2.2 Winter 2 1.0 51,750 510 2960 27600 4890 1600 40 14.5 3.4 1.75 54,380 16.0 4.2 1.75 18.2 4.8 2.25 60,960 800 4210 36290 5920 5030 400 2.5 71,120 830 4890 30920 7020 4460 410 16.8 4.7 2.65 64,240 780 4510 23570 6560 4050 380 2.8 66,020 920 4760 22410 8650 4020 320 14.2 4.2 2.9 96,040 770 2960 12070 5270 25100 310 Winter 3 1.5 54,730 730 2640 15400 12080 4170 300 1.75 47,430 660 2380 14050 16720 4000 250 2 17.3 3.9 Winter 4 1.0 64,220 1020 2840 11400 4920 4480 330 11.8 2.5 1.5 55,520 890 2390 9880 4940 3840 300 2.0 61,120 910 2480 10180 3150 4030 290 13.5 2.7 2.5 54,790 780 2280 10570 7730 3350 320 11.9 2.5 2.6 65,370 890 2740 16490 7630 4220 370 20.3 4.2 2.7 54,690 720 2480 15680 5940 3870 330 2.8 85,830 800 2010 8880 13920 20770 420 12.9 2.0 Spring 1 1.0 54,650 700 930 8670 3000 6030 100 9.8 1.7 2.0 71,280 900 1180 11150 4000 8910 150 9.2 1.7 3.0 61,080 800 1090 9740 3160 7400 110 8.0 1.5 Spring 4 1.0 67,680 940 1040 9340 3000 6780 130 7.5 1.3 1.0 61,990 950 1050 8880 2780 6360 120 2.0 61,500 920 1000 8690 2940 6470 390 6.9 1.2 3.2 58,450 890 910 7620 2300 5440 100 6.5 1.1 68 depth (Table 4.6). Profiles of particulate As species are characterized by large bottom water spikes which range from 111-120 ug-L"1 (Table 4.6). The spring period exhibited the greatest concentrations of metal-bearing particulate phases, with values for Fe, Mn, Ni, Cu, Zn and As averaging 649, 9, 11, 94, 29 and 69 |igL"l. In addition, the total burdens of these metals in the water column received relatively greater contributions from their respective particulate phases; percentages for Fe, Mn, Ni, Cu, Zn and As averaged, respectively, 89,3,3,33,31 and 23 % (Tables 4.6 and 4.7). 4.2 Interstitial Waters 4.2.1 Peepers 4.2.1.1 Nutrients and Sulphate Two pairs of tandem peepers, set approximately 15 m apart in Balmer Lake, were emplaced at stations 1 and 2 for the first deployment during the summer session. The use of two peepers at each site was designed to assess the reproducibility of the dialysis arrays, as well as the homogeneity of the sediments over short lateral distances. A coarse examination of the porewater profiles indicated that the samples collected by the arrays provided consistent representations of the porewater constituents (Fig. 4.12). The porewater results of all parameters are presented in Appendix C. Examination of the duplicate profiles acquired 15 m apart at each site during the summer period reveals marked similarity in the distributions of nitrate, ammonium and sulphate. Values for these constituents in the peeper- sampled bottom waters remain fairly uniform above the sediment-water 69 Fig. 4.12. Duplicate summer peeper profiles of dissolved NFL4+, NO3" and SO^- for station 1 (A-C) and station 2 (D-F), Balmer Lake, June, 1993. Replicate samples are represented by double symbols at specific single depths. 7 0 interface (Fig. 4.12) and closely match those concentrations measured from pumped samples drawn independently during the peeper deployment period (compare Table 4.1 with Appendix C). At both stations, steep porewater ammonium gradients are evident immediately below the sediment-water interface where they approach maxima of 600-700 umol'L"1 at ~ 10 cm (Fig. 4.12). Nitrate profiles show sharp spikes at, or just above the interface, below which values decrease precipitously to low levels at shallow sediment depths (Fig. 4.12). Larger interfacial increases and steeper porewater gradients characterize the distribution at station 1. As for nitrate, sulphate concentrations decrease with depth, although less abruptly; levels fall gradually throughout the sampled sediment column, the profiles being concave-downward (Fig. 4.12). The peeper samples collected in the fall yield nitrate, ammonium and sulphate profiles similar to those observed during the summer period. The duplicate peeper arrangement repeated at station 2 again demonstrates the good consistency between profiles 15 m apart (Fig. 4.14). At all three sites sampled (stations 1, 2 and 5), porewater nitrate values plummet dramatically at shallow sediment depths, reaching very low concentrations at ~5 cm (Figs. 4.13 and 4.14). Sulphate distributions for stations 1 and 2 exhibit good agreement; concave- down profiles decrease gradually throughout the sampled sediment column from ~3.6 to ~0.5 mmolL'l . Peeper data for the shallow site in "Tailings Bay" (station 5), however, show evidence for a bottom water sulphate enrichment before sharply decreasing at, or slightly above, the sediment-water interface to a minimum of ~2 mmolL"!. The ammonium profile obtained for this site relates closely to that observed for sulphate; concentrations are enriched in the lowermost bottom waters and are depleted at, or below, the sediment surface (Fig. 4.13D). Concentrations of ammonium do, however, increase in the porewaters below ~ 10 cm depth at this site. Stations 1 and 2 show similar 71 Dissolved [NH/] , u m o l L 1 Dissolved [N0 3 ] , a m o l L 1 Dissolved [ S 0 4 2 ] , mmolL- 1 Dissolved [NH 4 +], j imolL- 1 Dissolved [NO,"], umolL- 1 Dissolved [ S O / ] , mmo lL 1 Fig. 4.13. Fall peeper profiles of dissolved NH4+, NO3" and SO^-' for stations 1 (A-C) and 5 (D-F), Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. 72 Fig. 4.14. Duplicate fall peeper profiles of dissolved NH4+, NO3" and S04^- (A-C, respectively) for station 2, Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. 73 consistency with respect to ammonium distributions. Steep porewater ammonium gradients are evident immediately below the sediment-water interface where they approach maxima of 400-600 umolL"! (Figs. 4.13A and 4.14A). Peeper values for nitrate, sulphate and ammonium in the fall bottom waters closely match those concentrations measured from pumped samples drawn independently during the peeper deployment period (compare Table 4.2 with Appendix C). The bottom water distributions for nitrate, sulphate and ammonium determined from peeper sampling during the winter period (Figs. 4.15 and 4.16) accentuate nicely the values measured from independently drawn water column samples. At the deeper locales (stations 1, 2 and 4), nitrate profiles exhibit negligible inter-site variation, with values decreasing sharply in the lower bottom waters to minima 5-10 cm above the sediment surface (Figs. 4.15 and 4.16). At the shallow site in "Tailings Bay" (station 6, 2 m depth), nitrate levels decrease immediately below the interface; concentrations of the ion approach zero below 20 cm depth (Figs. 4.15 and 4.16). The deeper stations (stations 1, 2 and 4) exhibit marked similarity with respect to winter distribution of sulphate. Stations 1 and 2 show pronounced bottom water maxima of ~ 11-14 mmolL"! 5-10 cm above the sediment surface, below which their concave-down profiles decrease progressively to < 0.5 mmol-L'l in the lowest sampled horizons (Fig. 4.15). Station 4 is marked by a less intense bottom water enrichment and occurs in vicinity of the interface. Sulphate concentrations in the lower three decimetres of the shallow water column at station 6 vary insignificantly. Levels initially decrease at the sediment- water interface and continue to decline to a depth of 20 cm where values approach a minimum of ~2 mmol-L-1 (Fig. 4.16). 74 Winter W i n t e r p Winter Station 1 b Station 1 ^ Station 1 Dissolved [NH 4 + ] ,nmolL- 1 Dissolved [N0 3] , u m o l L 1 Dissolved [S0 4 2 1, mmolL" 1 Winter p Winter p Winter Station 2 t Station 2 ' Station 2 Dissolved [NH 4 +], u m o l L 1 Dissolved [N0 3] , u m o l L 1 Dissolved [ S 0 4 2 ] , mmolL" 1 Fig. 4.15 Winter peeper profiles of dissolved NH4+, NO3" and for stations 1 (A- C ) and 2 (D-F), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. 75 Winter Station 6 Winter Station 6 Winter Station 6 Dissolved [NH *], nmol L 1 100 200 300 400 500 600 ], j i m o l L 1 Dissolved [NO, Dissolved [S04z], mmolL" F i g . 4 .16 . Win te r peeper prof i les of d isso lved N H 4 + , NO3" and S04^~ fo r stations 4 ( A - C ) and 6 ( D - F ) , Ba lme r L a k e , M a r c h , 1994. Repl icate samples are represented by double symbo ls at speci f ic s ingle depths. 7 6 The winter bottom water and porewater ammonium profiles at stations 1, 2 and 4 are quite similar. Concentrations at all three stations increase substantially in the lower bottom waters to values ranging from ~ 300-580 mmolL"! (Figs. 4.15 and 4.16). Stations 1 and 2 are characterized by bottom water maxima -10 cm above the sediment-water interface. Ammonium concentrations at station 1 steadily decrease below this depth, while those at station 2 are variable near the interface; below the interface at station 2 levels increase gradually through the top decimetre (Figs. 4.15 and 4.16). Below the bottom water maximum at station 4, the ammonium content is invariant to ~15 cm in the sediments, and steadily declines to the base of the profile. Conversely, the ammonium distribution at station 6 is characterized by a pronounced bottom water depletion in the top decimetre above the sediment surface. Concentrations decrease gradually to shallow sediment depths, below which levels increase in a concave-down profile (Fig. 4.16). The duplicate peepers deployed at station 1 during the spring period yielded good precision with respect to dissolved nutrients and sulphate. The nitrate profiles are similar to those observed at this locale during the summer and fall; uniform bottom water values decline precipitously in the top 10 cm to minima of < 10 umolL"! (Fig. 4.17). Ammonium distributions are also similar to those seen during ice-free periods (Fig. 4.17). However, sulphate profiles differ greatly from those seen during other seasons. Concentrations increase sharply from uniform bottom water values of -3.5 mmolL"! to subsurface maxima of ~6 mmol-L"l at 5-10 cm depth, and decrease rapidly below this horizon to values that approach 1 mmol-L"l (Fig. 4.17). 77 A Spring Station 1 B Spring Station 1 Fig. 4.17. Duplicate spring peeper profiles of dissolved NH4"1", NO3" and S04^- (A-C, respectively) for station 1, Balmer Lake, May, 1994. Replicate samples are represented by double symbols at specific single depths. 78 4.2.1.2 Trace Metals Porewaters were analyzed for a suite of dissolved trace metals including Mn, Fe, Ni , Cu, Zn, As and Pb. Throughout the seasons and at all depths sampled, dissolved Pb concentrations generally hovered at or below its detection limit of ~ 0.5 ug'L"1. Consequently, plots of porewater Pb distributions are not presented; measured values, however, are reported in Appendix C. Examination of the respective trace metal profiles suggest that some discrete samples may have been contaminated. Contamination could have resulted from the addition of small amounts of tailings to the porewater samples during peeper subsampling; the subsequent addition of acid would then have dissolved such metal-rich particles, yielding high dissolved metal concentrations. Given that diffusion acts to reduce concentration differences relatively quickly over short spatial scales, the principal criterion used here to define suspected contamination of samples is an obvious lack of consistency of the high concentrations with lower values measured in adjacent horizons. Such single-sample "spikes" are rejected from subsequent interpretation. Trace metal distributions observed during the summer period at stations 1 and 2 are very similar (Figs. 4.18 and 4.19); reasonable agreement also exists between the intra-site measurements. Profiles are generally characterized by uniform distributions above the sediment-water interface and sharp sub-surface concentration gradients in the shallow porewaters. Dissolved Mn concentrations increase sharply below the sediment-water interface to concentrations as high as -800-1100 ug'L"1 between 10-15 cm depth (Figs. 4.18 and 4.19). Below depths of 10 cm, Mn levels remain fairly constant. Dissolved As closely tracks Mn at both stations. Concomitant increases in As levels result in large sub-surface maxima of 2-5 mg'L"1 within 10 cm of the sediment-water interface (Figs. 4.18 and 4.19). 79 Fig. 4.18. Duplicate summer peeper profiles of dissolved Fe, Mn, As, Ni, Cu anci Zn (A- F, respectively) for station 1, Balmer Lake, June, 1993. Replicate samples are represented by double symbols at specific single depths. 80 Fig. 4.19. Duplicate summer peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A- F, respectively) for station 2, Balmer Lake, June, 1993. Replicate samples are represented by double symbols at specific single depths. 8 1 As profiles also show evidence of depletion at greater depths. Note that at station 1 (Fig. 4.18), the magnitudes of the subsurface maxima differ significantly. Significant enrichments of dissolved Fe appear below those observed for manganese. At both stations, concentrations increase rapidly between 5-10 cm, below which gradual increases result in peeper bottom values up to 4 mg'L"1 (Figs. 4.18 and 4.19). Dissolved Zn profiles exhibit variable sub-surface enrichments in the top 5 cm ranging from 60-150 ug'L"1, which immediately fall off to concentrations less than half of the observed water column values. Dissolved Cu and Ni profiles exhibit marked similarity at the two sampled stations (Figs. 4.18 and 4.19). With two exceptions, the concentrations of both elements decrease sharply immediately below the sediment-water interface and remain generally constant below minima of ~5 ug'L"1 for Cu and -60 ug'L"1 for Ni. The exceptions are two dissolved Ni profiles, one at each site, which show evidence of slight subsurface enrichments in the first few centimetres below the interface (Figs. 4.18 and 4.19). Results from the summer period exhibited excellent agreement between bottom water values obtained from peepers and those from concurrently drawn pumped samples (compare Figs. 4.18 and 4.19 with Table 4-1). The trace metal porewater data for the fall period are comparable to the summer results (Figs. 4.20-4.22). The tandem peepers at station 2 yielded consistent profiles for all constituents (Fig. 4.21). Dissolved Mn and As porewater distributions are nearly parallel. Stations 1 and 2 are characterized by initially steep Mn gradients in the top 5 cm, below which concentrations gradually increase to values ranging from -700-1000 ug'L"1 in the deepest horizons (Figs 4.20 and 4.21). Mn levels in the porewaters of "Tailings Bay" (station 5) steadily increase from -200 ug'L"1 at the interface to 1800 ug'L"1 between 20 and 30 cm. Below this maximum, concentrations drop to -700 ug'L"1 82 0 200 400 600 800 1000 0 200 400 600 800 0 200 400 600 800 1000 Dissolved [Fe], ppb Dissolved [Mn], ppb Dissolved [As], ppb Fig. 4.20. Fall peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 1, Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. 83 Fig. 4.21. Duplicate fall peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 2, Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. 84 Dissolved [Fe], ppb Dissolved [Mn], ppb Dissolved [As], ppb Fig. 4.22. Fall peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 5, Balmer Lake, October, 1993. Replicate samples are represented by double symbols at specific single depths. 85 in the deepest samples (Fig. 4.22). At all stations, dissolved As concentrations begin to rise at shallow sediment depths, reaching maxima between ~900 and 3000 ug'L"1 (Figs. 4.20-4.22). Dissolved Fe increases in concentration with depth at all sites during the fall period; however, the depths and magnitudes of the enrichments vary. Levels at station 1 remain essentially uniform to a depth of 20 cm whereupon concentrations increase gradually to a maximum of ~1000 Ug'L"1 at the base of the profile (Fig. 4.20). Iron distributions seen at stations 2 and 5 are characterized by variable increases starting at 5-10 cm depth (Fig. 4.21 and 4.22), with concentrations rising to values as high as -3-4 mg'L"1 in the deepest samples at station 2, but only to 400-500 ug'L"1 at station 5. Dissolved Zn profiles from the fall at stations 1 and 2 exhibit pronounced, shallow sub-surface enrichments ranging from -130-300 Ug'L"1 (Figs. 4.20 and 4.21) . Below these maxima, concentrations decline dramatically to values less than their respective bottom water levels. In contrast to the other sites, dissolved Zn concentrations at station 5 generally decrease from the uppermost sampled bottom waters to shallow sediment depths (Fig. 4.22). Below a few centimetres in the sediment column at this site, dissolved Zn levels decrease marginally (Fig 4.22) . The bottom water and porewater distributions of dissolved Cu and Ni from the autumn sampling are similar to those seen in the summer, exhibiting minimal lake-wide variability. At stations 1 and 2 (Figs. 4.20 and 4.21), Cu concentrations decline precipitously at shallow sediment depths, approaching minima of <20 ug'L"1 -5 cm below the interface. Station 5 is characterized by a similar gradient; decreases, however, begin at or slightly above the sediment surface (Fig. 4.22). Steep sub-surface gradients are also evident for dissolved Ni distributions at all stations sampled. Nickel profiles at station 1, and to a lesser 86 extent at station 2, show distinct subsurface enrichments, before levels dramatically decline immediately below their respective maxima (Figs. 4.20 and 4.21). As seen for Ni, dissolved Cu concentrations at station 5 begin a steep decline at or above the sediment surface (Fig 4.22). The winter distributions of the various dissolved constituents demonstrate that a dramatic shift in the chemical environment of the bottom waters and porewaters occurs during periods of ice-cover (Figs. 4.23-4.26). The bottom water enrichments of dissolved Mn that were clearly depicted at all sampled stations from water column profiling are similarly represented in their respective peeper profiles. Stations 1, 2 and 4 are characterized by increasing dissolved Mn concentrations throughout the lowermost bottom waters to the interfacial porewaters (Figs. 4.23-4.25). Profiles at stations 1 and 2 continue to increase below the sediment surface, reaching maxima between 1200-1400 ug'L"1 at depths of ~15 and 5 cm, respectively (Figs. 4.23-4.24). At greater depths at these sites, and below the interfacial maximum at station 4, dissolved Mn concentrations decrease to ~600-700 ug'L"1 and remain constant to peeper-bottom (Figs. 4.23- 4.25). The bottom waters in "Tailings Bay" (station 6) show no evidence of a near- interface dissolved Mn enrichment; concentrations remain essentially constant to a sediment depth of ~ 10 cm then fall steeply to levels that approach 20-30 ug'L"1 (Fig. 4.26). Dissolved Fe distributions in the bottom waters and interstitial fractions observed during the winter period exhibit considerable lake-wide variability. Station 1 is characterized by relatively uniform levels of dissolved Fe to a sediment depth of ~5 cm, below which values increase erratically to ~2100 Ug'L"1 in the lowest horizons (Fig. 4.23). Concentrations at station 4 remain < 200 ug'L"1 in the bottom waters and porewaters to a sediment depth of -40 cm, but increase to -600 ug'L"1 at greater depths (Fig. 4.25). In contrast, the bottom waters 87 Winter E Winter F Winter Station 1 Station 1 Station 1 i u u ^ u u J U U « u u = u u o u u 0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 Q 5 0 1 0 0 1 5 0 2 0 0 Dissolved [Ni], ppb Dissolved [Cu], ppb Dissolved [Zn], ppb Fig. 4.23.. Winter peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 1, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. 88 Winter Station 2 B Winter Station 2 0 200 400 Dissolved [Fe], ppb 0 400 800 1200 Dissolved [Mn], ppb Winter Station 2 30 E o 8 2 0 & 10 ». CD s S-iof E CO E o 4= -30 CD O ca -40 CO b -20 -50 I - 1 — I I I—I I | T—i—i—r—i—i—r—i—r 0 200 400 600 800 Dissolved [As], ppb Winter Station 2 Winter Station 2 Winter Station 2 0 100 200 300 400 500 Dissolved [Ni], ppb 0 20 40 60 80 100 Dissolved [Cu], ppb 0 50 100 150 200 250 300 Dissolved [Zn], ppb Fig. 4.24. Winter peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 2, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. 89 Fig. 4.25. Winter peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 4, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. 90 Winter Station 6 B 40 E u <B o •g a> .c i— CD T5 c CD E (0 E 8 <D O c w 20 -20 -40 -60 Winter Station 6 0 100 200 300 400 Dissolved [Fe], ppb Winter Station 6 0 100 200 300 400 Dissolved [Mn], ppb -60 o 5 10 15 Dissolved [As], ppm D Winter Station 6 Winter Station 6 100 200 300 400 500 Dissolved [Ni], ppb 200 Winter Station 6 200 Dissolved [Cu], ppb Dissolved [Zn], ppb Fig. 4.26. Winter peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 6, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. 91 immediately above the sediment surface at station 2 are enriched in dissolved Fe. Below the interface, concentrations fluctuate significantly throughout the sediment column (Fig. 4.24). The strong similarities between the distributions of dissolved As and Mn observed for the summer and fall periods are not clearly obvious during the winter session. The As profiles at stations 1 and 2 are characterized by zones of generally increasing concentration, extending from the bottom waters, across the sediment-water interface, to depths of -20 and 10 cm, respectively (Figs 4.23 and 4.24). Dissolved As profiles at stations 4 and 6 exhibit uniform bottom water values and increased levels at shallow sediment depths (Figs. 4.25 and 4.26). A zone of relatively high As concentrations between -5 and 15 cm depth at station 4, overlies porewaters that contain variably lower concentrations (Fig. 4.25). Dissolved As at station 6 increases essentially monotonically throughout the sampled sediment column (Fig. 4.26). Dissolved zinc distributions during the winter period are similar to those observed for other seasons; profiles are characterized by variable increases near the interface at each site, below which levels decline sharply to minima in the top two decimetres (Figs. 4.23-4.26). The well-defined bottom water enrichment of dissolved Zn seen at station 1, corroborates that observed in the water-column sample set (Fig. 4.23). Similar maxima at stations 2 and 4, although evident from water column profiling, are not as clearly defined by the peeper data (Figs. 4.24 and 4.25). The winter distributions of dissolved Ni demonstrate significant lake-wide consistency and resemble profiles observed during other seasons (Figs. 4.23-4.26). Gradients in bottom waters are not distinct in the peeper profiles except perhaps at station 4, where the concentration decrease accords with the respective water column profile (see Fig. 4.1 ID). At all stations, dissolved N i concentrations 92 diminish abruptly from values of 300-450 Ug'L"1 in bottom waters to minima of < 40 ugL"1 in the first two decimetres (Figs. 4.23-4.26). Slight sub-surface maxima can be seen at stations 1,2 and 4 (Figs. 4.23-4.25). The gradients of dissolved Cu in winter bottom waters seen in the water column sample set (Figs. 4.8-4.11) are verified by corresponding peeper data (Figs. 4.23-4.26). Profiles at stations 1, 2 and 4 clearly illustrate the zone of decreasing dissolved Cu concentrations above the interface (Figs. 4.23-4.25). Below this horizon, values continue to decline at stations 1 and 4 to concentrations < 10 ug'L"1 (Figs. 4.23 and 4.25). Station 6 is characterized by a more intense sub-surface gradient where levels decrease from over 400 to < 20 ug'L"1 in the top 20 cm (Fig. 4.26). Sampling conducted in the spring following ice break-up yielded bottom water and porewater distributions of dissolved trace metals comparable to those seen during the well-mixed periods of summer and fall. Logistic and sampling problems restricted the deployment of only two peepers which were emplaced 20 m apart at station 1. Concentrations for all constituents exhibit uniform profiles in the bottom waters and agree well with independently measured water column values (compare Fig. 4.27 with Table 4.4). Reproducibility between the profiles was fair for dissolved Cu and Ni, but limited for Fe, Zn and As. Dissolved Mn contents in the spring porewaters at station 1 sharply increase at shallow sediment depths to maxima ranging from ~820-1100 ug'L"1 (Fig. 4.27). Below these peaks, concentrations steadily decline to values of -500- 700 Ug'L"1 at -40 cm depth. The dissolved Fe profiles are characterized by dissimilar sub-surface enrichments. Concentrations obtained from peeper 1 remain constant to a depth of -20 cm, at which point levels increase rapidly to -1 mg'L"1 in the deepest horizons (Fig. 4.27). Conversely, the profile of peeper 2 shows only slight enrichments up to -400 ug'L"1 between depths of 10 and 20 cm. 93 Dissolved [Fe], ppb Dissolved [Mn], ppb Dissolved [As], ppb Fig. 4.27. Duplicate spring peeper profiles of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) for station 1, Balmer Lake, May, 1994. Replicate samples are represented by double symbols at specific single depths. 94 The distributions of dissolved As and Zn are both characterized by sub- surface enrichments in the shallow porewaters at station 1. Variable As maxima occur between depths of 10-20 cm; concentrations below theses enriched horizons generally decrease to peeper bottom (Fig. 4.27). Profiles of dissolved Zn also exhibit broad maxima of variable magnitude between the sediment surface and depths of -10 cm (Fig. 4.27). Concentrations in one profile steadily decline at greater depths, while in the other (peeper 2), they increase again below a minimum confined between 10 and 20 cm depth. Although bottom water concentrations of dissolved Ni and Cu obtained from peeper 1 are significantly lower than those from peeper 2 (Fig. 4.27), their profiles match well at shallow sediment depths. Sharp decreases are evident for both elements in close proximity to the sediment surface, and minima are approached in the first decimetre. Dissolved Ni profiles exhibit subtle sub- surface maxima at depths of 5 cm. 4.2.1.3 pH In the spring sampling period, selected cells down the length of one peeper at station 1 were measured for pH by direct insertion of the probe into each chamber. The pH profile is fairly uniform, with values ranging from 7.32- 8.02 (Fig. 4.28). A slight increase in the vicinity of the interface contrasts with a subtle decrease towards deeper horizons. 95 Fig. 4.28. Spring profile of porewater and bottom water pH measured from peeper 2, station 1, Balmer Lake, May, 1994. 96 4.2.2 Core Porewaters 4.2.2.1 Nutrients and Sulphate Analytical results obtained from centrifuged core samples have been plotted against peeper-derived profiles in order to compare the two methods. Core porewaters were not analyzed for ammonium; consequently, peeper/core comparisons are limited to nitrate and sulphate concentrations. The interstitial nitrate and sulphate fractions from duplicate cores at station 1 demonstrate excellent inter-core coherence and agree well with peeper values (Fig. 4.29). Concentrations determined from the core supernatant, however, disagree with lowermost bottom water measurements acquired from peeper sampling. It appears the composition of the core-top waters are similar to values measured 20-30 cm above the sediment surface. Examination of the data revealed that the disparity is not due to any analytical difficulty, but instead reflects a design limitation of the corer. The upper portion of the sleeve valve of the corer has two large ports which allow upward-flowing water to escape during instrument descent. However, the flow is not free, in that the total cross-sectional area of the ports is less than that of the mouth of the core barrel. The resistance to the flow of water through the instrument ports during lowering causes some water inside the barrel to be carried down with the corer, for probably not more than 1-2 metres. In regions of fine-scale stratification, there is the potential for the supernatant water to consist of a mixture of waters from distinct layers. In addition, the top corer ports sit 30-50 cm above the sediment surface, a distance greater than the thickness of the compositionally distinct strata observed in the bottom waters. Since the core-top waters most likely represent an integrated sample of the lower bottom waters, data from these samples have been plotted 20 97 Winter Station 1 B Winter Station 1 Fig. 4.29. Distributions of dissolved NO3" and S O 4 2 " obtained from peeper (open circles) and core porewaters (diamonds) at station 1 (A-B) and station 6 (C-D), Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. 98 cm above the sediment-water interface. In the following discussion, however, only peeper-derived bottom water values will be used in comparisons with shallow porewaters. The core-derived nitrate profile in the sediments in "Tailings Bay" (station 6) exhibits a steeper sub-surface gradient than the analogous peeper profile; core values approach minima at a sediment depth of 10 cm as opposed to 20 cm (Fig. 4.29). Excluding one presumably contaminated core sample, the corresponding sulphate distributions parallel one another reasonably well. 4.2.2.2 Trace Metals The duplicate trace metal distributions obtained from core samples at station 1 are generally similar in form but exhibit variable congruence with the respective peeper profiles. Dissolved Fe concentrations at station 1 increase at depth, with core-derived profiles exhibiting steeper gradients (Fig. 4.30). The magnitudes of sub-surface dissolved Mn maxima differ markedly in the two cores; core lb is comparable to the maximum in the peeper profile, albeit shallower, while the corresponding peak in core la is -400 ug'L"1 less. Fundamental differences are evident in comparisons of dissolved As distributions. Core-derived duplicate profiles show dramatic sub-surface enrichments in excess of 1200 ugL"1 at 5 cm depth; such increases are not observed in the peeper profile (Fig. 4.30). Distributions for dissolved Ni, Cu and Zn at station 1 exhibit better agreement between their respective core and peeper profiles; decreases in their concentrations occur abruptly at shallow sediment depths. The core-derived sub-surface gradients for all three elements, however, are steeper than those seen in the peeper profiles (Fig. 4.30). 99 Winter Station 1 40 •' • • I • • •' I' • • • 11 • ' ' l • • • • I • • • •! • —e--Peeper 0 Core 1 • Core 2 0 1000 2000 3000 Dissolved [Fe], ppb B Winter Station 1 40 E 0 <D u •g S c 1 • • c CD E <o E e CD O c CO • CO b 20 20 -40 -60 t t i » t t t i i ) i 11 i i i i i i | i t i t n -e— Peeper o Core 1 • Core 2 i •, • • * t • •»»i <»* 0 500 1000 1500 Dissolved [Mn], ppb Winter Station 1 40 E o •g 20 0? * 0 c CD E 00 E o 8- c ca CO b 40 -60 1 T [ | T T I | T T T | T I I | I I I | 1 1 I [ I I I —e--Peeper o Core 1 • Core 2 0 400 800 1200 Dissolved [As], ppb Fig. 4.30. Dissolved distributions of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) obtained from peeper (open circles) and core porewaters (diamonds) at station 1, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. 100 The differences in peeper- and core-derived trace metal profiles at station 6 are similar to those at station 1 (Fig. 4.31). For most elements, gradients in the core porewaters are steeper. For example, decreases in dissolved Mn, Ni, Cu and Zn in core 2 occur over a sediment depth one-half that of decreases of similar magnitude in the peeper profile. Similarly, porewater As increases more rapidly with depth in the core porewaters (Fig. 4.31) 4.3 Sediments Three high quality cores were obtained using the lightweight gravity corer described by Pedersen et al. (1985). Duplicate cores la and lb were collected approximately 3 m apart in the more "natural" sediments at station 1, while core 2 was taken from a predominantly tailings region in "Tailings Bay" (station 6). Core logs are presented in Appendix D. Duplicate cores la and lb were visually and compositionally identical. Both cores were veneered by a 2 to 3 mm thick cap of brown noncohesive very fine-grained sediment. The top two centimetres were additionally characterized by alternating light/dark millimetre-scale laminations. These surface horizons were underlain by ~2 cm of gray, homogenous, less-cohesive ooze which graded into a black layer occupying a depth from 4 to 8 cm. Small bubbles, presumably methane, were evident in the latter. Below the 10 cm horizon, a homogeneous gray/chocolate brown ooze comprised the remainder of the core. Core 2, collected from station 6 in "Tailings Bay" was very different from the duplicate cores obtained at station 1. Most of the core comprised homogeneous, very fine grained, light-gray, silty tailings material; intermittent thin amber bands (~1 mm thick) occurred in the top 3 cm. 101 Fig. 4.31. Dissolved distributions of dissolved Fe, Mn, As, Ni, Cu and Zn (A-F, respectively) obtained from peeper (open circles) and core porewaters (diamonds) at station 6, Balmer Lake, March, 1994. Replicate samples are represented by double symbols at specific single depths. 102 The "natural" sediments at station 1 and the tailings deposits at station 6 are compositionally distinct, particularly below 10 cm depth, as shown by the solid-phase ratio data (Figs. 4.32-4.33). Below the 10 cm horizon, the distributions of most parameters remain essentially constant at both sites. The upper decimetre at station 1 has clearly received a significant input of anthropogenic material. XRD analyses indicate that the "natural" Balmer Lake sediments and the tailings material represent two distinct mineralogical assemblages; the major and minor elemental compositions of the three cores are reported in Appendix E. The natural deposits contain variable mixtures of quartz and feldspar (mainly plagioclase). These contrast with the tailings deposits, which are relatively feldspar-poor and replete with quartz, abundant clay minerals and halite. The clay fraction is composed primarily of chlorite, with minor contributions from montmorillonite and micaceous minerals (illite and biotite). Both natural sediment and tailings signatures are evident in the top horizons. Cores la and b consist of essentially tailings-free, organic-rich sediments below -10 cm depth, as shown by the Corg/ nitrogen, P, Ni , Cu, Zn and As distributions (Figs. 4.34-4.35). Extraordinarily high metal concentrations in the top several centimetres of both cores, reaching up to -0.5 wt. % Ni, 1.8 wt. % Cu, 0.6 wt. % Zn and 0.5 wt. % As (Fig. 4.35), demonstrate that the upper stratum is composed of a mixture of tailings material and natural deposits, which is capped by a 0.5 to 1 cm thick surface veneer of more natural sediments. The opposing CaC03 and organic carbon distributions in the upper 7 cm reflect dilution of natural sediments by relatively carbonate-enriched tailings (Fig. 4.34). The data in Figures 4.34E and 4.35A imply that S and Mn are also enriched in the tailings component at station 1. 103 B E 1 0 u £ O 15 u c Q. CO Q 25 30 35 - Core 1a •o- - Core 1b -Core 2 i m l i i i l l n n llllllllllllllllllllll 3 4 5 Si/AI Wt. Ratio 0 I i i i i 1.1 i i i | i i i Clfai \» i | i i i i 5 r E i o o 2 O 15 u _c £ 20 CD Q 25 t 30 35 •O—Core 1a •e -Core 1b —<•—Core 2 • • • . i • . • • i . 0.02 0.04 0.06 Ti/AI Wt. Ratio I 0 r E o CD O 15 O e £20 CD a 25 30 r 35 HJJJIIIIIU 0.1 0.3 0.5 0.7 Mg/AI Wt. Ratio D E 1 0 o 2 O 15 o CD o 25 r 30 35 1 1 1 1 1 1 1 1 1 m ^ f t - t | t t . 1 - < O."0 y "(3.°-. • 6b : * < • - \ * ^ - —O -Core 1a • Core 1b [ —•— Core 2 • • 1 • • • • 1 * . i — i . . . . 0.05 0.1 0.15 0.2 K/AI Wt. Ratio 0 25 0.02 0.06 0.1 0.14 Na/AI Wt. Ratio 0.18 10 15 I 20 a' c 25 30 35 • 1 1 1 1 1 i m i ' 1 1 1 1 .'.ici.y 1 ' 1 •O— Core 1a •Q—Core 1b —»—Core 2 1 * * 1 • * • * 1 • • • • ' • • 1 • ' 0 0.01 0.02 P/AI Wt. Ratio Fig. 4.32. Sedimentary weight ratio profiles of Si/AI, Ti/AI, Mg/AI, K / A l , Na/AI and P/Al (A-F, respectively) for station 1 (duplicate cores l a and lb , open symbols) and station 6 (core 2, closed symbols), Balmer Lake, March, 1994. 104 E 1 0 o 2 o 15 u c ct CD Q 25 30 35 i n | i i I»I U !C» I • " I ' —o- - Con 1a -a- - Con 1b -Con 2 9 "**.*H 9% ' •fa • 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 2 8 12 Zr/AIWt. Ratio (*10 4) B E 1 0 u 2 O 15 o c f 20 Q 25 30 35 -1 1 1 1 1 1 1 1 1 X - '"'•'•'°£*.... •v u • > 6 a V 6o : —O—Cora 1a —El--Core 1b di : —•—Core 2 . . . . i . . . 1 1 1 . . i • 0 10 20 Sr/AIWt. Ratio (*10 4) c rO-^EJ:..;...... •|3"--.0 • rjjr'' * f. '• oh '. K 9 W / 9 —O— Core 1a —a—Core 1b d : ! —•—Core 2 . — i — • — • — 0 2 4 6 8 Rb/AI Wt. Ratio (*10 4) Fig. 4.33. Sedimentary weight ratio profiles of Zr/Al, Sr/Al and Rb/AI (A-C, respectively) for station 1 (duplicate cores la and lb, open symbols) and station 6 (core 2, closed symbols), Balmer Lake, March, 1994. 105 Total Sulphur, wt.% Fe/AI Wt. Ratio Fig. 4.34. Sedimentary concentrations of organic carbon, calcium carbonate, total nitrogen, values for organic carbon/nitrogen, total sulphur and iron/aluminum weight ratios (A-F, respectively), for station 1 (duplicate cores la and lb, open symbols) and station 6 (core 2, closed symbols), Balmer Lake, March, 1994. 106 0 I i • i i I i i i • I O B 1 . L E o E o o CD Q 1 0 r 3 5 9" f o 6 1 5 2 0 i T I ; { 2 5 : h i i . . .< j— C o r e 1a 3 0 j : o •••O - C o r e 1b — « — C o r e 2 t I I « I I 1 I I I I I 1 1 I I • ' • I I I I » 0 1 0 0 0 2 0 0 0 Manganese, | igg ' 1 5 r 1 0 3 0 ft — o — C o r e l a — Q - C o r e 1b — • — C o r e 2 1 * • * • 1 • • • * 1 * * • ' 0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 ,-1 Nickel, ngg" 3 0 K 3 5 —o- - C o r e 1a •a -- C o r e 1b — » - - C o r e 2 I I I I I I 1 I I I I t I I I I 2 0 0 0 4 0 0 0 6 0 0 0 Zinc, ugg" 1 D E F Copper, jig-g"1 Arsenic, j igg- 1 Lead, jig-g"1 Fig. 4.35. Sedimentary concentrations of manganese, nickel, zinc, copper, arsenic, and lead (A -F , respectively) for station 1 (duplicate cores la and lb, open symbols) and station 6 (core 2, closed symbols), Balmer Lake, March, 1994. 107 In contrast to the "natural" sediments site, "Tailings Bay" sediments (core 2) are characterized by higher carbonate contents, low overall organic C, total S, and total N and P contents, and higher F e / A l weight ratios (Fig. 4.34). Paradoxically, the concentrations of total sulphur and trace elements (e.g., Ni , Cu, Zn and As) are significantly depressed at shallow sediment depths relative to maximum values observed at station 1 (Fig. 4.34 and 4.35). Profiles for Mn, Ni , Zn, Cu, As, total S, and Fe/Al are essentially invariant throughout the sampled sediment column, with the exception of slight surficial enrichments in Cu and As. The compositional contrasts between the two lake sites can be further assessed by means of element/Al ratios. Variations in such ratios can be interpreted in terms of the texture and mineralogy of sediments (Calvert, 1976). The ratio profiles in Figs. 4.32 and 4.33 show that there is some compositional similarity in the top horizons, but at depths greater than a few centimetres, there are major differences in mineralogy between the two locations. 108 V. DISCUSSION 5.1 Water Column Balmer Lake is a highly dynamic system which exhibits pronounced seasonal variability, both physically and chemically. The non-steady state nature largely results from the geographic position of this inland, boreal waterbody, and from variable loadings and dynamics of mining-related inputs. The annual freeze-thaw cycle of the terrestrial and aquatic regimes fosters significant seasonal signatures. For example, the formation of ice in early winter essentially cuts off atmospheric interactions for several months; subsequent reactions in such shallow "closed systems" can greatly affect water column chemistry. The freeze-up period is also characterized by reduced circulation, allowing the development of fine-scale stratification. In addition, the associated seasonal temperature swings of up to 20°C undoubtedly influence the kinetics of abiotic arid microbially mediated reactions in the water column and sediments. Pronounced seasonal pulses of fresh water and associated allochthonous inputs to Balmer Lake are evident in the spring. Spring-melt at these latitudes is rapid, and the combined inputs of terrestrial run-off and the melting of 1 m of surface ice, may lead to a 30 % increase in liquid-water volume. Furthermore, vigorous stream activity can result in increased particulate loads in spring. The nature of the tailings circuit design, in combination with the annual freeze/thaw cycle in Balmer Lake, results in a pronounced seasonal signal with respect to heavy metals and cyanide. Dramatic increases in the concentrations of these parameters are consistently observed at the Balmer Lake outlet (Fig. 2.1) in the spring, and have been shown to be strongly correlated with ice deterioration 109 (EIG, 1986). Moreover, the total inventory of heavy metals also increases at this time. Four independent controls are believed to account for the observed annual cyclicity. First, the absence of wind turbulence under the ice allows the denser, metal-rich effluents to collect in the bottom waters over the winter months. Prior to whole-lake turnover in the spring, the outfall drains the less contaminated surface waters. However, upon ice break-up and subsequent wind-mixing, the accumulated inventory of mine-effluents is dispersed throughout the water column; this can contribute to higher metal levels at the outfall. The total increased metal burden in Balmer Lake at ice-out can also be attributed to freeze/thaw processes. As will be mentioned in section 5.1.4 below, the resuspension of a non-cohesive flocculant layer (built up over the winter months) has been suggested to account for lake-wide increases in springtime metal levels (EAG, 1986). Greater dissolved loads at this time may also stem from the dissolution of cyanide compounds. Metal-cyanide complexes, such as ferricyanide [Fe(CN)6]"4, can form insoluble double-metal cyanide precipitates such as copper ferrocyanide [Cu2Fe(CN)6] (Higgs, 1979). Such complexes are stable in the absence of UV radiation and may potentially accumulate on the lake bottom over the winter months. Upon ice break-up, however, their dissolution may foster significant releases of CN" and associated metals into the water column. Finally, the processes described above also occur in the tailings ponds; however, similar increases in heavy metal concentrations and cyanide precede those in Balmer Lake due to the earlier thaw. Thus, seasonally elevated inputs from the tailings circuits also contribute episodically to the lake burden. 110 5.1.1 Physical Lirnnology The shallow water column of Balmer Lake is fairly well-mixed for the approximately 7-month ice-free period. During this time, the solubility of dissolved oxygen is temperature dependent. The generally windy conditions during the summer field session obviate detectable thermal stratification (Fig. 4.1). The high biological oxygen demand of the organic-rich sediments in the lake, (6-10 % organic carbon) in combination with restricted atmospheric exchange, fosters a gradual depletion of dissolved oxygen in the deeper lake areas over the duration of ice coverage (~ 5 months) (Fig. 4.3). Spring sampling, conducted less than 2 weeks after ice-off, revealed no traces of the previous stratification (Fig. 4.4). The rapid breakdown of chemical gradients and the pronounced warming of the water column over a short time period in the spring, suggest that wind and solar influences play a significant seasonal role. 5.1.2 Sulphate The oxidation of sulphide-bearing ore in the milling circuit liberates large quantities of SO42" which are delivered with the tailings to the clearwater ponds. Dissolved SO42" loads entering Balmer Lake average over 20 mmol/L. This has resulted in a lake-wide yearly average dissolved sulphate concentration of > 3 mmol/L, greatly exceeding natural concentrations in non-perturbed Canadian Shield lakes (Nriagu et al, 1982). Due to the scale of this input, month-to-month variations in the sulphate loadings from the two mines should ultimately determine any seasonal variability. I l l 2 While SO4 " distributions remain essentially constant through the spring, summer and fall, winter profiling revealed large bottom water enrichments in the deeper lake areas (Fig 4.7). Peeper profiles of bottom water and porewater sulphate, however, indicate that the sediments were not releasing sulphate during this period; interfacial gradients in fact suggest pronounced sedimentary SO4 " consumption (Figs. 4.15-4.16; porewater sulphate profiles will be discussed in greater detail in section 5.3.2). The bottom water sulphate increases (which are seen only at depths > 2.5 m) must therefore represent a lateral advective flow of dense, sulphate-rich water moving out across the deeper lake basin. Indeed, profiling conducted to 2.6 m at the shallowest water column site (station 3), and at the shallow peeper site (station 6) did not reveal any evidence of such enrichments (Fig. 4.7C and 4.16F, respectively); only a marginal signal was observed at the 3 m depth at station 4 (4.7D). Although the exact source and timing of such events cannot be determined accurately, the respective inputs from the Dickenson and Campbell tailings ponds represent two obvious potential sources of the high density discharge. During the winter sampling period in March, 1994, the Campbell mine was not discharging to Balmer Lake; complete cessation of discharge was in effect from Dec. 10/'93 to May 2/'94 (PDI, pers. comm.). However, waste inputs from Dickenson persisted throughout the winter months. The effluents potentially contributing to the observed chemical stratification averaged approximately 25 mmol/L SO42", 1.5 mmol/L N r V , 10.2 mg'L"1 C N , 2.5 mg'L"1 Cu, 1.5 mg'L"1 Ni, 1.0 mg'L"1 Zn, and 2.0 mg'L"1 As (Goldcorp, pers. comm.). The high dissolved salts content of the tailings effluents, reflecting largely the high sulphate concentrations, would impart considerably higher densities than associated lake waters. An episode of slumping and oxidation of sulphide-rich 112 tailings material from somewhere within the lake represents a less likely, indeed implausible, explanation for the bottom water SO42" increases. During the fall sampling period, a bottom water sulphate enrichment was also evident at the site most proximal to the Campbell Mine effluent discharge (Fig. 4.13F). In light of the typical windy conditions, it is reasonable to assume that during ice-free periods such profiles would only be manifested within a short distance from the input point. The winter water column distributions of dissolved Fe, As and Cu suggest that sulphide precipitation occurs in the seasonally reducing bottom waters of Balmer Lake (Figs. 4.8-4.11). Due to the large excess of S042", however, direct evidence for sulphate reduction in the water column profiles is probably obscured. Bottom water SO4 "distributions seen in peeper profiles at stations 1 and 2, and to a lesser extent at station 4, suggest a mechanism for sulphate removal near the sediment-water interface at these sites (Figs. 4.15 and 4.16). Considering the trace metal profiles, the abundance of reducible sulphate and the reducing conditions prevalent in the winter, the seasonal precipitation of authigenic mineral sulphides in the water column is a tenable hypothesis that will be further assessed in the trace metal sections below. 5.1.3 Ammonium and Nitrate The burdens of ammonia and nitrate in the water column of Balmer Lake greatly exceed concentrations observed in non-contaminated fresh water bodies. The observed enrichments can be partially attributed to mining-related by- products delivered to the lake via drainages from lake-side clearwater polishing ponds. More specifically, excess levels of NH 4 * and N 0 3 appear to be derived from the breakdown of cyanide compounds. 113 The addition of sodium cyanide in the tailings circuit facilitates the removal of gold, and other metals, in carbon-in-pulp milling operations. The long-term use of such concentration techniques has resulted in significant additions of cyanide to the waters of Balmer Lake, with levels in the past sometimes exceeding 5 mg/L. Historically, mitigation of cyanide inputs has entailed degradation that occurs naturally in tailings ponds. In such systems, the volatilization of H C N to the atmosphere has been shown to be primarily responsible for cyanide removal (IEC, 1979). More sophisticated treatments emerged with stricter environmental standards in the late 1970's, at which time hydrogen peroxide (Vickell et al, 1989) and Inco-S02 (Devuyst et al, 1982) methods became more widespread. The large ammonium burdens in Balmer Lake are partly attributable to the hydrolysis of cyanate (CNO") and cyanide (CN") compounds. Cyanate is generated in the milling processes due to reaction of free cyanide and cupric ions. It may additionally stem from the treatment of cyanide-bearing wastes upon oxidation with hydrogen peroxide or hypochlorite (Higgs, 1993). Cyanate can hydrolyze in basic mill effluents to produce ammonia via: CNO" + OH" + H 2 O -> C O 3 2 - + N H 3 (1) At pH's < 9.3 (i.e., Balmer Lake), however, N H 4 + is the thermodynamically favoured product of cyanate hydrolysis (Stumm and Morgan, 1981). Cyanide will also hydrolyze in neutral waters to form ammonium via: CN" + 3H20 -> HCOO- + N H 4 + <2> 114 The large bottom water ammonium enrichments in the winter water column in Balmer Lake are believed to be derived from both sediment sources and from mining-related lateral advective flows. Ammonium is generated by heterotrophic bacteria as a primary end-product of organic matter decomposition, either directly from proteins or from other nitrogenous organic compounds (Van der Weijden, 1992). The remineralization of organic matter in Balmer Lake sediments, and the concomitant release of N H 4 + to the porewaters, results in steep gradients at shallow sediment depths. Ammonium released during aerobic respiration into oxic horizons will be oxidized by nitrifying bacteria (nitrification) to nitrate (Sholkovitz, 1973). Similarly, the same fate holds true for upward diffusing N H 4 that migrates into oxic sediments. However, the complete consumption of oxygen, and the reductive dissolution of Mn oxyhydroxide phases in the reducing winter bottom waters, effectively remove these potential oxidants from interfacial horizons. In this environment, remineralized N H 4 + is allowed to diffuse from permanently reducing horizons and accumulate in the anoxic bottom waters. Such effluxes have been described elsewhere: Stadelmann (1971) and Wetzel (1975), for example, report the presence of >500 umol'L"1 N H / in temperate, productive lakes in Michigan and Switzerland, respectively . This topic will be discussed in greater detail in section 5.3.1. The large N H 4 concentrations reported in the effluents discharged to Balmer Lake indicate that the winter bottom water enrichments can also be partially attributed to a lateral transport mechanism, as suggested for sulphate. Support for this argument can be seen in the fall bottom water and porewater N H 4 + distribution (Fig. 4.13D) in "Tailings Bay" (station 5). The ammonium profile is characterized by a well-defined bottom water enrichment, comparable to that observed for SO42". Like sulphate at this site, ammonium concentrations 115 decrease dramatically at shallow sediment depths demonstrating that the + sediments are not releasing N H 4 , but are serving as a diffusive sink. Ice formation in Balmer Lake may also indirectly contribute to the large winter N H 4 + inventories. Free cyanides exist in solution as cyanide (CN") and hydrocyanic acid (HCN) in the equilibrium (Stumm and Morgan, 1981): H C N <-> H+ + CN" (pKa = 9.3) (3) At the neutral p H of Balmer Lake, 95% of free cyanide exists as molecular H C N . As a result of its relatively high vapour pressure, H C N tends to volatilize to the atmosphere, even in stagnant solutions (Higgs, 1993); the presence of an ice cap during the winter months, however, prevents H C N release. Thus, the increased residence time of cyanide compounds during the winter months, and their + subsequent breakdown, may contribute to the increased winter N H 4 concentrations in Balmer. Lake. During the well-mixed periods of the spring, summer and fall, nitrate concentrations remain fairly constant (Tables 4.1, 4.2 and 4.4). Pronounced decreases beginning 10-20 cm below the top of the oxycline at all sites in the winter, however, suggest that denitrification is occurring in the bottom waters during this period below depths of 2.4-2.7 m (Fig. 4.5). Denitrification becomes important at low O2 concentrations (<6 umol/L) and involves the bacterially- mediated reduction of one or both ionic nitrogen oxides (nitrate or nitrite) to their gaseous oxides and eventually to N 2 (Froelich et al., 1979). Peeper bottom- water profiles at stations 1,2 and 4 suggest that nitrate consumption (i.e., defined where d2[N03"]/dz2 > 0) begins in winter at depths approximately 10,15 and 5 cm above the sediment-water interface, respectively (Figs. 4.15B, E and 4.16B); peeper data were not obtained from station 3. Similar profiles demonstrate that 116 denitrification does not occur in the shallow waters (~1 m) at station 6 in winter (Fig. 4.16E). Nitrate is produced in most lake systems primarily by the fixation of molecular N 2 by cyanobacteria and via the bacterial-mediated oxidation of ammonium diffusing upwards into oxygenated horizons in the sediments or water column (Wetzel, 1975). The abundance of oxidizable N H 4 + suggests that nitrification is responsible for a high proportion of the N 0 3 burden in Balmer Lake. In general, the ratio of N 0 3 -N to N H 4 + - N is a highly variable function of lake trophic status, hydrogeology, and natural and pollution-derived sources (Wetzel, 1975). Well-buffered, oligotrophic lakes receiving drainages from calcareous landforms, for example, can have N 0 3 -N: N H 4 + - N ratios of greater than 25:1. Conversely, in permanently anoxic, eutrophic lakes, ratios of 1:10 are common (Wetzel, 1975). The values of N 0 3 - N : N H 4 + - N in Balmer Lake progressively increase over the winter, spring, summer and fall, with ratios averaging <1, 1.8, 3.5 and 6.2, respectively. This hierarchy largely reflects a decrease in N H 4 + concentrations from the maximum observed during the winter period. The relative removal of N H 4 + can be attributed to biological nitrification, phytoplankton uptake and sorption to clay minerals. 5.1.4 Suspended Particulates The considerable intra-station variability of particulate trace metal concentrations in the well-mixed water column of Balmer Lake warrants consideration of potential sampling and analytical errors (Tables 4.6 and 4.7). Inaccuracies in the data set may have arisen from two sources. First, the over- pressure filtration units, on occasion, leaked. In such circumstances, the total volume of water filtered could not be accurately determined, and as a result, 117 errors would be manifested where the total mass of particulate metal per unit volume is computed. Such error should not be introduced, however, into the determinations of total mass of metal per gram of suspended material. Yet, the coarsely paralleled variability of mass: mass calculations with those of mass: volume, suggest that some degree of particulate heterogeneity is evident in the water column. Additional error may have been introduced into the mass:mass values since the absolute particulate fraction, determined by subtracting the pre- filtered from post-filtered weights, was very small. A third source of error, with respect to Zn concentrations, appears to have originated from contamination either during cleaning of the polycarbonate filters or during the filtration procedure. Values determined from filtration blanks ranged from 0.3-4.8 ppb (Appendix B). Such levels exceed some of the low-end water column particulate values. In order to obtain the most representative estimates, averages of the respective filter blanks for each sampling session were subtracted from the total. It should be noted that filters taken out of their wrappers and processed (i.e., not acid-cleaned) exhibited negligible levels of all metals (Appendix B). Zn enrichments most likely arose from water contact with zinc-rich rubber fittings in the filter apparatus, despite the O-rings and stoppers being wrapped in teflon tape prior to use. Concentrations of total suspended solids in the water column of Balmer Lake are high, and are fairly typical of levels encountered in productive, nearshore lake zones characterized by higher rates of sediment resuspension (Nriagu et al, 1981). The particulate trace metal levels greatly exceed values reported for other non-perturbed inland waters (Table 4.6). In general, values in Table 4.6 are high in comparison to suspended metal concentrations observed in lacustrine and estuarine waters receiving pollutant metal inputs (Kubota et al, 1974; Jackson, 1978; Sholkovitz, 1979; Nriagu et al, 1982; Jackson and Bistricki, 118 1995). Moreover, the lake-wide averages for the observed concentrations in suspended particulates of approximately 5 wt.% Fe, 850 mg'kg"1 Mn, 1500 mg'kg"1 Ni, 7000 mg'kg"1 Cu, 5000 mg'kg"1 Zn and 4000 mg'kg"1 As also surpass reported values of other polluted systems (see Nriagu et al, 1982). The relatively large magnitudes of the particulate values demonstrate that the settling of suspended material may present an important transport mechanism of metals to bottom sediments. Aquatic microorganisms exert controls on the geochemical cycles of many trace elements via active and indiscriminant associations (Morel and Hudson, 1985). The interactions of organic particles with metal ions and other reactive elements can occur via two mechanisms: metal ions can become coordinated to particulate surfaces of cells and/or fecal material in accordance with surface coordination or sorption equilibria, or alternatively, they can be incorporated into living cells and physiologically assimilated (Sigg, 1985). The high affinity of metal ions for biological surfaces suggests that biogenic particles may play a prominent role in the binding and transfer of metals to bottom sediments (Nriagu et al, 1981; Sigg, 1985; Morfett et al, 1988; Reynolds and Hamilton- Taylor, 1992). Particulate organic matter is abundant in the water column of Balmer Lake and constitutes a large percentage of the total suspended solids (Tables 4.5 and 4.7); concentrations are similar to levels measured in other Ontario lakes (e.g., Nriagu et al, 1981, Nriagu et al, 1982). Several studies have indeed shown that concentrations of particulate trace metals are strongly influenced by the production of POC (Sigg, 1985; Nriagu et al, 1981 and 1982). They attributed more efficient scavenging of trace elements to larger productivities and higher particle sedimentation rates. Except for Zn, however, particulate metal levels in Balmer Lake do not exhibit a clear correlative relationship with concentrations of POM. 119 Other than associations with organic matter, variable fractions of the particulate trace metal burden are likely to be associated with metal oxyhydroxide precipitates (e.g., Fe, Mn and Al) and clay minerals in the lake waters. The relative distribution of trace metals between these fractions is unknown. However, some inferences can be made based on seasonal comparisons of POM, TSS, metal per volume and metal per gram particulate. For example, the highest particulate metal concentrations evident in the spring water column coincided with the highest levels of total suspended solids (Tables 4.5 and 4.6). The relatively lower organic content of these particulates, however, implies that large fractions of the trace metals are associated with inorganic matter. Furthermore, the winter water column was characterized by less than half the average POM content of the summer, but contained comparable or greater concentrations of all trace elements (Tables 4.5 and 4.6). From the above discussion, it is evident that POM does not exert the principal control on particulate metal levels in the water column of Balmer Lake. However, given the abundance of particulate iron in lake waters, control by Fe- oxide phases may be important. Indeed, existing reports on trace metal scavenging by Fe and Mn oxyhydroxides in lakes polluted with mine and smelter wastes illustrate the importance of such metal-particulate associations (Jackson and Bistricki, 1995). Seasonal concentrations of particulate Fe species in Balmer Lake range from 107-649 |ig'L-l which correspond to 3 to 6 wt.% of suspended material (Tables 4.6 and 4.7). The average seasonal concentrations of particulate Ni, Cu and As track mean suspended iron concentrations; for all four elements (including Fe) the seasonal hierarchy follows the order: spring > winter > summer and fall (Table 4.6). Variations in iron-trace metal relationships may be influenced by the relative proportion of less reactive Fe species, such as those in the lattices of resuspended clay minerals. Inorganic particulate fractions may 120 also be present as CaC03, quartz, chlorite, illite and feldspars; these minerals are also capable of sorbing both organic substances and metal ions, but they have relatively small specific surface areas and thus relatively little capacity for metal ion binding (Sigg, 1985). A basic mass balance analysis conducted during a period of elevated metal concentrations in Balmer Lake in the spring of 1986, suggests that an increase in the discharges from the adjacent tailings ponds could not alone account for the total increased metal levels (Masala, 1995). The latter point, in conjunction with observations of greater particulate metal fractions during this period, leads to the conclusion that wind-suspension of fine-particulate material were contributing to such loadings. More specifically, it was proposed that in the quiet waters under the ice in winter, finer particulates settle out gradually and form a non-cohesive flocculant layer susceptible to resusupension upon ice break-up (PDI, 1986). This argument is consistent with the high TSS and particulate metal levels observed in Balmer Lake observed during the study period. 5.1.5 Trace Metals The dynamic nature of Balmer Lake is marked by considerable variability in the seasonal distributions of the measured trace elements. Two principal mechanisms are invoked to account for the observed variability: seasonal shifts in the redox geochemistry of the sediments and bottom waters, and the effect of mining-related discharges. It will be demonstrated below that an understanding of the timing and relative influence of such mechanisms is critical in the evaluation of the spatial variability observed in the winter water column. 121 To constrain fully the mechanisms controlling the behaviour of trace metals in the water column of Balmer Lake, both interfacial and porewater distributions must be examined. A detailed assessment of the seasonal porewater chemistry is presented in section 5.3, and an interpretation of trace metals in the water column is offered immediately below. The latter is divided into three sections: 1) descriptions of the respective trace element geochemistries; 2) presentation of a water column model; and 3) application of the model to four zones in Balmer Lake. 5.1.5.1 Iron The transport of Fe and Mn in water bodies and sediments has received a great deal of attention because of the central role that these abundant metals play in the geochemical cycling of other elements. The surface properties of Fe oxyhydroxides and associated trace metal sorption mechanisms, in particular, have been well studied (Davis and Leckie, 1978a; Davis and Leckie, 1978b; Vuceta and Morgan, 1978; Benjamin and Leckie, 1982; Tessier et al, 1996). The redox geochemistry of Fe has been repeatedly stressed as a principal mechanism governing the distribution of many trace elements in lacustrine systems (Laxen, 1985; Tessier et al, 1985; Belzile and Tessier, 1990; Davison, 1993; Hamilton- Taylor and Davison, 1994; Jackson and Bistricki, 1995). Iron (II) and iron (III), represent the two oxidation states of importance to the aquatic geochemistry of Fe. Ferrous iron is stable in anoxic water, exists predominantly as a simple hydrated aquo ion, and is soluble with respect to most inorganic ions; however, in some environments, the solubility products of siderite (FeC.03), amorphous iron sulphide (FeS) and vivianite [Fe3(PC>4) 2] may be exceeded (Davison, 1993). In well oxygenated waters, Fe(III) is the stable 122 oxidation state. This acidic cation occurs as completely hydrolyzed oxides at neutral pH, with its solubility being controlled by amorphous iron hydroxide (Sturnm and Morgan, 1981). Numerous field and experimental works have been carried out to characterize iron and manganese oxide phases in lacustrine waters and sediments (Davis and Leckie, 1978b; Laxen and Chandler, 1983; Davison et al., 1992). For the most part, the nature of iron oxide particles in natural waters is not completely understood. Although variations in particle size, composition and structure have been observed, general similarities exist. In oxygenated lake waters, ferric iron has been predominantly found as amorphous particles and colloids, which are negatively charged due to adsorption of humic substances (Tipping, 1981). The spherical or ellipsoidal particles are presumed to be poorly crystalline ferrihydrite with mean diameters typically in the range of 0.05-0.5 urn. The general presence of these particles as amorphous phases makes them readily susceptible to dissolution under anoxic conditions. Tipping et al. (1981) used electron microscopy to examine iron-rich particles in Esthwaite Water surface sediments. The particles contained 30-40 % Fe (by weight), with P, N , Mn, Si, S, Ca and Mg accounting for up to 8 wt. %. Organic matter constituted up to 36 wt. % with one third of this being represented by humic substances. The data in Table 4.6 clearly illustrate that particulate Fe fractions dominate the total iron inventory in the waters of Balmer Lake. The ratio of particulate to filterable species exhibits considerable seasonal consistency with particulates comprising 60-89 % of the total Fe inventory. The Fe-particulates are believed to comprise variable mixtures of iron oxides and clay mineral fractions derived from resuspended sources. It should be noted that dissolved Fe species become more important during the winter months. 123 5.1.5.2 Manganese As for iron, the biogeochemical cycling of Mn has been shown to govern the behaviour and mobility of redox sensitive elements in freshwater environments (Davison and Woof, 1984; Agett and O'Brien, 1985; Sigg et al, 1991; Young and Harvey, 1992; Davison, 1993). Of the three naturally occurring oxidation states, Mn(II) is stable in anoxic waters, where the predominant species are simple hydrated aquo ions. Manganese (II) is very soluble with respect to most inorganic ions, although in certain environments, such as interstitial waters, it is possible to exceed the solubility product of rhodochrosite (MnCOa) (Carignan and Nriagu, 1985). Conversely, in well-oxygenated waters, Mn02 (Mn IV) is the thermodynamically stable form. In practice, a range of metastable oxidation products exist in natural waters due to kinetic limitations (Stumm and Morgan, 1981); Mn(III) is only stable in a lattice of mixed oxidation state (Davison et al, 1993). Numerous works have attempted to describe manganese oxidation products and characterize them using electron microscopy and X-ray diffraction (De Vitre and Davison, 1993). However, the basic chemistry of both the Mn(IV) and Mn(III)-bearing phases are not well understood. Natural manganese particles are difficult to identify because of their extreme microcrystallinity which renders them X-ray amorphous. In addition, due to their tendency to coexist and coat many other minerals and organic material, sometimes only electron microscopy can provide identification (Chiswell and Mokhtar, 1986). The most common forms of Mn oxides identifiable in oxic lake sediments are y-MnOOH and birnessite (Mn02). The structure of the latter can be complex. Birnessite is thought to contain Mn(III) and possibly Mn(II) or other divalent metal cations (e.g., Pb, Cu, Zn) in addition to predominant tetravalent Mn species. The humic 124 content of these phases is low (~1 %), but bacteria and/ or algae may contribute 20-30 % by weight (Davison,1993). In well-oxygenated, neutral waters (pH 6-8), the half-life of Mn(II) is typically on the order of days to months; by contrast, Fe(JJ) has a typical half-life of ~4 h under the same conditions (Stumm and Morgan, 1981). The variable reaction rates of Mn have been attributed to bacterial-mediated oxidation kinetics; various strands of evidence have been introduced to support the importance of microbial interactions (Davison,1993). The differing oxidation kinetics have corresponding consequences with respect to the distribution of soluble and particulate species. Examination of the partitioning of Mn in the water column of Balmer Lake indicates that particulate phases are insignificant relative to the dissolved fraction (Tables 4.6 and 4.7). During all sampling periods, dissolved manganese counted for at least 95 % of the total inventory. These results agree well with a recent study of Mn speciation in a Precambrian Shield lake, where predominantly non-particulate forms were also observed (LaZerte and Burling, 1990). As for Fe, the term "dissolved" must be used with caution, as colloidal phases have been shown to contribute significantly to filterable fractions (De Vitre et al, 1988; LaZerte and Burling, 1990). In natural systems, the proportion of Mn in particulate form may vary considerably. For example, Laxen et al (1984) showed that in various British rivers and streams the particulate fraction varied from 0 to 100%. The wide range has been attributed to the relative contributions of Mn from two distinct sources: 1) weathering of Mn-bearing minerals which yields suspended particles; and 2) inputs of soluble manganous species from oxygen-deficient sediments and ground waters (Davison, 1993). The relative influences of these sources, and therefore the particulate proportion in a particular lake or stream, depends on the 125 local hydrogeology. It will be shown below that the dominance of filterable species in the Balmer Lake water column can be attributed to the direct input of dissolved manganese. 5.1.5.3 Arsenic The inherent toxicity of many As compounds and their transformations in biogeochemical cycles has stimulated considerable interest in As speciation in the environment (Ferguson and Gavis, 1972; Sanders, 1983; Reimer and Thompson, 1988; Cullen and Reimer, 1989; Anderson and Bruland, 1991; Bowell, 1994). Although total dissolved As was measured in Balmer Lake, predictions of speciation can be made using relevant thermodynamic constraints and results of previous studies. Dissolved As can occur in natural waters in both inorganic and organic forms. In oxygenated waters at neutral pH, the anionic arsenate species (H2ASO4- at 3.6<pH<7.3) is the thermodynamically stable form of As. Conversely, neutrally charged arsenite (As(OH) 3 °) is the predicted form in reducing environments (Cullen and Reimer, 1989). The large predicted As(V)/As(III) ratio of 10 1 5 -10 2 6 suggests that As(III) should contribute a negligible fraction of total inventories in neutral waters. However, the observation of significant amounts As(III) in oxygenated waters, and the presence of As(V) in reducing regimes, suggests that the two are rarely in thermodynamic equilibrium. Arsenic is also host to a suite of organic compounds, of which monomethylarsonic acid (MMAA) and dimethylarsinic acid (DMAA) contribute significantly (-10%) to total As concentrations in freshwaters (Andreae, 1979; Anderson and Bruland, 1991). Because arsenate and phosphate are chemical analogues there is little discrimination between the two species during biological 126 uptake. This has led to the suggestion that organisms have developed mechanisms to isolate and detoxify As by producing organoarsenicals (Wood, 1974). However, the processes that promote biological methylation of As are still poorly understood. Factors including high A s i P O ^ - ratios in ambient waters, and the nature of biotal communities, have been invoked to explain greater abundances of methylated compounds (Anderson and Bruland, 1991). Arsenic concentrations in the water column of Balmer Lake greatly exceed natural baseline values observed in non-perturbed Canadian Shield lakes. The abundance of arsenopyrite in the milled ore suggests that the oxidation of this phase during the milling process is responsible for the large inventory. Arsenic is largely dominated by filterable fractions throughout the sampled seasons, contributing over 95% of the total burden (Table 4.6). Large bottom-water particulate spikes observed during the winter season (27-32% of the total inventory) represent the only deviations from the typically low suspended particulate As values. Considering that Balmer Lake is productive and hosts excess As with respect to phosphate, there is a high probability that concentrations of methylated species are substantial. In all likelihood, however, these species probably represent an insignificant fraction of the total arsenic in the lake waters. Arsenic species in the water column of Balmer Lake are most likely dominated by arsenate and arsenite, with their respective contributions fluctuating in accordance with seasonal oxic/anoxic cycles. 127 5.1.5.4 Nickel Nickel exhibits a nutrient-type distribution in the oceans and is characterized by both shallow and a deep water regeneration. Vertical profiles in the top 800 m appear similar to phosphate, while below this depth nickel exhibits a deep maximum similar to silicate (Bruland, 1983). Concentrations of dissolved nickel increase from ~2 nmol'L"1 in surface central gyre waters of the North Pacific to a deep maximum of 11 nmolL-1 (Sclater et al, 1976; Bruland, 1980). Similarly shaped profiles determined for the North Atlantic exhibit maxima one half the magnitude of those observed in the deep North Pacific (Bruland and Franks, 1983). The non-steady-state nature of lacustrine systems combined with the relative dearth of appropriate data make generalizations of nickel distributions in lakes difficult. The micronutrient-like behaviour observed for Ni in marine systems, for example, has not been clearly demonstrated for freshwaters. Available evidence suggests that vertical variability is largely dictated by the redox transformations of Fe, Mn and S. Concentrations of dissolved nickel in the hypolimnion of Lake Sammamish, for example, were shown to increase in response to bottom water anoxia (Balistrieri and Murray, 1992); the supply of dissolved Ni to the bottom waters was attributed to the dissolution of nickel- bearing Mn-oxide phases. Indeed, Fe and Mn oxides have been repeatedly inferred to control the distribution of Ni in lake waters, sediments and porewaters (Green et al., 1989; Tessier et al, 1996). Conversely, the removal of nickel from reducing bottom waters (Green et al, 1989) and porewaters (Carignan and Nriagu, 1985) has been largely attributed to the precipitation of authigenic Ni-sulphides. 128 Nickel concentrations in Balmer Lake greatly exceed those in non- perturbed Canadian Shield lakes. Loadings from the tailings circuit, which average over 1 mg-L"l, contribute to the high values. The nickel inventory in the water column is dominated by dissolved species throughout the sampled seasons, which is consistent with measurements of nickel partitioning in other lake systems (Nriagu et al, 1981; Balistrieri and Murray, 1992). Speciation calculations of freshwaters suggest that nickel exists largely as the free aquo ion N i 2 + and as carbonato complexes during well-mixed periods in Balmer Lake (Florence, 1982; Carignan and Nriagu, 1985). 5.1.5.5 Copper In the oceans, copper exhibits a distribution intermediate between that of nutrient-type elements and that of scavenged elements such as Pb. The vertical distribution of Cu in the Pacific, for example, is characterized by an approximately linear increase with depth (Bruland, 1980). The deviation from a nutrient-type element (e.g., cadmium) is due to in situ scavenging by particulates in the intermediate and deep waters. The remineralization of Cu-bearing phases at the sediment-water interface, however, results in a major portion of the total benthic Cu-flux being recycled back into the bottom waters (Bruland, 1980). Deep water concentrations in the Atlantic (Bruland and Franks, 1983) are one- half to two-thirds of deep Pacific waters due to the progressive addition of dissolved Cu during deep water migration. The compressed depth scales and dynamic nature of lacustrine systems prevent generalizations of the vertical distributions of copper in lakes. However, the available data suggest that redox transformations and particle scavenging represent two important controls. The dissolution of sinking Mn and Fe oxides, 129 for example, can yield peaks of dissolved Cu at oxic/anoxic boundaries (Green et al, 1989). Conversely, hypolimnetic depletions can result from the precipitation of copper-bearing sulphide phases in reducing bottom waters (Green et al., 1989; Balistrieri et al., 1992) and porewaters (e.g., Carignan and Nriagu, 1985). Direct evidence linking Cu distributions with biological cycles is rare, although algal decomposition has been invoked to account for Cu enrichments at the sediment- water interface in a British lake (Morfett et al., 1988). Filterable copper species dominate the Balmer Lake water column throughout the sampled seasons, comprising between 65 and 90% of the total pool. Speciation calculations suggest copper exists predominantly as C u O H + and CuC03 in aerobic fresh waters (Stumm and Morgan, 1981), with a significant portion of the dissolved fraction potentially bound to organic ligands. Inorganic and organic colloids have also been shown to contribute significantly to total copper burdens in freshwaters (Dai et.al, 1995). 5.1.5.6 Zinc Zinc is a group lib element which has well-established associations with biological cycles in both marine and lacustrine environments. In the oceans, zinc exhibits a nutrient-type distribution similar to that of silicic acid; dissolved concentrations in the central North Pacific increase from -0.2 nmol-L"1 in the surface mixed layer to -8 nmol-L - 1 in the intermediate and deep waters (Bruland et al, 1994). Although biologically-mediated Zn-nutrient associations are often obscured by other processes in lakes, two convincing data sets clearly illustrate a biological control. In productive Lake Windemere, for example, dissolved zinc, phosphate and silicate were positively correlated throughout the water column during a rapid growth phase of a spring diatom bloom (Reynolds and Hamilton- 130 Taylor, 1992). Similarly, acid-soluble zinc concentrations were shown to correlate with silicate distributions during summer stratification in Lake Sammamish, Washington (Balistrieri et al , 1992). Although there is little doubt that Fe and Mn oxides also act as carriers for Zn in lacustrine systems, observations indicate that redox related phenomena have a relatively small effect on the vertical distribution of dissolved Zn in lakes. Conversely, numerous examinations of the early diagenesis of zinc in oxic lake sediments have demonstrated the importance of Fe and Mn oxide phases with respect to zinc mobility (Carignan and Tessier, 1985b; Tessier et al, 1989; Young and Harvey, 1992; Tessier et al, 1996). In reducing sediments, dissolved zinc is commonly removed from pore-solution via the precipitation of sphalerite (ZnS) and/or mixed Fe-sulphide phases (Carignan and Tessier, 1985b; Pedersen et al, 1993). The zinc concentrations in the water column of Balmer Lake greatly exceed those in unpolluted freshwater systems. Although dissolved fractions dominate the zinc burden, a considerable portion is present in suspended particulates. The filterable zinc species are predicted to exist predominantly as Z n 2 + and ZnC03, with inorganic and organic colloids potentially hosting a significant portion (Florence, 1982). 5.1.6 Water Column Model The following section presents a model to explain the water column profiles of trace elements over the fall-winter-spring transition. The scheme, presented in Fig. 5.1, represents a time series, and encompasses the broad range of profiles observed in Balmer Lake. 131 A B C Fig. 5.1. Time series of progressive trace metal water column profiles (A-E) over the fall- winter transition in Balmer Lake: A) well-mixed water column, B) influence of remobilized bottom-water source, C) influence of interfacial sulphide sink, D) greater extent of sulphide removal, and E) influence of metal-rich lateral advective flow. Depth arid concentration axes are in arbitrary units. 132 As described in section 1.2.3, the degradation of organic matter is mostly an oxidative process, for which electron acceptors must be present. Remineralization follows a sequence of oxidation reactions in which the redox couple that liberates the highest free energy yield acts before the next most efficient couple. Bacterial assemblages, specialized in using the energy released in each of these redox reactions, obey this sequence (Froelich et al, 1979). The thermodynamic sequence of oxidant depletion follows the order: O2, NO3" and Mn(m,IV)-oxides, Fe (Ill)-oxides, SO42" and finally C 0 2 . Deviations of this sequence are possible, however, and relate to such factors as porewater p H and the stability of the iron oxide (Postma and Jakobsen, 1996). During the well-mixed, oxygenated periods of the spring, summer and fall in Balmer Lake, the anoxic-oxic redox boundary (as defined by the depth where SO42" reduction commences) lies below the sediment-water interface. The formation of ice in early winter, however, creates the conditions necessary to promote the gradual consumption of the oxidant pool, and the evolution of bottom water anoxia. The SO4 2" redox boundary is predicted to migrate upwards towards the sediment surface over the duration of ice cover, and in some areas, cross the interface into the bottom waters. Examination of the water column and interfacial metal distributions measured in late winter suggests that extensive sulphide precipitation had already occurred earlier that season in the deeper lake areas. Authigenic sulphide formation has been suggested to present a significant control on trace element distributions during anoxic periods in other dimictic lakes (Morfett et al, 1988). Discharges to Balmer Lake from the adjacent tailings circuits present another important consideration. Measurements made on tailings pond effluents suggest that significant burdens of trace metals, sulphate, cyanide and ammonia are associated with these discharges. The influence of mining-related effluents in 133 the lake appears to be related to water depth, and to the proximity to the point of discharge. Due to its large signal in mine discharge waters, sulphate is considered to be the most appropriate tracer of these inputs. During well-mixed periods, homogenous conditions resulting from extensive wind-mixing of a shallow water column produce uniform depth profiles of dissolved constituents as was shown in Chapter 4, which are summarized schematically in Fig. 5.1 A. The formation of ice in early winter removes any atmospheric influences and permits the development of water column stratification. As anoxia develops over the course of ice-cover, the predicted reducing swing in the redox environment at the sediment-water interface promotes the reductive dissolution of labile Mn-bearing solid phases. Significant input to bottom waters subsequently yields the generalized profile shown in Fig. 5.IB. After the consumption of NO3" and Mn-oxides, the oxidation of organic matter proceeds via Fe(III) reduction. The most reactive phases are those with the highest solubility, i.e., amorphous and poorly crystalline Fe-oxyhydroxides (Van der Weijden, 1992). Dissolved Fe distributions during such periods resemble the profile shown in Fig. 5.1B. The remobilization of Fe and Mn oxyhydroxide phases is often associated with the concomitant release of sorbed trace elements which may yield similar enrichment patterns of such elements in bottom waters (e.g., Green et al, 1989; Aggett and Kriegman, 1988). As conditions become more reducing at the sediment-water interface, and reactive Fe(III)- and Mn(IV)-bearing solid phases have become exhausted, sulphate reduction is expected to become important (Froelich et al, 1979). The introduction of an interfacial metal-sulphide sink to the situation shown in Fig. 5.1B, may yield a profile similar to that shown in Fig 5.1C. More extensive sulphide precipitation, in conjunction with diffusional processes, may remove 134 any evidence of prior metal release (Fig. 5.1D). The final profile in the model sequence results if a dense, metal-rich lateral flow displaces or mixes into metal depleted bottom waters (Fig. 5.1E). This five-step order of events is used below to assess the intra-lake variability during the winter sampling period. 5.1.7 Comparison of Four Lake Zones To reiterate, the trace metal distributions observed during the spring, summer and fall sampling periods exhibited little variability with depth; windy conditions prevented the development of any thermal stratification and, in effect, inhibited the detection of bottom water gradients associated with sediment consumption or release. The relatively stagnant conditions during winter sampling, however, permitted the development of chemical stratification. The fact that Balmer Lake has a large surface area to volume ratio enables benthic exchanges to be more easily manifested in the water column structure. The •following discussion attempts to resolve the winter variability of trace metal distributions with respect to four lake areas (stations 1-4, Fig. 2.1). Evidence for interfacial sulphide precipitation throughout Balmer Lake implies that the reductive dissolution of labile sedimentary Mn- and Fe-oxides, and associated metal releases, had occurred some time prior to sampling. It should be noted that at any time during the development of reducing conditions, the redox intensity at the sediment-water interface may exhibit significant lake- wide variation; differential rates can be attributed to such major variables as water depth and the concentration of organic matter. This point will become relevant in the following interpretations of water column profiles. Steep porewater manganese gradients evident during the other seasons imply that the upward diffusion of remobilized Mn(II) species across the 135 interface is at least partially responsible for enrichments in the reducing winter bottom waters. The fact that stations 3 and 4 exhibit minimal or negligible signatures from pollutant effluents (i.e., insignificant SO4 " signal) and are still characterized by bottom water Mn maxima, supports this premise (Figs. 4.7,4.10 and 4.11). The decoupling of the soluble and particulate Mn fractions seen in Balmer Lake results from the relatively slow oxidation kinetics of its reduced form; this results in greater stability and thus longer residence time of reduced Mn species in oxygenated waters. However, in aquatic systems where Mn oxidation is. substantially accelerated by bacterial processes, or where residence times are long (i.e., some lakes), the interconversion of the two forms may be an important process. Indeed, the pronounced decrease in the dissolved Mn content, which occurred over the four week sampling hiatus between the winter and spring sessions, implies a removal mechanism from the water column. The oxidation of M n 2 + and subsequent precipitation as M n 0 2 provides a probable mechanism. Bottom water concentrations of Mn exceed 1 mg/L in some zones of Balmer Lake. The release is predicted to be predominantly present as reduced Mn(II) hydrated aquo ions, since speciation studies suggest that inorganic and organic complexes represent insignificant fractions (Chiswell and Mokhtar, 1986; LaZerte and Burling, 1990). Manganese enrichments in the anoxic hypolimnia of lakes in excess of 1 mg'L"1 have been reported by many authors (Davison et al, 1982; Sigg, 1985; Cornwell, 1986; Agett and Kriegman, 1988; Morfett et al, 1988; Anderson and Bruland, 1991). The seasonal anoxia in these studies, however, occurs in the summer, and results from the development of thermal stratification in combination with enhanced organic transport to the hypolimnion. Sulphate reduction in the bottom waters also implies that the reductive dissolution of reactive Fe phases had occurred prior to winter sampling. 136 Interfacial enrichments of dissolved Fe derived from remobilized sediment sources have been reported in several studies of seasonally anoxic lakes (e.g., Davison et al, 1982; Cornwell, 1986; Anderson and Bruland, 1991; Sigg, 1985; Agett and Kriegman, 1988). The extremely high metal levels in Balmer Lake, and thus their increased ability to compete for surface sorption sites, may suggest that Fe and Mn amorphous phases in the top few centimetres host significant burdens of exchangeable metals. Therefore, the reductive dissolution of Fe and Mn oxides may promote a pronounced release of sorbed constituents (Belzile and Morris, 1995). Moreover, trace metals associated with conspicuous porewater maxima (i.e., Fe, Mn, As, Zn and to a lesser extent Ni) could diffuse into the bottom waters in the absence of an interfacial oxide trap. In addition to Mn(II), Fe(II), and associated trace elements remobilized from reducing sediments, the reductive dissolution of sinking oxide particles in reducing horizons has also been invoked to support enrichments of metals in the bottom waters of stratified lakes (Green et al, 1989; Anderson and Bruland, 1991; Balistrieri and Murray, 1992). This pattern of supply to the surface waters and remobilization at depth, may lead to maxima in the concentration of soluble constituents towards the end of seasonal stratification (Davison, 1993). The abundance of metal particulates in the winter water column of Balmer Lake, suggests that trace metals remobilized from sinking particles may present an important input to the bottom waters. Because the lower stratified zone in Balmer Lake is < lm thick the residence times of particulates in this zone are predicted to be brief, suggesting that remobilization occurs predominantly at the sediment-water interface. In general, considerable difficulty arises when trying to distinguish between diffusion across the sediment water interface and dissolution of settling 137 oxide particles as the source of increased concentrations. In an earlier review, Sholkovitz (1985) demonstrated that enhanced redox-related effluxes were accountable for increased bottom water concentrations of Na, K, Mg, Ca, Ba, Fe, Mn, P, Si and NH3 in some seasonally anoxic lakes. However, evidence showing similar effects with other elements is scarce. Visible dissolved trace metal gradients associated with bottom water sources have also been attributed to the remineralization of organic matter (Morfett et al., 1988; Hamilton-Taylor and Willis, 1990; Reynolds and Hamilton- Taylor, 1992). In the oceans, many trace elements (e.g., Cu, Zn, Ni , Cd ) display vertical concentration distributions strongly correlated with those of the major nutrients; they are depleted in the eutrophic zone by incorporation onto or into planktonic material, carried out of the surface waters by settling, and partly regenerated at depth through decomposition of the organic particles (Bruland, 1980). However, distinct correlations evident between concentrations of P and Si and several trace elements that have been observed in the oceans are rare for lake systems (Balistrieri, 1992). The absence of such correlations may reflect the contrasting environments of marine and freshwater systems, but may also reflect a lack of appropriate lacustrine data (Reynolds and Hamilton-Taylor, 1992). Metal inputs to terrestrial receiving waters are typically greater than loadings to seawater, and given the relative shallowness of lakes, little vertical segregation may develop between the biological scavenging of metals in near surface waters and their partial release at depth during respiration (Sigg 1985). Sigg (1985) and Morel and Hudson (1985) have suggested that algae incorporate trace elements stoichiometrically, resulting in the approximate formula C106N16P1 (Fe, Zn, Mn)o.oi (Cu, Cd, Ni , Co, etc.)o.ooi- Although this concept of Redfield ratios may be applicable to the metal content of algae, the dynamic nature of lakes limits its usefulness because only under severely 138 restricted conditions will the stoichiometry be reflected in dissolved metal concentrations (Reynolds and Hamilton-Taylor, 1992). Furthermore, the scheme proposed here remains hypothetical, as some elements that have shown correlations with nutrients have no clear biological role. The mechanisms described above (i.e., trace element remobilization from sedimentary and settling particulate oxide phases, and organic matter remineralization) are proposed to account for the Ni, Cu, As and Zn enrichments observed in the winter bottom waters at station 3 (Fig. 4.10). The relative partitioning of these elements will be discussed in the porewater sections below. Station 3 has the shallowest water column (2.8 m deep), and appears to lie above the depth of influence of dense mining-related effluents (Fig. 4.7C). Several of the dissolved trace metal profiles at this site show evidence of a bottom water source and generally resemble Fig. 5.1B. The slight contrasts among bottom water gradients of dissolved Mn, Ni, As and Zn, may reflect kinetic variability with respect to the oxidation and precipitation of upward diffusing species (Fig. 4.10). Although corresponding peeper data were not collected from this site, sulphide precipitation appears to be a less significant control on trace metal distributions. Indeed, the less pronounced nitrate removal with depth suggests that reducing conditions are in a less advanced stage of development as compared to deeper lake areas (Fig. 4.5C). The removal of trace metals as authigenic sulphides in anoxic lake systems has been reported for several elements, including Fe, As, Cd, Cu, Ni , Pb and Zn (Davison et al, 1982; Morfett et al, 1988; Green et al, 1989; Davison et al, 1992; Balistrieri, et al, 1992). Evidence for metal precipitation stems largely from calculated ion activity products using measured values of dissolved constituents in the water column and porewaters. Although sulphide was neither measured nor observed (by smell) in Balmer Lake, the abundant reducible sulphate in the 139 anoxic winter bottom waters, and the observed trace metal distributions, suggest that sulphide removal presents a principal control on trace element distributions in the deeper lake sites. In a more reducing area of Balmer Lake (station 4), sulphide precipitation mechanisms become more relevant. This site (3 m deep) is slightly deeper than station 3, and appears to be only marginally affected by mining-related advective flows (Fig. 4.7D). The reduced influence of such discharges is thought be a result of the shallow depth and its relatively large distance from effluent inputs (Fig. 2.1). Profiles of dissolved Fe, As, Ni and Cu at this site exhibit pronounced decreases below deep water column maxima (Fig. 4.11) and, with two exceptions (Mn and Zn), generally resemble Fig. 5.1C. The removal zones for Ni and Cu appear to be up to 5-10 cm above the sediment surface (Figs. 4.25D and E). The insignificant decreases in dissolved Fe over these horizons in comparison to other trace elements suggest that co-precipitation with FeS is not significant; rather, precipitation as their respective sulphides (i.e., AS2S3, NiS and CuS,) presents a more likely mechanism. Prior to the onset of sulphate reduction, the near-bottom distributions of dissolved Fe, As, Ni and Cu may have been similar at this site. However, it is suggested that sulphide removal has since consumed part of the excess of remobilized dissolved trace metals to produce the observed profiles. The fact that diffusional processes have not eroded the prominent dissolved maxima indicates that a steady input of dissolved species (presumably remobilized from sinking particulates) is being supplied to these horizons. Similar water column profiles of various dissolved constituents in permanently stratified Lake Vanda, Antarctica (60 m deep), were attributable to such processes (Green et al., 1989). However, due to the extreme compression of depth scales in Balmer Lake, and the relatively shallow depth of the bottom stratified 140 layer, particulate metal distributions analogous to those in Lake Vanda are not observed. Throughout the lake during the winter period, dissolved Mn concentrations exhibit no evidence for sulphide removal, as expected given the high solubility of MnS (Emerson, 1976). Indeed, the Mn distributions suggest a continual supply of dissolved species, which are presumably remobilized from sediments and settling particulates. Although dissolved Zn profiles exhibit pronounced lake-wide removal at shallow sediment depths (Figs. 4.23-4.25), authigenic sulphide precipitation above the sediment-water interface does not appear to be relevant. Strong associations with organic matter have been suggested previously to explain the decoupling of Zn from the redox cycling of Mn and Fe in a seasonally anoxic lake (Morfett et al, 1988). The apparently strong associations of POM with Zn in Balmer Lake (section 5.1.4; Tables 4.5 and 4.7) may suggest that organic matter remineralization is contributing to the persistent bottom water enrichments. The reductive dissolution of zinc-bearing Mn oxides may also serve as an input mechanism of dissolved Zn (Sigg et al, 1987). Examinations of the deeper sites (stations 1 and 2) illustrate the greater extent of sulphide reduction in the bottom waters as well as the local importance of deep lateral advection. At station 1, dissolved Fe, As and Cu gradients begin at the 2.5 m depth horizon (Fig. 4.8). The general similarity of this trio of profiles suggests similar controls on the distributions of the three elements in the near- bottom waters. Peeper profiles imply that Cu removal begins 10-15 cm above the sediment-water interface (Fig. 4.23E). While Cu and Fe concentrations continue to decrease to the sediment surface (Fig. 4.8, summarized in Fig. 5.1 D), As shows an opposite distribution below 2.8 m (Fig. 4.23C). Such an increase, as well as those for Fe, As, Ni at station 2, can be perhaps attributed to the influence of 141 relatively metal-rich advective flows (Figs. 4.23 and 4.24; see Fig. 5.IE). At both stations 1 and 2, bottom water SO42" increases to values over 12 mmol'L"1 below depths of 2.7 m (Fig. 4.7). Further evidence for lateral flows lies in the observation that sulphide precipitation alone could not account for the water column minima; i.e., sulphide removal acting near the sediment-water interface could not produce a minimum 40-70 cm above the sediment surface. Furthermore, if the bottom sediments were providing a metal source, and the water column minima were the result of trace metal scavenging at an oxic front, one should see evidence of metal release from the porewaters. However, near- interface trace metal profiles for Fe, As and Ni at station 2, and As at station 1, offer little support for diffusive effluxes fostered by sub-interface release (Fig. 4.24 and 4.23). In addition, profiles of the respective dissolved and particulate metal fractions do not support the removal of upward diffusing species by incorporation into Fe- or Mn-oxide precipitates (Tables 4.6 and 4.7). 5.2 Sediment Geochemistry The sediments in Balmer Lake represent complex mixtures of lithogenic, biogenic and authigenic phases which vary in time and space. For example, the composition of settling particulate matter varies over the course of a year due to seasonal fluctuations in primary productivity, ice coverage and terrestrial runoff. In addition, significant heterogeneity has been introduced via episodic events of lake-wide tailings deposition. Organic matter in the deposits stems both from in situ lake production of phytoplankton and macrophytes, and from external inputs via streams and wind-borne transport which vary with the extent of run- off. Prior to the onset of mining activity, allochthonous inputs from Balmer Creek (Fig. 2.1) represented the only significant lithogenous contributions to the 142 lake. However, since the commencement of mining, tailings have become the major detrital input. Profiles of several compositional parameters illustrate that below approximately 10 cm, the sediments at stations 1 (cores la and lb) and station 6 (core 2) represent two discrete assemblages and can be clearly seen in the weight ratios of Mg/AI (Fig. 4.32C), Na/AI (Fig. 4.32E), R b / A l (4.33B) and Z r / A l (4.33C). Those deposits in the lower horizons at station 1 consist of pre-industrial natural deposits, while sediments in core 2 are predominantly mining-derived tailings material throughout. Above the 10 cm transition horizon, the deposits at both locations appear to have received variable contributions from both anthropogenic and natural sources. The organic carbon content in Balmer Lake sediments of 8-9 wt.% (Fig. 4.34A), agrees with previously measured values and fall within the ranges observed in other productive Canadian Shield lakes (Conroy and Keller, 1976; Semkin and Kramer, 1976; Carignan and Tessier, 1988). Biological activity in Precambrian Shield Lakes has been shown to be influenced by morphology (surface area/volume), lithology (surficial and bedrock geology) and atmospheric loadings (Conroy and Keller, 1976). Water bodies such as Balmer Lake, with high surface area to volume ratios and situated in calcareous terrain, tend to exhibit higher productivity. Lower production characterizes acid lakes (pH<5.5) that have been impacted by acid rain, are poorly buffered, and have high volume: surface area ratios. Total nitrogen is enriched in the surface sediments of both stations 1 and 6 (Fig. 4.34C) and is illustrated in the depth distributions of the respective C 0 rg:N wt. ratios (Fig. 4.34D). Lower Corg/N values in the upper horizons are most likely the result of increased inputs of inorganic-N compounds and not the result of variations in the C:N ratio of introduced organic matter. Increased inputs of 143 terrigenous organics, which are commonly associated with lake-side industry (e.g., forest clearing, etc.), would have resulted in a higher C:N signal. The observed distribution of N can be explained by a combination of two relatively recent inputs. First, the hydrolysis of cyanide introduced by the adjacent ore processing operations has increased the abundance of nitrate and ammonium in the lake. Second, accidental dam breaches have supplied significant loadings of fine-grained, chlorite-rich material to the inherently clay-deficient natural deposits (PDI, personal communication). The exchange capacity of chlorite for cations such as NH4+, C s + , R b + and K + is comparable to that of illite (Grim, 1953). Therefore, the mineralogical shift induced by the addition of chlorite could have contributed to the relative abundance of particulate nitrogen in the upper decimetre of the Balmer Lake deposits. The presence of clay-rich deposits in the upper of horizons was established by semi-quantitative X-ray diffraction and examinations of element/Al wt. ratios. XRD peak-height ratios suggest that the quartz content in the upper stratum at station 1 is higher than at depths below 10 cm at this site. However, an increased quartz content cannot represent the sole mineralogic contrast in the upper decimetre because Si /Al wt. ratios decrease in this stratum (Fig. 4.32A). The recent addition of a relatively Al-rich mineral phase to station 1 presents the only plausible explanation that can account for the lower Si:Al ratios. XRD analyses indeed confirm the presence of abundant clay minerals, notably chlorite, in the upper horizons. The data collectively imply that the deposition of detrital chlorite (Si:Al = 1.3) has overwhelmed the input of quartz. The high Mg content in chlorite and the higher Mg:Al wt. ratios in the upper stratum at station l(Fig. 4.32C) provide further support for a chlorite input. In contrast to station 1, Si:Al ratios in the upper horizons at station 6 imply a 144 relatively higher quartz content. This is supported by the XRD data: quartz/clay peak height ratios in deposits at station 6 are higher than those at station 1. The locations of historic tailings discharges with respect to the coring locations could be expected to have exerted significant control on the composition of mining-derived material that reached the respective locales. Station 6 is situated tens of metres.from the major points of tailings discharge while station 1 sits over 1 km away. Deposits immediately proximal to such outputs would be expected to be relatively coarser-grained than those more distal. Such textural influences could contribute to the mineralogical differences in the tailings-containing deposits in the two lake areas. Below depths of 10 cm at station 1, sediment concentrations of all trace metals are low, and are comparable to values typical of Canadian Shield lakes (Semkin and Kramer, 1976). In contrast, the lowest concentrations of Ni, Zn, Cu, As and Pb observed in the tailings deposits are still 4, 2, 2, 20 and 10 x greater, respectively, than the natural contents (Fig. 4.35). The upper horizons at both locales, which are characterized by overall increased metal inventories, are punctuated by dramatic metal spikes; however, those at station 1 (i.e., most distal from the tailings discharge) are an order of magnitude greater than comparable enrichments at station 6. This implies that higher metal contents are associated with the more widely dispersable "slimes" fraction of the tailings. Pedersen and others (1993) proposed a textural argument to account for metal distributions observed in Anderson Lake, Manitoba, into which tailings have been deposited since 1979. In their study, a coring site over 1 km distant from the tailings discharge was characterized by bands of metal-rich sediments in the upper facies that had Zn, Cu and Pb concentrations an order of magnitude greater than those observed in pure bulk tailings deposits. These horizons hosted ~2 wt.% Cu, -12 wt.% Zn and -15 wt.% S, and were marked by high 145 Fe /Al ratios ( - 4 : 1 ) , suggesting metal-pyrite and metal-sulphide associations. The sulphur distribution at station 1 in Balmer Lake sediments is not clearly matched with the observed metal spikes (Fig. 4 . 3 4 E ) . A broad S enrichment of 1.5 wt.% between 3 and 8 cm, and a minor surficial increase do correlate coarsely with Ni, Zn, Cu and As profiles. Assuming sulphide stoichiometries of C u 2 S , ZnS, NiS and A S 2 S 3 , the sulphur inventory present is more than needed to account for the metal enrichments if they occur as their respective sulphides. Fe/Al profiles at station 1, although characterized by general increases in the top stratum, correlate poorly with the observed metal spikes (Fig. 4 . 3 4 F ) . However, the large Fe signal contributed by the clay fraction of the tailings input would potentially overwhelm any contribution of Fe associated with the observed metal spikes. A coarse calculation comparing the S accumulation rate and the downward diffusive flux of sulphate was performed in order to test whether the observed distribution could possibly represent an authigenic sulphide precipitate (Appendix E). The lake-wide averaged diffusive influx of sulphate-S was calculated to be within an order of magnitude of the S accumulation rate, suggesting that in situ sulphide precipitation may partially account for the observed S distribution at station 1. Carignan and Tessier (1988) also observed sedimentary S maxima between 2 and 8 cm below the interface in eight sulphate-enriched Candadian Shield lakes. Such values, which ranged from 0 . 2 - 3 . 4 wt.%, were attributed to diagenetic enrichments arising from the downward diffusion of SO4 " followed by its fixation as various S species. Reduced inorganic (e.g., Fe-mono sulphides, pyrite, and elemental S), rather than organic sulphur compounds constituted the major proportion of the total S in these sediments. Organic forms of S were also 146 shown to be insignificant in polluted lakes near Sudbury, Ontario (Nriagu and Soon, 1985). Seasonal profiles of porewater Mn distributions (Section 5.3) indicate that considerable diagenetic cycling occurs in the near-surface sediments of Balmer Lake. Manganous ions remobilized in the reducing porewaters migrate upwards where they may precipitate as manganese oxide in aerobic sediments. At steady- state, the reducing horizon will migrate upward at the rate of linear sediment accumulation. Under such conditions, the cycle of dissolution and precipitation results in progressive solid-phase Mn enrichment above the Mn oxide redox boundary (Klinkhammer, 1980). In the organic-rich, highly reducing sediments in Balmer Lake, however, the thin veneer of oxic sediments limits the extent to which Mn can accumulate. For this reason, the broad, deep, solid-phase Mn maximum observed at station 1 does not likely represent a diagenetic profile. The Mn enrichment probably stems from the addition of a tailings-derived carbonate or sulphide phase. The sedimentary geochemistry of Mn will be discussed in greater detail in section 5.3.3. 5.3 Porewater Chemistry Examination of trace metal porewater distributions (Appendix C), in conjunction with the various metabolite profiles associated with the degradation of organic matter, provides an effective way to study metal behaviour and mobility in sediments. The dialysis-array samplers used in the Balmer Lake project yielded high resolution profiles of dissolved constituents across the sediment-water interface, and these are usually sufficient to allow reasonably accurate determinations of the direction and extent of diffusive trace metal fluxes. Furthermore, the high-resolution information provides good support for 147 the interpretation of post-depositional processes of trace metal consumption and remobilization. Various tools can be employed to assist interpretations of porewater profiles. Saturation indices, for example, are often used to elucidate solution/precipitation chemistry in marine and lacustrine sediments (e.g., Carignan and Nriagu, 1985; Balistrieri et al., 1992). However, caution must be exercised when comparing the measured ion activity product with the solubility product to deduce the tendency of a solid phase to form, as the kinetics of solid phase formation are often slow, and the measurement of soluble species (e.g., by filtration) may include colloidal forms (Davison, 1993). Furthermore, metal complexation with ligands in solution or those associated with particle surfaces can also promote deviations from predicted equilibria. Due to the lack of reliable thermodynamic data, most studies consider simple mineral phases in solubility calculations. This can result in further misinterpretation, as the formation of mixed phases has been proposed to control the concentration of numerous species (Suess, 1979). Various programs, including MINEQL, and HYDRAQL (Papelis et al, 1988), are often used to constrain chemical speciation and equilibrium in such solubility calculations. Examination of the dissolved trace metal and metabolite porewater profiles in Balmer Lake sediments demonstrate that dramatic changes in the chemical environment occur across the sediment-water interface. Sharp gradients for most constituents appear at or shortly below the sediment surface throughout the lake during all seasons. The shapes and depths of trace metal and metabolite profiles, in combination with solid phase distributions, are used below in the interpretation of the porewater chemistry in Balmer Lake. In particular, zones of non-linear concentration gradients delineate the depths of trace metal consumption and 148 remobilization. The absence of porewater HS", PO42" and HCO3" measurements prevented calculation of the solubility of various authigenic trace metal precipitates. However, due to the high spatial resolution of the profiles and the well-constrained thermodynamic transitions defined by NO3", Mn(II), Fe(TJ) and SO4 " data, a reasonable degree of confidence can be drawn in the predictions of trace metal behaviour. Diffusive fluxes, and hence estimates of reaction rates, were calculated from linear concentration gradients (Appendix F) using Fick's First Law: Jz = <t> dc F dz 2 1 where Jz = flux; D ° = in situ diffusion coefficient in cm 'sec (Li and Gregory, 1974); F = formation factor = 1.4 for silty clay (Manheim, 1970); this takes into account the tortuous diffusion path of an ion in wet sediments; 0 = porosity = 0.87, calculated using the measured water content and an average particle density of 2.65 gem"3; and dc/dz = concentration gradient in gem"4. It is assumed that porosity and diffusion coefficients are invariant with depth. Diffusion coefficients were not corrected for any random transport mechanisms such as biodiffusion, gas ebullition or wave mixing. 5.3.1 Ammonium and Nitrate The cycles of the essential elements, C, N , and P, are coupled because they are incorporated into the living biomass of primary producers in relatively constant ratios. The complete mineralization of this biomass must therefore lead + 2- to the regeneration of the nutrients, H C 0 3 , N H 4 and H P 0 4 in a similar stoichiometric proportion (Redfield, 1958). The generally concave-downward 149 profiles in Balmer Lake sediments suggest that N H 4 + is actively being liberated throughout the sampled sediment column (Figs. 4.12-4.14; 4.17) The large porewater N H / gradients evident during the spring, summer and fall, support upward diffusive fluxes towards the sediment water interface. Efflux estimates 0 1 over these periods range from 0.2-1.2 mmolm" d" (Appendix F). The bottom water mining-derived enrichment in the shallow waters of "Tailings Bay" (station 5) in the fall, however, fosters a downward influx of NFL; + into the sediments (Fig. 4.13D). The combination of a deeper remineralized sediment source at this site results in bi-directional diffusion toward the interface. The nitrification of upward diffusing NFL;+ at oxygenated horizons during the spring, summer and fall, represents the principal NFL;* sink in the interfacial sediments and bottom waters (Froelich et al., 1979). Oxidation is usually considered to proceed via the reduction of molecular O2 and Mn(IV) by nitrifying bacteria (Van der Weijden, 1992). From N H / profiles, it is generally assumed that NH4 + is not oxidized, directly or indirectly, by Fe(III) oxide (Froelich et al., 1979). Lower C:N ratios in the upper sediment stratum, however, suggest a portion of the remobilized NFL;* sorbs to the abundant clay fractions or other phases in these horizons (see section 5.2). Over the fall-winter transition, considerable shifts occur with respect to the distribution of porewater and bottom water NH4 + (see section 5.1.3). When the lake becomes ice-covered and anoxia develops at depth, NH4 + is predicted to diffuse across the sediment-water interface into the bottom waters. Diffusive effluxes are expected to be greatest in the earliest stages of anoxia, and to wane as diffusive equilibrium develops between interstitial and bottom waters. At the time of sampling in late winter, development of such an equilibrium was anticipated to be in an advanced stage. If the largest diffusive NFL;* flux calculated above of 1.2 mmolm" d" remained constant over time, it would take 150 approximately 170 days to produce a 200 uiriol'L"1 enrichment in the lowermost metre of the water column. Although this calculation represents a maximum, it suggests that diffusive processes significantly affect bottom water concentrations during the seasonally anoxic period. The relatively minor influences of mining inputs at station 4 make this site the most suitable to assess winter benthic exchanges due to natural processes (Fig. 4.16A). Gradients normally observed in the sediments migrate above the sediment surface during the winter at this site; the N H / profile reflects diffusive equilibrium across the interface. The concave-upward profile between 5 and 20 cm above the sediment surface implies oxidative removal of N F I / in this zone. Similarly, a concave-upward profile for N0 3" mirrors that of NFL;"1", suggesting active nitrification (4.16B). The seasonal N H / distributions at stations 1 and 2 have been plotted together to illustrate the variability over the course of a full year (Fig. 5.2). Note that spring profiles were not obtained for station 2. The spring, summer and fall profiles clearly illustrate ammonia production below the interface; these contrast strongly with the generally invariant N H i + distribution in the winter. The zone of oxidative removal migrates above the sediment surface in winter, and is characterized by a dramatic decrease in the concentration gradient. An unusual winter maximum above the sediment-water interface is proposed to reflect a non-steady-state input associated with mining-related discharges. Closer inspection of the winter N H 4 + distributions at stations 1 and 2 (Figs. 4.15A and D) illustrates the interfacial variation at these sites. It is suggested that variations with respect to the timing, frequency and magnitude of mine-related inputs can account for such variability. The fact that the N H 4 + concentrations are slightly higher in bottom waters at station 1 in winter implies that a minor ammonium influx to the sediments exists temporarily under the ice at this location. 151 Fig. 5.2. Seasonal peeper profiles of ammonium for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 152 Nitrate profiles during the spring, summer and fall exhibit remarkable inter-season and inter-site consistency (Fig. 5.3). Vigorous nitrate reduction begins 2-5 cm below the sediment-water interface, resulting in downward fluxes ranging from 0.3-1.3 mmol N'm" d" during the well-mixed periods (Appendix F). Such implied denitrification rates are similar to those observed in organic-rich sediments of other temperate lakes (Rudd et al., 1986), and lie intermediate to 2 1 denitrification rates observed in aerobic deep-sea sediments (3-10 umol N'm" d") 2 1 and eutrophic coastal marine systems (7 mmol N'm" d") (Jenkins and Kemp, 1984). The oxidation of upward diffusing NFL;+appears to account for the interfacial NO3" spikes seen particularly in summer at sites 1 and 2 (Fig. 5.3). Coastal marine sediments commonly have a sharp NO3" peak close to the sediment surface, from which NO3" is lost to the bottom waters via upward diffusion (Emerson, 1976). In this case, the peaks apparently occur at or immediately above the interface, implying that the sediment surface is the site of extremely active nitrification in summer. The winter data indicate that NO3" reduction begins 7-15 cm above the sediment-surface at stations 1,2 and 4 (Figs. 4.15 and 4.16; see section 4.1.3). The near-bottom anoxia implied by these profiles is not illustrated by the nitrate profiles at the shallow (2m) station 6 (Fig. 4.16E), indicating that anoxia is restricted to deeper lake areas. 5.3.2 Sulphate After aerobic respiration, SO4 " reduction is, on a global scale, the most important process in the diagenesis of organic matter in marine systems. However in lakes, sulphate reduction is relatively insignificant given the dearth 153 Fig. 5.3. Seasonal peeper profiles of nitrate for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 154 of dissolved SO42" (usually < 100 u.mol'L'1). This generalization does not apply to Balmer Lake, however, where relatively recent anthropogenic additions of SO42" have undoubtedly shifted the respective proportions of organic matter oxidized by O2 and SO4 ". This becomes particularly relevant during periods of bottom water anoxia, during which the absence of other oxidants at the sediment-water interface implies that organic matter remineralization proceeds via SO42" reduction alone for possibly several months per year. The dialysis-array data nicely depict the temporal mechanisms that govern the seasonal distribution of SO42". In winter, large external SO42" inputs impart a diffusive influx of SO42" into the sediments; the topmost porewaters host in excess of 8 mmol'L"1 during such periods (Figs 4.15 and 4.16). In spring, average water column values are greatly reduced (to ~ 3.5 mmol'L"1) as a result of water column mixing and an increase in lake volume. The subsurface maximum in spring represents the remnant of the interfacial SO4" excess seen in the late winter (Fig. 5.4). The reduction of SO42" in the porewaters, combined with diffusional transport, indicate that the sub-surface peak is in a state of progressive decay. Indeed, early summer profiles at station 1 exhibit a highly attenuated SO4 " maximum (Fig. 4.12C), and no evidence of the peak remains by fall (Fig. 4.13C). Diffusive influxes of sulphate over the course of the sampling year ranged O -I from 0.5-2.4 mmol'm" d" (Appendix F). The lower end of this range agrees well with Rudd et al. (1986), who reported reduction rates ranging from 0.1-0.4 mmolm" d" in organic-rich sediments in several temperate lakes. The upper limit in Balmer Lake sediments represents enhanced transport across the interface due to the large gradients imposed by sulphate-rich effluents. The depth where sulphate reduction commences in the Balmer Lake deposits, defined where d 2[S0 4 ]/dz 2 > 0, varies seaonally. The reduction horizon appears to 155 Fig. 5.4. Seasonal peeper profiles of sulphate for stations 1 and 2 (A and B , respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 156 migrate towards the interface through the summer-fall-winter transition, 2- indicating that S0 4 is being consumed at progressively shallower depths. The 2- presence of abundant electron acceptors other than S0 4 (e.g., reactive Fe(IJJ) and Mn(IV)) may sustain microbial processes in the upper sediments during the early summer. However, enhanced production and build-up of organic matter over the summer would cause labile Fe and Mn species to become progressively 2- depleted, thus favouring the consumption of S0 4 at shallower sediment depths. 2- Although dissolved Cu profiles suggest S0 4 is being reduced several centimetres above the sediment-water interface in winter (Fig. 4.23-4.25), non- 2- linear S0 4 concentration gradients only become obvious below the sediment 2- surface (Fig. 5.4). The S0 4 -rich advective inputs most likely overwhelm 2- consumption signatures in the hypolimnion in this season. The S0 4 reduction horizon deepens again in the spring as Mn and Fe oxides precipitate at and just below the reoxygenated interface. In addition to the reduction of sulphate, sulphide is also introduced to porewaters by the hydrolysis of proteins in organic detritus; fresh plankton has an organic-S content ranging from 1-2 wt.% (Jorgensen, 1977). Although dissolved trace metal profiles point to the precipitation of solid-phase sulphides as a sink for H2S, the respective organic and inorganic contributions to the overall accumulation of sulphur remain unclear (see section 5.2). 5.3.3 Iron and Manganese Diagenetic Fe and Mn enrichments are commonly observed at, or near the surface of oxic lake sediments. Such accumulations normally result from the upward diffusion of porewater Fe(II) and Mn(II) in anoxic sediments, followed by precipitation as their oxyhydroxides in surficial oxidized horizons (Carignan 157 and Tessier, 1988). In neutral lake waters, the oxidation of Fe is believed to proceed chemically, although its removal below the oxycline indicates that other oxidants in addition to O2, such as NO3" and Mn(II), may be involved. The extent of diagenetic iron oxyhydroxide enrichments in Balmer Lake can not be determined from the solid-phase data due to the large mining-derived signals of Fe and Mn in the top 10 cm of the sediment column (see section 5.2; Figs. 4.34F and 4.35A). However, the respective remobilization depths of porewater Fe and Mn during oxygenated periods provide an indication of the thickness of surficial oxide enrichments and hence help assess the sorptive capacity for trace metals. Dissolved Mn distributions in porewaters during the spring, summer and fall periods suggest that Mn oxyhydroxides are remobilized at depths ranging from ~2 to 8 cm (Figs. 4.18-4.27). The shallower depths evident in the fall suggest the Mn reduction horizon migrates upwards throughout the summer months, as alluded to in Section 5.3.2. The addition of dissolved Fe to the summer porewaters is thermodynamically consistent with the Mn distribution; significant gradients are evident 2-3 cm below comparable increases for Mn (Figs. 4.18 and 4.19). Assuming that the Fe oxyhydroxide dissolution front migrates in accordance with Mn, it can be proposed that the sorptive capacity of the surficial sediments for trace elements decreases through the summer, fall and winter, and increases again in the spring. Although dissolved iron distributions in Balmer Lake sediments exhibit considerable inter-site and inter-season (i.e., non-steady-state) variability, the available evidence indicates that authigenic sulphide precipitation presents a significant control on the distribution of Fe and other trace elements. During the summer period, concavity in the Fe profiles between ~8 and 20 cm depth at Stations 1 and 2 implies removal from pore solution (Figs. 4.18A and 4.19A). Since these zones lie immediately below the nitrate-zero boundary, sulphide 158 precipitation is implicated as a likely Fe sink. Sulphide precipitation was proposed to account for similarly shaped profiles in a Canadian Shield lake (Carignan, 1984; Carignan and Nriagu, 1985). The non-steady-state effects on the porewater sulphate distributions in Balmer Lake limit the use of the sulphate profiles in precisely defining dissolved sulphide production zones. In general, however, the sulphate distributions are thermodynamically consistent with the respective sulphate reduction horizons (Figs. 4.12C and F). The migration of the SO4 " reduction horizon towards the interface over the summer implies that authigenic sulphides precipitate at shallower sediment depths in the fall. The relative absence of Fe above 20 cm at station 1 in the fall (Fig. 5.5) can be attributed to two factors: (i) the reducible iron oxide inventory in the upper sediment column in spring is exhausted by late autumn; and (ii) sulphide precipitation removes iron from porewaters in the upper two decimetres as conditions become more reducing through the late summer. Deep sources of dissolved Fe can also be seen in profiles from stations 1 and 2 (Fig 5.5). Such increases at depth may represent the historical addition of dissolved iron to porewaters prior to the mining operations, when the natural dissolved sulphate concentration in the watershed was very low. Under such conditions, limited production of dissolved sulphide during diagenesis allows high Fe 2 + concentrations to persist in porewaters. Winter Fe profiles exhibit considerable inter-site variability (Fig. 5.6). The distribution at station 1 is fairly consistent with the summer and fall profiles at this site. The high concentrations of dissolved Fe observed in the deep porewaters at station 2 during the summer and fall (3-4 mg'L"1) are absent in winter (Fig. 5.6). At stations 4 and 6, dissolved Fe levels generally remain < 500 (ig'L"1. Whether such seasonal disparity at station 2 represents a sampling artifact, or a combination of diffusion and consumption mechanisms, remains 159 Fig. 5.5. Seasonal peeper profiles of dissolved Fe for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 160 Winter iron 0 500 1000 1500 2000 2500 Dissolved [Fe], ppb Fig. 5.6. Winter peeper profiles of dissolved Fe for stations 1, 2, 4 and 6 in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 161 unclear. The rapid oxygenation of Fe(II) in the presence of oxygen does make the sampling of reducing porewaters more susceptible to potential error. The biological processes leading to sulphide mineral formation are generally characteristic of anoxic sediments where the bacterially-mediated oxidation of organic matter is accomplished via reduction of porewater sulphate and production of hydrogen sulphide (Berner, 1984). Under such conditions, metal ions remobilized by reductive dissolution of oxide minerals can react with H2S and other sulphur compounds to form a variety of sulphide minerals. Due to the abundance and ubiquitous nature of Fe, these authigenic precipitates are often dominated by Fe-sulphides. The major iron sulphide minerals found in lacustrine sediments are mackinawite (Feo.995-i.023S), greigite (Fe3S4), pyrite (FeS2) and amorphous-FeS (Morse et ah, 1987). As an assemblage, amorphous-FeS, mackinawite and greigite are generally termed acid volatile sulphides (AVS) because they are readily decomposed by acids. In the presence of HS" at concentrations above pyrite saturation, transformation of these metastable phases usually proceeds to pyrite, a more thermodynamically stable mineral (Berner 1984). However, pyrite can precipitate directly in environments where the more soluble monosulphides are undersaturated with respect to sulphide (Perry, 1993). The amount of pyrite that may form in sediments is controlled by the rates of supply of labile organic matter, reactive detrital iron minerals and dissolved sulphate (Berner, 1984). In most marine systems (those with oxygenated bottom waters), the amount and reactivity of organic matter buried in the sediment represent the primary control on pyrite formation because it is this organic supply which limits the rate of in situ sulphate reduction and thus hydrogen sulphide production (Berner, 1984; Calvert and Karlin, 1991). Such sediments deposited in oxygenated 162 environments generally show a linear relationship between total or pyritic sulphur and organic carbon (Raiswell and Berner, 1985). In contrast to marine systems, pyrite formation in freshwater regimes is generally inhibited by low concentrations of dissolved sulphate (Berner, 1984). Therefore, much higher concentrations of pyrite typically form in organic-rich marine sediments; this has led to the use of C /S ratios to distinguish freshwater and marine sedimentary rocks (Davison et al., 1985). Experimental data suggest that limitations on pyrite formation are not important at sulphate concentrations greater than 5 mmol L"1 (in Berner, 1984, from Westrich, 1983). Concentrations of ~3 mmol L"1 in Balmer Lake deep waters during the summer and fall imply that authigenic sulphide formation may be somewhat limited by sulphate abundance. During the winter and spring periods, however, a plentiful supply of both organic matter and reducible sulphate (i.e., > 6 mmol'L"1) suggest that the amount of reactive Fe-phases limits pyrite formation during these periods. From the above discussion, it can be proposed that extensive precipitation of authigenic sulphides in Balmer Lake sediments is likely a relatively recent phenomenon, reflecting the current unnatural abundance of SO4 " in the lake waters. Recent sedimentary sulphide additions to S-polluted lakes in the Sudbury region of southern Ontario have similarly been attributed to anthropogenic sulphate loadings (Carignan and Tessier, 1988). Diagenetic sulphur accumulations in excess of 3 wt.% (i.e., greater than those enrichments observed in Balmer Lake) have been reported in such lakes. Since the potential SO42" supply has exceeded the labile-Fe (i.e., non-silicate) supply in these lakes, S fixation as inorganic sulphides appeares to be iron limited (Carignan and Tessier, 1985a; Nriagu and Soon, 1985). Reports of trace metal sulphide phases generally make no clear distinctions between precipitation as the pure metal sulphide or co-precipitation 163 with Fe sulphide. Experimental investigations have revealed that adsorption and coprecipitation of many trace elements with mackinawite (FeS) and pyrite (FeS2) are important processes in marine anoxic sediments (Huerta-Diaz and Morse, 1992; Arakaki and Morse, 1993; Morse and Arakaki, 1993). Furthermore, analyses of lacustrine iron sulphide particles collected from the anoxic bottom waters in a soft-water lake in the U.K. revealed Cu and Zn concentrations in excess of 4,000 and 6,000 mg'L"1, respectively; such levels were higher than those observed in iron oxide precursors (Davison et al, 1992). Control by phases other than oxide and sulphide precipitates may be affecting the behaviour of Fe species in Balmer Lake, although this is difficult to constrain with the available data. Iron carbonate and phosphate minerals, for example, are potentially important authigenic components of lacustrine sediments. Iron carbonate phases form in anoxic environments, where the metabolic generation of inorganic C species leads to supersaturation, favouring the formation of siderite (FeC03) (Suess, 1979). Such solubility control has been invoked to account for Fe distributions in several lacustrine studies (Emerson, 1976; Matisoff et al., 1980; Cook, 1984; Carignan and Nriagu, 1985). Although no definite conclusions can be drawn for Balmer Lake, the abundance of reducible sulphate suggest that dissolved Fe species in the anoxic porewaters would be scavenged by free sulphide. The seasonal porewater Mn distributions at stations 1 and 2 suggests that non-steady-state inputs of Mn-rich water are contributing to the temporal variability (Fig. 5.7). The summer and fall profiles are consistent with diagenetic processes; i.e., remobilization in reducing porewaters and precipitation in the oxic surficial horizons. However, the large winter maximum at ~15 cm depth at station 1 cannot represent a post-depositional signature. The exhausted inventory of reducible Mn-oxides in the surface sediments in late autumn 164 Fig. 5.7. Seasonal peeper profiles of dissolved Mn for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 165 effectively excludes the dissolution of such phases as a potential source of the dissolved input. Furthermore, the depth of the dissolved maximum lies within the zone of postulated FeS precipitation, suggesting that labile Mn-oxides are not present at these depths. The above argument implies that an influx from the bottom waters is contributing to the observed winter profile (Fig. 5.7). Implicit in this argument is that at some point between the fall and winter sampling periods (Oct. to May), the sediments at station 1 were in contact with bottom waters hosting Mn concentrations in excess of 1400 jxgU1. The large mining-derived sedimentary accumulations of Mn in the upper fades (Fig. 4.35) do suggest that tailings pond effluents host abundant manganese. Figure 5.8 illustrates the proposed temporal sequence. An instantaneous input of dissolved Mn to the fall bottom waters (5.8B), followed by diffusion of Mn into the sediments would foster increased concentrations in the upper porewaters (Fig. 5.8C). By late winter, the strength of the bottom water Mn source had decreased. However, due to the longer residence time of dissolved species in pore solution, evidence of the previous input persists (Fig. 5.8D). The high porewater concentrations support an efflux of Mn into the bottom waters during this period (Fig. 5.7). A similar deep porewater maximum was not evident for sulphate during winter sampling at stations 1 and 2 which may suggest that either: i) the consumption of sulphate in the interfacial horizons inhibits accumulation in the porewaters; or ii) the mining-related inputs of manganese and sulphate are temporally decoupled. Dissolved Mn concentrations progressively decrease from the winter maximum through the spring, summer and fall (Fig. 5.7). The re-establishment of an interfacial Mn sink upon re-oxygenation of the water column in the spring suggests that the progressive consumption of the dissolved Mn near the 166 A [Mn] B [Mn] bottom waters sediments c [Mn] D [Mn] Figure 5.8. Time series of progressive bottom water and porewater profiles (A-D) over the fall-winter transition in Balmer Lake: A) fall profile, B) instantaneous input of Mn to bottom waters, C) profile after some time of equilibration, and D) late winter profile. Depth and concentration axes are in arbitrary units. sediment-water interface, in conjunction with diffusional transport, are responsible for the declining values. The minimal influence of mining-related bottom waters at station 4 make this site perhaps best suited to examine the winter Mn behaviour (Fig. 4.25B). The Mn-redoxcline has clearly migrated above the sediment surface at this site, and temporary diffusive equilibrium has been established across the sediment- water interface. The concave-upward profile between 8 and 25 cm above the lake floor is consistent with the oxidative precipitation and removal of upward diffusing Mn(JJ) species. Upward diffusive effluxes determined for Mn ranged from 54-700 ug'm" 2 'd _ 1 (Appendix F). The low end represents fluxes calculated from the much shallower gradients observed during the late-winter period. This is consistent with the hypothesis that once the oxidative sink for dissolved Mn is removed 167 during winter anoxia, diffusional processes will cause a progressive decrease in the concentration gradient. Efflux estimates for the spring, summer, and fall exhibit reasonable consistency (Appendix F). Calculated effluxes were used to test the potential influence of such mechanisms on the bottom water composition. At station 4, for example, an excess of approximately 1 x 105 p.g has been added to the lower 0.5 m of the winter water column (Fig. 4.11). Calculations using a constant maximum gradient suggest that diffusive processes alone would take on the order of 140 days to account for such an accumulation. Similar calculations for hypolimnic concentrations at station 1 (~2 x 105ug excess in the lower 0.5 m) and station 2 (~ 3.5 x 105 u.g excess in the lower 0.5 m), demonstrate that time scales on the order of 280 and 500 days, respectively, 2 1 would be required. Since the maximum efflux of 700 u.g'm~ d~ was used in the above calculations, the estimated time periods most likely represent minimum values. The calculations suggest that mining-related inputs may have contributed to the bottom water enrichments at stations 1 and 2. Indeed, bottom water sulphate values are significantly higher at these two stations. The contrasting dissolved Mn profiles observed in the winter at station 6, indicate that Mn is being consumed as opposed to being remobilized in the upper sediments at this site (Fig. 4.26B). Sulphide precipitation as MnS (Suess, 1979) and sorption with mackinawite (FeS) (Heurta-Diaz and Morse, 1992; Arakaki and Morse, 1993) have been suggested to control Mn distributions in marine sediments. The relatively soluble nature of MnS (alabandite, pK = 0.40) suggests that the latter is more favourable in Balmer Lake sediments. Laboratory investigations have demonstrated that adsorption of Mn(U) with fine-grained mackinawite is more important than coprecipitation, and has been attributed to a combination of the high specific surface area of FeS, and a small partition coefficient that limits Mn incorporation into the FeS lattice (Arakaki and Morse, 168 1993). The tailings-rich deposits at station 6 perhaps favour the removal of Mn from porewaters via such adsorption mechanisms. The precipitation of Mn-carbonate phases has been suggested to govern the behaviour of Mn in the reducing porewaters of several lake sediments (Matisoff et al., 1980; Carignan and Nriagu, 1985). Supersaturation with respect to carbonate minerals can be achieved in anoxic environments due to the pH- buffering effect by proteolytic formation of H 2S, other weak acids, and possibly H +-ion exchange (Suess, 1979). However, solubility considerations indicate that the porewaters at station 6 are likely undersaturated with respect to rhodochrosite. 5.3.4 Nickel Peeper profiles suggest that two factors control the distribution of nickel in Balmer Lake porewaters: (1) release to solution from labile Ni-bearing phases at or near the sediment-water interface; and (2) the precipitation of authigenic Ni- bearing sulphide phases at depth. Each will be discussed in turn. Surface or near-surface maxima of dissolved Ni can be observed throughout the year in almost all profiles collected from Balmer Lake sediments. Although the enrichments are generally small in relation to extent of removal at deeper depths, the peaks are well-constrained and appear to represent real signatures (Figs. 4.18-4.27). The magnitudes of these signals above the respective water column background concentrations average ~40 ppb and range from 30- 200 ppb. Concentration profiles suggest that Ni is simultaneously diffusing from the interface to the sediments, and from the interface to the overlying water. The fact that these profiles exhibit reasonable seasonal consistency suggests that the spikes are representative of steady-state conditions. 169 Remobilization of trace metals at the sediment-water interface has been typically attributed to remineralization of organic matter and/or dissolution of Fe and Mn oxyhydroxides (Kh'nkhammer, 1980). In the case of Balmer Lake, the presence of dissolved Ni maxima 2 to 5 cm above the predicted horizon of Mn(IV) reduction during the summer sampling session suggests that Ni is released from the oxidation of organic matter, and not by dissolving oxides (Figs. 4.18D and 4.19D). Enrichments derived from organic matter oxidation would be expected to be observable at or near the sediment-water interface due to the higher rates of remineralization in the surface fades that stem largely from the greater reactivity of freshly deposited organic matter (Middleburg, 1989). In the fall period, the Mn and Fe redox horizons are immediately proximal to the sediment-water interface, and as such, a biogenic origin cannot be uniquely isolated (Figs. 4.13-4.14). The cycling of organic matter has been invoked as a control on the distribution of Ni because the metal has a nutrient-like distribution in the oceans (Bruland, 1980). The principal mechanisms of incorporation into particles are active algal uptake and indiscriminant complexation by high-affinity surface ligands. It has been proposed that some non-essential constituents function as chemical analogues to those that are essential, and hence, are sequestered and transported in a similar manner (Morel and Hudson, 1985). Although reports of Ni concentrations in porewaters are rare, the existing information sheds important light on the early diagenetic behaviour of the element. Westerlund et al. (1986), for example, used benthic flux chambers to assess trace metal transport across the sediment-water interface in Swedish coastal deposits. They showed that Cu, Ni, Zn and Cd, were released to the bottom waters in spite of the presence of abundant sulphide at shallow sediment depths. These trace metals were shown to be associated with phases from which 170 they were released via oxidation, rather than by reduction as was the case for Mn, Fe and Co. Similarly, Klinkhammer et al (1980) and Tsunogai et al. (1979) attributed post depositional migration of Ni and Cu to release from labile organics during early diagenesis in pelagic sediments. Interfacial enrichments on the order of ~ 10 ppb greater than immediately overlying water column values have been observed for some of the more biologically important elements (e.g., Zn; Morfett et al., 1988). However, oxidative additions in excess of 30 ppb for elements having less strong associations with organic matter (e.g., Ni), have not been previously reported. It is proposed that the high concentrations of Ni in the water column of Balmer Lake (averaging >300 ppb) facilitates its sorption onto and/or incorporation with labile particulates in concentrations great enough to support interfacial releases of such large magnitude. Indeed, the particulate Ni concentrations observed in Balmer Lake (Section 4.1.4.2) are comparable to the highest ever reported (see Nriagu et al, 1982). In addition to its affiliation with organic matter, nickel has been shown to have a high sorptive affinity for oxide surfaces (Vuceta and Morgan, 1978; Young and Harvey, 1992), and is commonly associated with diagenetic Mn and Fe enrichments in lake sediments (Carignan and Nriagu, 1985; Cornwell, 1986). For these reasons, diagenetic accumulations of Fe and Mn in Balmer Lake sediments may be expected to liberate exchangeable Ni species at their respective reduction horizons. Indeed, the large shallow spike of dissolved Ni in the fall at station 1 (Fig. 4.20D) might reflect contributions from both organic and oxide sources. In addition, the slightly deeper Ni enrichments observed in the spring (at ~5-7 cm), correspond more closely with the apparent remobilization depths of Fe and Mn (Fig. 4.27D). Remobilized porewater Ni maxima in excess of 8 ppb have been reported in coastal marine sediments (Shaw et al, 1990). Rapid sulphide 171 complexation/precipitation of liberated Ni may, however, prevent the formation of a dissolved Ni excess at deeper depths in the summer. Indeed, the reducing environment in Balmer Lake sediments suggests that the zones of trace metal remobilization (due to Mn(IV) and Fe(IH) reduction) and sulphide precipitation almost overlap. The downward concavity below ~5 cm seen in all profiles indicates that the sediments of Balmer Lake provide a sink for dissolved Ni throughout the year. Two observations support the precipitation of Ni-bearing sulphide phases as the most probable removal mechanism. First, the Ni removal zones in general correspond with SO4 " reduction horizons. Second, authigenic sulphide precipitation has been reported for Ni in several marine and lacustrine sediments (Morse and Arakaki, 1993; Green et al., 1989; Belzile and Morris, 1995), including reports from Canadian Shield lakes (Carignan and Nriagu, 1985). Peeper profiles of dissolved Ni exhibit remarkable inter-site and inter- season consistency (Fig. 5.9). While sulphate distributions follow a seasonal pattern consistent with the development of anoxia in winter (i.e., reduction depths are shallowest in late winter and deeper in spring and fall), Ni removal profiles do not obey this sequence. In general, the predicted zones of Ni removal occur deeper in the winter than in the fall and summer. Reasons for this are not clear; however, a decoupling of Ni and sulphate during the winter months might be expected considering the variable rates of discharge and poorly constrained composition of mining inputs. Non-steady-state Ni profiles associated with variable mining discharges like those seen for NFL;+ and Fe, are not easily discernible from the winter data. The predominantly uni-directional diffusive transport perhaps makes this element less susceptible to such influences. Solubility calculations indicated that the porewaters in Canadian Shield lakes studied by Carignan and Nriagu (1985) were supersaturated with respect to 172 Fig. 5.9. Seasonal peeper profiles of dissolved Ni for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 173 millerite (NiS). Diffusional influxes of Ni, which ranged from 10-51 u.g'cm~2 y"1, represented 59-161 % of the calculated Ni accumulation rate. Similar influxes are 2 1 calculated for Balmer Lake sediments: values range from 2.6-14.2 p-g'cm" y" (Appendix F). On average, concentration gradients for nickel in Balmer Lake porewaters are steepest in the summer and fall; due to temperature-dependent diffusivities, this translates to the greatest diffusive influxes being observed in 2 1 the summer. Assuming an average influx of 6 u,g'cm" y" over the entire lake basin, a mean lake depth of 2.4 m, and an average water column concentration of 380 ug'L"1, it would take on the order of 15 years for the sediments to consume the whole-lake inventory of dissolved nickel. This corresponds to -1% sediment retention of the total dissolved nickel burden with respect to a lake residence time of -230 days, and suggests that sulphide precipitation has little effect on reducing the large burden of Ni in Balmer Lake. Furthermore, downward diffusion alone cannot account for the large sedimentary Ni accumulations. The accumulation rate, based on 0.3 wt.% Ni, a dry bulk sediment density of 0.35 gem"3, and a sedimentation rate of 0.3 em'y"1, is estimated to be ~400 jj.g'cm '̂y"1 (Appendix G). Nevertheless, the gradients observed in the porewaters will support significant accumulation of N i in periods when tailings over-flow is absent. 5.3.5 Copper The general harmony between the porewater and bottom water distributions of Cu and Ni, suggest that their behaviours are influenced by similar controls (Figs. 4.18-4.27). Unlike Ni and Zn, however, dissolved Cu profiles exhibit no evidence of any interfacial remobilization above the depth of sulphide precipitation. For all seasons, the behaviour of Cu appears to be largely 174 dictated by authigenic sulphide precipitation at shallow sediment depths during the spring, summer and fall, and in the anoxic bottom waters during the winter period. During the well-mixed periods of the spring, summer and fall, peeper profiles of dissolved Cu exhibit good inter-season and inter-site agreement (Figs. 4.18-4.22,4.27). The minimal deviations reflect slightly different removal depths, bottom water concentrations and diffusive influxes. Because there is no evidence of any regeneration from labile solid phases in the upper sediment horizons during these three seasons implies that either particulate Cu is transported to the sediments in refractory phases that are not subject to remobilization, or Cu liberated to the shallow porewaters is rapidly precipitated. The biogeochemistry of Cu, and the conditions in Balmer Lake, suggest that the latter hypothesis is most tenable. Copper is considered to be biologically more important than Ni with respect to algal metabolic requirements and is characterized by nutrient-like distributions in the oceans (Bruland, 1980; Sanders, 1983). Copper associated with labile organic matter appears to be released during early diagenesis in a wide range of sedimentary environments including lacustrine sediments (Morfett et al, 1988; McKee et al, 1989b; Reynolds and Hamilton-Taylor, 1992; Young and Harvey, 1992), coastal marine deposits (Elderfield, 1981; Westerlund et al, 1986; Shaw et al, 1994; Kerner and Geisler, 1995), pelagic sediments (Tsunogai et al, 1979), suboxic hemipelagic sediments (Klinkhammer, 1980) and marine mine- tailings deposits (Pedersen, 1985). Examinations of Cu partitioning in various sedimentary phases have demonstrated that up to 80-90 % of the Cu burden may be organically bound (Elderfield, 1981; Young and Harvey/1992). The rapid scavenging of liberated species may account for the lack of dissolved Cu additions to pore and bottom waters at the sediment-water 175 interface. The extremely insoluble nature of CuS (pK = 35.4; Stumm and Morgan, 1981) as compared to NiS (pK = 26.7), may favour the preferential removal of Cu- sulphide phases in the interfacial sediments. This proposal is consistent with the geochemistry of Cu in the Balmer Lake water column. The chemical environment in the winter hypolimnion favours the removal of Cu presumably as sulphide precipitates; such a mechanism is not obvious for Ni. A fundamental criticism of the previous argument is the fact that the postulated Cu removal is occurring at depths shallower than those predicted for sulphate reduction. However, sulphide precipitation has been implicated for trace element removal in sub-oxic sediments. The removal of porewater Cd at a suboxic front in abyssal Atlantic sediments, for example, was attributed to the precipitation of authigenic CdS (Rosenthal et al., 1995). They suggested that the diffusion of free-sulphide from underlying anoxic sediments was sufficient to support sulphide formation. In addition, Sorenson and Jorgenson (1987) found that sulphate reduction can take place within the microenvironment of organic aggregates in the oxic zone. It is possible that the organic-rich sediments in Balmer Lake promote such development of heterogeneous redox conditions in the "oxic" horizons. The tendency of a metal to form ligand complexes is usually explained in terms of ligand field crystal-stabilization theory (Stumm and Morgan, 1981). This led to the Irving-Williams order for complex stability of plus II cations, which increases in the order: M n 2 + < Fe 2 + < C o 2 + < Z n 2 + < N i 2 + < C u 2 + . The sequence presented here reflects increases in the capability of the cation to take up electrons (i.e., increasing ionization potential of the metal). Complexation, steric hindrances, entropy effects and site specificity can distort this hierarchy (Benjamin and Leckie, 1981). Based on analyses of porewaters collected in Anderson Lake, Manitoba, Pedersen et al. (1993) suggested that the high sorptive 176 affinity of Cu for dissolved organic ligands, which would tend to keep Cu in solution, could account for Cu being removed at deeper depths than other metals such as Zn and Cd. This suggestion appears not to apply to the porewater Cu and Ni distributions in Balmer Lake, however, as the depth of Cu removal is always shallower than that for Ni. The anoxic sediments in Balmer Lake clearly provide a permanent sink for dissolved Cu throughout the year. Comparisons of seasonal profiles for stations 1 and 2, for example, suggest active sulphide precipitation occurs at depths between 2-10 cm from spring through fall (Fig. 5.10). Unlike dissolved Ni, the seasonal migration in the depth of Cu removal is consistent with that of sulphate; precipitation zones migrate upwards over the summer-fall-winter transition and descend again in the spring. Sulphide precipitation has been frequently proposed to limit Cu concentrations in marine and lacustrine porewaters (Elderfield, 1981; Pedersen, 1985; Gaillard et al, 1986; Pedersen, 1988; Pedersen et al, 1990; Hamilton-Taylor and Davison, 1994) , including reports for shield lakes (Carignan and Nriagu, 1985; Pedersen et al, 1993). Carignan and Nriagu (1985), for example, observed dramatic depletions of dissolved Cu at depths consistent with sulphide production in one acidic (pH ~ 4.5) and one slightly alkaline Canadian Shield lake (pH ~ 7.5); solubility calculations indicated that the porewaters were supersaturated with respect to covellite (CuS). Similarly, Pedersen and others (1993) attributed the precipitous decline of Cu in the shallow porewaters of contaminated Anderson Lake, Manitoba, to the precipitation of authigenic discrete or mixed metal sulphide phases. Indeed, framboidal pyrite was observed by these authors to be ubiquitous in the surface sediments. Diffusional influxes clearly support significant accumulations of Cu in some lacustrine sediments. In the Shield lakes studied by Carignan and Nriagu 177 Station 1 40 B Station 2 -60 Spring 0 50 100 150 200 250 300 Dissolved [Cu], ugL"1 E o of o ca t 0) CO $ i +-* c e d) CO E o M — CD O c CO ^—• CO b -60 0 50 100 150 200 250 300 Dissolved [Cu], fig L"1 Fig. 5.10. Seasonal peeper profiles of dissolved Cu for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 178 (1985), for example, diffusive influx estimates ranged from 0.5-5.1 ug'cm '̂y"1 and represented 1-52 % of the estimated Cu accumulation rate. In Balmer Lake, diffusive Cu influxes remained reasonably constant over the four seasons, with values ranging from 2.3-6.7 M-g'cm" y" (Appendix F). In general, no obvious spatial or temporal trends are evident with respect to the inter-site and inter- season flux estimates. The rate of Cu removal by sediment influxes appears to have little effect on the total lake burden of dissolved Cu in Balmer Lake. Estimates assuming an average influx of 4 ug'cm" y", a mean lake depth of 2.4 m, and an average water column concentration of 200 M-g'L'1, suggest that -5% of the total dissolved Cu is fixed within the lake residence time of 230 days. Downward diffusive fluxes of Cu in Balmer Lake do not appear to represent the principal depositional mechanism responsible for the large sediment concentrations. As suggested for Ni, the calculated rate of diffusion of dissolved Cu species from the bottom waters into the sediments is not sufficient to account for the bands of Cu-rich sediments (~1 wt.%) at station 1 (Fig. 4.35). The accumulation rate implied by the enriched horizons is estimated to be on the order of 1000 ug'cm '̂y"1 (Appendix G). However, since the last ephemeral episode of accidental tailings overflow (~1970), diffusional transport has most likely played a significant role in Cu accumulation. Indeed, the steep negative concentration gradients are comparable to the magnitudes observed in the cited previous studies in which the importance of such mechanisms has been stressed. 5.3.6 Zinc The bottom water and porewater distributions of dissolved Zn in Balmer Lake appear to be controlled by two principal mechanisms: 1) intense remobilization from labile, Zn-rich sedimentary phases in the first 5 cm below 179 the interface; and 2) precipitation of Zn-bearing sulphide minerals at greater depths. Each will be described in turn. Peeper profiles of Zn are characterized by shallow pronounced porewater maxima during all seasons (Figs. 4.18-4.27). Steep concentration gradients either side of the spike promote bi-diffusional transport away from the interfacial horizons. The concentrations of the surficial peaks average 75 ug L"1 greater than concentrations in the overlying bottom waters (Fig. 5.11). Typically, the largest interfacial spikes and gradients were observed during the fall period. Reasons for this are not fully clear, although the degree of enrichment will be related to the rate of supply and the rate of removal of dissolved Zn species. Specifically, such processes will be influenced by the supply of organic matter, rates of organic matter remineralization and rates of authigenic sulphide formation. Diffusive effluxes, which ranged from 11-330 ug'm^'d"1, were computed in order to assess the potential contribution of remobilization on the Zn enrichment in the winter hypolimnion (Appendix F). The following discussion ignores potentially significant influences from advecting mining-flows. From water column profiles, bottom water enrichments in the lower metre of the water column at stations 1, 2 and 4, were estimated to be ~ 5 x 104, 3.5 x 104, and 1.5 x 104 ug, respectively. Assuming a constant efflux of -300 ug'm '̂d"1 as seen in the fall, upward diffusion of Zn species at these stations would take -170,120 and 50 days, respectively, to produce the concentrations observed in the bottom waters. These values represent minimum time estimates, as effluxes are expected to decrease as gradients lessen during ice-cover. They do, however, illustrate that benthic effluxes can have a significant impact on the metal content of the water column during the seasonally anoxic period. The remobilization of Zn from the oxidation of organic matter is considered to be the most important source of labile Zn species in Balmer Lake 180 Fig. 5.11. Seasonal peeper profiles of dissolved Zn for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 181 sediments. Four strands of evidence are offered to support this. First, zinc appears to be generated at shallower horizons than the respective remobilization depths of Fe and Mn during the summer period. The putative organic source is more difficult to distinguish in the more reducing fall sediments. Second, of the elements measured, zinc exhibits the best correlation with POM in water column particulates. Third, a pronounced surficial dissolved Zn spike was evident at station 1 during the winter at a time when Fe and Mn oxide sources should have been exhausted. And fourth, reports of strong Zn-organic associations abound in the literature. With respect to the latter point, compositional analyses of phytoplankton demonstrate that Zn is typically the most abundant trace metal in both marine and lacustrine algae; extensions of Redfield ratios to include micro- nutrients have revealed stoichiometrics of Cn3N15P1Zno.03-o.06 (Sigg, 1985; Reynolds and Hamilton-Taylor, 1992). However, porewater profiles that provide direct evidence of surficial enrichments derived from remobilized organic matter are scarce in the literature. Data from seasonally anoxic Esthwaite Water demonstrated that elevated dissolved Zn concentrations at the sediment surface were attributable to a remobilized source from decaying algal material (Morfett et al., 1988). Transient interfacial Zn spikes coincided with large accumulations of organic material at the sediment surface, and were characterized by constant Zn:Cu ratios in a stoichiometry consistent with that of phytoplankton (Sigg, 1985; Reynolds and Hamilton-Taylor, 1992). Similarly, effluxes of Zn in Swedish coastal sediments were attributed to releases during the breakdown of labile organic matter at or near the sediment-water interface (Westerlund et al., 1986). Comparisons of the magnitudes of near-interface dissolved Zn contents between Balmer Lake and the studies cited above reveal considerable disparities. The greatest enrichments reported by Morfett and others (1988) and Westerlund (1986) were respectively ~ 12 and 20 ug'L"1 above background water column 182 concentrations; those observed in this study ranged from 15-200 ug'L"1. It must be pointed out, however, that Balmer Lake is unique in that it hosts dissolved Zn concentrations over an order of magnitude higher than either Esthwaite Water or Swedish coastal waters. More importantly, particulate Zn concentrations are among the highest recorded anywhere (Table 4.6). Therefore, one can argue that potentially large burdens of labile, Zn-bearing particulates are deposited in Balmer Lake; the remobilization of such phases may facilitate the formation of the anomolously large Zn spikes in the surficial horizons. Due to the compressed redox depth scales in Balmer Lake, dissolved Zn contributed by near-interface reductive dissolution of Fe and Mn oxides is difficult to distinguish from Zn released during organic matter remineralization. A spatial decoupling of the two sources is evident only during the summer. Previous studies which have assessed the partitioning of zinc between the water column and the sediments in lakes suggest that the binding of Zn by Fe and Mn oxides may be important in the seasonally oxic horizons in Balmer Lake (McKee et al, 1989a; Tessier et al, 1989; Williams, 1992; Young and Harvey, 1992). An examination of eight Canadian Shield lakes by Tessier et al (1989) demonstrated a strong pH-dependence with respect to the distribution coefficients of Zn between particulates and solution phases (i.e., 1Q increased with pH). Their results agree well with theoretical and experimental complexation models of cation sorption to oxide surfaces (Benjamin and Leckie, 1981). For the lakes assessed (pH values ranged from 4 to 8.4), the pH interval 5.5 to 6.5 defined a region of maximum potential change in % Zn sorbed to Fe oxyhydroxides; the results suggested that a large portion of the potentially available or mobile Zn present should be associated with the surficial sediments above p H 7, and be present in the water column below p H 5. A similar study of oxic lake sediments from seven other shield lakes examined the 183 sediment/porewater partitioning (i.e., Ka) of Zn together with the speciation of the bound fraction (Young and Harvey, 1992). Enhanced sorption between p H 5.0 and 6.5 was also observed in this work. Although Zn concentrations in Balmer Lake are complicated by mining inputs, the p H range of 7-7.4 (i.e., the upper end of the sorption edge) suggests that a significant portion of the sedimentary Zn inventory should be associated with Fe and/or Mn oxides in oxic sediment horizons. However, the role of these phases in the diagenetic cycling of Zn in the lake sediments is not clearly reflected in the porewater zinc profiles. Below the horizons of near-interface remobilization, dissolved zinc species are rapidly consumed from Balmer Lake porewaters presumably by the precipitation of authigenic sulphides. The precipitation of Zn-sulphides is a widely recognized aspect of Zn behaviour in lacustrine systems (Pedersen, 1983; Carignan and Tessier, 1985b; Pedersen, 1988; Tessier et al, 1989; Davison et al, 1992; Williams, 1992; Pedersen et al, 1993), and has been the subject of a fairly recent review (Hamilton-Taylor and Davison, 1994). For all seasons and sites, Zn and Ni are removed at very similar depths, which may reflect their relatively comparable solubility products with respect to sulphide. As for Ni, sulphide precipitation in the hypolimnion does not appear to present an important removal mechanism during winter. During the summer, winter and spring periods, Zn concentrations in the interstitial waters are at least a factor of two lower than in the overlying bottom waters (Fig. 5.11), suggesting there is a net diffusive transport into the sediments. In the fall, however, diffusive equilibria between the sediments and water column are evident at stations 1 and 2 (Fig. 5.11). Diffusive influxes of dissolved Zn below the near-interface peaks exhibited little inter-season variability with seasonal averages ranging from 2.2-3.9 ug'cm^y"1. Except for the fall period, 184 concentration gradients associated with downward diffusion were greater than those corresponding to effluxes (Appendix F). Tessier et al. (1989) examined a large number of oligotrophic, shield lakes of varying p H and degrees of Zn contamination. In the more polluted lakes (pH<6), downward influxes driven by pronounced gradients (0.39-1.7 x 10"8 gem"4) suggested dissolved Zn species were being efficiently trapped as sulphide phases at depth; these gradients agree well with those seen in Balmer Lake deposits, which range from 2.2-5.3 x 10~8 gem"4 (Appendix F). In lakes of pH>6, Tessier et al. (1989) observed that the vertical concentration gradients were generally much lower (0.65-1.9 x 10"9 gem"4). The greater Zn inputs to the acidic lakes in combination with more vigorous sulphide precipitation were suggested to contribute to the larger gradients. In a related study, porewater Zn profiles were obtained by Carignan and Tessier (1985) in order to assess the relative importance of diffusion mechanisms in two acidic lakes. Loci of Zn removal at depths of 2 cm coincided with the presence of measurable porewater sulphide; in both sediments, the porewaters were slightly supersaturated with respect to sphalerite. Carignan and Tessier estimated that the diffusive influxes in the lakes (1.0-1.7 ug'cm'̂ y"1) accounted for 76 and 52 % of the respective sedimentary Zn accumulations. The importance of diffusive influxes to the overall Zn accumulation in Balmer Lake deposits cannot be established so rigorously given the significant seasonality in the porewater distributions. Nevertheless, the data collectively point to an important conclusion: early diagenetic reactions in Balmer Lake sediments render the deposits below 10 cm depth a significant and permanent 2 1 sink for the metal. Calculations using an average influx of 3 M-gcm" y" over the entire lake benthic area suggest that 13% of the total dissolved Zn inventory is retained with respect to the lake residence time of -230 days. 185 5.4 Factors Controlling the Diagenetic Behaviour of As The ubiquity of Fe and Mn oxyhydroxides, and their potential role in regulating trace element concentrations in natural waters, has stimulated numerous investigations on As behaviour in the presence of these substrates. The importance of Fe-bearing oxides has been repeatedly inferred to exert the principal control on the distribution and behaviour of As in marine and lacustrine systems (Crecelius, 1975; Edenborn et al., 1986; Agett and Kriegman, 1988; Belzile and Tessier, 1990; De Vitre et al., 1991; Widerlund and Ingri, 1995) while Mn has received relatively little attention (Takamatsu et al., 1985; Anderson and Bruland, 1991). The apparent strong associations between As and Mn observed in Balmer Lake sediments imply that under certain conditions, manganese may also play a significant role in As geochemistry. The section below assesses the importance of Mn and Fe with respect to the diagenetic behaviour and mobility of As in sediments, and is presented within the contexts of experimental and theoretical considerations. The argument that is developed will be then applied to the seasonal geochemistry of As in Balmer Lake. 5.4.1 Experimental and Field Observations Theoretical and experimental studies were reviewed in order to determine the geochemical influences of Fe and Mn on the diagenetic behaviour of As. A number of questions receive particular attention in the following paragraphs: 1) how do the sorptive properties of Fe and Mn oxide phases compare? 2) what are the As species predicted to be associated with these phases? and 3) what is the nature of the adsorption mechanism? 186 As discussed in sections 5.1.5.1 and 5.1.5.2, natural Fe and Mn oxides represent a diverse assemblage of phases that encompass a broad spectrum of sorptive characteristics. Due to variability in the chemical environment during their formation and their ability to incorporate foreign ions into their structures, significant natural heterogeneity can develop in oxides with respect to reactive surface areas, types of binding sites, site densities and steric effects. A comprehensive body of literature has been devoted to their sorptive properties. Although most experiments commonly use relatively pure Fe and Mn oxide phases, which may exhibit considerable compositional and morphological differences compared to natural particles, useful comparisons can be drawn. Poorly ordered Mn and Fe oxides are typically characterized by loosely hydrated structures which are permeable to ions (Davison, 1993). The incorporation or adsorption of minor amounts of foreign ions in the formation of these phases appear to restrict crystal development, thus contributing to more random growth (Taylor, 1987). The permeability means that sorption reactions are not restricted to external sites as is the case for more crystalline solids. Thus, the reactive surface areas for the mostly amorphous phases can be large. BET analyses of synthetic M n 0 2 and Fe(OH)3 suggest that they share similar specific 2 1 surface areas. Estimates for the former range from 230-320 m g" (Vuceta and Morgan, 1978), while those for amorphous Fe-oxide range from 182-600 ir^'g"1 (Davis and Leckie, 1978b; Davis and Leckie, 1978a; Benjamin and Leckie, 1982). Similarly, good agreement has been observed between Mn and Fe solid phases with respect to experimentally derived site densities and chemical free energies of adsorption for trace elements (Vuceta and Morgan, 1978). More definitive contrasts between Fe and Mn oxide phases are evident upon comparison of the nature of their respective surface charges. The surfaces 187 of many oxides become relatively more negatively charged with increasing p H (and vice versa) due to variations in the adsorption of potential-determining ions (H 3 0 + and OH") (Stumm and Morgan, 1981). The p H at which the surface assumes a net zero surface charge (the point of zero charge, or pzc) can be used estimate whether a surface is likely to be positively or negatively charged at a given p H (Kinniburgh, 1981). The pzc for goethite for example is 7.5-8.6 while that for birnessite is 1.5-2 (Kinniburgh, 1981); goethite (a-FeOOH) and birnessite (5-Mn02) represent two of the more abundant oxides of Fe and Mn in natural systems. Therefore, at neutral pH, Mn oxide is predicted to be strongly negatively charged and thus will repel anions. This argument has been proposed by several groups to support stronger associations of Fe oxides with arsenate (H2As04") at natural p H levels (De Vitre et al, 1991). To constrain further the diagenetic behaviour of As in sediments, it is necessary to define the chemical species involved and compare their behaviour. Thermodynamic arguments suggest that inorganic As in oxidized aquatic systems should be present as arsenate [As(V)] (Ferguson and Gavis, 1972); arsenic acid (H3ASO3, pK a l = 2.2, pK a 2 = 6.9) is predicted to be largely dissociated at neutral pH's. Conversely, data in the literature show that arsenite [As(Hl)] typically predominates in anoxic interstitial waters (Edenborn et al., 1986; Peterson and Carpenter, 1986; Agett and Kriegman, 1988); arsenous acid (H 3 As0 4 , p K a l = 9.2) is expected to be predominantly neutrally charged in natural waters. For simplicity, organo-arsenicals will be ignored in the following discussion. Both arsenite and arsenate are removed from solution in the presence of amorphous Fe-oxides, although As(V) is more readily sorbed than As(IH) (Pierce and Moore, 1982; Bowell, 1994). The contrast has been attributed to charge 188 differences between the two complexes. Although accounts of solid-phase As speciation are rare, it is generally assumed that As(V) predominates in oxic sediments (Belzile and Tessier, 1990). Dissolution experiments using diagenetic Fe and As samples collected from 12 Canadian lakes, for example, demonstrated that only arsenate was present (De Vitre et al, 1991). From the above discussion, it is apparent that the scavenging of As (III) by oxide surfaces involves an oxidation step. For example, As(III) removal from the porewaters in oxic horizons may involve oxidation to As(V) prior to adsorption. Alternatively, the oxidation of As(III) to As(V) may occur during or after its adsorption (Gulens et al, 1979). Indeed, the sorption of As has been shown to occur simultaneously with the oxidation of As(III) to As(V) in several Saskatchewan lake sediments (Oscarson et al, 1980). Molecular O2 and Fe and Mn oxyhydroxides are the most likely oxidants of arsenite in sedimentary porewaters. However, oxidation by molecular O2 is kinetically slow, and thus is unlikely to be important (Cherry et al, 1979). In contrast, Oscarson et al (1980) observed that synthetic, poorly crystalline birnessite (5-Mn02) was very effective in oxidizing As(III). Microbial inhibition via the addition of HgCl2 did not significantly retard the reaction, indicating that such oxidation was largely an abiotic process. Despite the thermodynamic favourability, a redox reaction between As(III) and Fe(III) oxide was not observed within three days in a similar experiment (Oscarson et al, 1981a); the results were attributed to slow kinetics. However, in another study Fe oxyhydroxides (both natural and synthetic) were found to oxidize As quite rapidly; typically, 50-75% of spiked As(III) was oxidized with two days, irrespective of pH (De Vitre et al, 1991). The apparent discrepancy between the two experiments may be explained by the much higher Fe:As ratios (3,400 vs. 10) 189 in the latter study. Thus, it appears that both Fe(IIJ) and Mn(IV) serve as effective oxidants of As(IU) given suitable conditions. Experiments have shown that both coulombic attraction and specific adsorption are important in the uptake of arsenate by Fe-oxyhydroxides. In this context, specific adsorption refers to uptake that cannot be accounted for solely by electrostatic interactions, regardless of the sign of the surface charge (Stumm and Morgan, 1981). "Specific adsorption" which is a term equivalent to "ligand exchange", thus reflects the specificity in bonding of different ions due to their charge, size and polarizability (Taylor, 1987). Several strands of evidence have been offered to support specific associations of arsenate with Fe-oxyhydroxides. First, the specific adsorption of anions induces a greater negative charge on oxide surfaces which results in a shift in the point of zero charge (pH p z c) to a lower pH value. The amount of shift depends on the particular ion sorbed, its concentration, and the nature of the solid surface (Stumm and Morgan, 1981). Several groups have demonstrated p H p z c shifts of several p H units for Fe-oxyhydroxide surfaces with sorbed arsenate, indicating specific interactions (Pierce and Moore, 1982; Fuller et al, 1993; Bowell, 1994; Hsia et al, 1994). Second, adsorption due to electrostatic processes is usually very rapid (i.e., seconds). Equilibration periods on the order of hours observed for As-Fe systems indicate formation of chemical bonds between the As species and the adsorbent (Oscarson et al, 1980; Pierce and Moore, 1982; Fuller et al, 1993). Additional evidence for direct coordination of arsenate has been derived from various other chemical and physical techniques including energy dispersive X-ray analysis (EDAX) and Fourier transform infrared (FTIR) methods (Hsia et al, 1994). Laboratory demonstrations which have established the importance of Fe in As cycling are consistent with other studies of marine and lacustrine 190 sediments. A very convincing data set of six Canadian lakes (pH between 4.0 and 8.4) demonstrated the importance of Fe on the behaviour of As (Belzile and Tessier, 1990). Tightly correlated porewater profiles for all lakes indicated that As was mobilized during the reduction and dissolution of Fe-oxyhydroxides. Moreover, binding constants derived from the field data compared well with those obtained in the laboratory for the adsorption of arsenate on amorphous Fe- oxides. Similar conclusions were drawn for comparable associations of dissolved Fe and As in a Canadian Shield lake (De Vitre et al., 1991). In general, field studies which have alluded to an As-dependence on Mn have not differentiated between Mn and Fe-oxide control. Lake Ohakuri in New Zealand, for example, is a seasonally anoxic lake that receives unusually high As inputs from geothermal sources. The accumulation of dissolved Fe, Mn and predominantly As(III) in the hypolimnion over the course of anoxia was attributed to sediment release via the reductive dissolution of Fe(III)- and perhaps Mn(IV) oxides (Agett and O'Brien, 1985; Agett and Kriegman, 1988). Indeed, the large interfacial gradients supported sediment effluxes for all three elements. Relatively fewer laboratory studies of arsenate sorption onto Mn-oxide phases exist in the literature. Oscarson et al. (1981) demonstrated that synthetically prepared birnessite could efficiently oxidize arsenite to arsenate; the reaction was accompanied by increases in the concentrations of remobilized Mn(II) and K + from the oxide lattice. However, while a significant portion of As(V) was removed from solution by Fe(III) oxide in an identical experiment, very little of the added As was sorbed by Mn02. Since the pzc of birnessite is close to p H 2 (Oscarson et al, 1981b), it was assumed that the high negative charge density at experimental p H (pH=6.9-7.7) inhibited the sorption of the anionic arsenate species (largely H2ASO4" and H A S O 4 "). 191 In a similar experiment with synthesized hydrous MnG^, Takamatsu et al. (1985) demonstrated that arsenate was not appreciably adsorbed in the p H range 6 to 8. However, in the presence of large concentrations (20 mg'L"1) of a divalent cation, a substantial increase in the amount of adsorbed arsenate was observed; the percent adsorbed increased from <20% to 100 % at p H 6 with each of the added trace elements (Mn 2 + , Sr 2 + , Ba 2 + and Ni 2 +). The results were explained in terms of a shift in the surface charge of the Mn-oxide surface in the presence of abundant cations. Divalent cations such as M n 2 + and N i 2 + readily sorb to Mn- oxides, and in the process, displace protons to solution from the oxide surface (Stumm and Morgan, 1981). In this study, the amount of metal adsorbed relative to the extent of proton release was approximately 1:1 on a molar basis. Takamatsu and others (1985) suggested that since the ratio of charge equivalents released to charge equivalents adsorbed was <1, cation adsorption would increase the pzc of the oxide surface, and hence, enhance arsenate sorption. In another experiment, a slurry of oxidized As-bearing surface sediment and seawater was incubated with added plankton material to simulate the sequential depletion of oxidants that occurs during early diagenesis (Edenborn et al., 1986). The sequential release of reduced Mn and Fe species, as well as the timing of SO42" reduction, were thermodynamically consistent with measured redox potentials. The release of As(IU) to the porewaters occurred concomitantly with the release of Fe(II), suggesting strong Fe-As associations. It should also be noted that the As release came after the dissolution of almost all of the reducible Mn. The thermodynamics of reductive dissolution of Fe and Mn oxyhydroxide phases has further implications with respect to their relative influences on the diagenetic chemistry of As. Theoretical examinations of organic matter diagenesis predict that Mn(IV) should be reduced at a higher redox potential 192 than Fe(III) (Froelich et ah, 1979). This implies that in reducing sediments, the addition of manganous ions to porewaters should occur at shallower sediment depths than corresponding ferrous ion inputs. In mildly reducing sediments where thermodynamic gradients are shallow (i.e., the redox horizons for the various interstitial metabolites extend over many centimetres), such spatial decoupling of Fe and Mn reduction can be observed (Edenborn, et ah, 1986). Therefore, even assuming As is associated with Mn oxyhydroxides, it is possible that any release of As to the porewaters upon Mn dissolution could be masked by re-adsorption onto Fe oxyhydroxide phases. Hence, the final expression of As release to the porewater would be more closely associated with increases of dissolved Fe . Such an argument was put forth by Peterson and Carpenter (1986) to account for strong associations observed between Mn and As solid phases in surface sediments, and the concomitant release of dissolved Fe and As in deeper horizons. Because intensely reducing sediments are less likely to foster a Fe/Mn redox decoupling, the mechanism of Fe control described above may be less important in such regimes. In summary, contrasts in the surface charge characteristics of Fe and Mn oxide phases influence the sorption of charged species. The heterogeneity of these particulates makes comparisons of other parameters (e.g., specific surface areas, site densities, etc.) more difficult to assess. In addition, experimental data demonstrate a pH-dependence of arsenate sorption for which both coulombic and specific mechanisms are involved; however, the exact nature and variability of the chemi-sorptive processes are not completely understood. Speciation studies suggest that arsenate is the dominant species in oxic sediment horizons and that arsenite predominates in reducing interstitial waters; the removal of As(III) from solution appears to involve its oxidation by Mn(IV) and probably Fe(III) oxides. Finally, limited experimental data demonstrate weak associations 193 of As with Mn oxides; significant sorption has only been reported for high cationic-strength solutions in which the charge properties of the oxide surface have been modified. Conversely, similar studies offer conclusive evidence that Fe oxide phases exert significant control on the behaviour of As. 5.4.2 The Behaviour of As in Balmer Lake Sediments Seasonal porewater profiles suggest that two principal controls are governing As distributions in Balmer Lake sediments. First, intense remobilization at shallow sediment depths indicate a redox-controlled release from labile, As-bearing particulates; and second, lower concentrations in the deeper horizons suggest the consumption of As as authigenic sulphide phases. Each of these processes will be described in turn. During the well-oxygenated periods of the spring, summer and fall, porewater As profiles are characterized by steep gradients that begin just below the sediment surface (Figs. 4.18-4.22; 4.27). The depths of remobilization typically overlap with corresponding dissolved Mn production horizons. This suggests that a labile portion of As is associated with Mn-oxyhydroxide phases in the upper horizons, both of which follow a redox-pattern of dissolution in the suboxic zone, upward diffusion, and precipitation near the sediment-water interface. Where dissolved Fe gradients are clearly defined, corresponding As slopes are shallower (Fig. 5.12), suggesting that the dissolution Fe solid phases is not responsible for the porewater additions of dissolved As at sediment depths above than the Fe(III) redox boundary. The chemically analogous behaviours of phosphate and arsenate, and hence the potential for biological incorporation, might suggest that the remineralization of organic matter may contribute to the dissolved As enrichments; however, the decoupling between the oxidation of 194 S u m m e r S t a t i o n 2 D i s s o l v e d [ M n ] , ug L " 1 200 400 600 800 1000 0 1000 2000 3000 4000 D i s s o l v e d [ F e ] a n d [ A s ] , ug L " 1 Fig. 5.12. Summer peeper profiles of dissolved Fe, Mn and As for station 2 in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 195 organic material and the dissolution of oxide phases as seen for Zn during the summer period is not clearly evident for As. A linear regression was performed to assess further the dependence of As on the redox behaviour of Mn in Balmer Lake porewaters. Molar concentrations were compared at depths extending from the interfacial horizons to depths of ~20 cm in order to encompass the remobilization depth range (Fig. 5.13). Coefficients of determination (r ) can be interpreted as the proportion of the variation in Y that is "accounted for" by linear regression of Y on X. In other words, an r 2 value of 0.9 means that 90% of the variation in porewater As concentrations can be accounted for by variations in Mn. The r coefficients for the spring, summer and fall average 0.85 (0.70-0.96), indicating that the diagenetic cycling of Mn is significantly affecting the behaviour of As. Corresponding values for Fe and As over these periods are slightly lower (mean = 0.78(0.42-0.96)). The fact that the A s / M n slopes in Fig. 5.13 are not linear demonstrates that mechanisms other than the dissolution of Mn-oxide phases are involved; for each profile, uniform A s / M n ratios of <1 are evident to ~3-4 cm and then generally increase with depth to ~2 in the deeper horizons. The preferential removal of Mn from pore solution or the addition of As (e.g., via the dissolution of Fe-oxide phases) may account for the observed increases. It should be stressed that the weaker correlations between As and Fe do not imply that Mn is exerting greater control on the distribution of As in Balmer Lake porewaters. Instead, the dearth of ferrous-Fe in the shallow porewaters may suggest that Fe is rapidly consumed as authigenic sulphide phases. Indeed, the zones of Fe(III) and SO4 2" reduction are predicted almost to overlap. It is therefore possible that As is remobilized upon the reductive dissolution of Fe(IH) oxides, but Fe consumption from porewaters hinders conclusive interpretations. 196 60 F- | 40 «F 20 Summer T—|—i— I I | I I I | I I I | I—r • St. 2, ^=0.87 • SI. 2, ^=0.96 A 81.1.^-0.92 O St. 1,̂ =0.90 B Fall A o o • A O 25 • 1 1 i 1 1 1 1 i 1 1 1 ' i 1 1 1 1 i 1 • 1 ' i 'd ' • i 1 1 1 1 20 f- D .. n " "5 1 5 f - E =t 10 < 5 •1-1-*-1. .1. i • • 0 ^ D * D St. 1,̂ =0.71 • • St. 5, r'-o.ea • • 1 • • • • 1 • • • • 1 • • • • 1 • • • • 8 1 0 1 2 1 4 1 6 1 8 [Mn], umol L"1 5 1 0 1 5 2 0 2 5 3 0 3 5 -1 [Mn], umol L" Fall ^ 4 0 V 3 0 _J o E 2 0 CO 10 1 1 1 1 1 1 1 1 1 1 1 St. 2, ^=0.94 • - • St. 2, ̂ =0.70 Q • • ^< * 1 1 1 I 1 • • O • O % •S» CP . . i . . . i . . . i • . . . i . . . i . . . : 2 4 6 8 1 0 1 2 14 1 6 [Mn], umol L'1 D 10 o E 3. In < 6 Winter i * * * i * 1 1 i ' * * i * ' * i * 1 1 i 1 1 ' o <o°o @% 0 St. 2, ^=-0.07 • St. 1,̂ =0.50 1 St. 4, =̂0.01 • • • i • • •—i—i—i—i 8 1 2 1 6 2 0 2 4 2 8 -1 [Mn], umol L" 30 O 20 yf 10 < Spring - I — I — I — I — | — I — I — I — I — | — ' — I — I — I — | — I — I I I—|—I—'—i— r - • St. 1.1̂ =0.85 • St. 1.1̂ =0.81 I I I I I I I • • ' I I 1- .1 1 1 1 1 1 1 I I 1 0 1 5 2 0 2 5 [Mn], umol L"1 Fig. 5.13. X - Y scatter plot illustrating the dependence of dissolved As on the distribution of dissolved Mn. The values (molar concentraions) represent a depth interval spanning from ~3 cm above the sediment surface to ~20 cm below. Coefficients of determination (r ) from linear regression analyses for the respective stations are included. 197 Similar correlative examinations performed for the winter period (r values of -0.01-0.5) illustrate much weaker associations of As with Mn (5.12). This decoupling is consistent with the seasonal development of anoxia and the concomitant change in the redox chemistry at the sediment water interface. The sulphate reduction evident in the winter hypolimnion (section 5.3.2) implies that reactive Mn(IV)-oxides were absent in surface deposits during the time of sampling. Hence, the capacity of the surface sediments to sorb upward diffusing As species should have been greatly reduced. Under such conditions, the diagenetic cycle would have been interrupted; both As and Mn species would have diffused freely across the sediment-water interface. Comparison of the porewater plots for Mn (Fig. 5.7) and As (Fig. 5.14) suggests that different mechanisms influence the seasonal distributions of the two elements. Mn porewater maxima, for example, progressively decrease from a winter maximum through the spring, summer and fall. The winter enrichments were suggested to stem from inputs of dissolved Mn to the porewaters from the overlying bottom waters (see section 5.3.3). Conversely, arsenic profiles decrease from maximum porewater values in the summer to minima in late winter (Fig. 5.14). The disparity implies that the non-steady-state inputs of Mn-rich water to the winter bottom waters did not host significantly elevated concentrations of dissolved As. Instead, the distribution of arsenic may reflect seasonal variability in the kinetics of remobilization and consumption of dissolved As in the porewaters. More specifically, variable rates of reductive dissolution of As-bearing oxides, consumption of dissolved As in the oxic horizons, and authigenic sulphide formation at depth, all contribute to the observed seasonality. Large sub-surface porewater gradients in Balmer Lake sediments support year-round upward diffusion of dissolved As towards the sediment-water 198 Fig. 5.14. Seasonal peeper profiles of dissolved As for stations 1 and 2 (A and B, respectively) in Balmer Lake. Replicate samples are represented by double symbols at specific single depths. 199 interface. Rates ranged from 5-175 ug'cm"2'y"1; the large gradients observed in the fall account for the upper limit of this range (Appendix F). These values are much greater than those reported for the Laurentian Trough (0.07-0.24 ug'cm '̂y"1; Belzile, 1988) a temperate estuary (2.2 ug'cm '̂y"1; Widerlund and Ingri, 1995) and various coastal and lake sediments (0.37-2.4 ug'cm '̂y"1; Peterson and Carpenter, 1986). The largely invariant concentrations of As across the sediment-water interface imply that upward migrating species are consumed by Mn and/or Fe oxides in the surface sediments (Figs. 4.18-4.27). However, the reduced sorptive capacity of the surface sediments during periods of bottom water anoxia might suggest that dissolved As species are periodically allowed to diffuse into the bottom waters. In addition, the uncertainty in the position of the sediment-water interface (± 2 cm), and the proximity of As gradients to the sediment surface, suggests that the upper gradient boundary at some sites may extend slightly into the bottom waters. By taking into account a maximum observed As gradient, and neglecting consumption mechanisms of upward migrating species, an upper measure of bottom water input can be estimated. For example, an efflux of 175 ug'cm" 'y~, entering an average water column depth of ~3 m, would result in the addition of -600 ug'L"1 in one year. Although this level represents a maximum, the result indicates that diffusive transport mechanisms have potential significance within the residence time-scale of Balmer Lake (-230 days). The removal of porewater As below sediment depths of 20 cm at some sites in Balmer Lake is probably due to precipitation of As-bearing sulphide phases (Figs. 4.18-4.19; 4.27). Although the precipitation of AS2S3 and mixed Fe- As sulphides have generally not been distinguished, incorporation of As into pyrite has been clearly shown to be an important removal mechanism in several anoxic marine sediments (Huerta-Diaz and Morse, 1992). Similarly, the precipitation of a mixed-sulphide phase was proposed by Peterson and 200 Carpenter (1986) to account for pronounced decreases in dissolved As below the sulphate redox transition in near-shore marine sediments. In summary, the geochemical cycle of Fe has a dominant influence on the diagenetic behaviour of As in marine and lacustrine systems, but the influence of Mn on As behaviour is less clear. The lack of understanding partially stems from the dearth of appropriate data in addition to the inherently complex nature of natural systems. It must be realized that the high ionic strength of the polluted water column makes Balmer Lake unique in comparison to "pristine" water bodies. More specifically, the combined burdens of dissolved metal cations (Mn, Ni, Cu, Zn and Fe) exceed 2.5 mg L"1 in near-interface oxic porewaters. Given such circumstances, sorption of the metals to Mn-oxide phases may shift the surface charge sufficiently to encourage the sorption of anionic As species (Takamatsu et al, 1985). 201 VI. Summary and Conclusions The distribution and behaviour of trace metals in Balmer Lake are influenced by both natural biogeochemical processes and inputs of mining- derived effluents. The formation of ice in early winter in conjunction with the high oxidant demand of the organic-rich sediments fosters the gradual development of stratification over the 5 month period of ice cover. Such seasonality results in spatial and temporal changes in the redox potential in the bottom waters and sediments. Non-steady-state inputs from the tailings circuits hosting elevated levels of trace elements, sulphate, cyanide and ammonium also contribute to the dynamic character of the system. Dissolved and particulate metal concentrations in Balmer Lake are high, and compare with or surpass values reported for other polluted lacustrine environments. Homogenous distributions of all measured parameters evident during the summer, fall and spring are in marked contrast with the stratified water column conditions in late winter. The trace metal, nutrient and sulphate profiles during this period reflect the combined influences of redox-mediated reactions and inputs of dense mine-waters which migrate laterally across the lake bottom. Although the non-steady-state nature of the latter prevented definitive interpretations of the winter water column, it appears that variability with respect to the duration and intensity of reducing conditions in the bottom waters, and the proximity to effluent discharges contribute to the inter-station contrasts. At sediment depths shallower than 10 cm, the deposits in Balmer Lake comprise contributions from both organic matter- and feldspar-rich natural detritus, and carbonate-, chlorite- and metal-rich tailings inputs. Enrichments of solid phase Cu, Ni and Zn at shallow depths were significantly lower immediately adjacent to the tailings discharge than at a site situated over 1 km 202 distant. The distribution likely reflects textural sorting of the tailings material, with the finer grain-size fractions hosting higher metal concentrations. Both remobilization and scavenging mechanisms influence the behaviour of trace metals in the sediments and associated porewaters. Early diagenetic reactions in Balmer Lake sediments, for example, render the deposits below 5 cm depth a significant and permanent sink for dissolved Cu, N i and Zn. Consumption horizons occur at depths consistent with zones of sulphate reduction, suggesting these elements are sequestered as sulphide precipitates. The extent of authigenic sulphide formation likely reflects the relatively recent addition of mining-derived sulphate to the lake waters. Diffusive transport mechanisms have relatively little effect on the lake concentrations of Cu and Ni. Such processes can account for the removal of only 5 and 1% of their respective inventories with respect to a lake residence time of 230 days. The lake sediments provide a relatively greater diffusive sink for Zn; approximately 13% of the Zn inventory is consumed in this manner. Moreover, sulphide precipitation for these three elements contributes little to their overall accumulation rates. Episodic inputs of tailings material and the deposition of metal-rich particulates represent the principle contributors to the total benthic flux of these elements. Inputs of remobilized Zn and Ni from labile particulates result in pronounced spikes near the sediment-water interface. Since such peaks are shown to lie well above the Mn(IV) reduction horizon at some sites, it is suggested that the enrichments stem largely from the oxidation of labile organics, and not from the reduction of Fe and Mn oxides. Contrary to most reports in the literature, the distribution of As appears to be largely dictated by the redox behaviour of Mn. In porewater profiles where the redox horizons for Fe(III) and Mn(IV) are clearly decoupled, dissolved As gradients closely match those of Mn. It is suggested that the large concentrations 203 of divalent cations in the shallow porewaters may significantly alter the surface charge properties of Mn oxides to encourage the sorption of anionic arsenic species. The largely invariant concentrations of As across the sediment-water interface during well-mixed periods imply that upward migrating species are consumed by presumably Mn oxides in the surface sediments. However, the reduced sorptive capacity of the surface sediments during periods of bottom water anoxia might suggest that dissolved As species are periodically allowed to diffuse into the bottom waters. It is a reasonable expectation that once milling effluents are no longer discharged to Balmer Lake, concentrations of Cu and Ni should quickly diminish. This conclusion stems from the predominantly uni-directionally diffusive transport of these elements and the relatively short residence time of the basin. The same can be perhaps implied for Zn, although a fraction of the total benthic Zn-flux will continue to be regenerated back into the bottom waters. In addition, the settling of abundant organic and detrital particulates will continue to provide an effective transport mechanism of trace metals to bottom sediments. The apparent involvement of As in the redox cycling Mn has important ramifications with respect to the future water quality in the basin. The concomitant remobilization of As and Mn at shallow sediment depths produces large porewater gradients that will continue to support the upward migration of dissolved arsenic. However, the significance of potential effluxes to the lake waters and consumption mechanisms in the deeper porewaters remain unclear. 204 VII. Bibliography Agett, J. and Kriegman, M. R. 1988. The extent of formation of arsenic(DI) in sediment interstitial waters and its release to hypolimnetic waters in Lake Ohakuri. Wat. Res. 22(4): 407-411. Agett, J. and O'Brien, G. A. 1985. Detailed model for the mobility of arsenic in lacustrine sediments based on measurements in Lake Ohakuri. Environ. Sci. Technol. 19: 231-238. Aimer, B., Dickson, W., Ekstrom, C. and Hormstrdm, E. 1978. Sulfur pollution and the aquatic ecosystem. In: Sulfur in the Environment,F'art 1, pp. 271-311. Wiley-Interscience. Anderson, L. C. D. and Bruland, K. W. 1991. Biogeochemistry of As in natural waters: the importance of methylated species. Environ. Sci. Technol. 25(3): 420- 427. Andreae, M . 0.1979. Arsenic speciation in seawater and interstitial waters: The influence of biological-chemical interactions on the chemistry of a trace element. Limnol. Oceanogr. 24: 440-452. Arakaki, T. and Morse, J. W. 1993. Coprecipitation and adsorption of Mn(II) with mackinawite (FeS) under conditions similar to those found in anoxic sediments. Geochim. Cosmochim. Acta 57: 9-14. Baccini, P. 1984. Regulation of trace metal concentrations in freshwater systems. In: Circulation of metals in the environment., pp. 239-288. Balistrieri, L. S. and Murray, J. W. 1992. The biogeochemical cycling of trace metals in the water column of Lake Sammamish, Washington: response to seasonally anoxic conditions. Limnol. Oceanogr. 37(3): 529-548. Belzile, N . and Tessier, A. 1990. Interactions between arsenic and iron oxyhydroxides in lacustrine sediments. Geochim. Cosmochim. Acta. 54:103-109. Benjamin, M . M . and Leckie, J. O. 1981. Multiple site adsorption of Cd, Cu, Zn and Pb on amorphous iron oxyhydroxide. /. Colloid Inter. Sci. 79: 209-221. Benjamin, M . M . and Leckie, J. O. 1982. Effects of complexation by CI", S O ^ - , and S2O32- on the adsorption behaviour of Cd on oxide surfaces. Environ. Sci. Technol. 16(3): 162-170. Berner, R. A. 1984. Sedimentary pyrite formation: un update. Geochim. et Cosmochim. Acta 48:1-42. 205 Bischoff, J. L. , Greer, R. E. and Luistro, A. O. 1970. Composition of interstitial waters of marine sediments: temperature and squeezing effects. Science 167: 1245-1246. Bowell, R. J. 1994. Sorption of arsenic by iron oxides and oxyhydroxides in soils. Appl. Geochem. 9: 279-286. Boyle, E. A. and Edmond, J. M . 1977. Determination of copper, nickel, and cadmium in seawater. Anal. Chim. Acta. 91:189-197. Brandl, H . and Hanselmann, K. W. 1991. Evaluation and application of dialysis porewater samplers for microbiological studies at sediment-water interfaces. Aquatic Sci. 53(1): 55-73. Bruland, K. W. 1980. Oceanographic distributions of cadmium, zinc, nickel, and copper in the North Pacific. Earth Planet. Sci. Lett. 47:176-198. Bruland, K. W. 1983. Trace elements in seawater. In: Chemical Oceanography, pp. 157-220. Academic Press. Bruland, K. W. and Franks, R. P. 1983. Mn, Ni, Cu, Zn and Cd in the western North Atlantic. In: Trace Metals in Seawater: NATO Conf. Series 4, pp. 395-414. Bruland, K. W., Franks, R. P., Knauer, G. A. and Martin, J. H . 1979. Sampling and analytical methods for the determination of copper, cadmium, zinc and nickel at the nanogram per liter level in sea water. Anal. Chim. Acta 105: 233-245. Bruland, K. W., Orians, K. J. and Cowen, J. P. 1994. Reactive trace metals in the stratified central North Pacific. Geochim. Cosmochim. Acta 58(15): 3171-3182. Calvert, S. E. 1976. The minearlogy and geochemistry of near-shore sediments. In: Chemical Oceanography (ed. R. Chester) pp. 187-280, Academic Press. Calvert, S. E. and Karlin, R. E. 1991. Relationships between sulphur, organic carbon and iron in the modern sediments of the Black Sea. Geochim. Cosmochim. Acta 55: 2483-2490. Campbell, P. G. C , Lewis, A. G., Chapman, P. M. , Crowder, A. A., Fletcher, W. K., Imber, B., Luoma, S. N. , Stokes, P. M . and Winfrey, M . 1988. Biologically available metals in sediments. National Research Council Canada. NRCC No. 27694. Carignan, R. 1984. Interstitial water sampling by dialysis: methodological notes. Limnol. Oceanogr. 29(3): 667-670. 206 Carignan, R. and Nriagu, J. O. 1985. Trace metal deposition and mobility in the sediments of two lakes near Sudbury, Ontario. Geochim. Cosmochim. Acta 49: 1753-1764. Carignan, R., Rapin, F. and Tessier, A. 1985. Sediment porewater sampling for metal analysis: A comparison of techniques. Geochim. Cosmochim. Acta 49(1): 2493-2498. Carignan, R., St-Pierre, S. and Gachter, R. 1994. Use of diffusion samplers in oligotrophic lake sediments: effects of free oxygen in sampler material. Limnol. Oceanogr. 39(2): 468-474. Carignan, R. and Tessier, A. 1985a. The co-diagenesis of sulphur and iron in acid lake sediments of southwestern Quebec. Geochim. Cosmochim. Acta. 52:1179-1188. Carignan, R. and Tessier, A. 1985b. Zinc deposition in acid lakes: The role of diffusion. Science 228:1524-1526. Carignan, R. and Tessier, A. 1988. The co-diagenesis of sulphur and iron in acid lake sediments of southwester Quebec. Geochim. Cosmochim. Acta 52:1179-1188. Cherry, J. A., Shaikh, A. U., Tallmann, D. E. and Nicolson, R. V. 1979. Arsenic species as an indication of redox conditions in groundwater. /. Hydrol. 43: 373- 392. Chiswell, B. and Mokhtar, M . B. 1986. The speciation of manganese in freshwaters. Talanta 33(8): 669-677. Colley, S. and Thomson, J. 1984. Post-depositional migration of elements during diagenesis in brown clay and turbidite sequences in the North East Atlantic. Geochim. Cosmochim. Acta. 48:1223-1235. Colvine, A. C , Andrews, A. J., Cherry, M. E., Durocher, P., Fyon, A. J., Lavigne, M . J., MacDonald, A. J., Poulsen, K. H. , Springer, J. S. and Troop, D. G. 1984. An integrated model for the origin of Archean lode gold deposits. Ontario Geological Survey. Open-file report 5524. Conroy, N. and Keller, W. 1976. Geological factors affecting biological activity in precambrian shield lakes. Canadian Mineralogist 14: 62-72. Cook, R. B. 1984. Distributions of ferrous iron and sulfide in an anoxic hypolimnion. Can. J. Fish. Aquat. Sci. 41: 286-293. Corfu, F. and Andrews, A. J. 1987. Geochronological constraints on the timing and magmatism, deformation and gold mineralization in the Red Lake greenstone belt, northwestern Ontario. Can. J. Earth Sci. 24:1302-1320. 207 Corfu, F. and Wallace, H . 1986. U-Pb zircon ages for magma in the Red Lake greenstone belt, northwestern Ontario. Can. J. Earth Sci. 23: 27-42. Cornwell, J. C. 1986. Diagenetic trace-metal profiles in Arctic lake sediments. Environ. Sci. Tech. 20: 299-302. Crecelius, E. A. 1975. The geochemical cycling of arsenic in Lake Washington and its relation to other elements. Limnol. Oceanogr. 20(3): 441-451. Cullen, W. R. and Reimer, K. J. 1989. Arsenic speciation in the environment. Chem. Rev. 89: 713-764. Dai, M . , Martin, J. and Cauwet, G. 1995. The significant role of colloids in the transport and transformation of organic carbon and associated trace metals (Cd, Cu and Ni) in the Rhone delta (France). Mar. Chem. 51:159-175. Davis, J. A. and Leckie, J. O. 1978a. Effect of adsorbed complexing ligands on trace metal uptake by hydrous oxides. Environ. Sci. Technol. 12(12): 1309-1315. Davis, J. A. and Leckie, J. O. 1978b. Surface ionization and complexation at the oxide/water interface. II. Surface properties of amorphous iron oxyhydroxide and adsorption of metal ions. /. Colloid Inter. Sci. 67(1): 90-107. Davison, W. 1993. Iron and manganese is lakes. Earth-Sci. Rev. 34:119-163. Davison, W., Grime, G. W. and Woof, C. 1992. Characterization of lacustrine iron sulphide particles with proton-induced X-Ray emission. Limnol. Oceanogr. 37(8): 1770-1777. Davison, W., Lishman, J. P. and Hilton, J. 1985. Formation of pyrite in freshwater sediments: Implications for C/S ratios. Geochim. et Cosmochim. Acta 49:1615-1620. Davison, W. and Woof, C. 1984. A study of the cycling of manganese and other elements in a seasonally anoxic lake. Water research 18(6): 727-734. Davison, W. and Woof, C. 1982. The dynamics of iron and manganese in a seasonally anoxic lake; direct measurement of fluxes using sediment traps. Limnol. Oceanogr. 27(6): 987-1003. De Vitre, R., Belzile, N.and Tessier, A. 1991. Speciation and adsorption of arsenic on diagenetic iron oxyhydroxides. Limnol. Oceanogr. 36(7): 1480-1485. De Vitre, R. and Davison, W. 1993. Manganes particles in fresh water. In: Environmental Particles (edLewis) pp. 440. Boca Raton. 208 De Vitre, R. R., Buffle, J., Perret, D. and Baudat, R. 1988. A study of iron and manganese transformations at the 0 2/S(-H) transition layer in a eutrophic lake (Lake Bret, Switzerland): A multimethod approach. Geochem. Cosmochim. Acta. 52:1601-1613. Devuyst, E. A., Mosoiu, A. and Krause, E. 1982. Oxidizing properties and applications of the S02-air system. At Proc. 3rd Intl Symp. on Hydromet. Eadie, B. J. and Robbins, J. A. 1987. The role of particulate matter in the movement of contaminants in the Great Lakes. In: Sources and Fates of Aquatic Pollutants, pp. 319-364. American Chemical Society. Edenborn, H . M . , Belzile, N. , Mucci, A., Lebel, J. and Silverberg, N. 1986. Observations on the diagenetic behaviour of arsenic in a deep coastal sediment. Biogeochemistry 2: 359-376. EIG 1986. Spring melt studies relating to the operation of an improved control structure for Balmer Lake. Environmental Applications Group. Elderfield, H . 1981. Metal-organic associations in interstitial waters of Narragansett Bay sediments. Amer. J. Sci. 281:1184-1196. Elinder, C. G. 1984. Metabolism and toxicity of metals. In: Changing metal cycles and human health., pp. 265-274. Springer-Verlag. Emerson, S. 1976. Early diagenesis in anaerobic lake sediments: chemical equilibria in interstitial waters. Geochim. Cosmochim. Acta 40:925-934. Ferguson, J. F. and Gavis, J. 1972. A review of the arsenic cycle in natural waters. Wat. Res. 6:1259-1274. Florence, T. M. 1982. The speciation of trace elements in waters. Talanta 29: 345- 364. Forel, F. A. 1892. Le Leman: Monographic Limnologique. Tome I. Geographic, Hydrographie, Geologie, Climatologie, Hydrologie. Forel, F. A. 1895. Le Leman: Monographic Limnologique. Tome II. Mecanique, Hydraulique, Thermique, Optique, Acoustique, Chemie. Forel, F. A. 1904. Le Leman: Monographic Limnologique. Tome III. Biologie, Histoire, Navigation, Peche. Francois, R. 1990. Marine sedimentary humic substances: structure, genesis, and properties. Reviews in aquatic sciences 3(1): 41-80. 209 Froelich, P. N. , Klinkhammer, G. P., Bender, M. L., Luedtke, N. A., Heath, G. R., Cullen, D., Dauphin, P., Hammond, D. and Hartman, B. 1979. Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis. Geochim. et Cosmochim. Acta. 43:1075-1090. Fuller, C. C , Davis, J. A. and Waychunas, G. A. 1993. Surface chemistry of ferrihydrite: Part 2. Kinetics of arsenate adsorption and coprecipitation. Geochim. Cosmochim. Acta. 57: 2271-2282. Gaillard, J. F., Jeandel, C , Michard, G., Nicolas, E. and Renard, D. 1986. Interstitial water chemistry of Villefranche Bay sediments: trace metal diagenesis. Mar. Chem. 18: 233-247. ^ Green, W. J., Ferdmelman, T. G. and Canfield, D. E. 1989. Metal dynamics in Lake Vanda (Wright Valley, Antarctica). Chem. Geol. 76: 85-94. Grim, R. E. 1953. Clay Mineralogy. McGraw-Hill Book Company, Inc. pp. 384. Gulens, J., Champ, D. R. and Jackson, R. E. 1979. Influence of redox environments oh the mobility or arsenic in ground water. In: Chemical Modelling in Aqueous Systems., pp. 81-95. American Chemical Society. Gulson, B. L., Mizon, K. J. and Atkinson, B. T. 1993. Source and timing of gold and other mineralization in the Red Lake area, northwestern Ontario, based on lead-isotope investigations. Can J. Earth Sci. 30: 2366-2379. Hamilton-Taylor, J. and Davison, W. 1994. Redox-driven cycling of trace elements in lakes. In: Lakes II, pp. 129-152. Hamilton-Taylor, J. and Willis, M. 1990. A quantitative assessment of the sources and general dynamics of trace metals in a soft-water lake. Limnol. Oceanogr. 35(4): 840-851. Harrison, G. I. and Morel, F. M . M . 1983. Antagonism between Cd and Fe in the marine diatom Thalassiosira weissflogii. ]. Phycol. 19:495-507. Hesslein, R. H. 1976. An in situ sampler for close interval porewater studies. Limnol. Oceanogr. 21:912-914. Heurta-Diaz, M . A. and Morse, J. W. 1992. Pyritization of trace metals in anoxic marine sediments. Geochim. Cosmochim. Acta. 56: 2681-2702. Higgs, T. W. 1979. Factors affecting natural degradation of free and metal-complexed cyanides from gold mill effluents. Report prepared for Fisheries and Environment Canada, Burlington, Ontario. 210 Higgs, T. W. 1993. Technical guide for the environmental management of cyanide in mining. Mining Association of British Columbia. Hsia, T. H. , Lo, S. L., Lin, C. F. and Lee, D. Y. 1994. Characterization of arsenate adsorption on hydrous iron oxide using chemical and physical methods. Colloids and Surfaces. 85:1-7. Huerta-Diaz, M . and Morse, J. W. 1992. Pyritization of trace metals in anoxic marine sediments. Geochim. Cosmochim. Acta. 56:2681-2702. Hutchinson, G. E. 1957. A treatise on limnology I. Geography, Pysics and Chemistry. Wiley Inc. 1015. IEC 1979. Factors affecting natural degradation of free and metal-cyanide complexed cyanides from gold mill effluent. Report prepared for Fisheries and Environment Canada. Jackson, T. A. 1978. The biogeochemistry of heavy metals in polluted lakes and streams at Flin Flon, Canada, and a proposed method for limitin heavy-metal pollution of natural waters. Environ. Geol. 2(3): 173-189. Jackson, T. A. and Bistricki, T. 1995. Selective scavenging of Cu, Zn, Pb, and As by Fe and Mn oxyhydroxide coatings in lakes polluted with mine and smelter wastes: results of energy dispersive X-ray micro-analysis. /. Geochem. Explor. 52: 97-125. Jaworowski, Z., Bysiek, M . and Kownacka, L. 1981. Flow of metals into the global atmosphere. Geochim. Cos. Acta. 45: 2185-2199. Jenkins, M . C. and Kemp, W. M . 1984. The coupling of nitrification and denitrification in two estuarine sediments. Limnol. Oceanogr. 29: 609-619. Jenne, E. A. 1986. Chemical species in freshwater and terrestrial systems. In: The imprortance of chemical "speciation" in environmental processes., pp. 121-148. Springer-Verlag. Jorgensen, B. B. 1977. The sulphur cycle of a coastal marine sediment (Limfjorden, Denmark). Limnol. Oceanogr. 22: 814-832. Kerner, M and Geisler, C. D. 1995. Dynamics of Cu release during early aerobic degradation in aggregated seston from the Elbe estuary. Mar. Chem. 51:133-144. Kinniburgh, D. G. 1981. Cation adsorption by hydrous metal oxides and clay. In: Adsorption of Inorganics at Solid-Liquid Interfaces, pp. 91-160. Ann Arbor Science. 211 Klinkhammer, G. P. 1980. Early diagenesis in sediments from the Eastern Equatorial Pacific, II. Pore water metal results. Earth and Planetary Science Letters 49: 81-101. Kubota, J., Mills, E. L. and Oglesby, R. T. 1974. Pb, Cd, Zn, Cu and Co in streams and lake waters of Cayuga Lake Basin, New York. Environ. Sci. Technol. 8: 243- 248. Lantzy, R. J. and Mackenzie, F. T. 1979. Atmospheric trace metals: global cycles and an assessment of man's impact. Geochim. Cosmochim. Acta. 43(4): 511-525. Laxen, D. 1985. Trace metal adsorption/coprecipitation on hydrous ferric oxide under realistic conditions. Water Res. 19(10): 1229-1236. Laxen, D. and Chandler, I. M . 1983. Size distribution of iron and manganese species in freshwters. Geochim. Cosmochim Acta. 47: 731-741. LaZerte, B. D. and Burling, K. 1990. Manganese speciation in dilute waters of the precambrian shield, Canada. Wat. Res. 24(9): 1097-1101. Leckie, J. O. 1986. Adsorption and transformation of trace element species at sediment/water interfaces. In: The imprortance of chemical "speciation" in environmental processes, pp. 237-254. Springer-Verlag. Li, Y. H . and Gregory, S. 1974. Diffusion of ions in sea water and in deep-sea sediments. Geochmim. Cosmochim. Acta. 38: 703-714. Manheim, F. T. 1970. The diffusion of ions in unconsolidated sediments. Earth Planet. Sci. Lett. 9: 307-309. Martens, C. S. and Klump, J. V. 1980. Biogeochemical cycling in an organic-rich coastal marine basin: 1. Methane sediment-water exchange processes. Geochim. Cosmochim. Acta 44:471-490. Masala, C. 1995. Water and effluent management of Balmer Lake tailings management. M.Sc. Thesis, University of British Columbia. Matisoff, G., Lindsay, A. PL, Matis, S. and Soster, F. M . 1980. Trace metal mineral equilibria in Lake Erie sediments. /. Great Lakes Res. 64(4): 353-366. Mayer, L. M . 1976. Chemical water sampling in lakes and sediments with dialysis bags. Limnol Oceanogr. 21: 909-912. McKee, J. D., Wilson, T. P. and Long, D. T. 1989a. Geochemical partitioning of Pb, Zn, Cu, Fe and Mn across the sediment-water interface in large lakes. / . Great Lakes Res. 15(1): 46-58. 212 McKee, J. D., Wilson, T. P. and Long, D. T. 1989b. Porewater profiles and early diagenesis of Mn, Cu, and Pb in sediments from large lakes. / . Great Lakes Res. 15(1): 68-83. McKelvey, B. A. 1994. The marine geochemistry of zirconium and hafnium. Ph.D. Thesis, University of British Columbia. Middleburg, J. J. 1989. A simple rate model for organic matter decomposition in marine sediments. Geochim. et Cosmochim. Acta. 53:1577-1581. Morel, F. M . M . and Hudson, R. J. M . 1985. The geobiological cycle of trace elements in aquatic systems: Redfield revisited. In: Chemical Processes in Lakes, pp. 251-282. Wiley-Interscience. Morfett, K., Davison, W. and Hamilton-Taylor, J. 1988. Trace metal dynamics in a seasonally anoxic lake. Environ. Geol. Water Sci. 11(1): 107-114. Morse, J. W. and Arakaki, T. 1993. Adsorption and coprecipitation of divalent metals with mackinawite (FeS). Geochim. Cosmochim. Acta. 57: 3635-3640. Morse, J. W., Millero, F. J., Cornwell, J. C. and Rickard, D. 1987. The chemistry of hydrogen sulfide and iron sulfide systems in natural waters. Earth-Science Reviews 24:1-42. Mortimer, C. H . 1941. The exchange of dissolved substances between mud and water in lakes. /. Ecol. 29: 280-329. Mortimer, C. H. 1942. The exchange of dissolved substances between mud and water in lakes (Parts III and IV, summary and references). /. Ecol. 30:147-201. Mortimer, C. H. 1951. The use of models in the study of water movement in stratified lakes. Verh. Int. Ver. Limnol. 11: 254-260. Murray, J. W. 1987. Mechanisms controlling the distribution of trace elements in oceans and lakes. In: Sources and Fates of Aquatic Pollutants, pp. 91-130. American Chemical Society. Nriagu, J. O. 1989. A global assessment of natural sources of atmospheric trace metals. Nature 338:47-49. Nriagu, J. O., Lawson, G., Wong, H. K. T. and Azcue, J. M . 1993. A protocol for minimizing contamination in the analysis of trace metals in Great Lakes waters. /. Great Lakes Res. 19(1): 175-182. 213 Nriagu, J. O. and Soon, Y. K. 1985. Distribution and isotopic composition of sulphur in lake sediments of northern Ontario. Geochim. Cosmochim. Acta. 49: 823-834. Nriagu, J. O., Wong, H. K. T. and Coker, R. D. 1981. Particulate and dissolved trace metals in Lake Ontario. Water Res. 15: 91-96. Nriagu, J O, Wong, H . K. T. and Coker, R. D. 1982. Deposition and chemistry of pollutant metals in lakes around the smelters at Sudbury, Ontario. Environ. Sci. Technol. 16: 551-560. Oscarson, D. W., Huang, P. M. , Defosse, C. and Herbillon, A. 1981a. Oxidative power of Mn(IV) and Fe(III) oxides with respect to As(III) in terrestrial and aquatic environments. Nature 291:50-51. Oscarson, D. W., Huang, P. M . and Liaw, W. K. 1980. The oxidation of arsenite by aquatic sediments. /. Environ. Qual. 9(4): 700-703. Oscarson, D. W., Huang, P. M. and Liaw, W. K. 1981b. Role of manganese in the oxidation of arsenite by freshwater lake sediments. Clays Clay Minerals 29: 219- 225. Papelis, C , Hayes, K. F. and Leckie, J. O. 1988. HYDRAQL: A program for the computation of chemical equilibrium composition of aqueous batch systems including surface-complexation modeling of ion adsorption at the oxide/solution interface. Stanford Univ. Dep. Civ. Eng. Tech. Rep. 306. Parsons, T. R., Maita, Y. and Lalli, C. 1984. A manual of chemical and biological methods for seawater analysis. Pergamon Press, pp. 173. PDI, 1986. Spring melt studies relating to the operation of an improved control structure for Balmer Lake. Environmental Applications Group Ltd. PDI 1990. Balmer Lake Management program. Water quality and sediment coring results. Pedersen, T. F. 1983. Dissolved heavy metals in a lacustrine mine tailings deposit, Buttle Lake, British Columbia. Mar. Poll. Bull. 14: 249-254. Pedersen, T. F. 1985. Early diagenesis of copper and molybdenum in mine tailings and natural sediments in Rupert and Holberg Inlets, British Columbia. Can. Jour. Earth. Sci. 22:1474-1484. Pedersen, T. F., Malcolm, S. J. and Sholkovitz, E. R. 1985. A lightweight gravity corer for undisturbed sampling in soft sediments. Can. J. Earth Sci. 22:133-135. 214 Pedersen, T. F., Mueller, B. and McNee, J. J. 1993. The early diagenesis of submerged sulphide-rich mine tailings in Anderson Lake, Manitoba. Can. ]. Earth Sci. 30:1099-1109. Pedersen, T. F., Mueller, B. and Pelletier, C. A. 1990. On the reactivity of submerged mine tailings in fjords and a lake in British Columbia. At G A C / M A C Symposium on Acid Mine Waste, Vancouver, B.C. Pedersen, T.F. and Losher A. J. 1988. Diagenetic processes in aquatic mine tailings deposits in British Columbia., pp. 238-258. Springer-Verlag. Perry, K. A. 1993. The chemical limnology of two meromictic lakes with emphasis on pyrite formation. Ph.D. Thesis, University of British Columbia. Peterson, M . L. and Carpenter, R. 1986. Arsenic distributions in porewaters and sediments of Puget Sound, Lake Washington, the Washington coast and Saanich Inlet, B.C. Geochim. Cosmochim. Acta. 50: 353-369. Pierce, M . L. and Moore, C. B. 1982. Adsorption of arsenite and arsenate on amorphous iron hydroxide. Water. Res. 16:1247-1253. Postma, D. and Jakobsen, R. 1996. Redox zonation: Equilibrium constraints on the Fe(IH)/S04-reduction interface. Geochim. Cosmochim. Acta 60(17): 3169-3175. Raiswell, R. and Berner, R. A. 1985. Pyrite formation in euxinic and semi-euxinic sediments. Amer. J. Sci. 285: 710-724. Redfield, A. C. 1958. The biological control of chemical factors in the environment. Amer. Sci. 46: 206-226. Reimer, K. J. and Thompson, J. A. J. 1988. Arsenic speciation in marine interstitial water. The occurrence of organoarsenicals. Biogeochemistry.6: 211-237. Reynolds, G. L. and Hamilton-Taylor, J. 1992. The role of planktonic algae in the cycling of Zn and Cu in a productive soft-water lake. Limnol. Oceanogr. 37(8): 1759-1769. Rosenthal, Yair, Lam, P. and Boyle, E. A. 1995. Authigenic cadmium enrichments in suboxic sediments: precipitation and post-depositional remobilization. Earth Planet. Sci. Lett. 132: 99-109. Rudd, J. W. M. , Kelly, C. A. and Furutani, A. 1986. The role of sulphate reduction in the long term accumulation of organic and inorganic sulphur in lake sediments. Limnol. Oceanogr. 31(6): 1281-1291. 215 Salomons, W. 1995. Environmental impact of metals derived from metal mining: processes, predictions, prevention. /. Geochem. Explor. 52: 5-23. Sanders, J. G. 1983. Role of marine phytoplankton in determining the chemical speciation and biogeochemical cycling of arsenic. Can. J. Fish. Aquat. Sci. 40(Suppl. 2): 192-196. Schaule, B. K. and Patterson, C. C. 1981. Lead concentration in the Northeast Pacific: evidence for global anthropogenic perturbations. Earth Planet. Sci. Lett. 54: 97-116. Schindler, P. W. 1981. Surface complexes at oxide-water interfaces. In: Adsorption of Inorganics at Solid-Liquid Interfaces, pp. 1-50. Ann Arbor Science. Sclater, F. F., Boyle, E. and Edmond, J. M . 1976. On the marine geochemistry of nickel. Earth Planet. Sci. Lett. 31:119-128. Semkin, R. G. and Kramer, J. R. 1976. Sediment geochemistry of Sudbury-area lake. Can. Mineral. 14: 73-90. Shaw, T. J., Gieskes, J. M . and Jahnke, R. A. 1990. Early diagenesis in differing depositional environments: the response of transition metals in pore water. Geochim. Cosmochim. Acta 54:1233-1246. Shaw, T. J., Sholkovitz, E. R. and Klinkhammer, G. 1994. Redox dynamics in the Chesapeake Bay: The effect on sediment/water uranium exchange. Geochim. Cosmochim. Acta 58(14): 2985-2995. Sholkovitz, E. 1973. Interstitial water chemistry of the Santa Barbara Basin sediments. Geochim. et Cosmochim. Acta 37: 2043-2073. Sholkovitz, E. R. 1979. Chemical and physical processes controlling the chemical composition of suspended material in the River Tay estuary. Estuar. Coastal. Mar. Sci. 8:523-545. Sholkovitz, E. R. 1985. Redox-related geochemistry in lakes: alkali metals, alkaline earth elements, and ^7QS In: Chemical Processes in Lakes., pp. 119-142. Wiley-Interscience. Sigg, L. 1985. Metal transfer mechanisms in lakes: the role of settling particles. In: Chemical Processes in Lakes., pp. 283-310. Wiley-Interscience. Sigg, L., Johnson, A. and Kuhn, A. 1991. Redox conditions and alkalinity generation in a seasonally anoxic lake (Lake Greifen). Mar. Chem. 36: 9-26. 216 Sigg, L., Sturm, M . and Kistler, D. 1987. Vertical transport of heavy metals by settling particles in Lake Zurich. Limnol Oceanogr. 32:112-130. Sorenson, J. and Jorgensen, B. B. 1987. Early diagenesis in sediments from Danish coastal water: microbial activity and Mn-Fe-S geochemistry. Geochim. Cosmochim. Acta. 51:1583-1590. Stadelmann, P. 1971. Stickstoffkreislauf und Primarproduktion im mesotrophen Vierwaldstattersee (Hower Bucht) und im eutrophen Rotsee, mit besonderer Berucksichtigung des Nitrats als limitierenden Faktors. Schweiz. Z. Hydrol. 33:1- 65. Stumm, W. and Morgan, J. J. 1981. Aquatic Chemistry, pp. 583. Wiley- Interscience. Stumm, W. and Sulzberger, B. 1992. The cycling of iron in natural environments: considerations based on laboratory studies of hetergeneous redox processes. Geochim. Cosmochim. Acta. 56: 3233-3257. Suess, E. 1979. Mineral phases formed in anoxic sediments by microbial decomposition of organic matter. Geochim. Cosmochim. Acta£3: 339-352. Takamatsu, T., Kawashima, M . and Koyama, M . 1985. The role of manganous- rich hydrous Mn-oxide in the accumulation of arsenic in lake sediments. Wat. Res. 19(8): 1029-1032. Taylor, R. M. 1987. Non-silicate oxides and hydroxides. In: Chemistry of Clays and Clay Minerals, pp. 129-202. Wiley-Interscience. Tessier, A. and Campbell, P. G. C. 1988. Partioning of Trace Metals in Sediments. In: Metal speciation: theory, analysis, and application., pp. 183-200. Lewis Publishers. Tessier, A., Carignan, R., Dubreuil, B. and Rapin, F. 1989. Partitioning of Zn between the water column and the oxic sediments in lakes. Geochim. Cosmochim. Acta. 53:1511-1522. Tessier, A., Fortin, D., Belzile, N., De Vitre, R. R. and Leppard, G. G. 1996. Metal sorption to diagenetic iron and manganese oxyhydroxides and associated organic matter: narrowing the gap between field and laboratory measurements. Geochem. Cosmochim. Acta 60(3): 387-404. Tessier, A., Rapin, F. and Carignan, R. 1985. Trace metals in oxic lake sediments: possible adsorption onto iron oxyhydroxides. Geochim. Cosmochim. Acta49:183- 194. 217 Tipping, E. 1981. The adsorption of aquatic humic substances by iron oxides. Geochim. Cosmochim. Acta^S: 191-199. Tipping, E. and Cooke, D. 1982. The effect of adsorbed humic substances on the surface charge of geothite. Geochim. Cosmochim. Acta. 46: 75-80. Tipping, E., Woof, C. and Cook, D. 1981. Iron oxide from a seasonally anoxic lake. Geochim. Cosmochim. Acta. 45:1411-1419. Tsunogai, S., Yonemaur, I. and Kusakabe, M . 1979. Post depositional migration of Cu, Zn, Ni, Co, Pb and Ba in deep sea sediments. Geochim.}. 13: 239-252. Turner, D. R. 1984. Relationships between biological availability and chemical measurements. In: Circulation of metals in the environment., pp. 137-164. Van der Weijden, C. H . 1992. Early diagenesis and marine pore water. In: Diagenesis III, pp. 13-133. Elsevier. Vickell, G. A., Davies, D. T. and Gee, R. 1989. Hydrogen peroxide treatment of gold mill wastes. At Gold Mining Effluent Seminar Proceedings., Feb. 15-16, 1989, Vancouver, B. C. Vuceta, J and Morgan, J. J. 1978. Chemical modelling of trace metals in fresh waters: role of complexation and adsorption. Environ. Sci. Technol. 12(12): 1302- 1308. Westerlund, S. F. G., Aderson, L. G., Hall, P. O. J. and Iverfeldt, A. 1986. Benthic fluxes of cadmium, copper, nickel, zinc and lead in the coastal environment. Geochim. Cosmochim. Acta 50:1289-1296. Wetzel, R. G. 1975. Limnology. W. B. Saunders Company, pp. 741. Wetzel, R. G., Rich, P. H. , Miller, M . C. and Allen, H. L. 1975. Wintergreen Lake: a study in hypereutrophy. Arch. Hydrobiol. 73: 31-56. Widerlund, A. and Ingri, J. 1995. Early diagenesis of arsenic in sediments of the Kalix River estuary, northern Sweden. Chem. Geol. 125:185-196. Williams, T. M . 1992. Diagenetic metal profiles in recent sediments of a Scottish freshwater loch. Environ. Geol. Water. Sci. 20(2): 117-123. Wood, J. M . 1974. Biological cycles for toxic elements in the environment. Science 183:1049-1052. 218 Young, B. L. and Harvey, H. H. 1992. The relative importance of manganese and iron oxides and organic matter in the sorption of trace metals by surficial lake sediments. Geochim. Cosmochim. Acta 56:1175-1186. Zoller, W. H . 1984. Anthropogenic perturbation of metal fluxes into the atmosphere. In: Changing metal cycles and human health., pp. 27-42. Springer- Verlag. 219 VIII. APPENDICES Appendix A . Typical ICP/MS operating conditions. Parameter Typical value RF power (kW) 1.35 Argon gas flow rate (L/min): Cooling gas 13.83 Auxiliary 0.983 Nebulizer 1.032 Sampling Position (mm from load coil) 13 Sampler Cone (nickel) from orifice (mm) 1.0 Skimmer Cone (nickel) from orifice (mm) 0.8 Ion lens settings (volts): Extraction -118 Collector -2 L I -2 L2 -45 L3 +5 L4 -45 Pole Bias -3 Operating Pressure (mbar): Expansion 2.2 Interface < x l O - 4 Analyzer 2 x l O " 6 Resolution: A M 5.3 220 Appendix B. Quality Assessment/Quality Control (QA/QC) The QA/QC program included the analysis of quality assurance samples to assess the precision and accuracy of measurements of the various parameters. Several types of samples were included: laboratory blanks (DDW with added reagents), field process blanks, laboratory replicates and certified reference materials. Overall, the results for the QA/QC samples are indicative of good reproducibility, accuracy and contamination control. I. Dissolved Trace Metals Laboratory blanks for all trace metals were very low, and indicates good contamination control. Blank values were used to determine the detection limits which are reported as six times the standard deviation of the blank results. Table A l . Detection limits and machine precision for trace metal determinations. Detection limits represent six standard deviations of 10 process blank values. Element Method Detection Limit Precision (ppb) (%) Fe G F A A S 1.8 2.0 M n I C P / M S 0.06 2.5 M I C P / M S 0.10 2.7 Cu I C P / M S 0.14 2.5 Z n I C P / M S 0.84 2.6 A s I C P / M S 0.22 2.0 Pb I C P / M S 0.50 2.0 221 Table A2. Analytical results for field blanks. Values represent the average of four samples. Fe Mn Ni Cu Zn As Pb Water column filter blank Core porewater blank <1.8 <0.06 <0.1 <0.14 1.1 <0.22 <0.5 <1.8 <0.06 <0.1 <0.14 <0.84 <0.22 <0.5 Table A3 . Expected (Exp) and Observed (Obs) values for standard reference materials supplied by the National Research Council (SLRS) and the National Water Research Institute (TM Series). Observed values correspond to the mean of 10 measurements. The error in the observed values represents two standard deviations. Element SLRS-2 TM-21 TM-02 Exp Obs Exp Obs Exp Obs Fe 50.0 ± 1 6 50.3±2.0 Mn 10.1 ±0 .3 9.46±0.38 6.2 ± 2.0 6.1±0.14 M 1.03 ±0.10 1..17±0.10 6.6 ± 2.4 6.47± 0.30 Cu 2.76 ±0 .17 2.80±0.16 7.6 ±2 .2 6.93±0.28 Zn 3.33 + 0.15 3.63*0.32 7.5 ± 2.2 7.80*0.40 As 0.77 ±0 .09 0.81±0.14 Pb 0.129*0,01 0.148±0.01 5,5 ± 1 , 6 5,20±0,20 222 II. Particulate Trace Metals Table A4. Trace metal determinations of the sediment reference material BCSS. The analysis of seven replicates included digestion followed by ICP/MS (Mn, N i , Cu, Zn, As and Pb) and G F A A S (Fe). The error in the observed values represents ± two standard deviations. Digestion Fe Mn M Cu Zn As Pb 1 32,755 225 78 21 101 19 23 2 37,523 204 69 21 100 17 23 3 33,290 203 64 20 97 17 24 4 177 65 20 95 17 22 5 216 65 18 96 17 23 6 28,073 152 60 21 89 16 21 7 34,273 202 64 19 104 17 24 Mean 33,183 ± 6,800 197 ± 4 6 67 ± 1 1 20 ± 2 98 ± 12 17 ± 2 23 ± 2 Expected 32,900 ± 980 229 ± 15 55 ± 4 19 ± 3 119 ± 1 2 11 ± 2 23 ± 3 Table A5 . Laboratory blanks for digested polycarbonate filters. A l l values are expressed as p.g L"'. Digestion Fe Mn Ni Cu Zn As Pb 1 0.02 0.02 0.01 0.01 0.25 0.04 0.00 2 0.02 0.01 0.01 0.00 0.52 0.00 0.00 3 0.07 0.03 0.06 0.00 1.13 0.13 0.03 4 0.06 0.02 0.01 0.00 4.85 0.09 0.00 5 0.06 0.02 0.02 0.00 3.92 0.10 0.01 6 0.07 0.02 0.09 0.08 0.68 0.10 0.23 7 0.02 0.03 0.06 0.09 2.42 0.10 0.01 Mean 0.05 0.02 0.04 0.03 1.97 0.08 0.04 223 III. Nitrate and Sulphate Table A 6 . Detection limits and machine precision for nitrate and sulphate determinations. Detection limits represent six standard deviations of 10 process blank values. Anion Method Detection Limit Precision NO3- IC 0.1 pmol-L~l 1.0% S O 4 2 - IC 0.1 umol-L"! 1.0% Table A7. Expected (Exp) and Observed (Obs) values for standard reference materials supplied by the National Research Council and the National Water Research Institute. Observed values correspond to the mean of 10 measurements. The error in the observed values represents two standard deviations. Ani-04 Ion-96 Dionex N0 3 - 4 Exp Obs 145 ± 43 141 ± 6 1170 ± 110 1126 ± 66 Exp Obs 335 ± 3 8 347 ± 24 1044 ± 1 2 2 1084 ± 8 4 Exp Obs 1629 ± 10 1665± 100 1550 ± 10 1626 ± 120 224 Appendix C. Porewater data Porewater concentrations of dissolved Fe, Mn, Ni , Cu, Zn, As, Pb, N H 4 + , N0 3 " and S0 4 2 " in Balmer Lake for the summer, fall, winter and spring field seasons. A l l values are reported ugL"1, except for N H 4 + (umolL 1) N0 3 " (pmolL 1) and S0 4 2 " (mmolL 1 ). Positive values for sampled horizons (cm) for both peepers and cores (CI, C2 and C3) refer to depths above the sediment-water interface. Note, that due an angled peeper insertion, the depth scales for the metal and nutrient data differ for the last spring peeper profile. 225 Appendix C cont. Summer-Station 2 Depth Mn Fe NI Cu Zn As Pb NH4 N03 S04 (cm) C9L' ugL' pgL' |igL' ugL1 f»molL>molL"'mmoiL" 35.5 45 415 173 45 0.5 48 350 3.4 30.3 429 174 49 0.4 23.9 281 57 417 169 43 279 0.2 18.7 421 175 51 1.0 79 354 3.4 18.7 341 3.3 14.8 285 35 403 289 0.0 90 348 3.1 13.5 45 408 163 48 0.5 ^ 3 37 413 163 62 0.4 11.0 44 402 160 53 0.3 98 342 3.3 9.7 60 417 161 63 0.2 8.4 290 39 405 162 47 297 0.3 7.1 42 408 161 49 0.5 87 354 3.4 7.1 57 350 3.3 5.8 297 54 407 163 51 301 0.9 5.2 392 153 42 0.2 4.5 47 410 162 45 0.3 4.5 33 408 163 46 0.2 3.2 306 34 413 167 51 309 0.4 79 346 3.2 2.6 40 77 373 3.3 1.9 71 404 155 52 0.6 77 358 3.3 1.3 90 424 158 49 0.3 80 0.6 306 66 421 169 59 320 73 433 3.3 0.0 313 57 402 157 55 0.3 70 371 3.2 0.0 374 3.3 -0.6 44 64 400 3.2 -1.3 72 381 3.3 -1.9 338 49 407 148 63 390 0.3 128 329 3.2 -1.4 380 396 132 73 591 0.4 209 194 -3.2 412 75 437 139 107 567 0.2 269 180 3.1 -3.9 460 366 108 99 983 0.4 348 68 3.1 -4.5 494 100 429 120 124 992 0.4 354 92 3.0 -4.5 91 -5.8 680 110 244 50 69 2430 0.8 -7.1 820 490 154 15 72 4080 1.1 558 51 2.6 -8.4 940 1050 110 5 30 4960 0.5 574 77 2.6 -9.7 986 1400 101 7 30 5044 0.6 660 35 2.4 -11.0 961 1890 96 20 40 4862 1.9 618 41 2.2 -14.8 979 1510 76 4 26 3896 0.9 622 36 1.8 -18.7 908 1410 59 4 20 3028 1.0 642 15 1.2 -18.7 14 1.2 -22.6 881 1630 49 2 13 2561 0.6 655 23 1.0 -26.4 908 2150 54 1 17 2828 0.8 671 15 0.8 -31.6 888 2860 56 1 13 2405 1.0 647 14 0.6 -36.8 909 3240 67 11 27 3175 1.5 622 26 0.6 -36.8 27 0.6 -41.9 986 3850 86 6 33 3854 1.1 666 13 0.3 Summer-Station 2 Depth Mn Fe NI Cu Zn As Pb NH4 N03 S04 (cm) eg!-' cgL' egL' figL"1 pmotL''pmolL'mmolL 31.0 289 65 421 164 47 285 0.5 72 344 3.3 27.1 25.8 30 75 24.5 23.2 21.9 20.6 19.4 288 26 415 161 45 266 0.4 366 3.3 19.4 285 414 154 42 278 0.4 341 3.3 18.1 16.8 15.5 14.2 72 10.3 298 60 430 171 52 296 0.3 78 343 3.2 9.7 42 7.7 54 7.1 33 6.5 291 36 421 151 46 285 0.4 100 340 3.3 5.8 37 5.2 35 4.5 34 3.9 48 3.2 32 2.6 287 40 423 160 48 286 0.4 105 343 3.3 1.3 299 56 293 0.0 304 47 412 156 61 297 0.4 90 -1.3 361 57 485 147 86 372 0.3 93 368 3.3 -1.9 366 38 460 143 125 384 0.4 95 360 3.3 -2.6 421 79 484 135 90 523 0.5 87 401 3.2 -2.6 414 489 131 88 530 0.5 385 3.1 -3.2 446 43 461 123 143 446 0.5 85 364 3.2 -3.9 451 58 427 112 88 451 0.4 85 327 3.1 -4.5 506 60 322 88 100 885 0.6 67 344 2.3 -5.2 517 78 295 81 52 1049 0.6 128 179 2.8 -5.8 549 197 218 51 60 1502 0.5 272 194 2.8 -8.5 756 381 209 28 34 2135 0.5 333 87 2.6 -7.1 757 744 171 18 45 2536 0.5 360 80 2.4 -7.7 858 837 167 13 28 3003 0.5 354 35 2.5 -8.4 844 1598 142 10 32 3319 0.4 456 71 2.3 -9.0 916 1457 123 6 41 3525 0.3 441 52 2.3 -10.3 908 1490 116 18 20 3985 0.2 439 56 2.0 -10.3 918 113 3895 54 2.0 -11.6 891 1677 111 7 23 3794 0.3 510 59 2.1 -12.9 855 1544 105 6 21 3798 0.1 580 54 1.9 -14.2 494 29 1.7 -15.5 869 1688 105 29 38 3725 0.2 609 30 1.7 -19.4 806 1460 79 7 22 2946 0.3 628 32 1.4 -23.2 800 2056 72 2 23 2808 0.5 636 10 1.0 -23.2 812 2767 14 1.0 -27.1 779 1464 72 1 26 2099 0.4 686 14 0.8 -31.0 764 1535 67 1 23 1870 0.3 650 9 0.6 -36.1 814 2923 59 1 16 1977 1.2 610 8 0.5 -41.3 845 3175 73 0 22 2208 0.2 588 9 0.4 -41.3 828 74 2192 8 0.4 -46.4 909 4230 129 12 79 2355 0.5 606 4 0.2 226 Appendix C cont. Summer-Station 1 Depth Mn NI Cu Zn As Pb Fe (cm) (igL1 M9L' pgL' C9L' MSL' 34.8 237 423 154 57 285 0.0 36 33.5 32.3 31.0 29.7 237 427 150 44 287 0.0 31 29.7 243 441 28 23.2 242 426 155 49 295 0.0 32 23.2 154 47 292 18.1 242 421 153 48 294 0.0 31 18.1 35 14.2 256 455 167 72 316 0.1 342 14.2 330 12.9 253 428 157 67 298 0.1 191 12.9 178 11.6 253 434 158 51 304 0.0 151 11.6 134 10.3 244 437 162 50 307 0.0 76 9.0 237 428 160 50 310 0.0 41 9.0 37 7.7 234 438 155 48 295 0.0 33 7.7 31 6.5 245 428 164 45 317 0.0 35 5.2 239 443 155 52 302 0.1 35 4.5 233 432 156 47 303 29 4.5 163 48 311 0.0 3.9 240 424 149 46 301 0.0 31 3.2 234 431 161 45 313 29 3.2 230 438 0.0 2.6 229 418 150 44 304 0.0 31 1.9 224 429 159 51 294 0.0 29 1.3 242 443 151 47 300 0.0 30 0.6 256 434 156 50 288 0.0 29 0.0 260 397 144 60 298 0.0 34 -0.6 34 -0.6 333 469 143 60 315 0.0 -1.3 344 376 114 54 335 0.0 35 -1.3 -1.9 395 382 114 58 481 0.0 31 -1.9 104 59 433 0.0 31 -2.6 505 213 690 0.0 -2.6 513 220 51 34 689 162 -3.2 537 205 42 30 718 0.0 75 -3.2 65 -3.9 670 106 12 18 1875 0.0 475 -3.9 462 -4.5 613 118 13 20 1481 0.1 300 -4.5 267 -5.2 752 86 8 15 2663 0.0 1023 -5.2 980 -6.5 765 85 6 13 2809 0.0 1307 -6.5 84 6 14 0.0 -7.7 768 81 6 15 3086 0.1 1423 -9.0 787 75 7 26 2720 0.1 1445 -9.0 1365 -10.3 793 106 49 57 2685 0.3 1845 -11.6 799 78 17 21 2419 0.2 1536 -12.9 812 2501 0.1 1620 -12.9 815 86 16 39 2561 1596 -14.2 803 88 13 23 2598 0.1 -14.2 86 13 22 0.1 1633 -15.5 810 84 12 26 2871 0.1 1510 -16.8 819 77 11 24 2757 0.1 1467 -19.4 764 70 14 24 2794 0.1 1619 -19.4 -21.9 740 2673 0.1 1792 -21.9 739 67 6 17 2705 -23.2 -24.5 701 67 6 17 2066 0.1 1442 -24.5 1523 -28.4 688 56 2 14 1424 0.1 1775 -28.4 57 2 26 0.0 -33.5 693 54 2 14 1391 0.1 2482 -37.4 699 1390 0.1 2173 -37.4 687 67 5 21 1375 -42.6 740 122 86 82 1694 0.6 3848 -42.6 3963 Summer-Station 1 N04 S04 NH4 Depth Mn Ni umolL" mmolL jimolL' (cm) ugL' (igL 350 3.4 63 30.3 291 459 29.0 250 392 23.9 22.6 276 428 360 3.5 77 22.6 310 479 17.4 3.4 17.4 272 420 16.1 354 3.4 77 14.8 13.5 312 467 401 3.4 85 12.9 12.3 288 448 11.6 11.0 296 447 10.3 9.7 270 426 8.4 305 472 390 3.5 91 7.7 303 451 7.7 322 463 7.1 355 581 6.5 414 3.4 83 6.5 337 589 111 5.8 355 602 110 5.2 329 489 4.5 297 450 447 3.3 115 3.9 295 439 108 3.9 305 460 3.2 279 412 403 3.4 107 2.6 364 514 115 2.6 100 1.9 356 510 596 3.4 101 1.3 307 431 396 3.4 100 0.6 297 414 406 3.4 87 0.0 301 414 412 3.4 0.0 365 3.5 130 -0.6 324 422 356 -1.3 397 511 360 3.5 185 -1.3 384 457 -1.9 440 488 197 3.6 298 -2.6 443 394 199 3.5 -3.2 599 368 149 3.6 397 -3.9 637 304 -4.5 702 169 53 3.4 450 -5.2 712 176 -5.2 44 447 -5.8 838 126 3.2 -7.1 861 102 27 3.7 511 -7.1 861 105 27 -8.4 890 88 611 -8.4 984 -9.7 76 29 3.1 655 -9.7 1115 127 13 3.6 624 -11.0 3.6 -11.0 1017 79 15 3.4 682 -12.3 14 3.3 684 -12.3 1085 76 31 3.2 660 -13.5 1012 76 31 -14.8 1101 73 32 3.0 642 -16.1 -16.1 570 -17.4 31 2.6 570 -18.7 1106 77 33 2.2 646 -20.0 2.2 -22.6 959 -23.9 1038 84 -27.7 987 60 28 1.8 565 -29.0 -32.9 -34.2 844 66 28 1.2 567 -38.1 975 -39.3 48 23 0.8 593 -41.9 71 17 0.6 553 227 Cu Zn As Pb Fe N04 S04 NH4 pgL' egL' cgL' cgL' &*gL"1 (*molL"'mmotL>molL" 166 57 276 0.2 59 71 0.1 32 355 87 161 53 290 0.1 159 47 281 0.6 44 392 3.3 161 52 279 32 349 3.3 89 159 48 284 0.2 364 3.5 218 83 339 2.2 366 3.4 92 162 48 285 0.2 43 154 46 278 0.2 32 162 49 288 0.2 36 352 3.4 93 169 49 287 0.2 161 50 290 34 159 47 275 0.2 166 49 289 0.2 33 32 157 47 276 0.2 165 49 281 0.2 32 365 3.4 97 163 49 283 1.6 43 157 49 279 0.2 50 83 163 49 291 0.2 34 85 166 48 288 162 49 294 0.4 93 389 3.4 95 0.8 97 170 50 279 0.8 87 189 52 325 0.5 135 560 3.4 87 167 53 299 0.2 39 84 163 60 295 0.2 41 421 3.4 85 169 59 306 0.2 35 414 3.4 82 418 3.5 147 65 308 1.4 36 358 3.4 135 152 80 344 0.6 321 3.4 190 163 85 342 133 91 449 0.2 44 198 3.6 277 129 113 494 0.2 40 115 3.6 355 74 83 625 0.2 73 3.7 423 67 77 610 0.2 79 34 3.7 459 18 30 844 0.2 694 21 3.7 488 16 36 942 5.4 541 29 3.7 542 3.8 7 20 1469 0.3 18 3.7 580 6 21 1731 0.2 1799 18 3.7 600 7 20 1683 1726 7 40 1858 1.4 1826 19 3.7 529 5 0.3 1689 18 3.5 646 5 16 1962 0.3 63 107 1115 5.7 27 3.3 641 15 31 1679 1.1 1650 35 3.2 649 1628 3.2 6 17 1415 0.3 1432 65 3.0 620 5 18 1415 0.3 1370 59 2.4 601 4 17 1381 0.2 1394 67 2.7 528 626 4 18 1362 0.2 1311 71 2.4 558 71 1835 1151 82 1.8 489 4 18 1742 0.3 4 14 1066 0.2 1499 80 1.4 545 1.4 2092 73 1.1 510 3 15 1096 0.3 5 1265 0.4 2615 68 0.9 505 2 14 0.4 2 15 0.4 2437 68 0.9 503 Appendix C cont. Fall-Station 2 Depth Mn NI Cu Zn As Pb Fe N03 S04 NH4 (cm) MflL' C9L-' egL' f*9t-"1 (imolL mmolL pmolL 30.3 205' 285 106 55 202 0.3 102 304 3.6 58 18.7 200 288 106 45.8 209 0.2 90 301 3.7 62 18.7 202 289 107 45.2 212 0.2 13.5 206 285 108 55.2 203 0.4 90 302 3.6 64 12.3 11.0 9.7 201 271 100 40.5 211 0.2 95 289 3.6 62 5.8 202 263 97 41.1 201 0.4 102 289 3.5 62 5.2 4.5 205 271 100 43.5 195 0.3 109 268 3.2 62 4.5 275 3.2 3.9 3.2 201 272 100 42.3 206 0.3 293 3.6 63 2.6 86 285 3.4 63 1.9 198 281 105 44.6 205 0.3 93 291 3.6 60 1.3 196 194 103 295 3.6 61 0.6 203 285 107 49.7 215 0.6 100 292 3.6 60 0.0 215 265 98 52.7 222 0.5 93 294 3.6 62 0.0 288 3.5 -0.6 217 256 92.5 53.7 233 0.9 112 287 3.5 57 -0.6 258 92.1 54.3 0.8 -1.3 474 203 24.4 123.5 561 0.6 63 292 3.5 52 -1.9 426 216 29.9 131 565 1 103 -2.6 508 156 27.1 101 745 0.6 75 162 3.0 85 -3.2 445 166 28 130 591 1.1 105 128 2.9 95 -3.9 517 144 30.2 84.5 714 0.8 91 39 2.6 139 -4.5 439 109 20.2 47.8 859 1.2 186 53 2.5 140 -4.5 2.6 -5.2 594 122 31.5 67.2 1302 3 476 19 2.2 166 -5.8 632 98.2 16.6 38.5 1712 1.7 579 20 2.1 170 -5.8 20 . -6.5 754 128 28.2 76.1 1769 1.1 224 35 1.9 185 -6.5 745 1763 -7.1 736 79.8 18.6 28.9 2183 2.5 959 20 1.9 204 -7.7 681 113 27.2 67 1689 2.1 404 23 1.7 225 -8.4 822 75.9 16.3 24.1 2202 2 986 25 1.8 220 -8.4 822 2153 1.8 -9.0 689 101 19.7 42.3 1735 2.2 460 21 1.6 259 -9.7 775 67 10.2 26.6 2098 2.3 1237 29 1.7 262 -9.7 28 -11.0 815 90.9 17.6 54.7 2250 1.1 662 18 1.6 332 -11.0 92.6 17.7 48.5 1.1 -12.3 823 67.4 10.3 23.4 2364 1 931 16 1.5 364 -12.3 822 2410 -13.5 832 58.4 9.6 21.7 2646 0.6 946 16 1.4 384 -14.8 791 64.5 24.7 32.8 2255 0.6 810 2 1.2 408 -14.8 69 26 33.7 0.7 1.2 -16.1 759 82 38 56 2066 0.8 645 1 1.1 418 -17.4 697 110 28.6 97.5 1946 0.8 427 4 1.1 436 -17.4 690 1938 4 -20.0 741 61.5 18.3 39.3 1983 1.1 632 2 0.9 458 -23.9 876 58.2 13 37 2848 1.1 1010 2 0.7 502 -23.9 60.2 13 34.2 1.1 -27.7 943 32.3 3.8 13.5 2791 1.2 2108 2 0.6 513 -27.7 947 2840 -32.9 957 34.5 4.6 14.5 2781 1.4 2455 13 0.6 491 -32.9 0.6 -39.3 958 64.7 7.2 26.6 2797 0.8 2495 14 0.5 497 -45.8 999 51 7.4 29 3257 1.8 3009 8 0.3 -45.8 2997 8 0.3 507 Fall-Station 2 Depth Mn Ni Cu Zn As Pb Fe N03 S04 NH4 (cm) MgL1 C9L' cgL' cgL' **gL"' f*motL""mmolL>molL" 30.3 202 65 238 92.4 311 3.5 67 18.7 206 295 116 56 235 0.4 94.9 312 3.6 66 18.7 198 290 117 71 238 0.4 93.7 9.7 196 284 109 51 230 0.3 97.3 302 3.4 67 5.8 191 283 112 50 226 0.3 94.6 304 3.5 66 3.2 191 296 114 51 229 0.3 84.4 300 3.3 64 2.6 200 285 109 54 229 0.3 92.2 309 3.4 65 2.6 310 3.4 1.9 207 289 111 55 240 0.3 91 63 1.3 223 308 109 80 239 0.3 78.9 310 3.4 60 0.6 280 106 56 0.4 106.9 313 3.4 62 0.6 198 274 107 55 218 0.4 0.0 331 3.4 45 -0.6 416 326 63 203 320 0.5 50.7 335 3.6 34 -0.6 47.6 -1.3 431 302 44 217 353 0.5 57.1 258 3.3 78 -1.3 50.3 255 3.2 -1.9 459 223 49 170 546 0.5 42.2 230 2.9 79 -1.9 448 518 -2.6 451 198 43 238 578 0.5 39.4 155 2.4 135 -3.2 398 135 38 77 767 0.6 51.9 126 2.5 142 -3.2 134 37 77 0.6 -3.9 354 122 26 90 747 0.5 41.6 70 2.1 174 -4.5 376 106 29 48 1005 0.8 62.1 56 2.1 158 -5.2 319 84 19 37 975 1.1 92.5 27 1.9 185 -5.2 27 1.9 -5.8 389 84 17 34 1197 0.7 67.4 25 1.9 181 -6.5 46 13 5 1595 0.1 18 1.7 213 -7.1 426 70 20 25 1768 0.9 119.6 15 1.7 219 -7.1 416 1757 -7.7 502 55 15 20 2222 2.3 216.5 22 1.6 254 -7.7 212.1 -8.4 543 62 17 29 1988 1 121.8 19 1.6 244 -8.4 123.9 -9.0 -9.7 648 97 29 63 2048 1.7 447.9 21 1.5 287 -9.7 423 20 1.5 -11.0 629 98 80 63 1903 1.4 337.5 20 1.5 346 -11.0 612 100 81 64 1900 1.4 -12.3 609 91 38 59 1950 1.3 421.2 16 1.4 356 -12.3 89 37 58 1.3 -13.5 13 1.3 382 -14.8 52 0.3 6 1.1 421 -14.8 -16.1 687 83 21 63 1941 1.1 453.9 6 1.1 445 -16.1 671 84 20 61 1893 1 -17.4 7 0.8 463 -17.4 7 0.8 -18.7 -20.0 695 47 16 28 1912 0.9 650.7 11 0.9 489 -21.3 -22.6 -23.9 756 42 8 28 3287 0.8 919.2 8 0.7 537 -27.7 793 39 8 22 2049 1.7 1301 17 0.6 526 -32.9 830 23 1 10 2164 0.8 2417 14 0.5 547 -39.3 839 24 2.3 9 1749 0.9 3504 15 0.5 533 -39.3 823 24 2.5 11.5 1730 0.9 0.5 -45.8 934 34 4 15 4062 1.8 3780 10 0.3 581 -45.8 3934 11 0.3 228 Appendix C cont. Fall-Station 1 Depth Mn Ni Cu Zn As Pb Fe N03 S04 NH4 (cm) CSL-' M9L' C9L' ugL' ugL1 i»molL mmolL umolL 33.5 240 280 117 85 214 0.17 119 357 3.8 63 21.9 203 284 121 48 214 0.48 93.7 352 3.7 63 21.9 207 288 122 46 218 0.53 92.8 350 3.7 12.9 208 282 120 49 218 0.17 90.5 351 3.7 64 9.0 209 287 121 51 218 0.17 90.3 349 3.8 64 8.4 7.7 7.1 8.5 204 274 117 44 216 0.23 94.8 350 3.8 63 202 283 118 48 212 0.18 95.2 354 3.8 63 6.5 97 5.8 251 339 136 55 264 0.18 99.8 355 3.8 62 5.2 204 282 117 46 211 0.17 96.2 355 3.7 59 5.2 349 3.7 4.5 205 290 120 58 212 0.11 88.4 350 3.6 66 3.9 206 211 86.4 349 3.7 64 3.9 202 269 113 52 211 0.2 3.2 207 281 119 56 205 0.2 91.7 357 3.7 65 2.6 213 286 118 64 214 0.17 91 355 3.7 65 1.9 210 283 116 62 205 0.26 94.5 353 3.7 64 1.3 207 280 118 57 212 0.16 94.9 358 3.8 63 1.3 210 286 118 56 215 0.16 93.3 350 3.7 0.6 205 295 118 60 209 0.31 120.4 348 3.7 57 0.0 214 291 114 63 208 0.17 88.4 354 3.8 65 -0.6 221 297 111 71 204 0.17 86.9 330 3.7 56 -1.3 466 529 60 277 190 0.19 37.6 234 3.7 130 -1.3 59 275 0.2 -1.9 431 463 57 265 195 0.26 34.4 147 3.3 126 -2.6 477 366 45 133 287 0.25 35.1 95 3.8 193 -2.6 3.7 -3.2 482 370 39 121 327 0.37 45 70 2.9 173 -3.2 43.2 70 -3.9 439 292 41 88 392 0.27 34.3 25 2.9 207 -3.9 439 288 42 89 383 0.3 -4.5 465 267 29 62 431 0.41 37.96 30 2.7 232 -4.5 30 -5.2 509 235 31 50 430 0.42 45.4 2.7 226 -5.8 523 232 25 40 476 0.53 40.1 19 2.5 275 -6.5 622 211 38 68 477 0.38 54.8 16 2.7 271 -6.5 2.6 -7.7 536 242 42 84 676 0.36 38.9 13 2.4 290 -7.7 537 251 42 84 685 0.33 34.1 -9.0 570 236 28 74 712 0.48 45.9 46 2.3 279 -10.3 505 223 36 83 686 0.4 51.3 23 2.1 290 -10.3 23 -11.6 627 181 19 62 834 0.34 62.3 11 2.0 314 -12.9 617 179 33 81 894 0.24 70.44 8 2.0 329 -12.9 69.28 -14.2 597 185 30 83 852 0.31 74.24 8 1.9 354 -14.2 590 177 29 76 845 0.24 1.8 -15.5 -16.8 561 164 42 78 811 0.4 93.24 19 1.9 388 -18.1 -19.4 -20.6 513 207 47 93 710 0.36 45.9 10 1.5 430 -20.6 46 -24.5 564 90 27 42 829 0.58 152.8 10 1.4 404 -24.5 570 94 28 42 830 0.66 145.3 11 -29.7 714 38 12 14 801 0.65 494.4 10 1.1 408 -29.7 530 -38.1 680 32 5.8 10 775 0.31 803.1 11 0.8 398 -42.6 722 36 29 19 813 0.94 969 7 0.6 365 -42.6 719 36 29 17 816 1 1002 7 0.6 Fall-Station 5 Depth Mn NI Cu Zn As Pb Fe N03 S04 NH4 (cm) egt-' cgL' cgL' CSL"' cgL' eg!-' (igL-1 (imolL'mmotL>motL 36.1 196 292 118 189 213 1.3 502 313 3.7 63 24.5 184 294 109 188 204 0.3 120 326 3.7 65 24.5 179 279 104 181 209 0.3 19.4 186 296 110 89.7 215 0.3 98 313 3.7 67 15.5 192 298 110 73.7 223 0.4 96 323 3.7 67 11.6 204 307 110 57.5 255 0.3 97 313 3.6 75 11.6 300 107 55.6 0.3 7.7 235 309 96 80.3 338 0.3 110 319 4.4 94 7.7 104 319 4.4 7.1 6.5 264 324 100 91 382 0.3 108 318 4.7 105 6.5 257 370 5.8 254 323 96 117 369 0.3 99 319 4.7 106 5.2 259 323 94 79.5 376 0.3 112 320 5.1 110 4.5 254 324 92 75 376 0.1 112 318 5.1 109 4.5 100 3.9 250 323 93 72 371 0.2 121 317 5.0 111 3.9 252 327 94 71 376 0.2 3.2 244 321 91 67 359 0.2 103 317 5.1 109 3.2 317 5.1 2.6 256 324 92 65 365 0.2 97 317 5.2 110 1.9 260 321 91 70 346 0.2 98 320 5.0 112 1.3 253 305 82 60 306 0.2 102 313 5.1 114 1.3 256 304 83 62 312 0.2 104 0.6 310 307 57 37 277 0.1 77 312 5.0 114 0.6 74 0.0 352 199 43 42 234 0.3 74 309 4.8 92 -0.6 353 125 21 19 220 0.3 66 295 4.7 107 -0.6 300 4.8 -1.3 577 78 14 26 406 0.7 142 159 3.1 38 -1.3 146 -1.9 377 60 9.7 12 266 0.4 69 146 3.3 58 -1.9 376 60 9.7 12 272 0.4 68 -2.6 -3.2 771 75 11.5 18 428 0.5 96 -3.9 383 69 11.2 42 375 0.6 100 9 2.3 39 -5.2 744 79 13 25 590 1 9 2.3 50 -5.2 747 598 -6.5 619 69 13 43 725 0.9 209 5 2.1 52 -7.7 1008 91 17 35 873 1.1 191 5 1.9 55 -7.7 2.0 -9.0 1312 106 20 31 886 1.2 210 4 2.3 51 -9.0 4 -10.3 1384 112 22 32 1265 1.2 227 4 2.7 68 -10.3 222 -11.6 1470 120 25 35 1539 1.1 206 3 2.1 92 -11.6 -12.9 729 102 18 29 1285 1.1 2 2.4 125 -14.2 967 124 18 34 1258 1.2 211 2 2.3 129 -15.5 -16.8 1639 151 24 44 1388 1.4 264 2 2.4 158 -16.8 1745 1441 257 2.4 -18.1 -18.1 -19.4 1468 133 18 36 1808 1.1 203 2 2.8 218 -19.4 2 -24.5 1784 84 23 20 3041 1 221 1 2.6 243 -29.7 904 43 5 9 2862 0.9 219 2 2.3 207 -29.7 2.3 -34.8 683 40 5 15 2498 1.8 485 2 2.2 199 -34.8 449 -40.0 2 2.4 187 -40.0 2.4 229 Appendix C cont. Winter-Station 2 Depth Mn NI Cu Zn As Pb Fe N03 S04 NH4 (cm) ugL1 C9L' C9L' i*gL' cgL' MgL1 umolL mmolL (imolL' 27.7 487 432 95 187 194 0.6 66 249 6.9 288 27.7 486 439 97 190 193 0.6 22.6 671 483 74 141 209 0.5 57 193 9.0 284 16.1 812 501 56 181 238 0.3 71 16.1 74 14.8 34 10.1 393 14.8 389 9.7 942 420 24 197 414 0.2 132 7 10.3 457 9.7 129 8.4 7.1 218 7.1 954 410 21 198 488 0.1 220 5 10.5 449 6.5 5.8 941 417 22 218 512 0.2 259 6 10.3 401 5.8 970 402 21 216 516 0.1 8 9.8 5.2 937 402 20 224 488 0.1 270 10 10.5 413 4.5 977 399 21 212 553 0.3 306 10 10.2 393 4.5 969 408 22 219 551 0.3 3.9 930 395 20 251 497 0.1 268 7 10.0 413 3.2 988 404 20 211 565 0.1 329 14 9.3 433 2.6 930 404 21 269 504 0.2 281 2.6 285 1.9 972 406 21 270 589 0.2 374 11 9.2 1.3 965 400 21 280 535 0.2 311 9 9.5 429 1.3 9 9.5 0.6 970 408 21 256 616 0.1 398 7 8.9 413 0.6 978 404 22 285 606 0.2 0.0 978 399 23 275 530 0.4 337 10 9.1 393 -0.6 915 405 19 241 606 0 455 9 8.6 377 -0.6 926 403 19 245 622 0 -1.3 961 469 19 257 468 0.2 234 8 8.1 393 -1.3 219 -1.9 949 418 17 248 532 0.1 286 -1.9 291 -2.6 8 7.8 417 -2.6 425 -3.2 1044 446 17 243 536 0.05 217 23 4.5 401 -3.2 24 4.5 -3.9 -4.5 1117 365 16 223 580 0.06 163 8 3.9 433 -5.2 -5.8 1125 293 15 180 647 0.1 158 8 3.1 461 -5.8 1083 298 15 181 630 0.1 -6.5 -7.1 1003 220 11 132 780 0.1 287 6 2.8 453 -7.7 -8.4 1001 154 9.7 95 835 0.4 333 -8.4 154 9.5 95 0.3 -9.0 -9.7 885 115 11 78 736 0.3 182 11 2.2 489 -1Z3 731 68 19 54 610 0.2 63 13 1.7 481 -12.3 65 -13.5 668 43 23 33 632 0.2 66 -13.5 67 -14.8 10 1.5 477 -14.8 9 1.5 -16.1 656 37 28 38 703 0.2 72 -17.4 12 1.1 485 -17.4 473 -18.7 590 43 26 52 594 0.2 43 -18.7 560 42 25 52 557 0.2 -21.3 608 43 35 42 613 0.2 40 9 0.8 481 -25.2 485 -26.4 649 40 25 32 579 0.4 337 6 0.7 -26.4 667 48 13 1.6 -31.6 641 43 24 41 363 0.7 204 9 0.6 501 -31.6 201 -36.8 644 39 23 33 300 1.1 192 8 0.4 501 -36.8 643 39 23 33 295 1.1 193 -41.9 573 36 31 143 272 0.3 75 7 0.3 493 -48.4 651 46 22 52 314 0.9 241 5 0.3 497 -48.4 653 46 21 50 323 0.9 Winter-Station 1 Depth Mn NI Cu Zn As Pb Fe N03 S04 NH4 (cm) (jgL1 MgL' cgL-' cgL' MgL1 (.gL1 MgL'1 (imolL_mmolL>molL- 25.2 531 367 67 88 159 0.27 41 256 10.3 437 20.0 641 407 65 116 172 0.23 42 259 11.3 487 13.5 868 452 59 147 198 0.51 48 13.5 843 422 50 149 204 0.14 40 12.3 132 13.0 524 12.3 528 9.7 887 414 39 161 255 0.15 39 52 12.8 582 8.4 7.1 910 425 41 164 266 0.14 42 46 12.6 578 5.8 4.5 986 462 37 162 275 0.14 46 48 12.1 536 4.5 49 12.3 3.9 3.2 981 457 37 158 280 0.25 71 51 12.1 549 3.2 996 458 37 157 283 0.22 68 2.6 891 422 260 49 35 10.4 528 1.9 967 425 34 156 271 0.17 60 46 11.6 549 1.3 938 444 41 169 268 0.15 45 33 10.4 520 0.6 981 408 34 153 268 0.13 65 37 10.8 528 0.0 947 441 39 154 260 0.14 44 31 10.4 520 0.0 31 10.1 -0.6 1046 465 36 161 298 0.11 66 27 10.4 528 -0.6 1058 483 36 163 298 0.11 65 -1.3 989 495 40 185 269 0.12 45 25 9.9 520 -1.9 1067 482 34 174 301 0.12 72 24 8.3 495 -2.6 1055 571 45 248 297 0.13 49 -3.2 1073 466 36 200 295 0.12 81 21 7.5 499 -3.9 1024 447 49 223 278 0.12 53 19 8.9 528 -4.5 1090 421 35 174 297 0.17 87 -4.5 83 -5.2 19 8.2 536 -5.2 19 8.2 532 -5.8 110 495 -5.8 1143 342 28 136 297 0.17 115 14 6.3 -6.5 -7.1 1182 264 19 98 369 0.2 300 12 5.5 499 -7.7 -8.4 1234 204 14 70 411 0.22 477 20 4.7 507 -8.4 1232 206 14 68 405 0.22 467 -9.0 -9.7 1258 176 10 59 394 0.13 502 15 4.2 503 -10.3 -11.0 1342 136 9 45 385 0.4 532 -11.6 -12.3 1309 115 8 40 369 0.23 400 16 3.4 507 -14.8 1372 93 7 30 410 0.21 359 15 2.7 495 -14.8 16 2.7 -16.1 1325 72 7.4 34 408 0.11 183 -16.1 171 -17.4 7 2.7 495 -18.7 1214 41 4 17 441 0.27 493 -18.7 1215 42 4.2 20 450 0.31 -20.0 7 1.6 483 -20.0 491 -21.3 789 31 5 22 401 0.11 134 -21.3 126 -23.9 776 24 2.9 13 387 0.63 666 30 1.1 491 -27.7 8 0.9 478 -29.0 718 25 1.9 ias 365 0.43 683 -34.2 739 24 2.4 9.8 476 0.58 1476 5 0.7 474 -34.2 1438 -39.3 697 29 4 17 386 0.53 1017 2 0.5 445 -39.3 2 -44.5 628 33 6.3 21 250 0.59 692 5 0.4 416 -44.5 655 -51.0 718 33 4 16 491 491 2081 3 0.2 404 -51.0 2186 230 Appendix C cont. Winter-Station 4 Depth Mn Ni Cu Zn As Pb Fe N03 S04 NH4 (cm) M9L' cgL' MSI."' MgL' cgL' MgL' (igL-1 MmolL mmolL-MmolL 25.2 356 491 290 97 368 0.4 176 477 4.6 130 20.0 371 450 256 94 354 0.3 141 424 4.7 118 20.0 372 443 255 96 353 0.3 16.1 454 441 209 121 330 0.3 132 364 4.9 142 12.3 556 425 145 138 315 0.3 115 290 5.5 171 12.3 283 9.7 229 5.9 200 9.7 200 8.4 897 320 43 141 302 0.1 67 5.8 966 315 39 151 305 0.3 151 55 6.8 5.8 963 281 35 144 299 0.3 55 6.8 286 4.5 956 316 38 147 287 0.1 76 31 7.0 291 3.9 998 299 39 298 0.3 128 24 6.6 291 3.2 967 304 36 152 317 0.1 80 24 6.8 295 3.2 974 316 2.6 950 274 41 157 277 0.3 147 6.4 274 2.6 144 1.9 75 33 6.7 291 1.3 959 275 39 161 274 0.2 124 21 6.7 286 1.3 22 6.6 0.6 1023 304 36 151 308 0.1 74 27 7.2 291 0.6 1009 306 36 153 324 0.1 0.0 972 291 36 167 282 0.2 125 20 6.8 291 -0.6 1028 353 40 159 342 0.2 78 21 6.9 291 -1.3 1005 331 36 181 339 0.3 151 20 6.8 291 -1.9 1047 343 36 147 357 0.2 71 -2.6 987 356 33 168 405 0.4 153 23 6.7 307 -3.2 1024 336 31 123 409 0.2 75 17 6.7 295 -3.2 1021 338 31 124 416 0.2 65 17 6.5 -3.9 989 338 43 181 489 1.3 141 21 6.7 295 -4.5 13 5.7 295 -5.2 957 280 24 98 501 0.3 77 17 315 -5.8 1009 257 26 94 463 0.2 55 12 5.9 -5.8 1017 255 26 94 459 0.1 303 -6.5 -7.1 988 211 27 81 480 0.3 64 11 5.3 299 -7.7 -8.4 978 146 27 86 472 0.3 71 31 5.1 311 -8.4 991 141 27 64 491 0.3 30 -9.0 -9.7 958 107 15 44 481 0.2 79 12 4.5 303 -10.3 -11.0 992 104 11 41 480 0.2 77 -12.3 13 3.2 303 -13.5 972 79 4 32 502 0.3 162 -14.8 15 2.5 344 -16.1 940 50 2 29 524 0.4 125 -17.4 27 1.9 303 -18.7 921 45 3 25 436 0.2 125 -18.7 47 3.1 26 0.2 -20.0 -21.3 841 45 7.2 32 283 0.2 71 25 1.6 299 -25.2 21 1.4 282 -25.2 1.4 -26.4 684 70 14 65 191 0.2 64 -30.3 3 1.0 254 -30.3 278 -31.6 729 53 7.2 43 211 0.2 80 -31.6 726 55 7.5 45 206 0.2 -35.5 4 0.8 237 -36.8 623 45 12 46 235 0.1 52 -41.9 741 25 6.8 19 294 0.5 597 -41.9 615 -43.2 4 0.6 196 -47.1 705 26 7.1 20 134 0.5 436 -47.1 696 26 7 19 131 0.5 -51.0 4 0.4 155 -52.2 712 112 7.2 64 244 0.7 471 -52.2 715 110 6.8 57 249 1.3 Winter-Station 6 Depth Mn Ni Cu Zn As Pb Fe N03 S04 NH4 (cm) MgL1 MgL' MgL' MgL' cgL' MgL' MgL' (»molL mmoiL MTIOIL 26.4 317 410 185 125 325 0.6 129 436 4.1 98 21.3 287 365 148 81 387 0.5 90 432 4.0 94 20.0 18.7 325 418 185 85 382 0.5 18.7 315 408 179 80 355 0.5 17.4 473 4.8 98 14.8 313 414 182 83 375 0.5 112 13.5 427 4.0 94 11.0 314 410 185 86 379 0.4 112 11.0 284 376 163 81 365 0.5 112 9.7 454 4.4 98 9.7 94 8.4 313 410 178 89 408 0.4 108 7.1 439 4.1 90 7.1 434 4.2 5.8 302 397 164 94 465 0.4 101 440 4.2 90 5.2 311 409 168 91 435 0.5 435 4.4 82 4.5 317 419 156 104 552 0.6 93 459 4.5 86 4.5 310 404 157 98 540 0.5 94 3.9 332 435 170 94 458 0.5 463 4.5 78 3.2 314 415 157 112 612 0.5 99 42 2.6 170 99 0.6 453 4.4 66 2.6 457 4.2 1.9 314 419 157 145 695 0.4 87 457 4.3 7o 1.9 318 419 162 137 696 0.4 1.3 321 426 163 114 545 0.6 465 4.5 58 0.6 311 409 162 118 732 0.5 91 475 4.4 54 0.0 314 413 154 122 613 1 90 464 4.4 50 -0.6 327 421 158 131 879 0.5 84 -0.6 298 431 172 135 578 0.7 85 -1.3 321 407 149 116 721 0.6 440 4.4 50 -1.9 324 409 157 144 952 0.4 83 426 3.9 46 -2.6 334 378 135 111 839 0.4 97 417 3.7 50 -3.2 324 422 148 126 1085 0.4 86 405 4.1 50 -3.2 151 128 0.4 84 -3.9 334 355 126 101 998 0.7 132 385 4.2 62 -4.5 327 376 138 186 1334 0.6 123 379 4.0 58 -5.2 343 341 126 147 1158 0.7 134 -5.8 325 348 130 124 1468 0.6 84 312 3.6 66 -5.8 129 124 0.6 84 -6.5 378 313 97 82 1350 1 141 -7.1 324 311 101 114 1714 0.6 75 221 3.3 86 -7.1 221 3.3 -8.4 -9.0 -9.7 302 253 72 84 1974 1 236 4.0 82 282 211 56 70 2264 1 81 -11.0 191 3.6 98 -12.3 198 120 25 36 3028 0.7 60 -13.5 108 3.1 113 -14.8 186 108 23 19 3507 1 88 -14.8 155 103 26 19 3237 89 -16.1 66 2.4 125 -16.1 129 -17.4 126 76 17 15 4875 1 158 -18.7 -20.0 95 61 14 7 6034 1 317 32 2.1 141 -23.9 20 1.9 149 -23.9 20 1.9 -25.2 50 18 4 4 8536 0.5 127 -25.2 50 18 4 3 8483 0.6 129 -29.0 24 1.8 153 -29.0 161 -30.3 45 14 4 3 9314 0.5 108 -34.2 7 1.9 165 -35.5 39 10 3 6 10160 0.3 101 -40.6 36 10 3 2 10428 0.3 112 -41.9 2 1.9 161 -45.8 36 12 4 2 11813 0.3 129 -45.8 11478 -49.7 5 2.1 173 -51.0 35 13 4 3 13909 0.6 216 231 Appendix C cont. Winter-Core 1 Depth Mn Ni Cu Zn As Pb Fe N03 S04 (cm) C9L' H9L"' c9L' M9L' Hfl-L"' (imolL mmoll̂ 1.0 570 456 107 142 223 1.5 83 267 10.6 1.0 583 458 104 135 223 1.5 82 278 11.2 1.0 86 -0.3 740 491 67 126 345 0.3 50 58 12.0 -1.3 839 621 25 122 419 0.1 31 6 11.0 -1.3 863 634 25 126 434 0.1 -1.8 874 617 20 179 446 0.4 56 14 11.1 -1.8 57 -2.5 1082 288 66 54 438 0.6 76 2 10.7 -2.5 1057 283 69 56 448 0.6 3 10.9 -3.5 986 121 12 23 1021 0.5 104 3 8.2 -4.5 1017 42 1.3 45 1343 0.1 387 7 7.6 -4.5 408 -5.5 1033 13 1.3 4.6 1020 0.03 461 2 7.6 -5.5 1023 12 2 8.6 1035 0.08 -7.0 990 9.4 1.1 3.3 709 0.05 252 2 6.0 -7.0 3 6.0 -9.0 989 6.6 1.1 3.8 533 0.02 222 2 5.8 -11.0 943 6.3 1.1 4.8 434 0.02 221 3 4.7 -11.0 2 4.8 -13.0 848 7 1 16 688 0.07 116 2 4.0 -13.0 873 6.5 0.8 12 685 0.02 116 -15.5 848 4.8 0.6 6.9 423 0.02 436 3 3.2 -18.5 743 4.5 0.8 323 320 0.2 794 3 2.6 -18.5 737 4.5 0.9 318 321 0.2 2.5 -24.5 682 3.4 0.6 26 83 0.08 2411 2 1.1 -24.5 2469 -30.5 650 2.9 0.6 20 17 0.2 3074 6 0.5 -30.5 743 17 2963 6 0.5 Winter-Core 2 Depth Mn Ni Cu Zn As Pb Fe N03 S04 (cm) (•SfL-1 CSL' C9L' eg<-' pgL' cgL1 pgL' fimouVmmolL 1.0 755 561 129 145 234 1 102 272 10.3 1.0 692 561 163 141 237 1.4 104 308 8.9 -0.3 1083 663 71 129 386 0.4 44 54 12.5 -0.8 1020 676 68 142 374 0.4 44 19 12.3 -0.8 1071 694 70 154 383 0.4 -1.3 1041 694 51 173 400 0.5 47 12 10.9 -1.3 49 11 -1.8 1096 686 40 114 434 0.6 53 6 11.3 -1.8 1139 744 40 114 445 0.6 11.2 -2.5 1354 586 35 74 472 0.8 60 7 10.2 -2.5 7 -3.5 1348 167 22 28 740 0.3 90 3 8.7 -3.5 91 3 -4.5 1365 44 5 7.9 1020 0.2 469 3 8.1 -4.5 6.7 -5.5 1284 13 4 7.7 735 0.2 460 12 7.1 -7.0 1319 6 4 3.9 483 0.2 440 1 6.5 -7.0 1401 6 3.6 3.9 505 0.2 1 -9.0 1346 5 4 4.4 358 0.3 324 2 5.2 -9.0 331 5.3 -11.0 1216 8 5.5 4.1 309 0.2 291 1 4.9 -13.0 1103 4 3.5 4.1 268 0.2 316 1 3.9 -13.0 1039 6 4.8 4.2 254 0.2 -15.5 897 4 3.9 3.9 271 0.2 390 1 3.1 -15.5 408 -18.5 897 4 3.6 4 207 0.2 837 3 2.2 -18.5 3 -24.5 722 2.7 4.4 5.4 44 0.3 2197 5 0.9 -24.5 735 3 4.6 5.7 44 0.3 5 -30.5 897 3.3 4.2 4.3 9 0.3 3094 4 0.4 -30.5 3121 4 0.4 232 Appendix C cont. Winter-Core 3 Depth Mn Ni Cu Zn As Pb Fe N03 S04 (cm) egL' C9L' MgL' (jgL' egL' cgL' jimotLmmolL 1.0 193 243 6 9 1888 0.5 56 340 4.5 1.0 190 252 3 7 1878 0.4 46 418 4.7 -0.3 281 321 95 37 2658 0.7 45 -0.8 397 516 295 147 776 0.9 105 467 4.4 -0.8 -1.3 364 395 147 88 1815 0.5 57 421 4.3 -1.3 55 -1.8 264 300 47 23 3291 0.5 24 4.4 -1.8 259 301 46 25 3263 0.5 -2.5 420 502 65 114 2151 1 83 365 4.1 -3.5 457 457 66 96 1289 0.8 80 318 3.9 -3.5 456 452 65 94 1290 0.9 -4.5 451 276 41 49 1365 0.5 45 230 -4.5 229 -5.5 308 100 21 17 1780 0.4 31 124 3.4 -5.5 30 126 3.0 -7.0 186 54 11 8 2202 0.4 40 42 -7.0 183 2212 -9.0 72 13 2 73 5017 0.3 74 15 1.1 -9.0 13 2.8 -11.0 33 8.2 1 22 7553 0.3 82 21 2.8 -11.0 81 2.7 -13.0 32 6.9 1 4 7933 0.3 86 24 2.7 -15.5 32 6.3 0 2 9424 0.2 108 28 2.6 -15.5 27 -18.5 27 5.1 1 2 9928 0.1 131 30 2.7 -18.5 27 4.9 1 3 10491 0.1 -21.5 10142 118 25 2.7 -21.5 30 5.9 1 2 0.4 -24.5 37 6.1 1 3 11649 0.4 109 28 -24.5 35 5.9 1 2 11303 0.5 106 Spring-Station 1 - Depth Fe Mn NI Cu Zn As Pb N03 S04 NH4 (cm) cgL' MgL' MgL'1 MgL1 MgL' MgL' nmolLmmolLnmolL 31.5 82 290 409 266 61 224 1.7 265 3.5 187 31.5 78 26.4 77 295 405 259 57 233 1.8 263 3.5 185 19.9 82 290 397 247 59 228 1.5 19.9 295 415 260 60 233 1.4 18.6 269 3.6 191 18.6 187 13.5 75 301 401 241 60 231 1.3 262 3.6 185 12.2 10.9 72 288 400 240 60 229 1.1 10.9 72 290 403 243 60 227 1.1 9.6 259 3.5 185 9.6 261 3.5 8.3 73 287 401 236 61 226 1.1 7.0 265 3.6 5.7 72 288 415 246 222 0.96 256 3.4 189 4.5 73 291 400 238 61 222 0.97 260 3.4 185 3.8 80 302 401 239 63 230 0.12 255 3.5 187 3.2 68 307 400 237 63 233 1.3 266 3.6 185 2.5 82 302 404 238 64 228 1.3 259 3.4 191 1.9 68 303 407 241 63 227 1.1 260 3.6 185 1.9 70 1.2 82 305 426 254 70 236 1.7 262 3.6 185 1.2 264 3.8 189 0.6 74 299 417 249 232 1.2 259 3.5 178 0.6 300 411 246 69 230 1.2 -0.1 84 310 420 249 70 244 1.5 267 3.5 178 -0.7 74 309 415 242 69 237 1.5 178 -0.7 306 424 249 70 233 1.4 -1.4 305 399 240 86 251 2.2 283 3.7 180 -2.0 91 312 409 236 78 241 1.1 268 3.7 185 -2.0 314 420 241 89 238 1.1 272 3.7 -2.6 104 432 420 219 88 284 1.7 262 3.9 193 -3.3 85 415 421 220 97 260 1.3 175 4.2 230 -3.9 88 505 426 156 95 435 1.9 148 4.7 277 -4.6 79 528 446 150 111 424 1.6 118 5.0 331 -4.6 77 -5.2 70 724 426 85 93 657 0.89 -5.9 52 830 474 66 114 701 1.1 65 5.7 410 -5.9 5.7 -7.2 54 966 444 42 129 877 1.4 30 5.9 453 -8.5 59 963 360 29 120 1022 1.2 20 6.0 485 -9.7 185 1094 203 14 61 1747 1.6 8 5.9 503 -9.7 189 1071 204 15 61 1743 1.5 -11.0 277 1104 107 10 28 2090 0.75 10 5.5 511 -12.3 133 1067 128 21 42 1769 1.6 7 4.7 -12.3 127 511 -14.9 366 1015 107 7.2 26 2316 0.87 8 3.9 -14.9 366 4.0 503 -16.2 -17.5 201 972 136 24 56 2001 3.3 2 3.4 515 -17.5 967 134 23 46 1995 3.2 -20.1 11 2.9 511 -21.4 146 887 98 25 44 1532 2.7 -21.4 913 89 14 21 1.7 -22.6 9 2.6 498 -23.9 -25.2 3 2.4 505 -25.2 498 -26.5 111 762 109 54 68 706 3.4 -26.5 112 -29.1 9 1.8 466 -29.1 9 -31.7 175 785 90 31 55 637 2.2 -34.3 11 1.3 440 -34.3 1.3 -36.8 169 731 101 31 65 298 2.6 -36.8 714 100 30 63 282 2.2 -39.4 g 1.1 412 -42.0 239 800 124 30 93 298 2.7 -42.0 238 -44.6 4 1.1 397 -45.9 89 513 191 56 117 889 1.2 -45.9 501 192 57 112 879 1.2 233 Appendix C cont. Spring-Station 1 Depth Mn Ni Cu Zn As Pb Fe PH (cm) cgL' fgL1 egL' pigL' MgL' (igL' 31.8 277 360 198 100 199 0.7 74 30.7 7.61 27.2 274 355 194 56 203 0.5 72 27.2 272 363 197 58 207 0.5 26.0 7.77 21.4 272 335 180 53 200 0.4 74 21.4 76 20.3 7.81 15.6 276 335 186 201 0.8 72 15.6 272 352 196 62 211 0.6 14.5 7.80 9.9 278 380 204 60 213 0.7 74 8.7 7.70 4.7 7.78 4.1 280 363 199 64 215 0.8 72 7.95 3.5 2.9 8.02 1.8 275 352 200 66 203 0.9 69 7.99 1.2 0.6 275 360 197 74 192 0.9 69 8.00 0.0 -0.5 7.80 -1.7 278 345 194 69 201 1 76 -2.3 -2.8 346 364 184 85 216 1 82 -3.4 7.80 -4.0 401 396 173 85 262 0.9 79 -4.0 79 -5.2 402 373 173 77 266 1 83 -5.2 409 379 176 87 277 1 -6.3 596 369 132 89 349 1 64 7.70 -6.3 -7.5 747 339 97 81 448 0.8 50 -7.5 47 -8.6 858 259 53 66 637 0.6 46 7.54 -9.8 842 209 44 61 765 1.3 70 -9.8 834 209 44 62 747 0.14 70 -10.9 7.53 -12.1 792 180 47 75 742 1.4 83 -13.2 776 142 42 60 748 1.5 65 7.57 -14.4 792 112 27 40 916 0.9 74 7.54 -14.4 816 112 27 39 914 0.8 -15.6 735 114 26 63 812 1.3 80 -15.6 75 -16.7 720 109 23 57 848 1.4 93 -17.9 755 96 19 55 939 1.3 92 7.46 -19.0 -20.2 762 92 13 35 900 1 144 7.57 -20.2 767 93 13 37 917 0.9 138 -21.3 7.35 -22.5 747 75 11 34 664 0.7 188 -22.5 -23.6 7.56 -24.8 699 67 6.3 42 562 1 311 -26.0 7.41 -29.4 662 64 5.2 33 402 1.1 513 -30.6 7.32 -31.7 -32.9 651 57 1.7 18 451 0.3 834 -32.9 662 54 1.7 22 457 0.3 829 -37.5 682 67 7.5 19 738 0.5 925 7.56 Depth N03 S04 NH4 (cm) |*motL" mmolL" jimoU"' 39.3 260 3.6 188 34.7 254 3.5 190 28.9 254 3.6 188 23.1 239 3.2 23.1 185 17.3 181 15.0 254 3.5 185 14.4 13.9 253 3.5 183 13.9 247 3.5 13.3 12.7 245 3.5 185 12.1 11.6 252 3.6 185 11,0 248 3.5 185 10.4 253 3.5 188 9.8 251 3.6 185 9.8 243 3.5 9.2 251 3.6 183 8.7 249 3.6 181 8.1 253 3.6 181 7.5 250 3.6 179 6.9 286 3.6 183 6.4 253 3.6 179 6.4 252 3.6 5.8 280 3.6 183 5.2 251 3.5 177 4.6 247 3.5 177 4.0 3.5 244 3.5 173 3.5 2.3 255 3.7 188 2.3 1.2 265 3.8 192 1.2 260 3.8 0.0 267 3.7 198 0.0 -1.2 -2.3 186 4.7 296 -2.3 -3.5 -4.6 57 5.9 437 -5.8 -6.9 -6.9 -8.1 12 520 -8.1 -9.2 -10.4 -11.6 5 5.2 529 -12.7 -12.7 -13.9 -15.0 19 2.9 488 -15.0 3.0 -16.2 -17.3 -19.6 23 2.0 458 -24.3 78 1.4 392 -28.9 17 1.4 439 -28.9 1.4 234 Appendix D. Sediment Data Compositional sediment data arranged by sample depth for each core. Major elements, C a C 0 3 , Organic C and N are in wt.%; minor elements are in ppm. Depth C a C 0 3 N Si Ti Al Fe Ca Mg Na K P S As (cm) (wt.%) (wt.%)(wt.%)(wt.%)(wt.%)(wt.%)(wt.%)(wt.%)(wt.%)(wt.%)(wt.%)(wt.%) ppm Corel station i 0.25 0.65 5.80 1.015 48.14 0.56 11.67 8.77 4.15 3.57 1.16 1.96 0.22 0.32 0.75 0.96 e.85 1.595 0.75 0.43 1.25 1.25 3.41 0.859 48.44 0.59 11.26 10.54 5.08 4.71 0.72 1.63 0.16 0.26 1.75 0.98 2.28 0.739 50.19 0.63 13.05 11.87 5.05 4.95 0.60 1.63 0.15 0.28 4744 2.50 0.92 1.45 0,360 50.45 0.65 13.93 11.63 5.15 5.06 0.49 1.67 0.16 0.39 5029 3.50 0.82 2.97 0.630 48.54 0.58 13.28 10.77 4.45 4.47 0.58 1.73 0.19 1.20 4677 4.50 0.85 3.08 0.567 49.47 0.59 13.59 10.83 4.64 4.57 0.64 1.85 0.15 1.19 4450 5.50 0.49 4.29 0,730 48.89 0,56 13,65 10.16 3.84 3.81 0.76 1.87 0.17 1,43 4053 7.00 0.04 8.78 0.860 55.37 0.55 12.48 5.98 3.14 2.01 1.28 2.23 0.25 0.93 1272 9.00 0.01 8.84 0.787 46.79 0.51 9.80 5.09 2.84 1.59 1.37 2.02 0.22 0.57 318 11.00 0.00 8.64 0.755 55,96 0.54 12.34 4.30 2.94 1.84 1,41 2,15 0.26 0.47 173 13.00 0.00. 8.59 0.740 53.40 0.54 11.56 4.27 2.88 j 1.77 1.42 2.16 0.23 0.43 115 15.50 0.00 8.48 0.725 53.27 0.55 11.86 4.46 2.89 1.77 1.46 2.16 0.26 0.39 74 18,50 0.687 54.57 0.57 12.55 4,46 2.81 1.87 1.53 2.33 0.25 0.29 25 24.50 0.00 7.75 0.676 54.72 0.59 13.00 5.44 2.70 2.00 1.61 2.35 0.23 0.20 0 30.50 0.00 7.54 0.663 55.62 0.59 13.17 4.61 2.70 1.76 1.62 2.36 0.24 0.16 0 Core 2 0.25 0.67 Station 1 0.75 0.98 1.25 1.18 I. 75 0.98 2.50 0.93 3.50 0.80 4.50 0.76 5.50 0.46 7.00 0.05 9.00 0.01 II. 00 0.01 13.00 0.01 15.50 0.00 18.50 0.00 24.50 0.00 30.50 0.00 Core 3 0.25 2.44 Station 6 0.75 1.78 1.25 2.13 1-75 2.70 2.50 2.36 3.50 2.23 4.50 2.41 5.50 2.56 7.00 2.64 9.00 3.06 11.00 3.18 13.00 3.14 15.50 3.20 18.50 3.25 21.50 24.50 5.93 1.059 6.23 1.379 46.09 0.56 3.66 0.956 47.94 0.58 2.06 0.661 48,93 0.63 1.53 0.370 50.38 0.65 2.74 0.577 3.04 0.593 49,43 0.58 5.44 0.744 48.18 0.55 8.55 0.774 53.13 0.54 8.81 0.754 52.78 0.53 8.78 0.750 55.14 0.54 8.68 0.730 51.67 0.53 8.43 0.697 51.73 0.54 8.09 0.672 49.22 0.54 7.86 0.658 51.78 0.57 7.89 0,685 53,33 0.58 1.43 0.150 51.02 0.51 3.48 0.515 50.92 0.54 2.91 0.388 49.81 0.57 0.97 0.108 49.64 0.56 1.14 0.170 50.32 0.57 0.87 0.101 49.55 0.57 0,64 0.050 48,94 0.58 0.60 0.053 49.09 0.58 0.56 0.037 47.39 0.56 0.33 0.017 48.73 0.56 0.35 0.019 42.62 0.52 0.06 0.011 49.92 0.58 0.05 0.007 48.88 0,57 0.03 0.002 49.47 0.56 0.018 52.23 0.59 0.014 48.05 0.57 11.16 9.91 4.73 4.07 11.24 10.62 4.95 4.73 12.65 11.52 5.02 4.67 13.81 11.71 5.13 5.00 13.65 10.69 4.44 4.33 12.37 8.76 3.78 3.49 12.25 5.13 2.98 2.06 11.55 4.41 2.88 1,75 12.30 4.50 2.90 1.96 11.15 4.32 2.85 . 1.73 11.34 4.31 2,85 1.77 10.95 4.45 2.76 1.65 12.11 4.65 2.68 1.85 1Z60 4.52 2,62 2,07 9.15 10.26 8.02 6.32 9.96 9.80 6.33 4.93 10.32 10.75 7.23 5.95 9.40 10.19 8.27 6.45 9.93 10.88 7.53 6.37 10.53 11.07 7.02 6.23 10.56 10,99 7.50 6.66 10.71 10.86 8.01 7.04 10.06 10.88 8.22 6.98 10.25 11.16 9.16 6,97 8.28 10.41 9.08 5.68 10.43 10.97 9.26 7.70 9.50 10,36 9.22 7.20 9.30 10.19 9.34 7.13 10.46 11.64 9.05 7.45 9.39 11.12 8.55 6.59 1.15 0.37 0.81 1.65 0.23 0.42 0.75 1.66 0.16 0.24 4196 0.55 1.70 0.15 0.33 5014 0.56 1.68 0.18 0.42 5021 0.55 0.92 4751 0.59 1.87 0.15 1.23 4415 0.86 1.92 0.18 1.34 3505 1.32 2.04 0.23 0.67 751 1.36 2.08 0.23 0.48 265 1.37 2.07 0.25 0.45 181 1.40 2.05 0.22 0.42 114 1.48 2.24 0,23 0,35 54 1.54 2.34 0.22 0.36 30 1.58 2.44 0.21 0.20 0 1.55 2.30 0,23 0.15 0 0.43 1.08 0.07 0.32 2192 0.64 1.24 0.11 0.33 3756 0.44 1.23 0.10 0.19 0.30 1.14 0.07 0,07 2527 0.40 1.15 0.08 0.09 3193 0.34 1.23 0.07 0.11 3570 0.31 1.28 0.07 0.11 3104 0.25 1.24 0.06 0.04 2813 0.20 1.24 0.06 0.14 2387 0.25 1.19 0.04 0.05 2022 0.26 1.18 0.05 0.07 1978 0.22 1.14 0.05 0.06 1976 0.23 1.05 0.06 0.06 1827 0.25 1.06 0.06 0.04 2072 0.23 1.16 0.07 0.14 2217 0.24 1.13 0.05 0.07 2451 235 Appendix D cont. Core 1 Station 1 Core 2 Station 1 Core 3 Station 6 Depth Ba Co Cr Cu Mn Ni Pb Rb Sr V Y Zn Zr (cm) ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm 0.25 322 130 273 8280 1136 2177 132 68 158 167 22 2509 83 0.75 209 310 372 6893 1457 4615 129 67 117 196 22 4781 62 1.25 189 179 478 2687 1503 4105 105 45 119 220 27 2959 59 1.75 154 98 399 1530 1600 2759 171 32 103 247 27 5856 57 2.50 177 73 352 1412 1850 1748 196 28 99 277 31 2753 60 3.50 200 317 334 18136 1645 4516 146 0 112 249 58 2222 64 4.S0 201 208 325 4060 1601 3302 159 0 114 253 56 1922 67 5.50 237 203 285 3782 1391 2793 209 7 127 250 43 2902 76 7.00 453 64 115 708 718 827 29 58 192 110 23 1013 111 9.00 492 27 95 132 690 206 12 72 211 96 17 255 132 11.00 522 24 90 58 683 127 12 78 220 94 15 184 142 13.00 546 21 96 49 694 102 13 76 225 97 18 158 145 15.50 533 18 91 39 654 86 16 74 224 95 18 145 143 18.50 565 18 95 30 651 69 14 82 233 102 21 127 150 24.50 585 17 96 22 624 52 15 83 228 105 19 129 148 30.50 558 17 97 14 620 42 12 87 237 104 19 122 156 0.25 335 140 291 8539 1215 2413 122 68 162 177 26 2837 81 0.75 239 253 393 6122 1438 4370 123 60 126 200 28 4095 64 1.25 197 165 467 2928 1462 4045 118 42 116 217 24 3769 58 1.75 150 91 378 1443 1705 2411 187 35 104 259 26 5596 59 2.50 184 74 366 1128 1860 1775 205 23 99 275 25 2797 59 3.50 198 226 336 11739 1694 3498 165 0 111 252 48 2101 62 4.50 204 238 324 6433 1629 3780 174 0 113 264 56 2112 64 5.50 280 170 257 2939 1261 2368 117 22 142 199 43 2610 83 7.00 478 40 117 438 732 468 42 65 202 111 19 492 118 9.00 525 26 95 122 681 200 24 72 213 96 19 225 130 11.00 515 22 92 64 676 125 13 75 217 90 17 180 134 13.00 547 25 93 44 684 93 18 71 216 92 17 152 140 15.50 527 18 89 30 649 75 17 77 225 89 20 136 142 18.50 549 21 93 23 644 64 14 84 223 96 17 131 146 24.50 601 16 100 15 648 41 10 90 228 108 21 124 150 30.50 575 19 97 17 627 43 8 84 229 100 17 123 145 0.25 89 66 609 1163 2060 406 90 12 99 205 19 428 37 0.75 131 141 511 5160 1677 882 77 27 111 196 23 875 51 1.25 110 111 586 3525 1922 780 107 12 99 221 24 770 42 1.75 70 58 634 1041 2128 448 107 12 98 225 21 480 36 2.50 74 52 599 1873 1901 764 126 13 88 220 22 994 40 3.50 77 47 610 1097 1958 768 141 12 91 244 20 1145 41 4.50 76 33 637 436 2124 549 161 12 90 256 22 822 38 5.50 76 38 648 423 2166 420 159 14 91 247 21 727 40 7.00 68 34 663 250 2212 313 162 13 90 246 22 534 38 9.00 65 24 676 64 2313 196 154 8 89 249 20 319 34 11.00 56 23 681 24 2255 176 153 14 91 231 25 282 32 13.00 52 26 675 36 2315 194 131 9 93 238 24 305 37 15.50 42 29 663 28 2354 178 115 B 89 228 20 276 36 18.50 40 26 658 26 2289 174 117 4 89 222 19 260 31 21.50 49 27 658 69 2240 204 144 11 93 237 21 359 37 24.50 44 40 644 57 2169 237 124 4 96 234 22 373 34 236 Appendix E. Core Logs Core BL-1 Notes: - Collected 2 April 1994 - Clear skies, light breeze, -3°C - Core collected with Pedersen corer through 1 m of ice - Station 1,4.0 m - Core length 38 cm Core Log: Whole core: Very fine-grained, silty clay Top 3 mm: Brownish floe, very fine-grained. 0 to 2 cm: Finely laminated alternating light/forest green; millimetre scale or less. ~0.9 cm: Distinct dark lamina. ~ 1 to 2.5 cm: Alternating light/dark fine laminations, cohesive clay. 2.5 to 4.0 cm: Homogeneous grey zone, less cohesive. 4.0 to 8.0 cm: Black layer, evidence of methane bubbles. 8.0 to bottom: Homogeneous, grayish/ chocolate brown. Core BL-2 Notes: - Collected 3 April 1994 - Overcast skies, 15 knot breeze, blowing snow, -5°C - Core collected with Pedersen corer through 1 m of ice - Station 1,4.0 m - Core length 41 cm 237 Core Log: Stratigraphy identical to Core BL-1 Core BL-3 Notes: - Collected 3 April 1994 - Clear skies, 15 knot breeze, blowing snow, -5°C (-15 to -20 with wind chill) - Core collected with Pedersen corer through 1 m of ice - Station 6, 3.0 m - Core length 54 cm Core Log: 0 to 3 cm: Fine silty gray tailings material 1 to 1.5 cm: Separation plane 1 mm thick with < 1 mm brownish layer on separation surface. 2.5 cm: Small separation 2-3 cm wide with thin brownish layer on planar separation surface. Myriad of tiny bubbles (< 1 mm) appeared progressively in the top several em's of core during core logging (temperature effect?). ~3 to 4 cm: Very fine-grained irregular amber/ chocolate/ gray brown layer. Layer thickness unclear due to smearing down sides of core barrel. 4 to 25 cm: Very fine-grained light gray silt. ~26 cm: Separation plane ~ 1 mm thick, discontinuous laterally, water-filled. ~27 cm: Separation plane < 1 mm, with ~1.5 cm thick dark gray layer (medium silt) immediately above (i.e., ~25-27 cm). 27 to 41: Variously gray-coloured medium silt to very fine silt tailings material. Irregular coarsening below 35 cm. Slight variations in colour but no obvious stratification. 238 Appendix F. Diffusive Fluxes The diffusive flux (Jz) can be estimated using Fick's First Law (ignoring both bioturbation and advective transport): J z = Dj (j) dc F dz where: J z = flux Dj = in situ diffusion coefficient (Li and Gregory, 1974) F = formation factor & 1.4 for silty clay (Manheim, 1970) (j) = porosity & 0.87, assuming an average particle density of 2.65 g-cm~3 dc = concentration gradient dz The following data tables present diffusive influx estimates (positive values) for dissolved Zn, Cu, N i , N0 3 " and S0 4 2", and effluxes (negative values) for dissolved Zn, N i , Mn, As and N H 4 . Diffusion coefficients were calcuated from in situ surface sediment temperatures measured during the summer (17°C), fall (3°C), winter (3°C) and spring (10°C) sampling periods. For the sake of convenience, fluxes are expressed in both grams or moles per cm2 y"1 (Flux 1) and grams or moles per m 2 d"1 (Flux 2). 239 Appendix F cont. Downward Influxes Season Station Element Dj Gradient F lux l Flux 2 g/crrr4 i4j/crrr7y ng/m2/d Summer 2 Zn 5.98E-06 2.4E-08 2.8 77.4 2 Zn 5.98E-06 3.3E-08 3.8 105.1 1 Zn 5.98E-06 2.1 E-08 2.5 67.4 1 Zn 5.98E-06 2.3E-08 2.7 75.3 Fall 2 Zn 3.81 E-06 2.4E-08 1.8 48.3 2 Zn 3.81 E-06 5.3E-08 3.9 108.0 1 Zn 3.81 E-06 1.1E-07 8.3 226.6 5 Zn 3.81 E-06 2.3E-08 1.7 46.3 Winter 2 Zn 3.81 E-06 3.3E-08 2.4 66.9 1 Zn 3.81 E-06 3.3E-08 2.5 68.2 4 Zn 3.81 E-06 3.1 E-08 2.3 63.9 6 Zn 3.81 E-06 2.6E-08 2.0 53.5 Spring 1 Zn 4.89E-06 3.7E-08 3.5 96.6 1 Zn 4.89E-06 8.0E-09 0.8 21.0 Summer 2 Cu 5.74E-06 2.1 E-08 2.3 63.3 2 Cu 5.74E-06 2.7E-08 3.0 81.9 1 Cu 5.74E-06 4.6E-08 5.2 141.8 1 Cu 5.74E-06 5.9E-08 6.7 183.3 Fall 2 Cu 3.82E-06 5.7E-08 4.3 116.7 2 Cu 3.82E-06 3.3E-08 2.5 68.0 1 Cu 3.82E-06 3.5E-08 2.6 71.2 5 Cu 3.82E-06 3.3E-08 2.4 66.9 Spring 1 Cu 4.78E-06 7.1 E-08 6.7 182.4 1 Cu 4.78E-06 3.6E-08 3.4 92.8 Summer 2 Ni 5.66E-06 7.3E-08 8.0 220.5 2 Ni 5.66E-06 7.2E-08 8.0 218.2 1 Ni 5.66E-06 1.3E-07 14.2 389.0 1 Ni 5.66E-06 8.6E-08 9.5 261.0 Fall 2 Ni 3.56E-06 3.1 E-08 2.2 59.9 2 Ni 3.56E-06 7.4E-08 5.2 141.2 1 Ni 3.56E-06 9.1 E-08 . 6.4 174.2 5 Ni 3.56E-06 1.2E-07 8.4 230.4 Winter 2 Ni 3.56E-06 5.8E-08 4.0 110.8 1 Ni 3.56E-06 6.0E-08 4.2 115.4 4 Ni 3.56E-06 3.7E-08 2.6 70.9 6 Ni 3.56E-06 3.7E-08 2.6 71.6 Spring 1 Ni 4.61 E-06 4.0E-08 3.6 97.8 1 Ni 4.61 E-06 5.7E-08 5.1 139.9 240 Appendix F cont. mol/crrv4 mmol/cm2/y mmol/m7d Summer 2 N03 1.57E-05 1.3E-07 0.04 1.1 2 N03 1.57E-05 1.3E-07 0.04 1.1 1 N03 1.57E-05 1.5E-07 0.05 1.3 1 N03 1.57E-05 1.3E-07 0.04 1.1 Fall 2 N03 1.08E-05 9.8E-08 0.02 0.6 2 N03 1.08E-05 8.3E-08 0.02 0.5 1 N03 1.08E-05 8.9E-08 0.02 0.5 5 N03 1.08E-05 7.8E-08 0.02 0.5 Spring 1 N03 1.33E-05 4.2E-08 0.01 0.3 1 N03 1.33E-05 5.6E-08 0.01 0.4 Summer 2 S04 8.68E-06 1.3E-07 0.O2 0.6 2 S04 8.68E-06 1.1E-07 0.02 0.5 1 S04 8.68E-06 1.2E-07 0.02 0.6 1 S04 8.68E-06 1.1E-07 0.02 0.5 Fall 2 S04 5.65E-06 3.6E-07 0.04 1.1 2 S04 5.65E-06 4.7E-07 0.05 1.4 1 S04 5.65E-06 3.9E-07 0.04 1.2 5 S04 5.65E-06 7.8E-07 0.09 2.4 Winter 2 S04 5.65E-06 3.3E-06 0.37 10.0 1 S04 5.65E-06 6.4E-07 0.07 1.9 4 S04 5.65E-06 4.2E-07 0.05 1.3 6 S04 5.65E-06 2.1E-07 0.02 0.6 Spring 1 S04 7.17E-06 2.9E-07 0.04. 1.1 1 S04 7.17E-06 2.9E-07 0.04 1.1 pward Effluxes Station Element Dj Gradient Flux 1 Flux 2 g/cm" jxg/crrf/y ng/m2A Summer 2 Zn 5.98E-06 -1.7E-08 -2.0 -53.5 2 Zn 5.98E-06 -1.8E-08 -2.1 -56.2 1 Zn 5.98E-06 -1.6E-08 -1.9 -51.8 Fall 2 Zn 3.81 E-06 -4.1E-08 -3.1 -84.0 2 Zn 3.81 E-06 -7.8E-08 -5.8 -159.3 1 Zn 3.81 E-06 -1.6E-07 -12.0 -328.3 Winter 2 Zn 3.81 E-06 -2.1E-08 -i.5 -42.0 1 Zn 3.81 E-06 -2.9E-08 -2.2 -60.3 6 Zn 3.81 E-06 -1.6E-08 -1.2 -33.0 Spring 1 Zn 4.89E-06 -8.2E-09 -0.8 -21.5 1 Zn 4.89E-06 -4.3E-09 -0.4 -11.4 Fall 1 Ni 3.56E-06 -3.6E-07 -24.9 -682.2 241 Appendix F cont. Summer 2 Mn 5.60E-06 -1.3E-07 -13.8 -377.0 2 Mn 5.60E-06 -1.6E-07 -17.8 -487.4 1 Mn 5.60E-06 -1.2E-07 -13.6 -372.4 1 Mn 5.60E-06 -1.1E-07 -11.6 -318.0 Fall 2 Mn 3.50E-06 -2.0E-07 -13.6 -371.5 2 Mn 3.50E-06 -1.4E-07 -9.3 -256.0 1 Mn 3.50E-06 -3.8E-07 -25.9 -708.3 5 Mn 3.50E-06 -1.4E-07 -9.5 -259.3 Winter 2 Mn 3.50E-06 -6.5E-08 -4.4 -121.4 1 Mn 3.50E-06 -2.9E-08 -2.0 -54.0 Spring 1 Mn 4.55E-06 -1.7E-07 -15.0 -411.5 1 Mn 4.55E-06 -1.3E-07 -12.0 -327.6 Summer 2 As 7.50E-06 -1 2E-06 -174.6 -4782.7 2 As 7.50E-06 -7.8E-07 -115.0 -3151.5 1 As 7.50E-06 -9.7E-07 -142.9 -3916.1 1 As 7.50E-06 -4.2E-07 -61.0 -1672.2 Fall 2 As 4.79E-06 -5.1E-07 -47.8 -1309.6 2 As 4.79E-06 -3.9E-07 -36.4 -998.3 1 As 4.79E-06 -8.3E-08 -7.8 -213.3 5 As 4.79E-06 -1.1E-07 -10.6 -289.3 Winter 2 As 4.79E-06 -5.2E-08 -4.9 -132.9 1 As 4.79E-06 -4.4E-08 -4.1 -112.8 4 As 4.79E-06 -4.2E-08 -4.0 -108.3 6 As 4.79E-06 -2.4E-07 -23.0 -629.9 Spring 1 As 6.08E-06 -1.8E-07 -21.9 -599.6 1 As 6.08E-06 -1.4E-07 -16.4 -449.9 mol/cnr4 mmol/cm2/y mmol/m2 Summer 2 NH4 1.64E-05 -8.8E-08 -0.03 -0.8 2 NH4 1.64E-05 -1.3E-07 -0.04 -1.2 1 NH4 1.64E-05 -1.1E-07 -0.03 -0.9 1 NH4 1.64E-05 -1.1E-07 -0.03 -0.9 Fall 2 NH4 1.10E-05 -2.9E-08 -0.01 -0.2 2 NH4 1.10E-05 -2.7E-08 -0.01 -0.2 1 NH4 1.10E-05 -3.2E-08 -0.01 -0.2 5 NH4 1.10E-05 -1.6E-08 0.00 -0.1 Spring 1 NH4 1.37E-05 -7.0E-08 -0.02 -0.5 1 NH4 1.37E-05 -6.3E-08 -0.02 -0.5 242 Appendix G. Accumulation Rates The following section provides very rough approximations of accumulate rate for various sedimentary components in Balmer Lake. The accumulation rate at a sediment depth Z (AR Z ) , can be defined by: A R Z = Wt.% constituent x Dry Bulk Density (DBD) (g-cnr 3) x Sedimentation Rate(cm-yr~l) Where: i) DBD = (1- (j)) psediment particles -3 <j> = porosity a 0.87 assuming an average particle density of 2.65 g-cm ii) Sedimentation Rate approximated to be ~ 0.3 cm-yr' 1) Sulphur S ^ = 1.5 wt.% x 0.35 g.cm"3 x 0.3 cm y"1 = 2 x 10 3 gcm'V 2) Nickel Nim = 0.4 wt.% x 0.35 g.cm 3 x 0.3 cm y"1 = 4 x 10"7 kg cm"2 y 1 3) ;Gopper C u ^ = 1 wt.% x 0.35 g.cm 3 x 0.3 cm y"1 = 10 x 10"7 kg cm '~yl 243

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
Canada 7 0
United States 6 1
United Kingdom 4 0
India 2 0
Malaysia 2 2
Brazil 2 0
Tunisia 1 0
City Views Downloads
Unknown 14 4
London 2 0
Kuala Lumpur 2 2
Mountain View 2 1
Wilmington 1 0
Sunnyvale 1 0
Vancouver 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items