UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An examination of trace metal contamination and land use in an urban watershed McCallum, Donald Wayne 1995

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

Item Metadata

Download

Media
831-ubc_1995-0265.pdf [ 13.24MB ]
Metadata
JSON: 831-1.0050348.json
JSON-LD: 831-1.0050348-ld.json
RDF/XML (Pretty): 831-1.0050348-rdf.xml
RDF/JSON: 831-1.0050348-rdf.json
Turtle: 831-1.0050348-turtle.txt
N-Triples: 831-1.0050348-rdf-ntriples.txt
Original Record: 831-1.0050348-source.json
Full Text
831-1.0050348-fulltext.txt
Citation
831-1.0050348.ris

Full Text

AN EXAMINATION OF TRACE METAL CONTAMINATION AND LAND USE IN AN URBAN WATERSHED by Donald Wayne McCallum B.Sc, Queens University at Kingston, 1985 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Civil Engineering We accept this thesis as conforming to the required standard The University of British Columbia April 1995 © Donald W. McCallum, 1995 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 The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT The Brunette River Watershed is intensely urbanized, occupying 7200 hectares in the geographic centre of the Greater Vancouver Regional District in British Columbia. This study examines lake, stream, and street sediment trace metal contaminant history in the watershed in relation to changing land uses. Contamination of Burnaby Lake surface sediments with Pb, Cu, Zn, and Cd is indicated by their respective enrichment factors of 10, 6, 3.8, and 1.5. Chromium, Ni , Mg, Fe, and M h are not enriched in the surface sediments. Calculated fluxes of Pb, Cu, and Zn entering Burnaby Lake are from 3 to 10 times higher than measured in two comparable urban lakes. Stream contamination is indicated by the failure of all 33 stream stations to meet watershed sediment criteria for Pb, Cu, and Zn and 13 stream stations for Hg. Trace metal concentrations in Burnaby Lake sediments increased steadily from 1950, during a period of rapid urbanization. A sharp peak in Cu, Cr, Cd, and Ni concentrations in 1970 is related to industrial discharges in the Still Creek area at the time. Lead has decreased over the past 20 years in lake, stream, and street sediments as a result of the removal of the Pb additive, TEL, from gasoline. During this time, Zn, Cu, Mn, and Hg increased in stream sediments by 45 , 81, 130, and 290 percent respectively. Land use changes since 1973 have been small compared to demographic and traffic changes. Lake sediment contaminant profiles and spatial patterns of stream and street sediment contamination indicate that traffic contributes a large proportion of the Pb, Cu, and Zn loading to the watershed. The large Mn increases in stream sediments since 1973 may be related to in-stream chemical transformations, possibly resulting from usage of the gasoline additive, M M T . Increases in stream sediment Hg levels may be related to point-source emissions originating outside the watershed. Permeable land cover in the upland areas has mitigated some of the effects of non-point source pollution generated in more densely populated lower river reaches and has contributed to a relatively healthy aquatic habitat in the eastern region of the watershed. Much of the recent reduction in permeable land cover has occurred in these upland areas. i i TABLE OF CONTENTS Abstract • . . . i i Table of Contents iii List of Tables vi List of Figures x Acknowledgements xiv 1. INTRODUCTION 1 2. TRACE METAL CONTAMINATION IN URBAN WATERSHEDS: A REVIEW '. 4 2.1 Using sediments as an indicator of aquatic pollution 4 2.1.1 Particle-metal binding interactions 5 2.1.2 Sediment particle size 7 2.1.3 Availability to Organisms 8 2.1.4 Sediment transport 11 2.1.5 Trace metal background levels 11 2.2 Urban runoff 12 2.2.1 Impact of impervious areas 14 2.2.2 Urban non-point sources of trace metals 15 2.2.3 Automotive sources of trace metals 16 3. THE BRUNETTE RIVER WATERSHED 19 3.1 Geology and hydrology 19 3.2 Vegetation and wildlife 23 3.3 Demography and land use 23 3.4 Environmental stewardship 27 3.5 Contaminant record 30 3.5.1 Trace metals 30 3.5.2 Organic and microbial contaminants 33 3.5.3 Nutrients 35 3.5.4 Spills 35 3.6 Ecosystem Health 36 4. METHODS 39 4.1 Field methods 39 4.1.1 Streambed sediments collection 39 4.1.2 Street sediment collection 41 4.1.3 Lake sediment cores 41 4.2 Sample preparation and analysis 46 4.2.1 Stream and street sediments 46 4.2.2 Lake cores 47 4.3 Sediment analytical techniques 50 4.3.1 Moisture content 50 4.3.2 Loss in ignition (LOI) 50 4.3.3 Acid digestions (excluding Hg method) 50 iii 4.3.4 Metals detection techniques (excluding Hg) 50 4.3.5 Mercury analysis 51 4.3.6 2 1 0 P b Radioisotope dating analysis 52 4.4 Land use and traffic analysis 53 4.4.1 Sub-catchment areas.. 53 4.4.2 Land activity and cover 53 4.4.3 Traffic analysis .. 56 4.4.4 Demographic data 58 4.5 Statistical analysis 58 5. DISCUSSION OF RESULTS 61 5.1 Variability in methodology and environment 61 5.1.1 Precision and accuracy of methods 61 5.1.2 Variability among analytical methods 63 5.1.3 Variability within sampling stations 66 5.1.4 Repeatability of environmental conditions 67 5.2 Trace metals in lake sediments 71 5.2.1 Burnaby Lake . 71 5.2.1.1 Background metal concentrations 71 5.2.1.2 Sedimentation rates 74 5.2.1.3 Trace metals in recent sediments 76 5.2.2 Deer Lake 83 5.2.3 Gwendoline Lake ..87 5.2.4 Comparison to other urban lakes 90 5.3 Land use changes from 1973 to 1993 93 5.4 Trace metal loading from automotive sources 101 5.5 Trace metals in streambed and street sediments 103 5.5.1 Overview of results 103 5.5.2 Temporal analysis of trace metal data 121 5.5.2.1 Iron 123 5.5.2.2 Lead 123 5.5.2.3 Copper 123 5.5.2.4 Zinc 127 5.5.2.5 Nickel 129 5.5.2.6 Mercury 129 5.5.2.7 Manganese 131 5.5.3 Spatial analysis of trace metal data 135 5.5.3.1 Street sediment analysis 135 5.5.3.2 Stream sediment analysis 138 5.5.4 Trace metal inter-relationships 142 6. SUMMARY AND CONCLUSIONS 144 6.1 Extent and severity of trace metal contamination in aquatic sediments 144 6.2 Changes in sediment trace metal levels since 1830 145 6.3 Land use in 1973 and 1993 146 6.4 Relationships between land use and trace metal contamination 147 7. RECOMMENDATIONS .151 7.1 Implications for further research 151 7.2 Management implications 152 8. REFERENCES 154 iv APPENDICES 163 APPENDIX A S T R E A M A N D STREET SEDIMENT S A M P L I N G LOCATIONS 163 APPENDIX B L A N D C O V E R TEST A R E A S 166 APPENDIX C Q U A L I T Y A S S U R A N C E D A T A 168 APPENDIX D COMPARISON OF A N A L Y T I C A L METHODS . 172 APPENDIX E L A K E SEDIMENT CORE D A T A 181 APPENDIX F JUSTIFICATION OF L A K E SEDIMENT N O R M A L I Z A T I O N 193 APPENDIX G L A N D USE D A T A 196 APPENDIX H S T R E A M B E D SEDIMENT RESULTS 205 APPENDIX I STREET SEDIMENT RESULTS 213 APPENDIX J S P E A R M A N R A N K CORRELATION MATRICES 214 APPENDIX K C A L C U L A T I O N OF A U T O M O T I V E T R A C E M E T A L L O A D S 216 v LIST OF T A B L E S 2.1 Objectives, guidelines and criteria for trace metals in freshwater sediments 10 2.2 Effect of a rainfall event on the quality of combined sewage in a residential area 15 3.1 Population and employment in Burnaby and the Brunette River Watershed since 1892 24 3.2 Zoned land use areas in the Brunette River Watershed in 1973 and 1987 26 3.3 Studies examining trace metal levels in the Brunette watershed 31 3.4 Studies examining organic and microbial contaminants in the Brunette watershed 34 3.5 Bioassay results from the Brunette watershed 38 4.1 Lengths of lake sediment cores and lake depth at each sampling site 46 4.2 Preparation and analysis of dried street and stream sediment sub-samples.. ...47 4.3 Preparation and analysis of preliminary dried stream sediment sub-samples 48 4.4 Sub-sampling locations of long lake core sediments 49 4.5 Method detection limits for metals analyses 52 4.6 Size and description of each sub-basin in the Brunette watershed... 55 5.1 Measurement of method accuracy using the N R C marine sediment reference material, BCSS-1.. . 62 5.2 Precision of analytical methods measured by the median ratio of duplicates and coefficient of variation (CV) 64 5.3 Stream and street sediment analytical techniques used in current study and 1973 baseline study 65 5.4 Daily rainfall at Burnaby Mountain weather station prior to 1993 and 1994 stream sediment sampling 67 5.5 Background metal levels and surface enrichment factors observed in Burnaby Lake core #2 74 5.6 Comparison of actual and predicted metal concentrations in Gwendoline Lake surface sediments 90 5.7 Burnaby Lake and Deer Lake surface sediment trace metal concentrations compared to three urban lakes 92 5.8 Burnaby Lake recent trace metal fluxes compared to two urban lakes 92 vi LIST O F T A B L E S (continued) 5.9 Land use activities in the Brunette River Watershed as a proportion of total area in 1973 and 1993 93 5.10 Land cover in the Brunette Watershed in 1973 and 1993 96 5.11 Traffic density in the Brunette River Watershed in 1973 and 1993 100 5.12 Estimated trace metal loadings from automobiles and in stormwater runoff in the Brunette Watershed 101 5.13 Median stream and street sediment metal concentrations 114 5.14 Number of stream stations exceeding sediment quality criteria in 1973 and 1993 120 5.15 Median concentration differences and significance of ranked t-test for comparison of 1993 versus 1973 metal concentrations in stream bed sediments (n=32) 121 5.16 Median concentration differences and significance of ranked t-test for comparison of 1993 versus 1973 metal concentrations in street sediments (n=25) 122 5.17 Median streambed total metal concentrations in the Brunette River and Still Creek sub-basins for 1973 and 1993 141 5.18 Land use and traffic density in the Brunette River and Still Creek catchment areas in 1973 and 1993 142 A - l Stream sediment sampling sites 163 A - 2 Street sediment sampling sites 165 B - l Determination of low density residential area permeability 167 B-2 Determination of industrial/commercial area permeability 167 C - l Quality assurance measurements of the Hg method 168 C-2 Quality assurance measurements of the nitric acid/flame A A method 169 C-3 Measurement of the precision of the nitric acid/ICP-AES method 170 D -1 Median concentration differences and significance of ranked t-test for comparison of aqua regia versus nitric/perchloric acid digestions (n=15) 173 D - 2 Median concentration differences and significance of ranked t-test for comparison of aqua regia versus nitric acid digestions (n=14) 173 D-3 Median concentration differences and significance of ranked t-test for comparison of flame A A versus ICP-AES detection techniques (n=12) 174 vii LIST O F T A B L E S (continued) D-4 Median concentration differences and significance of ranked t-test for selected element comparison of wet and dry-sieving methods (n=21) 177 D-5 Comparison of aqua regia and nitric/perchloric acid digestions 179 D-6 Comparison of aqua regia and nitric acid digestions 179 D-7 Comparison of the <180 um and < 63 um sediment particle fractions 180 E -1 Organic matter and metal concentrations in two Deer Lake (short) cores 182 E - 2 Organic matter and metal concentrations in two Gwendoline Lake (short) cores. ....182 E-3 Sediment texture and metal concentrations in Burnaby Lake core #2 183 E-4 Sediment texture and metal concentrations in Burnaby Lake core #1 185 E-5 Sediment texture and metal concentrations in Burnaby Lake core #3 185 E - 6 Sediment texture and metal concentrations in Deer Lake (long core) 186 E-7 Pb-210 radioisotope dating results for Burnaby Lake core #2 187 G-1 Land cover permeability in 1973 197 G-2 Land cover permeability in 1993 198 G-3 Land activity in 1973 199 G-4 Land activity in 1993 200 G-5 Employment within the Brunette Watershed in 1971 201 G-6 Employment within the Brunette Watershed in 1981 202 G-7 Employment within the Brunette Watershed in 1991 ....203 G-8 Population and dwelling units within the Brunette Watershed between 1971 and 1991 204 H-1 Sediment texture and total metal concentrations in streambed sediments 205 H - 2 Extractable metal concentrations in streambed sediments 209 I-1 Sediment texture and total metal concentrations in street sediments 213 J-1 Spearman rank correlation matrix for total metal concentrations in streambed sediments 214 viii LIST O F T A B L E S (continued) J-2 Spearman rank correlation matrix for extractable metal concentrations in streambed sediments 214 J-3 Spearman rank correlation matrix for metal concentrations in street sediments 215 ix LIST OF FIGURES 2.1 Adsorption of Pb, Cu, Zn, Co, N i , and Mn on iron oxides as a function of pH 6 2.2 Adsorption of Pb, Cu, Zn, Co, N i , and Mn on manganese oxides (birnessite B38) as a function of pH 6 2.3 Estimated automotive lead emissions, 1928-2000 17 3.1 The Brunette River Watershed 20 3.2 Horizontal profile of Still Creek and the Brunette River 22 3.3 Employment distribution in the Brunette River Watershed in 1971, 1981, and 1991 25 3.4 Automobile registrations in British Columbia from 1912 to 1993 26 3.5 Locations of M O E monitoring sites, effluent discharges, and flow gauge sites 29 4.1 Location of streambed sediment sampling stations 40 4.2 Location of street sediment sampling stations 42 4.3 Location of sediment core sampling stations in Burnaby Lake and Deer Lake 44 4.4 Location of sediment core sampling stations in Gwendoline Lake 45 4.5 Sub-basins in the Brunette watershed 54 4.6 Peak hour (A.M.) traffic in the Brunette Watershed 57 4.7 Traffic zones in the Brunette Watershed 59 4.8 Components of a box-whisker plot 60 5.1 Standard Error of the mean of metal concentrations at stream stations illustrated using box-whisker plots (n=33) 66 5.2 Mean and standard deviation of selected metals at all street sampling locations (n=25) compared to one location (n=3) 67 5.3 Trace metal results from 3 stream stations sampled during low flow periods in 1993 and 1994 68 5.4 Sediment texture and metal concentrations in Burnaby Lake core #2 72 5.5 Age profile of Burnaby Lake core #2 determined using 210Pb radioisotope dating 75 5.6 Sediment accumulation over time in Burnaby Lake (core#2 location) calculated using 210Pb radioisotope dating 75 x LIST OF FIGURES (continued) 5.7 Sediment texture in Burnaby Lake core#2 sediments (0-48 cm depth) 77 5.8 Metal concentrations in Burnaby Lake core#2 sediments (0-48 cm depth) 78 5.9 Historic trace metal fluxes to Burnaby Lake 81 5.10 Metal concentrations and sediment texture in Burnaby Lake core #3.... 82 5.11 Metal concentrations and sediment texture in Deer Lake (long core) 84 5.12 Metal concentrations and organic matter in Deer Lake (short cores) 86 5.13 Metal concentrations in Deer Lake core taken in 1979 87 5.14 Metal concentrations and organic matter in Gwendoline Lake sediments 88 5.15 Historic fluxes of Pb, Cu, and Zn to the sediments of a small lake in Melbourne, Australia 91 5.16 Land use activity in the Brunette Watershed in 1973 94 5.17 Land use activity in the Brunette Watershed in 1993 95 5.18 Land cover permeability in the Brunette Watershed in 1973 97 5.19 Land cover permeability in the Brunette Watershed in 1993 98 5.20 Reduction in land cover permeability in the Brunette Watershed between 1973 and 1993 99 5.21 Lead distribution in streambed sediments in the Brunette Watershed 104 5.22 Copper distribution in streambed sediments in the Brunette Watershed 105 5.23 Zinc distribution in streambed sediments in the Brunette Watershed 106 5.24 Nickel distribution in streambed sediments in the Brunette Watershed 107 5.25 Chromium distribution in streambed sediments in the Brunette Watershed 108 5.26 Manganese distribution in streambed sediments in the Brunette Watershed 109 5.27 Iron distribution in streambed sediments in the Brunette Watershed 110 5.28 Magnesium distribution in streambed sediments in the Brunette Watershed I l l 5.29 Organic matter (LOI) distribution in streambed sediments in the Brunette Watershed 112 xi LIST OF F I G U R E S (continued) 5.30 Silt and Clay distribution in streambed sediments in the Brunette Watershed.. 113 5.31 Lead distribution in streambed sediments in the Brunette Watershed in 1973 and 1993 115 5.32 Mercury distribution in streambed sediments in the Brunette Watershed in 1973 and 1993 116 5.33 Manganese distribution in streambed sediments in the Brunette Watershed in 1973 and 1993 117 5.34 Sediment trace metal concentrations in the Salmon (n=19) and Brunette River (n=33) watersheds illustrated using box-whisker plots 119 5.35 Changes in iron concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots 124 5.36 Changes in lead concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots 125 5.37 Changes in copper concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots 126 5.38 Changes in zinc concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots 128 5.39 Changes in nickel concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots 130 5.40 Changes in mercury concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots 130 5.41 Location of G V R D incinerator and monitoring sites 132 5.42 Mercury concentrations in vegetation at G V R D monitoring sites from 1987 to 1989 132 5.43 Changes in manganese concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots 133 5.44 Metal concentrations in street sediments illustrated using box-whisker plots 136 5.45 Still Creek and Brunette River sub-catchment areas 139 5.46 Largest Spearman rank correlation coefficients in stream and street sediments 143 xii LIST OF FIGURES (continued) D - l Linear relationships between ICP-AES and flame A A detection techniques 175 D - 2 Linear relationships between the ratio of metal concentrations in 2 sediment fractions (<180 |im fraction / < 63iim fraction) and the percentage of silt and clay in the <180|im fraction 178 E - l Qualitative descriptions of lake sediment (long) cores 188 F -1 Metal concentrations and sediment texture in Burnaby Lake core #1 194 xiii ACKNOWLEDGEMENTS This research was made possible through funding received from the Tri-Council Secretariat, representing the National Science and Research Council (NSERC), Medical Research Council (MRC), and the Social Sciences and Humanities Research Council (SSHRC). The support I have received while undertaking research at U B C has been unrelenting. I wish to thank Dr. Ken Hall for providing the research opportunity, a stimulating and friendly work environment, and continuing encouragement. The support I received from Dr. Hans Schrier, especially in the analysis of land use, has enhanced my research and my experiences at the university. The scope of this research would not have been possible without the laboratory support I received. I would like to acknowledge Susan Harper for laboratory guidance, Dr. Martin Grosjean and Dr. Dan Engstrom for their assistance in lake core analysis, and Rose Pinheiro, Zoe Redenbach, and Ron Macdonald for their uncompromising attention to detail. I received a great deal of support in the area of land use analysis. Yao Cui provided technical guidance and excellent suggestions for the use of the Terrasoft GIS program. Historical land use analysis was facilitated by aerial photographs provided on loan from the Planning Department within the City of Burnaby. Moral encouragement, understanding, and, of course, editing assistance, were provided by Erin Rosen. xiv 1. INTRODUCTION The growth of population and intensive human activity within the Brunette Watershed is typical of many urban areas in North America. The area existed in a near pristine state as recently as 130 years ago and has since quickly developed into a metropolitan landscape. Accompanying these cultural changes have been what seems an inevitable modification and contamination of the urban streams and lakes. Contaminants which have generated the greatest amount of concern and which have prompted many studies in this watershed over the last 25 years include pathogenic bacteria, oxygen-consuming organic matter, excessive nutrient inputs, hydrocarbons, and trace metals. This study examines current trace metal contamination in the watershed in the context of the historical contaminant record, and explores land activities which may have caused the contamination. Following the rapid industrial expansion after the second world war, industry was responsible for much of the pollution observed in urban areas. Over the last two decades pollution abatement efforts have identified, regulated, and mitigated many of these point waste sources to our waterways. Increasingly non-point source pollution, in the form of urban runoff (stormwater), is being recognized as the major continuing source of contaminants to urban streams (Gibb et al. 1991). The most coordinated response to contaminated urban runoff was undertaken by the United States, beginning in the 1970's, with the formation of the Nationwide Urban Runoff Program (NURP). The goal of this program was to define the nature and extent of urban non-point source pollution and to evaluate treatment options. According to the findings of the NURP, heavy metals had the greatest potential of the surveyed pollutants to adversely affect aquatic life (Whipple 1987). Internationally, urban runoff has been the focus of several new research discussions including a N A T O workshop in France in 1985 (Marsalek 1986) and the first international conference on diffuse (non-point) pollution in Illinois in 1993 (Olem 1993). This research examining urban runoff indicates that aquatic environments are generally polluted in densely populated areas and some form of runoff treatment is necessary to protect life in streams and 1 lakes. A focussed approach to the problem is necessary since complete treatment of urban runoff for a typical large city using best management practices is currently prohibitively expensive (Lee and Jones-Lee 1993). Two alternatives to complete treatment of urban runoff are available which may accomplish the desired goal of protecting aquatic habitat. The first option is to identify the pollutant source and eliminate it before it is released. The second is to identify the areas which contribute the bulk of contaminants, and collect and treat the stormwater from those areas only. Both options require a great deal of watershed-specific land use information — the collection and interpretation of which is greatly facilitated by the use of Geographic Information Systems (GIS). This study has utilized two GIS systems to analyze actual land activity and land permeability over time and space in comparison with contaminant levels. Estimating the relative contribution of traffic to the total contaminant load has been a particularly effective use of GIS. This study is part of a larger "Eco-Research" project examining changes in the ecosystem health of the Lower Fraser River region in British Columbia. Funded jointly by the Natural Sciences and Engineering Research Council (NSERC), the Medical Research Council (MRC), and the Social Sciences and Humanities Research Council (SSHRC), the project is taking an inter-disciplinary, ecosystem approach to examining the sustainability of this region's development (Westwater Research Centre 1994). In support of the larger project's goals, this case study of an urbanized watershed will examine the human activities which have contributed both to the contamination and remediation of urban waterways. The specific objectives of this study are listed below. Study Objectives (i) Identify the extent and severity of trace metal contamination in streambed and lakebed sediments in the watershed; 2 (ii) Quantify changes in sediment trace metal concentrations and loading rates which have occurred over the last 130 years; (iii) Quantify changes in land activity, land cover, and traffic density in the watershed over the last 20 years using geographic information systems (GIS); (iv) Identify relationships between land use (land activity, land cover, and traffic) and trace metal contamination in aquatic sediments. 3 2. TRACE M E T A L CONTAMINATION IN URBAN WATERSHEDS: A REVIEW The magnitude of trace metal contamination on a global scale has been estimated by Forstner and Miiller (1973) using a ratio of the annual mining output of a metal to its concentration in uncontaminated soils. This "Index of Relative Pollution Potential" was particularly large for Pb, Hg, Cu, Cd, and Zn and 10 to 30 times lower for Fe and Mn. A dramatic example of the effects of trace metal contamination on humans has been observed from studies on lead levels in Europeans and Americans. These studies show blood Pb levels which are two to three orders of magnitude higher than pre-industrial humans (Forstner 1990). This section reviews the characteristics of sediments which enable information on past and present aquatic trace metal contamination to be derived. Research examining the nature of urban runoff contamination and non-point sources of trace metals is summarized briefly. 2.1 Using sediments as an indicator of aquatic pollution For several reasons, sediments, rather than surface water, are often chosen as the medium to investigate aquatic contamination. Contaminant levels in the water tend to be quite transient because most of the contaminants enter an urban stream only during a rain-storm event. Once entering a stream, many contaminants, including trace metals, quickly partition out of the water column and onto sediment particles. The sediments which remain in place then provide a good indicator of local pollution sources. Lake sediments provide a particularly good historical contaminant record by examining in-place sediments at depths below the water surface. While sediments provide a unique opportunity for examining aquatic trace metal pollution, proper interpretation of the contaminant data can only be made with an understanding of particle-metal binding interactions, availability to living organisms, sediment transport, and natural background metal levels. 4 2.1.1 Particle-metal binding interactions Research on the transport and fate of metals in the hydrosphere has strongly indicated the importance of metal-sediment interactions. Warren and Zimmerman (1993) have provided a comprehensive review of the literature regarding the factors influencing the dominant sorption reactions and the influence of particle grain size. Oxides are considered extremely important in the sequestering of metals from solution because of their abundance in aquatic environments and ion exchange properties. Metals such as iron, aluminum, and manganese readily precipitate as oxides from the water column within the pH and pE environments encountered in most riverine systems. The sorption of metal ions by these oxides is understood to involve a number of possible complexing mechanisms including : (i) metal ion complexed by oxygen atom of a surface hydroxyl group, (ii) metal-ligand complex sorbed by the oxygen atom of a surface hydroxyl group, and (iii) the underlying metal ion exchanging its surface hydroxyl group for a ligand-metal complex. Particulate organic material (POM) is also a very important factor in trace metal adsorption although less well understood than oxide adsorption (Warren and Zimmerman 1993). Large organic molecules have very low solubilities in water and would be expected to be attracted to the particulate surface-water interface. P O M is known to be an efficient adsorber of trace metals — comparable to oxide adsorption. While the exact nature of the competition for metal ions isn't known, it is believed that the addition of an organic coating to an oxide surface will increase the overall sorptive capacity. In both oxide and P O M adsorption, the binding intensity has been shown to be cation specific. Figure 2.1 illustrates the adsorptive capacity of iron oxides for 6 different metals. The relative differences in adsorption can be readily explained by the differing abilities of the metals to hydrolyze in solution (Campbell and LaZerte 1988). Adsorption onto manganese oxide surfaces, as is illustrated by Figure 2.2, does not retain the same relative relationships. Relative to the other metals, Co and M n are much more readily adsorbed on manganese oxide surfaces compared to iron oxide surfaces. The attraction of M n and Co is believed to occur as Figure 2.1 Adsorption of Pb, Cu, Zn, Co, N i , and Mn on iron oxides as a function of pH; 20 (imol/g added (a) hematite; (b) goethite. (adapted from Campbell and LaZerte 1988). Figure 2.2 Adsorption of Pb, Cu, Zn, Co, N i , and Mn on manganese oxides (birnessite B38) as a function of pH (a) 2 mmol/g added; (b) 1 mmol/g added, (adapted from Campbell and LaZerte 1988) 6 a result of their oxidation and retention at the manganese oxide surface (Campbell and LaZerte 1988). Laboratory demonstrations of metal adsorption onto P O M (summarized by Campbell and LaZerte (1988)) show a similar behaviour to iron oxide adsorption as is shown in the following relative adsorption relationships : (i) Cu > Zn > N i > Co > Mn on sediment humic acid (pH 7.0); and (ii) Hg > Fe > Pb > Cr > Cu > Cd > Zn > N i > Co > Mn on soil humic acid (pH 5.8). The sorption reactions described above are influenced by sediment grain size for two reasons: first, P O M and oxides tend to concentrate in the smaller size fractions; second, smaller sediment size fractions have relatively higher adsorptive capacities due to their higher surface area to volume ratios (Fergusson 1990, Hakanson and Jansson 1983, Salomons and Forstner 1984, Warren and Zimmerman 1993). 2.1.2 Sediment particle size Various correction procedures have been proposed to account for the disproportionate concentration of trace metals in the finest grain size fractions. Salomons and Forstner (1984) suggest either: (1) comparing with conservative elements such as Fe, A l , Mg, or Cs, (2) separation of grain size fractions, (3) extrapolation from regression curves, (4) determining the mobile fraction using dilute acids, or (5) correcting for inert mineral constituents. A sixth option is available for correcting lake sediments because of the inverse correlation between moisture content and dominant grain size. Hakanson and Jansson (1983) suggests expressing metal concentrations in lake sediments on a wet volume basis as a means for normalizing grain size effects. They also consider normalizing concentrations using loss on ignition (LOI). This, however, appears to be a valid technique only over a narrow range of LOI values. Many researchers now advocate using only the < 63 um (silt and clay) sediment fraction to characterize site contamination (Hakanson 1984, Salomons and Forstner 1984). Advantages to this method include lower within-site variability and comparability to 7 numerous studies which have analyzed this fraction. Several studies have shown that using size fractions larger than 63 |im is preferable when : (i) mining wastes are a pollution source (Moore et al. 1989), (ii) dry sieving is used to fractionate sediment (Krumgalz 1989) , or (iii) when a total contaminant load must be estimated (Wilber and Hunter 1979). Recent studies of Don River sediments in Ontario by Warren and Zimmerman (1993) showed that a consistent increase in trace metal concentration with decreasing particle size does not always occur. Their work suggested that a significant percentage of the trace metal load could exist in particle sizes greater than 63 lim and was highly dependent on the degree and nature of particle coatings. The most important comparison being evaluated in this study is to a 1973 trace metal survey (Hall et al. 1976) in which particle sizes less than 177 |i.m were examined. For this reason the < 177 |i.m particle size fraction has been analyzed in this analysis. Comparisons to other studies which have analyzed different particle sizes have been made using particle size correlations derived from this watershed. 2.1.3 Availability to organisms Heavy metals at higher than normal environmental concentrations have been found to be toxic to a wide variety of organisms. It is believed that the most relevant mechanism of toxicity is the inactivation of enzymes. Divalent metals react with different protein functional groups and may compete with essential elements such as zinc for binding sites (Forstner 1990). Contamination of surficial sediments poses risks immediately to a community of algae, bacteria, protozoans, invertebrates, and bottom fish. Contamination of this benthic community can in turn affect organisms further up the food chain such as predatory fish. The bio-availability of metals bound to sediment is highly dependent on the speciation of the metal (Campbell and Tessier 1991, Huang 1993, Salomons and Forstner 1984). Experimental evidence has indicated that bio-accumulation of metals is dependent on the level of iron oxides and organic matter associated with the sediment as well as the specific 8 geo-chemical phase to which the metals are attached (Huang 1993). Partial extraction procedures have been developed in order to selectively extract trace metals from sediments. Cold 0.5 M HC1 acid has been shown to effectively release metals bound to the organic and oxide coatings while leaving the metals in the residual or detrital form (Pickering 1981). More recently, sequential extraction schemes have been devised which attempt to fully characterize the sediment trace metal geo-chemistry (Campbell and Tessier 1991). While providing promise of better characterizing sediments, work by numerous researchers (summarized by Fergusson (1990)) has shown that this method is unable to extract metals according to discrete geo-chemical phases and suffers from the problem of metals re-adsorbing onto another solid phase after selective extraction. These draw-backs do not necessarily condemn the procedure as the scheme can be considered operationally defined by extraction method rather than reflecting discrete sediment phases (Campbell and Tessier 1991). A major disadvantage with this method, which has yet to be overcome, is its inability to selectively extract mercury (Forstner 1990). The feeding habits of benthic invertebrates are an important consideration when assessing the bio-availability of sediment trace metals because some species achieve their nutritional requirements by processing significant volumes of sediment particles through their gut. Both the particle sizes utilized and the geo-chemical nature of the particle coatings on different particle sizes should be determined when considering the impact of contaminated sediment on stream invertebrates. According to Cummins and Klug's (1979) functional classification of stream invertebrates, a group called filtering collectors ingest the entire size range of suspended sediment particles between 0.2 um - 800 um. Examples of this functional group include net-spinning caddisfly and blackfly larvae. Another group, called gathering collectors, processes deposited sediment in the same size range; individual species are limited to narrower particle size ranges by the morphology of their mouth parts. Examples of this group include midge larvae and stony-cased caddisfly larvae. Warren and Zimmerman's (1993) study of sediments in the Don River (Toronto) showed that oxide and organic coatings occurred across the whole grain size spectrum (1 to 9 >102 (lm) and were not concentrated in the finer fractions as expected. Combined with knowledge of benthic invertebrate feeding habits, this finding suggests that when considering sediment toxicity it may be necessary to examine particle size fractions larger than 63 -Lim. Although national sediment toxicity criteria have not yet been established, reference to site and region-specific guidelines (Table 2.1) can provide an indication of toxicity. Campbell et al. (1988) suggest that a rational method for assessing sediment toxicity would involve examining the ecology and feeding habits of the aquatic organism of interest and finally devising an appropriate scheme to collect and analyze the sediment. Table 2.1 Objectives, guidelines and criteria for trace metals in freshwater sediments Pb Cu Zn Hg Ni Cr Mn mg/kg (dry whole sediment) Brunette Watershed Objectives3 5 30 70 0.070 U.S. EPA guidelines for Great Lakes harbour sedimentsb non-polluted < 40 < 25 < 90 < 20 < 25 < 300 moderately polluted 40-60 25-50 90-200 20-50 25-75 300-500 heavily polluted > 60 > 50 > 200 > 50 > 75 >500 Ontario provincial sediment quality guidelines13 lowest effect 31 16 120 16 26 460 severe effect 250 110 820 75 110 1100 Ontario guidelines for dredged material 50 25 100 25 25 Wisconsin sediment criteriab 50 100 100 100 100 a- rationale for objective setting explained in Swain (1989) b - compiled by Bennett and Cubbage (1991) from various sources 1 0 2.1.4 Sediment transport Understanding the origin and fate of sediment particles in a riverine system is essential for determining linkages between sediment contaminant levels and land use. In general, sediments are eroded and transported in the upper reaches of a river system where current velocity is high and are then deposited lower down the system at lower stream velocities. The settling process is adequately explained by Stokes Law which shows a sediment's settling velocity to be proportional to the square of its diameter (Fergusson 1990). This relationship indicates that fine particles, which are of significance in contaminant studies, will remain suspended longer than coarser sediments and thus travel further from their source. As an illustration of the difficulty in assigning causes to observed contamination, Salomons and Forstner (1984) observed elevated Cd concentrations in the Rhine River over 1000 km downstream of the known source. Some sediment contaminant-land use studies have recognized these limitations and only make direct comparisons between streams of similar order (Colman and Sanzalone 1992). A similar approach is taken in this study to ensure only hydrologically valid comparisons between stream contamination and land use are made. 2.1.5 Trace metal background levels Natural variation in trace metal concentrations caused by geological differences must be accounted for when assessing contamination levels. Background levels determined within the watershed are ideal for assessing the presence or absence of trace metal enrichment. Enrichment factors — the ratio of the trace metal concentration to the background concentration measured in a deep Burnaby Lake core — have been used in this study to provide an indication of relative contamination in lake sediments. These factors are less useful for comparing stream sediment contamination because of the variability in organic matter and grain size throughout streams in the watershed. The same background measurements from the lake core do provide, however, an indication of the presence or absence of trace metal enrichment in stream sediments. 1 1 2.2 Urban runoff Determining the amount of toxic substances introduced to the hydrosphere by urban runoff is a complex problem because of the diffuse and intermittent nature of the pollution. Specific runoff events do not necessarily define the problem. In an effort to determine the contaminant sources and ecological effects of urban non-point source pollution, research has focussed on examining stormwater runoff, street dirt, and stream sediments. Characterization of stormwater runoff is perhaps the most direct method for evaluating the contribution of contaminants from different land uses. Relationships between land use categories (ie. residential, commercial, industrial, undeveloped), surface permeability, and traffic density to contaminant load become apparent by carefully selecting sampling locations with small catchment areas and well defined land use information. From stormwater studies in the Great Lakes region, Marsalek and Shroeter (1988) have determined that approximately 90% of the metal loading across all land categories in urban runoff is comprised of Pb, Cu, and Zn. However, because of the extreme variability in pollutant loads within and between storm events, universal relationships to land use have not been found. The variability within a storm event has been described by the term "first flush effect". It describes the higher pollutant concentrations observed during the early parts of storms — a phenomenon caused by the rapid suspension of contaminated sediments deposited since the last storm. Between-storm variability can also be significant and is highly dependent on the antecedent dry period and the total event runoff volume. Variability between years has been observed and has largely been attributed to differences in annual rainfall cycles. The final report of the NURP concluded that runoff quality could not be predicted solely by adjacent land use because of the extreme event to event variability (Athayde et al. 1983). In their comprehensive review of the topic, Gibb et al. (1991) suggest that long-term, catchment-specific studies are required to fully characterize stormwater runoff and estimate contributions from specific land uses. Studies characterizing urban street dirt have also been used to examine urban non-point source pollution. Hall et al. (1976) found significant enrichment of trace metals in 1 2 industrial/commercial area street dust relative to undeveloped areas, despite large contaminant variability within land use areas. A study of a watershed near Champaign, Illinois (Rolfe et al. 1977) found very high Pb street dirt concentrations which varied according to traffic volume. They also found that 54 % of total lead emissions from automobiles quickly settled from the air as dustfall and this dustfall was concentrated within 20 metres of the roadway. Several recent studies have attempted to characterize the geo-chemical properties of urban dust (Dempsey et al. 1993, Ramlan and Badri 1989). Fewer studies have used in-place stream sediments as an indicator and evaluator of urban non-point source pollution. One of the main reasons for this is that it is harder to localize pollution sources using sediments. For example, sediments analyzed at a main river stem location could potentially have been transported to that location from a nearby source or, in the extreme case, from headwater areas of the watershed. Modelling the origin, transport, and fate of sediments is extremely complex and site-specific given the number of variables involved. While integration of models with GIS has simplified the task (Engel et al. 1993), a robust sediment transport model for urbanized areas has yet to be developed. In the absence of a detailed model, variables such as channel slope and stream order can be used as a rough indicator of sediment transport. In their study of streambed sediments in the Upper Illinois river basin, Colman and Sanzolone (1992) were careful to make direct spatial comparisons of contaminant data only between streams of similar order. In-place sediments are an efficient means to determine changes in pollution sources over time because they act as an accumulator of contaminants. Lake sediments, in particular, have been used extensively to determine pollution records because of their relatively stable sedimentological processes (Hakanson and Jansson 1983). A recent study of a densely urbanized watershed in India (Vaithiyanathan and Ramanathan 1993) utilized suspended and bed sediments, along with a sediment core in the lower river reaches, to determine anthropogenic influences and spatial differences in contamination. While not identifying specific sources, their study identifies an increasing trend in trace metal contamination. In contrast, a similar study in New Jersey (Bonnevie et al. 1993) analyzed surface bed sediment 1 3 and long river sediment cores, observed a declining trend in trace metal contamination from 1960 onwards, and attributed the improvement to waste discharge regulations and improvements in the structure and functioning of combined sewer outfalls. Lake core examination has also been used to evaluate the effects of changes in land cover and permeability in catchment areas (Engstrom and Wright Jr. 1984). The following sub-sections summarize research which has evaluated the influence of impervious land cover and various urban land uses on urban runoff contamination. 2.2.1 Impact of impervious areas The process of urbanization severely alters an area's land cover through the construction of roads, buildings, and parking areas. These changes reduce the land's permeabilty which reduces infiltration of precipitation and thus increases surface runoff. Apart from any consideration of contaminants, the change in urban stream hydrology — as a result of higher surface runoff during storm events — can be extremely disruptive to aquatic communities. Stream contamination may increase near impervious areas because contaminants which might normally be trapped in the soil are quickly and easily washed off impervious surfaces. The specific effect of urbanization on the hydrology of a stream system has been predicted quantitatively through the use of an effective impervious area (EIA) measure. EIA measures the impervious area in a watershed which is directly connected to the drainage network. Dinicola (1990) has developed a series of factors for different types of urban land uses in Washington State relating EIA to total impervious area. These factors were used by Rood and Hamilton (1994) in their analysis of salmon streams in the Fraser River delta area. While the current study will not quantify hydrologic changes, the concept of EIA will be utilized to examine linkages between stream contamination and surrounding land use. 14 2.2.2 Urban non-point sources of trace metals Trace metals entering the urban aquatic environment can come from a wide variety of sources. Prior to the development and enforcement of waste discharge regulations, inputs from industry could be quite significant. In the absence of industrial point sources, the most significant contaminant sources include sewage discharged as a result of combined sewer outfalls (CSO), or faulty connections, metal-alloy corrosion, automobile-related activities, and the leaching of paints and stains. Sewage infiltration of stormwater systems can be a significant contaminant source. While the immediate concern relates to transmittal of pathogens and depression of stream oxygen levels, trace metal toxicity is also a concern. A study by Koch et al. (1977) of metal concentrations in a combined sewer before, during, and after a rainfall event (summarized in Table 2.2) illustrated the relative trace metal concentrations of sewage and runoff. Runoff, a dominant component in the sewer during a rainfall, appears to provide much of the N i , Pb, and Zn while Cu is more concentrated in the sewage. A lake core study which examined the effects of eliminating sewage discharges to Lake Washington (Seattle) revealed a decrease in Cu and Zn concentrations in lake sediments of 28 % and 20 % respectively, as a result of sewage diversion (Spyridakis and Barnes 1976). Table 2.2. Effect of a rainfall event on the quality of combined sewage in a residential area Sampling Period Cu Fe Mn Ni Pb Zn Flow cfs During rainfall event 62 600 49 2 127 207 14.1 After rainfall event 56 458 53 1 11 94 5.7 Summer dry weather 120 652 70 1 18 80 2.4 Source : (Koch et al. 1977) 15 Very high zinc concentrations have been found in the runoff from rooftops and have been attributed to the solubilization of galvanized gutters (Bannerman et al. 1993, Good 1993, Hey and Schaefer 1983). Through their detailed sampling program and runoff modelling, Bannerman et al. (1993) determined that roofs from commercial and industrial areas contributed significant loadings of zinc to urban runoff. 2.2.3 Automotive sources of trace metals A recent comprehensive review of urban runoff contamination (Gibb et al. 1991) identified the operation of motor vehicles to be the most significant non-point source of trace metals to the urban environment. This conclusion is supported, indirectly, by a recent study of urban runoff in Madison, Wisconsin (Bannerman et al. 1993) which demonstrated that streets and parking lots produced the largest trace metal runoff loads of Pb and Cu. A comprehensive source identification analysis conducted in the Santa Clara Valley (California) also found that motor vehicles were the largest contributor of many trace metals to urban runoff (Woodward-Clyde Consultants 1992). Contaminants contributed from automobiles have been attributed to 5 main sources : exhaust emission, tire wear, oils and greases, corrosion, and breakdown of the road surface. Exhaust emissions have changed in composition over the last 20 years due to regulatory controls on additives in fuels. Tetraethyl lead (TEL) has been used as a gasoline additive to enhance octane ratings since the introduction of the car early in this century. At its peak usage in 1973, Canadian automotive emissions of lead were 14,360 tonnes — approximately 70% of total national lead emissions to the atmosphere (Poon 1989). Recognizing that Pb levels in the Canadian urban environment were reaching dangerously high levels, the phase-out of T E L began in 1974 and was fully completed in 1990. The rapid rise in automotive lead emissions since 1928 and the even more rapid decline since 1973 is charted in Figure 2.3. Although Pb has been completely removed as an additive from gasoline, natural levels of lead in gasoline (approximately 10 mg/1; (Lee and Jones-Lee 1993)) may still pose a threat to the urban environment. 16 Figure 2.3 Estimated automotive lead emissions, 1928-2000. i5;000| "~ 10.000 H in <D C C .2 5000 h 1930 1940 1950 1960 1970 1980 1990 2000 Year Source: adapted f rom Commiss ion on Lead in the Environment (1985) Starting in 1974, methylcylopentadienyl manganese tricarbonyl (MMT) began to be used in gasoline and is now the primary octane enhancing additive used by Canadian refineries (Forget et al. 1994). Using a 1990 gasoline sales estimate of 34 billion litres (Lafleur 1994), an average current additive concentration of 11 mg Mn/1 (Alberta Research Council 1994), and a tailpipe emission factor of 30% (Loranger et al. 1994), manganese emissions from transportation in Canada total approximately 112 tonnes per year. While this amount is small relative to total Canadian emissions (6625 tonnes; (Puckett et al. 1988)), it is the next largest source of Mn after steel and alloy production. A few studies have examined the effects of increased urban Mn levels; Joselow et al.(1978), Forget et al. (1994), and Loranger et al. (1994) have all discovered significant correlations between environmental M n concentrations and traffic density. The Forget study examined M n uptake by blue spruce trees in high traffic areas but did not observe enhanced levels. Zinc emissions from traffic result from background levels in fuel, the degradation of lubricating oils, and the gradual wear of tires. Based on typical Zn concentrations in each of 17 these components (fuel emission rates and wear rates summarized in Appendix K) , tire wear and fuel combustion are the most significant automotive sources of Zn. Using a similar analysis, the wear of brake linings has been determined to be the most significant automotive source of Cu (Armstrong 1994). Armstrong's study of the composition of different brake linings found significant variation in Cu levels between models. This result suggests an opportunity for abatement since Cu is not an essential component of all brake linings. This study will utilize watershed-specific land use information in conjunction with previous urban runoff research to determine the origin of the surveyed trace metal contaminants in the Brunette Watershed. 1 8 3. T H E B R U N E T T E R I V E R W A T E R S H E D The Brunette Watershed, or Central Valley Watershed as it is referred to in Burnaby, is a small tributary system to the lower Fraser River. Located in the geographic centre of the Greater Vancouver Regional District (GVRD), the watershed is intensely urbanized and includes major traffic corridors which links Vancouver — the financial and employment centre — with growing population centres in the eastern regions of the district (Figure 3.1). The 7200 hectare watershed falls within the municipalities of Vancouver, Burnaby, New Westminster, Coquitlam, and Port Moody. The central portion, containing both Burnaby and Deer Lakes, falls entirely within the City of Burnaby. The purpose of the following section is to provide context to the objectives of the study. Past surveys of aquatic health and environmental contaminants in the watershed are reviewed in order to properly judge the relative risk of trace metal levels encountered in this study. Demographic changes are summarized to illustrate the direction and rates of urban development, while the desire and capacity to change is surveyed under the heading of environmental stewardship. 3.1 Geology and hydrology Surficial material in the lowland areas surrounding both the Brunette River and Still Creek is composed of peat sediments which were deposited following a rise in the sea level approximately 8000 years ago. This sea level rise caused the Fraser River to temporarily connect to Burrard Inlet through the Still Creek corridor (Gardner Dunster Associates Ltd. 1992). Overlaying bedrock in the rest of the watershed are glacial till and gravels. Humo-ferric podzols have since developed on top of the glacial material. The natural stream flow characteristics are highly dependent on the region's topography and climatic conditions. Being a lowland stream system, the hydrograph closely follows the precipitation cycle: maximum discharges occur during the high precipitation winter 19 Figure 3.1 The Brunette River Watershed months and minimum discharges in the summer. Within the watershed, variation in stream velocities occur as a result of the catchment topography. Steep gradients in the upper reaches of both Still Creek and the Brunette River (Figure 3.2) cause high stream velocities while the lower reaches of Still Creek from Gilmore Avenue to Burnaby Lake experience near-quiescent conditions during dry periods. Low dry-weather stream velocities in the Brunette River from Brunette Avenue to the Fraser River are also observed due to moderate gradients and tidal effects. Evident from the contour map in Figure 3.1, streams flowing off Burnaby Mountain in the north-east corner of the watershed follow very steep gradients and experience high stream velocities. A partial record of stream flows in the watershed has been kept by the Water Survey of Canada for the following 4 stations (Environment Canada 1991): (i) Brunette River near Buena Vista (1926-1935) : 08MH021 (ii) Brunette River @ Sapperton (1934-1971): 08MH026 (iii) Stoney Creek (1965(summer)) : 08MH117 (iv) Still Creek (1958-1978) : 08MH061 Human influence over the past century has altered the natural hydrology in a number of ways. A permanent dam was constructed on the Brunette River downstream of Burnaby Lake in 1935 for the purposes of flood control in New Westminster. Also for flood control, a relief channel was constructed from the Brunette River to the Fraser River at Braid Street in 1982. The greatest impact on the hydrological system was caused not by dams and diversion channels, however, but from engineering improvements which have been made to Still Creek. Stormwater containment has been the primary human use of Still Creek since the beginning of urbanization in the region. The upper reaches of Still Creek were completely enclosed in the Collingwood storm sewer by 1924 and channelization of much of the remainder of the creek soon followed. Recently, through the efforts of the Rupert-Cassiar Neighbourhood Association and others, recreational values associated with Still Creek have become more prominent in the last decade (Dawson, Flaherty, and Gang 1985). This 21 Figure 3.2 Horizontal profile of Still Creek and the Brunette River (adapted from Hall et al. 1976). anusAy aiiaurua peo^iriiON -urep ooquB^ 35fBq XqBoing a n u s A V Suiiiads anusAy sBjgnoQ anusAV uopSunnAV anusAy ajounir) pBO^ AjBpunog — ABAVqSlH A\9IApUW9 peo^ uipy o o I oo u on o CM CM E o 3 o co E 8 a o c CO <n Q (ui) uo|ieAa|3 2 2 growing community interest has prompted commitments from both the cities of Vancouver and Burnaby to keep the remainder of the stream open. 3.2 Vegetation and wildlife A recently completed Environmentally Sensitive Areas (ESA) study provides an excellent enumeration of the native and introduced species in the watershed (Gardner Dunster Associates Ltd. 1992). Significant changes have occurred in wildlife populations following the clearing and settlement of the region: the Coastal Blacktail Deer is the only remaining large mammal resident, while salmon populations have been completely eliminated in Still Creek. Salmon populations had also disappeared from the Brunette River during the 1960's and 1970's; however, efforts by the Sapperton Fish and Game Club to clear river debris, eliminate point sources of contamination, and re-stock from hatcheries reversed the trend and by the late 1980's 400-500 Coho spawners returned to the Brunette. The last spawning year (1994) proved disappointing as the returning Coho spawners numbered less than 100 (Rudolph 1995). 3.3 Demography and land use Prior to the arrival of European immigrants in the 1860's, the watershed was used solely as the hunting and fishing grounds of the Squamish and Kwantlen Indians. From the period of initial immigration to 1900, much of the area had been logged and agriculture became the dominant land use activity (Gardner Dunster Associates Ltd. 1992). The extension of the Great Northern railroad and the introduction of streetcar service from New Westminster to Vancouver, in 1904 and 1912 respectively, allowed residential development to proceed in the watershed while still retaining the rural character. The first zoning of industrial land in the Still Creek area in 1947 marked the beginning of a prolonged commercial and industrial expansion (Dawson et al. 1985). 23 Absolute growth in population peaked in the decade between 1951 and 1961 (Table 3.1), effectively defining the general urban land use structure of today. While population growth rates have since declined, employment in the area has grown disproportionally. In 1981, for the first time since urban expansion began, employment in Burnaby equalled the total resident work force. Since that time employment has continued to grow more rapidly than population and the area has become a regional employment centre (City of Burnaby 1987). The nature of employment in the watershed has changed over the last 20 years, as illustrated in Figure 3.3. Manufacturing now contributes a much smaller share of total employment, while the services sector has increased its share. Although the relative importance of manufacturing has declined, total employment in the sector has in fact increased since 1971. Growth in population and industry is guided by a municipality's official community plan (OCP). Burnaby's OCP, produced in 1987, advocates continued growth under certain constraints. Industrial and commercial growth is expected to be accommodated within the same land base, utilizing in-filling and other densification techniques. Expected modest Table 3.1 Population and employment in Burnaby and the Brunette River Watershed since 1892 Year Burnaby Watershed Watershed Population1 Population2 Employment2 1892 250 1921 12,900 1951 58,400 1961 100,200 1971 125,700 120,000 40,200 1981 136,500 133,000 68,000 1991 155,200 156,000 88,300 1. 1892:(Northcote and Luksun 1992);1921-1981: (City of Burnaby 1987); 1991:(Statistics Canada,1991) 2. Amalgamated from traffic zone data used in GVRD traffic study (Krajczar 1994).) 2 4 Figure 3.3 Employment distribution in the Brunette River Watershed in 1971, 1981, and 1991 i. &utilities Year Source: Amalgamated from traffic zone data used in GVRD traffic study (Krajczar 1994). increases in population (projected to reach 163,000 people by 2001) wil l occur by converting some institutional and open areas to apartment and group housing while leaving existing low-density residential areas intact. An examination of growth rates and land zoning over the last 20 years shows the results of Burnaby's growth strategies. Population and employment have grown by 30% and 120% respectively, yet only minor changes in land zoning have accompanied this growth (Table 3.2). Actual land use, rather than zoning, may have changed more significantly and will be examined in this study. Automobile traffic has rapidly increased throughout British Columbia since 1950 (Figure 3.4), particularly in the heavily populated lower mainland region of the province. The associated problems of roadway congestion and air and water pollution are a major concern in the Brunette watershed as a result of large traffic volumes throughout. 25 Table 3.2. Zoned land use areas in the Brunette River Watershed in 1973 and 1987 Land Use Zone 1973 % 1987 % residential 51.6 49.6 industrial 15.3 13.9 commercial/ comprehensive 5.0 11.6 agricultural 8.0 0.3 park 13.5 19.5 open space 6.3 3.8 Source: Duynstee (1990) Figure 3.4 Automobile registrations in British Columbia from 1912 to 1993 3000 1910 1930 1950 1970 1990 Year Source: Statistics Canada (1993) 26 3.4 Environmental stewardship The natural areas within the watershed are highly valued by the surrounding communities for several different purposes. Deer Lake Park, surrounding much of the 32 hectare lake, provides a recreational and cultural focus for the City of Burnaby. The park contains a museum, art gallery, and outdoor amphitheatre and provides access for picnics, fishing, and boating. Once a very popular swimming area, the lake has been closed to swimming in recent years because of high fecal coliform levels. Burnaby Lake also provides recreational opportunities in the form of rowing, picnicking, and wildlife viewing. Trail systems surrounding Still Creek and Brunette River and on Burnaby Mountain provide opportunities for experiencing nature in an urban setting. Complementing the recreational values associated with the natural areas are values associated with maintaining and enhancing aquatic habitat. Community groups, educational institutions, and many different levels of government have become involved in watershed stewardship with the goal of protecting these highly valued natural components of the area. Urban wastes have the potential to degrade the remaining natural areas to the detriment of all public uses and therefore receive the attention of government agencies from municipal to provincial. While neither raw or treated sewage is directly discharged within the watershed, mistaken or illegal cross-connections between the sewage and stormwater systems offer the potential for stream contamination. The G V R D is responsible for monitoring storm sewers to detect this contamination and alert the affected municipalities. Individual municipalities each manage their urban runoff and the necessary storm sewer systems. As part of this responsibility, the Burnaby Health Department regularly tests the waterways and lakes for fecal coliform contamination. Burnaby's 1993 State of the Environment Report recognized the aquatic hazards posed by urban runoff and pledged to work towards improving the runoff quality by supporting urban runoff research and contributing to the GVRD's Liquid Waste Management Plan (City of Burnaby 1993). Permitted industrial discharges are a relatively minor contaminant source to the watershed. At present, two petroleum tank farm operations discharge their treated 27 stormwater to Eagle Creek (locations noted on Figure 3.5) and are not considered a risk to aquatic life (Swain 1989). Solid waste is transported for disposal outside the watershed — either to a landfill or incineration at the GVRD's Burnaby facility. A large closed, regional landfill located near the Brunette River at Braid Street discharges collected leachate outside the watershed to the Annacis sewage treatment plant. Hazardous waste generated within the watershed is either stored, transported out of the province for disposal, or recycled. Emissions to the air are recognized as a large problem in the entire lower Fraser region. Control of these emissions is particularly challenging because the largest source of emissions emanate from the operation of motor vehicles (Environment Canada 1992). Within the Brunette watershed, all levels of government are addressing the problem, with the G V R D assuming the lead coordinating role. Responsibility for the protection of aquatic habitat is partially held by government agencies, but community groups and educational institutions also play a major role. The provincial Ministry of Environment, Lands and Parks monitors stream chemistry and contaminant levels at 5 stream and lake locations which are marked on Figure 3.5. Annual reports comparing measured contamination to objectives have been issued since 1990. Physical restoration of aquatic habitat has been accomplished mainly by non-governmental groups over the past 30 years. The Sapperton Fish and Game Club has taken a lead role in this area through their work in the Brunette River. As a result of their actions, which have recently been supported by the federal government's Salmonid Enhancement Program, populations of salmon and trout have been re-established in the lower end of the watershed. Most recently, the British Columbia Institute of Technology has embarked on an ambitious project of wildlife enumeration and habitat enhancement in the watershed (British Columbia Institute of Technology 1994). Throughout the entire region, "green spaces", which provide wildlife habitat, recreational opportunities, and a respite from urban congestion, are highly valued — 28 Figure 3.5 Locations of M O E monitoring sites, effluent discharges, and flow gauge sites 29 especially in the face of rapid urban development. Protection of the remaining green spaces in this watershed is supported by Burnaby's OCP, GVRD's Green Zone plan (Greater Vancouver Regional District 1989), and two environmentally sensitive area studies (Federation of B .C . Naturalists 1994, Gardner Dunster Associates Ltd. 1992). 3.5 Contaminant record A large database of contaminant information has been collected in this watershed over the last 25 years. The provincial Environment Ministry has been responsible for much of this information through their routine monitoring program and selected studies. University studies in the watershed are another significant source of contaminant information. Because of its proximity to the University of British Columbia and Simon Fraser University, this highly urbanized watershed has served as an ideal research case study of land use-contaminant relationships. The following sections summarize studies which have added to the contaminant record and briefly describes some of the major findings. 3.5.1 Trace metals Regular monitoring of the streams and lakes in this watershed over the last 20 years has found that surface water criteria designed to protect aquatic life have often been exceeded for Pb, Cu, Zn, and Cr, and infrequently exceeded for Hg (Swain 1989). Recognizing the presence of significant trace metal contamination, there have been at least 16 studies examining the problem within the watershed (Table 3.3). The most comprehensive survey of the watershed was conducted by Westwater Research in 1973 (Hall et al. 1976). While this study found that street dirt across all land uses was a significant contributor to stream contamination, it also implicated industry as the cause of the most severe contamination in the middle reaches of Still Creek. This study provided important baseline data which have been used for direct temporal comparison by Duynstee (1990) and in this report. Duynstee's work, which attempted to correlate trace metal changes with land use using a Geographic 30 Table 3.3 Studies examining trace metal levels in the Brunette watershed Report Medium Relevant Focus and Findings Studied 1. Benedict et al. 1973 surface water, stream sediments 2 sites sampled for each media; significant metal contamination detected. 2. Hall et al. 1974 surface water 3 sites monitored; high Pb, Cu, and Zn levels. 3. Hall et al. 1976 surface water, stream sediments, street sediments stream sediments (36 sites) contaminated relative to Fraser River; middle reach of Still Creek (industrial) most heavily contaminated; street dirt (26 sites) a major contributor to stream contamination. 4. Bindra et al. 1977 benthos, stream sediments 13 sites in watershed; related sediment concentrations to invertebrate concentrations. 5. McNeill 1978 surface water 12 sites monitored once/week for 6 months; insufficient sampling frequency to relate contamination to land use. 6. Anderson 1982 stormwater stormwater monitored at 12 sites; most contaminated in commercial areas, followed by industrial, residential, and open space. 7. Mathewes et al. 1982 Deer lake sediment core high Pb, Cu, and Zn at surface, Cu peaked 20 cm below surface - related to algicide treatment. 8. Bindra 1983 stream sediments 2 contaminated sites on Still Creek; potential for metal release if sediments disturbed. 9. Lawson et al. 1985 stormwater, street sediments 1 Still Creek (industrial) location; Pb, Zn higher in runoff than sewage; dustfall is a significant source. 10. Munteanu 1987 Deer Lake sediment cores examined lake restoration techniques, including dredging; large areal variability in sediment metal concentrations. 11. Swain and Walton 1988 stream sediments Still Creek (1 site) contaminated: Pb, Cu, Zn, Cd, Hg; Brunette River (2 sites) contaminated: Pb, Hg, Zn. 12. Duynstee 1990 stream sediments re-sampled Hall et al. 1976 streams sites; Pb decreased in some sites, Hg and Mn increased throughout. 13. BCIT 1992 fish livers 3 Carp livers from Burnaby Lake; MOE (fish muscle) objectives exceeded for Pb and Hg. 14. Matthews 1994 surface water 3 sites (2-Still Creek and 1-Eagle Creek) monitored during 2 storms. 15. B.C. Environment 1992, 1993a, 1993b, 1994 surface water, stream sediments, fish tissue water: Pb, Cu, Zn above criteria; Cr, Hg below, sediments: Pb, Cu, Zn above criteria, Hg incomplete, fish tissue: Hg above criteria (11 fish); Pb below. 16. Smith, 1994 stream sediments 5 sites on Still Creek (also toxicity studies). 3 1 Information System (GIS), discovered significant increases in Mn, Zn, and Hg, and large decreases in Pb at some stations. Nevertheless, the land use analysis was inconclusive. Three studies examining stormwater and/or surface water during storms have been executed in the watershed (report #'s 6, 9, 14 in Table 3.3). Anderson (1982) found correlations between stormwater quality and general land use but suggested more valuable information could be obtained by examining the degree of imperviousness and traffic density in the catchment areas. Lawson's (1985) study discovered higher metal loadings in an industrial area than predicted and also higher than a residential area in Vancouver he previously studied. The most recent study by Matthews (1994) noted contaminant changes from previous studies, including lower Pb loadings. The Westwater Research Centre has continued storm monitoring of Matthews' 3 stream locations and 4 additional street locations (Hall 1995). A history of trace metal contamination has been provided by Mathewes (1982) pollen and geo-chemical analysis of a 60 cm Deer Lake sediment core. Munteanu's (1987) analysis of 8 Deer Lake cores confirms the significant contamination of recent sediments in the deepest part of the lake where Mathewes core was taken. The other 7 cores, however, exhibit significant differences in sediment texture and trace metal content. The cores in shallower areas, which were not directly influenced by entering streams, exhibited higher organic contents, lower detrital material, and lower trace metal concentrations. This work suggests that trace metals from the catchment area, bound to detrital minerals, become very concentrated in a small deep area of the lake bottom. Beginning in 1990, the British Columbia Ministry of Environment began an annual water and sediment monitoring program (report #15 in Table 3.3) which includes 5 sites in the Brunette watershed (see Figure 3.5). The sediment analysis is performed on whole samples, making it difficult to compare to many of the previous studies which have examined fractionated sediments. With 4 years of monitoring completed, this program has confirmed the presence of significant Pb, Cu, and Zn sediment contamination and possible Hg 32 contamination. Most importantly for the future of the watershed, long-term objectives have been set through this program as a remediation goal. 3.5.2 Organic and microbial contaminants Detectable levels of chlorinated organic compounds such as polychlorinated biphenyls (PCB), l,l-bis(4-chlorophenyl)-2,2,2-trichloroethane (DDT), and chlorinated phenols have been found throughout the watershed (reports #1, 2, and 5 in Table 3.4). While the overall watershed contamination appears to be less severe than in other locations in the Fraser River estuary (reports #5, and 6), high concentrations have been detected in Still Creek. Polycyclic aromatic hydrocarbons (PAH) comprise a large number of compounds which are formed during combustion processes and can be extremely toxic at low concentrations. Morton's (1983) study found elevated levels in street dirt relative to stream sediments, suggesting street material may be a significant contaminant source. Stream sediment concentrations were comparable to values found in a heavily urbanized watershed in Boston. Although Hall et al. (1984) detected lower levels in one Still Creek site, Morton's more extensive study, which presents evidence of P A H bio-accumulation, indicates further monitoring may be necessary. Fecal coliforms are a group of bacteria which indicate the presence of fecal material and the possibility of pathogens. As a result of regular M O E monitoring, it has been recognized for many years that streams and lakes throughout the watershed contain fecal contamination which prevents primary contact recreation activities in the water (reports #1, 7, 8, and 9 in Table 3.4). Urban runoff is a significant contributor of fecal material to streams (Swain 1989), but extremely high values in Still Creek are likely caused by cross-connections between the sewage and stormwater systems (Coastline Environmental Services 1987). The G V R D is currently conducting a monitoring program in an attempt to identify the problem areas and alert the responsible municipalities (Greater Vancouver Regional District 1994b). 3 3 Table 3.4 Studies examining organic and microbial contaminants in the Brunette watershed Report Contaminant / Relevant Focus and Findings Medium 1. Hall et al. 1974 chlorinated pesticides, bacteria / surface water 3 sites sampled; all pesticides below d.l. (0.005 ug/1) except lindane : 0.013ug/l at one site; fecal coliforms high at each station, very high in Still Creek: 6670 MPN (ave.) / 100 mis. 2. Hall et al. 1976 chlorinated hydrocarbons / stream and street sediments 26 stream stations, 26 street stations; PCB's ranged from <d.l. - 780 ug/kg d.w. in stream sediments, D D T ranged from <d.l. - 135 |ig/kg d.w. in stream sediments; comparable to heavily industrialized basins in New Brunswick and Lake Michigan area. 3. Anderson 1982 hydrocarbons / stormwater 12 sites monitored; correlated hydrocarbons to traffic volume and vehicle residence time. 4. Morton 1983 PAH / stream and street sediments, benthos 5 stream stations (Still Creek), 8 street stations; individual compounds ranged from 55 ug/kg to 39 ug/g; street dirt significant source of PAH to streams; indication of bio-accumulation in oligochaete worms. 5. Hall et al. 1984 PAH, chlorinated phenols, phthalate esters / stream sediments 1 site sampled (Still Creek); levels lower than several other sites in Fraser estuary, highest PAH compound in Still Creek was pyrene : 108 ug/kg d.w. 6. Swain and Walton 1988 chlorinated hydrocarbons / stream sediments Still Creek (1 site), Brunette R.(2 sites); PCB measurements below d.l.( 0.03 ug/g d.w.), chlorinated phenols below d.l. (0.0005 txg/g d.w.). 7. B.C. Environment 1992, 1993a, 1993b, 1994 bacteria, dissolved organic matter / surface water fecal coliforms high throughout watershed, not suitable for primary contact recreation; dissolved oxygen objectives are not met periodically throughout the watershed. 8. GVRD, 1994b bacteria / storm sewers 52 sampling sites in sewers discharging to Still Creek; very high dry weather fecal coliform levels indicates sewage contamination. 9. City of Burnaby, 1994 bacteria / surface water fecal coliforms high throughout watershed, not suitable for primary contact recreation. 34 A problem often associated with high fecal coliform numbers is the depression of dissolved oxygen levels in streams. Bacterial degradation of dissolved organic matter, introduced from fecal contamination or excessive algal growth, depletes stream oxygen levels producing stressful or toxic conditions to fish. Monitoring over the previous 20 years has shown that Still Creek has often (7 of 38 measurements) experienced dissolved oxygen levels below 6 mg/L — a level known to impair fish habitat (Swain 1989). While the problem is less severe in other streams, the Brunette River occasionally fails to meet the M O E long-term objective of 8 mg/1 (6 of 43 measurements between 1973 and 1984; Swain 1989). 3.5.3 Nutrients Major nutrients such as nitrogen and phosphorus — essential for aquatic productivity — can, in excess, promote algal growth which is particularly harmful to salmonid populations (Swain 1989). Nitrogen can also exist in two chemical forms (ammonia and nitrite) which are toxic to fish under certain pH conditions. Nitrite and ammonia objectives have been met at all 5 M O E monitoring stations over the last four years. The objective for total phosphorus in Burnaby Lake was exceeded each year. Chlorophyll-a has been measured in the Brunette River in 1992 and 1993 as an indicator of periphyton growth — the M O E objective was met in 1992 and exceeded in 1993 (B.C. Environment, (1992, 1993a, 1993b, 1994)). Excessive algal growth (eutrophication) has occurred in Deer Lake in recent years; interfering with summer recreational activities associated with the lake. Northcote and Luksun (1992) present a review of studies pertaining to the eutrophication of Deer Lake and identify urban runoff as a major external nutrient source. 3.5.4 Spills Hazardous products and wastes are transported through the watershed every day on either the rail or transportation corridors. While the probability of accidental releases may be low, the consequence of one single release to aquatic life can be catastrophic. Two recent examples highlight this fact. Spawning grounds in Stoney Creek were temporarily destroyed in 1989 as a result of a 13,000 litre diesel fuel spill caused by a rail collision in 1989 (Jackson 35 1989). Prior to this spill, a fish kill was observed in the Brunette River in 1985 — believed to be caused by the dumping of chlorophenols into a storm drain (Swain 1989). The risk of an accidental or deliberate spill is always present in the Brunette Watershed. The water supply system serving the Greater Vancouver area also has the potential to harm aquatic habitat in the watershed in the event of a spill. The risk comes not from the water source but rather from the disinfecting chemicals added to prevent bacterial growth in the distribution system. At present chlorine is used as the chemical agent and could potentially harm fish in the event of a line break. Other disinfecting chemicals such as chloramine, with a much longer disinfecting life, have been considered as a replacement for chlorine and could severely reduce fish populations i f released through the distribution system. A large water main running under Burnaby Lake is an obvious potential risk area. 3.6 Ecosystem Health A large body of literature which has examined contaminant levels in the Brunette Watershed is accumulating (summarized in the previous section) but much less information is available regarding the actual effects contaminants may be having on aquatic organisms. An early preliminary study by the Westwater Research Centre found benthic communities dominated by a very few oligochaete worm species in Still Creek and suggested the low community diversity was the result of high contaminant levels known to exist in the sediments (Hall 1976). This work has not been expanded upon to determine whether the low community diversity is a result, primarily, of contaminant levels, or the altered flow regime of the urban streams. The bio-accumulation of contaminants in the benthos is an indication of weakened ecosystem health and presents the possibility of bio-magnification i f transferred through the food chain. Bindra and Hall (1977) examined the relationships between metals in different geo-chemical phases of the sediment, and metal levels in benthic organisms in the Brunette Watershed, Salmon (Langley) watershed, Ladner Sidechannel (Fraser River) and Sturgeon 3 6 Banks. Bio-accumulation factors for Pb in benthos were almost 20 times higher in the Brunette watershed than the other areas while the factors for zinc and copper were approximately 5 times higher. Bio-accumulation in the periphytic algae was even higher than in the invertebrates. The study concluded that the geo-chemical form of the trace metals examined (Pb, Cu, Zn) was an important factor in predicting bio-accumulation rates. A subsequent study examining P A H in Still Creek observed bio-accumulation in oligochaetes (Morton 1983). Bioassays provide controlled and comparable procedures for measuring the effects of contamination on organisms. Table 3.5 provides a summary of three studies performed in the watershed. The first study (Anderson 1982) performed acute toxicity tests on water fleas (Daphnia pulex) using the stormwater from 12 sites across the various land uses in the watershed (commercial(C), industrial (I), residential (R), and open space (O)). His results showed commercial areas to be the most toxic with aggregate toxicity values following the sequence C > I > R » 0 . Anomalies to this general pattern existed at some sites and suggested specific factors such as traffic volume may be more directly related to toxicity than general land use categories. Trace metal concentrations in stormwater were a significant factor affecting toxicity. Lawson et al. (1985) utilized luminescent bacteria (Microtox®), water fleas {Daphnia magna) and rainbow trout as assay organisms in their study of stormwater in an industrial area surrounding Still Creek. No acute toxicity was measured in stormwater during wet weather discharges however some toxicity in dry weather discharges was observed with Microtox® and Daphnia. Smith (1994) examined the relationships between sediment and pore water contaminants (trace metals, P A H , ammonia) and bioassay results for Daphnia magna, sediment dwelling invertebrates (chironomids) and Microtox®. Five stations on Still Creek were compared to reference sites in a relatively undisturbed Vancouver stream (Musqueam Creek). No measured differences in toxicity between Still Creek and the reference site were found for Daphnia magna, while slightly higher chironomid toxicity was observed in Still 37 Creek sediments compared to the reference sediments. All 5 Still Creek sediments (and 2 of 5 reference sites) induced toxicity in the Microtox® test. High ammonia levels at one site appeared to affect toxicity, yet no statistically significant relationships between sediment contaminants and bioassay results were found. Several studies examining the ecosystem health of the watershed are currently being undertaken as a contribution to UBCs Eco-Research project. These studies include: an examination of fish communities, an investigation of fish organ abnormalities, sediment bioassays, and metal uptake by benthic invertebrates (Westwater Research Centre 1994). This work by the Westwater Research Centre will help to put the results from this trace metal-land use study in a biological perspective. Table 3.5 Bioassay results from the Brunette watershed Report Medium Bioassay Rainbow Trout Daphnia Chironomids Microtox® Anderson 1982 Storm water 96hr (wet weather) L C 5 0 - 10% Lawson et al. Storm water non-toxic 48hr 50 sec 1985 (dry weather) L C 5 0 - 26% EC 5o- 61% Storm water non-toxic non-toxic non-toxic (wet weather) Smith 1994 Streambed 14 day 14 day 5min sediments Survival Survival E C 5 0 - 0.3% 73% (ave.) 37% (ave.) Streambed 15min pore-water EC50 - 38% Note - maximum toxicity is reported except where averages are noted. 38 4. METHODS The methods for this study have been chosen to provide data which can be directly compared to earlier studies in the watershed. A large number of sample preparation methods and analytical techniques have been utilized in this study. In order to minimize confusion, sub-section 4.2 (sample preparation and analysis) outlines all sample preparation steps and notes the analytical technique used for each sediment category. Sub-section 4.3 (sediment analytical techniques) can then be used as a reference to obtain details of each analytical technique. 4.1 Field methods 4.1.1 Streambed sediments collection Thirty-three sampling stations, indicated in Figure 4.1, are located in the same locations with the same numeric identifiers as those used by Hall et al. (1976) with the following exceptions : (i) stations 12, 18, 22, and 36 could not be sampled because stream is now enclosed, (ii) station 29 was sampled approximately 400 metres upstream of Hall's location, (iii) station 37 was added to compare with the study by Duynstee (1990). The closest street intersections and location co-ordinates are listed for each station in Appendix A . Preliminary samples of 2 stations on Still Creek (#33, and #35) were taken on July 21, 1993 to provide an indication of the optimum number of samples per site and particle fraction to use for metals analysis. Five separate surface sediment grab samples per site were collected using an aluminum pot affixed to a 3 metre wooden pole (Hall et al. 1976). Pool areas within a 20 metre stream span were sampled to ensure an adequate amount of fine fraction material was collected. Each sample was first passed through a coarse (2 mm) plastic sieve prior to placing in high wet-strength plastic bags. Bagged samples were 39 Figure 4.1 Location of streambed sediment sampling stations transported back to the laboratory and stored for up to 2 days at 4 °C prior to sample preparation. Streambed sediments were collected at each station during the period September 9-13, 1993. Except at station #1, three separate grab samples of streambed sediment were collected at each station using the preliminary sampling procedure noted above. Samples at station #1 were obtained using an Ekman grab. Bagged samples were transported back to the laboratory and stored in a freezer for up to 3 weeks before further processing. Stations #10, #15, and #37 were resampled on August 25, 1994, using the same procedures, to provide an indication of short-term temporal variations in sediment contamination. 4.1.2 Street sediment collection Street sediments were also collected at the same locations using the same identifiers as those used by Hall et al. (1976) with one exception. Hall's green space location "G3" was not sampled because of a change in land use since the original study. The 25 locations shown on Figure 4.2 (street locations given in Appendix A) were sampled during the period September 9-14, 1993. One sample from each area was collected using a broom and small plastic shovel, sieved to retain particles less than 2 mm, and finally placed in plastic bags. Three separate samples were taken from different corners of the intersection at North and Lougheed Road (station C4) to provide an indication of variability within areas. Bagged samples were returned to the laboratory and stored in a freezer for up to 3 weeks before thawing for sample preparation. 4.1.3 Lake sediment cores Two different methods were used to take lake cores. Short cores were taken from Deer and Gwendoline Lakes during the fall of 1993 while long cores were taken from Burnaby and Deer Lakes in 1994. 4 1 Figure 4.2 Location of street sediment sampling stations Short cores Core samples, obtained using a 2 inch (5.1 cm) gravity corer, were taken from two lakes on October 2, 1993. Two cores each, taken within two metres of each other, were retrieved from Deer Lake (Figure 4.3) and Gwendoline Lake in the U B C Research Forest (Figure 4.4). The plastic core tubes were removed from the core barrel and kept upright in the boat until reaching shore. On shore, the cores were extruded vertically with sections captured in plastic bags. The Deer Lake cores were extruded in 2 cm sections while the Gwendoline Lake cores were extruded in 2 cm sections to a core depth of 20 cm, and then 5 cm sections for the remainder of the cores. The bagged samples were then transported to the laboratory and frozen. Long cores A modified Livingstone 2 inch (5.1 cm) piston corer was used to obtain long cores from both of Burnaby and Deer Lakes on July 7, 1994 (Figure 4.3). Supported at the lake surface by a metal collar and chain hoist fastened between two small, linked fiberglass boats, the corer was attached to steel rods which were used to force the sampler into the sediments. The core barrel travelled through the water column in an external casing which was set into the surface lake sediments to ensure the sampler re-entered the hole cleanly after each extraction. Faegri and Iverson (1989) provide a detailed description of the sampling mechanism. Each retrieved core section, of approximately one metre length, was extruded whole from the core barrel onto a piece of plastic tubing which had first been cut in half longitudinally and then lined with plastic wrap. A preliminary examination of the stratigraphy was completed and the core was then wrapped in two layers of plastic and a final layer of aluminum foil. The section was protected for transport by enclosing in the two half sections of plastic tubing. The protected core sections were then transported to the laboratory and stored for several weeks at 4 °C prior to sample preparation. Table 4.1 provides a summary of the core lengths and lake depth for both the long and short core sampling sites. 4 3 Figure 4.3 Location of sediment core sampling stations in Burnaby Lake and Deer Lake 4 4 Figure 4.4 Location of sediment core sampling station in Gwendoline Lake ( 4) Table 4.1 Lengths of lake sediment cores and lake depth at each sampling site Location Lake Depth (m) Total Core Length (cm) # of Sections Burnaby Lake #1 1.5 138 2 Burnaby Lake #2 0.4 563 6 Burnaby Lake #3 0.5 92 1 Deer Lake (long) 5 153 2 Deer Lake (short) #1 5.5 14 1 Deer Lake (short) #2 5.5 14 1 Gwendoline Lake #1 18 60 1 Gwendoline Lake #2 18 30 1 4.2 Sample preparation and analysis 4.2.1 Stream and street sediments The frozen samples were thawed and then passed through a stainless steel (S.S) 180 iim sieve with the addition of a small amount of distilled water to increase the recovery of fine material. Sediment passing through the sieve was dried at 104 °C for 24 hours and then disaggregated. Sub-samples of this dried material were then taken for various determinations as outlined in Table 4.2. The nitric acid digests from the streambed samples collected in 1994 were analyzed by both flame AA (Civil Engineering laboratory, UBC) and ICP atomic emission spectroscopy (Soils Science laboratory, UBC) for comparative purposes. Additional sample preparation steps were included for the analysis of the 10 preliminary streambed sediment samples. Each sample was split in the laboratory in order to obtain two separate particle fractions for comparative metals analysis. One of each of the split samples were passed through a 180 ixm S.S. sieve while the remaining samples were passed through a 63 \im S.S. sieve. Small amounts of distilled water were added to increase 46 Table 4.2 Preparation and analysis of dried street and stream sediment sub-samples Sub-sample Preparation Analyses 1. heated in furnace at 550 °C for 2 hrs. Loss on Ignition (LOI) 2. digested with nitric acid and analyzed with flame Atomic Absorption (AA) spectroscopy in Civi l Engineering laboratory, U B C . Fe, Mn, Mg , Cu, Pb, Zn, Cd, C r , N i 3. digested with sulphuric and nitric acid and analyzed with cold vapour A A in Civi l Engineering laboratory, U B C . Hg 4. digested with cold, weak HC1 and analyzed with flame A A in Civi l Engineering laboratory, U B C (streambed samples only). Fe, Mn, Mg , Cu, Pb, Zn, Cd, C r , N i 5. digested with aqua regia solution and analyzed with flame A A in Civi l Engineering laboratory, U B C (15 samples only). Fe, Mn, Mg , Cu, Pb, Zn, Cd, C r , N i 6. dry sieved through a 63 |im sieve. a b fraction of silt and clay in the < 180 |im sediment fraction a a composite of the 3 samples from each stream station was analyzed. b silt and clay fraction from 6 samples were retained and analyzed for metals at Chemex Labs using an aqua regia digest and ICP-AES. the recovery of the fine fractions. Sediments passing through the sieves were dried at 104 °C for 24 hours and then disaggregated. Sub-samples of this dried material were then taken for various determinations as outlined in Table 4.3. 4.2.2 Lake cores Different sample preparation techniques were used for the long and short lake cores because of the differing field collection techniques. The long cores were better characterized as a result of arriving at the laboratory intact. 4 7 Table 4.3 Preparation and analysis of preliminary dried stream sediment sub-samples Sub-sample Preparation Analyses 1. heated in furnace at 550 °C for 2 hrs. Loss on Ignition (LOI) 2. digested with nitric/perchloric acid and analyzed by both flame AA and ICP atomic emission spectroscopy at Chemex Labs, North Vancouver. Fe, Mn, Mg , Cu, Pb, Zn, Cd, C r . N i 3. digested with aqua regia solution and analyzed by both flame AA and ICP atomic emission spectroscopy at Chemex Labs, North Vancouver. Fe, Mn, Mg , Cu, Pb, Zn, Cd, C r . N i 4. dry sieved through a 63 |im sieve (< 180 tim samples only). fraction of silt and clay in the < 180 iim sediment fraction Short lake cores Samples collected using the gravity corer were thawed and then dried at 104 °C for 24 hours. The dried samples were disaggregated and 3 sub-samples were taken for the following determinations: (i) LOI, (ii) nitric acid digestion and metals analysis using flame AA spectroscopy in the Civi l Engineering laboratory, U B C , and (iii) nitric / sulphuric acid digestion and Hg analysis using cold vapour AA in the Civi l Engineering laboratory, U B C . Long lake cores Each core section was carefully unwrapped and cut into longitudinal halves. One half was re-wrapped in plastic and aluminum foil and remains in storage at 4 °C. A detailed description of the stratigraphy was noted prior to removing samples from the remaining half. 48 Sub-samples were taken throughout the cores at sampling frequencies listed in Table 4.4. One sub-sample at each location within the cores was taken to determine percent moisture, LOI, and metal content. The samples which were combusted during the LOI determination were digested for metals analysis using nitric acid. This digestion method was used to avoid potential analytical interferences which could be caused by the high organic content of the lake cores. Metals determinations were performed at the Soils Science Department using ICP — atomic emission spectroscopy. One sub-sample was also taken every 2 cm throughout Burnaby Lake core #2 and sent to the Limnology research Centre at the 210 University of Minnesota for Pb radioisotope dating analysis. Table 4.4 Sub-sampling locations of long lake core sediments Core Location Core Interval (cm) Sampling Frequency (cm) Burnaby Lake #1 0-138 2 Burnaby Lake #2 0-85 2 85-180 10 180-563 20 Burnaby Lake #3 0-49 5 49-92 10 Deer Lake (long) 0-153 5 4 9 4.3 Sediment analytical techniques 4.3.1 Moisture content Approximately 3 to 5 grams of wet sample were initially weighed, dried at 104 °C for 24 hours and then re-weighed again. The percent moisture was calculated as the weight of water relative to the total wet weight. 4.3.2 Loss on Ignition (LOI) LOI was determined to provide a measure of the organic content of the samples. Empirical studies have shown a very strong correlation between LOI and organic carbon content — especially when LOI is greater than 10% (Hakanson and Jansson 1983, Reiberger 1992). The weighed dry samples were heated at 550 °C for two hours to burn off the organic material (APHA 1989) and LOI was calculated on the basis of sample dry weight. 4.3.3 Acid digestions (excluding Hg method) Sediments were digested over a hot plate using either nitric acid, nitric/perchloric acid solution, or aqua regia acid solution as described in Method 3030 of "Standard Methods" (APHA 1989). The precision and accuracy of the method was determined by performing duplicate digestions and analyzing certified reference material, BCSS-1. The weak acid digestion technique was based on the method used by Hall et al. (1976). 20 ml of 0.5 M HC1 acid was added to a test tube containing 2 grams of dried streambed sediment. The test tubes were sealed and placed in a mechanical shaker at room temperature overnight. After settling, the samples were filtered and the digest was made up to 40 ml with distilled water. The method precision was determined by performing duplicate digestions. 4.3.4 Metals detection techniques (excluding Hg) The metals Fe, Mn, Mg, Cr, Cd, Cr, Pb, Cu, and Zn were detected using atomic absorption (AA) spectroscopy with an air/acetylene flame at the Civi l Engineering laboratory at U B C and Chemex Labs, North Vancouver. The same metals were detected using ICP 50 atomic emission spectroscopy (AES) at Chemex Labs, North Vancouver and the Soils Science laboratory at U B C . Direct comparisons between detection techniques were performed at Chemex Labs and between the two U B C laboratories. Method detection limits are listed in Table 4.5. Limits were calculated by measuring the standard deviation produced from repeated determinations of blanks and very low concentration standards. For metals which were found at least 3 orders of magnitude above typical detection limits, the manufacturer suggested detection limit was used. The Cd detection limit on the flame A A (Civil Engineering laboratory) was very high which has unfortunately censored almost the entire stream and street sediment dataset. 4.3.5 Mercury analysis Samples were digested and analyzed using a technique modified from E P A Method 7471 (EPA 1986). Triplicate digestions of each sample were performed using approximately 0.1-0.2 grams of dried sediment which were weighed into 40 ml glass reaction tubes. 2 ml of acid solution (consisting of 35% concentrated H 2 S 0 4 , 15% concentrated H N 0 3 and 50% distilled water) and 2 ml of 5% K M n 0 4 (in distilled water) were added to the reaction tubes. The tubes were covered with aluminum foil and autoclaved for 15 minutes at 121 °C. The samples were allowed to cool and then 5 ml of a prepared NaCl-hydroxylamine sulphate solution was added. This solution was prepared by dissolving 120 grams of NaCl and 120 grams of (NH 2 OH)2H 2 S04 in 1 litre of distilled water. Samples which contained a high amount of organic matter (>20% LOI) were then sparged with compressed air for 2-3 minutes to remove potential interferences. Finally, 0.2 ml of 10% stannous chloride was added and the tube was immediately attached to the purging apparatus. The mercury vapour passed through a 30 cm dual glass cell where the absorbance at 254 nm was measured with a Pharmacia U V atomic absorption spectrophotometer. The method detection limit was 30 lig/kg dry weight. 5 1 Table 4.5 Method detection limits for metals analyses METHOD / SITE METALS (mg/kg dry weight) Fe Mn Mg Pb Cu Zn Cd Cr Ni Flame AA (Thermo Jarrell Ash 22)/Civil Engineering UBC 0.05t 0.031 0.005f 15 1.5 0.05t 3 3 6 Flame AA (Unicam UX9100X)/ Chemex Labs 10 1 10 1 1 1 0.1 1 1 ICP-AES (Thermo Jarrell Ash ICAP 61) / Chemex Labs 100 5 100 2 1 2 0.5 1 1 ICP-AES (Thermo Jarrell Ash ICAP 61) / Soils Sciences UBC 2 0.2 0.004t 3 1 0.1 0.2 0.3 0.6 t calculated using manufacturer (Jarrel-Ash) estimates. 4.3.6 2 1 wPb Radioisotope dating analysis This sediment dating method is applicable for sediments deposited within the last 150 to 200 years. The method, as performed by D. Engstrom at the Limnological Research Centre, University of Minnesota, is modified from Eakins and Morris (1978). The samples were digested using concentrated HC1 acid and plated onto silver planchets from a 0.5 N HC1 acid solution. The activity of 2 1 0Po (a granddaughter product of 210Pb) was then measured with Si-depleted surface barrier detectors and an Ortec Adcam™ alpha spectroscopy system. After subtracting the supported activity from each sample, the sediment dates and sedimentation rates were calculated using the constant rate of supply model (Appleby and Oldfield 1978). Confidence intervals were calculated by first-order analysis of counting uncertainty (Binford 1990). 5 2 4.4 Land use and traffic analysis The analysis of changing land use and traffic activity was accomplished with the aid of Geographic Information Systems (GIS). The Terrasoft GIS program was used for the analysis of land activity and land cover for both the 1973 and 1993 time periods. Digitized 1:20000 T R I M (Terrain Resource Inventory Management) map sheets 92G.025 and 92G.026 were purchased from the Surveys and Resource Mapping Branch, Province of British Columbia, and imported into the Terrasoft program for use as a base map. The T R I M map contains such features as lakes, streams, transportation networks, large buildings, and forested areas. The Maplnfo GIS program was used for traffic analysis and graphical presentation of trace metal data. 4.4.1 Sub-catchment areas In order to examine the relationship between land use activities and the spatial pattern of contamination, the watershed was divided into smaller sub-catchment areas. Catchment areas for each stream sampling station could not be defined because of the alteration of the natural drainage by the storm runoff collection system. Twenty-nine catchment areas, defined on stormwater drainage maps by the municipalities of Burnaby and Coquitlam and the Greater Vancouver Regional District, were digitized into the Terrasoft GIS program (Figure 4.5). Table 4.6 lists the size of the drainage area and the receiving watercourse for each of the catchment regions. 4.4.2 Land activity and cover Land activity and cover information was digitized into the Terrasoft GIS program for the 1973 and 1993 time period using aerial photos with reference to land zoning maps. The 1973 information was derived from 1:2400 black and white aerial photos while the 1993 information was derived from a 1:12,000 colour aerial photo flown in 1991. Recent land use changes in the Oakalla prison lands (bordering Deer Lake) made since 1991 were digitized 53 Figure 4.5 Sub-basins in the Brunette watershed Table 4.6 Size and description of each sub-basin in the Brunette Watershed Sub- Area Stream stations Receiving basin (hectares) in Sub-basin Watercourse 1 687 33, 34, 37 Still Creek 2 522 Still Creek 3 387 31,32 Still Creek 4 370 29, 30 Still Creek 5 839 21,22 Deer Lake 6 42 Still Creek 7 41 Still Creek 8 66 Still Creek 9 228 25, 26, 27, 28, 36 Still Creek 10 46 Still Creek 11 28 35 Still Creek 12 175 23 Still Creek 13 121 Burnaby Lake 14 87 19, 20 Burnaby Lake 15 443 24 Still Creek 17 75 Burnaby Lake 18 68 Burnaby Lake 19 568 13, 14,15, 16 Burnaby Lake 20 118 18 Burnaby Lake 21 87 17 Burnaby Lake 22 64 Burnaby Lake 23 57 Burnaby Lake 24 251 Brunette River 25 70 Brunette River 26 137 11, 12 Brunette River 27 789 7, 8,9 Brunette River 28 34 10 Brunette River 29 741 1,2, 3, 4 ,5,6 Brunette River Note : Burnaby Lake (#16) has an area of 132 ha. but is not considered a sub-basin. using Burnaby zoning maps. Land activity was designated using the British Columbia Land Use (BCLU) classification system as described by Sawicki and Runka (1986). Land cover designations were modified from the B C L U system to account for the nature of this study. A l l land was classified as either permeable, impermeable or water. Permeable land was sub-divided into grass, gravel, and forested areas, while impermeable areas were subdivided into buildings or paved surfaces. It was impractical to digitize land permeability information for all low-density residential areas and industrial/commercial areas, 55 therefore bulk permeability estimates were derived using test areas. Five different low-density residential test areas, approximately 2 blocks square, were digitized from the 1973 1:2400 aerial photos to derive an average permeability estimate. Three industrial/commercial test areas were digitized from both 1973 and 1992. No significant difference was found between years, therefore one average industrial/commercial permeability estimate was derived. Details of the test areas and permeability estimates are included in Appendix B. 4.4.3 Traffic analysis Traffic analysis was facilitated with the use of G V R D ' s E M M E / 2 transportation model (Greater Vancouver Regional District 1992b). The regional model employs 445 traffic zones as well as the road and transit network. Traffic zones, chosen to correspond with census tract boundaries, contain demographic data used for model input. This study utilized one model output, representing A . M . peak hour traffic in the fall, 1992. The road network and associated traffic data and traffic zone demographic data were located within the Maplnfo GIS program. Modifications to the G V R D model output were required for the purposes of this study. Bi-directional traffic estimates were combined into one volume estimate for each road segment. Figure 4.6 graphically illustrates the combined traffic volumes for the road network within the study watershed. Daily traffic estimates were obtained by multiplying the A . M . peak hour volumes by peak hour expansion factors determined from screenline studies (Greater Vancouver Regional District 1994a). Expansion factors were extrapolated from an adjacent street if screenline information was not available. Finally, a traffic density index — vehicle kilometres per day — was calculated for each road segment by multiplying the segment length by the traffic volume. Using the watershed and sub-basin overlays exported from the Terrasoft program, cumulative traffic indices were calculated for the whole watershed and for each sub-basin. 56 The same traffic indices were generated for 1973 using the identical GIS road network. Road segment traffic volumes were updated from municipal and provincial highway data (summarized by Duynstee, 1990) when available. A traffic estimate for roads which were not monitored in 1973 was derived by applying the average traffic zone 73/92 traffic ratio to the 1992 traffic volume. 4.4.4 Demographic data Demographic data for each traffic zone (Figure 4.7), compiled originally by Statistics Canada, was also received as part of the traffic model output from the G V R D Strategic Planning Department. Each traffic zone contains the population from 1971 to 1991 in 5 year intervals and the employment within the zone from 1971 to 1991 in 10 year intervals. Using the sub-basin overlay exported from the Terrasoft program, the demographic data in traffic zones was combined to provide population and employment information within the watershed sub-basins. 4.5 Statistical analysis Non-parametric comparative statistics were used in this study to detect differences in sample populations because the sample distribution deviated significantly from normality. The Wilcoxon signed-rank test was used to detect changes in one sampling area over time. Independent populations were compared using the Mann-Whitney U test and relationships between variables were analyzed using Spearman rank correlation coefficients. Each of these methods are adequately described in Iman and Conover (1983). Graphical techniques can be highly useful for illustrating population variability and changes over time and space. Box-whisker plots are an especially useful graphical tool and have found many applications in this study. Essentially an illustrative, non-parametric technique, the plot design tempers the effects of skewed data without a loss of information. 58 Figure 4.7 Traffic zones in the Brunette Watershed Figure 4.8 labels all of the components which may exist in the plot. The box encompasses 50 percent of the samples with the vertical line inside the box representing the median value. The absolute value of the difference between the ends of the box is termed the hspread. The whiskers extend to the range of values which are within 1.5 hspreads measured from the end of the box. Values which are between 1.5 and 3 hspreads from the ends of the box are plotted with an asterisk and values outside that range are plotted with an open circle. Figure 4.8 Components of a box-whisker plot whiskers o Source : adapted from Cook (1994). 60 5. DISCUSSION OF R E S U L T S For the purpose of clarity, the discussion of results has been presented in a segregated manner according to the distinctions of lake sediments, stream sediments, street sediments, and land use; the order of appearance of each sub-section was chosen to facilitate subsequent discussion. In many cases, stream and street sediment results are necessarily presented concurrently to illustrate relationships between the two. A l l metal concentrations are presented on a dry weight basis unless otherwise noted. The beginning discussion of methodological and environmental variability defines the degree of certainty which can be applied to the entire collection of results. 5.1 Variability in methodology and environment This sub-section provides the context in which to judge the significance of observed spatial and temporal differences in trace metal concentrations. Differences within and among methods as well as the within-station environmental variability have been quantified. 5.1.1 Precision and accuracy of methods Over 100 additional acid digestions and analytical determinations were made in order to determine the accuracy and precision of the metal results (complete listing of quality assurance data can be found in Appendix C). The accuracy of analyses was estimated from digestions and analysis of sediment reference material obtained from the National Research Council (NRC) (National Research Council of Canada 1981) and is summarized in Table 5.1. Results from the nitric acid digestion show Zn and N i within the 95 % N R C confidence limits and Pb, Cu, and Mg only slightly outside the limits. This suggests that a near total digestion was achieved for these elements. The nitric acid digestion significantly under-predicted Fe, Mn, and Cr, which suggests total digestion was not achieved. Mercury results were slightly above the N R C confidence limit. 61 Table 5.1 Measurement of method accuracy using the NRC marine sediment reference material, BCSS-1 Element Method # Mean Expected Mean 95% Confidence Limits Mean / Expected mean Mean within expected (mg/kg except Fe and Mg in %) limits ? Pb 1. 17 23 19-27 0.74 N-Cu 1. 23 19 17-21 1.21 N+ Zn 1. 110 119 107-131 0.92 Y Ni 1. 59 55 51-59 1.07 Y Mg 1. 1.30 1.47 1.33-1.61 0.88 N-Fe 1. 2.29 3.29 3.19-3.39 0.70 N-Mn 1. 180 229 214-244 0.79 N-Cr 1. 60 123 109- 137 0.49 N-Cd 1. < 3 0.25 0.21-0.29 - -Hg 2. 0.152 0.129 0.117-0.141 1.18 N+ Methods : 1. HN0 3 digestion / flame AA (Civil Engineering, UBC) average of 2 determinations, 2. H2S04-HN03-KMn04 digestion / cold vapour AA (Civil Engineering, UBC) average of 8 determinations. 62 Replicate samples were subsampled from the dried (and sieved in the case of stream and street samples) sample and independent acid digestions were performed to provide an indication of method precision. Two different measures have been used to indicate precision. When only 2 replicates were taken the ratio of the high to the low value is calculated. When 3 replicates of the same sample were taken, a more descriptive — coefficient of variation (CV) — measure is used. Table 5.2 provides the median values of these measures for each analytical method. A l l median replicate ratios are 1.25 or less and median CV's are 15 % or less, indicating a high degree of precision for each method. The lowest precision was exhibited by those metals found closest to the method detection limits (ie. Cd, Ni). 5.1.2 Variability among analytical methods A wide variety of analytical techniques, outlined in section 4.2, have been utilized in this study in order to make valid comparisons with the 1973 baseline study (Hall et al. 1976) and other sediment trace metal studies. Deviations from analytical methods used in the 1973 study have been avoided where possible, but in instances where they do occur a limited number of comparative analyses were performed. Table 5.3 lists all the differences in stream and street sediment analytical techniques used in this study compared to the baseline analysis. Details of the comparative analyses are presented in Appendix D and summarized in this section. Sediment Pb, Cu, Zn, and N i concentrations differ by less than 10 % between the 1973 study and current study digestion methods. Iron and Mn are under-predicted by approximately 20 % using the nitric acid digestion technique, relative to the nitric/perchloric acid digestion used in 1973. This result, and the estimate of nitric acid digestion accuracy (Table 5.1), suggests that the combination of nitric and perchloric acid completely solubilizes the sediment Fe and Mn. Comparative results were not obtained for Mg and Cd. No direct comparisons were made between the 1973, "qualitative" (Fletcher 1993) Cr analytical method and the current study methods. 63 Table 5.2 Precision of analytical methods measured by the median ratio of duplicates and coefficient of variation (CV) Element Method # Ratio of C V % high: low (median) value (median) Pb 1. 1.14 2. 1.11 10.6 Cu 1. 1.09 2. 1.10 7.5 Zn 1. 1.17 2. 1.06 7.3 N i 1. 1.25 2. 1.06 7.8 M g 1. 1.17 7.3 2. 1.08 Fe 1. 1.10 2. 1.07 7.7 M n 1. 1.05 2. 1.07 7.4 Cr 1. 1.16 2. 1.12 7.7 Cd 1. < detection limit 2. 1.21 15.1 Hg 3. 1.08 Methods : 1. H N 0 3 digestion / flame A A (Civil Engineering, UBC) - 10 independent duplicates, 2. HNO3 digestion / ICP-AES (Soils Sciences, UBC) - 17 independent duplicates, 5 independent triplicates, 3. H 2 S 0 4 - H N 0 3 - K M n 0 4 digestion / cold vapour A A (Civil Engineering, U B C ) -10 independent duplicates. 64 Table 5.3 Stream and street sediment analytical techniques used in current study and 1973 baseline study Measurement Technique Measurement 1973 Baseline Study Current Study Fe, Mn, Mg, Cd, Pb, Cu, Zn, N i H N O 3 - H C I O 4 / flame A A H N O 3 / flame A A Cr direct analysis / DC-arc Spectrography H N O 3 / flame A A Hg H 2 S 0 4 - H 2 0 2 - K M n 0 4 -Hydroxylamine / cold vapour A A H 2 S 0 4 - H N 0 3 - K M n 0 4 -Hydroxylamine / cold vapour A A Silt & Clay fraction of sediment < 53 um fraction of sediment < 63 um An analysis of detection techniques indicated that the ICP-AES method (Soils Sciences, UBC) consistently under-predicted metal concentrations relative to the flame A A method (Civil Engineering, UBC), and was unreliable for the detection of Pb and Cu below 20 mg/kg . The results obtained from the two methods were linearly correlated — allowing for direct comparisons between data collected using the different detection techniques. The Silt and Clay fraction was measured slightly differently in this study in order to provide trace metal information for the now commonly evaluated < 63 um sediment fraction. As a result of this difference, the percent Silt and Clay measurements in 1973 are expected to be lower than those reported in this study. No direct comparison between analytical method has been performed. A small comparative analysis was performed to provide information on the < 63 um sediment fraction. Linear correlations were identified which enable estimates of trace metal content in the < 63um fraction to be calculated using the measured metal concentrations in the < 180 um sediment fraction and the fraction of Silt and Clay in the sediment. 65 5.1.3 Variability within sampling stations Three separate sediment samples were taken at each stream station to obtain an estimate of the environmental variability within a site. Standard errors, calculated for each of the 33 stream stations, are shown in Figure 5.1 using box-whisker plots. While some stations exhibit very high variability, the standard error of the majority of stream stations is between 10 and 30 % of the mean for each metal analyzed. Except for station C4, only 1 sample was collected at each of the street locations. At station C4, 3 separate samples were collected from curbs surrounding the busy intersection (North Road at Lougheed Highway) in order to provide an indication of variability within a local area. The large local variability relative to the variability across all sites — illustrated in Figure 5.2 — suggests the limited information that a single street sampling location can provide. Figure 5.1 Standard Error of the mean of metal concentrations at stream stations illustrated using box-whisker plots (n=33). (Nitric acid digestion) i -z UJ O CC LU CL 80 60 \-40 20 0 Ni Zn C u C r Pb Hg Mn Fe LOI M E T A L 66 Figure 5.2 Mean and standard deviation of selected metals at all street sampling locations (n=25) compared to one location (n=3) (Nitric acid digestion) Zn Stations A l l C4 Stations 1200 -1000 -2 800 -™ 600 -400 -200 -0 -A l l C4 Stations 5.1.4 Repeatability of environmental conditions Three stream stations were re-sampled in 1994, one year after the initial sampling, to obtain an estimate of the repeatability of trace metal measurements at each location. The expectation of repeatability from one year to the next is reasonable because surrounding land use changes did not occur and sampling occurred during low-flow conditions in both years. Antecedent rainfall, summarized in Table 5.4, provides an indication of hydrological conditions during both sampling periods. Figure 5.3 illustrates the station means and site variability for the 3 stations for both time periods. The relative relationship between the stations was maintained for both sampling periods. Stations #10 and #37 are both enriched with Pb, Cu, Cr,and Ni relative to Table 5.4 Daily rainfall at Burnaby Mountain weather station prior to 1993 and 1994 stream sediment sampling (Source: Environment Canada 1994) Rainfall (mm) 18 17 • • • Days prior to sampling 5 4 3 2 1 0 1993 9 0 0 0 0 0 0 0 0 1994 11.2 0 0 5.4 1.6 1.0 0 0 0 67 -H—II-I I o iri o o co o CM h co ft . m c I- - o <0 o o d I—•—I CM I O CO I ' I Tl- CM h co I- - o ft *-a CO o ft 6>|/6UJ 6>)/6UJ CL co CD O) <3> CD • • i—n—m o o 00 o o o o co -r - r-o o in h co ft 10 = o ft ~ a o ft o o o o co l • l • o o o o o CM T -6>|/6ui 68 H - O - I W co 6>)/6LU I I • I O CO CD CD O) CD ' r " B • H W H —1 1 1 o o O co CM 1— h co . ^ c a o 6>)/6iu 69 station #15 for each time period. Station #15, on the other hand, is enriched with Mn relative to the other stations for both years. Absolute differences between years are difficult to detect due to high within-site variability. Nickel measurements at station #15 were lower in 1993, but the results are too near the detection limit to support a finding of significant difference. Organic matter (indicated by LOI) in sediments at station #37 (Still Creek at Atlin Rd.) was lower in 1994 and likely accounts for the lower trace metal measurements. Contamination from sewage in this section of Still Creek (Coastline Environmental Services 1987) is a probable source of variable organic matter content in the stream sediments. 70 5.2 Trace metals in lake sediments Sediment cores from three different lakes were taken to obtain a record of trace metal contamination, a record which spans the time period prior to European colonization and industrialization to the present day. Two of the lakes are within the study watershed while the third, located in a relatively remote area, was chosen to provide an indication of the atmospheric transport of contaminants. Comparisons between cores and lakes have been made using trace metal concentrations, fluxes, and enrichment factors. Complete trace metal and textural data, qualitative core descriptions, and radioisotope measurements are included in Appendix E. 5.2.1 Burnaby Lake Three separate core locations were used to obtain the sediment contaminant history of Burnaby Lake. The placement of the three cores (shown in Figure 4.3) enables the impact of the Still Creek contaminant load on the entire lake and downstream waterways to be evaluated. The analysis of these cores represents the first detailed contaminant examination of Burnaby Lake sediments spanning the present day to pre-industrial times. 5.2.1.1 Background metal concentrations Results obtained from the 6 metre core, taken at location #2 (Figure 5.4), show quite clearly the beginning of watershed disturbances at the 50-60 cm depth. Organic matter and water content rapidly decrease above this level, indicating increased coarse mineral input from the catchment area. Below this level, changes in sediment texture and metal concentrations occur very gradually. An exception to this pattern was observed at the 374 cm level. The sediment above this level is characterized by decomposed peat while organic silt dominates below. Corresponding with this change in stratigraphy, the organic matter abruptly decreases by half and Fe and M g both increase by more than a factor of 2. A temporary rise in the sea level approximately 8000 years ago which caused the Fraser River 7 1 Figure 5.4 Sediment texture and metal concentrations in Burnaby Lake core #2 Figure 5.4 continued Cd Cr o 50 H 100 150 200 250 300-350-400-l 450-500-550-600 • > — i — 1 — i — 1 — r 0 1 2 3 mg/kg (dry) Mn o 50 -100-150-200-250-300-350 400 -\ 450 H 500 550 H 600 1* Mg T 1 1 1 1 1 • -0 10 20 30 40 mg/kg (dry) 0 -50 -100-• • 0 -• 50 -100-1 150- • 150- • 200-• • 200-• • 250- • 250- • 300- • 300- • 350-400-• • • • 350-400-• • • • 450- • 450- • 500- • 500- • 550- • • 550- • • 600- 600- , r— 100 200 300 mg/kg (dry) 400 0 2500 5000 mg/kg (dry) Ni T 0 10 20 mg/kg (dry) 73 to connect to Burrard Inlet through the Still Creek/Burnaby Lake/Brunette River corridor and provided the peat soil material of much of Burnaby (Gardner Dunster Associates Ltd. 1992, Johnston 1921) is a potential explanation for this change in sediment texture. Most importantly for this study, a range of background trace metal values is contained in this lake core which can be used to evaluate the presence of current trace metal contamination in the watershed. Background concentrations are defined as the maximum metal concentrations observed in the core below the 65 cm level. Using these values, surface enrichment factors (surface sediment concentration / background concentration) were calculated for Burnaby Lake core #2 (Table 5.5). 5.2.1.2 Sedimentation rates The dates and rates of sedimentation throughout the upper section of core #2 were 210 determined using the Pb radioisotope dating method (Figures 5.5 and 5.6). The location of the core was chosen to reflect, specifically, the effects of land use changes in the Still Creek drainage area. An estimate of the accuracy of the method was determined by examining the probable causes of the high sedimentation rate calculated to extend from 13 to 19 cm depth and deposited in 1970 ( +/- 2 years). This high depositional zone correlates precisely to a well-washed sand layer in the core — indicating that a storm event which induced high stream flows and erosion is a probable cause. The highest daily precipitation ever measured in New Westminster (1874-1957) or Burnaby Mountain (1958-1994) was recorded in 1968 at Table 5.5 Background metal levels and surface enrichment factors observed in Burnaby Lake core #2 Pb Cu Zn Cd Cr Ni Mn Fe Background concentrations 18 14 71 0.6 20 23 202 1.57 (mg/kg except Fe - %) Surface enrichment factors 10 6.0 3.8 1.5 0.7 0.7 0.8 0.7 7 4 Figure 5.5 Age profile of Burnaby Lake core #2 determined using Pb radioisotope dating (error bars represent standard deviation) E o Q. O T3 2 8 ' I ' I ' I ' I 1 I 1 I 1 I ' I ' I 25 50 75 100 125 150 175 200 225 250 275 210Pb Age (yr) Figure 5.6 Sediment accumulation over time in Burnaby Lake (core#2 location) calculated 210 using Pb radioisotope dating (error bars represent standard deviation) (0 Q •Q 0. O 10 Sediment Accumulation (g/cm2/yr) 7 5 147 mm of rain (Environment Canada 1994). This date is within the confidence interval of the predicted high depositional period and therefore provides a measure of confidence in the dating method. The increase in lake sedimentation rates after 1880 corresponds to widespread clearing and tilling of the watershed. Increased sediment loads after 1915 are probably the result of increasing urbanization and stormwater improvements made to Still Creek. The peak calculated to occur in 1928 does not correspond to a sand layer in the core which suggests it was not the result of high stream flows. Instead, the high sediment load may be related to the enclosure of Still Creek's headwaters in the "Collingwood trunk" sewer in 1924. Stream channelization, carried out between 1914 and 1935 (Dawson et al. 1985), would also contribute unusually large sediment loads to Burnaby Lake. Sediment loads have remained very high following the storm event in 1968. These high sedimentation rates can be attributed to stream-bank erosion caused by high flows during storms and development surrounding Still Creek. 5.2.1.3 Trace metals in recent sediments Interpretation of the dated core (#2) is complicated by the variable sediment texture (illustrated in Figure 5.7) observed in the most recent sediments. Both organic matter (indicated by LOI) and moisture content have steadily decreased in core #2 since 1950. This trend indicates increasing particle sizes (moisture content is inversely proportional to particle size) and can be attributed to higher peak stream flows caused by urban runoff from impervious surfaces. Trace metals tend to concentrate on smaller particle sizes rich in organic matter, therefore corrections for both particle size and organic matter must be employed for valid comparisons between core depths. To normalize the effects of differing particle sizes, trace metal concentrations have been expressed on a wet volume basis (a numerical technique described by Ffajkanson and Jansson (1983)). 76 Figure 5.7 Sediment texture in Burnaby Lake core #2 sediments (0-48 cm depth) Empirical evidence in support of this normalization technique was recently published from a study of 52 remote Quebec and Ontario lakes (Rowan and Kalff 1993). This study showed that moisture content was the most significant determinant of trace metal concentrations in surface sediments and together with lake depth accounted for over 80% of the between-site and between-lake variability. A comparison with results from Burnaby Lake core #1 (detailed in Appendix F) further justifies this normalization technique. Trace metal concentrations, expressed both on a dry and wet ("normalized") basis, are shown for core #2 in Figure 5.8. An examination of the 9 trace metal concentrations reveals that Pb, Cu, and Zn levels have risen significantly since 1830, while increases in Cr, Ni , Mn, Fe, and Mg have been moderate. Cadmium concentrations were very high 77 Figure 5.8 Metal concentrations in Burnaby Lake core #2 sediments (0-48 cm depth) 1990 J isnn | ^ ^ 1830 0 100 200 300 10 20 30 40 • dry (mg/kg) o wet (u.g/cm3) 0 100 200 300 400 Ni 1990 J 10 20 30 40 78 Figure 5.8 continued 1990 1970 1950 0) +-> CO 1930 Q .Q a. o 1910 N 1890 1870 1850 1830 i—r~i—r 0 10C 200 300 400 • dry (mg/kg) <" wet (ug/cm3) 200( 400( • dry (%) * wet (mg*10/cm3) • dry (mg/kg) * wet (ug/cm3) approximately 140 years ago, yet current levels are only slightly above the background concentration. Lead concentrations began to increase as a result of increased human activity around 1880. After 1940, Pb concentrations increased dramatically — corresponding to increased automobile use and urbanization. Although the phase-out of the lead additive from gasoline began in 1973 and was completed in 1990, Pb levels in the sediments (on a normalized basis) have only slightly decreased within the last 5 years (the sharp decline for all metals in 1968 is the result of one storm event). These results imply that Pb introduced into the environment from car emissions may persist in soils for some time prior to entering the aquatic environment. Other continuing urban non-point sources may also be a cause of high surface Pb concentrations. 79 Rapid increases in both Cu and Zn concentrations began in 1940. Zinc concentrations — on a wet basis — in the lake sediment reflect the steady increase in automobile use (Figure 3.4) and point to traffic as a possible cause of the contamination. High Cu levels in urban areas have also been attributed to automotive causes, but the Burnaby Lake (Cu) concentration profile does not reflect the automobile trend. Rather, the peak experienced in 1970 was probably the result of industrial sources in the Still Creek area. This conclusion is supported by a sediment survey conducted by the Westwater Research Centre in 1973 which identified industrial sources to be the likely cause for extremely high Cu levels and slightly elevated N i and Cd in an industrial area surrounding Still Creek (Hall et al. 1976). Copper levels in lake sediments have since decreased from 1970 levels — indicating abatement of the industrial discharges. Surface Cu concentrations are still very high (note enrichment levels in Table 5.5) and indicate continuing sources need to be identified. Trace metal fluxes to Burnaby Lake from Still Creek (Figure 5.9) illustrate the combined results of increasing concentrations and sedimentation rates. Data from core depths 13-19 cm (1968) are not shown because the large loading represents only one single storm event. While this representation does not normalize sediment particle sizes, the similarity in the loading profiles of Cu, Cd, Cr, and N i suggests they all reached peaks in 1970 due to industrial contamination in the Still Creek area. Core #3, located in the central portion of Burnaby Lake (shown in Figure 4.3), provides no indication of any trace metal contamination throughout the core (Figure 5.10). A brief increase in mineral content, denoted by a high Fe concentration, did occur at 29 cm. Assuming this is related to the similar sharp rise in Fe concentration in core #2 at 29 cm (1931), the sedimentation rate over the last 60 years for both cores is approximately 0.5 cm per year. Most significantly, the results of this core show that trace metal loads from Still Creek, which contributes over 50 % of the lake input, do not affect the entire length of Burnaby Lake. 80 Figure 5.9 Historic trace metal fluxes to Burnaby Lake (calculated from core #2, 0-48 cm depth) 1990 J 1830 * i i i i i 0 .2 A .6 Note : metal fluxes from one extreme storm event (core depth : 13-19 cm) not shown 8 1 Figure 5.10 Metal concentrations and sediment texture in Burnaby Lake core #3 LOI H20 Fe 0 5 10 15 20 0 5 10 15 20 0 10 20 30 40 50 60 mg/kg (dry) mg/kg (dry) mg/kg (dry) Note : Pb, Cu, and Cd all less than the detection limit. 82 5.2.2 Deer Lake Results from Deer Lake varied considerably between core sampling locations. The long core, taken in a slightly shallower area of the lake (location shown in Figure 4.3), did not exhibit any trace metal contamination (Figure 5.11). The higher organic content of this core, relative to the short cores (Figure 5.12), suggests the long core was taken in an area of the lake which does not accumulate catchment-derived (contaminated) particles. Munteanu (1987) also found, in her examination of 8 cores in Deer Lake, that sediment texture and trace metal deposition were quite variable and dependent on lake depth. While surface trace metal contamination was detected in each of the 8 cores, Munteanu's results could not be confirmed because the top 4 cm of the long core was not analyzed. The 2 short cores, taken in a slightly deeper section of the lake, exhibit considerable Pb, Cu, and Zn contamination. N i , Cr, and Mn concentrations are very similar to background concentrations found at 119 cm depth in the long core. Background Hg levels are not available for either of the lakes in the watershed to evaluate the concentrations in this core. Assuming a 2 cm per year sedimentation rate (estimated by Northcote and Luksun (1992)), the contaminant profiles extend to a 1986 sedimentation date at the base of the cores. Relative to metal concentrations measured in a Deer Lake core taken by Mathewes and D'Auria (1982), Pb has increased since 1979 by approximately 40% while Cu and Zn have held constant (Figure 5.13). The Cu peak observed in the 1979 core was assumed to be caused by the addition of an algicide (copper sulphate) to the lake in 1957. The high Zn levels at the base of the 1979 core are unexplained. 8 3 Figure 5.11 Metal concentrations and sediment texture in Deer Lake (long core) LOI H20 Fe 0 200 400 600 800 0 10 20 30 40 20 40 60 80 100 120 mg/kg (dry) mg/kg (dry) mg/kg (dry) Note : Pb less than detection l imit . 84 Figure 5.11 continued Cd Cr Mg mg/kg (dry) mg/kg (dry) mg/kg (dry) Cu 0 10 mg/kg (dry) 85 Figure 5.12 Metal concentrations and organic matter in Deer Lake (short cores) Hg N C r E 3 ^ 5 £ 7 Q. <S 9 11 13 100 200 300 400 ug/kg (dry) 0 10 20 30 40 mg/kg (dry) 50 mg/kg (dry) LOI • Core #1 SI Core #2 % (dry) 0 200 400 600 mg/kg (dry) 86 Figure 5.13 Metal concentrations in Deer Lake core taken in 1979 Pb Cu Zn 0 50 100 150 0 100 200 300 400 0 100 200 300 mg/kg (dry) mg/kg (dry) mg/kg (dry) Source: Mathewes and D'Auria (1982) Note: Percent carbon at surface = 5 %; equivalent (approximately) to 10 % LOI. 5.2.3 Gwendoline Lake Gwendoline Lake is located in an uninhabited, forested watershed which, since being clearcut logged between 1958 and 1969 (Power 1994), has experienced only minor human disturbances. Trace metal input to this lake from sources within the watershed should not be above background levels. Results from the 2 short cores taken in this lake indicate increases in Pb levels in the upper 4 cm of the cores (Figure 5.14). This is a strong indication that this area has been affected by atmospheric emissions transported from the population centres in the Greater Vancouver area. Assuming the beginning of the sharp Pb rise at 5 cm corresponds to the rise in vehicle registrations in 1950 (Figure 3.4), the approximate recent sedimentation rate has been 0.13 cm per year. Other metal increases, such as Cu, N i , and Zn, are difficult to confirm 87 Figure 5.14 Metal concentrations and organic matter in Gwendoline Lake sediments 0 100 200 300 mg/kg (dry) Zn 100 200 300 400 mg/kg (dry) 200 mg/kg (dry) Hg Ni C r 0 100 200 300 0 10 20 30 40 50 60 0 10 20 30 ug/kg (dry) mg/kg (dry) mg/kg (dry) <D o 1 3 5 7 9 11 13 15 17 19 22 LOI Fe 60 20 % % (dry) 2500 5000 7500 mg/kg (dry) Core #1 Core #2 88 because of differences observed in the replicate cores. The surface layer in core #2 was disturbed during the extrusion process and could account for the discrepancies between cores. Surface enrichment factors, calculated for this lake using the data from core #1, are 12, 2.3, 1.8, 5.3, 1.0, 1.0, and 1.7 for Pb, N i , Zn, Cu, Cr, Fe, and M n respectively. While some of these enrichment factors are similar or greater than those observed in the Burnaby Lake core (Table 5.5), trace metal loading is much greater in Burnaby Lake because of higher sedimentation rates. The surface sediments of Gwendoline Lake were analyzed for trace metals (Reiberger 1992) as part of a study of 390 lakes in the province from 1982 to 1987. Although a direct comparison with the current study is not possible because of the sampling method (Ekman dredge) employed, high Hg levels observed throughout the core in the current study are confirmed by Rieberger's results (Table 5.6). A comparison can be made between Gwendoline Lake sediment metal levels and surface sediment concentrations recorded in a study of 52 remote Quebec and Ontario lakes (Rowan and Kalff 1993). This study found significant correlations between elevated concentrations of Pb, Cu, and Zn and the lake sediment moisture content and water depth. The relationships were independent of geological differences between catchment areas, suggesting the correlations may be valid for Gwendoline Lake. Utilizing these correlations (a range of moisture contents was used since measurements were not taken), the predicted values of Pb and Cu were lower than actual, while the Zn prediction was higher (Table 5.6). High background concentrations at Gwendoline Lake are a probable cause of the Cu discrepancy. The Pb and Zn discrepancies between actual and predicted values may be the result of different atmospheric trace metal inputs for Gwendoline Lake than for remote lakes in Eastern Canada. 89 Table 5.6 Comparison of actual and predicted metal concentrations in Gwendoline Lake surface sediments (all concentrations in mg/kg dry weight) Metal 1993 core (range of replicates) Rieberger core (Reiberger 1992)1 Predicted (Rowan and Kalff 1993 correlations) 80 % H 2 0 | 9 0 % H 2 O Pb 152 - 247 53 91 143 Cu 73 - 315 62 27 40 Zn 109-138 51 225 240 N i 25 - 55 10 Cr <3 - 8 26 Hg 0.18 - 0.22 0.27 1. one sample from the top 5-10 cm, sampled with an Ekman dredge. 5.2.4 Comparison to other urban lakes A comparison with three other urban lakes in the United States and Australia is provided to put the Burnaby Lake and Deer Lake results into a wider context. Surface metal concentrations and loading rates (for 2 lakes) have been extracted from the literature. The historic record of trace metal fluxes (Pb, Cu, and Zn) in a small lake in Melbourne, Australia (Botanic Gardens Lake) is illustrated in Figure 5.15. The sharp increase in the Pb flux just prior to 1950 is similar to Burnaby Lake (Figure 5.9) and was attributed to emissions from automobiles. There is no decrease in Pb at the surface since the core was taken in 1983 while the Pb gasoline additive — T E L — was still in use. Zinc loading increased in the Australian lake at an earlier date compared to Burnaby Lake. This could be the result of the rapid urban development of Melbourne following the gold rush in 1860 (Smith and Hamilton 1992). Surface metal concentrations for the Melbourne lake as well as Lake Washington (Seattle), Lake Ellyn (Chicago), and the Brunette Watershed lakes are summarized in Table 5.7. The lower Pb levels in Burnaby and Deer Lake can partially be attributed to the removal 90 Metal Flux (mg m ^ y 1 ) 1830 1790 -J Figure 5.15 Historic fluxes of Pb, Cu, and Zn to the sediments of a small lake in Melbourne, Australia (Adapted from Smith and Hamilton 1992) of T E L from gasoline in recent years. Copper levels are high in the two Brunette Watershed lakes compared to the Seattle and Melbourne lakes. High Cu levels in the Chicago lake are attributed to additions of the algicide, copper sulphate (Striegl and Cowan 1987). Higher sedimentation rates in Burnaby Lake result in trace metal fluxes which are much higher than in the Seattle and Melbourne lakes (Table 5.8). The large urban catchment area of Still Creek (3035 hectares), relative to the small depositional area in the western portion of Burnaby Lake, is the probable cause of the very high trace metal fluxes. 9 1 Table 5.7 Burnaby Lake and Deer Lake surface sediment trace metal concentrations compared to three urban lakes (all concentrations in mg/kg dry weight) Element Burnaby Deer Lake Botanic Lake Lake Ellyn Lake (average of 2 Gardens Washington (Chicago, (core # 2) short cores) Lake (Seattle, Illinois)3 (Melbourne, Washington)2 Australia)1 Pb 183 173 280 192 1590 Cu 84 103 70 46 250 Zn 266 213 360 192 210 Sources: 1. measurements taken in 1983 : Smith and Hamilton (1992) 2. measurements taken in 1975 : Spyridakis and Barnes (1976) 3. measurements taken in 1980 : Striegl and Cowan (1987) Table 5.8 Burnaby Lake recent trace metal fluxes compared to two urban lakes (all fluxes in g/cm2/year ) Element Burnaby Lake 1 Botanic Gardens Lake (Melbourne, Australia)2 Lake Washington (Seattle, Washington)3 Pb 69 23 11 Cu 32 6 3 Zn 100 34 14 Notes and sources : 1. representative of western portion of lake only 2. Smith and Hamilton (1992) 3. average from 3 lake sediment traps from Spyridakis and Barnes (1976) 92 5.3 Land use changes from 1973 to 1993 Changes in land activity, land cover, and traffic density have all been quantified with the aid of GIS to examine relationships between land use and trace metal contamination (complete results are tabulated for each sub-basin in Appendix G). Land activity, spatially illustrated for 1973 and 1993 in Figures 5.16 and 5.17, has changed only marginally over the last 20 years (Table 5.9). These small changes in land activity confirm Duynstee's (1990) study examining changes in land zoning between 1973 and 1987. While residential areas now occupy 5% more area than in 1973, this should be considered a modest increase considering the 30% increase in population over the same period (Table 3.1). The combined land area increase of 1.5% for the commercial, industrial, and institutional activities is even more surprising since employment in the watershed has increased 120% during this time. The shift in the employment structure away from a manufacturing base to a less land-intensive services sector is a likely explanation for this result (Figure 3.3). The City of Burnaby has recognized this trend of densification and is currently promoting it through the use of group housing development and lot in-filling (City of Burnaby 1987). Table 5.9 Land use activities in the Brunette River Watershed as a proportion of total area in 1973 and 1993 Land Use 1973 1993 Change % % % residential 40.8 45.7 + 4.9 industrial 11.9 13.2 + 1.3 commercial 3.6 4.1 + 0.5 institutional 6.6 6.4 -0.3 transportation* 2.7 2.7 + 0 agricultural 1.4 0 - 1.4 open spacef 32.9 28 - 5 * - includes Trans-Canada freeway and rail only t - includes parks, recreational and conservation areas, and undeveloped lands 9 3 Figure 5.16 Land use activity in the Brunette Watershed in 1973 (Note: undeveloped lands included in recreation category) 94 Figure 5.17 Land use activity in the Brunette Watershed in 1993 (Note: undeveloped lands included in recreation category) 95 Land cover permeability is an important predictor of the quantity of urban runoff (Rood and Hamilton 1994) and has also been found to be directly correlated to urban stormwater contamination (Bannerman et al. 1993). Detailed spatial permeability information (Figures 5.18 and 5.19) is also useful for examining the quality of buffer regions surrounding the lakes and streams in the watershed. A small decrease in permeable areas throughout the watershed has occurred since 1973 (Table 5.10). Although the overall changes may be modest, changes at the local scale can have consequences for individual streams. Figure 5.20 highlights those areas which in 1973 were either covered by woods or grass and have now been developed with significant portions of the areas impermeable to precipitation. Considerable reductions in land permeability surrounding the two creeks (Eagle and Stoney) originating on Burnaby Mountain have occurred over the last 20 years. Automobile traffic is a well known source of trace metal contamination and is increasing at a rate even greater than population (Table 5.11). Table 5.10 Land cover in the Brunette Watershed in 1973 and 1993 Land Cover 1973 1993 % % Permeable 66 59 Impermeable 34 41 Effective impermeable1 26 32 Notes: 1. Effective impermeable areas (Dinicola 1990) approximate the amount of impermeable area which is connected directly to the constructed drainage system. 96 Figure 5.18 Land cover permeability in the Brunette Watershed in 1973 9 7 Figure 5.19 Land cover permeability in the Brunette Watershed in 1993 98 Figure 5.20 Reduction in land cover permeability in the Brunette Watershed between 1973 and 1993 99 Table 5.11 Traffic density in the Brunette River Watershed in 1973 and 1993 Vehicle kilometres / day (x 103) Sub-basin 1 1973 1993 % Change 1 430 670 56 2 248 380 5 3 3 209 308 4 7 4 1 74 259 4 9 5 202 272 3 5 6 22 29 32 7 1 2 17 41 8 47 6 6 3 9 9 81 114 41 1 0 1 8 2 6 43 1 1 20 2 7 3 7 1 2 1 05 1 50 43 1 3 7 8 1 06 35 1 4 113 158 40 1 5 1 54 1 99 29 1 7 6 8 30 1 8 67 97 44 1 9 1 1 8 145 23 20 37 62 69 21 1 0 1 6 58 22 1 1 6 1 76 51 23 27 45 64 24 25 32 26 25 25 31 26 26 28 39 38 27 156 224 44 28 32 39 23 29 444 633 42 Whole Basin 3000 4300 44 1. Sub-basin boundaries delineated in Figure 4.5. 100 5.4 Trace metal loading from automotive sources Numerous studies have linked urban trace metal pollution to automobile use (summarized in sub-section 2.2). The traffic estimates summarized in Table 5.11 allow contaminant loadings in this watershed resulting from automobiles to be calculated. These estimates are intended to provide an indication of the magnitude of metal loadings only, and do not account for variability in fuels, automobile components, and wear rates. A comprehensive urban contaminant identification and control program in Santa Clara, California — begun in 1990 — has identified vehicle exhaust emissions, tire wear, and disk brake pad wear as the most significant non-point sources of Pb, Cu, and Zn in their study area (Armstrong 1994, Woodward-Clyde Consultants 1992). Fuel metal levels and consumption rates, and tire and brake pad compositions and wear rates used to calculate trace metal loadings in the Brunette watershed were extracted from the Woodward-Cyde (1992) report and other sources detailed in Appendix K . A comparison between the calculated trace metal loadings from automobiles and the total estimated metal load in storm water (Table 5.12) provides some indication of the Table 5.12 Estimated trace metal loadings from automobiles and in stormwater runoff in the Brunette Watershed Metal Automotive Source Calculated Automotive Trace Metal Loading (kg / day) Estimated Stormwater Trace Metal Metal Loading (kg / day)1 1973 1993 Pb fuel 132 3.1 5.6 Cu brake pad wear fuel 0.5 0.9 0.8 1.3 1.7 total 1.4 2.1 Zn tire wear fuel 2.0 2.2 2.9 3.1 2.0 total 4.2 6.0 M n fuel additive 0 3.4 2.7 Sources: 1. Swain (1989) 101 automobile's impact on stormwater quality. Zinc emissions resulting from the wear of tires and fuel emissions appears to be a major source contributing to stormwater contamination. Automotive sources of Cu are comparable to total stormwater loadings. This would suggest, assuming only a fraction of the vehicle-generated contaminants become associated with stormwater runoff, that additional Cu sources of equal significance may be present in the watershed. The stormwater loading calculations (Swain 1989) utilized stormwater quality data obtained during the period when T E L was gradually being phased out of gasoline. For this reason, no conclusions can be made relating automotive Pb and Zn to estimated stormwater loadings. 1 0 2 5.5 Trace metals in streambed and street sediments A graphical presentation of stream sediment metal concentrations begins this sub-section to provide an overview of obvious spatial patterns of contamination. The street sediment contamination risk is evaluated by comparison to stream concentrations while the severity of stream contamination is assessed by comparison to another similar watershed and sediment quality criteria. The remaining sections examine spatial and temporal contaminant patterns and the possible relationships to land use. The presentation and tabulation of trace metal data in this sub-section follows a few consistent conventions: trace metal concentrations refer to the results of nitric acid sediment digestion unless explicitly stated otherwise; "total" 1973 and 1989 concentrations refer to the results of a nitric/perchloric acid sediment digestion; "total" 1993 concentrations refer to the results of the nitric acid sediment digestion. The complete streambed and street sediments trace metal datasets are located in Appendices H and I respectively. 5.5.1 Overview of results Considerable variations in contaminant levels have been detected between stream stations in the watershed (illustrated in Figures 5.21 - 5.30). Concentrations of Pb, Cu, and Zn are high throughout the watershed, but the Still Creek area is the most highly contaminated region. The extractable concentrations of these metals are a significant fraction of the total which indicates possible anthropogenic sources and potential bio-availability. Manganese, a less toxic element, also exists predominantly in the extractable form in the sediments although exhibiting a much different spatial distribution throughout the watershed . A very small fraction of the total M n concentrations were in the extractable form in 1973 (Hall et al. 1976) which suggests an anthropogenic source may be responsible for the high 1993 values. Natural sources are likely responsible for much of the Ni , Cr, and M g since little spatial variability is observed and only a small fraction of these metals exist in the extractable form. Although Fe does vary across the watershed , this may be the result of 1 0 3 104 Figure 5.22 Copper distribution in streambed sediments in the Brunette Watershed Figure 5.23 Zinc distribution in streambed sediments in the Brunette Watershed 106 Figure 5.24 Nickel distribution in streambed sediments in the Brunette Watershed 107 Figure 5.25 Chromium distribution in streambed sediments in the Brunette Watershed 1 0 8 1 1 0 Figure 5.28 Magnesium distribution in streambed sediments in the Brunette Watershed 1 1 1 Figure 5.29 Organic matter (LOI) distribution in streambed sediments in the Brunette Watershed i 1 1 2 differing redox conditions in the streams. Contaminant concentrations can often increase in downstream streambed sediments as a result of fine-grained sediments accumulating in lower energy stream sections. With the exception of Zn (Figure 5.23), this has not been observed along the 7 stream stations located in Still Creek (Figures 5.21 - 5.30). Results from both the 1973 and 1993 sampling periods are illustrated (Figures 5.31 -5.33) for the three metals which have experienced the greatest changes in concentration (Pb, Hg, and Mn). Lead has decreased remarkably — mainly in the Still Creek area — while Hg and M n have increased throughout the entire watershed. No obvious spatial patterns for the metal increases can be detected (a more detailed analysis of changes in metal concentrations since 1973 is presented in sub-section 5.5.2). Street dirt can form a significant portion of the total sediment load to urban streams when the proportion of impervious land cover is high. If the trace metal concentrations are elevated in the street dirt, then this material becomes a major contaminant source to streams. The ratio of the measured median street sediment concentrations to stream concentrations (Table 5.13) reveals that street dirt is an obvious source of Pb, Zn, Cu, N i , and Cr to streams. Table 5.13 Median stream and street sediment metal concentrations (all concentrations in mg/kg except Fe in percent) Element Median Street Sediment Concentration (n=25) Median Stream Sediment Concentration (n=33) Ratio of median street concentration to median stream concentration Pb 222 63 3.5 Zn 373 143 2.6 Cu 129 56 2.3 N i 30 14 2.1 Cr 48 28 1.7 M g 3871 3424 1.1 Fe 1.9 2.0 1.0 Hg 45 98 0.5 M n 315 807 0.4 114 115 Figure 5.33 Manganese distribution in streambed sediments in the Brunette Watershed in 1973 and 1993 117 A useful comparison to the Salmon River watershed in Langley, British Columbia can be made based on the results of Cook's (1994) contaminant analysis. The Salmon River watershed is of similar size and located in a similar geologic setting to the Brunette watershed. Agricultural and undeveloped forested lands are the dominant land uses in the Salmon watershed although rapid growth in the residential sector (estimated population of 16,000) is a concern. Lead, Cu, and Zn stream sediment concentrations in the Brunette watershed were converted to a < 63 um sediment fraction basis (using a method outlined in Appendix D) in order to make a valid comparison between watersheds. A l l three metals were significantly higher (tested using the Mann-Whitney U test; a = 0.01) in the Brunette watershed, as is apparent in Figure 5.34. This comparison to another, less urbanized, Fraser River tributary confirms the contaminated nature of the Brunette watershed. Comparing both the 1973 and 1993 stream sediment results to various sediment contaminant criteria (listed in Table 2.1) provides some indication of the watershed aquatic health. A n absolute risk to aquatic life cannot be determined from this analysis because the sediment criteria do not account for differences in particle size, organic matter, or metal bio-availability. In spite of these limitations, the comparison reveals a large increase in the number of stream stations exceeding the different criteria levels for Cu, Zn, and Hg (Table 5.14). Although Pb levels have significantly decreased, the number of stations exceeding criteria has decreased only slightly. This comparison suggests the aquatic health in the Brunette watershed is poor and in many cases getting worse. 118 Figure 5.34 Sediment trace metal concentrations in the Salmon (n=19) and Brunette River (n=33) watersheds illustrated using box-whisker plots(concentrations in mg/kg) 6 0 0 o 5 0 0 4 0 0 3 0 0 2 0 0 100 Brinette Sarnon Brinette Safrion RIVER RIVER 6 0 0 Brunette Sarnon RIVER Notes: 1. Brunette data converted to < 63 um sediment fraction basis using method in Appendix D 2. Salmon watershed data from Cook (1994) - August 1991 sampling, sediments digested using HN0 3 / HC1/ HF acids 1 1 9 Table 5.14 Number of stream stations exceeding sediment quality criteria in 1973 and 1993 NUMBER OF STREAM STATIONS EXCEEDING CRITERIA (total stations: 36 in 1973, 33 in 1993) T R A C E Brunette Watershed Objectives Wisconsin sediment criteria EPA heavily polluted criteria M E T A L 1973 1993 1973 1993 1973 1993 Lead 35 33 25 22 20 20 Copper 20 31 4 10 11 24 Zinc 20 32 16 28 5 9 Mercury 2 20 n/a n/a n/a n/a n/a: criteria not available 120 5.5.2 Temporal analysis of trace metal data Comparisons between years are only valid i f the sediment texture, as described by organic matter (LOI) and silt and clay content, varies little between sampling periods. There is no significant difference in LOI in streambed sediments between years, yet the silt and clay content is higher in 1993 (Table 5.15). Different definitions of silt and clay (1973: < 53ixm, 1993: < 63 \im) may account for the observed differences. Organic matter was slightly higher in 1993 street sediments, but no differences in silt and clay were observed (Table 5.16). Table 5.15 Median concentration differences and significance of ranked t-test for comparison of 1993 versus 1973 metal concentrations in streambed sediments (n=32) Element Digestion Method Median percent difference, 1993 minus 1973 concentrations. p Value* M n total 131 0.00 extractable 2600 0.00 Pb total - 35 0.04 extractable - 16 0.05 Cu total 81 0.01 extractable 49 0.01 Zn total 45 0.00 extractable 33 0.10 N i total - 7 0.56 extractable n.c. Hg total 294 0.00 Cd total n.c. extractable n.c. Fe total - 21 0.02 extractable - 19 0.20 Silt and Clay 92 0.00 LOI 10 0.27 * - p values indicate the probability that the means are identical, n.c. - no comparison made because of high 1993 detection limit. 121 A statistical comparison between trace metal concentrations in stream and street sediments in 1973 and 1993 reveals many significant changes (Tables 5.15 and 5.16). Pb has decreased in each medium while Mn has similarly increased. Copper, Zn, and Hg increased in stream sediments but not in the street samples. No significant changes in N i levels were detected. Box-whisker plots (Figures 5.35 - 5.43) have been utilized in this analysis to provide both spatial and temporal variability information for each metal. Data obtained in 1989 from the same stream sampling stations by Duynstee (1990) have been included in the analysis. Table 5.16 Median concentration differences and significance of ranked t-test for comparison of 1993 versus 1973 metal concentrations in street sediments (n=25) Element Median percent p Value* difference, 1993 minus 1973 concentrations. M n 43 0.00 Pb - 67 0.00 Cu - 43 0.01 Zn 6 0.74 Ni - 13 0.22 Hg - 4 0.69 Cd n.c. Fe 1 0.40 Silt and Clay - 2 0.25 LOI 40 0.02 * - p values indicate the probability that the means are identical, n.c. - no comparison made because of high 1993 detection limit. 1 2 2 5.5.2.1 Iron Lower Fe levels in stream sediments are likely the result of differing digestion techniques between years (summarized in 5.1.2). The box-whisker plots for the 1973, 1989, and 1993 data illustrate no apparent trends in Fe levels (Figure 5.35). 5.5.2.2 Lead While a few stream stations have much lower Pb concentrations relative to 1973, the box-whisker plot (Figure 5.36) reveals that the majority of stream stations have experienced only modest changes. Large decreases in street dirt concentrations can be directly attributed to the phaseout and removal of T E L from gasoline in 1990. While Pb levels in street dirt are now only a fraction of their former values, the current median level is still high (222 mg/kg) relative to background levels, and presents a continued risk to streams . 5.5.2.3 Copper The box-whisker diagram illustrating total stream sediment metals (Figure 5.37) is dominated by three very contaminated (1973) stream sites in the industrial area surrounding Still Creek (#30, #31, and #35). The extremely high Cu levels are no longer present, implying abatement of the 1973 industrial source has occurred. The accompanying stream plot which excludes the 3 sites is more illustrative of the Cu increase in the remaining stations. Extractable Cu has also significantly increased (a very small fraction of the high Cu levels in 1973 were in the extractable form). The slight decrease in street Cu levels becomes apparent after removing three highly contaminated 1973 industrial street sites from the box plot. The decrease in street concentrations, coupled with increases in streams, suggest other significant non-street related Cu sources may be present in the watershed. The Cu concentrations in street dirt are still several times higher than background values, however, and present a risk to streams through urban runoff. 123 Figure 5.35 Changes in iron concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots (units in percent) Stream sediments (total) 20 May73 May89 Sept93 YEAR Stream sediments (extractable) V 2 Street dirt (total) 1 2 4 Figure 5.36 Changes in lead concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots (units in mg/kg) Stream sediments (total) 1000 800 600 400 200 Stream sediments (extractable) 800 600 f 400 200 h May73 May89 SeptSS May73 8ept93 YEAR YEAR 6000 4000 3000 2000 1000 Street dirt (total) May73 Sept93 YEAR 125 Figure 5.37 Changes in copper concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots (units in mg/kg) Stream sediments (total) Stream sediments (excluding stations #30,#31,#35) (total) 2000 1500 O 500 O 400 300 200 100 May73 May89 Sept93 May73 MayBQ Sept93 YEAR Street dirt (total) YEAR Street dirt (excluding stations 12,13,14 from May73) (total) 4000 3000 g ,2000 1000 400 300 O 200 100 May73 Sept93 May73 Sept93 YEAR YEAR 126 Figure 5.37 continued Stream sediments (extractable) 150 100 O 60 r May73 Sept93 YEAR 5.5.2.4 Zinc Zn levels in stream sediments have slowly increased over the last 20 years despite little or no increase in street dirt levels (Figure 5.38). This suggests another non-street related significant source may be present, or higher volumes of the street dirt (which is very contaminated relative to background levels and stream sediment levels) are reaching the streams through urban runoff. The most contaminated site in the 1973 study (#36) has been enclosed by a culvert and wasn't available for sampling in 1993. 127 Figure 5.38 Changes in zinc concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots (units in mg/kg) Stream sediments (total) Stream sediments (extractable) 1000 800 h 600 400 200 300 200 \-100 h May73 May89 Sept93 May73 Sept93 YEAR YEAR Street dirt (total) 1500 1000 \-600 May73 Sapt93 YEAR 128 5.5.2.5 Nickel The highest 1973 stream sediment N i concentrations can been attributed to discharges from electroplating industries surrounding Still Creek (Hall and Wiens 1976). These discharges have since been eliminated, resulting in decreased N i levels at the most contaminated stream stations (Figure 5.39). A comparison of extractable metal concentrations is not shown because only seven stations had concentrations above the detection limit (6 mg/kg) in 1993. The majority of stream sediment N i concentrations are within background levels. 5.5.2.6 Mercury Sampling in 1989 and 1993 confirms that Hg has increased significantly in stream sediments from 1973 (Figure 5.40). While the increase is on average small — similar to background levels measured in Gwendoline Lake (Figure 5.14) — any significant, measurable increase at the parts per billion level warrants concern. Mercury, especially when methylated, is an extremely toxic compound at low levels (Clarkson 1972). Increased traffic through the watershed is a possible source of the additional Hg, as crude oil can contain significant Hg concentrations (Fergusson 1990). In Valkovic's (1978) review of trace elements in petroleum, he found that Hg concentrates mainly in the highest boiling fractions. Asphaltenes and residual fuel oil contain the bulk of crude oil Hg, while a typical concentration for a distillate fuel oil is 0.005 ppm. Although typical concentrations are not available for gasoline, it is expected to contain much less Hg than fuel oil because it is a more volatile petroleum fraction. Extremely low concentrations in gasoline, and the low concentrations in street dirt suggest other sources may be more significant. Other potential sources of Hg to the environment include its use as a fungicide in lawn and garden applications and paint, industrial processes (chlor-alkali plants, pulp and paper, etc.), dentistry, batteries, switches, and refuse incineration (D'ltri 1972, Rasmussen 1994). A municipal waste incinerator in south Burnaby, operated by the G V R D since March 1, 1988, is a Hg source potentially affecting the Brunette watershed. 129 Figure 5.39 Changes in nickel concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots (units in mg/kg) Stream sediments (total) 100 80 60 40 20 Street dirt (total) 80 60 h Z 40 h 20 May73 May89 Sept93 May73 Sept93 YEAR YEAR Figure 5.40 Changes in mercury concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots (units in ug/kg) Stream sediments (total) 1000 800 600 \-400 200 400 300 g 200 100 Street dirt (total) May73 May89 Sept83 May73 Sept93 YEAR YEAR 130 Incinerator Hg releases peaked in the 3rd quarter of 1989 at 1.8 kg/day and have since decreased to 0.18 kg/day in the 3rd quarter of 1994 (calculated using G V R D stack emission data (Babensee 1994, Greater Vancouver Regional District 1992a)). Soil and vegetation have been monitored for contaminants each year since 1987 (prior to incinerator startup) at the sites shown in Figure 5.41 to determine i f incinerator emissions have adversely affected the surrounding area. The monitoring of Burnaby sites indicated an increasing trend in Hg vegetation levels from 1987 to 1989 (Figure 5.42). Although this suggests a direct impact caused by incinerator emissions, caution must be exercised in the interpretation. The method used for analysis during these years (hydride/ICP) was found unreliable for soils analysis in comparison with the cold vapour technique. The same comparison of techniques with 1989 vegetation samples found a very strong correlation between methods (r 2 = 0.86), with the hydride/ICP method showing a small negative bias (Soilcon Laboratories 1990). These data suggest that the incinerator may have caused an increase in Hg vegetation levels north of the incinerator. 5.5.2.7 Manganese The increase in Mn from 1973 to 1993 is most dramatically illustrated in the plot of extractable stream sediments (Figure 5.43). Insignificant quantities of the extractable metal were found in 1973 while the 1993 total increase is mainly in the extractable form. The gasoline additive, methylcyclopentadienyl manganese tricarbonyl (MMT), is an obvious new source of M n to the environment which may have caused this increase in stream concentrations. Assuming the decrease in street dirt Pb concentrations and the increase in M n concentrations are both caused by the change in gasoline additives, the relative street dirt concentration changes should reflect the difference in additive levels. The ratio of the peak Pb decrease (4365 mg/kg) to peak Mn increase (136 mg/kg) in street sediments equals 32 — very similar to the ratio of 1973 Pb additive loadings to 1993 M n additive loadings (summarized in Table 5.12) which equals 38. While changes in street dirt concentrations can 131 Figure 5.41 Location of G V R D incinerator and monitoring sites Burnaby Mountain • #5 Kilometer* 1 Legend A GVRD Incincerator ir Soil and vegetation monitoring sites Figure 5.42 Mercury concentrations in vegetation at G V R D monitoring sites from 1987 to 1989 • 1987 1988 11 1989 > E G O Site/Vegetation measurements below detection limit are shown at detection limit (0.05 mg/kg) 132 ;ure 5.43 Changes in manganese concentrations in Brunette watershed sediments from 1973 to 1993 illustrated using box-whisker plots (units in mg/kg) Stream sediments (excluding one 1989 outlier - 8794 mg/kg) (total) Stream sediments (extractable) 5000 4000 3000 2000 1000 3000 2000 1000 May73 May89 Sept93 May73 Sept93 YEAR YEAR Street dirt (total) 500 400 300 200 100 May73 Sept93 YEAR 133 be simply explained by gasoline additive levels, the much larger increases in stream sediment M n levels cannot. A more detailed examination of the geochemical properties of M n is required to explain the observed increase. Unlike many of the other trace metals which have been examined in this study, Mn can exist in several different oxidation states. The most common forms observed in the aquatic environment are the +2, +3, and +4 states. The divalent state is extremely soluble while both of the other states are relatively insoluble. In near neutral surface waters the insoluble +4 state is favoured, however, in slightly reducing conditions, the equilibrium is shifted to the soluble M n + 2 ion. Further complicating this element's chemistry are the kinetics of equilibrium and adsorption reactions. A considerable amount of soluble divalent M n exists normally in oxygenated riverine systems because the M n oxidation is kinetically limited (Campbell and LaZerte 1988). Like the other metals examined, the partitioning of the soluble M n ion in the water column is affected by surface adsorption to particulate matter. Laboratory analysis has shown manganese oxide particles preferentially adsorb manganese ions over other strongly binding ions (ie. N i , Zn, Cd) (Gadde and Laitinen 1974, Murray 1975). Hem (1981) has shown that the adsorption of M n + 2 onto manganese oxides, and subsequent rapid surface-mediated oxidation, can act as a "catalytic" reaction by continually providing new surface area for adsorption. This interaction between soluble and insoluble forms of M n is a possible clue to the behaviour of Mn in the Brunette system. The 1973 stream sediments contained very small quantities of weak acid extractable Mn. This indicates very little of the sediment M n existed as oxide particles in 1973 since metal oxides are solubilized with this digestion procedure (Pickering 1981). In contrast to the 1973 results, a large fraction of the Mn in 1993 sediments occurred in the extractable form. This indicates a large fraction of oxides or adsorbed ions. The introduction of M n from automotive emissions, precipitating as an oxide after combustion (Hotz 1988), may have provided the catalyst which has scavenged soluble stream M n and produced the high concentrations found in the 1993 sediments. 1 3 4 5.5.3 Spatial analysis of trace metal data 5.5.3.1 Street sediment analysis Street sediment sampling stations were originally categorized in the 1973 study according to the surrounding predominant land activity : Commercial(C), Industrial(I), Residential(R), and Green space(G). This grouping, illustrated in Figure 5.44, shows that street sediment trace metal levels are consistently lower in green space areas relative to all other land uses — a result also found by Hall et al. (1976). No other consistent land activity/contaminant patterns are apparent. The street stations have also been categorized according to traffic level using the median traffic volume of 7425 vehicles per day as the division between the high and low groups. While an analysis based on traffic volume is useful, some limitations are inherent. Traffic estimates are not as useful or available in parking lot areas (identified as a significant source of trace metals in the Bannerman et al. (1993) study). In addition, the large within-station variation measured at station C4 (Figure 5.2) implies uncertainty of individual station results. Significant differences (tested using the Mann-Whitney test; a=0.05) between traffic groupings for Pb and Cr indicate a direct relationship between these contaminants and traffic volume. Iron and Zn are also significantly different between traffic groupings when the significance levels is raised to 10 %. These results do not necessarily imply that Cu, N i , Mn, and Hg stream contamination is unrelated to traffic volume. A study of nonpoint pollution in Wisconsin (Novotny et al. 1985) observed differences in sediment accumulation as great as 10 times (by weight) between high and low traffic volume streets. If total trace metal load had been measured at each sampling location, the effects of traffic groupings would have been more evident. 135 Figure 5.44 Metal concentrations in street sediments illustrated using box-whisker plots(units in mg/kg except Fe in percent and Hg in Lig/kg) Activity Grouping Traffic Grouping C Q I R C Q I R # of stations in each group : C-commercial (6), G-green space (3), I-industrial (8), R-residential (8); high traffic (13), low traffic (12) 1 3 6 Figure 5.44 continued Activity Grouping Traffic Grouping 5 *° [ C Q I fl U! 20 # of stations in each group C-commercial (6), G-green space (3), I-industrial (8), R-residential (8); high traffic (13), low traffic (12) 137 5.5.3.2 Stream sediment analysis Relating the spatial differences in stream sediment trace metal concentrations to land use and traffic patterns is a major goal of this study. Quantifying land use influences at each individual sampling station, however, is an unrealistic objective for the following reasons. (i) The standard error of station metal concentrations, estimated by analyzing 3 samples per station, is too large to attempt to differentiate individual stations by subtle land use differences (Figure 5.1). (ii) Contributing drainage area land use information is not available for each station because of the complexity of the urban drainage area (discussed in section 4.4). (iii) Differences in hydrology make direct comparisons between each station invalid. Higher order streams can be expected to accumulate more contaminants simply as a result of the stream physics (Colman and Sanzalone 1992). There have been no studies which have convincingly shown or predicted the upstream origin of a streambed sediment — a necessary precondition for assigning land use causes to individual station contaminants. These analytical problems can be largely overcome by aggregating stream stations at a larger scale. The focus of this analysis is on the Still Creek and Brunette River sub-catchments areas (Figure 5.45), chosen because of their similarity in hydrology and sampling coverage. A statistical comparison of sub-basins at a smaller scale is not possible because of an inadequate number of sampling stations. Both of the Still Creek and Brunette River sub-catchment areas contain high energy, swiftly flowing streams, as well as quiescent regions near the mouth. They contain approximately the same number of sampling locations and an adequate number to provide statistical significance. Point industrial sources, which would obscure other land use relationships, were prevalent in the Still Creek area in 1973, however have abated at the present time. Finally, the analysis of Burnaby Lake core #3 (Figure 5.10) indicates that Still Creek trace metal loads remain in the western portion of the lake and are unlikely affecting the Brunette River: providing evidence for the independence of the two sub-systems. 1 3 8 Figure 5.45 Still Creek and Brunette River sub-catchment areas 1 3 9 A comparison of the indicators of sediment texture show no significant differences between sub-basins (Table 5.17), confirming the validity of this sub-basin analysis. The Still Creek basin shows significant enrichment of Pb, Cu, and Zn relative to the Brunette system. Several differences in land use between these catchment areas (Table 5.18) offer potential explanations for the higher contaminant levels in the Still Creek area. The difference in traffic volumes between basins is a potential cause of this enrichment based on the results of studies which have shown traffic to be a significant source of all three of these metals to urban runoff (Gibb et al. 1991, Woodward-Clyde Consultants 1992). Large building roofs and parking lots, characteristic of commercial and industrial areas, can also contribute large Pb, Cu, and Zn loads to runoff (Bannerman et al. 1993, Good 1993, Hey and Schaefer 1983). The results from two stations in the upper part of Still Creek (#37: near Atlin St. and #33: near Grandview), however, suggest commercial and industrial land usage may not be responsible for the bulk of trace metal contamination. These stations receive stormwater drainage from the enclosed headwater area of Still Creek (sub-basin #2 in Figure 4.5) which has a low percentage of commercial and residential area (10 %) yet has a high percentage of impermeable area (49 %) and high traffic volumes (Table 5.11). The concentrations of Pb, Cu, and Zn measured at these stations are above median values in the Still Creek catchment area, implicating traffic and land cover as more significant land use factors than commercial/industrial usage. Contamination from sewage, known to be present in the stormwater system above Atlin Street, may also be a significant source of Cu to Still Creek in this area. Very high traffic density (Table 5.11) and a large percentage of impermeable areas (44 %) in the sub-basin surrounding the lower reaches of the Brunette River (sub-basin #29 in Figure 4.5) have not caused Pb, Cu, and Zn levels in the surrounding stream stations (#1 -#6) to even reach the median levels encountered in the Still Creek catchment area. The same stream sediment stations, in 1973, were only slightly enriched in Pb compared to Still Creek stations (Figure 5.31), even though the surrounding traffic density was similar to many Still 140 Table 5.17 Median streambed total metal concentrations in the Brunette River and Still Creek sub-basins for 1973 and 1993 (Differences between sub-basins compared using the Mann-Whitney U test. Differences between years compared using the Wilcoxon signed-rank test). Element Year Still Creek sub-basin median (n=12) Brunette R. sub-basin median (n=13) Ratio Still: Brunette Mn 1973 318 360 0.8 (mg / kg) 1993 576 b 807 b 0.7 Pb 1973 324 50 6.5 a (mg / kg) 1993 130 b 55 2.4 a Cu 1973 58 17 3.4 a (mg / kg) 1993 130 51 b 2.6 a Zn 1973 154 56 2.8 a (mg / kg) 1993 251 128 b 2.0 a Ni 1973 19 10 1.9 a (mg / kg) 1993 17 12 1.4 Cr 1973 113 88 1.3 (mg / kg) 1993 32 25 1.3 Hg 1973 34 17 2.0 a (ug / kg) 1993 141 b 103 b 1.4 Fe 1973 2.4 2.3 1.0 (%) 1993 2.1 2.1 1.0 Silt and Clay 1973 29.7 20.4 1.5 (%) 1993 40.5 40.9 b 1.0 LOI 1973 6.2 3.7 1.7 (%) 1993 6.2 5.0 1.2 a - significant difference between sub-basins at the 5% probability level. b - significant difference between years within sub-basins at the 5% probability level. - cannot evaluate difference between years for Cr because of a difference in analytical methods 141 Table 5.18 Land use and traffic density in the Brunette River and Still Creek catchment areas in 1973 and 1993 Sub-basin Year Traffic Density (Vehicle km / day) Impermeable Area (%) Commercial & Industrial Area (%) Population Employment Brunette 1973 710,000 26 13 20,700 8690 River 1993 998,000 35 15 34,900 18,500 Still 1973 1,520,000 46 20 72,400 22,530 Creek 1993 2,240,000 52 22 86,000 50,050 Creek areas and use of the additive T E L was at it's peak. Although this appears contradictory to the Still Creek results previously discussed, these results are probably caused by the dilution of contaminated sediments with large volumes of clean sediment generated in upstream areas. These observations suggest the importance of maintaining highland and upstream areas in relatively natural states (with permeable land cover). Neither Hg or Mn exhibit significant differences between sub-basins, suggesting land uses within the watershed may not be responsible for increases observed between 1973 and 1993. However, i f particulate Mn from automotive exhausts is indeed catalyzing the removal of soluble stream Mn, sediment levels may be more dependent on ambient soluble concentrations rather than traffic density. 5.5.4 Trace metal inter-relationships An examination of the relationships between metals provides some indication of the nature of the pollutant source. The four relationships exhibiting the largest Spearman rank correlation coefficients in each of the stream and street sampling sets (full correlation matrices in Appendix J) are displayed in Figure 5.46. Trace metals solubilized with the weak acid digest, most indicative of anthropogenic sources, exhibit strong relationships between each of Pb, Cu, and Zn. The same relationships are observed in the strong acid digest, in 1 4 2 addition to a Ni-Cr relationship. Given the low indication of contamination of these two metals this correlation is likely indicative of a natural linkage in the sediment mineral structure. The street sediment relationships are quite different — characterized by the presence of Fe in all of the most significant interactions. This quite likely indicates the introduction of Fe as a corrosion product onto the streets. Fe may be scavenging Cu, Zn, and M n ions as a result of the surface-binding properties of iron oxides and hydroxides. Figure 5.46 Largest Spearman rank correlation coefficients in stream and street sediments Street Sediment Stream Sediment Stream Sediment (cone, nitric acid) (weak HC1 acid) (cone, nitric acid) 0.88 Ni Cr 0.72 Mn Pb Pb Zn Cu Zn Cu Mg 0.69 Zh Cu 0.82 0.78 0.90 143 6. SUMMARY AND CONCLUSIONS Results from each of the sampling areas — lake, stream, and street sediments — and land use studies are summarized and synthesized in this section for the purpose of explicitly addressing the research objectives. Measured contaminant levels are compared to areas with similar geographic and land use characteristics to provide a relative indication of the magnitude of contamination. The observation of spatial and temporal changes in trace metal levels illuminates relationships between land use and water quality. This, in turn, provides an indication of probable future environmental conditions based on land use decisions made today. 6.1 Extent and severity of trace metal contamination in aquatic sediments In common with many other urban areas, the lake and stream sediments in the Brunette Watershed are contaminated to varying degrees with Pb, Cu, and Zn. Significant enrichment of these three elements in stream sediments has been measured relative to the Salmon Watershed in Langley, British Columbia. Surface lake sediments in the western end of Burnaby Lake are enriched in these three elements above background levels by factors of 10 (Pb), 6.0 (Cu), and 3.8 (Zn). Other elements measured in the Burnaby Lake surface sediments were only slightly enriched ( Cd - enrichment factor of 1.5) or at background levels (Cr, N i , Mg , Fe, Mn). Both Hg and Mn are enriched in stream sediments relative to past surveys in the watershed. Analysis of sediments from Gwendoline Lake — a lake within in a forested catchment area 30 kilometres to the north-east of the Brunette watershed — revealed enrichment factors (at the surface) of 12, 2.3, and 1.8 for Pb, N i , and Zn respectively. These results indicate the effects of atmospheric transport of urban-generated contaminants. Although the enrichment factors are similar, or even greater at Gwendoline Lake, the trace metal flux is many times greater in Burnaby Lake sediments. The fluxes of Pb, Cu, and Zn to Burnaby 1 4 4 Lake are from 3 to 10 times higher than measured in comparable urban lakes in Seattle (Lake Washington) and Melbourne, Australia (Botanic Gardens Lake). With the exception of two stations, all of the sediments collected from the 33 stream stations failed to meet watershed objectives for Pb, Cu, and Zn. Mercury objectives were met at only 13 of the stations. Other, less stringent, contaminant criteria were exceeded for Pb, Cu, and Zn at fewer stations (ranging from 9 to 28). Spatial patterns of contamination were evident for many metals within Burnaby and Deer Lakes, and throughout the streams in the watershed. Trace metal enrichment in lake sediments was focussed in relatively small areas of the lakes which accumulated sediments derived from the catchment areas. A spatial contaminant pattern in stream sediments was only apparent for Pb, Cu, and Zn. With very few exceptions, the highest Pb, Cu, and Zn sediment concentrations were recorded in the streams located throughout the Still Creek catchment area. Stations located on Eagle Creek, Stoney Creek, and the Brunette River contained lower concentrations of the three metals, yet some stations still exceeded sediment criteria and objectives. 6.2 Changes in sediment trace metal levels since 1830 Analysis of lake core sediments in Burnaby Lake was relied upon to determine the longer term changes in watershed trace metal levels while stream and street sediments provided more detail of changes which have occurred since 1973. With the exception of Cd, which experienced a brief sharp peak around 1850, trace metals (Pb, Cu, Zn, Cr, and Ni) began slowly increasing above background levels in the 1890's. Increases in the concentration of each of the metals accelerated following 1950. Peak concentrations and fluxes of Cu, Cd, N i , and Cr were reached in 1970, while the Pb peak occurred approximately 15 years later. Zinc concentrations have continued to steadily increase in Burnaby Lake sediments. 1 4 5 Many changes in metal concentrations between 1973 and 1993 were observed by comparing stream and street sediment metal levels with values recorded at the same stations twenty years earlier. Of all the metals examined, Pb was the only element which exhibited a significant decrease across the entire watershed. The statistically significant stream sediment differences across the watershed, summarized here as the median of the percentage differences between years at each station, were recorded as -35 %, + 81 %, + 45 %, + 294 %, + 131 %, for Pb, Cu, Zn, Hg, and M n respectively. Some large decreases in metal concentrations occurred at individual stream stations between 1973 and 1993 while the overall watershed concentrations increased or did not change. Copper has generally increased throughout the watershed yet three Still Creek stream stations, in an area surrounded by industry, have experienced average reductions of 80 %. In the same area, two stream stations on Still Creek have experienced an average N i reduction of 78 %. The statistically significant street sediment differences across the watershed, summarized here as the median of the percentage differences between 1973 and 1993 at each station, were recorded as -67 %, - 43 %, and + 43 %, for Pb, Cu, and M n respectively. No significant changes in Zn, N i , Fe, and Hg street sediment concentrations were detected. 6.3 Land use in 1973 and 1993 Land activity, land cover, traffic density and demographic data were collected for 1973 and 1993 to explore possible causes for spatial and temporal patterns of trace metal contamination. Changes in land use across the entire watershed and spatial land use differences in 1993 have been examined in the most detail. Overall changes in land activity and land permeability between 1973 and 1993 have been relatively minor. The largest activity changes have occurred in residential areas (increased in area by 5 %) and open space areas (reduction in area by 5 %). Land permeability, an important measure because impervious areas efficiently transport 1 4 6 contaminants to streams and affect stream hydrology, has been reduced by 7 %. More significant demographic and traffic changes have occurred which have potentially contributed to changes in trace metal levels. Population and employment have increased by 30 % and 120 % respectively throughout the watershed since 1973. These increases, which have occurred on approximately the same land base, signal a trend towards a densification of development. This densification is facilitated, in part, by a shift in employment away from a manufacturing base towards one dominated by the services sector. Accompanying these demographic changes has been an increase in traffic density by 44 %. To provide more insight into the effects of land use on stream quality, the present land uses in the Still Creek and Brunette River sub-catchment areas have been compared and contrasted. The Still Creek area is more highly developed, with 2.5 times the population, and 2.2 times the traffic density of the Brunette area. Impervious areas account for 52 % of the Still Creek basin and less than 35 % of the Brunette River sub-basin. Commercial and industrial uses occupy 15 % of the Brunette River sub-basin area and 22 % of the Still Creek basin. An understanding of land use-water quality relationships wil l allow informed management decisions to be made regarding the protection of relatively healthy urban aquatic habitats, which can be found in the Brunette River sub-basin, and the rehabilitation of degraded environments, such as the Still Creek area. 6.4 Relationships between land use and trace metal contamination Variability, inherent in non-point source pollution and stream environments, dictates the use of several different types of analysis to obtain confidence in identifying a land use cause for a stream effect. Lake core analysis, stream and street sediment results, and compiled land use information have all been used to explore the relationships between land use and stream quality. 1 4 7 The analysis of the Burnaby Lake core has provided evidence of past land use-water quality relationships which have now been modified. A very large peak in Cu loading, and smaller peaks in Cd, N i , and Cr loadings, occurred in 1970 and are related to industrial sources in the Still Creek area. These sources are no longer apparent in the watershed. The rapid increase in Pb concentrations from 1950, also observed in many other urban areas, is related to emissions generated from automobile use. The gradual decline in Pb concentrations from the peak in the 1980's corresponds to the phaseout and removal of the gasoline additive — TEL. Recent sediments deposited in the Burnaby Lake core also provide information of current and continuing land use affects. Zinc concentrations have continued to increase over time at the same rapid rate which began in 1950; implying a land use cause which has also experienced rapid and unabated growth. Copper levels have decreased from 1970 but still remain very high, unlike Cd, N i , and Cr which have returned to near background levels. This observation suggests that although the 1970 industrial contaminant sources have been tempered, additional Cu sources persist in the watershed. The high surface sediment concentrations of Pb also implies a continuing source which is independent of the now eliminated gasoline additive. The analysis of the stream and street surface sediments builds on the lake core results to provide some indication of land uses contributing to the problem of trace metal contamination. Differences in the stream contamination experienced in the Still Creek sub-basin relative to the Brunette River sub-basin have provided some indication of land uses contributing trace metal loads. Lead, Cu, and Zn were higher in the stream sediments in the Still Creek basin, while no significant difference between basins was detected for Hg, Mn, Cr, Ni , Fe, and Mg. High Pb, Cu, and Zn levels in Still Creek sediments upstream of industrial and commercial areas suggest this land use is not the cause of differences observed between sub-basins. Differences in traffic density and land permeability are potential causes of the higher levels of the three metals in the Still Creek sub-basin. 148 Another indication of the nature of the trace metal sources was provided through a statistical analysis of the relationships between trace metal levels (Spearman Rank correlation analysis). This analysis showed that Pb, Cu, and Zn were significantly correlated to each other in all of the collected stream samples, indicating that a single source may be responsible for the enrichment of these metals in the watershed. Relationships between street and stream sediment metal levels provide an indication of the significance of street-generated contaminants. The ratios of median street sediment concentration to the corresponding median stream sediment concentration, in descending order, are 3.5, 2.6, 2.3, 2.1, 1.7, 1.1, 1.0, 0.5, and 0.4 for Pb, Zn, Cu, N i , Cr, Mg , Fe, Hg, and M n respectively. These ratios suggest street generated contaminants are at least partially responsible for enrichment of Pb, Zn, Cu, Ni , and Cr in stream sediments. Traffic volumes at each of the street sampling locations were used to group stations for the purposes of assessing the traffic density effect on trace metal generation. The comparison, only partially complete because total loadings were not measured, indicated significantly higher concentrations of Pb and Cr on streets with higher traffic levels. Zinc and Fe levels were also significantly higher when the statistical significance level was increased from 5 % to 10 %. The influence of traffic was also evaluated by comparing estimated loadings of automotive-derived Cu and Zn (2.1 and 6.0 kg/day respectively), with an estimate of total Cu and Zn loading in stormwater (1.7, and 2.0 kg/day respectively). This comparison indicates the importance of the traffic source to Cu and Zn loadings in stormwater. A l l of the preceding analysis supports a conclusion, reached in several other studies of urban watersheds, that traffic density is responsible for a large part of Pb, Cu, and Zn contamination in urban streams. Stream and lake sediment concentrations of Pb and Zn , in particular, are directly affected by the density of surrounding traffic. Some conflicting evidence, such as the large increase (between 1973 and 1993) in stream sediment Cu levels while street sediment concentrations decreased, implies other sources also contribute to high environmental Cu levels. Sewage infiltration to stormwater systems is one possible source. 1 4 9 Land use causes of the observed increases in Hg and Mn stream sediment concentrations are less readily apparent. Increases in street sediment M n concentrations are explained by the now widespread usage of the gasoline additive, M M T , but the magnitude of the increase does not explain the larger increases experienced in stream sediments. It is possible that M n particles, introduced into streams from automotive exhaust, are scavenging soluble stream M n from the water column, thus increasing sediment concentrations by many times the initial input from automobiles. The large increase in stream sediment Hg concentrations are a concern because of the toxic properties of Hg at very low concentrations. Variability in Hg concentrations cannot be correlated to any of the land use factors examined, which suggests the source may be external to the watershed. A solid waste incinerator which began operation in 1988 (located in Burnaby, 3 kilometres south of the Brunette Watershed) is a possible source of the observed Hg increase. The degree of land impermeability is a land use factor which may not contribute large trace metal loads of itself, but allows other sources to efficiently transport metals unimpeded to stormwater drains and ultimately streams. The spatial distribution of impermeable areas can also affect the extent of contamination throughout the watershed. Relatively permeable, undeveloped lands in the upland areas of the Brunette River sub-catchment region have tempered the effects of contaminants generated in the densely developed, lower reaches of the catchment area. The spatial distribution of permeable land areas in this watershed is changing, though, as the most significant reductions in land permeability over the last 20 years have occurred in the upland (Burnaby Mountain) areas surrounding Eagle and Stoney Creek. Protection of the remaining viable aquatic habitats, such as Stoney Creek and the Brunette River, wil l be dependent on controlling both the traffic increases and conserving permeable land cover. 1 5 0 7. R E C O M M E N D A T I O N S The conclusions drawn from this study can be used to inform management decisions concerning the remediation and conservation of natural areas in the Brunette Watershed, and urban watersheds in general. A need for further research which would enable better understanding of the source, transport, and fate of trace metals in urban environments has also been identified by these results. 7.1 Implications for further research The increase in stream sediment Hg concentrations is cause for concern, yet very little is known about the implications for aquatic health at the levels measured. Further research to define the degree of bio-availability of the Hg in stream sediments and to assess whether bio-magnification has occurred, will help to define the current risk to aquatic organisms. Research is also required to confirm the Hg source and if point-related, to define the extent of impact. Copper increases, at levels which pose a known threat to aquatic health, cannot be accounted for by traffic increases alone. Additional research is necessary to identify and test the significance of other potential sources. The implications of the increase in stream sediment M n — the largest absolute increase of all the metals examined — is uncertain because of naturally high background levels. Most of the increase occurred in the extractable form which indicates a high degree of bio-availability. The first step in defining the risk present would involve an examination of M n bio-accumulation in benthic organisms. An automotive source of the additional Mn has been postulated yet there is little direct evidence to support the conclusion. If particulate Mn, introduced by automobile exhaust, is indeed scavenging soluble stream Mn, then these particles may also be scavenging additional soluble Pb, Cu, and Zn. An examination of current and historical water quality data from the watershed wil l help to test the source hypothesis. Future policy 151 changes which may eliminate the addition of M M T to gasoline will provide an opportunity for examining sediment-particle interactions in a natural environment. Quantifying the effect of impervious land surfaces on stream contamination will assist decision makers in justifying the necessary preservation of green space areas within urban developments. This quantification could be accomplished through an extensive simultaneous monitoring of stormwater runoff from several different small Brunette watershed sub-basins. Detailed land use information, already digitized into a GIS for this study, could then be used to explore the relationships between stream contamination and land permeability. 7.2 Management implications While many uncertainties still remain, two conclusions from this study suggest management actions in the watershed are necessary. The first conclusion — that many streams in the watershed are heavily contaminated with trace metals at levels hazardous to aquatic health — demands a remediation response. The second conclusion — that increasing rates of trace metal contamination are beginning to threaten cleaner stream sections which now support many levels of aquatic life — demands a conservation and prevention response. Still Creek is an obvious choice for remediation work because of its contaminated nature and increasing community interest. The first step of remediation in this case involves drastically reducing the trace metal load to the stream through the use of stormwater collection and treatment. A low cost pilot project, using Best Management Practices (BMP's), would be a practical first step. Instituting a pilot project in the headwater area of the creek would ensure that the effects of the project on stream contaminant levels could be assessed. In conjunction with a pilot treatment project, further work needs to be done to identify and eliminate sewage connections to the stormwater collection system in the Still Creek area. 152 Increasing contaminant levels across the entire watershed warrant efforts to control the sources and reduce their impact. Traffic is a major source of trace metals to streams and needs to be managed to avoid further stream contamination. Reducing traffic volume is an obvious solution which will require cooperation throughout the entire regional district. A secondary, yet still important solution to this problem, lies in examining and reducing the specific contaminant sources in automobiles. Much of the trace metal load emitted from automobiles derives from background levels in fuel, and is not amenable to abatement measures. The decomposition of tires (Zn source) and brake lining (Cu source) is a metal source which could be better managed. Adjustment of manufacturing processes could conceivably eliminate these sources of Cu and Zn to the urban environment. The results of this study also show that conservation of green space areas in upland and headwater areas can serve to mitigate contaminant sources located lower in the stream and river systems. Incremental development in areas such as those surrounding Eagle Creek and Stoney Creek should be evaluated in the context of preserving the aquatic health of the entire river system. 1 5 3 8. REFERENCES Alberta Research Council. 1982. Alberta motor gasoline survey 1980. Alberta Research Council. 1994. Composition of Canadian summer and Winter gasolines (sulphur, manganese. T90) 1993. Canadian Petroleum Products Institute. Anderson, B.C. 1982. Toxicity of urban stormwater runoff. M . S c , University of British Columbia. A P H A . 1989. Standard methods for the examination of water and wastewater. 17th ed. Washington, D . C : American Public Health Association. Appleby, P.G., and F. Oldfield. 1978. The calculation of lead-210 dates assuming a constant rate of supply of unsupported 210Pb to the sediment. Catena 5 : 1-8. Armstrong, L.J . 1994. Contribution of heavy metals to storm water from automotive disc brake pad wear. Prepared for the Santa Clara Valley Non-Point Source Pollution Control Program. Athayde, D.N. , P.E. Shelley, E.D. Driscoll, D. Gaboury, and G. Boyd. 1983. Results of the Nationwide Urban Runoff Program. Vol.LFinal report. NTIS, PB84-185552. B.C. Environment. 1992. The attainment of ambient water quality objectives in 1990. Ministry of Environment, Lands and Parks, Province of British Columbia. B.C. Environment. 1993a. The attainment of ambient water quality objectives in 1991. Ministry of Environment, Lands and Parks, Province of British Columbia. B.C. Environment. 1993b. Water quality in British Columbia : objectives attainment in 1992. Water Quality Branch. B.C. Environment. 1994. The attainment of ambient water quality objectives in 1993. (unpublished). Ministry of Environment, Lands and Parks, Province of British Columbia. Babensee, C. 1994. Burnaby incinerator - 03 1994 Stack Monitoring. Greater Vancouver Regional District. Bannerman, R.T., D.W. Owens, R.B. Dodds, and N.J. Hornewer. 1993. Sources of pollutants in Wisconsin stormwater. Wat. Sci. Tech. 28 (3-5) : 241-259. Baxter, D. 1994. Personal communication. G V R D Development Services, Burnaby, B.C. Benedict, A . H . , K.J . Hall, and F. Koch. 1973. A preliminary water quality survey of the lower Fraser River sytem. Westwater Research Centre, University of British Columbia. Bennett, J. and J. Cubbage. 1991. Summary of criteria and guidelines for contaminated freshwater sediments. Environmental Investigations and Laboratory Services, Washington State Department of Ecology. 1 5 4 Bichler, H . 1994. Manager. Geochemistry. Chemex Labs. North Vancouver. B .C. Bindra, K.S . 1983. Mobilization of selected trace metals in the aquatic environment. PhD, University of British Columbia. Bindra, K.S. , and K.J . Hall. 1977. Geological partitioning of trace metals in sediments and factors affecting bioaccumulation in benthic organisms (unpublished). Westwater Research Centre, University of British Columbia. Binford, M . W . 1990. Calculation and uncertainty analysis of 210Pb dates for PIRLA project lake sediment cores. J. Paleolimnology 3 : 253-267. Bonnevie, N.L . , R.J. Wenning, S.L. Huntley, and H . Bedbury. 1993. Distribution of inorganic compounds in sediments from three waterways in Northern New Jersey. Bul l . Environ. Contam. Toxicol. 51 : 672-680. British Columbia Institute of Technology. 1992. Trace metals in fish (Carp) livers taken from Burnaby Lake (unpublished). BCIT, Renewable Resources. British Columbia Institute of Technology. 1994. The Burnaby Lake system : A proposal for a community project to protect, enhance and rehabilitate the area's fish, wildlife and recreation resources. BCIT, Fish, Wildlife and Recreation Program. Campbell, P.G.C., and B. LaZerte. 1988. Environmental Chemistry - General Overview. In Manganese in the Canadian Environment. Edited by P. M . Stokes. 1-21. Ottawa Canada: National Research Council of Canada. Campbell, P.G.C., A . G . Lewis, P .M. Chapman, A . A . Crowder, W.K. Fletcher, B . Imber, S.N. Luoma, P . M . Stokes, and M . Winfrey. 1988. Biologically available metal in sediments. National Research Council of Canada, N R C C 27694. Campbell, P.G.C., and A . Tessier. 1991. Biological availability of metals in sediments: Analytical approaches. In Heavy metals in the environment. Edited by J. P. Vernet. 161-173. New York: Elsevier Science Publishing Company. City of Burnaby. 1987. Official community plan for Burnaby. British Columbia. Burnaby Planning and Building Inspection Department. City of Burnaby. 1993. The state of the environment report (SOER) for Burnaby. Environment and Waste Management Committee. City of Burnaby. 1994. City of Burnaby : Water quality - bacteriological data (unpublished). City of Burnaby, Environmental Health Department. Clarkson, T.W. 1972. Biological effects of mercury compounds. In Environmental Mercury Contamination. Edited by R. H . a. B. D. Dinman. 341-346. Ann Arbor, Michigan: Ann Arbor Science. Coastline Environmental Services. 1987. Greater Vancouver liquid waste management plan -1987. Prepared for the Greater Vancouver Regional District. 155 Colman, J.A., and R.F. Sanzalone. 1992. Geochemical characterization of streambed sediment in the upper Illinois River basin. Water Res. Bull . 28 (5) : 933-949. Commission on Lead in the Environment. 1985. Lead in gasoline: A review of the Canadian policy issue. The Royal Society of Canada. Cook, K . E . 1994. An examination of water quality and land use in the Salmon River watershed, Langley, B C , using GIS techniques. M . S c , University of British Columbia. Cummins, K .W. , and M.J . Klug. 1979. Feeding ecology of stream invertebrates. Ann. Rev. Ecol. Svst. 10 : 147-172. D'ltri, F . M . 1972. Sources of mercury in the environment. In Environmental Mercury Contamination. Edited by R. H . a. B. D. Dinman. 5-25. Ann Arbor, Michigan: Ann Arbor Science. Dawson, L . , M . Flaherty, and M . Gang. 1985. Still Creek Inventory. Part I. Rupert-Cassiar Neighbourhood Association. Dempsey, B .A. , Y . L . Tai, and S.G. Harrison. 1993. Mobilization and removal of contaminants associated with urban dust and dirt. Wat. Sci. Tech. 28 (3-5) : 225-230. Dinicola, R.S. 1990. Characterization and simulation of rainfall-runoff relations for headwater basins in Western King and Snohomish Counties. Washington. U.S. Geological Survey, 89-4052. Duynstee, T. 1990. Evaluating metal contamination in an urban watershed using Geographic Information Systems. B .Sc , University of British Columbia. Eakins, J.D., and R.T. Morrison. 1978. A new procedure for the determination of lead-210 in lake and marine sediments. Appl. Radiat. Isot. 29 : 531-536. Engel, B .A. , R. Srinivasan, J. Arnold, C. Rewerts, and S.J. Brown. 1993. Nonpoint source pollution modeling using models integrated with geographic information systems (GIS). Wat. Sci. Tech. 28 (3-5) : 685-690. Engstrom, D.R., and H.E. Wright Jr. 1984. Chemical stratigraphy of lake sediments as a record of environmental change. In Lake sediments and environmental history. Edited by E. Y . Haworth and J. W. G. Lund. 11-67. Leicester University Press. Environment Canada. 1991. Historical streamflow summary in British Columbia to 1990. Inland Waters Directorate, Water Resources Branch, Water Survey of Canada, Ottawa, Canada. Environment Canada. 1992. State of the environment for the lower Fraser River basin. SOE Report No. 92-1. Environment Canada. 1994. Digital archive of Canadian climatological data. Atmospheric Environment Service, Environment Canada. 156 EPA. 1986. Test methods for evaluating solid waste. 3rd Ed. Vo l . 1A. United States Environmental Protection Agency. NTIS, SW-846. Faegri, K . , and J. Iverson. 1989. Textbook of pollen analysis. 4th ed. John Wiley and Sons. Federation of B.C. Naturalists. 1994. Environmentally important sites in the G V R D - Draft report. F .B.C.N. , Land for Nature . Fergusson, J.E. 1990. The heavy elements: chemistry, environmental impact and health effects. 1st ed. Pergamon Press. Fletcher, K . 1993. Personal communication. Professor of Geological Sciences, University of British Columbia. Forget, E. , F. Courchesne, G. Kennedy, and J. Zayed. 1994. Response of Blue Spruce (Picea pungens) to manganese pollution from M M T . Water. Ai r and Soil Pollut. 73 : 319-324. Forstner, U . 1990. Inorganic sediment chemistry and elemental speciation. In Sediments: Chemistry and toxicity of in-place pollutants. Edited by R. Baudo, J. P. Giesy and H . Muntau. 61-105. Chelsea, Michigan: Lewis Publishers. Forstner, U . , and G. Miiller. 1973. Metal accumulation in river sediments: A response to environmental pollution. Geoforum 14 : 53-61. Gadde, R.R., and H.A. Laitinen. 1974. Studies of heavy metal adsorption by hydrous ion and manganese oxides. Anal. Chem. 46 : 2022-2026. Gardner Dunster Associates Ltd. 1992. The nature of Burnaby: A n Environmentally Sensitive Areas strategy. Prepared for the City of Burnaby. Gibb, A. , B . Bennet, and A. Birkbeck. 1991. Urban runoff quality and treatment: A comprehensive review. British Columbia Research Corporation. Good, J.C. 1993. Roof runoff as a diffuse source of metals and aquatic toxicity in stormwater. Wat. Sci. Tech. 28 (3-5) : 317-321. Greater Vancouver Regional District. 1989. The livable region : A strategy for the 1990's. G V R D , Development Service. Greater Vancouver Regional District. 1992a. Burnaby Incinerator: Summary of stack monitoring data. Greater Vancouver Regional District. 1992b. Metropolitan Vancouver transportation model. G V R D , Development Services Department. Greater Vancouver Regional District. 1994a. 1992 Greater Vancouver Travel Survey. Report 5: Vehicle and transit volumes. G V R D , Strategic Planning Department. Greater Vancouver Regional District. 1994b. Summary Report - Still Creek sampling project: Coliform counts 1993-1994 (unpublished). G V R D , Sewerage and Drainage Department. 157 Hakanson, L . 1984. Sediment sampling in different aquatic environments: Statistical aspects. Water Resour. Res. 20 (1) : 41-46. Hakanson, L . , and M . Jansson. 1983. Principles of Lake Sedimentologv. Springer-Verlag Berlin Heidelberg New York Tokyo. Hall, K .J . 1976. A benthic invertebrate survey of the Salmon and Brunette watersheds (unpublished). Westwater Research Centre, University of British Columbia. Hall, K . J . 1995. Unpublished data. Westwater Research Centre, University of British Columbia. Hall , K.J . , V . K . Gujral, P. Parkinson, and T. Ma. 1984. Selected organic contaminants in sediments and fish from the Fraser River estuary. Westwater Research Centre, University of British Columbia. Hall, K .J . , F. Koch, and I. Yesaki. 1974. Further investigations into water quality conditions in the lower Fraser River system. Westwater Research Centre, University of British Columbia, Technical report #4. Hall, K .J . , and J.H. Wiens. 1976. The quality of water in tributaries of the lower Fraser and sources of pollution. In The uncertain future of the lower Fraser. Edited by A . H . J. Dorcey. 49-83. Vancouver: Westwater Research Centre, University of British Columbia. Hall, K.J . , I. Yesaki, and J. Chan. 1976. Trace metals and chlorinated hydrocarbons in the sediments of a metropolitan watershed. Westwater Research Centre, University of British Columbia, Technical Report No. 10. Harper, S. 1993. Personal communication. Environmental Engineering laboratory manager, University of British Columbia. Hem, J.D. 1981. Rates of manganese oxidation in aqueous systems. Geochim. Cosmochim. Acta 45 : 1369-1374. Hey, D.L. , and G.C. Schaefer. 1983. An evaluation of the water quality effects of detention storage and source control. NTIS, PB84-110980. Hotz, M.C .B . 1988. Methycyclopentadienyl Manganese Tricarbonyl (MMT). In Manganese in the Canadian environment. Edited by P. M . Stokes. 177. Ottawa: National Research Council of Canada. Huang, P . M . 1993. An overview of dynamics and biotoxicity of metals in the freshwater environment. Water Poll. Res. J. Canada 28 (1): 1-5. Iman, R.L. , and W.J. Conover. 1983. A modern approach to statistics. John Wiley & Sons. Jackson, B . 1989. Train diesel spill threatens Fraser. Vancouver Sun. May 26, 1989, B6. Johnston, W.A. 1921. Sedimentation of the Fraser River delta. Geological Survey of Canada. 158 Joselow, M . M . , E. Tobias, R. Koehler, S. Coleman, J. Bogden, and D. Gause. 1978. Manganese pollution in the city environment and its relationship to traffic density. Am. J. Public Health 68 (6): 557-560. Koch, F.A. , K.J . Hall , and I. Yesaki. 1977. Toxic substances in the wastewaters from a metropolitan area. Westwater Research Centre, University of British Columbia, Technical Report No. 12. Krajczar, K . 1994. Demographic data used for preparation of 1992 Greater Vancouver Travel Survey, unpublished. Strategic Planning Department, Greater Vancouver Regional District . Krumgalz, B.S. 1989. Unusual grain size effect on trace metals and organic matter in contaminated sediments. Mar. Pollut.Bull. 20 (12) : 608-611. Lafleur, R. 1994. Canadian Petroleum Products Institute (CPPD. Ottawa. Lawson, E . M . , G.B. Mitchell, and D.G. Walton. 1985. Stormwater monitoring in an industrial catchment basin located in Burnaby . British Columbia. Waste Management, Ministry of Environment, Province of British Columbia. Lee, G.F., and A . Jones-Lee. 1993. Water quality impacts of stormwater-associated contaminants: Focus on real problems. Wat. Sci. Tech. 28 (3-5) : 231-240. Loranger, S., J. Zayed, and E. Forget. 1994. Manganese contamination in Montreal in relation with traffic density. Water. Air and Soil Pollut. 74 : 385-396. Marsalek, J. 1986. Report on N A T O workshop on urban runoff quality. Proceedings of an Engineering Foundation conference. In Urban runoff quality. Proceedings of an Engineering Foundation conference. New England College. 15-28. Henniker, New Hampshire: American Society of Civi l Engineers. Marsalek, J., and H . Shroeter. 1988. Annual loadings of toxic contaminants in urban runoff from the Canadian Great Lakes basin. Water Poll. Res. J. Canada 23 (3) : 360-378. Mathewes, R.W., and J .M. D'Auria. 1982. Historic changes in an urban watershed determined by pollen and geochemical analyses of lake sediment. Can. J. Earth Sci. 19 : 2114-2125. Matthews, H . M . 1994. Trace metal loading from urban stormwater runoff in the Brunette River watershed. B . A . S c , University of British Columbia. McNeil l , B . 1978. A study of water quality relationships in the Brunette River basin. M A , University of British Columbia. Moore, J.N., E.J. Brook, and C. Johns. 1989. Grain size partitioning of metals in contaminated, coarse-grained river floodplain sediment: Clark Fork river, Montana, U.S.A. Environ. Qeol, Water Sci, 14 (2) : 107-115. Morton, T.A. 1983. Polycyclic Aromatic Hydrocarbons in Still Creek sediments. M . S c , University of British Columbia. 159 Munteanu, N . 1987. Deer Lake Restoration - 1987. Limnology Services, Prepared for The City of Burnaby. Murray, J.W. 1975. The interaction of metal ions at the manganese dioxide-solution interface. Geochim. Cosmochim. Acta 39 : 505-519. National Research Council of Canada. 1981. Marine sediment reference materials. Marine Analytical Chemistry Standards Program, Ottawa, Canada. Northcote, T.G., and B. Luksun. 1992. Restoration and environmental sustainability of a small British Columbia urban lake. Water Poll. Res. J. Canada 27 (2): 341-364. Novotny, V., H . M . Sung, R. Bannerman, and K . Baum. 1985. Estimating nonpoint source pollution from small urban watersheds. Journal WPCF 57 (4) : 339-348. Olem, H . 1993. Diffuse pollution. Proceedings of the IAWO 1st international conference on diffuse (nonpoint) pollution sources: Sources, prevention, impact, abatement, held in Chicago. Illinois. USA. 19-24 September 1993. Pergamon Press. Pickering, W.F. 1981. Selective chemical extraction of soil components and bound metal species. C R C Crit. Rev. Anal. Chem. Nov. : 233-266. Poon, D. 1989. 1988 Annual Report: Leaded and lead-free gasoline regulations monitoring program. Environment Canada, 88-04. Power, C A . 1994. Gwendoline Lake Watershed - recent history (memorandum). Puckett, K.J . , W.H. Schroeder, P .M. Stokes, B . LaZerte, and C. Trick. 1988. Geochemical Cycle. In Managanese in the Canadian Environment. Edited by P. M . Stokes. 23-64. Ottawa: National Research Council of Canada. Ramlan, M . N . , and M . A . Badri. 1989. Heavy metals in tropical city street dust and roadside soils: A case of Kuala Lumpur, Malaysia. Environ. Technol. Lett.. 10 : 435-444. Rasmussen, P.E. 1994. Current methods of estimating atmospheric mercury fluxes in remote areas. Environ. Sci. Technol. 28 (2) Reiberger, K . 1992. Metal concentrations in bottom sediments from uncontaminated British Columbia lakes. Water Quality Branch, Ministry of Environment, Lands and Parks, Province of British Columbia. Rolfe, G.L. , A . Haney, and K . A . Reinbold. 1977. Environmental contamination by lead and other heavy metals: Volume II Ecosystem Analysis. National Science Foundation Rann Program. Rood, K . , and R. Hamilton. 1994. Hydrology and water use for salmon streams in the Fraser Delta Habitat Management Area. British Columbia. Department of Fisheries and Oceans, Government of Canada, Manuscript Report of Fisheries and Aquatic Sciences, 187 p. Rowan, D.J., and J. Kalff. 1993. Predicting sediment metal concentrations in lakes without point sources. Water. Air . Soil Pollut. 66 : 145-161. 1 6 0 Rudolph, E. 1995. Personal communication. Sapperton Fish and Game Club. Sadiq, M . , I. Alam, A . El-Mubarek, and H . M . Al-Mohdhar. 1989. Preliminary evaluation of metal pollution from wear of auto tires. Bull . Environ. Contam. Toxicol. 42 : 749-753. Salomons, W., and U . Forstner. 1984. Metals in the Hydrocycle. Berlin, Heidelberg: Springer-Verlag. Sawicki, J., and G . Runka. 1986. Land use classification in British Columbia. Prepared for the Ministry of Agriculture and Food, Soils Branch and Ministry of Environment, Survey and Resources Mapping Branch. Victoria B.C. Shaheen, D.G. 1975. Contributions of urban roadway usage to water pollution. U.S. Environmental Protection Agency, EPA 600/2-75-004. . Smith, J. 1994. Sediment bioassays: Battery test evaluation of shallow urban streams and the effect of sampling method on toxicity. M . S c , University of British Columbia. Smith, J.D., and T.F. Hamilton. 1992. Trace metal fluxes to lake sediments in south-eastern Australia. Sci. Tot. Env. 125 : 227-233. Soilcon Laboratories. 1990. Greater Vancouver Regional District Soil and vegetation monitoring program. Prepared for the G V R D . Spyridakis, D.E., and R.S. Barnes. 1976. The effects of waste water diversion on heavy metal levels in the sediments of a large urban lake. Department of Civi l Engineering, University of Washington. NTIS, PB-253 796. Statistics Canada. 1991. 1991 Census. Government of Canada. Statistics Canada. 1993. Motor vehicle registrations. Government of Canada. Striegl, R.G., and E.A. Cowan. 1987. Relations between quality of urban runoff and quality of Lake EUvn at Glen Ellyn. Illinois. U.S. Geological Survey, Water-Supply Paper 2301. Swain, L . G . 1989. Coquitlam-Pitt River area, tributaries to the lower Fraser River along the north shore, water quality assessment and objectives (Technical appendix). B C Environment, Water Management Division. Swain, L . G . , and D.G. Walton. 1988. Report on the 1987 benthos and sediment monitoring program. Water Management Branch, Ministry of Env. and Parks, Province of British Columbia. Vaithiyanathan, P., and A . L . Ramanathan. 1993. Transport and distribution of heavy metals in Cauvery River. Water. Air . Soil Pollut. 71 : 13-28. Valcovic, V . 1978. Trace elements in petroleum. Tulsa, Oklahoma: The Petroleum Publishing Company. 161 Warren, L . A . , and A.P. Zimmerman. 1993. Trace metal-suspended particulate matter associations in a fluvial system: physical and chemical influences. In Particulate matter and aquatic contaminants. Edited by S. S. Rao. 127-155. Lewis Publishers. Westwater Research Centre. 1994. Fraser basin Ecosystem study : Annual Report 1993-1994. The Westwater Research Centre and The Sustainable Development Research Institute, University of British Columbia. Whipple, W. Jr. 1987. Implementing dual-purpose stormwater detention programs. J. Water Res. Plann. Manage. Div..(Am.Soc.Civ.Eng.') 113 : 779-792. Wilber, W.G. , and J.V. Hunter. 1979. The impact of urbanization on the distribution of heavy metals in bottom sediments of the Saddle river. Water Res.Bull. 15 (3): 790-800. Woodward-Clyde Consultants. 1992. Source identification and control report, prepared for the Santa Clara Valley Non-Point Source Pollution Control Program. 162 A P P E N D I X A S T R E A M A N D S T R E E T S E D I M E N T S A M P L I N G L O C A T I O N S Table A - l Stream sediment sampling sites Station Number Station Description General Remarks 1. Brunette R. at Spruce Ave. (bridge) At river mouth, wood products industries. 2. Brunette R. at Camphor Ave., near railway bridge. Wood products industries nearby. 3. Brunette R. at Braid St. (bridge) Wood products industries nearby. 4. Brunette R. at Brunette Rd. Potentially affected by high traffic volumes. 5. Small stream north of Trans-Canada Hwy., west of Hart St. between Roderick and Henderson St. Storm sewer feeds this stream directly upstream of sampling location. 6. Brunette R. at North Rd.(east side) Sampled within Hume Park. 7. Stoney Creek at Grandview Hwy., 100 m west of intersection of Hunter and Keswick Streets. 8. Stoney Creek at Beaverbrook Dr.and Noel Dr., samples obtained upstream and downstream of bridge. Residential area. 9. Stoney Creek at East Broadway, 50 m west of Norcrest Rd. Residential area. 10. Brunette R. at Cariboo Rd., samples obtained upstream and downstream of bridge. Potentially affected by high traffic volumes. 11. Small stream arising from a storm sewer, south side of Winston St., east of Brighton St. Light industrial and residential area. 13. Eagle Creek on Piper Avenue, south of Winston St. Located in Werner Loat Park. 14. Eagle Creek at East Broadway( south side), between Lake City Way and Lawrence Drive. Below golf course, iron hydroxide precipitates evident. 15. Tributary of Eagle Creek at Shellmont St. (north side), east of Arden Ave. Downstream of petroleum tank farm runoff detention facility. 16. Tributary of Eagle Creek at Woodbrook Place, east of Phillips Ave. Upstream of golf course, wooded stream buffer. 17. Robert Burnaby Creek, near park entrance at 4th St. Located within Robert Burnaby Park. 19. Deer Lake Brook at Glencairn Dr. (north side) North of freeway and south of Burnaby Lake. 20. Deer Lake Brook at Deer Lake Ave., south of Canada Way, upstream and downstream of bridge. Downstream of Deer Lake. 21. Small stream at Moscrop St. (south side), between Royal Oak Ave. and Oaktree Ct. Residential area downstream of cemetery. 163 Table A - l continued Station Number Station Description General Remarks 23. Still Creek at Sperling Ave. Street not accessible to traffic at this point. 24. Small creek at intersection of Sperling Ave. and Jordan Dr. Residential area. 25. Beecher Creek near Goring Ave., sampled on south side of railway tracks. Small tributary of Still Creek. 26. Beecher Creek at Lougheed Hwy. (south side) Upstream of station #25. 27. Beecher Creek at Springer Ave. (east side) Upstream of station #26. 28. Beecher Creek at Westlawn Dr. (north side) Upstream of station #27. 29. Small stream in Westburn Park along Gilpin Cr. 400 m upstream of 1973 location. 30. Still Creek on Still Creek Dr., west of Willingdon Ave. Industrial area, heavy traffic. 31. Still Creek at Gilmore Ave. (east side) Industrial area. 32. North branch of Still Creek at Lougheed Hwy. (south side) Affected by site development and heavy traffic. 33. Still Creek at Grandview Hwy. (south side) and Renfrew St. (east side) Residential area. 34. Still Creek at Myrtle St., east of Boundary Rd. Industrial area. 35. Still Creek at Douglas Ave. Industrial area. 37. Still Creek at Atlin St. and 27th Ave. Wooded ravine. 1 6 4 Table A-2 Street sediment sampling sites Station Number Station Description Traffic Volume1 (vehicles/day) Traffic Category 11 Rupert St. (between Grandview and Broadway E.) 19980 high 12 Boundary Rd. and Myrtle Ave. 7020 low 13 Gilmore Ave. north of Still Creek. 11480 high 14 15 Willingdon Ave. north of Trans-Canada Hwy interchange. Douglas Rd. at Still Creek Way. 28350 7020 high low 16 Lake City Way north of Venture St. 12560 high 17 Spruce Ave., near stream station #1. 2970 low 18 Camphor Ave., near stream station #2. 5270 low Cl Canada Way at Boundary Rd. 21740 high C2 Parking lot of Brentwood Mall (Willingdon Rd. at Lougheed Hwy.) 28350 high C3 Parking lot of Lougheed Mall (Lougheed Hwy. at Austin Rd.) 19440 high C4 North Rd. at Lougheed Hwy. 37050 high C5 Canada Way at Sperling Ave.. 76990 high C6 Braid St. at Columbia St. 22140 high Rl E. 14th Ave., 1 block E. of Renfrew St. n/a low R2 E.16th Ave. between Renfrew and Rupert St. n/a low R3 Smith Ave. at Spruce Ave. 7430 high R4 Whitsell Ave. at Williams Ave. n/a low R5 Duthie St. (2400 block). 5670 low R6 Eglinton St. at Gilmon Ave. n/a low R7 Canada Way at May field St. 36590 high R8 Lee St. at 10th Ave. 14450 high Gl Forest Lawn Memorial Cemetary. n/a low G2 Robert Burnaby Park parking lot. n/a low G4 Phillips Ave. at Halifax St. n/a low 1. Traffic estimates modified from Greater Vancouver Regional District 1992b. 165 APPENDIX B LAND COVER TEST AREAS The degree of permeability of low density residential areas and commercial/industrial areas was determined using average permeabilities. These averages were determined by digitizing the land cover from 5 residential areas and 3 commercial/industrial test sites. Each test site was approximately 2 blocks by 2 blocks in area, chosen to reflect variability within the land use category across the watershed. The residential analysis, summarized in Table B - l , utilized 1973 black and white aerial photos (1:2400 scale) for the digitization process. The industrial/commercial test sites were digitized from the 1973 photos as well as from 1992 colour photos (1:12,000 scale) in order to determine if land cover changes had occurred within the industrial/commercial areas over the last 20 years. No obvious temporal trends were evident among the 3 industrial/ commercial sites (Table B-2), therefore the bulk permeability estimate was derived by averaging each of the results. 1 6 6 Table B - l Determination of low density residential area permeability Test Area Watershed Region Location Permeable Cover (percent) total (grass) Impermeable Cover (percent) total buildings pavement 1. western Bounded by Cassiar St, Skeena St., 23rd Ave.,and 25th Ave. 51.6 48.4 22.2 26.2 2. north-central Bounded by Napier St., Charles St., Kensington St., and Sperling St. 55.3 44.7 17.9 26.8 3. eastern Bounded by Astor Dr., North Rd., Sullivan St. and David Dr. 52.7 47.3 16.2 31.1 4. western Bounded by Smith SL, Macdonald St., Kincaid St., and Forest St. 54.5 45.6 20.8 24.8 5. south-central Bounded by 2nd Ave., 55.6 44.3 17.0 27.3 Newcombe St., Wedgewood St., and 19th Ave. Average land cover permeability for low density residential areas (in percent): 50 Table B-2 Determination of industrial/commercial area permeability Test Location Year Permeable Impermeable Area Cover (percent) Cover (percent) total (grass) total buildings pavement 1. Bounded by Boundary Rd., 1973 23.2 76.7 25.6 51.1 Douglas Rd., and 2nd Ave. 1992 15.1 84.9 24.7 60.2 2. Bounded by Lougheed Hwy., 1973 16.9 83.1 15.4 67.7 Railway tracks, Holdom Rd., 1992 16.5 83.5 17.0 66.5 and Douglas Rd. 3. Bounded by Lake City Way, 1973 31.1 68.9 17.8 51.1 Underhill Ave., Broadway Ave., 1992 24.7 75.3 20.5 54.8 and Lougheed Hwy. Average land cover permeability for 20 industrial/commercial areas (in percent): 167 APPENDIX C QUALITY ASSURANCE DATA Table C - l Quality assurance measurements of the Hg method Sample Replicate Analysis ? Hg Concentration (ug/kg dry weight) Ratio of Replicates Calibration curve Standard Addition #10-2 (max.) Y 80 1.05 #10-2 (min.) Y 76 #32-3 (max.) Y 45 1.13 #32-3 (min.) Y 40 #30-2 (max.) Y 72 1.06 #30-2 (min.) Y 68 #27-1 (max.) Y 74 1.07 #27-1 (min.) Y 69 #35-1 (max.) Y 169 1.06 #35-1 (min.) Y 159 #29-1 (max.) Y 77 1.00 #29-1 (min.) Y 77 #37-1 (max.) Y 142 1.08 #37-1 (min.) Y 131 #32-1 (max.) Y 169 1.10 #32-1 (min.) Y 154 13 (max.) Y <30 _ 13 (min.) Y <30 16 (max.) Y 37 1.23 16 (min.) Y 30 #35-2 N 144 199 #31-2 N 145 165 #29-2 N 248 209 #24-1 N 85 89 #5-2 N 109 123 #25-3 N <30 <30 #5-1 N 634 657 #25-2 N 423 476 #25-1 N 617 1008 #33-1 N 2115 2373 BCSS-1(94-01-31) N 149 BCSS-l(94-02-08) N 104 BCSS-l(94-02-12) N 137 BCSS-1(94-02-16) N 103 BCSS-l(94-02-18) N 122 BCSS-l(94-03-03) N 220 BCSS-l(94-03-09) N 178 BCSS-l(94-03-10) N 206 1 6 8 Table C-2 Quality assurance measurements of the nitric acid/flame A A method (all concentrations in mg/kg dry weight) Replicate Sample Analysis ? Fe Cu Cr Pb Ni Zn Mn Mg #14-3 Y 31962 49 18 28 7 105 689 2882 #14-3R Y 26533 27 16 45 <6 123 2132 554 Ratio 1.2 1.8 1.1 1.6 1.2 3.1 5.2 #16-3 Y 13716 40 19 < 15 11 115 514 3741 #16-3R Y 10961 20 <15 10 3554 Ratio 1.3 1.0 1.1 1.1 13 Y 16045 110 27 90 23 200 270 5098 I3R Y 16929 152 31 48 44 324 271 3995 Ratio 0.9 1.4 1.2 1.9 1.9 1.6 1.0 1.3 16 Y 19731 126 58 222 26 321 286 2817 I6R Y 21402 130 52 213 29 280 275 4335 Ratio 1.1 1.0 1.1 1.0 1.1 1.1 1.0 1.5 R6 Y 13594 40 27 91 24 133 252 3323 R6R Y 15419 75 32 84 52 224 276 3327 Ratio 1.1 1.9 1.2 1.1 2.1 1.7 1.1 1.0 DL2-2 Y 30838 130 41 177 37 225 500 7897 DL2-2R Y 32821 137 33 211 58 230 524 8328 Ratio 1.1 1.1 1.2 1.2 1.6 1.0 1.0 1.1 GL1-01 Y 140551 315 <3 247 55 138 2638 906 GL1-01R Y 123060 194 <3 246 69 224 2284 1099 Ratio 1.1 1.6 1.0 1.3 1.6 1.2 1.2 GL1-12 Y 127574 76 10 < 15 22 94 2379 397 GL1-12R Y 115264 82 12 < 15 28 97 2372 403 Ratio 1.1 1.1 1.2 1.3 1.0 1.0 1.0 GL2-01 Y 148504 73 8 152 25 109 2405 572 GL2-01R Y 143214 80 12 154 30 129 2528 710 Ratio 1.0 1.1 1.5 1.0 1.2 1.2 1.1 1.2 GL2-07 Y 126707 60 8 46 23 77 2228 338 GL2-07R Y 128875 62 9 37 22 81 2294 380 Ratio 1.0 1.0 1.0 1.2 1.0 1.1 1.0 1.1 BCSS-la N 24485 20 60 17 58 109 190 13792 BCSS-lb N 21349 25 60 17 60 110 170 12171 Note: all Cd determinations below detection limit (3 mg/kg). 1 6 9 Table C-3 Measurement of the precision of the nitric acid/ICP-AES method (all concentrations in mg/kg dry weight) Sample Fe Cu Cr Pb Ni Zn Mn Mg Cd B1-124C 16179 44 18 271 21 323 213 4391 1.4 B1-124C-2 16136 46 18 265 21 324 214 4445 1.4 Ratio 1.00 1.04 1.02 1.02 1.01 1.00 1.00 1.01 1.02 B1-44C 13123 86 17 235 20 283 237 3134 1.2 B1-44C-2 13459 83 18 242 21 300 243 3173 1.0 Ratio 1.03 1.03 1.07 1.03 1.01 1.06 1.03 1.01 1.16 B1-4C-1 15553 128 20 338 21 331 216 3995 0.9 B1-4C-2 14515 122 19 307 20 306 204 3723 0.9 Ratio 1.07 1.05 1.07 1.10 1.04 1.08 1.06 1.07 1.02 B1-85C 15585 51 21 242 17 304 220 3954 1.1 B1-85C-2 15775 49 21 240 16 306 222 3920 1.3 Ratio 1.01 1.03 1.01 1.01 1.04 0.99 1.01 1.01 1.16 B2-11C-1 11094 303 23 378 32 301 242 3433 1.4 B2-11C-2 10472 275 24 336 30 276 231 3222 1.4 Ratio 1.06 1.10 1.02 1.12 1.04 1.09 1.05 1.07 1.05 B2-25C 13314 106 25 234 22 259 306 4276 2.0 B2-25C-1 13386 105 26 225 23 264 309 4264 1.7 Ratio 1.01 1.01 1.02 1.04 1.03 1.02 1.01 1.00 1.21 B2-45C 10810 31 19 5 15 130 264 3433 0.4 B2-45C-1 10052 28 17 7 14 123 245 3178 2.6 Ratio 1.08 1.12 1.13 1.46 1.04 1.06 1.08 1.08 6.37 B2-420C 11544 <1 15 <3 16 45 162 4226 <.2 B2-420C-2 10736 <1 14 <3 15 43 151 3904 <.2 Ratio 1.08 1.06 1.06 1.04 1.07 1.08 B2-460C 13289 <1 20 <3 20 53 193 5302 0.5 B2-460C-2 11288 2 17 <3 18 47 164 4525 0.4 Ratio 1.18 1.16 1.16 1.12 1.18 1.17 1.26 B2-500C 12299 13 18 <3 20 50 170 4894 0.3 B2-500C-2 13213 4 19 <3 21 53 182 5262 0.3 Ratio 1.07 3.39 1.06 1.04 1.04 1.07 1.08 1.06 B2-540C 12798 <1 18 <3 18 50 181 5608 0.2 B2-540C-2 13957 <1 21 <3 22 56 197 6108 0.3 Ratio 1.09 1.14 1.18 1.12 1.09 1.09 1.48 DL-94-1 19357 <1 20 <3 20 66 416 5538 <.2 DL-94-2 15892 <1 15 <3 16 51 340 4575 <.2 Ratio 1.22 1.29 1.26 1.30 1.22 1.21 1 7 0 Table C-3 (continued) Sample Fe Cu Or Pb Ni Zn M n Mg Cd B2-9C 10755 155 22 260 23 246 243 3171 1.2 B2-9C-1 10825 158 21 272 25 253 242 3248 1.2 B2-9C-2 10362 155 20 259 22 243 234 3078 1.5 CV (%) 2.3 1.1 5.5 2.7 5.4 2.0 2.0 2.7 15.1 B2-13C 6838 131 15 123 17 133 147 2299 0.9 B2-13C-1 7231 144 15 143 18 132 158 2583 1.1 B2-13C-2 7267 140 14 130 18 136 156 2400 0.6 B2-13C-3 7291 136 13 137 18 133 159 2426 0.7 CV (%) 3.0 4.1 7.7 6.5 3.4 1.5 3.7 4.8 26.9 B2-17C 6927 64 12 89 12 116 144 2364 0.7 B2-17C-1 6616 52 13 75 13 100 142 2450 0.8 B2-17C-2 5958 45 9 67 12 92 126 2114 1.2 B2-17C-3 5955 47 9 70 10 92 125 2142 0.5 CV (%) 7.7 16.1 20.0 12.9 8.9 11.3 7.4 7.3 40.2 B2-19C 9270 120 19 212 18 196 214 2915 1.0 B2-19C-1 8072 106 16 181 18 162 185 2648 1.1 B2-19C-2 9875 129 21 239 21 209 231 3077 1.1 CV (%) 10.1 9.8 12.3 13.9 10.1 12.7 11.1 7.5 4.7 B2-21C 15564 193 34 431 32 335 349 4827 2.1 B2-21C-1 13326 171 30 384 30 303 300 4298 2.4 B2-21C-2 15942 197 34 475 35 349 362 5096 2.0 CV (%) 9.5 7.5 7.3 10.6 7.8 7.3 9.7 8.6 10.7 171 APPENDIX D COMPARISON OF ANALYTICAL METHODS Digestion Comparisons Perchloric acid (HC10 4) was not used throughout this study because of the risk to laboratory personnel from its explosive nature. A set of preliminary samples were digested and analyzed at Chemex Labs in North Vancouver, though, using both a nitric/perchloric and an aqua regia acid digestion. Each digest was also analyzed using both flame A A and ICP-AES detection techniques. Results from the nitric/perchloric digests run on the ICP-AES were not used for this comparison because perchloric acid was found to suppress metal concentration values relative to the aqua regia method. The suppression of ICP-AES results by perchloric and hydrofluoric acid is common and considerable effort to correct for matrix effects is necessary (Bichler 1994). The results from the Chemex comparison (Table D - l ) show very little difference between digestion methods for Pb, Cu, Zn, Ni , and Cd. Significantly more Mn is solubilized using the more reactive perchloric acid digestion. Another comparative study was performed in the U B C Environmental Engineering laboratory (flame A A ) to determine the differences between nitric acid and aqua regia digestions (Nitric acid is preferred to aqua regia in this laboratory because of the corrosive qualities of hydrochloric acid (Harper 1993)). Concentrations of Mg, Mn, Pb, Cu, Zn, Ni , and Cr differed by less than 10 % between methods (within the limits of method precision) while significantly less Fe was solubilized using the nitric acid digestion method (Table D-2). These results indicate that only minor differences (< 10 %) in metal concentrations between nitric acid and nitric/perchloric acid digestions are expected for Pb, Cu, Zn, and N i . Digestions using nitric acid will significantly under-predict Fe and Mn (< 20%) concentrations relative to the nitric/perchloric acid digest. The comparison was inconclusive for M g and Cd. 1 7 2 Table D-1 Median concentration differences and significance of ranked t-test for comparison of aqua regia versus nitric/perchloric acid digestions (n Element Median percent difference, Aqua Regia minus Nitric/ Perchloric p Value* Mn -17 0.00 Pb 1 0.49 Cu 0 0.48 Zn -6 0.00 N i -3 0.03 Cd -4 0.26 p values indicate the probability that the means are identical Table D-2 Median concentration differences and significance of ranked t-test for comparison of aqua regia versus nitric acid digestions (n=14). Element Median percent difference, Nitric Acid digestion minus Aqua Regia digestion p Value* Fe -15 0.00 M g 8 0.02 M n -3 0.06 Pb 8 0.01 Cu 5 0.18 Zn 2 0.02 N i -6 0.09 Cd below detection limit 'p values indicate the probability that the means are identical 173 Detection Techniques A small comparative study examining the flame A A (Civil Engineering, UBC) and the ICP-AES (Soils Sciences, UBC) detection techniques was performed using nitric acid digests in order to quantify the differences between methods. The results of this analysis (Table D-3) indicate no significant difference in methods for M n and Zn determinations, while Pb, Cu, Cr, and N i were significantly lower using the ICP-AES detection technique. Determinations of Pb, Cu, Cr, and N i were strongly correlated between methods as is evident in Figure D - l . These plots suggest that the ICP-AES method is unreliable for Pb and Cu below sediment concentrations of approximately 20 mg/kg. Table D-3 Median concentration differences and significance of ranked t-test for comparison of flame AA versus ICP-AES detection techniques (n=12). Element Range of Median percent Comparison difference, ICP minus (mg / kg flame AA) flame AA p Value* M n 159-2238 -4 0.12 Pb 24-211 -55 0.00 Cu 21 - 128 -59 0.00 Zn 84-440 0 0.48 N i 9 - 2 6 -19 0.00 Cr 14-30 -43 0.00 *p values indicate the probability that the means are identical, all concentrations expressed on a dry weight basis 1 7 4 Figure D-1 Linear relationships between ICP-AES and flame A A detection techniques (all concentrations in mg/kg dry weight) Note: 1. A l l determinations made on nitric acid digestions of stream sediment samples, 2. flame A A analysis performed in Civil Engineering laboratory, U B C , 3. ICP-AES analysis performed in Soils Sciences laboratory, U B C . 175 Grain Size Effects The <180 um sediment fraction was chosen to perform trace metal analyses on the stream and street sediments in order to provide a direct comparison with data collected by the Westwater Research Centre in 1973 (Hall et al. 1976). The sieving procedure differed slightly between studies, however, as the 1973 study utilized a dry technique and the current study used distilled water to enhance the process. Colman and Sanzolone (1992) examined the differences in metal concentrations caused by dry versus wet sieving and found that many elements were higher using the wet method (Table D-4). The magnitude of difference is small however in comparison to both the within and across-site variability measured in this study. During the past 20 years many river and lake sediment studies have utilized the <63 um fraction for trace metal analysis (Forstner 1990). In order to compare the results of this study to others using the <63 um sediment size fraction, additional analyses on a small subset (14) of the total samples in this study were performed. For each of these 14 samples, metal concentrations in both the <63 um and <180 um fractions and the amount of silt and clay (<63 um) present in the <180 um fraction were determined. Linear relationships between the ratio of metal concentrations in the two size fractions (<180 : <63) and the relative amount of silt and clay in the <180 um fraction is presented in Figure D-2. Using the strong, linear correlations for Pb, Cu, Zn, and M n , the <63 um metal concentrations can be estimated for each stream and street sediment station. 176 Table D-4 Median concentration differences and significance of ranked t-test for selected element comparison of wet and dry-sieving methods (n=21). Element Median Percent Difference, Wet Sieving Minus Dry Sieving p Value* Carbon, organic 2 0.00 Cr 17 0.00 Cu 12 0.00 Fe 0 0.09 Pb 3 0.33 Hg -10 0.01 N i 5 0.01 Zn 13 0.00 *p values indicate the probability that the means are identical. Source: Colman and Sanzolone 1992 177 Figure D-2 Linear relationships between the ratio of metal concentrations in 2 sediment fractions (<180 [im fraction / < 63pim fraction) and the percentage of silt and clay in the <180p:m fraction. 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Silt&clay(%) Silt&clay(%) 1.3 y = 0.79 + 1.9e-3x RA2 = 0.07 1.1 H co CC 0.3 • • • • • • • EB • • • • • 10 20 30 40 50 60 70 80 Silt&clay(%) 1.3 y = 0.74 + 3.3e-3x RA2 = 0.06 1 .H n CC o 0.5 H 0.3 • • • • • • • 10 20 30 40 50 60 70 80 Silt&clay(%) 178 Table D-5 Comparison of aqua regia and nitric/perchloric acid digestions, (all concentrations in mg/kg dry weight) Pb Cu Zn N i Cd M n Sample AR NP AR NP AR NP AR NP AR NP AR NP G l 340 365 235 260 320 355 27 27 3.3 3.7 390 490 G2 177 200 172 220 405 490 24 26 4.1 5.4 510 690 G4 197 200 226 230 500 555 29 30 4.1 4.3 2100 2200 G5 288 280 260 250 680 680 29 30 6.7 6.0 2100 2000 G1S 600 600 320 320 580 640 26 25 10.6 11.0 610 740 Dl 157 155 200 210 453 472 30 31 2.4 2.5 430 580 D2 180 180 227 226 510 540 31 31 2.7 2.7 520 630 D3 188 180 240 240 540 550 29 29 2.3 2.5 490 580 D4 192 190 214 218 510 540 30 31 2.1 2.1 450 560 D5 158 154 167 167 405 385 25 28 1.9 1.8 410 510 D1S 162 156 242 236 520 532 34 34 3.0 2.9 550 660 D2S 153 148 208 206 460 490 32 31 2.6 2.5 550 660 D3S 213 200 280 260 620 650 34 37 3.3 3.1 610 620 D4S 200 205 250 250 560 620 34 35 2.6 2.7 550 610 D5S 186 184 210 216 460 505 33 35 2.5 2.6 540 620 Notes and abbreviations: (i) AR= aqua regia digestion, NP=nitric perchloric acid digestion; (ii) analyses performed at Chemex Labs using flame AA detection technique. Table D-6 Comparison of aqua regia and nitric acid digestions, (all concentrations in mg/kg dry weight) Pb Cu Zn Ni M n Mg Fe Sample N AR N AR N AR N AR N AR N | AR N AR 101-3 85 55 32 63 140 137 21 21 320 329 4847 4337 27481 30910 102-2 30 20 32 26 75 73 34 38 260 280 6002 6499 20508 25995 105-3 55 50 32 35 112 114 15 16 370 395 3602 3298 25013 29479 110-3 879 817 250 235 547 533 31 30 1858 1814 4845 4585 37463 44358 121-1 105 95 30 27 94 90 15 14 225 239 3000 2743 10999 17953 126-3 90 85 49 50 172 168 18 17 999 1028 3597 3294 18985 23954 130-2 80 75 81 71 173 161 14 16 210 210 3044 3048 15469 17991 132-1 115 105 70 66 163 162 17 19 199 205 2491 2298 13951 17483 16 215 225 133 112 353 334 23 27 289 309 4340 4093 24945 29946 R6 100 90 33 43 138 146 16 16 255 240 3601 3050 16003 18498 BCSS1 <15 <15 13 13 106 105 47 61 164 180 9980 9987 24542 24558 DL1-4 175 165 87 83 227 222 26 28 495 509 6496 5992 30981 33952 DL2-2 169 164 90 85 230 223 26 26 494 503 6481 5979 31904 34380 GL1-13 20 20 70 67 100 96 18 16 2095 2216 249 354 114736 128244 Notes and abbreviations: (i) AR= aqua regia digestion, N= nitric acid digestion; (ii) analyses performed in Civil Engineering laboratory using flame AA detection technique; (iii) all Cd values below detection limit (3 mg/kg). 179 *5 bo e 8 'So § llz § ° c bo o E -—- o c 8 ,—* s bo O |S<* ° IN 2> c bo E 8 to o lio a J2* 8 p ^ 1I-J ^ 4 5 a £ a. E 1=2 t N t N C S m c S V O c O t n T t N N t N f O c n t — v o < N O \ o o o N t - ~ N O c o v © r - t N o o m O O O Q O O O Q « n > n > 0 > 0 0 0 Ol I - H »-H I - H o © o o N (S - t VO VO >0 I— § Q >n m >/•> o >n © m «/•» to v i I - H T t r~ to >/"> m tN r-» to —i tN I - I <N — rt.-HON'—ioor~vot~r~to<N(Na\vo m m M t n c N t N N m t N r H f H p H r t i H k t i - H t — > / - > i / " > > ^ © T t c o © 0 1 0 ) \ 0 ' - H loi o i tN oi i-< co >n r~ o i o ©' O >-H © a w H f . i q q m o o i n q q v i i ' i q oi oi co oi o i r-i r~ r~ cs >-» oi •-< M O O O m « 0 0 « » O O m M ^ T t i n i n i n t o t O T j - o o o i ' - i i - i i — • —t —• 0 ) 0 0 0 < O O O Q v O v O v O v O T t - 0 -f o o M n c N O ' t ^ « c r » | o o \ o o o \ ._ • » O r ~ o o m r H M t N M - H _ - . . . O r f v o >C m rt r H H H t O M CN r-< CN CN — i O O O N O O f - O O N O C N m CN CN CN r < r t r t r t N ' t i H i o S M O Q o m </-> NO in oo N C S M - H C S C N C i - H v O ' O O O v O O O O V O r - i O O v O N C O C 0 0 * 0 > 0 ' - H O ) C O ' H » C \ 0 ' > * C N O N C O CN OJ OI CN OI CO CO tO »-H i-l P H I - H w oo I -< - H > ce N N ^ co I-< m oq oi oi to co tN rf co T J - oi to o i oi to o o v o r ^ o O T f v - > o « ^ o o w ^ O N O r j ; o o oi o i co co co T t T t o i T f oi co co CN I - H CO O N O N VO ON oi P H O 0 0 iri | p oo >/"> to - H p CN - H O CO o i I - H CO T t (NcocNcooovocNTtr~-cNr— i—io>n r - r - v o t — v o c o c o t - c o T t i - i c o T t T t i - H C N c O T t i n i - H t N c o ^ i r i o o v c o o i i i i l I i , T T O p H C N C N ( N •s 1 Vi u bO '•3 £> Xi o Is ; o i x "S Sold C fe O N — 180 APPENDIX E LAKE SEDIMENT CORE DATA 181 I s «G *•—' I <U <U Q o CA a o •CJ at t l C CU u c o u -a •4-* CD s <D U CC] t l 1-. CD t! s u •f-H c W u 3 H CN bo M r * CN bo - >^  Z b£ bO bO CN bo bO b bo 6 * 00 00 CO ON cn ON o CO CN f-H cs cs 00 cs CS f-H CO m CO CS cs cs cs T t r— _ f-H _ f-H cn CO CO CO CO CO NO m T t CO NO cs 00 cn co CO CO CO CO CN V V V V V V V ON © 00 cs 00 en ON r~ r~ VO T t i n T t T t T t T t i n ,_, CO ON cs CO vo r— 00 t~ T t T t T t T t T t T t T t o r~ ^ NO cs o \ ON t— fH m r~ cn 00 co 00 m Q r-NO T t vo NO \o T-H 00 ON CS t— OO T t ON T t CO i n T t pH CO T t r- f-H T t i n r~ VO t~ NO NO >o T t 00 r- T t NO CO m CO ON l> r-CN 00 vo VO vo NO r~ o 00 NO ON CN CO CN cs CN cs 6 * bO bO bO 6.* £ b 0 _ <— f H es s & e 5-8^ CS 00 ON CS — i Tt ON CO CS TJ f-H Tt f-H oo - H i n co oo © r— ON ON o ON ON CN O CN CS CO CN CS CO CO co >n oo NO m NO f-H cs Q p p oo CS CS CN CS CS f-H co m oo r> co NO oo t-H O O O l-H f-H 00 CS CN CS CN CN CS I-H t— I-H i n T t r- co CO T t co co co co vo co NO >n oo vo T t co co co co cs co co oo r** NO vo co oo r» r- r- r- oo so 00 ON f-H CS ON O 00 vo m vo oo vo oo vo s m ON ON ON ON CN ON co vq vo p T t NO CO I-*' O f-H l-i ON |cs oo >n vo r~- ON vo PH' PH' o o I-H* o o PH co >n r- ON PH PH 8 u o Vi CD CD _G "o T3 C CD o o Vi c p ed t l G CD CJ G O u "c3 •4-1 CD 6 CD CJ ed t l •a 9-IH CD -*-» a e CJ • r H G a bfj Ui o CN i w cd H cs b O * K b p „ _ CN b O * z f PH e, * bO E 4*: bO es cs ^ * bo S ^ P -A * T t 00 NO NO CS 00 CS o CS cs CN CS PH m ,_, 00 NO cs f-H CS cs i n 00 cs o cs PH f-H es CN m cs f-H ON NO i n co cs PH CN V V o T t V PH T t T t T t >n ON m 00 00 o o CO CN vo T t o CO m ON cs CN CN cs cs 00 cs CO ^ o CO NO o i n •n NO CO T t o T t CN cs cs co CS T t 00 >n CO VO m CO CO CO CO VO o 0 \ ON 00 Q 00 NO i n 00 ON T t CO CO co bO bO ^ ^ P H e.S es ^ * bO £ ' | P -A * es bO t~ f-H es -1 s co m CS ON NO t-~ 00 3 r~ ON PH ON o 8 CO o ON NO CO CO m ON r~ O0 NO CS es CN —< i n 00 l-H i n VO NO vo CO PH T t es PH O cs T t CS CO es o es cs 5 g-E m PH es r~ o CS o\ PH oo PH as i n CS PH e s e s e s c s C S C S C S P H C S P H C S C S CO CO CO T t T t ON ON v v v vo so vo i n T t o q o ^ c o c o c o c o c o p r ^ p H C N c o CO CO CO CO CO T t NO r- O co o oo ON H ON r~ m ON O c~ Tt Tt T t T t co cs r- oo PH r~ cs T t oo —i co O 00 t~ PH ON i m r - T t r~ r~ Tt vo v p c o c s o Q e s N O O N e S i > - O N r ~ c o o T t c o c o c s p H c o c o c o e s c n c o c o c o iH p O O NO T t CO es r~ 00 T t r-•n NO r~ oo oo o i n r-H m CS vo r~ cs i n co >n vo PH es oo ~ CO r~ i n co o CO •n <n r~ ON vo CO o r~ PH r-~ r-T t m CO PH CN PH cs cs T t co co es PH PH PH PH pH f-H f-H l-H PH PH rt PH PH ON r » oo co vo o r - CS Os o vo o o l-H vo i n r~ r~ r~ r~ 00 oo 00 i n PH co ON o o T t 00 NO T t OS CO o r » vo vo oo oo 00 i n 00 ON 00 co T t r- ON oo NO NO OS O CO T t t"» t~- PH ON T> T t t ~ ON O NO cs os r-oo oo r» i> O O O O V O O O O N O O O O O O N O O N PH CS PH PH PH o o r - T t m N o r - N O O m — ' O O O O O O N O O O c s Q O N N O o o r — N O N O T t f - H i n m i n v O T t m T t c o T t c o e s e S p H p H PH v V r ~ c O T t r ~ o o o c N O m m r ^ m i n i n i - H i n m o l^ CN^ COTtTtCOCSP-PHpHl-HPHpHCSPHpHCS c o e s o Q o o i n Q v o v o o o o o o o r—vovovoininvovoinTtt~in i n N O Q i n p H r - - i n r ~ O N r - - c N v o i n i n o o T t o o o N p H r ^ » O N O N O i n i n m i n T t N O r ^ r ~ - m r ~ - r - N O i n co e s N O e s o o p p c o c s c o O p H T t T t C O T t C O T t T t T t T t T t l n T t C O i n c ^ O N i n c o o e S p H O e s o o i n T t e s r - - o r - - c o T t C O C O C O C O T t T t T t T t T t C O T t T t C O T t T t C O C O in in v i in i n v i in p< c n in r » » CM CMV N V ' N V " p H c o i n r - O N p H p H p H p H p H e s e s c o c o T t T t i n i n 182 CO s CN 4> * W> 2 = L o S3 C3 O p—) ^ >* t l X i c e u pq 8 O 00 I f u " SS o T - J 13 •*-» u e •#-» c 6 00 cn i W T-H X ) C3 H c a (U ^ X <D c a C o •1-H +-» £3 t l C ii U a o o t T3 Ir y i i - H r r i v o c o c s t ~ c S T t T t O N c o v o r ~ T - H o v o v o T t v o i n o o T t o N Q T - I C S T - H C S C S C S T - H I - H T - H C S C O C S C O C O C O C O C S C S C S C S C S C S C S C S C S C S C S C N 3 CS ON m r~- ON •H r— ON C O T t ON t C O -3 - T t CS t-i CUD 0 0 o ON T-H 0 0 ON o 3 r- T t o C S r- oo cs ON C O 0 0 T-H o r» m NO , - H cs C O r- r- r» ON cs ON C O T t i n ON co ON cs r~ o T t ON - H T-H 0 0 cs cs C O i-H vp C O co C O T t o T-H r~- ON ON T-H ro > cs ON © 0 0 0 0 ON cs m T t co r- ON C O T t r- T-H 0 0 0 0 cs 0 0 © m co r- T-H r~ T-H m ON m O ON cs o o T"H NO r~ NO ON m l-H CO T f r- T t O o cs ON ON o o C S — ' ON ON ON ON oo vo r- r- VO r—1 T-H T-H T-H » - H T-H T-H T-H T-H T-H T-H t-H T-H T-H T-H t-H 4> a O # X 5 9 a. 5 CU o u T - H T - H T — I T - H T - H T — I T - H I - H T - H T - H T - H T-H T-H 1—| T-H C S C S C O T - l T - H T - H f O C S C S C S C S C S T - H T - l T - H T - H T - H T - H T - H T - H T - H T - l , _ , , - H T-H T-H T-H T-H ON VO OO T-H VO oo r-; co ON VO p vq r-; r~; >n ON T t T t cs ^ t l ' ^ t ' l ' ' . 0 . ^ ^ P P ' l l ' * . 1 O O O T - H ' O O O O C J O T ^ ' O O O O O O O N r ~ - O N ^ c s T f O N t N t ~ ~ p t ^ T r p T r T - H C S O N O N o q O O O t - H T - H T - < O O O t - < c N T - < c S T - H C S ' 0 0 © C S T t © i n T t 0 0 T t T - H T t C S i n 0 N © T t v O T t C S C O C O V O C 0 0 0 O i n o o v o o o c o T t c s v o r - - c o r ~ i - H © i n O c o T - H © r ~ O N O r - - o o o c s T - H T — I T - H T - H T — I I - H T — I T - H T - H T - H T — I T - H T - H T — I T - H T - H T - H T - H T-H T—IT—I T t m © 0 N 0 N t ~ T t 0 N c s r -c^Ttr^vococoTtTtcocs ^3 VO Ti- VO oo r~- cs ro co ON >n T t 0 0 r- NO o ON cs cs T-H cs rt co cs T-H C O ON 0 0 r- C O ON co m T-H T t C S C O C O co co cs co i n 0 0 >n m cs m r-cs >n r » r--o 0 0 m vo >n 1-H NO Tt O NO ON T t CS m vo ON r- T t s VO T t cs T-H cs >n r- r~ NO 92 cs cs r~ vo ON 0 0 VO cs ON CS T t CO T t T t 85 r » CO VO ft NO O >n ON i n >n T t vo i n T-H ON CO CS ON 8 CO VO VO co NO m cs cs ? N T t >n 0 Q oo 0 0 CS m NO cs ON i n T t r- CS T t ON o m 0 0 ON T t 0 0 rt T-H T-H T-H C S , _ o 0 0 T t NO CO CO CO m o T t ON >n T-H C S s ON CO o VO i n m 0 0 r- C S T t oo T-H T-H T t ON 0 0 cs ON T t T t CO CO cs CO CO CO co T t CO cs cs cs ON C S s o r- co C S ^ ON m 0 0 i n m CO T-H co m T f 0 0 o ON vo ON i n 0 0 ON m CO o r~ T-H CO v© CO ON ON NO T t >n T t T t CO T t T t CO CO T t i n CS t—' cs ON O m cs cs i n ON m co cs T t cs ON oo o o o TJ- C S T t T-H oo r— T t t ~ oo T-H c o o o N v o r - v o N V O T t v o c s m m o r ^ T — I C S C S T - H T - H T — I T - H T ^ 0 0 ^ v 0 » N ' H C M ( » l ' H r H r ^ x O \ 0 \ ( > ^ V 0 ( S 0 0 X ' t M i H t ^ N ^ N M t ^ t ~ 0 \ O N 0 ( S t ^ i n o N i - n o c o o O ' - i c o T t c s v o T - i o o o N O O c s o o o N T t i n v o c O T t T t T t c s c s c s c S ' - i C S T - H T - H C S T - H C S C S - - H » - l T - H T - H T - H ^ H T - H r t T-H C S C S C S C S C S C O t - H T - H T - H C O C S C S C S C O C O T - H C S t - l - H T - H T - l T - H T - H T-H T-H ^ c « / ^ T H ^ ( f l l f l ^ N c ^ r ^ m l o ^ o o o ^ ^ ^ v O v 0 ^ c ^ o c ^ v ^ M l n ^ c c J - H ( ( l ( ^ l ' > t ^ H H H N T N N H ^ H ^ H C O C S C S C S C O C O T - H T - H T - H I — I T - H T - H T - H C S T - H T - H T - H T - H ^ H ^ H R T |r~ooopcooNVOTtoe>ooTtt~o-HTtooinOTtTtONr--cscocs T-H t ~csocscocsoinocOT - H »-Hooc- -Tt incscoT-Hcs » - H - H ( S ( N ( N - H M « H H W H ooinTtoovot-~cscooOT-Hcovoco>nr--'Ttinr»-ininTtcs v v v V V v V V V T-HCSCSeSCScOt-t N t t N N - H H H | N 3 I H m tn - H T H c ~ c o v o r ~ - v o o o v o o o r ^ O N V O f H O O c s i n r - o r - v o o c o T t T t T - H c s oo o cs oo oo T-H r - r - O N V o c o c o c s c S t - H ^ H , _ , cs T-H - H —. - H I S ^ ! Q S ^ C S 7 ! ^ Q ^ r t f i ^ » ^ w 9 N ^ * N ' H N r i N ' H ( T i T f C N i c ~ t H c o o o ( < i |oo^HOONO<nom V N O c S O N i n o v o t ^ v o c O T t c o c O T t r o c o c o c o c S cscs T-H cs ^  ^ H T-H T-H T-H f O T-H T-H T-H T-H TTH 0 0 T t m p i n T t t-H CO CO ON o 0 0 0 0 m 0 0 o i n 0 0 C-; T t o ON ON cs ON r-; NO VO O i n ON ON 0 0 T t vd T t T t T t 0 0 vd CO 0 0 i n ON ON T t ON o i n TH T-H CS -H cs T-H T-H T—I CO T-H cs cs T-H T-H cs CO cs cs CO CO cs CO VO CO o o o c o v o o o i n c o i n o > T t Q T t T - i c s r - - v o c s i n i n v o c s c o c o o o c o c s t - H T t t ~ T - H r ^ m ^ ^ t t « l n T r { S N i n V O V O t ^ V O l ^ V D ^ O N O I ^ V D N O I ^ t ^ V 0 ^ t ^ 0 0 t ^ t ^ 0 0 M ( » 0 O 0 0 0 0 T — i c o m r - O N t — i c o m r - O N T — i r o m r - O N i — i c o m t ~ - O N T - H c o i n r ~ O N T — i c o m r - O N T - H c o m r - O N ^ ^ r t T H t - H C S C S C s c s c S c o c o c o c o c O T t T t T ^ T t T t i n i n i n i n i n v o v o v o v o v o 183 c n E "§> ca OS ca C o •a CS b c u o C O u u E ca a -a <D ca ca 8 OH X <U ca C o •a a tl a Hi o c o u T3 <U C •a a 0 cn 1 W u I* •a \o & X 1.8 •a cu bl o U r— m o o cs T t O O ON CO T t VO O «-H cs T t m i-H 00 CO cs i-H cs o o IH •a in O NO o <0.2 <0.2 cs ? cs ? cs ? cs o V cs o cs o V CO © T t © T t d CO T t d d cs d V in d CO d cs m d d <u co cs cs <n >n T-H i - H m T-H >n T-H T—H r-T-H 00 T-H T-H T-H cs 00 T-H ON T-H 00 cs co CO co o T t in CO s >, 00 00 o T t S o cs CO cs cs CO T t 1 - H T t cs NO r-T t r~ VO 1 -H NO 00 CO T-H CS T t © <N NO CO ON cs 00 T-H VO 00 ON 1 -H co cs CO m 00 vo 00 ON r-VO T-H m m T-H CS ON 1 -H T-H r-o ON T-H o ON T-H m CO <N cs NO T-H in m o CO ON CO $ CO o T-H T-H ON T-H o T-H CO T t VO ON CO CO T-H T-H IH •o T-H o T t l - H cs 00 ON in oo i - H r-VO vo CO cs T t 8 co T-H rt m T-H in <n T t 85 VO T-H 00 3 oo o T-H cs t— VO ON T H CO 00 ON 1 -H 00 -H CS 00 CO 00 CS T t cs T t s CO m cs vo cs in 00 CO 8 © in vo 4-1 m co oo co cs in 00 8 CO T t 00 vo r-m m co o t~ ON VO 00 VO o NO o r~ ON VO cs 00 VO 00 ON Sr% cs vo • - H CS oo m CO cs m vo r-cs 00 m m cs §1 CO CO IH •o ON m o in CO m CO CO r» oo T f CO VO 00 cs cs cs r-cs >n o ON VO T t ON T t r~ m 1 -H m in co T t VO T-H T-H oo m O r~ r-vo r~ cs m ON o 00 cs S cs 00 T t in T-H ON 00 cs CO CO T-H CS CO oo m ON CO cs m «-» 1) 5* e 00 m m co m CO T t T t in cs co T t VO vo CO T-H ON T-H T-H o T-H CS CO I H 1 - H •S ON T t in ON CO i-i CO T t m cs in CO cs CO r-T t T-H cs Si CO T-H T t t— oo CO VO in T t CO in CO m O 00 m in a> CO T t T t T t m IH •a o 00 CS T t 00 oo cs T-H ON T-H T-H o o ON O ON m o cs ON oo o T-H CS Pb dry wet cs Pb dry wet V V V V 00 T-H V V V V V cs cs T t — I CS - H V O 1-H T-H ^ H S v v v CO T-H T-H CS H - 4 (Nl T-H 1 - H T-H V V V V V cs T t - 00 T t in T t T t T t vo ON CO in —^  oo vo vo vo T t co i—^  co vo in I-H T t vo cs ov oo co i/^ vo ci T t T t in Tt 'csoNcocs'csoNod^vd«nTtTtdincs ' i-HONco m t O N t O M W N N ' t ^ ^ ^ t t i n ^ ^ ^ M l t ^ m C i f ' i n t ' l C N l ' H T H T H T H T H p H - H T H T H H > n i n v o v o v o c ~ - v o T t o \ O N O N o o o o o N o o i ^ O N O N O N O N O O T t O N O N r ~ v o c o c S i - H T - H T - H r x i c s » oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooor-~r--r--i-H co m r- ON T-H r- r- r~ r- r~ — - H c r - i i o - H r t ^ ^ T - H ^ ^ ^ o O O O Q O O O O C D O O O O Q O O O O Q o o o o o o o N O i - H C S T t i n v o r ^ o o o c s T t v o o o o c s T t v o o o o c s T t v o o o o c ^ T t v o T H ^ T - H T - H T - H T - H T H i - H c s c s c s c s c s c o c o c o c o c o T t T t T j - T t T t i n i n i n i n 184 Table E-4 Sediment texture and metal concentrations in Burnaby Lake core #1 (metal concentrations expressed as mg/kg dry weight) Core depth (cm) H 2 0 % LOI % Cu Pb Cr Zn Fe Mg M n Cd Ni 4 49 11.1 128 338 20 331 15553 3995 216 0.9 21 8 63 15.8 118 110 17 302 13175 3188 207 0.8 16 14 55 13.0 157 360 27 432 15872 3927 239 1.2 23 19 58 14.4 135 306 20 319 14720 3615 214 1.4 20 23 52 13.8 157 449 22 362 17024 4226 229 1.3 21 30 37 6.2 73 153 11 158 9617 2744 134 0.6 12 36 20 0.7 <1 9 4 61 5483 2338 86 <.2 7 44 70 22.9 86 235 17 283 13123 3134 237 1.2 20 53 52 11.2 109 218 20 243 17975 5057 289 1.1 19 59 52 13.0 253 298 18 243 14819 4069 229 1.3 19 64 21 1.2 <1 9 5 64 5993 2432 89 <.2 7 68 29 3.8 17 65 6 97 7080 2459 97 0.3 7 75 25 2.6 <1 21 4 93 6382 2297 89 <.2 6 80 53 14.0 109 282 17 271 15749 4291 216 1.0 18 85 59 16.3 51 242 21 304 15585 3954 220 1.1 17 90 58 16.6 74 346 26 367 19422 5071 279 1.5 20 Recorded core depths below this point in error due to coring problems, true depth is noted within brackets1. 99 (63) 25 2.1 13 6 69 6755 2435 99 7 104 (68) 28 3.5 2 45 7 109 7547 2583 103 0.3 8 109 (73) 23 1.7 <1 15 4 74 6443 2396 92 <.2 6 114 (78) 58 15.2 64 294 25 355 17886 4605 257 1.4 19 124 (88) 58 15.8 44 271 18 323 16179 4391 213 1.4 21 129 (93) 79 35.8 25 108 13 200 9973 2378 148 1.1 11 136 (100) 22 0.9 <1 4 4 44 4533 2124 77 <.2 5 1. Correlation between contaminant profiles and stratigraphy in the 60-91 cm region of the upper core section with the 96-127 cm region in the lower core section confirms that the second core drive did not re-enter the hole (created by the first drive) cleanly. Table E-5 Sediment texture and metal concentrations in Burnaby Lake core #3 (metal concentrations expressed as mg/kg dry weight) Core depth (cm) H 2 0 % LOI % Cu Pb Cr Zn Fe Mg M n Cd Ni 4 89 30 <1 <3 7 28 5344 1360 92 <0.2 8 9 89 31 <1 <3 6 23 5040 1290 86 <0.2 7 14 89 31 <1 <3 6 23 4740 1224 84 <0.2 7 19 90 31 <1 <3 6 22 5104 1284 89 <0.2 8 23 89 31 <1 <3 7 27 5217 1298 93 0.2 8 29 89 32 <1 <3 15 54 10479 2764 176 <0.2 17 34 90 32 39 89 29 <1 <3 7 29 5627 1611 88 0.2 9 44 88 28 <1 <3 7 29 5459 1547 84 <0.2 8 49 88 26 <1 <3 7 30 5653 1634 85 <0.2 8 59 88 27 <1 <3 7 33 5747 1674 83 <0.2 9 69 88 26 <1 <3 " 8 37 6363 1811 89 <0.2 9 79 86 23 <1 <3 8 33 6016 1960 88 <0.2 8 185 Table E-6 Sediment texture and metal concentrations in Deer Lake (long core) (metal concentrations expressed as mg/kg dry weight) Core depth (cm) H 2 0 % LOI % Cu Pb Cr Zn Fe Mg M n Cd Ni 4 82 24 <1 <3 12 63 7531 2163 198 <0.2 9 9 80 25 <1 <3 12 51 11588 3516 261 0.2 13 14 81 27 <1 <3 12 57 12573 3760 294 0.5 14 19 79 25 <1 <3 10 49 10845 3261 243 0.3 12 24 73 52 <1 <3 7 34 7228 2132 171 0.2 8 29 80 25 <1 <3 11 49 11249 3427 251 0.2 12 34 79 23 <1 <3 12 54 11946 3624 260 0.2 13 39 66 11 17 <3 12 77 13783 3802 339 0.5 13 44 81 25 <1 <3 11 52 11739 3592 264 0.2 13 49 81 27 <1 <3 10 57 10406 3120 247 <0.2 11 54 81 26 <1 <3 11 53 11336 3441 267 0.4 13 59 80 24 <1 <3 12 54 11555 3533 259 0.2 13 64 79 22 <1 <3 11 54 11658 3646 254 0.3 13 69 78 21 <1 <3 19 96 12988 4014 303 0.2 15 74 77 22 <1 <3 12 52 12550 4014 266 0.3 14 79 78 24 <1 <3 11 50 11573 3689 251 0.2 13 84 79 22 <1 <3 11 51 11349 3567 252 0.3 12 89 79 22 <1 <3 12 52 11862 3805 255 0.2 13 94 76 22 <1 <3 20 66 19357 5538 416 <0.2 20 99 78 21 <1 <3 16 53 16223 4522 368 0.3 17 104 79 21 19 <3 23 79 13667 3943 336 0.2 22 109 79 22 <1 <3 16 56 16391 4571 374 0.2 17 114 79 22 119 78 21 <1 <3 33 112 31295 8827 709 0.7 33 124 77 19 <1 <3 16 52 16537 4793 353 • 0.2 16 129 75 18 <1 <3 17 51 16382 4851 332 0.4 17 134 77 17 <1 <3 17 56 17038 4983 359 0.3 18 139 74 17 <1 <3 18 57 17797 5409 358 0.3 18 144 75 18 <1 <3 18 59 18031 5593 367 <0.2 18 149 78 22 <1 <3 16 56 15999 4637 363 0.4 17 186 s a "a o a* 2 w s s o -2 a b S u |Ww<! a •*-» es .§ i «u cs 2 ** es u « S a CU < o cu I i* III <N S o Q 3 <2 « a es o — cs s s - -Si 5 U a o-l •a cu o o a. o 3 . W 3 ••a © .- « •3 u !* cn « • a u 3 «! cu n3£ a. « M U a, cs E Ml H CO VO fH (fl Ifl Tf O N T t T - H Q O O V O O O T t o c e o v O ^ O N P i O Id d . . - i d o o o CO CO 00 $ CO Tt- vo in CO t— cs VO o\ m s CS Tt cs 00 s o 00 ON CO cs cs Ov ON r~- r~ VO o 00 1-H O r- VO 00 in >n © o Q Q 1-H 1-H VO 1-H O o (—N T-H T-H cs O o o O 5 o o o o d o O o o o d d d d d d d d d d d d d d 00 CO m m 3 m 3 cs VO Tfr 00 r- cs cs 8 T t T t vo 5 VO oo VO in 00 VO s co 00 t-~- r~ cs o cs ON CS VO T t o CO VO r» <n r- o Ov r- 00 m CO CO T-H cs cs oo co ^ H I— CO r- T t T t cs co o cs m CS in cs 00 d 1-H d T-H t - H T-H T-H T-H T t T-H T-H O o o o o o d d d d d CO CO d d d d d d d d d d d d d d T t NO T t T t CO CS NO ON CO in CO in CN 00 in d ON T-H d ON CS in ON cs' NO oo 00 00 VO VO NO m T t T t CO ON ON ON ON ON ON ON ON ON ON ON ON ON c o r ~ T t r - T - H T - H T t c o NO od vd Tt -H* co cs' 2 t - H o r ~ v o m c o o N ON ON oo oo oo oo r~-| i n m o o r ~ i n r ~ c s c o c o c o m t-; t-^  r-; oo oq oq oo oq T t in 00 3 V O cs - H CS CO V O Tt 5 O N N O cs I"" 1 cs O N vd in cs cs" cs cs' T-H T-H O N 1-H T-H N O cs cs T-H 00 NO cs cs m ON s r» in CO T t CO T t 00 in 00 T t d T t CO 00 T-H ON p T t cs CO ON VO T-H t - H O od vq vd in in ON CO C-" d co T t T t vd cs ON in cs' in co vo 1 -H NO T-H CO T t T-H T-H cs cs cs CS cs CO CO T t in VO r» 00 1 -H T-H T-H T t r -- H cs , 00 T t 00 00 T t ' ' T-H CS ON OO 00 I • - H NO ON T-H T t © _ ^ H • VO CO T f s CN T t T-H ( - H NO NO m m T t cs cs T - H cs in T-H NO CO 3 o cs ON m ON ON cs T t 3 00 00 o T t cs ON 00 00 NO m T t T t CO in r- 1-H NO T t c— cs T t CO 00 CO 00 co cs cs ON CN CO ON T t in CO d d T-H o ON ON NO in T t CS cs T-H — H d d d 00 cs t - l T f ON ON m 5 cs cs NO ON o o 83 00 o 3 s q d cs o d d d d d d d d 00 VO CO 00 T T vo o NO m vo 3S3 O O O O O CO ON co vo in o T-H T t T t CN T t CO CO o 3 3 s o d cs o s d d d d d d cs NO cs 00 CO 1-H in r-CS co cs ON cs NO ON vq r - cs T-H o 1-H d T-H c s r - - H T t i n o o m v o r ~ vOONONTt—<»-HTtCNCN S r ~ r ~ T t e s o N v o c o T t p co NO in cs cs ON r-; T - H O O O O T - H T - H T - H T - H O O T - H CO N O 00 m N O O N CO T t O N CO in r- co m O N T"H 00 T-H d m d m V O m d T-H d d d d CO m CO 00 m T t T f T t NO 00 3 T t CO ON o o m m cs t - CN ON r- CO r- CN cs T t vq t~ T-H ON 1-H 00 T t r- ON cs p cs O CO CN CN CO in vd ON T-H CO in vd T-H00ONincsm—ivo r ~ T t m T t O N 0 0 T - H v o o o T t c s T t c o o N O O T-H ON NO CO T-H p t-; Tt oo od ON d T-H cs" cs* co c o i n ' r ~ O N t - * c o ' i n ' r - ' o N i - H c o i n r ~ O N T - H r o i / ON i-H cs CO m 00 CO CO T t CO T-H Cj cs ON NO CO T-H T t T f in vd r" 00 CS cs cs cs cs m m m in m CN T t vd 00 CN T t T t T t T t m in in in 00 in T-1 CO in C-" T-H t T f T t T t •n 187 Figure E- l Qualitative descriptions of lake sediment (long) cores (drawings not to scale) Burnaby Lake Core #1, section #1 Core Depth (cm) Lithology Contact Colour (Munsell) Remarks 0-7 ^ < ^ < ^ < ^ < ^ ^ ^ < ^ ' 10YR21 7-9 10YR21 9-11 10YR31 11-24 ^<^<^<^<^<^<^<^ ' ^ <^ <^ <^ * ^ <^ <^ <^ • JQr 1 10YR21 24-33 10YR21 33-42 5Y41 well-washed 42-50 \ ?\ ?\ /\ ?\ /<. /\ ?\ , 7.5YR20 50-60 ^ <^ <^ <^ <^ <^ • 10YR31 60-69.5 5Y41 well-washed 69.5-70.5 7.5YR20 70.5-72 <§><§><$•• & < & < & < ^ < ^ < ^ < ^ < & ' 10YR31 72-78 5Y41 stratified 78-82 10YR31 82-83 5Y41 83-91 10YR31 Legend Lithology sand itiisiu mixed sand & silt organic/ peat Contact distinct, i . j horizontal l ^ n indistinct wavy bottom of core section 188 Figure E - l continued Burnaby Lake Core #1, section #2 Core jj Colour Depth (cm) Lithology o (Munsell) Remarks 91-96 96-104 1 f\A A A T ' V V V V V V V1 04-107 ^ ^ ^ ^ ^ ^ ^ ^ ^ ft fit fit fit ft ft ^ 107-111 111-114 114-116 116-129 129-132.5 132.5-134 134-135 135-138 10YR21 appears contamianujd & unreliable 5Y41 10YR31 5Y41 10YR31 5Y41 5Y41 10YR31 5Y41 appears to be the same stratigraphy as section #1 below 60 cm well-washed well-washed Legend Lithology sand pi I^ J ^ < ^ < ^ < $ silt mixed sand & silt organic/ peat Contact distinct, horizontal indistinct wavy bottom of core section 1 8 9 Figure E-l continued Burnaby Lake Core #2, section #1 ts Core |2 Colour Depth (cm) Lithology o (Munsell) Remarks 0-12 10YR211 12-12.5 7.5YR20 12.5-13.5 5Y41 13.5-19 5Y41 well-washed 19-28.5 10YR31 28.5-29 7.5YR20 29-30.5 ^ ^ ^ ^ ^ ^ ^ ^ <§> & & 5Y31 less organic/peat 30.5-39 ^ ^ ^ ^ ^ ^ ^ ^ 5Y31 more organic/peat 39-40 5Y31 well-washed 40-45 5Y31 more organic/peal 45-48 ^ ^ ^ ^ ^ ^ ^ ^ 5Y31 less organic/peat 48-51.5 ^ ^ ^ ^ ^ ^ • ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ 5Y31 51.5-85 ^ ^ ^ ^ ^ ^ ^ ^ 10YR31 Legend Lithology Contact sand. silt mixed sand & silt organic/ peat distinct, \ - j horizontal i ^ H \ 1 indistinct ^ ^ ^ ^ wavy bottom of core section 190 Figure E - l continued Burnaby Lake Core #2, sections #2 - #6 Core Jj Colour Depth (cm) Lithology © (MunseU) Remarks 85-177 jS^t. jt£k jtfiW- JAL JAL JHHL jHrV. jib* Sg? Sgr Sgr SgP v V V V 10YR21 177-274 10YR21 274-372 10YR21 decomposed 372-374 10YR21 peat 374-472 ^^ ^^ ^^ ^^ ^^ ^^ ^^  ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^  10YR31 organic-rich silt 472-490 SgV S§V ^? "4i*jr ^ ^ & Q ^  ^ 4^  <gV ^ ^ <gV 10YR31 490-500 4 # - 4 4 4 ' 4 > 4 - 4 4 4 4 4 > 4 10YR21 500-562 4 4 4 4 4 4 4 4 -10YR31 Legend Lithology sand silt mixed sand & silt organic/ peat Contact distinct, I horizontal I E indistinct wavy bottom of core section 191 Figure E - l continued Burnaby Lake Core #3, section #1 Core Depth (cm) Lithology u a o Colour (Munsell) Remarks 10YR21 gyttja 5Y2.51 10YR21 5Y2.51 10YR21 Deer Lake Long Core, sections #1 - #3 Core | Depth (cm) Lithology o 0-41 41-76 76-77 77-98 98-145 145-153 Colour (Munsell) 10YR21 10YR21 5Y2.51 10YR21 10YR21 5Y2.51 Remarks gyttja Legend Lithology sand ^ ^ ^ ^ silt mixed sand & silt organic/ peat Contact distinct, horizontal indistinct E3 wavy bottom of core section 192 APPENDIX F JUSTIFICATION OF LAKE SEDIMENT NORMALIZATION The trace metal profiles of Burnaby Lake core #1 provide some justification for the normalization procedure used in the analysis of Burnaby Lake core #2. Obvious chemical and physical markers in core #1 are used to facilitate a direct comparison between cores. A more variable sediment texture is apparent in core #1 (Figure F- l ) ; however, unlike core #2 there is no consistent declining trend in LOI or moisture content with time. In the absence of such a trend, direct comparisons of trace metal concentrations (on a dry basis) within core #1 can be made at depths which contain approximately the same sediment texture (LOI and moisture content). Copper levels peaked briefly at the 59 cm level (253 mg/kg), suggesting this sediment was deposited in 1970 — the age of the peak copper level (303 mg/kg) in core #2. This hypothesis is supported by the presence of a 10 cm well-washed sand layer in core #1 directly beneath the Cu peak and can be attributed to an extreme storm event experienced in 1968 (discussed in sub-section 5.2.1.2). Using this marker, the approximate sedimentation rate at this location is 2.5 cm/year. The trend in Pb and Zn levels over the last 25 years in core #1 more closely resembles the same profiles in core #2 (Figure 5.8) which are expressed on a wet rather than dry basis. This information confirms that Pb concentrations have increased slightly following 1970 and have decreased — marginally — only in the last few years. It also confirms that Zn concentrations have increased since 1970. 193 Figure F - l Metal concentrations and sediment texture in Burnaby Lake core #1. LOI H20 Fe mg/kg (dry) mg/kg (dry) mg/kg (dry) 1 9 4 Figure F-l continued Cd 0.0 0.5 1.0 1.5 2.0 mg/kg (dry) Mn mg/kg (dry) Cr Ni 2000 3000 4000 5000 6000 mg/kg (dry) 195 APPENDIX G LAND USE DATA Permeability Land cover permeability information for each sub-basin is included in Tables G - l and G-2. "Effective Impermeable" area, a measure of the degree to which impervious areas are directly connected to the storm sewer system, was calculated based on factors determined by Dinicola (1990). According to this method, the following areas were included in the category of effective impermeable area : (i) 95 percent of building, pavement, and industrial/commercial impermeable areas, (ii) 66 percent of residential impermeable area. 1 9 6 CO O N r H C • l H > S 3 P H CD > o o 9 H H 0 CD 1 CU I. cs I-a B SS CU k> 05 —^' >> u C3 S E Ice <u I-a , *» l-c CO cu fa (U O. r»> u <u r» |5 03 cu s I. cu a. -*H a CU cv l _ II-oq CO rH ON oq T t T t CO r-; CO CS m rH PH © CO r i CO i n NO © T t CO r-" 00 T t cs' cs m 00 VO 00 cs PH CO O PH P H NO cs CO co cs rH rH cs rH CD r l c T3 <=>-c CN _ t» 3 •O G O g I I I 0 5 , 3 T3 I I 00 * CN rH T t 00 CS -H O co co CN r i © VI CO r l rH rH CN 00 00 CN rH i n CO ON rH r-; cs T t T t ON PH vo r-" © i n cs rH cs t - H rH m cs CN T f O i n © rH CO T t T t VO oq rH rH T t cs © r-' r i t> i n ON i n ON r-' CO rH rH r ~ cs rH rH CS rH CO CO O d d p CS oq o CO i n es O cs CO Tt P H l > CO CO 00 i n r H r H cs ON CO cs r l r H CO CS CS r l CS r l p CN CS CN CS H CS O NO CO cs m r -es cs. cs 3 !> co es OS Tt CO CO C> i n rH CO ON vo co es CO NO CO p CN CO T t pH i n PH CO i n NO T t i n CO CO d vp vo i n co CN CS CO i n r i CO d 00 VO ON m rH es CO es CO © ON m m co vo T t r - 00 NO i n CO PH T t m rH rH es rH T t CS o o o o e s o o o o o o o o o o NO o o o o o o o o o o o o o o o o p © © © d r l O 00 CO O r i o o Tt Tt p p p es © d © ri' d © © PH" © © © o d d 0 O O r l d o o m es T t cs' i n q vq h d in T t © rH CS CS CO t - i n ON © NO i n CO cs vq © © m CO NO NO Os i n T t oq es vq i n CO i n i n 00 d es' © 00 PH r—1 d es' NO i n T t i n OS NO 00 co CO T t fH CN es i n m es CO i—1 © m CO CO r l NO n r - NO r l CO vo Tt n T t T t oo i n co p p co i n T t T t co © od i n r i n ' oo i n cs CO CO n n rH i n 00 © 00 OS © 00 m ON r l NO NO T t T t VO 3 i n cs i n r l 00 r l d d CN NO* co © © T t cs" i n i n CO vo r l ON 58 © CS T t © VO ON ON 00 NO CS © co in © Tt co co © © NO es © Tt Tt Tt os Tt vq co in © l i n ' c o e s ' c ^ O N r H ' i n ' e s o o ' d r H i ^ r H ' c o c o d ' n ' o N ' 1^  _ _^ —. . CO m i H Tt CS CO rH r l C O © i n T t C S r H C S r H oo t~-n CO n n CS T f OS m T t © n T t CO rH r l T t o i n >n r-^ es NO cs os r-; es oo n" es' oo" vo" ON cs Tt es r o d o s o N T t c o r ^ © r ~ ~ O O l n r H O N V O C S r H r H C O © Tt 00 00 Tt O d i n i n T t © © © T t p o N r i o q © O N NO r l CO © © © CO © T t © r l CN* CS T f rH rH T t r~ © i n r ^ o q e s < n p p e N v q e s p v q r H o q r H © p T t v o ' c o r ^ T t ' c o d d c o d r H ' d d v o ' v o ' c N ' d d e s ' o q v q o q p v q p p c o © r ^ d c s ' d e s ' d d d d co oq i n CO r l r i VO OS CO CS* Tt r i n es © T t i n v q p p c ^ e s e s p p p T t T t © © es' Tt © © d d co oo d d PH r i es' i n © n O © O O T t © © O O O O O V O O ' H cs n c S c O T t « n N O r ~ O O O N © r H C N c O T t i n v o r ~ o o o N © p - H c s t O T t m v o r - o o o N r i P H n r i n e s e s c s c s c s c s c s e N e s c s i n Tt r l ON ON NO es m P oo Tt OO es' VO in r l Ico es T t T t ON es ON cs VO ON r l ON ON CS S3 o H 197 ON ON T-H a • TH >N co D P H T H > o 0 -a - H CN 1 o 3 cd H CQ cu h< £ E 3 es cu u v5 CU PJ CU O I P '•S cu E b l cu a . T W c eu CJ •v. eu l i b ( f i i O H f t i n T j - ' i d ^ C h H o o d u i n O v O O O N N O C N l r t - H N O O T H C N l N O N C N ' I C l CN T-H T-H T-H CN t - 1 i n c o c o o e NO O in* CO T-H CN* r~ in CN o O co CN CN CN co "5" 8 ea - 1 s* > § 8 5* c T 3 o - H CN £ 3 •a c o g I > lo g?l •a I o n * o d 00 t-; 00 CN t n 00 CN T f 1—1 r- r-; ON vd 00 d d NO ON T f r-" CO ON 00 CO CN T-H CN CO T-H CN CO CN r- T-H CN o CO T H o T-H NO tn CO CN o p T f T f CN t~~ i n m i n ON r-" 00 CO CN* d CN* T-H co CN* d d T-H vd i n I-H T f CO od i n od CN CO T-H T-H CN CN c~ CO CO CN CO CN m T-H T f CN T-H CN T f CN ON CO 00 Tf* r-° i n T-H T-H CN i n CN CO ON CN T f CN i n r~; vq i n vg co oo vd T f VH vq Tf' CO CN* _ N O V O C O T - H C N C O T - H C N CO CN T-H T-l m T-H O - H t—' vd CN CO p T f VO co* T-* i n CO CN o 00 m CO CO T-H CN T-H CO NO T—1 o T-H CN T f d T-H CN d 00 T f T f TH r - VO i n CN oo in T-H CN p T t r-CN T f 00 T f O ON T t r— T-H T-H m T t O O O O tzi c5 d o o o o o o o o o o o o o o o o o o o o o o o o O O O O T t p O N O p d d d d o d d T-H' d o T t o o o o o o o o o o o p p p p p p p T t T-H 00 T-H 00 CN CO T-l O T t T-H I co* d vd CO T-H o ON o - H m T-H oo m m cs CO o vq CN o CO CO CN CN ON m 00 o CN ON CO T t vq d CN —H CN o vd vd vd vd m* CO* d 00 m" in" m T-H m m CN m CN 1-H 1-H T t CO to T-H m i - H T f r - T f CN T"H CO T—I T f T t ON O CO T f NO T f ON O | o d d o N m ' r - * c o ' c o o d m ' c o d T f m T t CN O T-H TH co T t CN oo p vq co TH p p m p T t ON r - ' d m c N d c o ' c N c T ' d c N ' c N ' d i n T t r - ' T-H CN T-H l-H 00 CO CN t-; NO d T H vd r- T H ON T t o o v q v q m v q c N p c S P p c N c o r ^ v q p o N t ^ t ^ m o o ^ n ^ H i n N d d d ^ n n o J o ' S n i - H M o o i o o o i o o a O oo T f m cN T-H co T t i-H m CN ro >-H i - H T t o o v O T t c N i - i T t c o i - i O N oo T f CO TH T—I T f l—1 CO TH T H C N C N C N T t O v O O N C N N O N O T t r - T t f O O O C N TH NO CN ON vo m t-» r-r - m ON T t T t NO 00 IN T H T H TT T H T f TH CN o vq t n oq CO d r-* T-H ON T f CN NO T f CO 00 ON NO O O O T f r~" CN NO O O CN m c N N O C N p v q m o o T H c o ' d i - H ' d d r ^ N d r - * o o o o d d p p T t oq oq vq oo d d CN* m r~* d CN m p T t vq T t m CN co* d r*- T-H T t co* vd CN T-H r~- co TH m O CO O ON CN ON »H O vd O i - i CN ro T f TH CO m TH co CN vq ON cs* m* od CN* 0 T f t~-d CN in NO p p CO CN d d d i - H co* od o o o T f T t p m p p p p c s p p o d TH* T^' cs d d d ON d d d d Tt* T-H CN • H C N c o T t r n v o r - o o o N O T H C N c o T t m N O r~ oo ON o TH CN co TH TH TH CN CN CN CN in vo r - oo ON CN CN <N CN CN 198 CN o o CN < a o o _ w o a. a o CS rH o o 1 2 2 o o CN VO — \o S3 >-> a o '-8 8 - o .2 . «"H I "O rH o 1 - H 35 o > O O O 0 N O O O O O O O O T t O O O V 1 O O p p V N V 0 T t p c N a s 0 N i d d d t d d d d d d d d d d o o T f o d \ 0 rH c o i n o q c o v O c n o v c N c s i n q c s o ^ o q o i n q • t ' o d r H V d i n i n ' r t T t d d o v d T f N d d - H d d d d d d d d N d ^ CN CO CN rH T f rH rH rH <T) o o o o o o o o o o o o o o p p p p p p p p p p p p p p p 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 c o p c x j s q p s q c o i / ^ p p p o s O s T t r ^ p p i n T t c ^ d d ^ v d d N r t i n d d d v d - H ^ T t d o M T f V d i r i d w d d r l d r i rH rH rH rH CO rH ft"! o o o o o o o o o o o o o o o o o o o o o o p o p p p p c s 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TJ-O O r H r i c N O O O N O O O p p p p p O O O O O O O O O O p p p d o v c ' r ^ e s 0 0 ^ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 cN c o O O O O O O O O O O O O O O O O O O O O O O O O O O O O V N o o o ^ o d d c ' o o o o o o d d o d d d c ' o o o o o o o c N ' 0 \ O O C N O O O O O O O p p p p p p p p p p > r i p o q p p p p p c ^ o d r H ' o o o o o d d d o o o o o o d d d d d c o ' c ' d o d o CN c » p « o c » c 5 q o q p c » r r T f p p r H > r ; p p p r H ^ m d i < i ^ » H ' d ^ ' r t ' n d d N < i r » d d N ' n c N d d d m d d ^ d H CO CN NO CN rH rH rH CO o o o o o o o o o o o o o o o o o o o o o o o o o o o o o d o o o v o o d d o o o o o o o o o d d d o o o d d o d o o CN o o o o o o o o o c j N O O O T j - o r r o o o o o o p p p p p o q p r H r H o d d d o o d u S d o o v d c ' ^ o o o o o o o o ^ d c ' c ^ o i o i n o o m m c / i O O O ^ o o o o o M O O O o o o o o o o o q q p i ; O T f c o c s c N o d o v d o o o d O ' H ' o d c ' o 6 o o o o o o o » T-H rH T-H CN c o i*o s q s q I-H s q CN p rH v i p T t r-^  ON c o o s q CN ON T t T t T t o s T t NO s q c o p r i c » \ d o « S ' r " r - ' i r i c N \ d d r H c o d v o t ^ OS r~ T t O rH CO CO rH T f H M O " CO I f l ^  N r l (S) H OO >Tl C O C O r l r H T t r I CO rH CN T t o q o q v q t ^ T t v N v q o q v q i r i i ^ c N v q d c ' r H t > d c N w S T t C » O C N l ^ r H T t C » d r H ' \ O i ^ T t V N i n n r - -H CN CN 1 1 1 SO (S SO C O r H O N < O T t C O n t > r H t > O O r H l O rH CN -H T f rH p O T t o o o o o T t o p p p p v q p p p c o p p p p p p p p p O N d d o N o d d d o o N d d d d d c o d d o o o o o o o o o o o o o o o o c s o o o o o o o o o o v q o o o o o o o o o o o o o o o o o o o o o o o o o ^ o o e N o ^ o o o o o o o o o o c ' CO CO rH O O O T t O r H O O O O O O T t O O T t p O O > O p O p p p p p c N p o q c N T t r n ' o c o o d d o o d o o o c s o o o c N O o d r H C N C O T t > ^ V O r ^ O O O N O r H C N C O T t l n s O r ~ O O C > O r H C N C O T t l n s O C - - O O O N H H H H H H H H H H C S , N ( S | N ( S t S l N N C S ! ( S l ON I ON T t _ CN O vi ON oo CN s q i r ^ c ^ p s q T t r H p c s p r ^ c N i r i p T t o s q o s q o o o T t o r i T t ' s d o d i r ) T t c » c O T t > X c N T t v O C N d v i o c » 0 T t CN rH CO rH CN rH rH CN rH CN T t rH CO CN rH CN CN V I rH V© T f ' rH T t s CN ON SO V ) CN CN d SO O T t r-o o CN 00 CN 00 CO c o o CN 00 CN SO Tt o H 199 8 a © o _ CN o c o H 5 o o I. 38 1151 o CN V O — v o CS •-> s o "-8 8 ON e Q - CM . —> 1*1 o o o o o o o o o o o o o o o o o o o o o o o p o p p p p © d d d d © d d d d © d © d o ' o d d d o d d d d d d d d d CN T t CN 00 T t c o CO CN CN p CN O N T t © © © V } © © © © © p p © O N © 00 d r ^ d o I ^ w i ^ H ^ ' d d d v i i t d d d H ' d d d d d d d d N d w CN CO CN CN NO rH rH m l o o o o o o o o o o o p p p p p p p p p p p p p p p p p o q d d d d d d d d d d d d d d d d d d d d d d d d d d d d v i v N O o o v N p p p ^ p p p T t O N T t i n p p w ^ T t o q ^ c o p d d ^ v d d 6 d « i d d d d " T f ' M d d c n ^ ^ d i n ' d c n ' d d " O H rH C N rH i—1 C O rH C O o o o o o o o o o o o o o o o o o o o o o o o o o p p p c s d d d d d d d d d d d d d d d d d d d d d d d d d d d d T t CN V N i n o O C N C N V O C N O O V O r - r H C N c O O V O O V N O c o O O O c o O C O T t C N C N O v o v o — < o \ c o r H r > T t v o o r - v o T t o v o o c N O > n CN rH S O rH rH CN rH C N CN CN T t rH C O C O rH o o o m o H M O M n d CN CN r - CN CN v o O O r H p u - j O O O N O p p p p p p p p p p p p p p p p p p p p d o M - H o o d d r n d d d d d d d d d d d d d d d d d d d d d o o o o o o o o p p p p T t p p p p p p p p p p p p p p p p d d d d d d d d d d d o H ' d d d d d d d d d d d d d d d d l o o p o c o o o o o o o o o p p p p p p p p p u - i p v N p p p p p d d d v d d d d d d d d d d d d d d d d d d d d ^ d d d d d O O C O < n O r H O O O O O T t T t O p r H ^ o q p p r H C O T t p p O p p T t r H p C » ( o ^ ' i n ^ d H ' d ^ r l c n ^ d c s i s D ' c o ' d d c s i i n ^ ' d d d i n ' d n H ' d c o ' C O C N V O C O rH rH H H f ^ C O o o o o o o o o o o p p p p p p p p p p p p p p p p p p p d d d d d d d d d d d d d d d d d d d d d d d d d d d d d p p p p p p p p p p p p p p o o o o o o o o o o o o - H O N O o o o o o o o o o d d d d d d d d d d d d d d d d d d c o d j O O r H p o q - H O p p r H r H p o O O C N O O O O s O O O O O O O O O V N 2 f 2 S 2 0 0 0 2 ^ ' , " ° ° ° o o o O ' * d d d d d o ' d e N d » O s r H O T t c ^ s q c N p p o O C N c O O N C N O O N C N ^ O O H T t T t v n O s o r ^ O i n ^ 2 2 ^ H co c o - H m H ( f i « - H c o v N v O T t c N - H c o c o o o v o m H H r H f O r H r l CN oq r~~- sq T t sq oq co oq T t rH CN sq T t cq CO ON CN oq o CN CN VN CO T t sq t™*" C^"* T j " V ) P H r l C O ^ H 00 G) t"™* V " l T j " T j " G) V } VO I"*"" VO t^"* I*"1" CS l/S OS V * T t C O V N r H r H r H CN CS « 1 M H O T f M M H 1 ^ H ^ * CN CN CN r H C O r H p p c N p p p p p T t p p p o p v q p p o v o o o o o o o o o o o s d d r ' d d o ' d d ON d d d d d CO © d d co d d d d d d d d d oo' v o v o o p o o c N o p o o o o p o o o v q o o o o o o o o o o o o o d d d d d d d d d d d d d d d c N d d d d d d d d d d d d d T t o v o v o c o p p p p p p p o q p v q p p p p p p p p p p p r H p v q h T t ^ H r - d d d d d d d d d r t d d d d d d d d d d d d d d r H C N c o T t v n v o r ~ o o o N O r H c s c O T t i r i s o r ^ o o o N O - - c N c o T t > n s o r ^ o o o N rH rH rH rH rH rH rH rH rH rH CN CN CN CN CN CN CN CN CN CN o d oo Os CN 00 CN 00 00 CN co ON T t NO C O C O o d c o o o v N T t p o o p o N p p p r ^ T t ^ p p v N p p p p p o N O C N co oo d CN co CN d d d *o d d d od d r" o d T t co d d d d *-* o co os d n rH CN rH rH SO rH 00 O T t T t d V O ON CN ON NO CN 1/1 00 CN VO 00 CN o H200 i ON T3 <D X ) ca t H 5 c 2 OH-IO •s c •fl •i-H •* 4-* c 8 > N o i-H CH 6 w 6 >^ JO a H co Z o u > o o co o • s CO H \ 0 ' H ( ^ ^ c « M ^ ^ o c e l » c e c ^ o o \ 0 ' t H H T * H ; ^ H H O i O N O ^ c - H \ O M O i r i » o i f i m h i n M O \ ' t ' H r t i n O N ' H ^ T t <N m o ON CN r-NO t N iri cs Nf <H cs TH cs co in^(»i^NOWcNiNO^CiOinaiinc>ciNvi,otwooo\NO'tfs ONONOOONOOinTtTHONTtCOONincSCO ro cS to N H H rt h cs cs NO cs ^ i n r -c o c s N O c O ' - H O T t i n ' - H oo-i— l « i IH i n ON I-HON >-H -H i n •"Hcocoinr~0--icooOTtTtinONinr}-cSTj-oo H T f M I O H CO TH COCO NO O N r ^ o o i n c o ^ T t i n c c r ^ c s c s c > N O O N O - * O N r ~ o o - ^ o o c s r ^ i n r o ( N v o i n o > o o v o N O i n c s o N i n c o T r c S T H O O T H ^ ^ c o c O T H T H C S T H c o i n c S ' - H i n T t T t CN NO TH c o m T t o o o o s T H m c s o o c s r - o o c s c s v o r - - o o - H T H N O N o m o r ~ c o o m c s r-ocsmrtTHTHcom -H co oo cs so <-> cs cs CS H NO - t O N T f i - H O O m H ' I H . TH m TH m v o m o o o o o r - - T H r > r ^ c o o c N N O T H c o O N O o o O T t c o o c s r i - N q r ~ o o m oo T< NO i o t o (O m co cs TH TH -TH CS — H cs cs O t o i t H o m N O O o o o o o o o o C H h T f T H O c S T H O o r - r ^ T j - r t m m c S T H T H O T ) -T f CS TH CS TH TH CO TH TH ' ON T H C s o o o c s r--r~-cs T t T t CN o -H r— TH CO ON r - m <1 T f T f _ . c o c o r - TH c s m o N O N O O O N r n v o i O N O O N c o Q c s c o O o o T t O N m c s Q 5 T t c o r ~ c o c o c o cs p co c o m T t cs ON NO i n m r- c o c o oo o oo c o c o co o O N N O O O C O T t c S O O m O ' — I T H O O C O O N O O O O O C O O N O O O N O O ON CN TH i n T-I TH C O T H T H <—I T H C s c o T t m v o t ~ o o O N O T H c s c O ' T t m r — O O O N 201 cs - cs c s c o T t m N Q r - - o Q O N c s c s c s c s c s c s c s c s m c o s T t c o NO ON CO cs T t T— p. T t r~c~ c s o o o o o c o c s c o r~oooTtoooNCSTto--HOt~ooNOor~ o o o N O ' N O O O O ^ C O T t T t T H T t T t m T H T H T t - i m C S TH co c s T t NO co o NO T t i n * O C O C O C S C S » - ' C S T H TH I - H C O O O T H c o c o o o c o c o T H ^ v o o o m c o O T t o o o o m o o c N O o O T t m O T t m c o N Q O N N O c o c o c o c o c c N O c s o N N O T t T H i n o cs T t m m NO CN c o f N H f » H T f TH TH CS CS CS TH OO s T t m 00 c o 3 cd *-» o H as 1 T 3 C cd 13 CO C O E • ^ e •» .9 1 1 Wl 00 as <u -C co g <o a 5 CQ ID •s c • i - H • f l • r - H *—< a <u e O H w V O 6 u T-H - O crj H o co Z o o o o cj o cj Z cj H • X) CO > O i H i n o o r ^ » O v o r » r ^ o O ' H o o v O o o r ^ N » ^ r o ^ v p Q v D v O c > v i N H \ o ^ ^ o c c H O ^ ^ « c ^ T J o ^ ^ c ^ c > n ^ N H O ^ S ( < l c ^ H \ o i * ^ c ^ O M O c ^ M l n c ^ v 6 o o m 4 c ^ N ^ N c N H ^ N l ! 0 ^ m T f ^ ON vo v o m r- >—i TH T f VO m H r f M i H V O r f m H T f m c o O t N O v ^ H i n - - ' VO C l CO < H ( T | t-H T H T-H rH i m Tt <-H m CO > T H vo CN M N O \ 1 - ( N V O » O T H t r ) 0 \ i c l ^ r I T < T H r > c O O V O V O ' H o o v O H o o i O r H P i i n t N t N i N m m o v w T t -H TH i n H ( « l O T H CN T H H H H ( S ON VO CO TH CN T H T f T t CO T t , r - r ~ c N T t v o > n o o o N Q . « « . . . _ . H CN TH Tt Tt o vo 0 0 co i n co co m vo TH co THcocovoinr~THr~r-T t c s c o c N c o T H O c o r -TH ON CN r~vor~THr - - t--incNONTtoomTHor~ - c N c N o o i n T t c o r ~ O N O O O N O C N coTHvOcNt~-voTfinovocNON>nincocNONTtcNONcoTHONTH — — — CO 0 0 T t CN TH T H T H CN t-~ co i n CN Tt t - - m v o o N C N o r ~ o o v o o o O ' - H T t T t T f o o o A r - ~ O N O i n N O O N r ~ c N p ~ i n r - -c> TH co T t r— m T t co T f ro o NO TH CN i n TH TH >n CN CO co i n cN CN T t , H T H T H T H « n r ^ O c o c o C N O T H T H i n c N V O O v O O O t - - O N ' - H O > n T H T H O O N O N T t o O O O N ON^oocoTtcocNmONCNOoocN NomvoincomcNTHTHOTHOooco v o v o c o m ON c o c o CN r o <N CN T t H T H m m oo S3 CN CN O N C O i n i n c o - H T t c o v q c N i n c O T t r o c N r ^ i n c N o o o ON T t CO TH TH CN —< CO - H c o r ^ O c O N O c o i n i n O T f O N C O r O T - H c N T f CN TH 0 0 TH CN C O T H i n O N f - O N N O T f TH CN r -CN 0 0 ON 0 0 I CN o o T H m o o c N v o r ~ o c N Q r > o v o c N i n o t ~ T f i n c o c N T H T t c o i n T f o o N > — i t3vm ov oo m vo N m H cs i n NO oo CN ro ( V O C O T - H T H Tf TH T H CO t - H T f C N v o c N o o r ~ i n T H c o r ~ o o o N i n O N T t r - - v o r - c o c « t ^ v o c o c o c o t ~ c N C N c « c o c N r ^ T H c o N O c o c N o o T H T H C o c o o s T H T H c o T H T - t CN H o o o v o r ; NO CN CN CN CO TH TH TH i— i CN r- m o CO T f T t ro TH o T H <—I t-H H i n T H T t i n c O T H O O c o o o i n O N T H t ~ O O r - - O c o c o O N O O O c o o i n m i n co CN I-H CN Tf -HCNcoTf invor^ooosOTHCNcoTfmr^ooc^QTHCNcoTfm TH TH T H T H I - H T H T H T - H T H C N C N C N C N C N C N C N C N C N C N 202 CN 8 Tf 8v CO ON ON NO CO 85 CN CO 3 CN CN NO CN NO 0 0 3 OO ON ON CN o H 1 4 > O E • .2 •a 8 S3 0 0 •a O o g 60 a II B 8 CQ !* 3 ^ |8 s s ".§ 2 co <u "J o jf u « l . a O X! S bl « g 3 o 5 « p g o h o «H o • e|8 ON ON T3 O C3 c PQ 43 •5 a <u 6 >» 0 " E H a w 1 O <u i—H 4= cd H C/3 Z o > o o 2 o u o z vo co o oo - H r -rj rj- m " • C O i-H ON H 3 H H f i H H r / i H O O t N n o \ ' t | n * O H ( s o o \ o r— oo T t . _ . . _ _ T H r > T H O T t c o c N O N r » c N o o v o c N c o (N h O O H H THi-H CN T H —< CN CN >n 3 co T H co r> CN CN CN HH Tf c N c N c N 0 0 O i - H > n r -c N c o i r i T f c N c o c o i n i - H c o c N O c o n o N o r ~ T H CN NO r-~ T H ~ o c N r ~ T H N O c o T f o < - < < N c o m v o v o c o T H O N C N C N O N T f c N C N T H O T f i n O ^ T f c O c O C O N O i - H C O C N C N Tf i n C O C O —. NO C O T H T H T - I T H C N T H C N T H T H CN i—imococNr^cocNTfr-r-ooi-H r--i-Hinooo>nr~TfinNOTfONCN I - H T H O CN «-H T H I—t S O Tf m CN Tf O O i - H c o o r ~ - i - H r ~ c N > r i i n c o H T H T H co r— NO i n H H i n ^ o e o o n N O H N t C N ^ i ' i O o o o i o i n h o n o o o O H C O O O r ~ - < - H O O O O T f i - H O O C N O C N T f O N O O T f T f C N C O T H T f i - H N O ' n T H O N c o O N O N O N O N C N T H i - H c o i - H r - i m c N co co >n CN T H T H vovoooTHOcNcNTfONTfoooTfincoooi m T H i r i T H c o c o c N O N < n c o i - H U - i o o i r i T f CO ON CN Tf NO CN CN T H CN CN f—tf—ti-Ui-^ v O O N i n o O T f O N O C N C N C N C N Tf Tf o Tf vo ini-Hco T H r~ oo CN T H C O C O T H C « C O C O O T f c O C O N O T f c N C O C O O N O T f O O N O O N C N O O l r i C O > n T f c o c o i n c o v o i n i - H o o i n c o O N C N C N O N I N H H CN T f m i n oo vo ici vo ov • I - H I-I T H co r -r ^ c o o N i n o o o N O O o o c N C N T H T f c N c o r - c o T f v o c o t ~ - r ^ v o i - i T H C 5 C O C ~ V O T H O O O - H I — I C O I — i r — O N > n c N t — T f o o o c o m c N > n c o r ~ v o O t ~ l n t ~ T f T f T f l — I T f T f C N l — I T f l - H T H V - l I—I H 1/1 C N T H C N O N c o v o m o c o o o » n o o o r > 0 c o T f r ~ c o o p i - H c ^ t - - T f o o r > p i n f T  T H i n > ^ ^ f ~ T f C N i C N m c r - N C O N C N c o i m oo CN O CN CO I-H CO CO Tf i—I CN CN T f 00 T H 00 CN 00 O NO Tf T f ON T f m CN NO Tf r> CN O O vo i n cN T H co co co t— vo I - H CN O N V O T f c o v o T f T f T f v o c o o c o o o i n i n r - - c o c o c o o N c o T f c o c o O N i n i - i O N r - v o c N c o r — T f i n V O V O O O T H i — I H i - H Tf I—I CN T H CO T H c o c N i - H O O c o r ^ i n c o o v c o CN i n ON ON Tf Tf T H T H CN NO CN c—— co T H CJN I/"I i n t—— H I H CN I*— T f H OQ — H ON T f ON T f r—* co T f ^ f i*— CN ON CN CO o o O i - i T f O N t - - o o i n i n r ~ c N i n c « c N ' H C N oo c o T H ov r> r- co ' ' " " NO CNlnmCNTfTHTHTHTHTHTHTfTH Tf CN CN Tf Tf r- o ON - H vo H r » ON Tf Tf I—I T H I—I T H r - Tf ON r - Tf >n i—i CN T H T H i H C N T H O T H O i - H O O O T H T f T H T H CN T H C N C O T f i n N O t - O O O N O T H C N C O T f ^ ^ O O O j ; 203 ON CN 00 00 00 s T f 3 CO T f oo NO C O C O VO 00 Tf VO r-c o i n 3 ON 00 CN I I ON o oo Tf o $ Vi 73 3 Vi C eg c/3 el| 8 &z ii l a ON ON 0 0 a s u CS a. os -a <—i <u oo •to* ON a —: 4> ON a o o eu ON Os CS £> s C S v O t v O M ( S ™ T j o o t N r N v O T f f f i O ~ . - - • ON f O 00 i n N t N t N r n ( N H t N t— C N co C N N O ' CN O ' H ' J - i n t N ' H i n r H O M n ^ f l T t oo i n c n TJJ t O O V O i n T t O O O O ' - r ~ O N - H T j ; T f O l T f O O n i — I C N C N T t C N m O s i n m cs oo I H I H „ H C S - H T t cn C N C N p r l C N C N r H T f r H C N O C O r H S O T t C » O O C N r H r H O C O O N O r H t ^ r H O C O O O r- T t T t 0 0 -5 C N r t m r H o . oo >n rH C N C N C N r H r H T t C O r H c n vi T t N O V 1 0 0 t ^ V O r H S O C N O N t ~ O C ? \ r ^ t ~ r ~ S O O N r ~ r H r H i n T t N O C n r H S O O O - - •- • • - — •- — • •- -*• — osr-msomosmooNTt C O T t C N r H l / - > r H T t r ~ r H C N c n N t s o i ' l w r ^ s u N o c N O N t ^ o w r ' r - r ^ v m i O N t ~ V ^ T t S O 0 O T t C ~ 0 O r H 0 O T t T t V N T j Q T t C n c n T t 0O C O C N r H r H C N t*"" C N C N r n C O T}- C N C O ON SO NO C N C N T t r H C N r H > n s O c O C O C X 5 0 0 t ~ r H C N C N C N C O V N i n O N O C N C > r H C N V 1 r H C ^ N O O g N O N O O O O O r H S O O W ^ N O r H C N O r H C O C O t ^ O N C O l O V 1 C O C O r H r H T t C O C > l O O - " C N C ^ C S r H r H S O C N r H C O C S C O C N r H C O T t C O T t C N r H U - l r H r H Q 00 C N C O - H C N r H C N CN ON ON i n i n [ O N O S C O S O r n so r - oo N O m m T t T t so i n m C N r n c o r H r H r H i n o c o c o i n T t o o O T t r ^ c o r ^ o O T t O N O N r ~ - O N N O r - -s O O N O O s o r ^ i n c o c N T t c o c N i n c N r H O s c N s o o o o c O T t c N T t r H T t r H C O C N C N N O r H C N O O r H C N C N V I C N V } r - r H H H C N N O S O N O r H t ~ « n N O O O O T t r H T t t ^ O C N N O T t r - ~ r ~ > n C O C O - ' - s O T t < n < n T t c o s p t ~ c » O T t T t i n r H o o o o c o N O • c N c » r - O N < 5 o N T t r H s o c N c o o o s o s o c N T t ON ON >n _ T t CN CN rH T f ON C N C N r H l H O N . . oo C N so C N r -O O C O N O O N r H O C O O O O O r H C N r r ^ N o o o r ^ T t O N O N O N r H r ^ - r H r H r H r H O 0 0 r H C N c o T t i n s o t ^ o o o N O r H C N c O T t > n r ^ c » o s O r H ( N c o T t i n s o t ^ c x 3 c 3 N r H r H r H r H r H r H r H r H r H C N C N C N C N C N C N C N C N C N C N ON I C N 0 0 m oo i n c o 0 0 C N 0 0 C N 0 0 0 0 C O m c o m c o >n ON CN T t m CN NO i n O oo ON SO CN Os m C N •a I PQ e •r-< T J <D s xS E o o 3 'S S 1 s u "S I T J c 3 03 O N C r H CO • T J •a e o I -8" O N 32 PQ bom 2 C N < 3 H o 2 0 4 APPENDIX H STREAMBED SEDIMENT RESULTS Table H - l Sediment texture and total metal concentrations in streambed sediments (nitric acid digest, values expressed on a dry weight basis) Station LOI Silt & Fe Cu Or Pb Ni Zn Mn Mg Cd Hg Clay % % mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg Pg/kg #1-1 7.0 37478 68 42 58 23 195 1249 7146 <3 136 #1-2 7.8 35020 61 43 62 28 205 725 6376 <3 164 #1-3 4.1 25720 69 30 69 52 158 331 4526 <3 97 #1-Ave. 6.3 79 32739 66 38 63 34 186 768 6016 <3 132 #2-1 5.1 16860 46 27 59 14 119 515 4445 <3 113 #2-2 4.0 26790 38 65 32 42 113 338 3113 <3 86 #2-3 7.1 19139 70 31 95 17 170 750 5097 <3 198 #2-Ave. 5.4 57 20929 51 41 62 24 134 534 4218 <3 132 #3-1 7.9 22675 67 34 75 17 200 450 4845 <3 139 #3-2 5.4 29865 69 25 56 13 130 375 4102 <3 142 #3-3 4.3 28303 38 56 33 37 106 379 8087 <3 130 #3-Ave. 5.9 64 26948 58 38 55 22 145 401 5678 <3 137 #4-1 5.2 16461 48 19 61 12 123 1417 2807 <3 60 #4-2 3.0 25601 37 20 29 10 88 1148 3445 <3 32 #4-3 5.2 18535 40 22 56 10 135 1333 2855 <3 60 #4-Ave. 4.5 41 20199 42 20 48 11 116 1299 3036 <3 51 #5-1 2.2 12445 24 11 <15 6 40 1135 2509 <3 634 #5-2 8.8 32516 57 20 73 14 182 1040 3454 <3 109 #5-3 3.9 18180 33 23 49 7 110 1642 4247 <3 78 #5-Ave. 5.0 42 21047 38 18 43 9 111 1272 3403 <3 274 #6-1 2.6 5962 30 12 45 <6 65 288 4015 <3 46 #6-2 3.9 13485 38 18 53 8 110 704 2362 <3 63 #6-3 3.2 11864 25 15 45 10 85 691 2338 <3 61 #6-Ave. 3.2 32 10437 31 15 48 7 87 561 2905 <3 57 #7-1 4.6 15280 30 28 40 16 100 574 4095 <3 84 #7-2 6.2 31368 45 24 40 13 179 2131 4182 <3 95 #7-3 2.9 14213 20 19 30 9 65 320 2367 <3 77 #7-Ave. 4.6 29 20287 32 24 37 13 115 1009 3548 <3 85 #8-1 4.8 10918 30 17 41 8 109 268 1960 <3 54 #8-2 4.3 16966 30 23 30 7 92 722 2996 <3 45 #8-3 3.6 11739 30 22 25 10 85 535 3297 <3 54 #8-Ave. 4.2 35 13208 30 21 32 9 95 508 2751 <3 51 #9-1 6.4 22464 55 28 94 14 153 968 3645 <3 136 #9-2 2.6 11745 45 29 85 10 100 710 3049 <3 80 #9-3 5.1 18213 52 22 109 12 129 743 3365 <3 130 #9-Ave. 4.7 37 17474 51 26 96 12 128 807 3353 <3 115 #10-1 9.9 35721 114 32 245 18 220 1401 5103 <3 47 #10-2 6.3 23802 72 28 134 17 206 1301 4263 <3 78 #10-3 8.0 31769 237 46 843 23 503 1603 4220 <3 112 #10-Ave. 8.1 44 30431 141 35 407 19 310 1435 4529 <3 79 205 Table H - l (continued) Station LOI Silt & Fe Cu Cr Pb Ni Zn Mn Mg Cd Hg Clay % % mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg rlg/kg #10(94)-1 11.1 34473 72 22 211 18 227 2000 3656 <3 #10(94)-2 5.7 27530 47 30 81 17 182 884 6516 <3 #10(94)-3 5.6 22500 34 24 43 17 120 575 6503 <3 #10(94)- 7.4 29 28167 51 26 112 17 176 1153 5558 <3 Ave. #11-1 6.9 23609 122 29 68 10 161 1048 2971 <3 70 #11-2 6.3 26729 130 29 75 13 283 1133 3721 <3 141 #11-3 3.8 18211 50 20 43 10 110 743 3556 <3 98 #11-Ave. 5.6 31 22850 101 26 62 11 185 975 3416 <3 103 #13-1 3.7 13197 18 15 <15 9 50 332 2778 <3 53 #13-2 4.1 19066 31 19 29 7 135 1647 3294 <3 53 #13-3 4.6 18718 80 21 30 8 135 1348 3294 <3 30 #13-Ave. 4.2 31 16994 43 18 22 8 106 1109 3122 <3 45 #14-1 2.8 22756 26 19 22 10 88 885 2998 <3 32 #14-2 3.0 16897 27 18 66 9 84 800 2783 <3 39 #14-3 5.1 31962 49 18 28 7 105 689 2882 <3 78 #14-Ave. 3.7 38 23872 34 18 39 9 93 791 2888 <3 50 #15-1 8.0 47189 51 18 <15 7 155 3695 5343 <3 <30 #15-2 6.6 38332 35 19 45 6 163 3020 3345 <3 <30 #15-3 8.8 48651 49 21 20 <6 171 3491 4288 <3 <30 #15-Ave. 7.8 52 44724 45 19 24 6 163 3402 4326 <3 <30 #15(94)-1 5.6 25197 24 14 24 12 107 449 3890 <3 #15(94)-2 6.6 42740 28 17 28 13 441 2238 4065 <3 #15(94)-3 6.2 25951 29 20 31 12 113 2067 4580 <3 #15(94)-4 5.1 20740 21 15 31 10 84 1460 3845 <3 #15(94)- 5.9 38 28657 25 17 28 12 186 1553 4095 <3 Ave. #16-1 4.1 12933 23 16 25 8 115 544 3845 <3 70 #16-2 6.1 17818 87 25 64 52 267 1460 3390 <3 35 #16-3 4.0 13716 40 19 20 11 115 514 3741 <3 75 #16-Ave. 4.8 42 14822 50 20 36 24 166 839 3659 <3 60 #17-1 2.9 16149 37 30 30 15 62 218 4175 <3 42 #17-2 10.9 24052 141 50 220 16 205 750 5351 <3 75 #17-3 6.8 16042 113 39 197 11 215 455 4348 <3 81 #17-Ave. 6.8 35 18748 97 40 149 14 161 474 4624 <3 66 #19-1 6.7 10529 58 22 77 9 110 266 2688 <3 36 #19-2 6.0 9575 60 19 50 9 114 184 2580 <3 88 #19-3 2.5 8004 49 18 34 6 105 150 2346 <3 59 #19-Ave. 5.0 17 9370 55 19 53 8 110 200 2538 <3 61 #20-1 7.4 15652 120 19 114 11 232 2136 3630 <3 72 #20-2 2.6 6917 21 13 57 <6 104 573 2213 <3 <30 #20-3 13.4 16201 76 22 88 16 177 1237 3464 <3 120 #20-Ave. 7.8 31 12923 72 18 86 10 171 1315 3103 <3 69 206 Table H - l (continued) Station LOI Silt & Fe Cu Cr Pb Ni Zn M n Mg Cd Hg Clay % % mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg ug/kg #21-1 5.1 13072 46 24 93 14 127 283 3095 <3 96 #21-2 6.5 19810 54 34 30 19 144 1357 3566 <3 79 #21-3 7.2 22382 55 33 57 18 167 1077 3612 <3 109 #21-Ave. 6.3 49 18421 52 30 60 17 146 906 3424 <3 95 #23-1 10.7 25905 165 46 133 21 340 500 5851 <3 287 #23-2 14.1 22403 176 44 145 24 438 392 4776 <3 172 #23-3 11.4 22010 166 44 125 21 395 400 5025 <3 173 #23-Ave. 12.1 76 23439 169 44 134 22 391 431 5217 <3 211 #24-1 4.1 22060 89 39 127 20 203 1208 3554 <3 85 #24-2 6.8 38627 167 56 283 25 353 3600 4187 17.0 110 #24-3 4.1 21181 100 39 160 17 185 1204 3352 <3 110 #24-Ave. 5.0 38 27289 119 45 190 21 247 2004 3698 7.0 102 #25-1 23.2 15056 49 18 52 12 100 360 2301 <3 617 #25-2 13.0 15235 46 18 25 14 85 330 3297 <3 423 #25-3 23.4 11413 55 18 51 10 82 311 2270 <3 <30 #25-Ave. 19.9 77 13901 50 18 43 12 89 333 2622 <3 352 #26-1 3.2 19400 73 26 74 11 125 820 3850 3.5 67 #26-2 4.3 19870 51 27 68 14 145 939 4094 <3 44 #26-3 5.1 15265 41 19 75 7 115 848 3542 3.0 80 #26-Ave. 4.2 40 18178 55 24 72 11 128 869 3829 3.0 64 #27-1 5.7 20504 60 27 60 18 180 1125 3591 <3 72 #27-2 6.8 21943 54 29 63 15 175 1237 3461 <3 51 #27-3 7.0 21817 55 30 96 16 233 1458 3476 <3 81 #27-Ave. 6.5 53 21421 56 29 73 16 196 1273 3509 <3 68 #28-1 8.4 28355 89 28 85 12 139 970 3214 <3 54 #28-2 5.2 26821 80 27 135 15 130 681 3578 <3 41 #28-3 6.7 27352 84 28 89 13 139 1015 3357 <3 39 #28-Ave. 6.7 45 27510 84 28 103 13 136 888 3383 <3 45 #29-1 4.6 14397 29 19 39 10 72 576 2495 <3 77 #29-2 4.4 9147 24 15 23 7 46 193 1972 <3 257 #29-3 3.8 10745 24 18 17 8 63 303 2349 <3 128 #29-Ave. 4.2 36 11430 26 17 26 8 60 357 2272 <3 154 #30-1 6.8 29759 261 41 159 26 330 455 5102 <3 160 #30-2 2.7 16093 122 23 93 40 185 224 3340 <3 70 #30-3 4.4 23804 201 38 130 17 271 357 3506 <3 132 #30-Ave. 4.6 41 23219 195 34 127 28 262 346 3983 <3 121 #31-1 6.6 21114 256 30 137 15 342 1392 3110 5.7 184 #31-2 10.4 30512 368 41 199 20 465 1751 3937 <3 165 #31-3 4.3 18253 213 27 86 12 215 860 3393 <3 99 #31-Ave. 7.1 37 23293 279 33 141 16 341 1334 3480 3.0 149 207 Table H - l (continued) Station LOI Silt & Fe Cu Cr Pb Ni Zn M n Mg Cd Hg Clay % % mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg Pg/kg #32-1 9.3 13363 70 21 107 11 130 225 2803 <3 68 #32-2 4.8 12319 119 44 239 21 195 185 2710 <3 162 #32-3 17.7 10993 51 29 52 12 96 172 3126 <3 43 #32-Ave. 10.6 47 12225 80 31 133 15 140 194 2880 <3 91 #33-1 3.1 25490 174 33 562 16 163 369 2927 <3 2115 #33-2 5.7 21748 152 31 195 23 334 976 3975 3.5 341 #33-3 6.8 22107 159 35 164 17 337 2973 4079 <3 153 #33-Ave. 5.2 28 23115 162 33 307 19 278 1440 3660 <3 870 #34-1 5.2 22520 140 35 198 18 367 257 4127 <3 104 #34-2 2.8 19614 137 28 183 17 216 549 2937 <3 282 #34-3 3.1 21029 148 41 188 16 183 291 2886 <3 255 #34-Ave. 3.7 22 21054 142 35 190 17 255 366 3317 <3 214 #35-1 6.1 19720 155 37 120 15 280 339 4593 <3 164 #35-2 6.6 19288 139 38 117 20 304 284 4079 3.1 144 #35-3 4.8 17352 133 37 110 20 265 239 247 <3 104 #35-Ave. 5.9 49 18787 142 37 116 18 283 287 2973 <3 137 #37-1 8.5 16387 154 34 229 14 258 641 2791 <3 137 #37-2 14.2 26780 284 47 261 25 309 1076 5217 <3 185 #37-3 7.5 17676 159 32 129 18 246 448 4108 <3 114 #37-Ave. 10.1 40 20281 199 38 207 19 271 722 4039 <3 145 #37(94)-1 4.3 16362 115 28 111 18 131 282 3285 <3 #37(94)-2 1.9 11140 52 20 102 17 84 159 2491 <3 #37(94)-3 2.7 16449 128 26 146 9 120 288 2926 <3 #37(94)- 3.0 20 14651 98 25 120 15 111 243 2901 <3 Ave. Note : concentrations below detection limit were included in averages using 50% of detection limit value, : stations resampled in 1994 contain "(94)" in the sample description. : refer to Table A - l and Figure 4.1 for station locations 208 Table H-2 Extractable metal concentrations in streambed sediments (0.5 N HC1 acid digest, values expressed on a dry weight basis) Station Fe Cu Cr Pb Ni Zn Mn Mg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg #1-1 14094 32 3 46 <6 84 760 1020 #1-2 15212 38 4 51 7 104 500 1219 #1-3 8840 24 3 40 <6 56 168 720 #1-Ave. 12716 31 4 46 <6 81 476 986 #2-1 5001 24 3 52 <6 74 280 370 #2-2 4403 20 5 10 9 30 118 1641 #2-3 6098 37 5 78 <6 58 440 620 #2-Ave. 5167 27 4 47 <6 54 279 877 #3-1 8423 37 4 66 <6 112 158 620 #3-2 14034 23 3 50 <6 56 158 400 #3-3 6403 15 6 14 <6 42 168 972 #3-Ave. 9620 25 4 43 <6 70 161 664 #4-1 6798 22 3 42 <6 82 1300 410 #4-2 4800 14 3 26 <6 60 562 316 #4-3 8203 20 5 42 <6 78 1066 390 #4-Ave. 6600 19 4 37 <6 73 976 372 #5-1 6824 15 3 12 <6 16 1147 328 #5-2 19278 37 6 58 <6 114 880 502 #5-3 5597 17 3 18 <6 28 114 242 #5-Ave. 10566 23 4 29 <6 53 713 357 #6-1 2759 10 3 19 <6 70 190 210 #6-2 5039 18 4 39 <6 64 568 350 #6-3 3202 13 <3 23 <6 40 486 328 #6-Ave. 3667 14 3 27 <6 58 415 296 #7-1 6238 16 6 26 <6 52 482 576 #7-2 14607 29 5 39 <6 102 1785 674 #7-3 4799 11 <3 17 <6 30 288 218 #7-Ave. 8548 19 4 27 <6 61 852 489 #8-1 3201 14 4 24 <6 44 390 318 #8-2 3599 12 3 26 <6 50 588 338 #8-3 4202 19 3 32 <6 46 686 458 #8-Ave. 3667 15 4 27 <6 47 555 371 #9-1 7599 37 7 73 <6 78 714 640 #9-2 2599 19 4 57 <6 40 206 346 #9-3 4581 69 6 32 <6 72 556 570 #9-Ave. 4927 42 6 54 <6 63 492 519 #10-1 18192 46 7 194 6 156 1034 870 #10-2 9599 36 5 108 <6 138 1008 686 #10-3 10995 82 9 638 12 252 1042 620 #10-Ave. 12928 55 7 313 7 182 1028 725 #11-1 15402 94 6 56 <6 126 1192 640 #11-2 12199 78 8 48 <6 154 826 546 #11-3 8483 35 5 30 <6 82 734 738 #11-Ave. 12028 69 6 45 <6 121 917 641 209 Table H-2 (continued) Station Fe Cu Cr Pb Ni Zn M n Mg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg #13-1 4739 20 3 23 <6 22 304 400 #13-2 7920 29 4 26 <6 94 1580 460 #13-3 8577 33 6 26 <6 80 1379 460 #13-Ave. 7078 27 5 25 <6 65 1088 440 #14-1 10804 10 3 22 <6 56 620 400 #14-2 200 1 <3 <6 16 248 380 #14-3 19522 17 3 28 <6 48 580 420 #14-Ave. 10175 10 3 25 <6 40 483 400 #15-1 30392 14 5 10 <6 94 3199 578 #15-2 19591 16 <3 6 <6 66 2199 498 #15-3 30986 18 4 12 <6 98 2599 530 #15-Ave. 26990 16 4 9 <6 86 2666 535 #16-1 3181 17 <3 25 <6 70 272 210 #16-2 6080 29 4 36 <6 102 1368 558 #16-3 3002 <3 20 <6 66 312 364 #16-Ave. 4087 23 <3 27 <6 79 651 377 #17-1 2221 12 3 24 <6 12 40 360 #17-2 11619 93 13 181 <6 102 520 940 #17-3 3599 60 6 146 <6 66 202 656 #17-Ave. 5813 55 7 117 <6 60 254 652 #19-1 3022 37 <3 71 <6 72 174 438 #19-2 2019 28 <3 59 <6 70 84 360 #19-3 1300 14 3 33 <6 46 34 246 #19-Ave. 2114 26 <3 54 <6 63 97 348 #20-1 5823 94 <3 101 <6 142 2011 628 #20-2 1980 14 <3 36 <6 46 432 300 #20-3 5141 48 <3 76 <6 146 1150 666 #20-Ave. 4314 52 <3 71 <6 111 1198 531 #21-1 4498 18 3 90 <6 54 118 334 #21-2 3000 18 3 32 <6 56 936 260 #21-3 6001 26 5 54 <6 84 828 326 #21-Ave. 4500 21 4 59 <6 65 627 307 #23-1 6542 88 7 106 7 222 200 580 #23-2 7223 88 8 99 8 288 204 740 #23-3 6096 94 6 126 7 252 144 640 #23-Ave. 6620 90 7 110 7 254 183 653 #24-1 6921 51 7 109 6 136 1036 664 #24-2 17203 102 14 238 9 220 2600 618 #24-3 5823 55 7 132 8 136 1030 620 #24-Ave. 9982 70 9 160 7 164 1556 634 2 1 0 Table H-2 (continued) Station Fe Cu Cr Pb Ni Zn M n Mg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg #25-1 6803 29 3 36 <6 58 226 320 #25-2 4518 26 <3 23 <6 46 162 360 #25-3 6079 29 <3 40 <6 50 266 420 #25-Ave. 5800 28 <3 33 <6 51 218 367 #26-1 3000 28 3 48 <6 66 494 310 #26-2 3600 22 4 50 <6 76 272 546 #26-3 4801 28 5 51 <6 82 640 364 #26-Ave. 3800 26 4 50 <6 75 469 407 #27-1 6440 27 3 47 <6 92 848 366 #27-2 8196 27 5 52 <6 90 982 418 #27-3 10657 40 6 80 <6 124 1534 520 #27-Ave. 8431 31 5 60 <6 102 1121 435 #28-1 12129 58 5 72 <6 72 799 430 #28-2 9296 46 4 95 <6 54 440 520 #28-3 12539 52 5 82 <6 66 770 460 #28-Ave. 11321 52 5 83 <6 64 669 470 #29-1 4401 12 4 26 <6 38 436 206 #29-2 3798 14 <3 20 <6 42 164 158 #29-3 3401 12 <3 18 <6 34 206 260 #29-Ave. 3867 13 <3 21 <6 38 269 208 #30-1 10872 90 7 124 <6 190 214 620 #30-2 4299 40 3 54 <6 102 78 392 #30-3 6538 63 5 96 <6 132 170 460 #30-Ave. 7236 64 5 91 <6 141 154 490 #31-1 8794 163 3 117 <6 224 1289 520 #31-2 12722 174 8 140 <6 250 1278 679 #31-3 4966 97 4 70 <6 140 618 419 #31-Ave. 8827 145 5 109 <6 204 1062 539 #32-1 3602 36 4 76 6 104 90 362 #32-2 3581 53 8 172 8 100 78 320 #32-3 2141 25 3 42 <6 50 68 500 #32-Ave. 3108 38 5 97 6 85 79 394 #33-1 8199 132 5 414 6 100 242 618 #33-2 6725 90 6 190 <6 212 661 781 #33-3 9998 108 7 124 7 252 1222 1004 #33-Ave. 8308 110 6 243 6 188 708 801 #34-1 7697 70 8 150 <6 234 120 780 #34-2 5602 91 5 144 7 132 400 520 #34-3 5516 79 6 159 7 108 138 520 #34-Ave. 6272 80 7 151 6 158 219 606 2 1 1 Table H-2 (continued) Station Fe Cu Cr Pb Ni Zn M n Mg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg #35-1 4699 67 5 91 6 170 116 460 #35-2 4694 74 7 89 7 184 102 519 #35-3 4057 57 9 75 <6 150 96 460 #35-Ave. 4483 66 7 85 6 168 105 480 #37-1 4579 95 6 169 11 146 138 980 #37-2 6880 187 8 227 12 200 760 1040 #37-3 3581 99 6 92 8 156 220 780 #37-Ave. 5013 127 7 163 10 167 373 933 Note : concentrations below detection limit were included in averages using 50% of detection limit value, : all Cd determinations were below detection limit. (3 mg/kg). : refer to Table A - l and Figure 4.1 for station locations. 212 A P P E N D I X I S T R E E T S E D I M E N T R E S U L T S Table 1-1 Sediment texture and total metal concentrations in street sediments (nitric acid digest, values expressed on a dry weight basis) Station LOI Silt & Fe Cu Or Pb Ni Zn M n Mg Cd Hg Clay % % mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg rlg/kg C l 8.5 43 28685 224 71 250 40 742 408 5329 <3 <30 C2 11.8 50 19558 189 73 349 47 791 319 4569 <3 225 C3 6.3 17 18087 75 73 267 27 305 235 3822 <3 45 C4-1 10.0 20908 125 60 259 38 485 385 4592 <3 42 C4-2 5.7 22374 135 58 134 29 379 274 5581 <3 134 C4-3 7.8 27251 175 74 416 34 1383 464 5590 <3 70 C4-Ave. 7.8 32 23511 145 64 270 34 749 374 5254 <3 82 C5 7.2 33 27120 268 61 363 40 627 395 3871 4.1 42 C6 12.4 56 25688 170 69 358 43 728 444 5337 <3 43 G l 7.9 59 16271 80 28 40 23 199 384 3564 <3 <30 G2 6.4 25 10609 30 16 23 11 112 203 2275 <3 55 G4 12.7 29 16823 125 47 143 27 427 285 3819 <3 <30 11 14.2 57 23098 239 50 220 35 1110 294 3554 4.7 50 12 7.0 28 23441 140 46 221 25 603 370 5098 <3 <30 13 4.1 60 16045 110 27 90 23 200 270 5098 <3 <30 14 7.3 52 25429 244 63 351 44 588 369 5335 <3 43 15 4.9 42 20539 224 48 179 31 420 338 4819 <3 196 16 4.3 25 19731 126 58 222 26 321 286 2817 <3 33 17 8.5 42 20683 130 42 115 38 681 305 5108 <3 48 18 7.9 9 19402 205 48 156 76 372 315 3315 <3 85 R l 7.0 32 17859 85 33 246 30 300 295 4337 <3 36 R2 8.8 73 19650 55 35 66 31 140 400 4845 <3 46 R3 7.8 33 17228 129 65 455 24 336 283 3038 <3 58 R4 6.8 37 13734 100 35 229 29 250 245 2802 <3 61 R5 8.2 26 19377 180 55 457 28 439 380 3311 <3 124 R6 5.8 28 13594 40 27 91 24 133 252 3323 <3 41 R7 9.2 47 16312 77 34 144 28 241 305 4476 <3 <30 R8 8.8 17 18134 75 40 475 33 373 336 3321 <3 74 Notes : refer to Table A-2 and Figure 4.2 for station locations. 2 1 3 APPENDIX J SPEARMAN RANK CORRELATION MATRICES Table J - l Spearman rank correlation matrix for total metal concentrations in streambed sediments (Nitric Acid digestion, n=33) F e M g M n Cu Cr N i Hg Pb Zn LOI Silt& _ _ _ _ _ _ Clay Fe 1.00 M g 0.51 1.00 M n 0.41 0.16 1.00 Cu 0.43 0.38 -0.07 1.00 Cr 0.48 0.65 -0.14 0.75 1.00 N i 0.44 0.57 -0.17 0.70 0.88 1.00 Hg 0.13 0.27 -0.31 0.55 0.56 0.64 1.00 Pb 0.29 0.32 -0.04 0.87 0.75 0.71 0.51 1.00 Zn 0.56 0.54 0.17 0.90 0.78 0.70 0.45 0.80 1.00 LOI 0.22 0.39 0.09 0.53 0.31 0.39 0.27 0.43 0.48 Silt& 0.37 0.37 -0.04 0.03 0.24 0.37 0.16 -0.03 0.08 Clay Table J-2 Spearman rank correlation matrix for extractable metal concentrations in streambed sediments (weak HC1 acid digestion, n=33). Fe M g M n Cu Cr Pb Zn LOI Silt& Clay Fe 1.00 M g 0.50 1.00 M n 0.50 0.20 1.00 Cu 0.23 0.69 0.03 1.00 Cr 0.21 0.54 0.05 0.66 1.00 Pb 0.05 0.55 -0.07 0.87 0.68 1.00 Zn 0.27 0.67 0.23 0.78 0.59 0.72 1.00 LOI 0.21 0.41 0.02 0.48 0.33 0.46 0.39 Silt& 0.21 0.15 -0.16 -0.06 0.06 0.00 0.07 Clay Note: Ni not reported since only 7 stations above detection limit 2 1 4 Table J-3 Spearman rank correlation matrix for metal concentrations in street sediments (n=25). Fe M g M n Cu Cr N i Hg Pb Zn LOI Silt& Clay Fe 1.00 M g 0.60 1.00 M n 0.72 0.59 1.00 Cu 0.77 0.35 0.48 1.00 Cr 0.67 0.27 0.34 0.64 1.00 N i 0.70 0.44 0.54 0.66 0.56 1.00 Hg 0.04 -0.24 -0.01 0.20 0.26 0.38 1.00 Pb 0.44 0.05 0.30 0.44 0.71 0.43 0.36 1.00 Zn 0.81 0.45 0.48 0.82 0.71 0.70 0.19 0.54 1.00 LOI 0.28 0.20 0.42 0.20 0.25 0.49 0.09 0.16 0.49 1.00 Silt& 0.19 0.52 0.35 0.20 -0.04 0.18 -0.24 -0.22 0.14 0.31 1.00 Clay 215 A P P E N D I X K C A L C U L A T I O N O F A U T O M O T I V E T R A C E M E T A L L O A D S This section outlines the calculation methods and factors used to determine trace metal emissions from automobiles. Emissions from fuel and automotive components (tires, brake pads) have been calculated separately. (i) Fuel-related emissions (kg/day) were calculated as follows: Vehicle km/day x fuel consumption rate (L/km) • emission factor • concentration of metal in fuel (kg/L). (ii) Tire and brake pad emissions (kg/day) were calculated as follows: Vehicle km/day • component wear rate (kg/km) • # of components per vehicle (tires only) • concentration of metal in component (kg metal/ kg component) Factors and concentrations applied in calculations Fuel consumption • 0.24 Litres / km Tail-pipe emission factor (fraction of combusted metal in fuel which is exhausted) 0.30 (Loranger et al. 1994) Tire wear rates • 16.5 mg / km / tire (Woodward-Clyde Consultants 1992) Brake pad wear rates • 7.64 mg / car km (Woodward-Clyde Consultants 1992) 2 1 6 Lead concentration in gasoline and automotive components • 1973 average Canadian additive level (gasoline) : 0.6 g / L (Alberta Research Council 1982) • 1993 additive level: 0 mg / L • background concentration in gasoline : 10 mg / L (Lee and Jones-Lee 1993) • concentration in brake pads : < 0.01 % (Armstrong 1994) • concentration in tires : 30 pig / g (median - (Sadiq et al. 1989)) Copper concentration in gasoline and automotive components • concentration in tires: < 20 p:g / g (median - (Sadiq et al. 1989)) • concentration in break pads: 2.26 % (Woodward-Clyde Consultants 1992) • background concentration in gasoline : 4 mg / L (Shaheen 1975) Zinc concentration in gasoline and automotive components • concentration in tires : 1.018 % (Woodward-Clyde Consultants 1992) • background concentration in gasoline : 10 mg / L (Shaheen 1975) Manganese concentration in gasoline • additive concentration in gasoline in 1973 : 0 mg / L • average additive concentration in gasoline in 1993 : 11 mg / L (Alberta Research Council 1982) 2 1 7 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0050348/manifest

Comment

Related Items