UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An evaluation of water quality and land use in the Salmon River watershed, Langley, B.C, using GIS techniques Cook, Kathryn Emily 1994

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

Item Metadata


831-ubc_1994-0519.pdf [ 6.49MB ]
JSON: 831-1.0087692.json
JSON-LD: 831-1.0087692-ld.json
RDF/XML (Pretty): 831-1.0087692-rdf.xml
RDF/JSON: 831-1.0087692-rdf.json
Turtle: 831-1.0087692-turtle.txt
N-Triples: 831-1.0087692-rdf-ntriples.txt
Original Record: 831-1.0087692-source.json
Full Text

Full Text

AN EVALUATION OF WATER QUAUTY AI’Tt LAND USEIN THE SALMON RIVER WATERSHED, LANGLEY, BC, USING GIS TECHNIQUESbyKathryn Emily CookB.Sc., Queen’s University at Kingston, 1979A THESIS SUBMITrED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SaENcEllTHE FACULTY OF GRADUATE STuLJEs(Department of Soil Science)We accept this thesis as conformingto the required standardThe University of British ColumbiaOctober, 1994© Kathryn E. Cook, 1994Signature(s) removed to protect privacyIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives, it is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of o,,/ cg-The University of British ColumbiaVancouver, CanadaDate &b— I9- /q94IDE-6 (2188)Signature(s) removed to protect privacy11ABSTRACTThe Salmon River watershed is at the rural-urban fringe of the rapidly developingGreater Vancouver Regional District (GVRD) and has undergone substantial land-use changein the last 20 years. This study provides information on spatial and temporal patterns ofwater quality within the Salmon River, using trace metals in sediments, and nitrate and totalphosphorus in water as indicators of the quality of the aquatic environment. Relationshipsbetween water quality, surficial materials, and land use within the watershed are examinedusing a watershed approach combined with GIS techniques.Approximately 80 km2 in size, the watershed areas’s land is 50% agricultural, 7%residential, and 25% undeveloped. The glacial outwash deposits in the middle reaches ofthe river make up the Hopington Aquifer, which is an important source of ground-watersupply and baseflow during the low flow period of July through September.The trace-metal concentrations in fine-fraction (<63 sum) sediments from 19 siteswithin the stream system were compared with the background metal concentration insurficial materials from within the watershed. No evidence was found of elevated totalmetal concentrations in sediments for the metals Zn, Cr, Cu, Co, Ni, and Mn. At thestations on the glacial outwash parent materials, N03-N, specific conductance, and chlorideincrease in the downstream direction, and are likely due to anthropogenic additions. Thedifference in specific conductance, chloride, N03-N, and total phosphorus from the early1970s to 1991-93 is small, but localized increases in N03-N were observed in CoghianCreek at the confluence with the Salmon River and in Davidson Creek.Seven percent of the over 400 wells within the Salmon River watershed that havebeen tested during the past 20 years have N03-N concentrations above the Canadian111drinking water guideline of 10 mg U’. A spatial lag occurs between the location of wellswith elevated N03-N and the increase in N03-N in the surface water of the Salmon Rivermainstem. This is attributed to the northward flow of ground water from the HopingtonAquifer into the Salmon River.Residential land use, (1600 septic systems installed since 1970) and agricultural land-use were characterized by: i) a density index, an estimate of the land-use activityimmediately upstream of the water sampling station, and ii) a cumulative density index, thedensity of the land-use activity in the entire watershed upstream of the water samplingstation. High density indices of both agricultural and residential activities are present in theareas of high N03-N in surface and ground water, which suggests that both of these land-usetypes are sources of N03-N. The pattern of the cumulative density index matches thepattern of N03-N in surface water more closely than the density index, and illustrates thatthe water quality at a point in the river can reflect the land-use activities of the entirewatershed upstream. A spatial lag is observed between the higher density of land useactivities and the increase in N03-N in surface water.Examining relationships between land use and water quality in the Salmon Riverwatershed is complicated by multiple sources of N03-N in close proximity, changes in therelative importance of sources within different locations in the watershed, and the temporaland spatial lags between N03-N leaching from the land surface and its detection in theground and surface water. Using Spearman rank correlation, both agricultural activities andseptic systems are correlated to N03-N contamination of ground water. However, theapproach used in this study does not allow for the determination of the relative importanceof these two sources.ivAbstractTABLE OF CONTENTS11Table of Contents.ivList of Tables ixList of FiguresXIIList of AbbreviationsAcknowledgements1. Introduction1.1 Study goal1.2 ObjectivesXvixviii155781416203. The Salmon River Watershed Study Area3.1 Physical features3.1.1 Climate3.1.2 Surficial geology3.1.3 Ground water resources3.1.4 Drainage network3.1.5 River discharge3.1.6 Surface water quality . .3.1.7 Ground water quality3.2 Biological features3.2.1 Fish resources3.2.2 Natural areas3.3 Cultural features3.3.1 Population3.3.2 Land use3.3.3 Water and sewerage services3.3.4 Discharge permits3.3.5 Water licenses242424273033343440424243434345494950502. Indicators of Water Quality2.1 Trace metals in sediments2.2 Chlorides in surface and ground water2.3 Total phosphorus in surface water2.4 Nitrate in surface and ground water. use zoningEnvironmental management within the Salmon River watershed . . . 55V4. Methods4.1 Field methods4.1.1 Sediment sample collection4.1.2 Surficial material sample collection4.1.3 Water sample collection4.1.4 Surface water quality measurements4.2 Laboratory analysis4.2.1 Sediment and surficial material analysis4.2.2 Water analyses for nitrate+nitrite-N, chloride, and total4.3 Compilation of existing surface and4.3.1 Surface water quality4.3.2 Ground water quality4.4 Land-use mapping4.4.1 Water wells4.4.2 Septic systems4.4.3 Agriculture4.5 GIS analysis4.5.1 TRIM base map4.5.2 Contributing areas4.5.3 Surficial geology4.5.4 Land-use analysis4.6 Statistical analysis5. Results and Discussion5.1 Trace metals in fine-textured river-bed sediments5.1.1 Variability in analytical methodology5.1.2 Metal concentrations in soil parent materials 825.1.3 Spatial variability of metal concentrations in river sediments5.1.4 Comparison of trace metals in soils and river sediments5.1.5 Seasonal variation in trace metals in river sediments5.1.6 Temporal variation of trace metals in river sediments between 1970sand 1991 925.2 Surface-water chemistry5.2.1 Spatial and temporal variability in surface-water specificconductance5.2.2 Spatial and temporal variability in surface-water chloride5.2.3 Spatial and temporal variability in surface-water total phosphorus .5.2.4 Spatial and temporal variability in surface-water N03-N5.2.4.1 Spatial variability5.2.4.2 Temporal variability5.2.4.3 Other data sources of N03-N in surface water5.2.5 Relationships between surface-water quality variables5.3 Spatial and temporal variability in N03-N in ground water5.3.1 Spatial variability in N03-N in ground water5.3.2 Annual variability in N03-N in ground water5.3.3 Temporal variability in NO3-N in ground waterphosphorusground-water quality records5757575959616262656666676969707172727273767780808086879298100105106114114120123129131131135141vi5.3.3.1 1973 - 1981 . 1415.3.3.2 1973 - 1993 1435.3.4 Spatial and temporal distribution of N03-N in well water 1495.4 Comparison of surface water and ground water N03-N 1515.5 L.and use activities 1535.5.1 Installation of water wells 1535.5.2 Residential land use 1555.5.3 Agricultural land use 1585.6 Spatial relationships between land use and surface water quality 1645.6.1 Residential Land Use 1675.6.1.1 Relationship between septic systems and surface-waterN03-N 1705.6.2 Agricultural activities 1745.6.2.1 Relationships between agricultural activities and surface-water N03-N 1745.7 Relationships between land use and ground water quality 1795.8 Spearman rank correlation coefficients between land use and N03-N 1825.9 The effect of agriculture versus residential land use on N03-N 1856. Summary and Conclusions 1886.1 Spatial and temporal variability in total trace-metal content of streambedsediments 1886.2 Spatial variability in water quality 1896.3 Temporal variability in water quality 1916.4 Land-use activities in the Salmon River watershed 1936.5 Spatial relationships between land use and water quality 1937. Recommendations 1967.1 Recommendations for further research 1967.2 Management recommendations 1998. Literature Cited 202Personal communication 214Appendices 215Appendix 1 Description of the sediment and water sampling locations in theSalmon River watershed 216Appendix 2 Emission lines used and detection limits for ICP-AES used forthe determination of metals in sediment and soil samples 217Appendix 3 Surface water quality stations in Beale (1976) and the EQUISand SEAM databases, and the equivalent station numbers fromthis study 218VIIAppendix 4 Agricultural land activity codes from the 1989 land-use map,and the categories included in this study as intensiveagricultural operations 219Appendix 5 List of TRIM features imported into TerraSoft for the SalmonRiver Watershed and Hopington Aquifer 220Appendix 6 Generalization of soil types from 1:25000 soil map(Luttmerding 1980) for the production of the surficial geologymap, Figure 3.8 221Appendix 7 Measurement of the precision of the laboratory technique fortrace metal analysis, by comparison of the ratio of the duplicateanalyses 223Appendix 8 Summary statistics for trace metal concentrations in surficialmaterial and streambed sediments in the Salmon Riverwatershed 225Appendix 9 Total concentration of trace elements in river sediments andparent material types in the Salmon River watershed and in themarine reference standard, MESS-i 227Appendix 10 Sample preparation and digestion methods of sediment samplescollected in 1970’s and 1991 233Appendix ii Summary statistics for the streambed sediments collected in the1970’s and August 1991 in the Salmon River 234Appendix 12 Determination of pH in surface water in the Salmon River,August 1991 to August 1992 235Appendix 13 Determination of temperature in surface water in the SalmonRiver, August 1991 to September 1993 236Appendix 14 Determination of specific conductance in surface water in theSalmon River, August 1991 to September 1993 237Appendix 15 Determination of chloride in surface water in the Salmon River,August 1991 to September 1993 238Appendix 16 Determination of total phosphorus in surface water in theSalmon River, February 1992 to September 1993 239Appendix 17 Determination of nitrite+nitrate-N in surface water in theSalmon River, August 1991 to September 1993 240vii’Appendix 18 Measurement of the precision of the nitrate+nitrite-Ndeterminations in surface water in the Salmon River 241Appendix 19 Measurement of the accuracy of the nitrate+nitrite-Ndeterminations in the surface water in the Salmon River 242Appendix 20 Site number identifiers used for the 12 NAQUADAT well sites . 243Appendix 21 Summary of septic system installations in the Salmon Riverwatershed and Hopington Aquifer, derived from Ministry ofHealth records at the Central Fraser Valley Health Unit,Langley BC 244Appendix 22 Relationships between the density index of agriculturalactivities on all surficial materials and nitrate-N in surfacewater during low-flow conditions in the Salmon Riverwatershed 245Appendix 23 Relationships between the density index of agriculturalactivities on glacial outwash materials and nitrate-N in surfacewater during low-flow conditions in the Salmon Riverwatershed 246Appendix 24 Relationships between the cumulative density index ofagricultural activities on all surficial materials and nitrate-N insurface water during low-flow conditions in the Salmon Riverwatershed 247Appendix 25 Relationships between the cumulative density index ofagricultural activities on glacial outwash materials and nitrate-Nin surface water during low-flow conditions in the SalmonRiver watershed 248Appendix 26 Median nitrate-N values in well water tested between 1983 and1993 for each contributing area 249Appendix 27 Relationships between agricultural activities on all surficialmaterials and nitrate-N in well water 250Appendix 28 Relationships between agricultural activities on glacial outwashmaterials and nitrate-N in well water 251Appendix 29 Spearman rank correlation coefficients between land use andnitrate-N in water, when the Salmon River watershed isgrouped into 23, 13, and 10 land areas 252ixLIST OF TABLESTable 2.1 Trace element concentrations in rocks and soils 10Table 2.2 Background concentrations of trace elements in sediments 11Table 2.3 Guidelines and criteria for trace elements in freshwater sediment 15Table 3.1 Canadian water quality guidelines and BC water quality objectives forselected water quality parameters and water uses 38Table 3.2 The proportion of land use activities in the Salmon River Watershed in1971, 1979-1980, and 1989-90 48Table 4.1 Water and sediment quality measurements performed at selected stationson the Salmon River and tributaries 61Table 4.2 Sources of historic information on surface-water quality data for theSalmon River watershed 67Table 4.3 Sources of information on ground-water quality data for the SalmonRiver watershed 68Table 4.4 Size of each contributing area and the water quality sampling station foreach contributing area 75Table 5.1 Measurement of the precision and accuracy of the microwave digestiontechnique using the marine sediment reference material, MESS-i (NRC) . . 81Table 5.2 Median metal content, and significant differences in metal contentbetween the three dominant surficial material types in the Salmon Riverwatershed. Pairs of the surficial material types were compared using theMann-Whitney U test 86Table 5.3 Median metal content in the three dominant surficial material types andthe <63 ,um streambed sediments in the Salmon River watershed. Metalconcentrations significantly higher in the sediments than in the surficialmaterials are indicated with a letter. Pairs of the surficial material andsediment were compared using a one-tailed Mann-Whitney U test 88Table 5.4 Spearman rank correlation matrix for metal concentrations in SalmonRiver <634um streambed sediments. (nt38) 90xTable 5.5 Median metal content in the <63 1um streambed sediments collected inAugust and December, 1991, in the Salmon River watershed, andsignificant differences between August and December sediments. TheAugust and December sediments were compared using the Wilcoxonmatched-pairs signed-ranks test 93Table 5.6 Median metal content in the streambed sediments collected in the 1970’sand August, 1991, in the Salmon River watershed, and significantdifferences between 1970’s and August 1991 sediments. The 1970’s and1991 sediments were compared using the Mann-Whitney U test 97Table 5.7 Median metal:Fe ratio the streambed sediments collected in the 1970’sand August, 1991, in the Salmon River watershed, and significantdifferences between metal:Fe ratio in the 1970’s and August 1991sediments. The 1970’s and 1991 sediments were compared using theMann-Whitney U test 98Table 5.8 Mean daily flow measured for the Salmon River at the 72 Ave crossingat the gauge station 08MH090 for the surface water sampling dates in1974-75 and 1991-93 101Table 5.9 Spearman rank correlation coefficients between water quality variables,using all Salmon River watershed stations sampled at high and low flowconditions in 1991 to 1993 130Table 5.10 Spearman rank correlation coefficients between water quality variables,using all Salmon River stations located on glacial outwash materials,sampled at high and low flow conditions in 1991 to 1993 132Table 5.11 Number of wells tested for Nitrate-N in the Salmon River Watershed andHopington Aquifer Area between 1973 and 1993, and the concentrationof Nitrate-N found in the wells 133Table 5.12 Summary statistics for water wells in the Salmon River watershed andHopington Aquifer 153Table 5.13 Estimate of the percentage of dwellings in the Salmon River watershedrepresented in the septic system database, for selected Master Legal mapsheets within the watershed 157Table 5.14 Estimate of the accuracy of the septic system database for dwellings lessthan twenty years old 159Table 5.15 Summary of the subset of agricultural activities selected from the 1989land-use map for this study. Land-use information from Sawicki andRunka (1990) 162xiTable 5.16 Grouping of contributing areas for examining relationships between landuse and water quality 165Table 5.17 Septic systems installed in each contributing area between 1970 and1993. Data from CFVHU-MOH records 168Table 5.18 Agricultural land use in each contributing area, as the percentage eachcontributing area 175Table 5.19 Spearman Rank correlation coefficients illustrating associations betweenmeasures of land use activities and nitrate-N in water in the SalmonRiver watershed 183xiiLIST OF FIGURESFigure 3.1 Location of the Salmon River watershed within the Lower Fraser basin . . . 25Figure 3.2 Location of the Salmon River and major tributaries and residential areaswithin the watershed in the Township of Langley 26Figure 3.3 Surficial geology of the Salmon River watershed, generalized from the1:25000 soils map (Luttmerding 1980) 29Figure 3.4 The changes in depth to water table in three observation wells in theSalmon River Watershed, from 1982 to 1990 32Figure 3.5 Hydrograph of the Salmon River at 72 Ave crossing at the gauge station08MN090, based on discharge measurements from 1970-1993 35Figure 3.6 Average daily discharge measurements for the Salmon River at the 72Ave crossing at the gauge 08MH090. Triangles indicate water qualitysampling dates for this study 36Figure 3.7 Location of the urban areas within the Township of Langley and theSalmon River watershed 45Figure 3.8 The 1989-1990 land-use map of the Salmon River watershed 46Figure 3.9 Land-use zoning in the early 1970’s 53Figure 3.10 Land-use concept for the Township of Langley, from the 1993 Rural Plan . 54Figure 4.1 Location of the stream-bed sediment and surficial material samplingstations within the Salmon River watershed 58Figure 4.2 Location of the surface water sampling stations within the Salmon Riverwatershed 60Figure 4.3 Delineation of the land area contributing to each surface water qualitystation, within the Salmon River watershed 74Figure 4.4 Components of a box-whisker plot 78Figure 5.1 Box-whisker plots illustrating the variability in total metal concentrationsin Salmon River sediments and surficial materials 83Figure 5.2 Spatial variation in concentration of total zinc and carbon in SalmonRiver <63 am streambed sediments, sampled in August 1991 91xl”Figure 5.3 Box-whisker plots illustrating the variability in total metal concentrationsin Salmon River sediments collected in the 1970’s and 1991 95Figure 5.4 Scatterplots of trace metals versus Fe for Salmon River streambedsediments sampled in 1970’s and August 1991. A. Zn vs Fe. B. Cu vsFe 99Figure 5.5 Seasonal variation in specific conductance in the Salmon River andCoghian Creek 102Figure 5.6 Temporal variation in low-flow specific conductance in the Salmon R.and Coghlan Creek: comparison between 1974 and 1991-93 104Figure 5.7 Temporal variation in low-flow chloride in the Salmon R. and CoghlanCreek: comparison between 1974 and 1991-93 107Figure 5.8 Temporal variation in high-flow chloride in the Salmon R. and CoghlanCreek: comparison between 1974-75 and 1992 108Figure 5.9 Temporal variation in low-flow total phosphorus in the Salmon R. andCoghian Creek: comparison between 1974 and 1992-93 109Figure 5.10 Temporal variation in high-flow total phosphorus in the Salmon R. andCoghian Creek: comparison between 1974-75 and 1992 110Figure 5.11 Spatial variation in nitrate-N in the Salmon River and Coghlan Creekduring summer low-flow conditions 115Figure 5.12 Spatial variation in nitrate-N in the Salmon River in relation to thesurficial deposits 117Figure 5.13 Seasonal variation in nitrate-N in the Salmon River and Coghlan Creek . . . 119Figure 5.14 Temporal variation in low-flow nitrate-N in Salmon R. and CoghlanCreek: comparison between 1974 and 1991-1993 121Figure 5.15 Temporal variation in high-flow nitrate-N in Salmon R. and CoghlanCreek: comparison between 1974-75 and 1991-93 124Figure 5.16 Location of the sampling stations on the Salmon River represented in theEQUIS and SEAM databases 126Figure 5.17 Temporal variation in nitrate-N at five sampling stations in the SalmonRiver 127xivFigure 5.18 Relationship between discharge at 72 Ave (08MH090) and nitrate-N atfive sampling stations in the Salmon River 128Figure 5.19 Spatial variability in nitrate-N in well water in the Salmon Riverwatershed and Hopington Aquifer, from wells tested between 1970 and1993 134Figure 5.20 Location of the twelve water wells in the Hopington Aquifer monitoredregularly by Environment Canada from 1974 to 1981 136Figure 5.21 Seasonal and temporal variation in nitrate-N in the Hopington Aquifer,from monitored wells east of the Salmon River 137Figure 5.22 Seasonal and temporal variation in nitrate-N in the Hopington Aquifer,from monitored wells west of the Salmon River 138Figure 5.23 Temporal variation in nitrate-N in the Hopington Aquifer, from monitoredwells east of the Salmon River 144Figure 5.24 Temporal variation in nitrate-N in the Hopington Aquifer, from monitoredwells west of the Salmon River 146Figure 5.25 Temporal variation in nitrate-N in well water, from wells on the threedominant surficial geology types in the Salmon River watershed andHopington Aquifer 148Figure 5.26 Median well-water nitrate-N in 12 areas in the Salmon River watershed,from measurements made in 1972-1982 and 1983-1993. A. mediannitrate-N. B. median nitrate-N, with range indicated by error bars 150Figure 5.27 Comparison of nitrate-N in surface water during low-flow conditions andin well-water, in the Salmon River watershed. The well-water nitrate-Nis the median nitrate-N value from wells tested within the contributingarea upstream of the surface-water sampling station, from measurementsmade between 1983-1993 152Figure 5.28 Location of water wells within the Salmon River watershed andHopington Aquifer 154Figure 5.29 Location of septic systems installed in the Salmon River watershed andHopington Aquifer between 1970 and 1993 156Figure 5.30 Installation of septic systems from 1970 to 1992 in the Salmon Riverwatershed, Coghian Creek sub-watershed, and the Hopington Aquifer .... 160xvFigure 5.31 Selected intensive agricultural activities in the Salmon River watershedand Hopington Aquifer, derived from the 1989 land use map (Sawickiand Runka 1990) 161Figure 5.32 Location of septic systems installed in the Salmon River watershed andHopington Aquifer between 1970 and 1993 in relation to the delineatedcontributing areas 169Figure 5.33 Relationships between residential land use and nitrate-N in surface waterduring low-flow conditions in the Salmon River watershed. The measureof residential land use is the number of septic systems installed between1970 and 1993 171Figure 5.34 Relationships between agricultural land use and nitrate-N in surface waterduring low-flow conditions in the Salmon River watershed. The measureof agricultural land use is the number of hectares of selected agriculturalactivities 176Figure 5.35 Relationships between residential and agricultural land use and nitrate-Nin ground water in the Salmon River watershed. The well-water nitrate-Nis the median nitrate-N value from wells tested within each contributingarea, from measurements made between 1983-1993 181Figure 5.36 Selected agricultural and residential land use activities on the HopingtonAquifer 186xviLIST OF ABBREVIATIONSNamesALR Agriculture Land ReserveALC Agriculture Land CommissionBC British ColumbiaBCAA British Columbia Assessment AuthorityBCGS British Columbia Geological SurveyCFVHU Central Fraser Valley Health UnitCGDS Computerized Groundwater Database SystemDFO Department of Fisheries and OceansFRMP Fraser River Management ProgramFRAP Fraser River Action PlanGVRD Greater Vancouver Regional DistrictGVWD Greater Vancouver Water DistrictMOAFF Ministry of Agriculture, Fisheries and FoodMOELP Ministry of Environment, Lands and ParksMOEMPR Ministry of Energy, Mines, and Petroleum ResourcesMOH Ministry of HealthNAQUADAT National Water Quality Data BankNTS National Topographic SurveySEAM System for Environmental Assessment and ManagementTRIM Terrain Resource Inventory ManagementUBC University of British ColumbiaUSEPA United States Environmental Protection AgencyWQCP Water Quality Check ProgramxviiAbbreviationsAAS atomic absorption spectroscopyICP-AES inductively-coupled plasma atomic emission spectroscopyDTM digital terrain modelESA environmentally sensitive areaGIS geographic information systemOCP Official Community PlanUnitsBP before presentCV coefficient of variationTU tritium unitSymbolsHF hydrofluoric acidHCI hydrochloric acidHNO3 nitric acidHC1O4 perchioric acidN03-N nitrate+nitrite-NTP total phosphorusxviiiACKNOWLEDGEMENTSI am grateful for the financial support provided by the NSERC (National Science andEngineering Research Council) post-graduate scholarship.Many people have provided assistance throughout the course of my stay at UBC. Iwould like to thank Dr. Hans Schreier for his support, encouragement, and unflaggingenthusiasm throughout my research project. I appreciate the advice and encouragementprovided by my committee members, Dr. Ken Hall and Dr. Les Lavkulich. Dr. Lavkulichprovided guidance and technical advice throughout my studies, especially during the timeswhen it all seemed overwhelming.Information collected by government agencies was an important component of thisstudy and I appreciate the assistance given to me by the staff and the time taken from theirbusy schedules. I extend my thanks to the staff at municipal office of the Township ofLangley, the Central Fraser Valley Health Unit, MOH, the Agriculture Land Commission,the Groundwater Section, BCMOELP, and the Surrey Branch of BCMOELP.I would like to acknowledge the assistance provided by the foLlowing people:Roxanne Beale, for her careful compilation of water quality measurements taken in 1974through 1975; Dean Watts, for introducing me to the Salmon River watershed and initiatingthe GIS work within the watershed; Carmen Heaver, for instructing me in the intricacies ofthe TerraSoft GIS computer program; Sandra Brown, for technical support and the carefulattention to detail in the preparation of the maps from TerraSoft; and the students and staffin Hans Schreier’s group who assisted with sample collection, data entry, and the GISanalysis. The students and faculty in both the Department of Soil Science and ResourceManagement and Environmental Studies have enhanced my experience at UBC.For Andrew, my partner and best friend, I thank you for your constant support,encouragement, and assistance throughout our extended stay at UBC.11. INTRODUCTIONThe natural environment and land use activities within a watershed influence thequality of the water resources. The climate, topography, geology, and biota interact andaffect the chemistry of the stream sediments, surface water, and ground water, while humanactivities often result in transport of contaminants from the land surface into water bodies.Settlement during the last 100 years has converted the land resource in the Lower FraserBasin from natural forested land to agricultural, residential and industrial land uses, resultingin deterioration of water quality in many of the region’s rivers and streams (Dorcey 1976,Hall et al. 1991, Schreier et al. 1991, Environment Canada and BC Ministry of Environment1992). As the Lower Fraser Basin population continues to grow, from 1.9 to an estimated2.9 million people by the year 2020 (GVRD Development Services 1992), there will beincreasing demands on the aquatic resources. The purpose of this thesis is to examine thecumulative effect of small-scale land-use change on water quality in the Salmon Riverwatershed in the Lower Fraser Basin.Water contaminants from agricultural and urban land uses originate from eitherpoint or diffuse (nonpoint) sources (OECD 1986, Hagen 1990a, Meybeck and Helmer 1992).Due to its widespread origins, nonpoint-source pollution is much more difficult to identifyand control. Similarly, the cumulative effects of small changes in land use on the quality ofwater resources are difficult to measure, as is the contribution of each of the causativeagents. The presence of contaminants in a water body can have environmental, social, andeconomic costs, because they can negatively affect the aquatic ecology and result in thewater being unsuitable for a given use.Agricultural land-use activities can result in an increased input of sediments to2wetlands and watercourses, nutrient enrichment and pesticide contamination of surface andground water, increased temperature of surface waters, higher loads of pathogenicorganisms, and salinization of water bodies. An increased sediment load can change theplant productivity in water bodies by reducing light penetration, increase nutrient loadings,and thereby damage fish and wildlife habitat. Toxic or undesirable substances, such asnutrients, pathogenic organisms, and synthetic organic compounds, including pesticides, canbe adsorbed onto sediments. Addition of nutrients to water bodies, particularly nitrogen andphosphorus, can stimulate algal blooms and deplete the oxygen supply, resulting in fish kills(Anderson 1990, Hagen 1990b, Cooper 1993).Urban areas use watercourses as disposal sites for stormwater runoff, domestic andindustrial wastewater (either with or without treatment systems), and combined sanitary andstormwater sewer overflows. Atmospheric contaminants are also generated in the urbancentres and enter water bodies through wet and dry deposition. Urban stormwater runoffmay contain elevated concentrations of suspended solids, heavy metals, pathogenicorganisms, toxic organic compounds, and deicing salts. In addition to these substances,discharge of domestic and industrial wastewater may contain compounds that produce a highbiochemical oxygen demand and result in oxygen depletion in the watercourse. The highpercentage of impermeable surfaces in the urban environment results in high rates of surfacerunoff that is directed into the natural drainage system by storm sewerage, causing rapid anddirect transport of contaminants to watercourses and faster flood response that may result inincreased rates of streambank and streambed erosion (Anderson 1990, Meybeck and Helmer1992).The watershed is a useful land unit for examination of land-water interactions, since3the land activities upstream can potentially affect downstream water quality. A watershed isthe land area drained by a stream, and is delineated from the topography of the land surface.The relationships between land use and water quality can be interpreted by examining thespatial pattern of these parameters within a watershed. A set of parameters, or indicators, isneeded to provide inventory information about the state of the aquatic and land resourceswithin a watershed and allow for monitoring change. Environmental indicators can providean early warning of environmental problems and identify areas of stress to ecosystems fromhuman activities (Environment Canada, Indicators Task Force 1991).GIS (geographic information system) technology is a useful, integrating tool for theanalysis of water quality indicators and land resources within a watershed. A GIS iscomposed of computer software and hardware that is organized for the collection, storage,retrieval, analysis, and display of spatial information. The data in a GIS provides a modelof the land surface that can be transformed and manipulated to examine relationshipsbetween environmental variables (Burrough 1986). Geographic features, such as areal landuse, linear stream networks, and sampling stations, are represented as polygon, line, andpoint features, respectively. With the advent of personal computers (PCs) with greater datastorage capacity and faster data retrieval and manipulation, powerful, inexpensive machinesare being used for GIS applications.GIS is increasingly being applied, often in combination with external hydrologic andwater quality models, to the study and management of water resources. GIS has been usedfor the evaluation of changes in water quality and land-use management (Walsh 1985,Osborne and Wiley 1988, Kalkhoff 1993) and to predict nonpoint-source pollutant potential(Gilliland and Baxter-Potter 1987). In combination with other spatial modelling tools, GIS4has been used to model ground water pollution potential (Halliday and Wolfe 1991), soilerosion (Sivertun 1988), surface runoff (Steube and Johnson 1990), and nonpoint-sourcepollution (Ventura and Kim 1993, Tim and Jolly 1994). The integration of spatial modelswith the ability of GIS to manipulate large spatial data sets has produced powerful tools foruse in nonpoint-source pollution control (Tim and Jolly 1994).The Salmon River watershed is a small watershed of approximately 80 km2 inLangley Township, at the rural-urban fringe of the rapidly developing Greater VancouverRegional District (GVRD). The land use activities have been, and continue to be,dominantly agricultural and rural residential uses (Crawford 1993). With the rapidlygrowing population in the GVRD there is a demand to convert agricultural land into ruralresidential and hobby farm land uses. The preservation of productive agricultural land is animportant issue in an area subject to increasing population pressure. There is also theconcern for the need to preserve natural areas, aquatic and terrestrial, both for their inherentvalue, and as recreational facilities in the increasingly urban environment of the GVRD.The desire to preserve the important salmonid spawning and rearing areas for stocks of cohosalmon, steelhead trout, and cutthroat trout is an important factor driving the interest tomanage the watershed.A comprehensive strategy to manage the Salmon River watershed is needed, toensure maintenance of the quality of the watercourse and the fisheries resource (HowardPaish & Associates 1980, Watts 1992). Issues of land use, water use, water quality, and fishproduction must be addressed in the development of a management strategy. The SalmonRiver Watershed Management Partnership, with representatives from community groups andgovernment agencies, was formed in 1993 to establish a cooperative, community-based5stewardship of the watershed.Within this wider watershed management objective, this study provides informationon water quality and identifies relationships between surficial geology, land use, and waterquality. In the 1970s, studies conducted by Hall et al. (1974) and Beale (1976) used tracemetals and nutrients as indicators of water and sediment quality. Building on the workconducted in the 1970s, this study uses the indicators trace metals in sediments, and nitrateand total phosphorus in water, to document water quality changes during the past 20 years.A watershed approach, combined with GIS techniques, is used to examine theinterrelationships between water quality, surficial geology, and land-use activities.1.1 Study goalThe goal of this study is to identify spatial and temporal relationships betweensurface and ground water quality and land-use activities in the Salmon River watershed,using Geographic Information System (GIS) techniques.1.2 Objectives• To evaluate spatial and temporal relationships in the concentration of trace metal in thefine-fraction of streambed sediments (as an indicator of the quality of the aquaticenvironment) in the Salmon River;• To describe spatial trends in surface and ground water chemistry, and compare themwith the 1970s studies;• To examine the temporal changes in surface and ground water chemistry between 1974and 1992;• To quantify land use activities, including residential development and intensiveagricultural operations, that could potentially affect water quality; and• To identify spatial relationships between land use, surficial materials, and water qualityusing GIS.672. INDICATORS OF WATER QUAUTYA large number of measures of the water column, sediment, and biota may be usedto characterize the quality of an aquatic environment. These include the amount ofsuspended sediment; the concentrations of inorganic chemical constituents, plant nutrients,man-made organic chemicals; the presence of pathogenic organisms; the concentrations oftoxic substances in the biota; and the community structure of primary and secondaryproducers. It is usually not feasible or financially possible to completely characterize awater body, therefore indicators are needed that measure the effect of human activities andprovide an early warning of environmental problems. The indicators chosen must bedefendable scientifically, reflect the seasonal variability observed in the environment, permitevaluation of historic trends in environmental quality, and measure the resilience of thesystem to change. No single indicator will entirely characterize water quality problemswithin a watercourse, therefore a variety of indicators that reflect the effect of different land-use activities should be chosen.Because of the interest in examining temporal trends in water quality, the selection ofwater quality indicators for this study was constrained by the availability of existing historicwater-quality measurements on the Salmon River. The indicators chosen are trace metals insediments, and nitrate, chloride, and total phosphorus in water. Trace metals were used asan indicator of urban and residential activities. The nutrients and chloride are indicators ofboth agricultural activities and human waste disposal in residential areas. All theseindicators are naturally occurring substances that fluctuate seasonally with changes in riverdischarge, therefore detection of concentrations above natural levels is required to establish alinkage between land use and water quality. This chapter will outline the properties of the8indicators used in this study, their natural and anthropogenic sources, and the advantagesand disadvantages of using these variables as indicators.2.1 Trace metals in sedimentsTrace metals may be introduced into the riverine environment through the naturalprocesses of weathering and erosion or from human activities. The increase in use of metalsand metal-containing materials has resulted in widespread dispersal of metals in theenvironment. Routes of dispersal of metals include fine particle emissions to theatmosphere from the smelting or combustion of metal containing materials, emissions fromvehicles, disposal of effluent from industrial and urban activities into waterways, and thedeposition of metal-containing materials onto the land surface. Introduction of metals intothe aquatic environment can result from direct discharge of effluent, wet and dryatmospheric deposition, ground-water leachate, or surface runoff (Fergusson 1990, Meybeckand Helmer 1992, Thomas and Meybeck 1992).Trace metals are particle-reactive and therefore the concentration of trace metalsassociated with suspended or bed sediments can be many times greater than the dissolvedmetals in the water column (Salomons and Förstner 1984, Elder 1988, Horowitz 1988).During periods of low flow, particulate material accumulates in the streambed sediments,where they are both a sink of trace metals and also a potential source of metals to the watercolumn, and they can be remobilized during high flow conditions. Since sediments canaccumulate in the streambed over a period of time, they represent an integration of particlereactive materials introduced in the aquatic environment. Due to the higher concentration ofmetals in sediments compared to the water column and the accumulation of sediment in the9river bed, assessment of the trace metal concentrations in sediments is considered suitablefor monitoring metal pollution in aquatic environments (de Groot et a!. 1982, Salomons andFörstner 1984).In the natural environment, the trace metal concentration in sediments reflects theabundance of these metals in the rocks and unconsolidated materials within the catchmentarea of the river system. Some examples of concentrations of trace metals in rocks anduncontaminated soils and sediments are listed in Tables 2.1 and 2.2 and indicate a highdegree of natural variability in trace metals amongst geologic materials. Sediments areheterogeneous in nature and are made up of a number of constituents, including insolubleprimary and secondary minerals, and products of biological activity, such as insolubleorganic matter and skeletal material that may be high in carbonates, silica, or phosphates(Martin et al. 1987, Campbell et a!. 1988, Fergusson 1990). The natural concentration oftrace metals in the total sediment reflects the concentrations of trace metals in the individualconstituents of the sediment.As well as the geological source of the sediments, trace metal levels are also affectedby both the physical and chemical properties of the sediment. The fine fraction of thesediments, due to both the physical and mineralogical properties, has a much greater amountof particle reactive surface area and thus tends to be an accumulator of trace metals. Theorganic matter, iron and manganese oxides, and clay mineral constituents within thesediments are the most important accumulators of trace metals and these chemicalconstituents tend to be found in the finer size fraction (e.g. Ackermann 1980, Salomons andFörstner 1984, Horowitz 1988).The effect of particle size has been cited by many investigators as the most important10Trace element concentrations in rocks and soils.Material Zn Cr Cu Co Ni Pb Mn Femg kg1 %ROCKSultramafic’ 58 2980 42 110 2000 14 1040mafic3 100 200 90 35 150 3 1500granitica 52 4 13 1 0.5 24 400limestonea 20 11 5.5 0.1 7 5.7 620sandstone3 30 35 30 0.3 9 10 460shales/clays’ 120 39 39 19 68 23 850crustal average3 75 100 50 20 80 14 950average composition of 127 71 32 13 49 16 720 3.59surficial rocksSOILS, mean and rangew Id i c 1 s-9oo 84 25.8 12 33.7 29.2 760 3.2or soi S 0.9-1500 <1-390 0.3-200 0.1-1520 <1-888 0.01-21soils from various 3 5-770 54 13-24 7.9 22 32 437 05-5countriesd, 1-1500 1-300 0.1-275 0.2-450 3-189 7-9200Canadian soils ll-00 43 22 21 20 20(uncultivated)e, 10-100 5-50 5-50 5-50 5-50US soils and surficial 60 54 25 9.1 19 19 550inaterials <5-2900 1-2000 <1-700 <3-70 <5-700 <10-700 <2-7000 2.6 .01->10a. Alloway and Ayres (1993) using data from Krauskopf (1967) and Rose et al. (1979)b. compiled by Martin and Meybeck (1979) from various sourcesc. Ure and Berrow (1982)d. Kabata-Pendias and Pendias (1992)e. McKeague and Wolnetz (1980)f. Shacklette and Boerngen (1984)Table 2.111Table 2.2 Background concentrations of trace elements in sediments.MATERIAL Zn Cr Cu Co Ni Pb Mn Femg kg’ %MARINE SEDIMENTSdeep-sea clays’ 120 100 200 55 200 100 6000 6.0LAKE SEDIMENTSlacustrine sedimentsb 118 62 45 16 66 34 760 4.34Burnaby Lake, sub-recentsediments (80-560 cm)c 41 10 0.1 10 <0.6 161 7.26median and range 19-68 4-20 <0.08-13 3-23 <0.6-18 105-202 0.28-1.6RIVER SUSPENI)ED SEDIMENTSMackenzie R. suspended 126 8.5 42 14 22 24 600 3.65particulatesYukon River’ 115 416 40 1270 6.3world average river 350 100 100 20 90 150 1050 4.8suspended matter’STREAMBED SEDIMENTSsub-recent Rhine 115 47 51 16 46 30 960 3.23sedimentsupper Illinois R. basin,low order streams, <63- 100 56 23 26 27 2.9dam fractione median (241) (74) (35) (35) (53) (4.1)(90 percentile)Western US sediments, 49-510 20-210 0-110 9-52<63-pm fractionNTS 92G Vancouver map 322sheet (<177-pm fraction) 48 44 26 8 7 53-2100 2.02mean and range 10-1000 12-518 2-415 1-32 1-165 1-140 0.4-10.5a. compiled by Martin and Meybeck (1979) from various sourcesb. Förstner (1978), cited in Salomons and Förstner (1984)c. unpublished data, D.W. McCallum (pers. conun.)d. Förstner and Muller (1974) cited in Salomons and Förstner (1984)e. Colman and Sanzolone (1992)f. Severson et a!. (1987) cited in Combest (1991)g. BCMOEMPR (1990).12factor affecting trace metal levels in sediments (Salomons and Förstner 1984, Horowitz1988). Differences in metal concentrations within a site or between sites may simply be theresult of different particle size distributions between the samples. The effect of particle sizeis particularly important in a river system where a wide range of particle sizes may exist ina cross section of the river or in different reaches of the river from the headwaters to themouth. A number of methods have been proposed to take into account the effect of particlesize and permit comparison of samples with different particle-size distributions. Thesemethods include: separation of particle size fractions (de Groot et al. 1982, Salomons andFörstner 1984), extrapolation from regression curves of the fine fraction metal load (eg.Ackermann 1980, Ackermann et al. 1983, Salomons and Förstner 1984), correction for inertor organic material, and comparison of metal concentration with reference elements, i.e.elements unlikely to be enriched from human activities (e.g. Trefry and Presley 1976,Goldberg et al. 1979, Ackermann 1980, Schropp et a!. 1990, Pardue et a!. 1992).Salomons and FOrstner (1984) suggest that separation of the <63-sum fraction is thesimplest and most reliable method to avoid misinterpretation of results due to particle sizedifferences. The separation of the <63-sum fraction was chosen for this study because thisfraction can be obtained by sieving which is less time consuming than separation by settlingof finer fractions of <2 or <20-1um size. Other reasons for using the <63-4um fraction arethat trace metals tend to be found in the clay/silt particles, this fraction is similar to the sizefraction of material carried in suspension, and many studies have been done using this sizefraction (Salomons and Förstner 1984). If particulates flocculate or trace metals are presentas surface coatings on larger particle sizes, analysis of the finer fraction may exclude a largeportion of the contaminant load (Krumgalz 1989, Moore et a!. 1989). By evaluating only a13portion of the sediment it does not allow for estimation of the total contaminant load of thesediment (Wilber and Hunter 1979). Despite these limitations, this method is widely usedand allows for comparison with other studies.The use of a reference or conservative element, one that is unlikely to be enrichedfrom human activity and tends to be found in the finer fraction, has been suggested byseveral investigators as a method of distinguishing between anthropogenic affects on tracemetal concentrations and differences in metal concentration due to natural variation inparticle size. The advantage of this method is that particle size fractionation, a timeconsuming procedure that could result in sample contamination, is not necessary. Fe is usedas a reference element in this study to facilitate comparison of samples taken in the 1970s,when the size fraction collected varied between <177-aiim and <2-mm in size, with the <63-dum size fraction samples collected in 1991.In the use of trace metals as an indicator of anthropogenic contamination, theselection of a suitable trace metal concentration representing a background level isimportant. Investigators have suggested the use of the trace metal concentration in a worldaverage for soils, in shale, in fossil aquatic sediments, and in sediments that are upstream ofany influence from human land-use activities within the river system of interest (Salomonsand Förstner 1984, Luoma 1990, Thomas and Meybeck 1992). The use of a world averagemay not be suitable if the natural mineralogical content of the sediments is unusual, andoften there are no portions of the river system unaffected by human activities from whichuncontaminated sediments may be collected. In this study background levels were obtainedby sampling the three dominant surficial material types within the watershed. The sampleswere collected from the C-horizon to avoid possible surface contamination from land-use14activities.Due to the high natural variability in trace metal concentrations, and the difficulty ofobtaining a suitable background level, a high degree of enrichment is needed before elevatedtrace metal concentrations can be attributed to human activities. Therefore, theconcentration of trace metals in sediments is limited in its usefulness as an early warningindicator of human influence. However, trace metals in sediments can be used to monitorthe long term trends in trace metal contamination of water bodies.Preliminary guidelines or criteria have been set for trace metals in freshwatersediments in some provinces and US states and have been summarized by Hall (1992). Theproposed guidelines listed in Table 2.3 are for unfractionated sediments. Given the widerange of particle size distributions found in freshwater sediments, sediment quality criteriaare needed that take into account the particle size of the sediments.2.2 Chlorides in surface and ground waterChloride occurs naturally in freshwater systems, although the concentrations aregenerally low. The concentration of chloride in surface water varies with the climaticregion, with humid regions containing low concentrations usually less than 10 mg U1, andoften less than 1 mg U’ (Health and Welfare Canada 1989). The range of dissolvedchloride in surface waters was measured in the Canadian Pacific Region to vary from <0.1to 27 mg U’, using NAQUADAT data from 1980-81 (Health and Welfare Canada 1989).The guideline for chloride in drinking water is 250 mg U’ because NaCl in solution at thisconcentration produces a salty taste (Stednick 1991). Water sources with chloride up to15Table 2.3 Guidelines and criteria for trace elements in freshwater sediments.Cr Cu Ni Pb MnZn Fecompiled by Hall (1992) from various sources.2000 mg L’ are used in some areas of the world without adverse effects on human health(Sawyer and McCarty 1978).Entry of chloride into freshwater systems occurs from the weathering and leaching ofrocks and soils, recycling from the ocean through wet and dry deposition, and from hotsprings (Feth 1981, Hem 1985). Sources of chloride from rocks include igneous rocks withchloride bearing minerals or sedimentary rocks, particularly the evaporites. Porous rocksthat are submerged by the sea after their formation or marine sediments that have beenincompletely leached may also be a natural source of chloride to freshwater systems (Hem1985). Rivers that flow into the ocean have dense salt water wedges that mix with the freshwater above (Sawyer and McCarty 1978).Chloride salts, predominantly sodium chloride, potassium chloride, and calciummgkg1 %USEPA Guidelines for classification of Great Lakes Harbour sedimentsnon-polluted <90 <25 <25 <20 <40 <300 <1.7moderately polluted 90-200 25-75 25-50 20-50 40-60 300-500 1.7-2.5heavily polluted >200 >75 >50 >50 >60 >500 >2.5Ontario provincial sediment quality guidelineslowest effect 120 26 16 16 31 460 2severe effect 820 110 110 75 250 1100 4Ontario guidelines for dredged material100 25 25 25 50 1.0Wisconsin sediment quality criteria100 100 100 100 5016chloride, are used extensively in industry and agriculture. Sodium and calcium chloride areused for the removal of snow and ice, whereas potassium chloride is used primarily for theproduction of fertilizers. Chloride salts are also used in a variety of manufacturingindustries (Health and Welfare Canada 1989). The entry of chloride into surface and groundwater from human activities can result from road salting, sewage discharge, irrigationdrainage, and discharge of industrial effluent into watercourses (Health and Welfare Canada1989). The amount of chlorides in human excreta averages about 6 g per day, and increasesthe chloride concentration in sewage about 15 mg U1 above the carrier water (Sawyer andMcCarty 1978).The geochemical cycling of chloride is principally by physical processes. In general,chloride ions are not involved in oxidation-reduction reactions, do not form insoluble salts,are not significantly adsorbed onto mineral surfaces, and do not play a major role inbiochemical pathways. This anion is easily leached from the soil and permeable rock (Hem1985). The chloride ion has been used widely as a tracer, particularly in the pollution ofwells, due to the above properties, and it can be easily measured (Sawyer and McCarty1978). Since chloride salts are widely used and present in industrial, human, and animalwaste, this highly mobile anion is considered a useful indicator of human influence on waterquality.2.3 Total phosphorus in surface waterThe concentration of phosphorus in natural surface waters usually ranges from 0.005to 0.020 mg U1 (Chapman and Kimstach 1992) and is controlled by complex processes ofbiological uptake and chemical solubility. Phosphorus occurs in freshwater systems as the17fully oxidized phosphate form, with the anions H2PO; and HP042the predominant inorganicanions at the normal pH range of natural waters. Phosphorus is an essential nutrient and ispresent in soil and water bodies as inorganic and organic forms, in both the dissolved andparticulate fractions.The cultural eutrophication of surface waters resulting from nutrient additions is ofconcern because of stimulated algal and aquatic plant growth. Phosphorus is often thelimiting nutrient in freshwater systems. Although carbon, nitrogen, and phosphorus are allessential plant nutrients, carbon and nitrogen are often not the limiting nutrients because ofthe exchange of C and N from the atmosphere, and ability of some blue green algae to fixatmospheric nitrogen (Sharpley et at. 1994). Accelerated eutrophication of surface wateroften limits the use of the surface water for drinking, industry, recreation, and fisheriespurposes.The phosphorus that is available as an algal nutrient is of concern for eutrophication.Phosphorus as dissolved orthophosphate is readily available; however, some of thephosphorus associated with the organic and particulate forms may also be released asorthophosphate and be available for algal uptake. The mechanisms of phosphorus fluxbetween the various soluble and particulate forms is not completely understood and there isno chemical analytical procedure that measures the algal available phosphorus (Hegemannand Keenan 1985). The measurement of phosphorus fractions in water samples isoperationally defined, and involves the conversion of the phosphorus to orthophosphatefollowed by a colorimetric determination of the dissolved orthophosphate (APHA 1989).Fractions that are often measured in unfiltered water samples include the “reactivephosphorus” that can be measured with a colorimetric test without preliminary hydrolysis or18digestion; “acid-hydrolyzable phosphorus”, where dissolved and particulate condensedphosphates are converted to dissolved orthophosphate by acid hydrolysis at boiling watertemperature, and “total phosphorus”, where phosphorus forms are converted to dissolvedorthophosphate by oxidation with a strong oxidizing acid (APHA 1989). Both inorganic andorganic forms may be measured in the acid-hydrolyzable and total phosphorus fractions.Anthropogenic sources of phosphorus to aquatic systems include point-sourcedischarges of urban and industrial wastewater and nonpoint-sources of phosphorus fromurban and agricultural runoff and erosion (Chapman and Kimstach 1992). With the controlof point-source discharges of phosphorus in wastewater, agricultural activities remain a largecontributor to the phosphorus load in many river systems (USEPA 1990 cited in Sharpley eta!. 1994). Phosphorus is applied as a plant nutrient to the land surface as inorganicfertilizer, animal manure, or sewage sludge. Since phosphorus is strongly adsorbed in thesoil profile, the transport of phosphorus to aquatic systems from the land surface asdissolved and particulate-P is predominantly from surface runoff and erosion (Baker andLaflen 1983, Sharpley et a!. 1994). The longer-term application of phosphorus in excess ofcrop requirements results in a buildup of phosphorus at the soil surface and increases thepotential for loss from runoff and erosion (Sharpley et a!. 1994). Although losses ofphosphorus to surface water can occur from row-crop production, the greatest potential foraccelerated eutrophication is in areas of intensive animal production where manure isapplied to the land surface in excess of crop requirements (Duda and Finan 1983, cited inSharpley et at. 1994). Manure application rates are often determined based on the cropsnitrogen requirements. However, the ratio of N:P in manure is usually about four whereascrops requirements for N:P is usually about 8, resulting in excess phosphorus fertilization19(Sharpley et al. 1994).The objectives for the concentration of total phosphorus in fresh water are set fromthe perspective of protection of the water body from eutrophication. In British Columbia theobjective for total phosphorus for BC lakes is usually between <0.010 to <0.015 mg U1, butfor some lakes the objective is as high as <0.075 mg U1 (Nordin et a!. 1990, BCMOE1991). No objective for total phosphorus has been set for any BC river systems. ThePrairie Provinces Water Board has set an objective of 0.05 mg L1 for total phosphorus,whereas the Ontario guideline for rivers is 0.03 mg U’ (Environment Canada, IndicatorsTask Force 1991). A review of State water quality standards in the USA lists criteria valuesranging from 0.007 to 0.5 mg U’ for total phosphorus in freshwater, with most values forriver systems set at or below 0.1 mg U1 (USEPA 1980a). The objective set for lakes andreservoirs is usually lower than for rivers, between 0.02 and 0.05 mg U’ (USEPA 1980a),due to the longer residence time of lake water.Total phosphorus has been used extensively as an indicator of phosphorus enrichmentin surface waters. Although used as a surrogate of bioavailable phosphorus, it also includesthe less available forms of phosphorus such as the insoluble primary and secondaryphosphorus containing minerals. The interpretation of the measures of total phosphorus isalso complicated by the suspended sediment load at the time of sampling. Since the amountof total phosphorus present in native, undisturbed soil usually ranges from about 500 to 800mg kg’ (Stevenson 1986), the contribution of the suspended load to the total phosphoruscontent of the water column can be substantial.202.4 Nitrate in surface and ground waterThe biogeochemical cycling of nitrogen is controlled by complex interactions ofbiophysical processes, and pools of this element are found in the atmosphere, hydrosphere,lithosphere, and biosphere. Transformations of nitrogen in soil are controlled by microbiallymediated oxidation-reduction reactions. Under aerobic conditions, the bacterialdecomposition of organic matter in soil produces ammonium that is rapidly converted tonitrate (NO3-) by nitrifying bacteria. The small size and high mobility of the nitrate anionfacilitate its uptake by plants; however, it is also very susceptible to leaching from therooting zone. In undisturbed terrestrial ecosystems nitrogen is conserved; very little nitrogenis leached below the rooting zone into the deeper ground water (Keeney 1986). Inagricultural ecosystems, where nitrogen sources are added to the crop to promote growth,inappropriate timing or excess application of nitrogen above the crops requirements canresult in leaching of nitrate below the rooting zone and result in contamination of groundand surface water supplies.Nitrate can occur in freshwater systems from both natural and anthropogenic sources.Natural sources of nitrogen which can be transformed to nitrate and transported to waterbodies include nitrogen in geologic deposits, biologically fixed nitrogen, and nitrogenpresent in biomass or soil organic matter (Keeney 1986, Henry and Meneley 1993). Themineralization of soil nitrogen and nitrate losses to surface and ground water have beenrecorded after forest disturbance by fire or harvest and from native grasslands afterploughing (Likens et al. 1978, Vitousek and Mellillo 1979, Keeney 1989). Anthropogenicsources of nitrogen include organic wastes and inorganic fertilizer applied to the landsurface, disposal of human and household wastes below the land surface in septic systems,21and wet and dry deposition onto the land surface of nitrogen from atmosphericcontamination. The rate of addition nitrogen from an average household septic system isestimated at 33 kg yr1 and since this nitrogen is applied under the soil surface wherenitrogen sinks are limited, these disposal fields can locally be a significant source of nitrateto the ground water (Keeney 1989). Fertilizer-N application in excess of 150 kg hat yr1are commonly applied in intensive cropping systems (Anderson 1990, Henry and Meneley1993). Organic materials such a farm manures, sewage sludge, and food processing wastesare desirable as nitrogen fertilizers because the nitrogen becomes slowly available to theplants as the organic material is mineralized. However, these wastes are often applied to thesoil in excess of plant requirements and nitrate leaching can occur.Elevated nitrate in surface and ground water is of concern because of potentialnutrient enrichment of waters, the economic cost of loss of fertilizer-N, and the adverseeffect of high nitrate concentrations in drinking water on human health (Follett and Walker1989). The reported health effects of ingestion of water with a high nitrate concentrationinclude methaemoglobinaemia and cancer. The drinking water standard in both Canada andthe USA is set at 10 mg U1 of nitrogen as nitrate (equivalent to 45 mg U’ nitrate), since nocases of infant methaemoglobinaemia have been reported for concentrations of N03-N inwater supplies below this level (Walton 1941, cited in Follett and Walker 1989).Methaemoglobinaemia, also known as blue baby syndrome, occurs when nitrate is reducedto nitrite in the oral cavity and stomach, and is absorbed into the blood stream where itoxidizes haemoglobin to methaemoglobin. The methaemoglobin does not transport oxygenand death can result. Other concerns about high concentrations of nitrate in water suppliesinclude concerns about stomach cancer, particularly from the formation of the carcinogenic22N-nitroso compounds from nitrate. However, no reliable estimate can be made of the riskof human cancer from nitrate in drinking water and the possible formation of n-nitrosocompounds (Follett and Walker 1989).Nitrate is the most common chemical contaminant of ground water (Spalding andExner 1993). Widespread and increasing concentrations of nitrate contamination of surfaceand ground water have been well documented in North America and western Europe (e.g.OECD 1986, Hallberg 1989, Moody 1990, Spalding and Exner 1993). Research world-widehas shown that agriculture is the most extensive source of nitrate-N to surface and groundwater (Hallberg 1989). Higher concentrations of nitrate in ground water have beenassociated with intensive agricultural operations that include rain-fed row cropping, irrigatedgrain production, and animal feeding and holding operations (Keeney 1986, Hallberg 1989).Other more localized sources of contamination include land disposal of food processingwastes, on-site sewage disposal systems, and excessively fertilized turfgrass (Keeney 1986,Cogger 1988, Petrovic 1990). Examination of potential sources of nitrate is required at thelocal level to determine the cause of elevated nitrate levels.Nitrate transport to surface and ground water occurs through subsurface runoff anddeeper base flow (Baker and Laflen 1983). The natural levels of surface water nitrate-N instreams are usually below 0.1 to 0.2 mg U1 (Omernik 1977, Chapman and Kimstach 1992).Madison and Brunett (1985, cited in Spalding and Exner 1993) used 3 mg U’ as thebackground level for nitrate-N in US aquifers; however, this level may be too high becausemany samples collected across the US do not have detectable levels of nitrate-N (Spaldingand Exner 1993). The National Pesticide Survey conducted in the US estimated that 2.4percent of rural domestic wells and 1.2 percent of community water systems have nitrate-N23in excess of the maximum contaminant level of 10 mg U1 (USEPA 1990, cited in Spaldingand Exner 1993). Within BC, the areas of the Lower Fraser Valley (Langley/Abbotsford),Okanagan Valley (Osoyoos), and the Kettle River Valley (Grand Forks) have reported casesof nitrate-N in well water above the Canadian drinking water guideline of 10 mg U1(Freeze et al. 1993, Michael Wei pers. comm.).The climate and geology of a region, as well as the management of the supply ofnitrogen to the soil, affect the susceptibility of ground and surface waters to contaminationfrom nitrate. Aquifers that are particularly susceptible to contamination from surface land-use activities, are the shallow, unconfined sand and gravel aquifers in humid regions,overlain by thin, permeable soils, and with the water table near the land surface (Cherry1987, Moody 1990).Since nitrate is a highly mobile negatively charged ion, it is possible to use it as anearly warning system for the detection of contamination from human activities. Onedisadvantage of using nitrate as an indicator is that its concentration can change rapidly dueto changes in biological activity, and it also varies seasonally. This variability often makesit difficult to differentiate seasonal trends from longer-term temporal trends (Hallberg 1989).Despite these limitations, nitrate is relatively easy and inexpensive to measure, and has beenmonitored in many locations for a number of years, and therefore provides a historical basisfor comparison.243. THE SALMON RIVER WATERSHED STuJW AREA3.1 Physical featuresThe Salmon River Watershed is in the Fraser Lowland of southwestern BritishColumbia, within the Township of Langley and District of Matsqui municipalities (Figure3.1). The watershed is approximately 8020 ha in size, with flat to rolling terrain andelevations ranging from sea level to about 140 m. The Salmon River flows northwest andempties into the Fraser River west of Fort Langley (Figure 3.2). The Salmon River and itstributaries are deeply incised in the middle reaches of the watershed, with slopes rangingfrom 15 to 60 percent (Luttmerding 1980).3.1.1 ClimateThe inshore marine climate of the Lower Fraser is strongly influenced by the CoastMountains to the north of the Fraser River and the Cascade mountains to the east(Armstrong 1984). Due to the influence of these mountain ranges there is an increase inprecipitation from southwest to northeast, with rainfall varying from about 1100 mm yr1 inWhite Rock to about 1880 mm1 in Chilliwack (Dakin 1993). The average annualprecipitation measured at Langley City to the west of the Salmon River watershed is 1454mm yr’, from 1951-1980 data. Based on the precipitation isohyetal maps published byDakin (1993), the precipitation within the Salmon River watershed varies from about 1400mm yr1 in the southeast to about 1700 mm yr1 near Fort Langley at the Fraser River.About 70 percent of the total annual precipitation falls between October and March and anaverage of only 6 percent of total precipitation in July and August (Armstrong 1984, Dayton& Knight et at. 1994). The flow regime of the Salmon River is controlled by this pattern ofFigure3.1LocationoftheSalmonRiverwatershedwithintheLowerFraserbasin.Source:adaptedfromBeale(1976)1%.))1CANADAUSA05—------—---------.S0510biIOw,.l.rJChilli.acbI_Ok.Figure3.2LocationoftheSalmonRiverandmajor tributaries,andresidentialareaswithinthewatershedintheTownshipofLangley.MINSTEMSTREsMNETWORKROADNETWORK—WATERSHEDBOUNDARY10123Km(%j27seasonal precipitation.3.1.2 Surficial geologyThe landscape and deposits of the Lower Fraser basin are the result of the lastglaciation (Late Wisconsin) and post-glacial periods (Holocene) (Armstrong 1984). Withinthe Salmon River watershed, the surficial geology is made up entirely of unconsolidateddeposits with no rock outcrops at or near the surface (Armstrong 1984). The properties ofthese surficial deposits are important in defining the suitability of different land-useactivities, influencing the streamfiow, streambed material, and water chemistry of the surfaceand ground water in the watershed (Hall and Wiens 1976).The eastern portion of the watershed, containing the headwaters of the Salmon River,is composed of glacial-marine deposits of the Fort Langley Formation, dated fromapproximately 13000 to 11700 years before present (BP). These glacial-marine deposits arecomposed of marine sediments with some intermixed ice-rafted rock debris of glacial origin(Armstrong 1980, Armstrong 1984). The glacial-marine deposits of the Fort Langleyformation have some of the highest clay contents of these sediments within the FraserLowland (Armstrong 1984), with a moderately-fine texture and clay content of 20 to 40percent (Luttmerding 1980). These deposits have been overridden by ice (Armstrong 1984),producing gently undulating to gently rolling terrain with slopes from 0.5 to 9 percent(Luttmerding 1980).To the west of the glacial marine deposits and in the central portion of the SalmonRiver watershed are deposits of glacial-fluvial origin. These surficial deposits are of mixedorigin, including outwash sand and gravel and ice-contact gravel and sand containing till28lenses of clasts of glacial-marine material, both from the Sumas Drift (about 11,000 yearsBP), and proglacial deltaic sands and gravels of the Fort Langley Formation (Armstrong andHicock 1980). The deposits of the Sumas Drift, with inclusions of till or glacial marinematerials are more heterogeneous in nature than the gravels and sands of the Fort LangleyFormation. These coarse-textured sands and gravels, with the Sumas Drift deposits in thesouth and the Fort Langley Formation sands and gravels to the north, make up theHopington Aquifer, also called the Salmon River Aquifer (Dakin 1993). These glacialfluvial deposits are usually overlain by 20- to more than 50-cm-thick aeolian deposits ofHolocene origin and have produced a gently undulating to undulating terrain with slopesfrom 0.5 to 2 percent (Luttmerding 1980).The northwestern portion of the watershed is composed primarily of Capilanosediments of marine and glacial-marine origin that were not overridden by Sumas ice. Thisarea encompasses a portion of the Milner flats with nearly level terrain, and the gentlysloping to moderately rolling (0.5 to 15 percent slope) terrain to the north and west(Luttmerding 1980). In the northern section of the watershed, the Salmon River flowsthrough recent (Holocene) fluvial deposits of the Fraser River. This former distributarychannel of the Fraser is composed of laterally and vertically accreted deposits of moderatelyfine to fine textured materials (Armstrong and Hicock 1980, Luttmerding 1980). FortLangley, on the southern shore of the Fraser River, is located on a small upland areacomposed of proglacial deltaic gravel and sand from the Sumas Drift (Armstrong andHicock 1980).A map of the surficial geology of the Salmon River watershed, generalized from thesoils maps (Luttmerding 1980), is illustrated in Figure 3.3. Four general categories wereFigure3.3SurficialgeologyoftheSalmonRiverwatershed,generalizedfromthe1:25000soilsmap(Luttmerding1980).SN%N\VHopingtonForanexplanationofthemapping,seesection4.5.3.SURFICILDEPOSITSE121ALLUVIUMf:Z!L1F1RINEE:6LRCILMARINEGLCILDUTWSHE6LICILTILL—WATERSHEDBDUNDiiRY10123K30were used and include glacial marine deposits to the east and north-west (42 percent of thewatershed), glacial outwash deposits (30 percent), marine deposits (19 percent) and alluvialdeposits adjacent to the Fraser River (8 percent).3.1.3 Ground water resourcesGround water recharge occurs mainly from precipitation during winter months fromOctober through March, when 70 to 75 percent of the rainfall occurs and evaporation andevapotranspiration rates are low (Piteau Associates 1991). The three major aquifers withinthe watershed include the Fort Langley Aquifer, the Hopington Aquifer, and the AldergroveAquifer. Order of magnitude estimates of recharge rates for the three aquifers are 6 yearsfor the Fort Langley Aquifer, 8 years for the Hopington Aquifer, and 6.5 years for theAldergrove Aquifer (Dakin 1993). Tritiurn dating of the water in two wells in theHopington Aquifer was reported by Sather (1988). Low levels of tritium were found in bothground water samples, with less than 6 tritium units (TU) from a 235 ft well and 10 TU in a140 ft well depth (Sather 1988). Large quantities of man-made tritium were released intothe atmosphere from the testing of thermonuclear bombs between 1952 and 1962. Thenatural tritium content of precipitation prior to 1952 was in the range of 5-20 TU. Waterwith less than 5 to 10 TU must have entered the ground-water zone prior to 1953 (Freezeand Cherry 1979). Based on these estimates, the age of the ground water in the HopingtonAquifer probably varies from about 8 to greater than 40 years old.The Fort Langley Aquifer is an unconfined aquifer located south of Fort Langley,with portions of this aquifer well flushed by exfiltration from the Salmon River (Dakin1993). The Township of Langley operates a 20-rn-deep well within this aquifer for31municipal water supply. The Aldergrove aquifer is centred around the community ofAldergrove and extends about 35 km2. It is a mostly confined aquifer composed of thin (5to 20 m) sand and gravel deposits, capped in most of the area by silty clay sediments. Mostof the ground-water flow is towards the south and east, but there is also evidence thatground-water from this aquifer also flows northwest into the Salmon River and northeastinto Nathan Creek (Piteau Associates 1991).The Hopington Aquifer is the largest aquifer within the watershed, and covers anarea of about 40 km2 (Figure 3.3). Discharge areas for this aquifer include the upperNicomekl River Valley, tributaries of the Salmon River, West Creek, and Nathan Creek. Inmany areas of the aquifer, local perched water table aquifers are present (Dakin 1993). Thisunconfined aquifer is made up of sand and gravel deposits that can be up to 30 m thick(Halstead 1986), with an average depth of 8 m (Dakin 1993). For the Fraser Lowland area,fence diagrams of the hydrostratigraphic units, i.e. surficial deposits with the same hydraulicproperties, have been constructed based on well driller logs (Halstead 1986). From thesefence diagrams it is apparent that the sand and gravel deposits of the Hopington Aquifer arethinner south of 48 Ave and are often overlain by a less permeable layer that may containclay, stony clay and clay-like deposits (Haistead 1986). These less permeable layers protectthe ground water from contamination by activities on the land surface.The water level of the Hopington Aquifer has been monitored by the BC Ministry ofEnvironment, Lands and Parks (BCMOELP), Groundwater Section, at three locations sincethe 1960s. The water table level measured at the three observation wells from 1982 to 1990is illustrated in Figure 3.4. The drop in water level at Well No. 10 has been attributed tothe installation of high capacity irrigation wells in the Sperling area of North Central32Figure 3.4 The changes in depth to water table in three observation wells in theSalmon River Watershed, from 1982 to 1990. The approximate location ofthe wells are indicated by the Street intersections.29.0U)>U)C0I.0)ci).0ci,U)I-.a)EC0.a)029.6Well No. 5 (216 St and 80 Aye)1992 1699 1994 1965 1909 199V 1999 1 999 1990Source: data from Groundwater Section, BCMOELP.vN-30.230. 831.432.0±6 016 .8.616.419.220 . 0±4.0±4.8±5. 616 . 417. 2±8.01992 lOSS 1964 1099 1965 1SS 1969 1969 1990. Well No. 10 (240 St and 70 Aye)AOA900900 Iø.Y 19071952 1869 1904 1965 1995Well No. 7 (240 St and 32 Aye)I I33Langley (Dakin 1993).3.1.4 Drainage networkThe headwaters of the Salmon River flow through undulating terrain composed ofglacial-marine deposits. The upper reaches of the river have gently sloping stream banks,often with little stream-side vegetation, and the water is often stagnant during the summermonths. In the middle reaches, the river is deeply incised into sand and gravel deposits,with very steep stream banks. The middle reaches of the river, with medium-sized gravelsubstrate and extensive stream-side vegetation, are important salmonid spawning and rearinghabitat. Coghian Creek, the main tributary of the Salmon River, flows into the SalmonRiver about 14 km upstream from the Fraser River and also contains important salmonidhabitat (Watts 1992).Human modifications to the natural drainage network of the Salmon River includethe installation of the flood gate and pump house at the mouth of the river, the constructionof bridges and culverts for road crossings, and the addition of ditches to promote landdrainage. The pump station and flood gates were installed in 1949 and prevent flooding ofthe floodplain area surrounding Fort Langley during spring freshet of the Fraser River. Thepump is usually in operation from late March to July, but also operates when the FraserRiver water levels are high. Culverts have been used extensively throughout the watershedat road crossings and the major culverts have been mapped by Watts (1992). Increasedstream velocity and waterfalls can occur as a result of culvert installation and can act as abarrier to fish migration (Chilibeck et aL 1992, Watts 1992). At many road crossings,runoff from the street is channelled directly into the river through the use of drainage pipes.34The runoff may negatively affect the water quality of the river by carrying oils, metals, andparticulate material from the road surface into the river system.3.1.5 River dischargeThe discharge of the Salmon River has been measured by Environment Canada,Inland Waters Branch, at the gauging station located at the 72 Ave. bridge (station 6 thisstudy, Figure 4.2) since the early 1960s. The records are continuous from about 1970 to thepresent. The mean annual discharge, based on records from 1970 to 1993, is 1.4 m3s’. Theannual hydrograph for the Salmon River is calculated from the daily dischargemeasurements from 1970 to 1993 and is illustrated in Figure 3.5. High flow periods occurbetween the middle of October and the end of March, corresponding to the periods of highprecipitation, and low flow conditions between July and early September. The mean dailydischarge during July and August is 0.17 m3s1, compared with a mean daily discharge of2.69 m3s1 during the high flow periods from November through February. During the 24-year period of record, the minimum average daily discharge recorded was 0.10 m3s’ onOctober 1 and 2, 1975 in contrast to a maximum value of 39.3 m3s1 recorded on February24, 1986. The daily discharge measurements for 1991 to 1993, during which water andsediment samples were collected for this study, are presented in Figure Surface water qualityA number of intensive sampling programs were undertaken in the 1970s tocharacterize the water quality of the Fraser River, the Lower Fraser River, and its tributaries.Westwater Research Centre at UBC sampled the Lower Fraser River, including the Salmon35Figure 3.5 Hydrograph of the Salmon River at 72 Ave crossing at thegauge station 08MH090, based on discharge measurementsfrom 1970-1993. Discharge data was obtained fromEnvironment Canada.100-mad mum1JJ4\r0.1 -0.01—Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Janday of year36Figure 3.6 Average daily discharge measurements for the Salmon Riverat the 72 Ave crossing at the gauge 08MH090. Trianglesindicate water quality sampling dates for this study. Dischargedata was obtained from Environment Canada.25-1991Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan25-I: rIr r F1992IJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan25-19932- 20- .-00)I I I I ICu I I I I I I I I I I.c ID- ----0 I I I I I I I IU) I I I I I I I I I I1EIJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jandate37River, in July and August 1972 (Benedict et a!. 1973) and again from February through Mayin 1973 (Hall et al. 1974). The Salmon River was sampled over a ten-month period fromJune 1974 through March 1975, with biweekly sampling during the summer months andmonthly sampling in the winter (Beale 1976). As part of an assessment of the water qualityof the Fraser River, the Salmon River was sampled regularly in 1975 by the BC provincialgovernment, Waste Management Branch (Clark et a!. 1981). These sources, as well as otherwater sampling conducted by the BC provincial government and reviewed by Swain andHolms (1985), provide important historic information on the water quality of the SalmonRiver, characterize the water chemistry, and indicate that some water quality problems wereapparent at that time. The Canadian drinking water guidelines for selected water qualityparameters, and the provisional water quality objectives set by the BC Water ManagementBranch, MOELP, are listed in Table 3.1.The water in the Salmon River can be characterized as moderately soft water,indicated by water hardness measures with an average of about 40 mg L1 near the mouth ofthe river. The measured pH of the water ranges from 6.3 to 8, with most measurementsabove 6.8, within the guideline proposed for the protection of aquatic life. The water has anadequate buffering capacity for acidity, with an average total alkalinity ranging from 28 at232 Ave to 40 mg U1 CaCO3 at the mouth. In general, specific conductance and chlorideconcentrations increase toward the mouth of the river, with values usually ranging from 5.0to 20 mS m for specific conductance and from about 1 to 25 mg U1 for chloride (Benedictet a!. 1973, Hall et a!. 1974, Beale 1976, Clark et a!. 1981, Swain and Holms 1985).In a study of the water quality of 17 tributaries to the Lower Fraser River duringJuly and August 1974, the water quality in the four agricultural tributaries, including the38Table 3.1 Canadian water quality guidelines and BC water quality objectives forselected water quality parameters and water uses. All units are mg U1except where indicated.dissolved oxygenTotal dissolvedsolidsIronManganeseChlorideAmmonia, totalTotal coliform 10MPN/100 mLFecal coliformMPN/100 mLcold-water biota:9.5 early life stages6.5 other life stages2.2 at pH 6.5,temperature 10°C0.06100.0 avoid concentrationsthat stimulate weedgrowthno more than 10%of samples in 30 dperiod, no more thantwo consecutivesamplesoutside dilution zoneof effluent6 all times8 alevin, larvae, fish notin eye-to-hatch fish11.2 eggs in eye-to-hatchstage0.03 unionized, maximumper sample0.007 unionized, mean valuefor 5 weekly samplesduring 30 d period0.02 continuous exposure ofsalmonids40 protection of aquaticlife4000 maximum, forirrigation & livestockwatering1000 geometric mean over30 d period, forirrigationCanadian Water Quality Guidelines B.C. ObjectivesDrinking Livestock Freshwater Freshwater Aquatic LifeWater Aquatic Life in the Salmon R.Water Quality concentration comments concentr. commentsParameterpH (log scale) 6.5-8.5 6.5-9.0 6.5-8.530005000.30.052500.3Nitrite 1.0 10.0Nitrite+Nitrate 10.0presence0Data sources: Swain and Holms (1985), Health and Welfare Canada (1989)39Salmon River, Sumas River, Chilliwack Creek, and Nicomen Slough, generally ranked thelowest for 13 water quality parameters. Near the mouth of the Salmon River, during thisperiod of low flow conditions, measurements of high biochemical oxygen demand(maximum of >22 mg U1), low dissolved oxygen (minimum of 0.4 mg U1), total and fecalcoliforms up to 2200 and 1300 MPN/100 mL, respectively, were observed, indicatingorganic enrichment. The highest concentrations of Kjeldahl-N, nitrate-N, and totalphosphorus observed in the tributaries were measured in the Salmon River, with maximumvalues of 1.7, 2.9, and 0.26 mg U1, respectively, for the three measures of nutrients,indicating nutrient enrichment in the river system. Concentrations of total Fe and Mn werealso high, with values ranging from 0.3 to 0.45 mg U’ for Mn and 0.14 to 0.42 mg U’ forFe (Benedict et al. 1973).Other water quality studies of the Salmon River, over a wider range of flowconditions, indicate similar values in water quality parameters, with the exception ofdissolved oxygen and total phosphorus. Measures of dissolved oxygen reported by otherauthors range from 6.7 to 13.8 mg U’, with most of the measurements above 8.0 mg U1, thewater quality objective (Table 3.1) for the protection of fish in all but the eye-to-hatch stage(Hall et al. 1974, Beale 1976, Clark et a!. 1981, Swain and Holms 1985). Beale (1976)found some values of total phosphorus of up to 3.0 mg U’ and thus reported phosphorus tobe a potential water quality problem, whereas most other measurements reported forphosphorus are below 0.1 mg U’. Swain and Hoims (1985) stated the phosphorus valuesmeasured in the Salmon River indicate that it may be a low productivity river. The highconcentrations of nitrate-N measured by Benedict et a!. (1973) were observed in all otherstudies. Elevated nitrate-N was attributed to agriculture and septic systems by Hall and40Wiens (1976), and to agricultural activities by Swain and Hoims (1985). Levels of Fe andMn above Canadian drinking water guidelines were noted by Hall et at. (1974) and Beale(1976).The high levels of coliform bacteria and nitrate-N, as well as periodic low dissolvedoxygen levels and high total phosphorus, illustrate the effect of land-use activities on thewater quality within the watershed in the 1970s and early 1980s.3.1.7 Ground water qualityMost of the studies of ground-water quality in the Lower Fraser basin have focusedon the contaminant nitrate-N in unconfined, highly permeable sand and gravel aquifers.Concentrations of nitrate in ground water above the Canadian drinking water guideline havebeen reported in the Brookswood Aquifer in southeast Surrey and southwest LangleyTownship, the Hopington Aquifer in Langley Township, and Abbotsford Aquifer in Langley,Matsqui and Abbotsford municipalities (Kerr 1984, Halstead 1986, Kwong 1986, Sather1988, Zimmerman 1990, Liebscher et at. 1992, Gartner-Lee Limited 1992 and 1993). Theelevated nitrate-N has been attributed to leaching from stockpiled manure, disposal of animalwaste on the land surface, use of fertilizer-N, and disposal of human wastes in septicsystems (Halstead 1986, Kwong 1986, Kohut et at. 1989, Gartner-Lee Limited 1992). Lessdata has been collected on the presence of pesticides in groundwater; however, pesticides inwell water has been reported in the Lower Fraser basin by Liebscher et al. (1992) andGartner-Lee Limited (1992, 1993).The presence of nitrate-N in the Hopington Aquifer has been reported by Kerr(1984), Kwong (1986), Zimmerman (1990), and Gartner-Lee Limited (1992, 1993). Ground41water monitoring in the Fraser Lowland by Environment Canada during the 1970s wasreviewed by Kwong (1986). He identified three areas in the Hopington Aquifer withnitrate-N greater than 10 mg L1, the Canadian drinking water guideline. These areas are allwithin the area delineated by 232 St to the west and 256 St to the east and between 64 Aveto the north and around 46 Ave to the south. The maximum nitrate-N concentration duringthe period of record from about 1974 to 1981 was greater than 50 mg L1. Higherconcentrations of nitrate-N in wells less than 200 ft (61 m) suggests that the sources ofnitrate-N are from the land surface (Kwong 1986). Kerr (1984) collected thirty samples inJune, 1984 in the Hopington area of Langley, and reported that 13 percent of the sampleswere above 10 mg L1 and 57 percent had concentrations greater than 3 mg U1. Gartner-LeeLimited (1992) reviewed available data on wells tested for nitrate-N during the 1980s in theLower Fraser basin. Within the Hopington Aquifer they found high nitrate-N concentrationsin the areas identified earlier by Kerr (1984) and Kwong (1986).Some of the wells monitored by Environment Canada in the 1970s were againmonitored for nitrate-N by the Groundwater section of BCMOELP in the late 1980s to lookfor changes in the concentrations of nitrate-N. For the five wells monitored in theHopington Aquifer, Zimmerman (1990) reported that two wells showed an increasing trendand three showed a decreasing trend in nitrate-N. These studies illustrate that nitrate-N hasbeen detected in the Hopington Aquifer since the early 1970s and the problem persiststoday.423.2 Biological features3.2.1 Fish resourcesThe Salmon River contains important salmonid stocks of coho salmon(Oncorhynchus kisutch), steelhead (Oncorhynchus mykiss) and cutthroat trout (Oncorhynchusclarki clarki). In the middle reaches of the Salmon River and in its main tributary, CoghlanCreek, the stream morphology, the medium size gravel substrate and extensive stream-sidevegetation, provide excellent spawning and rearing habitat for salmonids. A detailedinventory of the salmonid habitat in the middle reaches of the Salmon River is provided byWatts (1992). The water quantity and quality are also important parameters that affect thevalue of the Salmon River as salmonid habitat.Within the Lower Fraser basin, the Salmon River is considered the most productivestream of its size for coho salmon. The 1981 to 1985 spawning escapement figures for cohoindicate that the Salmon River watershed produces about five percent of the coho in theFraser River tributaries below Hope (Farwell et al. 1987). The fish stocks in the SalmonRiver watershed are also important from the perspective of maintenance of the within-species genetic diversity of the Fraser River basin.Twelve other species of fish also use the Salmon River for at least part of their lifecycle (Watts 1992). The lower reaches of the river are used by juvenile chinook(Oncorhynchus tshawytscha) and chum (Oncorhynchus keta) salmon as a feeding and restingarea during their downstream migration to the ocean in late May (Barry Chilibeck pers.comm.). The Salish sucker (Catostomus sp.) is a species found in the headwaters of theSalmon River upstream of 256 Ave. The Salish sucker population in the Salmon river isthought to be declining, possibly as a result of the effect of human activities on their habitat,43such as addition of culverts, water withdrawal and removal of stream-side vegetation. Thisspecies is found in only four stream systems within BC and is considered to be endangered(Inglis et al. 1992).3.2.2 Natural areasMany natural areas remain within the Salmon River watershed, including wetlands,forested ravines and second-growth deciduous and mixed forests (Cook et al. 1993). Watts(1992) estimated that 25 percent of the land area within the watershed is in an“undeveloped” land use category, which includes both natural areas and idle land with noperceived management. Although a substantial percentage of the land cover within thewatershed, these natural areas are fragmented by surrounding residential and agriculturalland uses or are linear units adjacent to the stream network. The preservation andrestoration of natural vegetation along the stream network is important for the protection offish habitat and water quality.3.3 Cultural features3.3.1 PopulationThe population of the Township of Langley has grown from about 14600 in 1971 toover 66000 in 1991, with recent growth rates of over four percent per year (Crawford 1993,Corporation of the Township of Langley 1994). The population estimate for the Townshipis about 75000 for 1994 (Corporation of the Township of Langley 1994). The 1991population of the watershed was about 16000 people, estimated from 1991 Statistics Canadacensus data (Statistics Canada 1991). This figure is only approximate, since in many cases44the census tract boundaries include areas outside the watershed.The two principal urban areas within the Salmon River watershed are Fort Langleyand the Salmon River Uplands (Figure 3.7). The population estimate for 1994 for FortLangley is 2600 and 6230 for the Salmon River Uplands. Fort Langley is not growingrapidly, with an average of a 0.7 percent growth per year for 1986 to 1991 compared to a4.3 percent growth rate for the Salmon River Uplands over the same time period. Theurbanized areas of Fort Langley and the Salmon River Uplands represents about 55 percentof the population of the watershed (Statistics Canada 1991, Corporation of the Township ofLangley 1994).3.3.2 Land useThe first permanent European settlement on the Lower Mainland of British Columbiawas the establishment of Fort Langley at Derby in 1827 (Crawford 1993). At the time ofsettlement, the Salmon River watershed was predominantly covered by coniferous forest,with grassland and deciduous forest in the Fraser Floodplain area around the current locationof Fort Langley (North and Teversham 1984). Logging and farming activities spreadthroughout the Township of Langley in the late 1800s, with agricultural activities theeconomic base for Langley’s population (Crawford 1993).The current land uses within the Salmon River watershed are dominantly agriculture,undeveloped non-commercial forested lands and residential activities (Watts 1992). Watts(1992) produced a generalized land use map for the Salmon River watershed using eightcategories of land use (Figure 3.8) based on the land-use mapping of the Township ofLangley by Sawicki and Runka (1990). Ninety percent of the watershed is covered by the45Figure 3.756 AYELocation of the urban areas within the Township of Langley and theSalmon River watershed.urban areaSalmon Riverwatershed boundary102 AVE96 AVE88 AVE80 AVE72 AVE64 AVE48 AVEcnOFLANGLEY16 AVE8 AVESource: adapted from the Corporation of the Township of Langley (1994)Figure 3.8 The 1989-1990 land-use map of the Salmon River watershed.zIICr)cccw>IIcz0-jccC!)[KI— )-(j F:312z —. hiI—U J 12 :3 Z °-[K U .J Z CLD -. a: 12 F- -.12F- F- 12 —. —. (K F- 1 I-.J Z J LI F- 12 :3 F-2 U U (K C.) QF- a:Ci 12 > U a: cn -. U— -. U Z [K Z F- (K[K Ct) 12 E F— (n C.)CD U 12 X Ua: (K 2 C.) LI F —. z46aDIaaCI)CI- Ib 12 [ It) • N CI C)O LI) N 0 -a. I12— C) N D - N 12 C’) IC)C’) C’) N 12 C’) N0 CD C’) 1L12 U) 0 — — N 12 112- N 112-ja:I-12I-LIU,DCz-JC00 HU ltDSource: Watts (1992)47land use map, and illustrates 50 percent agriculture, 25 percent undeveloped land, and sevenpercent residential land use. The remaining 8 percent of the watershed has recreational,institutional, commercial, transportation, and gravel extraction land-use activities. Theagricultural activities are distributed throughout the watershed, whereas the concentration ofresidential land use is in Fort Langley and in the middle reaches of the watershed (Watts1992).Due to its diversity of soils and proximity to the urban market of Vancouver, a widevariety of agricultural crops is produced within the Township of Langley. Livestockoperations include dairy, beef, swine, sheep, veal, rabbits, poultry and mink production.Small fruit crops include strawberries, raspberries, blackberries, blueberries, and cranberries.Vegetables, flowers, nursery stock, mushrooms, and turf farming are other agriculturalactivities within the Township (Corporation of the Township of Langley 1993). Most ofthese agricultural activities are represented within the Salmon River watershed (Sawicki andRunka 1990).Land Use ChangeSeveral land use maps have been compiled for the Salmon River watershed atdifferent times and can be used to assess general trends in land-use activities. Land use inthe Salmon River watershed has been summarized into eight land use categories for threetime periods, 1971, 1979-1980, and 1989-1990, in Table 3.2. It is difficult to comparebetween 1971 and the other time periods due to differences in map scale, watershedboundaries used, and comparability of land-use categories; however; some general trends canbe observed. Over the past twenty years, agriculture is the dominant land use, followed by48undeveloped lands and residential activity.The apparent increase in agriculture and decrease in residential land use from 1971and 1979 may be due to different categorization of the low density rural residential activitiesin the production of the two maps. The rural residential population has grown with theincrease in small hobby farm development within the Township (Crawford 1993). Aminimum lot size of 1.7 ha (4.2 acres) has resulted in many small properties, with37 percent of rural lots within the Township smaller than this size. The allocation of someof the rural residential or small hobby farms to the agriculture category in the 1979 mapcould explain this discrepancy. These differences illustrate some of the difficulties incomparing land-use maps that were prepared at different times for different purposes.The largest change in land use between 1971 and 1989-90 is the decrease inTable 3.2 The proportion of land use activities in the Salmon River Watershed in1971, 1979-1980, and 1989-90.datemappingscalesource19711:50 000Beale (1976)%1979-19801:25 000Watts (1992)%1989-19901:10 000Watts (1992)%Land UseTypeagriculture 47.8 59 50residential 9.5 4 7undeveloped 37.7 21 25commercial/industrial 0.1 0.5 1extraction 0.2 0.5 0.5transport/utility 1.1 1 1.5institution 0.1 2 2recreation 0.5 2 3not mapped 10 10Total 97 100 10049undeveloped land, including vacant and non-commercial forested lands, from about 38percent to about 25 percent of the watershed (Table 3.2). This reduction in land with noperceived use illustrates the intensification of land use activities within the watershed overthe past two decades.3.3.3 Water and sewerage servicesResidents within the Salmon River watershed have on-site sewage disposal systemsfor wastewater treatment and disposal and rely primarily on domestic wells for water supply.Only a very small section of the watershed in the northwest is serviced with sewers. Theresidential areas of Fort Langley and Forest Knolls receive water from the GreaterVancouver Water District (GVWD), supplemented by water from municipal wells within thewatershed. Sixty-two percent of the water supplied to these areas is from the GVWDsupplemented by thirty-eight percent municipal well water (Joe Chaylt pers. comm.). A fewareas within the Salmon River Uplands receive water from community wells that service asmall number of houses or from municipal wells in the Aldergrove Aquifer (Joe Chaylt pers.comm.).3.3.4 Discharge permitsTrinity Western University, immediately south of the trans-Canada highway, has apermit to discharge treated domestic sewage into the river and began discharge in early1974. This permit allows discharge during months of high flow from October 15 throughMay 15 (Swain and Hoims 1985). Other waste discharge permits within the watershed arefor disposal of waste to the ground.503.3.5 Water licensesApproximately 50 water licences have been issued for the removal of water from theSalmon River and its tributaries, based on the records maintained at BCMOELP WaterResources Branch, Surrey, as of December 1990. Most of the water is used for irrigationpurposes during the months from April through September, with about 280 acre-feet yr1(344500 m3 yr’) used for irrigation compared to 71600 gallons day1 (118800 m3 yr1)allocated for domestic, conservation, and land improvement (such as ponds) purposes. Ofthe water licensed for removal for irrigation purposes, more than 75 percent was allocatedprior to 1960 and 94 percent allocated prior to 1970. Howard Paish & Associates (1980)estimated that during summer low flow conditions the water withdrawal in the middlereaches of the Salmon River, between 256 Ave and 72 St, may be up to 25 percent of thetotal flow, and up to 50 percent of the flow is withdrawn from Coghlan Creek.3.3.6 Land use zoningAgriculture Land ReserveThe Agriculture Land Commission Act was passed in 1973 to reduce theencroachment of urban land use onto productive agricultural lands. This Act takesprecedence over municipal plans and zoning-bylaws. When the Agriculture Land Reserve(ALR) was created most of the land within Salmon River watershed was designated asagricultural land, with the exception of Fort Langley, and a small area of land near thewestern watershed boundary near 232 St and a rural residential area within the SalmonRiver Uplands (Beale 1976). As of 1991, of the approximately 1070 ha that have beenexcluded from the ALR within the Salmon River watershed, about 680 ha are within the51Salmon River Uplands. The majority of these exclusions (80 percent) were the result ofreviews undertaken by the Agriculture Land Commission (ALC) in co-operation with TheCorporation of the Township of Langley in 1976 and 1979 to minimize the conflictsbetween farms and existing residential areas and to allow for infihl areas between thesubdivisions. The purpose of the reviews was to establish a rational and defendableboundary between urban and agricultural land uses (Anonymous 1991).In the Salmon River Uplands, built-up areas and subdivisions had occurred prior tothe creation of the ALR, with the result that a number of parcels existed that were 0.5 to 2.0acres (0.2 to 0.8 ha) below the 5 acre (2.0 ha) minimum lot size required within the ALR.Of the 750 acres (304 ha) excluded from the ALR in 1977, about 620 acres (251 ha) hadalready been subdivided, with the remaining 130 acres (53 ha) consisting of infill areasbetween existing small-lot subdivision areas. As a result of the 1976 review, it wasrecommended that the zoning within the Salmon River Uplands be amended to require aminimum parcel size of 15-20 acres in those areas where the lands are suited to berryproduction, to encourage viable farm units. The exclusion of an additional 95.5 ha in theSalmon River Uplands in 1981, as a result of the 1979 review, illustrates the continuingdemand for rural residential development in this area (Anonymous 1991). Although themajority of this land had very good agriculture potential, the Land Commission recognizedthe need to “fine-tune” the ALR boundaries and to clearly define the boundary betweenurban and agricultural uses.Zoning designationsThe 1966 Official Long Range Plan for the Salmon River watershed, from the Lower52Mainland Regional Planning Board, is presented in Figure 3.9 (Beale 1976) and illustratesthat areas of the watershed have been delineated as areas for urban growth for almost thirtyyears. The Salmon River Uplands area, Fort Langley, and a small portion in the NWportion of the watershed are designated as urban growth areas, with the remainder of thewatershed designated for rural, agricultural uses. The land-use zoning as of 1973 designatedthe Salmon River Uplands area as an urban area, with minimum lot size ranging from 0.5 to2.5 acres (0.2 to 1.0 ha), and most of the remaining watershed designated as rural with a 5acre (2.0 ha) minimum or 20 acre for the Salmon River floodplain area (Figure 3.9). TheLangley Official Community Plan (OCP), passed as Bylaw in 1979, delineates a largeportion of the Salmon River Uplands as a designated urban growth area (The Corporation ofthe Township of Langley 1979), although sections of this area remained in the ALR at thetime of the OCP development.The Rural Plan was undertaken by the Township of Langley in order to develop acomprehensive policy for the rural areas, to enhance the viability of the agriculturalindustries through the protection of agricultural land, and to retain the rural character of thelandscape (Crawford 1993). The land use concept provides a buffer of small farms andcountry estates, with a minimum lot size of 1.7 ha, between the urban areas to the west andthe agricultural activities (Figure 3.10). The agricultural/countryside designation proposes anincrease in the minimum lot size from 1.7 ha to 8 ha to discourage the development of nonfarm uses and to provide land use stability in agricultural areas (The Corporation of theTownship of Langley 1993, Crawford 1993). The Salmon River Uplands has not been53Figure 3.9 Land-use zoning in the early 1970’s.1973 Land-use zoning4 Symbol Description4 2 1 UR-1 Y2 acre minimum2 UR-2 1 acre minimum2 3 UR-3 2 1/2 acre or less mimimum4 4 RR-1 5 acre minimum4 4 5 RR-3 20 acre minimum(after Langley Municipality Zoning Bylaw No.3 4 i3Z,i4.i1970,1973)41966 Long range planFRASER RIVER3 Symbol Description1 URB-1 Established urban area4 2 URB-2 Developing urban area3 URB-3 Lowland rural area4 RRL-1 Upland rural areaS S5 RRL-2 Acreage rural area6 6 IND-1 Developing industrial area7 RSV-2 Institutional reserve area4 8 RSV-1 Limited use reserve area2(after Lower Mainland Regional Planning4 8Board, 1966)5______________________________________________________________2455Source: adapted from Beale (1976)54Figure 3.10 Land-use concept for the Township of Langley, from the 1993 RuralPlan.I102 AVECountryside/Agricultural96 AYECountry Estates/Small FarmsRP Regional ParkUrban Industrial Boundary88 AVEA Rural Commercial Centre• Agro-Service Centre:.::::::::::.i80 AVE——— Salmon R. watershed boundary72 AVE64 AVEh..56 AVEFFSalmon:E:.:.:.::.Li_______- ---Lds1rr48 AVE40 AVE32 AVE24 AVEIp.ciGAVEISource: adapted from the Corporation of the Township of Langley (1993)55included in the Rural Plan; however,“The Salmon River Uplands shall be maintained for rural residential and agriculturaluses. A more detailed plan will be prepared setting out policies for future growth,subdivision and agriculture in this area.”(The Corporation of the Township of Langley 1993).Most of the Salmon River Uplands is currently zoned as suburban-residential-i (SR-i), witha minimum lot size of about 0.37 ha, and a small portion zoned SR-3, with a minimum lotsize of about 0.18 ha (Corporation of the Township of Langley i987).3.3.7 Environmental management within the Salmon River watershedThe federal, provincial, and municipal governments have legislative authority forenvironmental protection and management within the Salmon River watershed. The federalDepartment of Fisheries and Oceans (DFO) is responsible for the protection of fish and fishhabitat, and the management of the Pacific salmon populations whereas the BC Ministry ofEnvironment, Lands and Parks (BCMOELP) is responsible for the management of steelhead,trout, and other freshwater species (Chilibeck et at. 1992). Other provincial watermanagement responsibilities include floodplain management, licensing of water withdrawal,and water quality monitoring and assessment. The control and regulation of landdevelopment are both a provincial and municipal responsibility. The preservation ofagricultural land is controlled by the provincial ALC, whereas the development of anOfficial Community Plan (OCP), zoning of land for designated uses, and the management ofdevelopment applications are within the jurisdiction of the municipal government.The goal of watershed management is difficult to achieve when each of thegovernment agencies is responsible for different portions of the land and water resourceswithin the Salmon River watershed. The Salmon River Watershed Management Partnership56was formed in 1993 in order to overcome the difficulties of different and overlappingjurisdictions between government agencies and to establish a cooperative, community basedstewardship of the watershed which balances economic, environmental and social needs.Current members of the partnership include the federal agencies DFO, Environment Canada,Fraser Basin Management Program (FRMP), the Fraser River Action Plan (FRAP); theprovincial agencies of MOELP and Ministry of Agriculture Fisheries and Food (MOAFF),Township of Langley; the education institutions, Kwantlen College and Westwater ResearchCentre at UBC; and the public organizations including Langley Environmental PartnersSociety, Langley Environmental Organization, the Langley Field Naturalists, and theMatsqui/Langley Soil Conservation Group. The goals of the partnership are to use anintegrated and comprehensive approach to develop a management plan and communitybased governance of the watershed.To further the goals of environmental protection and the promotion of the ruralcharacter of the Township of Langley, inventory and evaluation of environmentally sensitiveareas (ESAs) were conducted in 1992 (Cook et a!. 1993). This inventory included areas ofgeological hazard potential, ground and surface water resources, natural vegetation, wildlifeand fish habitat, and cultural features. The ESA evaluation provides summary informationon key environmental factors within the Township of Langley that should be considered forprotection, conservation, and selective management.The Salmon River Watershed is one of six watersheds selected by the FRMP asdemonstration projects to provide examples of partnerships among concerned citizens,community groups, and government agencies, working toward environmental, economic andsocial sustainability within the Fraser River Basin.574. METHODS4.1 Field methods4.1.1 Sediment sample collectionStreambed sediment samples were collected on August 21 and December 10, 1991,as representative of low flow and high flow conditions, respectively. The locations of the19 sediment stations are illustrated in Figure 4.1, with a description of their location inAppendix 1. Stations 1 through 15 were located at the sites of water sampling carried outby Beale (1976) from May 1974 through April 1975. These stations were originally chosenbased on changes in surficial geology or land-use patterns (Beale 1976, HaIl and Weins1976). These stations, plus an additional four stations (16, 17, 19, 20) were sampledbetween August 1991 and August 1992. Stations 16 and 17 are downstream from two areasof potential point source pollution, Trinity Western University and the Vancouver GameFarm. Stations 19 and 20 were chosen to give additional spatial information on CoghlanCreek, the major tributary of the Salmon River.Grab samples of the top several centimetres of bed sediments were collected using analuminum pot attached to a 2-m wooden pole. Sediments were collected, when possible,from small pools where fine sediments had accumulated. At stations where there were nosmall pools, sediments were collected from a number of locations along the river bankwhere fine sediments had collected around tree roots and rocks. The sediment collected ateach station was a composite of three to six samples collected along a 5 to 50 m stretch ofthe river. When possible, the December sample was collected from the same pool as theAugust sample. In many cases no fine sediment was present in the pool in December,illustrating the short residence time of the fine sediment in some reaches of the river. TheFigure4.1Locationofthestream-bedsediment andsurficialmaterial samplingstationswithintheSalmonRiverwatershed.•SEDIMENTSTATIONSURFICILGEOLOGYSThTIONSTREAMNETWORK—ROADNETWORK—WATERSHED8OUNDnRY1012IIISURFICIALMATERIALSGM=glacialmarineOW=glacialoutwashMA=marine0, Co591-2 kg composite sediment samples were stored in plastic bags, on ice, and then refrigeratedat 8 to 10°C in the laboratory for 1 to 5 days until the samples were sieved.4.1.2 Surflcial material sample collectionFive C-horizon soil samples were collected from each of the three dominant parentmaterial types within the watershed (Figure 4.1) during August and September, 1991. Thesesamples were used to obtain an estimate of the background levels of trace metals expectedto be found within the watershed. Samples were collected at a depth of 50-120 cm tominimize the possibility of contamination with metals of anthropogenic origin. Samples ofthe marine and glacial marine materials were collected with a soil auger. Since it was notpossible to collect samples of the coarse-textured glacial outwash materials using an auger,samples were collected from soil pits, excavated building sites, or road cuts. The soilsamples were transported to the laboratory in plastic bags and then air dried for storage untilanalysis.4.1.3 Water sample collectionGrab samples of surface water were collected from the Salmon River and itstributaries at selected locations within the watershed (Figure 4.2, Appendix 1). Watersamples from August 1991-1992 were collected at the 19 sediment sampling stations. Fouradditional stations for water quality were added in September 1993. Stations 22 and 24provided additional information on water quality in tributaries draining predominantly ruralresidential land use and stations 23 and 25 were additional stations on the Salmon River andCoghlan Creek, respectively. The water and sediment parameters measured at eachFigure4.2LocationofthesurfacewatersamplingstationswithintheSalmonRiverwatershed•SMPLINOSTATIONSTREAMNETWORKROADNETWORK—WATERSHEDSOUNDARY10123KmII61sampling date are listed in Table 4.1.Table 4.1 Water and sediment quality measurements performed at selected stationsfor the Salmon River and tributaries.Water Sampling DateWater quality parameter 21/08/91 10/12/91 05/02/92 24/08/92 29/09/93pH X X X Xconductivity X X X X XNitrate-N X X X X XChloride X X X XTotal-P X X Xtrace metals in sediments X XSample times were chosen to represent both low flow and high flow conditions.Low flow conditions were sampled August 21, 1991, August 24, 1992, and September 29,1993. High flow conditions were sampled December 10, 1991 and February 5, 1992.Water samples were collected in 250-mL, acid-washed high-density polyethylene bottles,then stored on ice and refrigerated in the laboratory until analysis.4.1.4 Surface water quality measurementsIn situ measurements were made of pH, temperature, and conductivity of the surfacewater. The pH readings were taken with a glass electrode (Fisher model 13-620-108) on aportable digital pH meter, Canlab model H5503-1. Conductivity measurements were madewith a Yellow Springs Instrument Company salinity-conductivity-temperature meter Model33, with temperature correction made at sampling time. Temperature was measured both62with a glass mercury thermometer and the Yellow Springs Model 33 meter.4.2 Laboratory analysis4.2.1 Sediment and surficial material analysisSample preparation and storageSediment samples were wet-sieved through stainless steel sieves to obtain the<63-pm fraction. River water collected at station 7 was used in order to maintain the sameionic strength solution during wet sieving. The amount of river water added to samplesduring sieving was kept to a minimum to shorten the sample drying time. First, the entiresediment sample was sieved through a 2-mm mesh to remove stones and large organicdebris, and then a sub-sample was taken and sieved through a 180-4um mesh to remove thecoarse sand fraction that could possibly abrade or damage the fine 63-pm mesh. The <180-pm fraction was then passed through the 63-pm sieve, and the resultant <63-pm fraction wasdried in glass beakers in a 60°C oven. The dried samples were then ground, using an agatemortar and pestle, to pass a 150-pm (100-mesh) sieve, and transferred to plastic containersand stored at room temperature until analysis.Surficial material samples were air dried for storage. Sub-samples were later taken,disaggregated in distilled water, then wet-sieved through stainless steel sieves to obtain the<63-pm fraction. Sample preparation and storage of the surficial material samples was asdescribed above for the river sediment samples.Sediment microwave digestionThe digestion of soil and sediment was carried out in a CEM microwave oven,63model MDS-81D, with a maximum power output at 100% of 650 watts at 2450 MHzfrequency. The samples were digested in sets of twelve, and each set included one blankthat contained only the acid mixture, one sample of the certified marine reference sediment,MESS-i, from the National Research Council of Canada, and one duplicate sediment orparent material sample. The precision and accuracy of the trace metal analyses weredetermined using duplicate digestions and certified reference material, MESS-i.The microwave digestion method used was a modification of a digestion procedureprovided by CEM for total dissolution of soil and sediment samples. A sub-sample of the<63-4um sediment fraction was oven dried at 105°C so that metal analyses could be reportedon an oven-dry weight basis. Five mL of nitric acid (HNO3)and 5 mL of deionized waterwere added to a 0.5 g sub-sample of oven-dried sediment in the digestion vessel and left topre-digest overnight at room temperature. Water was added to the acid mixture, since it hasbeen shown to improve the extraction of all elements by 20 to 30 percent (Millward andKluckner 1989). 2.5 mL of hydrochloric (HC1) and 6 mL of hydrofluoric acid (HF) werethen added to the vessels and the samples were microwaved for 3 minutes at 100% power,30 minutes at 50% power and 30 minutes at 30% power. The samples were then allowed tocool, manually vented to release any pressure build-up, then microwaved again for 3minutes at 100% power and 30 minutes at 50% power. The second microwave heating wasadded since particulate residue was visible in the digestion vessel. After cooling, thedigested sample was transferred to a 60-mL polyethylene bottle and brought up to 30.00 gweight with deionized water. Despite the long microwave times, there was still undigestedresidue on the sides and bottom of the vessel.The chemicals used were of analytical reagent grade. The HF (about 45%) was64BDH AnalarR; the 70% HNO3 acid was Baker Analyzed, and HC1 was either BDHassurance 36.5-38% or Merck Proanalysi 32%. The water used was distilled then deionizedusing a Milli-Q reagent water system.ICP atomic emission spectroscopy of sediment digestsThe element analysis of the digested sediment samples was performed by Dr.Graeme Spiers at the University of Guelph using ICP emission spectroscopy. Theinstrument used is a LECO Plasmarray, with a HF-resistant sample introduction systemmade of Teflon, and containing a sapphire injector tube in the torch and an Ebdon stylenebuliser (Graeme Spiers, pers. comm.). The emission line used for each element is listedin Appendix 2. The operational detection limits listed in Appendix 2 are the detection limitsobtained for solutions with complex matrices, such as produced from the digestion proceduredescribed above. The elements Fe, Mn, Al, Na, Ca, Mg, and Ti were measured on a 1:10dilution of the sample digest, whereas Zn, Cu, Cr, Ni, Co, Pb, Sr, Zr, P and V weremeasured using the undiluted sample digest. P and V results were obtained using anultrasonic nebuliser in the sample introduction system (Graeme Spiers, pers. comm.)Total carbon analysis of sedimentsTotal carbon was determined for the sediment and parent material samples using aLeco induction furnace analyzer, model no. 572-200 (Nelson and Sommers 1982), using asample size of 0.1 to 0.5 g.654.2.2 Water analyses for nitrate+nitrite-N, chloride, and total phosphorusDissolved nitrate+nitrite-N was analyzed using a QuickChemAE Lachat auto-analyzer, using the QuikChem Method No. 10-107-04-1-B provided by the manufacturer.With this method, nitrate is reduced to nitrite by passing the sample through a copperizedcadmium column. The nitrite (original plus reduced nitrate) is then diazotized withsulfanilamide and then coupled with N-(1-naphthyl) ethylenediamine dihydrochloride and theresulting dye read at 520 nm (Lachat Instruments 1990). Prior to analysis, the watersamples were centrifuged in a refrigerated Beckman J2-21M/E centrifuge at 10,000 rpm(15300 g at maximum radius in JA-14 rotor) to remove particulates that could potentiallyblock the tubing or reduction column of the auto-analyzer. The method outlined in theLachat manual was modified by the substitution of the microsample loop with a 20 cmsample loop, in order to improve the precision of measurement of low concentrations ofnitrate-N in the water samples. The combined nitrate-N plus nitrite-N measured in thisstudy will be referred to as N03-N in the remainder of this thesis.Dissolved chloride was analyzed using a Lachat auto-analyzer, using the QuikChemMethod No. 10-117-07-1-A provided by the manufacturer. With this method, chloride reactswith mercuric thiocyanate, releasing thiocyanate that then reacts with aqueous iron(III). Theamount of fenicyanide ion produced is determined by its absorbance at 480 nm (LachatInstruments 1990). The combined chloride colour solution for this analysis was purchasedfrom Sigma Diagnostics. The sample loops used for this method were 150 cm loop forsamples below 30 mg L’ and 20 cm sample loop for samples with chloride above 30mg U1.Total phosphorus analysis was performed at Zenon Environmental Laboratories in66Burnaby, BC. Their analysis method uses a sulfuric acid-persulfate digestion, followed by amolybdate/ascorbic acid reaction for colorimetric measurement at 885 nm.All laboratory water chemistry analyses, including chloride, nitrate+nitrite-N, andtotal phosphorus, for the September 1993 samples were performed by Zenon. Methods usedby Zenon are based upon those in “Standard Methods for the Examination of Water andWastewater”, 17th edition, APHA (1989), (Shawn Heier pers. comm.).4.3 Compilation of existing surface and ground-water quality records4.3.1 Surface water qualityExisting water quality information for the Salmon River was compiled from BCprovincial government records and from work carried out at UBC during the early 1970s(Hall et a!. 1974, Beale 1976). Environmental monitoring data for BC from about 1971 to1985 was maintained in the computer data storage and retrieval system named EQUIS andextracts of this system were stored on microfiche in 1985, when the system was retired(Clarke 1992). Selected water quality parameters for twelve stations within the watershedwere manually transcribed from the microfiche and entered into a computer database (Table4.2). The current environmental database for BC is the SEAM (System for EnvironmentalAssessment and Management) database maintained by MOELP in Victoria, and containssome water monitoring data that has been transferred from the EQUIS database (Clarke1992). A computer download from the SEAM database to dBASE IV format was made inApril 1993 for water quality information from 6 stations. These 6 stations included thewater quality measurements from the two Salmon River stations studied by Hall et at.(1974). Water chemistry data for selected variables for the 14 stations monitored by Beale67(1976) from June 1974 to March 1975 were manually entered into a computer database(Table 4.2).There is no established convention for naming surface water quality sampling sites.Stations that were sampled at different times by different agencies were matched based onthe station description, including cross streets and the latitude and longitude. The stationsthat are considered equivalent are listed in Appendix 3.Table 4.2 Sources of historic information on surface-water quality data for theSalmon River watershed.Sources of information on surface-water qualitySource Number of Date No. of days No. of station-Stations sampled sampledays*From ToEQUIS database 12 Jan 1972 Aug 1984 123 271Beale (1976) 14 Jun 1974 Mar 1975 15 189MOELP SEAMdatabase 6 Apr 1974 Aug 1992 51 58*number of station-days=number of days each station sampled, summed over all stations.4.3.2 Ground water qualityInformation on ground water quality obtained from water-well testing was obtainedfrom the Groundwater Section, MOELP, Victoria. Most of these wells were usuallysampled once, with some of these wells tested up to four times, and 12 wells with 20 to 40measurements. Information on groundwater quality was compiled from a number ofsources, listed in Table 4.3.Each data set was collected to meet specific objectives, and has associated bias in theselection of sample sites. The Environment Canada NAQUADAT (National Water Quality68Table 4.3 Sources of information on ground-water quality data in the Salmon Riverwatershed.Sources of information on ground-water qualityNumber of WellsHopingtonAquiferData Bank) database, with records from 1954 through 1986, contains records of wells thatwere selected to determine the extent of nitrate contamination in groundwater, therefore,areas of high contamination are better represented in the data set (Rod Zimmerman pers.comm.). The BC Water Quality Check Program (WQCP), with records from 1981 to thepresent, is a voluntary monitoring program that is subsidized by the BCMOELP and thewater analyses are conducted at Zenon Environmental Laboratories in Burnaby, BC. Anyprivate domestic well-owner wishing to have his/her water tested can do so at a subsidizedcost. The water is collected by the home-owner, so there is no control over how thesamples are collected, or the length of time between sample collection and analysis. TheSource Salmon R.WatershedDateSalmon R. and From ToHopingtonAquiferNAQUADAT, 97 84 144 Oct 1954 Nov 1986EnvironmentCanadaWater Quality 149 117 186 May 1981 Jul 1991Check Program(WQCP),BCMOELPSEAM database 27 34 43 Jan 1982 Feb 1993BCMOELPGroundwater 0 3 3 Nov 1987 May 1991Section FilesBCMOELPGartner-Lee 16 16 19 Mar 1975 Jan 1990(1992)69SEAM database contains groundwater monitoring records collected by the BCMOELP andcontains records for some of the same wells sampled by Environment Canada between 1973and 1981. The review of groundwater contamination in the Fraser Valley, conducted byGartner-Lee (1992), contains some Ministry of Health (MOH) records not represented in theabove data sets. The 1993 Gartner-Lee records are the samples collected as part of theFraser Valley Ground Water Monitoring Program, funded jointly by MOH and MOELP(Gartner Lee Limited 1993) (Table 4.3).4.4 Land-use mapping4.4.1 Water wellsThe Groundwater Section of BCMOELP maintains a computerized GroundwaterDatabase System (CGDS), which contains the location, depth, date of installation, lithology,and other attribute information about water-well installations. This information is providedon a voluntary basis to the MOELP by the well drillers (Michael Wei pers. comm.). Aswell as the database, the locations of the wells are maintained on paper maps. The wells areuniquely identified by the BCGS map sheet and map quadrant number, and each well withina quadrant is given a unique number. Within the Salmon River watershed, wells in theTownship of Langley are mapped on 1:5000 scale cadastral maps, whereas wells in MatsquiDistrict are mapped on 1:12000 maps. The well database and associated maps for the areawithin the Salmon River watershed were obtained from MOELP in April 1993 and wasconsidered by the staff to be up-to-date to about 1988. The attribute information is notcomplete for all the wells and wells remain in the database even if they are no longer inactive use.704.4.2 Septic systemsRecords of septic system installed in the watershed and Hopington Aquifer area sincethe early 1970s were obtained from the Central Fraser Valley Health Unit (CFVHU) inLangley. The office maintains two card catalogues with septic system applications filed bydate of application and by street address. They also archive the original application forms.The street card catalogue was used to locate applications for septic systems within theSalmon River watershed and Hopington Aquifer. Information on the street address, lot andplan numbers, file number, date of final permit issue, size of the septic tank, and size of thetile field (length of laterals), was transcribed from the paper card catalogue and entered in acomputer database. When several entries existed for an address, the record of the earliestapplication was recorded. Repairs to septic systems were not transcribed. For some of theearlier records, the date of final permit was not available and the date the permit was issuedwas used.Each septic system record was then given a unique identifier and the location wasplotted onto the address and assessment roll number maps (1:4000 scale) supplied by theTownship of Langley. When the address information was incomplete, or contained anaddress that did not match the address maps, the Master Legal Plan Maps (1:5000) wereused in conjunction with the legal parcel information, to locate the address. Recordswithout legal parcel information, and addresses that did not match the address maps, wereplotted adjacent to the address that best matched the transcribed address. The mappedlocation of the septic system onto the address map does not represent the location of thedisposal field on the property. The accuracy of the mapping is to within the propertyboundary.71An address is assigned to a lot by the Township when an application for a buildingpermit is requested. Since almost none of the Salmon River Watershed or HopingtonAquifer is serviced by sewers, all properties with addresses would have an on-site sewagedisposal system (septic system). The address maps provide a check on the quality of theseptic system database. An addressed property either had its septic system installed prior tothe early 1970s or the installation record is missing from the CFVHU card file system. Toobtain an estimate of the accuracy of the septic system database compiled from the CFVHUrecords, the number of septic systems on selected master legal map sheets, by Township andSection number, are compared with the British Columbia Assessment Authority informationon properties with on-site sewage disposal systems, reported by Dayton & Knight et aL(1994).4.4.3 AgricultureThe 1989 land use maps (Sawicki and Runka 1990) of the Township of Langley, at ascale of 1:10000, were used as the source of information on intensive agriculturaloperations. These land use maps were produced using the classification outlined in LandUse Classification in BC (Sawicki and Runka 1986). Polygons were delineated using 1984air photographs, supplemented with field mapping in August and September 1989 (Sawickiand Runka 1990). A subset of all mapped agricultural activities that is considered torepresent intensive agricultural operations was selected for this study. A list of theagriculture land use categories, and the subset considered as intensive agricultural operationsis presented in Appendix 4. Approximately 90% of the watershed was covered by theseland use maps, with the westernmost portion of the watershed and the area within Matsqui72District not covered by these land use maps (see Figure 3.8 for areas not covered by theland use mapping).4.5 GIS analysis4.5.1 TRIM base mapThe 1:20000 TRIM (Terrain Resource Inventory Management) maps were used asthe base map for this study. The Salmon River watershed and Hopington Aquifer are withinthe area covered by the TRIM files for map sheets 92G.007, 92G.008, 92G.017, and92G.018. The TRIM files were imported into the GIS program, TerraSoft, using the TRIMdata translation module of TerraSofi. The features selected for this study, from the muchlarger set of features within the TRIM files (Surveys and Resource Mapping Branch 1992),are listed in Appendix 5. These features include the road and railway network, the drainagenetwork, and the biological land cover features of marshes, swamps, and wooded areas.4.5.2 Contributing areasThe land area contributing to each surface water quality sampling station wasdelineated based on topography. Information on the topography of the watershed wasobtained from the 1:25000 NTS maps (92G/Old, 92G/02a, 92G/02g, 92G!Olh), with 10 footcontour intervals, and the digital terrain model (DTM), generated for the Township ofLangley using the spot heights in the TRIM data files. The TerraSoft DTM module wasused to create the raster elevation layer, using a 20 x 20 m raster cell size. The elevationlayer contains heights to the nearest metre, but is only accurate to ±5 metres. The 1:25000topographic maps were registered to the digital TRIM map using road intersections and the73contributing-area boundaries were digitized. Adjustments to the boundaries were madebased on the information in the elevation layer of the DTM.The delineated contributing areas are illustrated in Figure 4.3. Due to the accuracyof the topographic information, the boundaries are considered estimates based on availableinformation. In areas of the watershed where there is little difference in topography, theboundary was placed at the centre of an area of equivalent elevation. The accuracy oflocation of the contributing area boundary lines is estimated to be between 50 and 200 m.The size of each contributing area is listed in Table Surficial geologyThe surficial geology map was generalized from the 1:25000 soils maps(Luttmerding 1980), and from information in the 1:50000 Surficial Geology Maps 92G/1and 92G/2 (Armstrong 1980, Armstrong and Hicock 1980). The 1:25000 digital soils maps(Kenk et al. 1987) were translated by Agriculture Canada staff into TerraSoft format. Theywere then transformed from NAD27 to NAD83 projection for use with the TRIM base map,using the national transformation program, version 1.1 (Surveys and Resource MappingBranch 1991).The surficial geology within the Salmon River watershed was categorized into fourgeneral types: glacial outwash, glacial marine, marine, and alluvial deposits. Each soilpolygon was assigned a surficial geology type based on the parent material, as outlined inAppendix 6. Soils with parent materials of local stream deposits or organic material weregiven the surficial geology type of adjacent soil polygons. Polygons containing soilcomplexes with several parent material types, were assigned the parent material of theFigure4.3Delineationof thelandareacontributingtoeachsurfacewaterqualitystation,withintheSalmonRiverwatershed.Areasweredelineatedbasedonthetopography.14sil—.——/I..‘S...S1O/CO5’‘CO7_.4co4-‘——I•‘‘IC06‘co-._jCO2’7‘:,7•4.‘‘——IIcolk1•SIS __Ia ),‘%S02(‘...I /(SO5.S03......I,—_sLs061,‘S04/XISOOCONTRIBUTINGflREflBOUNORYSTREAMNETWORK—WATERSHEDBOUNDARY10123K,n-J75Table 4.4 Size of each contributing areaeach contributing area.and the water quality sampling station forContributing water size percent of land areaarea quality (ha) watershed upstream ofsampling samplingstation station (ha)Salmon River upper reachesSoo 650 8.1 650SOl 11 188 2.3 838S02 17 140 1.7 978S03 15 447 5.6 1425S04 10 209 2.6 209S05 09 339 4.2 1974S06 24 96 1.2 96S07 07 502 6.3 2571S08 08 91 1.1 91S09 04 169 2.1 2831Coghlan CreekCOl 12 288 3.6 288C02 21 17 0.2 304C03 20 288 3.6 592C04 19 346 4.3 938C05 25 114 1.4 1052C06 22 95 1.2 95C07 05 340 4.2 1487Salmon River Middle reachesSb 23 146 1.8 4464Sib 06 484 6.0 4948S12 03 585 7.3 585S13 16 561 7.0 6095S14 02 57 0.7 6152Davidson CreekDOl 13 124 1.5 124D02 14 279 3.5 279Salmon River lower reachesS15 01 1142 14.2 7697S16 324 4.0 8021Total 8021 100.076the surrounding polygons. Within the Salmon River watershed and Hopington Aquifer,polygons smaller than 50 ha and with a different parent material type than the surroundingpolygons, were generalized to the surficial geology type of the surrounding polygons.4.5.4 Land-use analysisInformation on land use, including the area-based information on agriculturalactivities, and the point locations of septic systems and wells, were digitized onto the1:20000 base map. The maps with land-use information were registered to the basemapusing road intersections. The area features were digitized using common boundaries ofroads, rivers, and other transportation corridors. Septic systems and wells were digitized aspoint features, each with a unique identifier assigned.Information on the location of water wells is present in both the database compiledfor ground water quality, and in the MOELP well location maps. Unfortunately, there is nounique well identifier code for water wells in BC, therefore the same well may berepresented by different identifiers in the NAQUADAT, SEAM, WQCP, and CGDS datasets. To minimize the number of wells that would be represented more than once in thedigital mapping, the well records in the different databases were matched based on addressor legal plan number and when possible, well depth. When a match occurred between arecord in the water quality database and the well location database (CGDS), the CGDS wellidentifier number and location on the 1:5000 maps was used for the digital mapping. Whenno match to the CGDS database was found, the well was mapped onto the property with thespecified address and given a unique identifier number. Since the information on addressesand well depths was incomplete for some of the well records, it is possible that some wells77are represented twice in the digital mapping.To quantify the areas of each land use type within the watershed and eachcontributing area, GIS vector overlay techniques were used. The location of the well andseptic system point features within polygons was also determined using GIS overlaytechniques (Digital Resource Systems Limited 1991).For the examination of relationships between water quality and agricultural activities,the agricultural operations were grouped into five general land-use categories. These generalland-use categories are berry operations, horse operations, poultry and fur bearers, otheranimal operations, and other agricultural activities. The grouping of the land activity codesinto these categories is listed in Appendix 4.4.6 Statistical analysisThe mean and median are used as measures of central tendency and the range as ameasure of variation for the water quality measurements. Means were used to characterizethe surface water quality measurements. For the larger data set for groundwater quality andsediment trace metal concentrations, the median was used as a measure of central tendencyto reduce the effect of unusual, and perhaps erroneously high, outlying values. Mostcomparisons between data gathered in this study and historic water quality records areexamined graphically for trends. Statistical comparisons between the data sets wereconsidered inappropriate due to the lack of a consistent sampling design between the studiesand the variation introduced from difference analytical methods.The variability observed in the concentration of trace metals in sediments isdisplayed graphically using box-and-whisker plots. The location of the median value is78Figure 4.4 Components of a box-whisker plot.outside value* 0far outside valueSource: adapted from Systat for Windows (1992).indicated by the vertical line inside the box, with the box representing the range of 50percent of the samples, 25 percent on either side of the median (Figure 4.4). The locationof the line within the box illustrates the skewness of the distribution of the samples (Sibley1987). The ends of the box are referred to as the hinges, and the absolute value of thedifference between the values of the two hinges is termed the Hspread. The whiskersextend to the range of values that are within 1.5 Hspreads measured from the hinges.Values that fall between 1.5 and 3.0 Hspreads from the hinges are plotted with an asterisk,while those outside 3.0 Hspreads of the hinges are plotted with an open circle (Systat forWindows 1992).Differences between populations of samples and relationships between variables havebeen examined in this thesis using non-parametric statistics. Non-parametric statistics areuseful when the assumption of a normal distribution for the sample population is not met79(Sprent 1989). The Mann-Whitney U test is used to estimate whether two independentsamples have been drawn from the same population, whereas the Wilcoxon matched-pairssigned-ranks test is used to test for differences between related samples (Siegal 1956). Thedegree of association between variables is tested using Spearman rank correlationcoefficients. For the statistical analyses with trace metal data, the detection limit was usedas the measured value when the metal content was below the detection limit.Mann-Whitney U tests and Wilcoxon matched-pairs signed-ranks tests for the tracemetal analyses on sediments were performed using the PC-based statistical software packageSPSS.-PC+ (Statistical Package for the Social Sciences) version 4.0. Summary statistics(median, mean, range), box-and-whisker plots, and Spearman Rank Correlation coefficientswere performed using SYSTAT for Windows, version 5.0.805. RESULTS AND DISCUSSION5.1 Trace metals in fine-textured river-bed sediments5.1.1 Variability in analytical methodologyAn assessment of some of the laboratory methods was conducted to determine thereliability of the analytical work. To obtain a measure of the accuracy and precision of themicrowave digestion technique and ICP atomic emission spectroscopy analysis, six replicateanalyses were performed using the marine sediment reference material MESS-i (NationalResearch Council of Canada i98i). The data from these replicate analyses are presented inTable 5.1. The mean concentration for the elements Al, Mn, Zn, Cr, and Cu is within thereported 95 percent confidence limits, which indicates that a complete digestion wasobtained for these elements and an accurate determination of the trace levels of theseelements can be made using this methodology. The precision of measurement of these 5elements is good, with the coefficient of variation (CV) ranging from 6 to 17 percent. Forthe element Fe the precision of measurement is good, with a 5 percent CV. However, themean value obtained is 7 percent higher than the expected mean and outside the reported 95percent confidence limits. For Co, Ni, and Pb, both the precision and accuracy ofmeasurement of these elements are poor. Co determinations were 1.5 times greater than theexpected mean while Ni and Pb were 0.5 times greater than the expected mean. The CVranges from 72 to 112 percent for these three metals. The data obtained for the otherelements analyzed in the MESS-i reference standard are also included in Table 5.1.Another measure of the precision of the laboratory technique was obtained byperforming duplicate analyses of the collected soil and sediment samples (Appendix 7). Theratio of the maximum:minimum for the two analyses provides a measure of the precision ofTable5.1Measurement oftheprecisionandaccuracyof themicrowavedigestiontechniqueusingthemarinesedimentreferencematerial,MESS-i(NRC).Sixreplicateanalyseswereperformed.ElementminimummaximummedianmeanCVexpected95%mean!meanor(%)meanconfidenceexpectedmedianwithin1mgkg’limitsmeanexpectedlimitsmgkgZn1.65E+021.90E+021.86E+021.81E+026.01.91E+021.70E÷0i0.95YCr6.OOE÷O11.03E-i-027.06E+O17.41E+0119.97.1OE+011.1OE+011.04YCu2.64E+013.23E+012.84E+O12.87E+016.92.51E+013.80E÷001.14YCo7.OOE-01*2.94E÷012.02E÷011.62E+0170.91.08E+O11.90E+0O1.50N+Ni7.OOE-01*2.70E+011.80E+011.45E+0169.32.95E+012.70E+000.49N-Pb1.20E-01*4.89E+011.39E+011.85E+01112.53.40E+016.1OE+000.55NFe3.04E+043.46E÷043.31E+043.27E÷044.93.05E+041.80E+031.07N+Mn4.69E+025.96E÷025.13E.i-025.23E+028.95.13E+022.50E+011.02YAl4.45E+047.04E+046.13E÷045.88E+0417.05.84E+042.OOE÷031.01YCa3.19E+036.78E+035.56E+035.37E+0324.14.80E+036.OOE+021.12YMg5.13E+031.11E+048.94E÷038.59E+0328.58.70E+035.OOE+020.99YP3.30E÷029.94E+024.28E+025.11E+0248.66.40E+026.OOE+010.80N-*notdetected,detectionlimitused82each metal determination. The average ratio for seven duplicate pairs analyzed is below1.25 for the elements Al, Mn, Fe, Zn, Cr, and Cu, indicating good precision of measurement(Appendix 7). The precision of measurement of Co, Ni, and Pb is poor, with an averageratio of 1.5 for Co, 2.7 for Ni, and 3.9 for Pb.5.1.2 Metal concentrations in soil parent materialsThe metal concentrations in the soil parent materials within the watershed provide ameasure of the natural background levels of metals expected to be found within the soilsand sediments. The variability in total metal concentration in the three parent materials ispresented in box-whisker plots in Figure 5.1, summary statistics are listed in Appendix 8,and the trace metal concentrations measured for each sample is listed in Appendix 9. Thevariability within a parent material type for the 9 metals (Zn, Cr, Cu, Co, Ni, Pb, Fe, Mn,and Al) is small for the glacial marine and marine parent material samples. The ratio of themaximum to minimum values is below 1.3 fold for the glacial marine samples and 1.7 foldfor the marine samples, for most elements. Pb is an exception for the marine samples, withconcentrations ranging from not detected to 15 mg kg’. For the glacial outwash samples,variability in total metal concentrations is much higher compared to the marine and glacialmarine samples, with up to an eight fold difference between maximum and minimum valuesfor some of the metals. A higher variability in the glacial outwash deposits is expectedbecause the samples exhibit a much greater range in particle sizes, indicating differentdegrees of sorting and washing in the depositional environment. More homogenousdepositional conditions occur in the marine and glacial marine environment (Armstrong1984).83Figure 5.1400300200D001000250200150100500150100500Box-whisker plots illustrating the variability in total metal concentrationsin Salmon River sediments and surficial materials. The units for all thedata presented are mg kg1 dry weight, except for carbon in percent.— o4s’0’N’ •Q c\— oe0001801601401201008050403020100\ .\— e Q O°c •\kac:3O’” — oeC200150100500s’’ •Q •Q0\ a: oeCs’ •gci p oe°84Figure 5.1 continued.150000 I I I I 7000* 6000 -00500010000040000 zw0 3000500002000I *10000I I 0 I I IkJ120000 I I 5000*1100004000100000 *00000 300080000 070000500002000600001000 -40000 I I 0 I Ici-e s’’0o •Q15 I I I*10z0 0ciJ05-0 — IIci c- oP)0Oe°85Differences between the total metal concentrations in the three parent material typesare compared using the Mann-Whitney U two-tailed test for independent samples. Themedian metal concentration is presented Table 5.2 and significant differences between pairsof parent material types are indicated with a letter. Since three two-sample comparisonsbetween parent material types were made (glacial outwash and glacial marine, glacialoutwash and marine, and glacial marine and marine) the a level was set at 0.017 rather than0.05. No significant differences were observed for either Zn, Cu, Pb, or Mn. Pb was notdetected in either the glacial outwash or glacial marine samples. Variable but detectable Pblevels were measured in the marine samples, with a median of 8.9 mg kg’ in the marinesamples. Significant differences were observed between at least one of the three pairs ofparent material types for the remaining five metals. Cr is significantly higher in the glacialoutwash compared with the marine samples; however, the magnitude of the difference issmall. Both Co and Ni are very low in the glacial outwash samples, with a median of 2.5mg kg1 for Co and 3.5 mg kg1 for Ni, and both metals are significantly lower than in theglacial marine and marine parent materials. A significantly higher concentration of Fe wasmeasured in the marine parent material than in the glacial marine material, with a 1.2 folddifference in means observed between the marine and glacial marine materials. Thedifferences in trace metal concentrations between the glacial outwash and the fine texturedmarine samples may be the result of the different depositional environments for these twosedimentary deposits. These results give an indication of the range of trace metalconcentrations found within the surficial deposits of the Salmon River watershed.86Table 5.2 Median metal content, and significant differences in metal contentbetween the three dominant surficial material types in the Salmon Riverwatershed. Pairs of the surficial material types were compared using theMann-Whitney U test.Element Glacial Outwasha Glacial marineb Marinen=5 n=5 n=5mg kg’ mg kg4 mg kg4Zn 1.12E+02 8.46E+O1 1.OOE+02Cr 1.42E+02c 1.39E+02 1.23E+02aCu 1.OOE+02 5.79E+O1 7.02E+O1Co 2.53E+OO 4.25E+Olac 317E+OlabNi 3.52E+OO 3.32E+Ola 2.99E+OlaPb 1.20E+OO 1.20E+OO 8.90E+OOFe 7.68E+04 5.19E+04c 6.18E+04”Mn 1.72E+03 6.43E+02 9.20E+02Al 7.22E÷04” 5.67E+04a 6.08E+04Ca 1.30E+04 1.33E+04 1.09E+04Mg 3.61E+03 5.86E+03c 7.54E+03bP 3.96E+O3 4.74E+02a 5.27E+02aC (%) 1.8E+OO’ 2.OEO1a 1.6E-O1letter indicates significant difference between pair at a=O.0175.1.3 Spatial variability of metal concentrations in river sedimentsThe variability in total metal concentrations in the river sediments collected duringlow flow conditions in August, 1991, and high flow conditions in December, 1991, iscompared with the three parent materials in box-whisker plots in Figure 5.1. Summarystatistics are listed in Appendix 8 and the trace metal concentrations measured in eachsample is listed in Appendix 9. Zn is more variable in the sediments than in the soil parentmaterial, while Cr is of similar variability and Cu is less variable (Figure 5.1). For Ni, Co,87and Pb, the range is greater in the sediment samples than in the soil samples, particularly forthe December sediment samples. For the major elements, Fe and P varied more in theglacial outwash samples than in the river sediments, and Mn and Al were of similarvariability. Much lower variability was observed for these major elements in the glacialmarine and marine parent material samples. The carbon concentration varied the most in theAugust sediment samples, whereas both low variability and low levels were observed in theglacial marine and marine sediment samples. These results indicate the marine and glacialmarine samples collected are more homogeneous than the river sediment samples.5.1.4 Comparison of trace metals in soils and river sedimentsTo test the null hypothesis that no trace metal enrichment has occurred in the SalmonRiver sediments (sediment metal concentrations are less than or equal to the concentrationsin the parent materials), the metal levels in August sediment samples are compared with theparent material samples using the Mann-Whitney U, one-tailed test. Since three two-samplecomparisons between the parent material types and sediment were made (glacial outwashand sediment, glacial marine and sediment, and marine and sediment) the a level was set at0.017 rather than 0.05. The sediment metal concentration must be greater than the metalconcentration in all three parent material types for the metal to be considered potentiallyenriched by human activity. The median metal concentrations for the soil and sedimentsamples is presented in Table 5.3. Metals significantly greater in the sediment as comparedwith each parent material type, are marked with a letter indicating the parent material type.Only the concentrations of Zn and Pb are significantly greater in the sediments than in theparent material types.88Table 5.3 Median metal content in the three dominant surficial material types andthe <63 1um streambed sediments in the Salmon River watershed. Metalconcentrations significantly higher in the sediments than in the surficialmaterials, are indicated with a letter. Pairs of the surficial material andsediment were compared using a one-tailed Mann-Whitney U test.Element Glacial Glacial Marinec Aug 1991Outwasha marin&’ n=5 n=19n=5 n=5mg kg’ mg kg4 mg kg4 mg kg4Zn 1. 12E+O2c 8.46E+O1c 1 .OOE+O2c 1 .54E÷O2&cCr 1.42E+02 1.39E+02 1.23E+02 1.21E+02Cu 1.OOE+02 5.79E+O1 7.02E+O1 4.1OE+O1Co 2.53E+OOd 4.25E+O1 3.17E+O1 2.63E+OlaNi 3.52E÷OOd 3.32E+O1 2.99E+O1 2.70E+OlaPb 1.20E+OOd 1.20E+OOd 8.90E+OOd 3.69E+O1Fe 7.68E+04 5.19E+04 6.18E+04 4.54E+04Mn 1.72E+03 6.43E+02d 9.20E+02 1.66E+03”Al 7.22E+04 5.67E+04 6.08E+04 6.35E+04Ca 1.30E+04 1.33E+04 1.09E+04 1.21E+04Mg 3.61E+03” 5.86E+03d 7.54E÷03 7.3OE÷O3’P 3.96E+03 4.74E+02d 52’0d 1.63E+O3C (%) 1.79 020d 3.99kletter indicates significant difference between pair at atO.017 for one-tailed testThe analytical precision of the Pb determinations is very poor, with a range of notdetected to 49 mg kg’ for the MESS-i reference sample compared to the expected value of34 mg kg’ (Table 5.1). Further analyses are required to obtain more precise determinationsof this element in the Salmon River sediments. The median values for Pb for the Augustand December sediments fall within the range of values cited for uncontaminated soils(Table 2.2 and 2.3; Ure and Berrow 1982, Kabata-Pendias and Pendias 1992). The Pb in89the sediment samples is significantly correlated with both Co and Ni; however, thesignificance of this result in unknown because of the poor accuracy of the Pb results.The precision and accuracy of the Zn determinations are high; therefore, it is unlikelythat the differences observed between the river sediments and the soil parent material typesare due to analytical error. However, the ratio between the median Zn concentration in thesediment and in the glacial outwash materials (the highest median level for Zn of the threeparent material types) is only 1.4 fold. The difference in median Zn in the soils comparedwith the sediment may be due to differences between their physico-chemical propertiesrather than due to anthropogenic additions of Zn into the river system. Zn in the riversediments is highly correlated with the carbon content (Spearman rank correlation coefficientof 0.64, significant at a=0.01, Table 5.4). The higher level of Zn in the river sediments islikely the result of the higher carbon content of the sediment samples compared to theparent material types (Figure 5.2).The trace metals concentrations measured in the <63-,um fraction of Salmon Riversediments for the elements Zn, Cr, Cu, Co, Mn, and Fe are all higher than world soilaverage (Table 2.1), whereas Ni and Pb are similar to reported soil values. It is expectedthat the fine fraction sediments will have higher concentrations of metals than inunfractionated soil samples. The concentrations of Zn, Cr, Cu, and Pb are all within thereported range reported for the <63-sum fraction from uncontaminated sediments in theWestern US (Table 2.2).Table5.4SpearmanrankcorrelationmatrixformetalconcentrationsinSalmonRiver<63umstreambedsediments.(n=38)FeMnAlZnCrCuCoNiPbCaMgPCFe1.00Mn0.55**1.00Al0.210.121.00Zn0.180.55**0.181.00Cr0.10-0.25-0.25-0.301.00Cu0.310.51**0.320.79**-0.081.00Co0.190.250.35**1.00Ni-0.24-0.180.03-0.040.38*0.130.54**1yJPb0.**0.54**1.00Ca0.38**0.12-**0.12-0.070.400.55**0.260.41*0.041.00P0.220.280.43**0.25-0.38*0.49**0.41*-**1.00C0.000.55**-0.130.64**-0.320.39*0.04-0.15-0.04-0.092-0.1180.351.00*signifintatct=0.05,**signifintata=0.010Figure5.2SpatialvariationinconcentrationoftotalzincandcarboninSalmonRiver<63jimstreambedsediments,sampledinAugust1991.Errorbarsindicaterangeofmeasurementsforsurficialmaterialcontrols.OW=glacialoutwash,GM=glacialmarine,MA=marine.C 0.0 Ca C)450--14400-•35°-300-0)250-.2 200-N::::.....OWGMMA12201951117151098746316213141waflsadwaasnuthSahronestationnumber0 H925.1.5 Seasonal variation in trace metals in river sedimentsRiver bed sediments were collected in August, 1991, after an extended period of lowflow conditions and again in December after several months of high flow conditions toexamine seasonal differences in sediment metal concentrations. The metal concentrationsand percent carbon from the two seasons are compared using the Wilcoxon matched-pairssigned-rank test, which compares the direction of differences as well as the magnitudebetween a set of paired samples (Siegal 1956). Significant differences are observed betweenthe August and December sediments for the elements Cr, Cu, Al, Mg, P, and carbon (Table5.5). The magnitude of the seasonal differences is small, as can be seen by the ratio of themedians of the two seasons. For all the metals that show significant seasonal differences,the August sediments have a higher median value than the December sediments, with theexception of Cr. These differences may be the result of small differences in the particle sizedistribution within the<63-dum fraction of the samples, since it would be expected that alarger percentage of fine silts and clays would be present during low flow conditions,compared to high flow conditions. It is also interesting to note the higher carbon andphosphorus content of the August sediment, which may be due to higher biological activityin the watercourse during the summer months.5.1.6 Temporal variation of trace metals in river sediments between 1970s and 1991Salmon River sediments were analyzed for trace metals in the summer of 1974 (25stations; Hall, unpublished data), May 1975 (12 stations; Beale 1976), and June 1977 (9stations; Bindra and Hall 1977). The sample preparation and element analysis methods usedin these studies are outlined in Appendix 10. The variability in trace metals in these93Table 5.5 Median metal content in the <63 4um streambed sediments collected inAugust and December, 1991, in the Salmon River watershed, andsignificant differences between August and December sediments. TheAugust and December sediments were compared using the Wilcoxonmatched-pairs signed-ranks test.Aug 91 Dec 91 Aug median:Elementn=19 n=19 Dec medianmg kg’ mg kg1Zn 1.54E÷02 1.51E+02 1.02Cr 1.21E+02 1.37E+02* 0.88Cu 4.1OE+01 3.92E+01* 1.05Co 2.63E+01 2.42E+01 1.09Ni 2.7E+01 2.24E+01 1.21Pb 3.69E+01 2.38E+01 1.55Fe 4.54E+04 4.78E+04 0.95Mn 1.66E+03 1.08E+03 1.53Al 6.35E+04 5.73E+04* 1.11CaMgPC(%)*signifint at a=0.05.samples are compared with 1991 sediment analyses using box-whisker plots in Figure 5.3,and summary statistics are in Appendix 11. The variability observed for the elements Zn,Cu, Ni, and Mn is similar for the 1970s and 1991 samples, whereas Cr, Co, and Fe variedmore in 1970s data sets and Pb varied more in the 1991 data. In general, the data fromBeale (1976) varied more than the 1974 data (Hall unpublished data). The larger <2 mmsize fraction was used for trace metal analysis by Beale (1976); thus more heterogeneity1.21E+04 1.18E+04 1.037.30E+03 S.95E+03* 1.231.63E÷03 9.33E+02 1.743.99 337* 1.1894between samples would be expected.The metal analysis for all three sampling times in the 1970s have been pooled andcompared to the August 1991 sediment samples using the Mann-Whitney U test forindependent samples (Table 5.6). Significant differences between the two time periods areobserved for all analyzed metals, with Zn, Cu, Co, Pb, Fe, and Mn significantly lower in the1970 analyses as compared with the 1991 samples. Since these metals are not elevatedabove background in the 1991 samples, it is most likely that the differences between the twotime periods are due to different sample preparation and analysis methodology. Inparticular, the size fraction used for sediment metal analyses varied between the studies. The1970 samples were either the <2-mm fraction (Beale 1976) or <177-dum fraction (Hall,unpublished data), (Appendix 10), compared to the <63-4um fraction used for the 1991samples. Higher metal concentrations in the finer sediment fraction has been welldocumented in the literature (e.g. Salomons and Förstner 1984, Helsel and Koltun 1986,Horowitz 1988), and these results illustrate the difficulty of making comparisons betweensediments with different particle size distributions.Normalization of trace metal values to a reference element has been suggested by anumber of authors to account for the natural variability of metals found in sediments withdifferent particle size distributions (Ackermann 1980, Trefry and Presley 1976, Schropp etal. 1990, Pardue et a!. 1992). Fe is unlikely to be enriched from human activity, is wellcorrelated with particle size (Trefry and Presley 1976), and was measured in both the 1970and 1991 sediment samples. The ratio of each trace metal:Fe for the two time periods iscompared using the Mann-Whitney U test (Table 5.7). When normalized to the Feconcentration, only Zn and Cu are significantly higher in the 1991 samples.95Figure 5.3 Box-whisker plots illustrating the variability in total metal concentrationsin Salmon River sediments collected in the 1970’s and 1991. The unitsfor all the data presented are mg kg dry weight, except for carbon inpercent.400 30030000*100200100*200D000100806040200150i i 11 p’eGI I I I0•i •; •i• .9—— 0eGa:000zMay 74 Aug 91 Dec 91I ILIS5+it -9’ .\ga1 p o010080604020020015010050010050 0FR.I I00-.1 ••TT • •9’3510 — Oec i .1 9—eG96Figure 5.3 continued.150000 I I I 40000 30001000000Lu 200005000000100: I•1A 9’ 9’ i i •1e’ e c oe° e200000 I 15150000 *10z0500001000000I I 0 IMay 74 Aug 91 Dec 91 May 74 Aug 91 Dec 9197Table 5.6 Median metal content in the streambed sediments collected in the 1970sand August, 1991, in the Salmon River watershed, and significantdifferences between 1970s and August 1991 sediments. The 1970s and1991 sediments were compared using the Mann-Whitney U test.Aug 19911974 - 1977mg kg1 1970s median:Elementmg kg Aug 1991 mediann= n=19Zn 46 7.1OE+01 1.54E+02* 0.46Cr 12 1.53E+02 1.21E+02* 1.26Cu 46 1.60E+01 4.1OE+01* 0.39Co 37 1.27E+01 2.63E+01* 0.48Ni 37 3.44E+01 2.70E+01* 1.27Pb 34 1.41E+01 3.69E+01* 0.38Fe 46 2.12E+04 4.54E+04* 0.47Mn 46 5.18E+02 1.66E+03* 0.31Al 12 1.21E+05 6.35E+04* 1.91C(%)*signifint at a=0.05.Scatterplots of Zn and Cu versus Fe are illustrated in Figure 5.4. A linearrelationship between Zn and Fe and Cu and Fe is observed for all the sediments, with theexception of the data from May 1974. A linear trend is characteristic of natural conditions,and anthropogenic addition of metals would result in trace metal values above the linearrelationship (Trefry and Presley 1976, Schropp et at. 1990). The May 1974 sedimentsdeviate from the linear relationship; however, the metal levels are below the line. The twohigh Zn values from 1991 are the tributary samples from stations 8 and 10 that are also veryhigh in carbon content (Figure 5.4).Although up to a five-fold difference in trace-metal concentration is observed12 .52 399* 0.1398Table 5.7 Median metal:Fe ratio the streambed sediments collected in the 1970s andAugust, 1991, in the Salmon River watershed, and significant differencesbetween metal:Fe ratios in the 1970s and August 1991 sediments. The1970s and 1991 sediments were compared using the Mann-Whitney U test.1974- 1977 Aug ‘91 1970s median:EJement 1mg kg mg kg Aug 1991 median11= n=1946 2.88E-03 3.40E03* 0.8512 2.79E-03 2.80E-03 0.9946 8.OOE-04 9.11E-04 * 0.8837 6.63E-04 5.94E-04 1.1237 1.S1E-03 5.36E04* 2.8234 7.71E-04 7.76E-04 0.99between sediments sampled in the 1970s and 1991, these differences likely reflect thenatural variability in trace metal concentrations due to the physico-chemical properties of thesediments. Additional variability is introduced with different sample preparation andanalysis methodology.5.2 Surface-water chemistryTo obtain a measure of the variability in surface water chemistry under different flowconditions, grab samples were collected at three times during summer low flow conditionsbetween August 1991 and September 1993, and twice during high flow conditions inDecember 1991 and February 1992. pH, temperature, specific conductance, chloride, nitrateand total phosphorus were measured (Appendix 12 - 17).To illustrate the spatial and temporal variability of selected surface water chemistryparameters, the Figures prepared for this section present the mean of the grab samplesZn/FeCr/FeCu/FeCo/FeNi/FePb/Fe*signifint at a=0.05.99Figure 5.4 Scatterplots of trace metals versus Fe for Salmon River streambedsediments sampled in 1970’s and August 1991. A. Zn vs Fe.B. Cu vs Fe. The line of best fit was generated using all the dataexcept the May 1974 samples.B+ltDAz +:VA ItDD/H DI I I60000 80000Fe (mg/kg)I400 -350 -300 -250 -200 -150-100-50 -0-100+:,-.AAz:0I I I I I I I I I I I I I I20000 40000 60000 80000 100000 120000Fe (mg/kg)0i: May74A Aug74o Jun77+ Aug91best fitD May74A Aug74o Jun77+ Aug91best fit80 -60-40-20 -I’— I I I I I I I I I0 20000 40000 100000 120000100taken at low flow or high flow conditions, with the range represented by error bars.The mean daily discharges for the sampling dates for three low flow periods, inAugust 1991, August 1992, and September 1993, were obtained from measurements made atthe Environment Canada hydrometric station 08MH090 at 72 Ave. (surface water samplesite 6), and are presented in Table 5.8. Discharge measurements for the August sampletimes are similar, with 0.26 m3s’ for 1991, 0.21 m3s1 for 1992, and 0.165 m3s1 for 1993.The mean daily discharge, during high flow sampling conditions, was 3.08 m3s1 forDecember 10 and 1.73 m3s1 for February 5. The discharge measurements recorded at thegauge station for the low flow and high flow periods sampled by Beale (1976) and used forcomparison with results from this study are also listed in Table 5.8. The daily dischargemeasurements recorded for 1991 to 1993 are presented in Figure Spatial and temporal variability in surface-water specific conductanceThe specific conductance of the surface water provides a general measure of the totaldissolved solids transported by the river system. The geology of the watershed is reflectedin the specific conductance of the surface water during low flow conditions, illustrated inFigure 5.5. The specific conductivity is generally quite low, with most stations measuredbelow 16 mS m1 during low flow conditions and below 8.0 mS m1 during high flowconditions. These values illustrate the low dissolved solid load in the river.A distinctive spatial pattern of specific conductance is evident at low flow conditions(Figure 5.5). The headwaters of both the Salmon River and Coghlan Creek flow throughglacial marine deposits and stations 11, 17, and 12 have specific conductance measurementsof greater than 15 mS m’. As the Salmon River flows through the glacial outwash deposits101Table 5.8 Mean daily flow measured for the Salmon River at the 72 Ave crossing atthe gauge station 08MH090 for the surface water sampling dates in 1974-75and 1991-93. Data from Environment Canada.surface water . average flow average flow for. average daily flowsampling date 3 1 previous week previous two weeksms 3-1 3-1ms ms1974 low flow31/7/74 0.22 0.23 0.2713/08/74 0.22 0.20 0.2027/08/74 0.22 0.23 0.2330/09/74 0.19 0.19 0.1928/10/74 0.27 0.23 0.221974-75 high flow28/11/74 0.65 1.31 1.7130/12/74 2.68 2.51 4.0730/01/75 0.91 1.71 4.3001/03/75 3.37 3.50 3.221991-93 low flow21/08/91 0.26 0.27 0.3124/08/92 0.21 0.22 0.2229/09/93 0.15 0.17 0.171991-92 high flow10/12/91 3.08 4.75 3.3702/05/92 1.73 6.74 7.32(sampling stations 15 through 6) a relatively constant specific conductance is expected;however the conductivity steadily increases from about 7.0 to 13 mS m’. The conductivitycontinues to rise from station 23 to station 2 as the river flow through marine deposits,which contain natural source of salts from the geological material. Within Coghlan Creek,the river stations in the glacial outwash material (stations 20 to 25) have specificconductance measurements of about 10 mS m1, increasing to about 15 mS m’ at station 5.The specific conductance measured at the tributary stations during low flow conditions areFigure5.5SeasonalvariationinspecificconductanceintheSalmonRiverandCoghlanCreek.Aug/Seplow-flowconditionsarerepresentedby3samplingtimes;trianglesrepresentthemeananderrorbarstherange.DecandFebarerepresentedbyonesamplingtime.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.§indicatesAug/Septvalueforstation3notshownbecauseitisoffscale.25-__________________________________________35-SalmonR.tributaries30AA5-I....—IIIIIIIIIII\1024853131420-I.ZZZ1117159742361621122019255headtersmouthheadwatersSalmonconfluencestationnumber—A—Aug/Sep91-93——Dec91——Feb92jj......SalmonRCoghianCr.H L’3103generally between 10 and 15 mS m1, with the exception of station 3 (not plotted on Figure5.5) with a maximum reading of 65 mS m1. Station 3 drains a sub-watershed of marinedeposits with predominantly agricultural activities.The increase in specific conductance at the glacial outwash sampling stations, 15 to23 on the Salmon River and station 5 on Coghlan Creek, is probably due to the introductionof salts from human activities. The most likely anthropogenic sources of salts within theSalmon River watershed include inorganic fertilizers, animal manures, and human wastesdisposed in septic fields.During high flow conditions, specific conductance is much lower in the river thanduring groundwater-fed baseflow conditions during the summer months (Figure 5.5).Specific conductance measurements during high flow conditions vary between about 5.0 and8.0 mS m1 within the watershed. A gradual increase in specific conductance is observedfrom the headwaters towards the mouth for both the Salmon River and Coghlan Creek.A comparison of specific conductance measurements taken in 1974-1975 with themeasurements from this study shows that the spatial variation in conductivity in the SalmonRiver is similar in 1991-1993 to the pattern observed in 1974 (Figure 5.6). For most of thestations the conductivity data in 1991-1993 are near or within the range observed in 1974,although the mean conductivity observed in 1975 is slightly lower. One exception is station5 in Coghlan Creek, where the 1991-93 mean conductivity measured at low flow conditionsis 1.6 times greater than the 1974 mean conductivity. During high flow conditions, thespatial pattern of specific conductance in 1974-1975 is similar to the present; however, meanconductance values are somewhat lower in 1974-1975 (data not shown). These differences inspecific conductance through time may simply reflect the higher frequency of sampling inFigure5.6Temporalvariationinlow-flowspecificconductanceintheSalmonR.andCoghlanCreek:comparisonbetween1974and1991-93.1974isrepresentedby5samplingtimes,1991-93bythreesamplingtimes.Errorbarsrepresenttherangeofmeasuredvalues.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.§indicateserrorbarnotshownbecauseitisoffscale.25-C,) a) C.)-Co V C 8 C) ‘I 0 G) 0 C’)headwatersstationnumbermouthheadwatersSalmonconfluence—A—Aug/Sep91-93—0---Jul/Oct197470SalmonRtñbutaiies*160-30---::____.1..0—.IIIIIIIIII108531314474715-410-5- a.’SalmonR,IIIIIIIIIIIIII111715974616211220195CoghlanCr.C1051974-5 during a wider range of flow conditions, or it may indicate an increase in intensityof land-use activities.5.2.2 Spatial and temporal variability in surface-water chlorideThe concentration of chloride in the surface water is quite low, with mostmeasurements below 20 mg U1. However, spatial patterns are observed that likely reflectboth the geology and human activities within the watershed. The spatial pattern of chloridewithin the Salmon River watershed is very similar to the spatial pattern of specificconductance. The chloride concentration of the surface water during low flow conditions isillustrated in Figure 5.7. Higher chloride concentrations, ranging from about 10 to20 mg U’, are observed in the headwaters at stations 11, 17, and 12, which occur in theglacial marine deposits. Chloride gradually increases from a mean value of about 2 to 9mg U1 between the glacial outwash stations 15 to 23, which is similar to the pattern ofincrease in specific conductance. The stations 16 and 2, that drain areas of marine deposits,show higher chloride concentrations during low flow conditions. Values range from 14 to 18mg U’. Within Coghian Creek, the mean chloride concentration drops from about 11mg U1 in the glacial marine headwaters to approximately 7 mg U1 at station 19 on theglacial outwash deposits and then gradually increases to above 8 mg U’ at station 5, at theconfluence with the Salmon River. The chloride concentrations at the tributary stationsduring low flow conditions are generally between 5 and 14 mg U’, with the exception ofstation 3 (not plotted on Figure 5.7) which has a maximum reading of about 140 mg U1.The gradual increase in chloride from station 15 through 23, similar to the pattern ofincrease in specific conductance, is likely due to anthropogenic sources of chloride such as106human and animal wastes and the inorganic fertilizer potassium chloride.A comparison of chloride measurements taken during low flow conditions in 1974with the measurements from this study shows similar spatial patterns in the two time periods(Figure 5.7). The range of the chloride values measured at a station in 1991-1993 is higherthan the range observed in 1974, with somewhat higher variability observed in themeasurements at a station in 1974. These differences may be due to the different analyticaltechnique used for chloride determination between the two time periods or may be indicativeof more intensive land-use activities.The chloride concentration in the river is lower during high flow conditions thanduring low flow conditions in the summer months, with high flow measurements varyingbetween about 4 and 14 mg L’ within the watershed (Figure 5.8). A gradual increase inchloride is observed from the headwaters towards the mouth for the Salmon River, whilethere is a gradual decrease in Coghian Creek from the headwaters to the confluence with theSalmon River. During high flow conditions, the spatial pattern of chloride in 1974-1975 issimilar to the present; however, mean chloride values were somewhat lower in 1974-1975(Figure 5.8).5.2.3 Spatial and temporal variability in surface-water total phosphorusThe concentration of total phosphorus measured during low flow conditions isusually below 0.075 mg U’ (Figure 5.9). High flow conditions were only sampled once onFebruary 5, 1992 and most of the values obtained are below 0.100 mg U’ (Figure 5.10).Whereas the specific conductance and chloride levels in the Salmon River are higher at lowflow conditions, the concentration of total phosphorus in the water is generally higher duringFigure5.7Temporal variationinlow-flowchlorideintheSalmonR.andCoghlanCreek:comparisonbetween1974and1991-93.1974isrepresentedby5samplingtimes,1991-93bythreesamplingtimes.Errorbarsrepresenttherangeofmeasuredvalues.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.20-______________________________________40-35-30--J C) E25- a) L.. 0 015W 10- 5--SalmonR.tributariesIIII114SalmonR.CoghlanCr.So-_I11 hNs171597461621mouthstationnumber1220has19—A—Aug/Sep91-93—e--Jul/Oct19745SanonFigure5.8Temporalvariationinhigh-flowchlorideintheSalmonR.andCoghianCreek:comparisonbetween1974-75and1992.1974-75isrepresentedby4samplingtimes,1992byonesamplingtime.Errorbarsrepresenttherangeofmeasuredvalues.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.§indicatesvalueorerrorbarnotshownbecauseitisoffscale.20-______________________________________40SalmonR.tbutaes1535 30-I0-IIIIIIIIIII?251085.311420-SalmonRCoghianCr.o 15-o:/10-stationnumberH C 0,1117headwaters1mouth1220headwaters19—A---Feb92——1974-755 SalmonconfluenceFigure5.9Temporalvariationinlow-flowtotalphosphorusintheSalmonR.andCoghlanCreek:comparisonbetween1974and1992-93.1974isrepresentedby5samplingtimes,1992-93bytwosamplingtimes.Errorbarsrepresenttherangeofmeasuredvalues.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.§indicateserrorbarnotshownbecauseitisoffscale.-J I111715headwatersmouthstationnumberSalmonconfluence0.5974616211220195headwatersAAug1992•Sep1993—-—medianJul/Oct1974CFigure5.10Temporalvariationinhigh-flowtotalphosphorusintheSalmonR.andCoghlanCreek:comparisonbetween1974-75and1992.1974-75isrepresentedby4samplingtimes,1992byonesamplingtime.Errorbarsrepresenttherangeofmeasuredvalues.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.§indicateserrorbarnotshownbecauseitisoffscale.0.25 0.2-J c,) E1(I) 0 0 0.1-as 4-’ 0 4-’0.05 0mouthstationnumberheadwatersconfluenceSalmonR.tributaes111715974616211220195headwaters—A--Feb1992—s-—1974-75SalmonH H111the high flow conditions. Since phosphorus is not a highly mobile element within the soil,its movement is usually associated with rainfall events that produce overland flow. Bothdissolved and particulate phosphorus is carried by overland flow into the watercourse (Bakerand Laflen 1983, Sharpley et aL 1994). Increased stream flow from storm events can resultin mobilization of stream-bed sediments and increase the total phosphorus load in the watercolumn. Higher total phosphorus measurements are expected at the stations that drain areasof predominantly marine or glacial marine deposits, since infiltration rates on these soilswould be lower and result in more frequent occurrences of overland flow.During low flow conditions, the total phosphorus levels are high in the very slowmoving headwaters at stations 11 and 17, well above 0.1 mg L’. Total phosphorusconcentration at station 17, immediately downstream from the Vancouver Game Farm, isabove 0.3 mg U’ at both low flow sampling times (Figure 5.9). There is little or no stream-side vegetation along the banks of the Salmon River as it flows through the VancouverGame Farm, and the water is highly coloured with a high suspended sediment load, factorswhich could account for the high total phosphorus levels at this location. For the stationsdownstream of the Game Farm on the Salmon River mainstem, the total phosphorus valuesmeasured are usually quite low, below 0.06 mg U’. A similar pattern of total phosphorus isobserved for Coghian Creek, with total phosphorus values below 0.026 mg U’ at all stations,except with a slightly higher value of 0.038 mg U’ measured in the slow moving headwatersat station 12. Tributary values, especially for stations 3, 10, and 13, are generally higherthan in the mainstem of the Salmon River, and range from 0.09 to 0.24 mg U’. Thesestations drain areas that are predominantly glacial marine or marine deposits.The range of total phosphorus values obtained during high flow conditions in the112Salmon River mainstem is small; however, the spatial pattern does reflect the surficialgeology found in the watershed (Figure 5.10). The headwater stations of 11, 17, and 15drain fine-textured glacial marine deposits and have total phosphorus concentrations of0.08 mg U1 or greater. Higher total phosphorus levels are expected at these stations than atthe glacial outwash stations, since overland flow conditions are more likely to occur in thesefine-textured materials than in the freely draining glacial outwash deposits. Totalphosphorus levels gradually fall at the glacial outwash sampling stations, to a minimum of0.045 mg U1 at station 4, and increase from station 6 through 2 which drain glacial outwashand glacial marine deposits. The range of total phosphorus values in Coghlan Creek is verysmall during high flow conditions. Values of total phosphorus in the tributaries 3, 10, and13 are higher than in the Salmon mainstem during high flow conditions, similar to thepattern observed during low flow conditions, indicating that these tributaries are a source ofphosphorus to the mainstem of the river.The total phosphorus measurements made in 1974 by Beale (1976) are compared tothe present low flow (Figure 5.9) and high flow conditions (Figure 5.10). Since the 1974results are quite variable, the median value obtained for low flow and high flow conditionsis presented as the representative value. For both low flow and high flow conditions, thecurrent measurements fall within the range of values observed in 1974, indicating nodetectable change in the concentration of total phosphorus in the river system. The 1974values vary more than the values from this study, and the median low flow values areslightly higher in 1974 than at present. These differences are likely due to several factors.The detection limit for the method used to determine total phosphorus is likely higher forthe 1974 values, resulting in a higher median value. Also, in 1974, the low flow period was113sampled five times and the high flow period sampled four times, compared with two lowflow and one high flow sampling dates in the present study. Since the movement of totalphosphorus is generally associated with suspended sediments, total phosphorus levels in thewater column are expected to be highly variable depending on storm events and resultingchange in flow conditions.Total phosphorus measurements obtained from the SEAM and EQUIS data sets areavailable for stations 23, 6, 2, Glover Rd., and station 1. Total phosphorus levels areusually below 0.1 mg U1, and values obtained during this study fall within the range ofvalues obtained in the 1970s (data not shown). Most of the values of total phosphorusabove 0.2 mg U’ measured by Beale (1976) in 1974 were obtained during two samplingtimes in late October and early November, during conditions of relatively low flow (between0.2 and 0.5 m3s’). Since the flow conditions were low during the sampling, it is possiblethat these high values are a result of laboratory error.Elevated concentrations of phosphorus in the aquatic environment is of concernbecause of the potential for eutrophication of waters. The concentration of total phosphorusmeasured in the Salmon River during the past 20 years is usually below 0.06 mg U1. Thehigher concentrations of phosphorus are observed during the winter high flow conditions,when lower temperatures would slow down the rate of algal growth. The levels of totalphosphorus measured during this study and in existing government records, indicates thelevel of phosphorus in the river system does not appear to contribute to eutrophication.There are localized areas, such as station 17 (Vancouver Game Farm) and the tributarystations 3 and 8 that have concentrations above the recommended level for river systems ofabout 0.1 mg U’ total phosphorus. Measurement of the more reactive forms of phosphorus,114such as dissolved orthophosphate or NaOH extractable phosphate, would provide a betterindicator of algal available phosphorus than the determination of total phosphorus (Sharpleyet at. 1994).5.2.4 Spatial and temporal variability in surface-water N03-NThe precision of the measurement of N03-N was determined by collecting replicategrab samples at approximately 6 of the 19 sample stations and this data is presented inAppendix 18. For values above 1.0 mg L1, duplicates are, on average, within 5 percent ofeach other, and range from 0 to 22 percent. For values below 1.0 mg U’ the measurementsvaried more, ranging between 16 and 41 percent of each other (Appendix 18). The accuracyof the measurement of N03-N is illustrated in Appendix Spatial variabilityLow flow conditionsThe spatial variation in N03-N during low flow conditions for the Salmon River andselected tributaries is illustrated in Figure 5.11. The N03-N concentration at a station duringlow flow exhibits very little variability between the three years, with a range about the meanof about 0.1 to 0.4 mg U’ N03-N. Station 1, near the mouth, and station 3, a tributary ofthe Salmon, are exceptions with a greater range about the mean of 0.9 to 1.7 mg U’ N03-N(Figure 5.11).The N03-N concentration varies spatially in the Salmon River mainstem, varyingfrom below 1 mg U1 to greater than 5 mg U’. The concentration of N03-N remains below1 mg L1 in the headwaters until station 9, then increases to a peak at station 23 (232 Aye)Figure5.11Spatialvariationinnitrate-NintheSalmonRiverandCoghianCreekduringsummerlow-flowconditions.Pointsrepresent themeanofthreevaluesmeasuredin1991-93anderrorbarsrepresent therangeofvalues.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.mouthSalmonconfluence-J C)z1117159742361621122019255headwatersheadwatersstationnumber(11116of greater than 5 mg U’ (Figure 5.11). The N03-N level decreases from station 23 towardsstation 1 at the mouth. Station 1 exhibits the highest between-year variability. CoghlanCreek exhibits a similar pattern in N03-N as compared to the headwaters and middle reachesof the Salmon River. The N03-N in the headwaters is low and gradually increases to 1.7mg L’ at station 25, then rapidly increases to almost 7 mg L1 at station 5, near theconfluence with the Salmon River. This high concentration of N03-N at the mouth of theCoghlan Creek may explain the increase in N03-N between station 4 and 23 on the SalmonRiver, since the confluence of the Salmon River and Coghlan Creek is downstream ofstation 4.The level of N03-N in the five small tributaries of the Salmon River varied frombelow 1 mg U’ to greater than 6 mg U’. Station 8, with a N03-N concentration of about 5mg U’, flows into the Salmon River immediately downstream of station 7 and may partiallycontribute to the increase in N03-N between station 4 and 7. Station 14 on Davidson Creekhas N03-N levels of greater than 5 mg U’. This tributary flows into the Salmon River atGlover Rd, between station 2 and station 1. The discharge of this creek is small incomparison with the Salmon River; therefore,may not have a large influence on the N03-Nconcentration in the lower reaches of the Salmon River.A comparison of summer low flow N03-N with the surficial geology in thewatershed is presented in Figure 5.12. Surface water stations located on the glacial outwashsurficial materials are expected to have higher concentrations of N03-N since these materialshave shallow, highly permeable soils; therefore, the ground water is more susceptible tocontamination from land use activities. The headwater stations 11 and 17 in the SalmonRiver and 12 in Coghlan Creek are located on fine-textured glacial marine surficial materialsFigure5.12Spatialvariationinnitrate-NintheSalmonRiverinrelationtothesurficialdeposits. 8 S 4 2 0N03mg/LSURFICIiLDEPOSITSALLUVIUMMARINELJGLACIALMARINEGLACIALDUTWASHGLACIALTILL•SAMPLINGSTATIONSTREAMNETWORK—WATERSHEDBOUNDARY10123I<nH H118and have low N03-N concentrations. The stations located on the glacial outwash materials,15 through 4 on the Salmon River, and 19 through 5 on the Coghlan Creek, show increasingN03-N values in a downstream direction and all have higher N03-N values compared to theglacial marine stations. The peak N03-N concentration in the Salmon River mainstem is notobserved in the glacial outwash materials but is observed downstream in the marine parentmaterial at station 23. The ground-water flow direction from the Hopington Aquifer isconsidered to be northward (Gartner-Lee Limited 1992), and may contribute to the highN03-N concentrations at station 23. The four Salmon River stations on the fine-texturedmarine surficial materials show a gradual decrease in N03-N as would be expected if thesource of N03-N is removed; however, the decrease is small. If little additional water issupplied from tributaries or ground water in this section of the river, little dilution of theN03-N would occur. Station 1, located on alluvial materials, has a much lowerconcentration of N03-N.High flow conditionsThe spatial variation in N03-N within the Salmon is much lower at high flow than insummer low flow conditions, with the range of N03-N values between 0.7 and 3.0 mg U1for the Salmon River mainstem (Figure 5.13). For both high flow sampling dates, thepattern of N03-N from the headwaters to the mouth, for both the Salmon River mainstemand Coghlan Creek, is similar to low flow conditions; however the pattern is not aspronounced (Figure 5.13). In the headwaters of the Salmon River mainstem, N03-N valuesare below 2 mg U1 until station 9, and then increase to between 2.5 and 3.0 mg U’ in themiddle reaches, and decrease gradually towards the mouth.Figure5.13 c,) zSeasonalvariationinnitrate-NintheSalmonRiverandCoghlanCreek.Aug/Seplow-flowconditionsarerepresentedby3samplingtimes; trianglesrepresent themeananderrorbarstherange.DecandFebarerepresentedbyonesamplingtime.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.10 8 6 4 2 01117159mouthstationnumberheadwatersSalmonconfluence10headwaters742361621122019255—A-—Aug/Sep1991-93—--Dec91—---Feb92H120The higher N03-N concentrations in the middle reaches of the Salmon River duringsummer low flow conditions suggest that ground water is the source of N03-N. The N03-Nconcentration is lower in the middle reaches during winter high flow conditions than duringlow flow conditions, most likely as a result of dilution from rainfall and overland flowduring winter rainfall events. N03-N concentrations are higher in the headwaters duringwinter high flow periods than during low flow; therefore, the N03-N source in this portionof the river is likely due to overland flow during winter storm events. Temporal variabilityLow flow conditionsTo determine if the concentration of N03-N in the surface water of the Salmon Riverhas changed during the past 20 years, the average low flow N03-N from 1991-1993 iscompared with measurements made by Beale (1976) in 1974. The 1974 sampling timeswere chosen for comparison based on the time of year the samples were collected and themean daily discharge at the time of sampling (Table 5.8). Five sampling dates between Julyand October 1974 were selected for comparison with the present study because the meandaily discharge at the hydrometric station for the two weeks prior to sampling was below0.30 m3s1.The mean and range of N03-N at each sample station for low flow conditions in1974 and 1991-1993 are plotted for the Salmon River, Coghian Creek, and selectedtributaries of the Salmon River (Figure 5.14). The N03-N concentration during low flowconditions in 1974 increases at station 7, reaches a peak at station 4, then gradually declinestowards the mouth. The pattern of N03-N from the headwaters to the mouth for the SalmonFigure5.14Temporalvariationinlow-flownitrate-NinSalmonR.andCoghianCreek:comparisonbetween1974and1991-1993.1974isrepresentedby5samplingtimes,1991-1993bythreesamplingtimes.Errorbarsrepresenttherangeofmeasuredvalues.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.IIIIIIIII1117159742361621headwatersmouthstationnumberheadwatersSalmonconfluence-J c) z10- 8- 6- 4- 2- 010- 8SalmonR.tñbutaiies A6 4: :[;SalmonRCoghlanCr.IIIII122019255—A-—Aug/Sep1991-93——Jul/Oct1974H t’J H122River mainstem during low flow conditions in 1993 is similar to the pattern observed in1993, with some notable differences.A higher level of N03-N at station 7 in 1974 than at present (3.3 versus 1.4 mg U1)may be due to a small difference in the location of station 7 between 1974 and 1991.Station 7 was sampled upstream of the small tributary represented by station 8 in 1991-1993, whereas in 1974 station 7 may have been located downstream of station 8, and wouldinclude flow from this small creek. Since the N03-N at station 8 is high, at about 5 mg U’in both 1974 and 1991-1993, a difference in sample location may partially explain thedecrease in N03-N at this station between 1974 and present.The N03-N concentration in Coghian Creek, immediately upstream from theconfluence with the Salmon River (station 5) is much greater at present, with 6.9 mg U’compared to 4.5 mg L1 in 1974. This higher concentration of N03-N in Coghlan Creekmay explain the slightly higher N03-N concentrations observed currently at station 6 than in1974. Another substantial increase in N03-N is observed at station 14 on Davidson Creek atRawlison Cres., increasing from an average of 1.5 mg U’ in 1974 to 5.6 mg U1 at present.In conclusion, the pattern of N03-N during low flow conditions has changed littlebetween 1974 and the present, except at two stations in the two tributaries, Coghlan Creekand Davidson Creek. The N03-N has changed dramatically at these tributaries, suggestinglocalized increases in N03-N within the watershed. In particular, the change in CoghianCreek could explain the increase in N03-N observed during low flow conditionsimmediately downstream of the Salmon-Coghian confluence at station 6.123High flow conditionsThe mean of high flow N03-N levels from the two sample periods in 1991-1992 iscompared with the mean of four sampling periods conducted during high dischargeconditions by Beale (1976) in 1974-75 (Figure 5.15). The mean daily flow is averaged forthe two weeks prior to sampling to characterize the high flow conditions sampled in 1974-75 and 1991-92. The two-week average of mean daily flow ranged from 1.7 to 4.3 m3s1 for1974-75, compared with 3.4 and 7.3 m3s1 for the two 1991-92 sampling periods (Table 5.8).The mean and range of N03-N at each sample station for high discharge conditions areplotted for the two time periods for the Salmon River, Coghlan Creek, and selectedtributaries of the Salmon River (Figure 5.15). N03-N is more variable at high flowconditions for both time periods than at low flow conditions. Different intensities of stormevents, resulting in different rates of transport of N03-N from the land surface, or dilution ofthe N03-N from ground water sources, would result in variable surface water N03-N.The range of N03-N is greater in the 1974-75 samples, represented by four high flowsampling times, than in 1991-92, where samples were collected only two times. For all butone station, the 1991-92 high flow samples fall within the measured range of the 1974-75samples. Therefore, no increase in high flow N03-N concentration in surface water between1974 and 1992 can be discerned from this data set. Other data sources of N03-N in surface waterAdditional measurements of N03-N concentration in the Salmon River was compiledfrom the BCMOELP EQUIS and SEAM databases and combined with data from Beale(1976) and data from this study to examine variability in N03-N at a station with time.-J z0headwatersmouthstationnumber122019headwatersSalmonconfluence1010Figure5.15Temporalvariationinhigh-flownitrate-NinSalmonR.andCoghianCreek:comparisonbetween1974-75and1991-93.1974-75isrepresentedby4samplingtimes,1991-1993bytwosamplingtimes.Errorbarsrepresenttherangeofmeasuredvalues.ArrowsmarktheapproximatelocationontheSalmonRivermainstemofthesampledtributaries.SalmonR.tributaries8- 6- 4: 2- f’.I64SalmonR.1013144 2CoghlanCr.111715974616215—A—1991-92-.-—1974-75F’.)125Five stations (Figure 5.16), between 232 Ave and 96 Ave have at least 15 measurementsbetween 1972 and 1993 (Figure 5.17). For all five stations, the range of N03-N levelsmeasured in this study between August 1991 and September 1993 are within the range ofN03-N values observed in the early 1970s.Since there is no apparent temporal trend in N03-N at these stations, the relationshipbetween N03-N at each station and discharge is examined (Figure 5.18). There is a goodrelationship between the discharge measured at the gauge station (station 6) and N03-N forstation 23, 6 and 2, with N03-N decreasing with an increase in discharge. This suggests thatN03-N source is from ground water during periods of low flow and is diluted duringflooding events. There is no significant relationship between N03-N and discharge in thelower reaches of the river at Glover Rd. and Station 1, which suggests N03-N sources fromboth ground water and flooding events.An increase in surface water N03-N with time has been widely documented in manyriver systems in North America and Europe and has been attributed to the large increase inthe use of nitrogen fertilizer since the 1940s (Smith et al. 1987, OECD 1986, Hallberg1989). In comparison, there appears to be only localized increases in surface water nitrateduring the past 20 years in the Salmon River watershed. Measurements at the mouth of theriver indicate no overall increase in surface water N03-N. Investigations of other riversystems have also shown little change in stream N03-N despite an increase in fertilizer usewithin the watershed (Thomas et al. 1992, Keeney and DeLuca 1993). In an investigationof surface water N03-N in the Des Moines River, Iowa, no temporal trend was observed inN03-N concentration despite a doubling of nitrogen fertilizer use within the watershed. Theinvestigators suggested that there was either little change in the N03-N loadings to the river,Figure5.16LocationofthesamplingstationsontheSalmonRiverrepresentedintheEQUISandSEAMdatabases.———MINSTEMSTREAMNETWORKROADNETWORK—WATERSHEDBOUNDARY10123KmH7-s2,wliCres.6C)-,..2: 2-$,c1_..4..4,0—IIIIIIIII1972197619801984198819927- 6-GbRDad_J5 C)z:03-.-,,,-:1-IIIIIIIIIIIII1972197619801984198819927- 6-0) d3.2---$---.-*.,1- 0—1•a-I1II197219761980198419881992dateFigure5.17Temporalvariationinnitrate-NatfivesamplingstationsintheSalmonRiver.LettersindicatethelocationofthesamplingstationsasshowninFigure5.16.DatafromEQUIS,SEAM,Beale(1976),andthisstudy.7- 6-25-2: &3-Cu-C1—A232St,Staüon34,,:. +4,44,,‘1IIIIIIIIIIJIIIIIIIII1972197619801984198819927- 6-BGfion25-,.‘::::4-2: cb3.j..eI’II:-.--C:1-;40IIIIIIIIIIII197219761980198419881992dateN)Figure5.18Relationshipbetweendischargeat72Ave(08MH090) andnitrate-NatfivesamplingstationsintheSalmonRiver.*Spearmanrankcorrelationcoefficientsignificant ata=0.05.LettersindicatethelocationofthesamplingstationsasshowninFigure5.16.DatafromEnvironmentCanada,EQUIS,SEAM,Beale(1976),andthisstudy.7- 6-z d3- 2- 1-A232St,Station23*17-Station2,RawIianCres.*6- 0—IIIIiijIIIIII0.11107- 6--J-1—+,.•0—IIIIIIIIIIIIII100.11107- 6ige,tizs...•.+,0-i•I•I4IIIIIII0.10.1 7- 6-25-I3) z d3-- 1— 0.LBGaeStatior 72Ae,Station6*0.1disctBrge(n1)10disclwge(m31)10H co129or that changes in N03-N sources to the river cancelled each other resulting in no differencebetween N03-N concentrations in the river system in 1945 compared to the present (Keeneyand DeLuca 1993). In the Hopington Aquifer where N03-N concentrations are highest, it ispossible that N03-N sources from agricultural activities in the 1970s have been replaced byN03-N in septic system effluent in residential areas, resulting in only a small change inamount of N03-N sources and relatively little change measured in N03-N in the surfacewater. Alternatively, the lack of a visible change in N03-N in surface water may be due tothe slow movement of N03-N contaminated ground water, and the effect of changes in landuse and N03-N loadings may not yet be measurable in the surface water.5.2.5 Relationships between surface-water quality variablesThe degree of association between the five surface water quality variables measuredfor the five sampling periods of this study are examined using Spearman Rank correlationcoefficients (Table 5.9). The specific conductance and pH are both positively correlatedwith the river discharge (measured at the gauge station), whereas total phosphorus innegatively correlated with discharge. Higher dissolved salts are expected during low flowconditions when ground water is the predominant water supply, whereas higher totalphosphorus is expected during higher flow conditions, when suspended sediment load ishigher. The negative correlations observed between total phosphorus and both pH andN03-N is likely due to the relationships of these variables with flow.The subset of the 13 sampling stations draining glacial outwash materials (stations15, 10, 9, 24, 7, 8, 4, 20, 19, 25, 22, 5, and 23) was examined separately to remove theinfluence of the surficial geology type on water quality and the Spearman Rank correlation130Table 5.9 Spearman rank correlation coefficients between water quality variables forall the Salmon River watershed stations sampled at high and low-flowconditions in 1991 to 1993.specific totalchloride nitrate-N dischargepHconductance phosphoruspH 1.000specificconductance 0.543** 1.000chloride 0.320* 0.692** 1.000nitrate-N 0.202 -0.022 0.184 1.000totalphosphorus 0.410** -0.082 -0.035 0.535** 1.000discharge atstation 6 0.536** 0.614** -0.204 0.181 0.399** 1.000* significant at a=0.05, ** significant at a= 0.01131coefficients are presented in Table 5.10. The associations between water quality variablesobserved when the entire set of sampling stations is considered are also observed for thesampling stations located on glacial outwash. In addition, a negative correlation betweentotal phosphorus and specific conductance, and a positive correlation between N03-N andchloride is observed. The negative correlation observed between total phosphorus andspecific conductance is likely due to the relationships of these variables with flow. Sincethe high concentrations of both N03-N and chloride measured at the glacial outwash stationsare considered to result from the effect of land use activities, the correlation between N03-Nand chloride suggests a common source of these contaminants.5.3 Spatial and temporal variability in N03-N in ground water5.3.1 Spatial variability in N03-N in ground waterApproximately 400 wells within the Salmon River watershed and Hopington Aquiferhave been tested for water quality during the past 20 years (Table 5.11). Seven percent ofthe wells tested have N03-N concentrations above the Canadian drinking water standard of10 mg L1 N03-N. Based on a conservative estimate for the natural background level forN03-N in ground water of between 1 and 3 mg L1 N03-N (Madison and Brunett 1985, citedin Spalding and Exner 1993) between 30 and 40 percent of the wells have N03-N levelsabove background and draw on water supplies that have been negatively affected by humanactivity. Considering only those wells located in the Hopington Aquifer, 12 percent of thesampled wells have N03-N concentrations that have been measured to be above theCanadian drinking water standard, 48 percent above 3 mg L1, and 58 percent above1 mg L1 N03-N.132Table 5.10 Spearman rank correlation coefficients between water quality variables forthe Salmon River watershed stations on glacial outwash materials, sampledat high and low-flow conditions in 1991 to 1993.totalspecificchloride nitrate-N dischargepHconductance phosphoruspH 1.000specificconductance 0.564** 1.000chloride 0.375* 0.712** 1.000nitrate-N 0.257 -0.165 0.434** 1.000totalphosphorus 0.555* .0.455** -0.281 .0.429* 1.000discharge atstation 6 .0.443** .0573** -0.210 0.184 0.513** 1.000* significant at ct=0.05, ** significant at cz= 0.01PHSP.CONDK CHL0DENITRATE133Table 5.11 Number of wells tested for Nitrate-N in the Salmon River Watershed andHopington Aquifer Area between 1973 and 1993, and the concentration ofNitrate-N found in the wells*.Salmon R. Watershed* Hopington Aquifer onlyNitrate-N Number of Wells Percent of Wells Number of Wells Percent of WellsmgL )>0 to <=1 236 58 89 41>1 to <=5 78 19 48 22>5 to <=10 61 15 53 25>10 29 7 26 12total 404 100 216 100>1 168 42 127 58>3 127 31 104 48§ For wells tested more than once, the maximum nitrate value recorded was used.Some of the wells in the Hopington Aquifer are included in the summary for the SalmonRiver watershedThe spatial variation in N03-N in well water from samples collected between the early1970s and 1993 is presented in Figure 5.19, with most areas of the watershed represented byat least a few samples. If the well was sampled more than once during the 20 year period,the maximum N03-N value measured was used. Information on wells was collated only forthose wells within or near the study area, and the wells were divided into four categoriesbased on the N03-N concentration. The variation in N03-N concentration is not evenlydistributed throughout the watershed. Wells which have N03-N concentrations measuredbelow 1 mg L’, illustrated by yellow circles, are found throughout the watershed. Wellswith N03-N levels above 1 mg L’ are located in three general areas: The HopingtonAquifer, Fort Langley adjacent to the Fraser River, and a small concentration of wells in theheadwater area of Coghian Creek. Both Fort Langley and the Hopington Aquifer are areasFigure5.19Spatialvariabilityinnitrate-NinwellwaterintheSalmonRiverwatershedandHopingtonAquifer,fromwellstestedbetween1970and1993.Datasource:GroundwaterSection,BCMOELP.NITRATE—N(mg/L)<1•1—5•5—10•>10AQUIFERBOUNDARYSTREAMNETWORKWATERSHEDBOUNDARY0123KH w135of glacial outwash surficial deposits whereas the Coghian Creek headwater area is composedof glacial marine deposits. Within the Hopington Aquifer, the highest concentration of wellswith N03-N levels above 5 mg U1 is found in the central section of the aquifer,approximately bordered by 232 St to the west and 248 St to the east, to the north by 58 Aveand the south by 48 Ave.5.3.2 Annual variability in N03-N in ground waterTo determine seasonal patterns in N03-N in ground water, the records from regularmonitoring of 12 wells between 1975 and 1981 are examined. These wells are located onthe Hopington Aquifer (Figure 5.20) and monitoring was carried out by EnvironmentCanada. The wells were not randomly selected, but were chosen where the N03-Nconcentration was near or above the Canadian drinking water standard of 10 mg U1 (RodZimmerman, pers. comm.). The wells range in depth from 10 to 170 ft (3.0 to 51.8 m) andeach well was sampled approximately 5 times per year. The N03-N concentration exceededthe Canadian drinking water standard in five of the 12 monitored wells at some time duringthe eight years of observation. The N03-N values in these monitoring wells was highlyvariable; however, some general trends can be observed.An annual variation in the concentration of N03-N was observed for all twelve wells.The annual range per well varied from 0.5 to 22 mg U’, with an median value of about 2mg U’. The greatest within-year ranges were observed in 1976 for the 17 ft (3.0 m) well(Figure 5.21A) and in 1977 for the 108 ft (32.9 m) well (Figure 5.22A). For the six wells,9012, 9013, 9015, 9018, 9020, and 9022, the within-year variation was 5 mg U’ or less.The wells deeper than 120 ft (9022, 9013, 9020, 9018, Figures 5.21A, 5.22B and C) showed136Figure 5.20 Location of the twelve water wells in the Hopington Aquifer monitoredregularly by Environment Canada from 1974 to 1981.• ‘JELLSTREAM NET4ORKOUIFER BOUNORY— WATERSHED BDUNDRYI C 1 2 3K,137Figure 5.21 Seasonal and temporal variation in nitrate-N in the HopingtonAquifer, from monitored wells east of the Salmon River. Datafrom NAQUADAT, supplied by Groundwater Section, BCMOELP.25 -A 9014, lOft.20 — 9017, 17ft. -9022, 158 ft._Jz----.5 I0— I I I I I I I I I I I I I I I I I I I I I I I I1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982DATE12-B: : : :9016, 88 ft.10----C)ci:.2- - -0— I I I I I I I I I I I I I I I I I I I I I I I I I1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982DATE138Figure 5.22 Seasonal and temporal variation in nitrate-N in the HopingtonAquifer, from monitored wells west of the Salmon River. Datafrom NAQUADAT, supplied by Groundwater Section, BCMOELP.40-— 9019, 69 ft.35-9011, 108 ft.30- ----- -C)Ez 20--• :10- I5—0- i I I I I I I I I I I I I I I I I I I I I I1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982DATE2- 9013, 131 ft. - -I-] — 9020, 170ff.0—••.——.—•••.i I I I I I • I I I I I I I I I I I I I I I I I I I1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982DATEFigure 5.22 continued.139IC. -10-_--J0)z 6-E4_______2-0-1972 1973 1974 1975 1976 1977 1978DATE1979 1980 1981 1982I..1-_— 9012,61 ft.— 9018, 121 ft.140the least within-year variation.An annual cycle in N03-N values was most noticeable in the shallow wells, 9014and 9017 (Figure 5.21A). Although the N03-N concentrations varied greatly from year toyear in the 10-ft well, an annual peak is discernable. In the 17-ft well two N03-N peaks peryear were observed, one in the first quarter, and the second in the third or fourth quarter ofthe year. Annual N03-N peaks were observed in many years for the other wells; however,the cyclic pattern was less regular. It is interesting to note that the month of the peakannual value varies from year to year for an individual well and between wells.N03-N in ground water can be highly variable within a year for a single well,particularly shallower wells, in this portion of the Hopington Aquifer. This high variabilityin ground water N03-N is also observed in the Brookswood Aquifer in Langley and theAbbotsford Aquifer in Matsqui (Kwong 1986, Zimmerman 1990, Dayton & Knight et a!.1994). Factors that influence the nitrate concentration in a given well include the types ofland use and nitrogen sources, the hydraulic properties of the overlying materials and of theaquifer itself, the depth to the water table, preferential flow paths through geologicmaterials, and the ground water flow systems (Hallberg 1989). The complex interaction ofthese factors can result in solute stratification within the ground water, and in combinationwith seasonal and temporal variation in recharge, can affect the timing of appearance ofground water with a given N03-N concentration in a well (Hallberg 1989). It is expectedthat the N03-N levels in the ground water in the Hopington Aquifer is affected by theseasonal cycle of the recharge of the aquifer from rainfall; however, the rate of delivery ofthe autumn season “dose” of nitrates to a particular well will vary. The above factors canexplain the variation in the date of peak annual N03-N concentration between nearby wells141and between years in the Hopington Aquifer.Based on the long term monitoring of these wells, there is no evidence that any onetime of the year would be most appropriate for sampling in order to determine the maximumN03-N concentration. Given the wide range of N03-N concentrations observed in several ofthe wells, a single sample is inadequate to determine whether the N03-N concentration isabove the Canadian drinking water standard, and the well suitable for use as a drinkingwater supply. To obtain a better understanding of annual fluctuations in N03-N seasonalvariation with time, more frequent sampling (biweekly or monthly) would be required,combined with an analysis of rainfall distribution. However, the results from an individualwell only provides information on the small portion of the aquifer tapped by the well. Thepresence of perched water tables (Dakin 1993) and the reworking of the unconsolidatedmaterials with successive glacial advances (Armstrong 1984) suggests a high degree ofheterogeneity in the Hopington Aquifer, therefore making it difficult to extrapolate from afew localized wells.5.3.3 Temporal variability in N03-N in ground water5.3.3.1 1973 - 1981To determine if the level of N03-N in the ground water of the Salmon Riverwatershed has changed with time, the records for individual wells sampled repeatedly willbe examined. The twelve NAQUADAT wells have been divided into groups basedadjacency of wells and on the similarity in N03-N profile during the ten year period.The four wells tested east of the Salmon River are illustrated in Figure 5.21A and B.Despite wide fluctuations N03-N in wells 9014, 9017, and 9022, all exhibit a gradual142increase in N03-N during the observation period (Figure 5.21A). The 10-ft shallow wellshows the largest variability in N03-N and the poorest match to the pattern of increasingN03-N with time. The highest values were observed in the first quarter of 1975, and 1976,exceeding the Canadian drinking water standard. Following these two annual maximumvalues, the N03-N concentration gradually increased from 1977-1981. Well 9016, althoughin close proximity to well 9022, shows wide fluctuation of N03-N and a general pattern ofdecrease in N03-N from 1975-1978 (Figure 5.21B)On the west side of the Salmon River, the wells 9013, 9015, and 9020 exhibit similarpatterns in their N03-N profiles (Figure 5.22B). The wells 9013 and 9015 are withinapproximately 0.5 km of each other. Both wells have similar patterns in their annual cycles;however, there is a lag of about one year in the deeper well and the magnitude of the annualcycle is also smaller in the deeper well. Both wells showed a pattern of decrease, thenincrease in N03-N levels, with no overall trend during the observation period. The deepestwell (Figure 5.22B), shows less annual cyclic variation, but also exhibits decreasingconcentrations in N03-N from 1975 to 1978. The sharp drop in N03-N seen in the othertwo wells, is observed in the deepest well in the first quarter of 1978. This well did notshow a subsequent increase in N03-N as was seen in wells 9013 and 9015.Wells 9019, 9011, and 9018, are near the western boundary of the aquifer (Figure5.20) Levels of N03-N above the Canadian drinking water standard were measured formany of the sampling dates for both wells 9011 and 9019 (Figure 5.22A). The slightlydeeper well, 9018, had NO3-N below the drinking water standard but with almost half of thesamples above 8 mg U1. A peak in N03-N concentration is observed between 1977-78 anda decrease from 1978 to 1981 for all three wells. Well 9012 fluctuates between 6.5 and 3143mg L1 N03-N during the period of record, with no increasing or decreasing temporal trend(Figure 5.22C). Well 9021 had N03-N levels at detection limit during the period of record(data not shown).A longer period of monitoring would be required to determine if an overall increasein N03-N in the aquifer has occurred. During the monitoring period, three of the four wellsmeasured to the east of the Salmon River exhibit an increasing trend in N03-N, whereas ofthe eight wells monitored to the west of the river, six show no increasing or decreasingN03-N trend, and exhibit high variability. High spatial and seasonal variability in N03-N inwell water from unconfined aquifers is the rule, rather than the exception, and makes thedetection of longer term temporal changes more difficult to observe (Hallberg 1989). Thesimilar patterns in N03-N values observed in wells separated by distances of up to severalkm (e.g. 9017 and 9022) suggests that either the N03-N sources are widespread, or thatlocalized sources of N03-N are rapidly dispersed in the ground water. 1973 - 1993Six of the 12 NAQUADAT wells were sampled again at least once in the late 1980s.The equivalent BCMOELP SEAM site numbers are listed in Appendix 20. Of the wellstested east of the Salmon River, N03-N levels have increased in 9017, decreased in 9022,and are within the 1972-81 range for 9014 and 9016 (Figure 5.23). West of the SalmonRiver only two wells have been re-sampled. N03-N levels measured recently for 9019 arewithin the 1972-91 range, whereas N03-N levels appear to have increased in well 9019(Figure 5.24). Wells monitored in the Brookswood and Abbotsford Aquifers show similarresults, with some wells showing a pattern of increasing and other decreasing N03-N with15--JC)z 10-I5-20-p15-144Figure 5.23 Temporal variation in nitrate-N in the Hopington Aquifer, frommonitored wells east of the Salmon River. Data fromNAQUADAT, SEAM, and FVGMP, supplied by GroundwaterSection, BCMOELP.20A A 9017, 17ftAA A A-----•---“-----------A4,A A£AA : : : : :A: : AAI I I I1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 19DATEB A A 9014, lOft.A:A: : : : : :4AA :: :4. : : : : : :A: : : : : I5-- -0-—1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994DATEFigure 5.23 continued.2015--J10-ci)5-I.’I I I I I I I ‘1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994DATE145-JzA 9016, 88ft.C A 9022,158ft.: : : AA•AAAAI&%ALA : : :£D2015-10-5-:A:AA0-i I ‘ I I1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994DATE146Figure 5.24 Temporal variation in nitrate-N in the Hopington Aquifer, frommonitored wells west of the Salmon River. Data fromNAQUADAT, SEAM, and FVGMP, supplied by GroundwaterSection, BCMOELP.40-A A I A 9011,108tL.35-1-A I : : : I :3025 AE A : : : : :•A Az20 A:k:4: A15 AA1:“AJA0-i I I I I I I I I1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994DATE40_________B A 9019,69ft. I35-A30-25-A Az2O--A. -C A : ,: : : : :10 AAA___A0— i I I I I I I I1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994DATE147time (Zimmerman 1990). From these results it is apparent that, given the dynamic andvariable nature of N03-N in these wells, insufficient wells have been monitored during thetwenty year period to use records from individual wells to determine if a change in N03-Nconcentration has occurred in the Hopington Aquifer.Although very few individual wells have been monitored regularly for N03-N duringthe past two decades, the approximately 400 wells within the study area that have beentested at least once for N03-N during this time period can be examined for temporal trends.The database of 400 wells tested for water quality is comprised of samples from theNAQUADAT data set, the Water Quality Check Program (WQCP), and the SEAM database.Despite the limitations introduced due to different biases in the sampling design of eachprogram, this collective database contains the best historic data available for the study area.The wells have been stratified into three surficial material types, marine, glacialmarine, and glacial outwash, and the N03-N concentration plotted against time in Figure5.25. For wells that were measured more than once, the median N03-N value was plotted.On the marine surficial material all but one measurement were below 3.0 mg U1, with mostvalues at or near the detection limit of 0.02 mg U’. On the glacial marine material all butfive measurements were below 5 mg L. For both glacial marine and marine parentmaterials there is greater variability in N03-N in the WQCP dataset collected after 1980 thanin the earlier NAQUADAT dataset (1972-1981). The range in median N03-N value in theglacial outwash material is much greater, from below detection limit to greater than 20mg U1. Elevated NO3-N values, particularly on the glacial outwash material, have beenrecorded since the early 1970s. However, the number of wells with N03-N recorded abovethe drinking water standard has increased with time. This may reflect an increase in NO3-N148Figure 5.25 Temporal variation in nitrate-N in well water, from wells on the threedominant surficial geology types in the Salmon River watershed andHopington Aquifer. See Figure 3.3 for the surficial geology map. Datasupplied by Groundwater Section, BCMOELP.30-glacial outwashA• A A20 ---—a----- A.A• Az :A A hA: Aci)’ • :A A•10 AAAAA±A””• AAEAAA. LA AAAA AA A A :AA AAA: Aj:44A4 :iA0- 4AAA. I AdLA I .1970 1975 1980 1985 1990 199530-___________glaciaimaiine : : A EasttWestEAzci) :• • ACu----•• A A:.:A AA.A A40- & A A& A A A AA i A1970 1975 1980 1985 1990 199530 - •marine:-Jz)Cu -• A •• AA:0- A A4 1A A AA A A1 j A%.A A1 AL A1970 1975 1980 1985 1990 1995date149in the Hopington Aquifer or may merely reflect biases introduced in sampling. Participationin the WQCP is voluntary and dependent on a household requesting water testing. It ispossible that more people would request testing in problem areas such as the HopingtonAquifer area, where elevated N03-N levels in well water have been documented since theearly 1970s, and may skew the well-water N03-N records towards higher values.5.3.4 Spatial and temporal distribution of N03-N in well waterThe spatial extent of wells sampled with elevated N03-N (above 3 mg L1) isexamined at four different time intervals, 1973-1977, 1978-1982, 1983-1987, and 1988-1992to determine if the spatial extent of contaminated wells has increased. It appears that thedifference in spatial extent of elevated N03-N in well water is small, when measured duringthe twenty year period (data not shown). Due to lack of representation of all areas of thewatershed by the samples collected in each five year interval, it is not possible to determineif the small differences observed are from the incomplete sample coverage or an increase inspatial extent of N03-N in the ground water.To examine the N03-N concentration in well water with time, the median N03-Nvalue in well water in each contributing area is compared for wells sampled for the 10 yearperiod before and after 1983 (Figure 5.26A). For wells that were sampled more than oncein the ten year period, the median N03-N value for the ten period was used. The medianN03-N value per contributing area is plotted for each ten year period, with the number ofwells measured plotted above each median. The median N03-N is lower pre-1983 than inpost-1983 for all but two contributing areas. For many of the contributing areas, the samplesize is small. However, for the three contributing areas S06-07, S08-09, and COl-SlO theFigure 5.26_JD)z 3-1—25-20 --J15-5-150Median well-water nitrate-N in 12 areas in the Salmon River watershed, frommeasurements made in 1972-1982 and 1983-1993. A. median nitrate-N. B.median nitrate-N, with range indicated by error bars. Numbers above the barsindicate the number of wells sampled. Data from NAQUADAT, SEAM, andWQCP, supplied by Groundwater Section, BCMQELP.A6-5-4I267L()CoB6ACNo 0C)(0 0o 00__J_ii I JII I I Ic’.j u r-o 0 0 0Cl)- c,o 0 0Co Co CoI I I I0) ‘- It) 0) L) r— cj9 .;. — ‘- 0 0 0 9a 0 C.J Co 0 —0 ‘-— 0 0 0Co Cl) Co 0 0 0 00contributing area• 1972-1982 1983-1993151sample sizes are greater than 15 and an increase in median N03-N is at least three fold forall areas. When the mean N03-N concentration per contributing area is compared, a similartrend of N03-N increase in the second decade is observed (data not shown). These dataindicate a substantial increase in median well N03-N concentration in middle sections of theSalmon River watershed, from pre-1983 to post-1983 records. Although the median haschanged from pre- to post-1982, the difference is small between the range of nitrate valuesobserved in each contributing area in the two decades. Given the multiple data sources usedand the sample biases in the different data sets, this observed increase in well water N03-Ncould be due to the more intensive sampling of problem areas during the second decade,when N03-N had been identified as a possible contaminant, compared to prior to 1993.5.4 Comparison of surface water and ground water N03-NThe median N03-N concentration in well water per contributing area is comparedwith the mean low-flow N03-N in surface water in Figure 5.27. There is high associationbetween ground water and surface water N03-N from the headwaters to the middle reachesof the Salmon River, contributing areas COl-SlO. However, further downstream the N03-Nsurface water remains high while the median N03-N concentration in well water drops tonear or below 1 mg U’. A lag of one contributing area is observed between the increase inN03-N in the ground water and the increase in N03-N observed in surface water. Thisdifference may be attributable to the northward flow of ground water.152-Jc,)z7Figure 5.278-6-5-4-3-2-1—o-jComparison of nitrate-N in surface water during low-flow conditionsand in well-water, in the Salmon River watershed. The well-waternitrate-N is the median nitrate-N value from wells tested within thecontributing area upstream of the surface-water sampling station, frommeasurements made between 1983-1993. Numbers above the barsindicate the number of wells sampled. Well data supplied byGroundwater Section, BCMOELP.contributing areaC) L() r—9 0 0th0 00 0 0• median well water 1983-93 low flow surface water1535.5 Land use activities5.5.1 Installation of water wellsThe spatial extent of the use of the ground water resources is illustrated by thedensity of water wells (Figure 5.28). Wells outside of the study area of the Salmon Riverwatershed and Hopington Aquifer are not include on this map. The number of wells locatedon each surficial material type and the mean and median depths of the wells are listed inTable 5.12. Wells located in the glacial outwash have a median depth of 80 ft, comparedwith 94 ft on glacial marine materials, and 220 ft on marine materials.Table 5.12 Summary statistics for water wells in the Salmon River watershed andHopington Aquifer. The surficial material categories used are as outlined inFigure 3.3.Salmon River Watershed HopingtonAquifer onlysurficial materialalluvium glacial marine glacial glacialmarine outwash outwash12 468 139 723 935number of wells*9number of wells withdepth recorded** 12 458 130 694mean depth (ft.) 184 117 245 99median depth (ft.) 69 94 220 80minimum depth (ft.) 44 5 2 4maximum depth (ft.) 475 484 1328 617Some of the wells in the Hopington Aquifer are included in the summary for the SalmonRiver watersheddata sources used for well locations: WQCP records and CGDS well location thtabase,Groundwater Section, BCMOELPDepth of well used: depth to the bottom of the well screen when available, otherwise,recorded well depth.895106974617154>- crCc1 z NO Dz C oD No I— 0Li 0z I— LiLi ILi Z U)LJ - Li 0 LiJ :3 <I F-Li C H 0 C,.) :I...ci04-c,)a0Cuci)U)I.ci)‘-0ci)-]U-isc-)CuC/)c04- ci)(J)--± Q)Cj)ci.4- C)0L..QcoQ4-ocoCL()ci)c)U-?p: i7JTT•, i ,5 /1 . 9\ PJ INJ 1, L% INJ‘<‘ ‘1/’,.N // .-h1••} I-1 ii.._/ LiArki(I! :41555.5.2 Residential land useSeptic system installations are used as an index of residential development within thewatershed and are considered a potential point source of water pollution. The location ofseptic systems within the study area are illustrated in Figure 5.29 for the time period of theearly 1970s to the present. Records were not collected for areas outside the study area ofthe Salmon River watershed and Hopington Aquifer or for the portion of the watershedwithin Matsqui District. The location of septic systems is not evenly distributed throughoutthe watershed; there is a high density of residential development in Fort Langley and in thesouthwest portion of the watershed over the Hopington Aquifer (Figure 5.29).To assess the completeness of the septic system database, septic system records from1970 to 1993 are compared with the number of properties with addresses. Since propertiesare given an address when a building permit is issued, and there are almost no areas withinthe watershed with sewer service, the number of addressed properties is a good measure ofthe number of septic system installations. This analysis is done for the 17 TownshipSections (Master Legal 1:4000 map sheets) that fall entirely within the watershed. Listed inTable 5.13 are the percentages of addressed properties per map sheet that also arerepresented in the septic system database. An average of 54 percent of the addressesproperties in the watershed, ranging from 26 to 88 percent per map sheet, are represented inthe CFVHU-MOH septic system records. The properties not represented in the MOHrecords are either properties with buildings constructed prior to the early 1970s or missingfrom their records.Information on the number of dwellings and age of residential properties,summarized by Township Sections from British Columbia Assessment Authority (BCAA)Figure5.29LocationofsepticsystemsinstalledintheSalmonRiverwatershedandHopingtonAquiferbetween1970and1993.Datasource:CFVHU,BCMOH.•SEPTICSYSTEMSTREAMNETWORKROADNETWORKWATERSHEDBOUNDARY10-123KH 4J1157Table 5.13 Estimate of the percentage of dwellings in the Salmon River watershedrepresented in the septic system database, for selected Master Legal mapsheets within the watershed. The septic system database was created fromCFVHU-MOH records from 1970 to 1993.Master Legal Total properties No. of septic system percent of propertiesMap No. per map records with a septic systemper map record10-35 40 26 6510-36 23 14 6111-01 40 19 4811-02 136 62 4611-03 365 223 6111-10 286 204 7111-11 181 121 6711-12 110 51 4611-15 54 36 6711-16 61 30 4911-19 36 13 3611-20 150 90 6011-21 12 5 4211-28 23 6 2611-29 66 35 5311-32 489 172 3513-31 8 7 88sum 2080 1114 54records, is reported by Dayton & Knight et a!. (1994). The number of properties pertownship section in the BCAA database and on the master-legal maps are compared inTable 5.14. There are large discrepancies between number of properties per map sheet forsome of the maps. For those maps where there is reasonable agreement between propertynumbers, an estimate of the number of records missing from the septic system database wasobtained by comparing the number of dwellings listed in the BCAA database that have been158built in the last 20 years with the number of records in CFVHU septic system database(Table 5.14). According to this estimate, less than about 10 percent of the dwellingsconstructed in the last twenty years are missing from the CFVHU-MOH records, whichillustrates that the CFVHU database is reasonably comprehensive.Temporal trends in the installation of septic systems for the Coghian Creek sub-watershed, Salmon River watershed including Coghian Creek, and the Hopington Aquifer(including the area within and outside the Salmon River watershed) are illustrated in Figure5.30 and summarized in Appendix 21. The rate of increase of septic system installations forthe entire Salmon River watershed is greatest between 1973 and 1980, with an average ofabout 100 per year, and decreases to about 65 per year from 1981 to 1992. In the CoghlanCreek sub-watershed, the rate of growth is about 16 per year between 1973 and 1979, andincreases to about 28 per year between 1980-1989. For the Hopington Aquifer, of which 75percent is within the Salmon River watershed, the rate of growth is somewhat higher from1973 to 1981 with about 65 per year and dropping to about 46 per year from 1981-1991.These graphs illustrate the small cumulative changes in waste load within the watershed andHopington Aquifer during the past 20 years.5.5.3 Agricultural land useThe distribution of selected intensive agricultural activities, derived from the 1989land use map, is illustrated in Figure 5.31 and summarized in Table 5.15. The agriculturalactivities selected for this study include berry operations, and the animal operations ofpoultry, horses, beef, dairy, swine, sheep and mixed livestock fanns. Land-use informationwas not available for the 8 percent of the watershed in Matsqui District and small portion in159Table 5.14 Estimate of the accuracy of the septic system database for dwellings lessthan two years old. The septic system database was created from CFVHUMOH records from 1970 to 1993.Master Master BCAA database SepticLegal Legal systemMap No. Maps databaseno. of no. of no. of no. of percent of <20properties properties, properties, properties year old BCAAall years <20 years properties inold septic systemdatabase10-35 40 41 12 26 21710-36 23 14 6 1411-01 40 27 3 1911-02 136 159 54 62 11511-03 365 360 241 223 9311-10 286 309 235 204 8711-11 181 189 135 121 9011-12 110 84 29 51 17611-15 54 48 33 36 10911-16 61 58 15 30 20011-19 36 0 0 1311-20 150 2 1 9011-21 12 3 1 511-28 23 0 0 611-29 66 1 1 3511-32 489 5 1 17213-31 8 6 2 7 350sum 2080 1306 769 1114160Figure 5.30 Installation of septic systems from 1970 to 1992 in the Salmon Riverwatershed, Coghian Creek sub-watershed, and the Hopington Aquifer.Data from CFVHU-MOH records.—s— cumulative number1ci)EDC200175150125100755025075‘70 72 74 76 78 80 82 84 86 88 90 ‘92ranCoghian50Lci).0EDC70fl 76:8082 IPIt2501ci).02DCci)>D2DC)a).02DCci)>D20I-a).02DCD2D0I-ci).02C125100755025070 72 74yearFigure5.31SelectedintensiveagriculturalactivitiesintheSalmonRiverwatershedandHopingtonAquifer,derivedfromthe1989landusemap(Sawicki andRunka1990).ADRICULTURALLANDUSE18d1POULTRY,GAME,FURBEARERSHORSESOTHERANIMALSBERRIESOTHERAGRICULTUREAQUIFERBOUNDARY—STREAMNETWORK—WATERSHEDBOUNDARY10123KmH HTable 5.15 Summary of the subset of agricultural activities selected from the 1989 land-use map for this study. Land-use information from Sawicki and Runka(1990).Fruit, berry, and nut productionTree fruit productionBerry productionTree fruit or grape, nursery stockproductionNut productionA240 Growing plants in greenhousesA241 Growing vegetablesA270 Growing mushroomsTotal ha selected agricultural activitiesTotal Area (ha)131369228021162Land use Land Use Description Salmon R. Coghian Cr. Ropingtoncode watershed sub-watershed Aquifer(ha) (ha) (ha)8 83 1 3229 89 20318A130A131A133A134A135A200A210A211A212A213A214A215A220A221A222A223A225A231A232A2332 025 7 514 17 117 2 24 1 27 1 2Site Agricultural ActivitiesHousing livestockHousing dairy animalsHousing beef animalsHousing horsesHousing swineHousing sheepOutside animal feeding andholding areasDairyBeefHorseSheepHousing poultryHousing game birdsHousing fur bearing animals21822298142911641936706752179161711124 0282 5301487 2927Percent of Area 11 19 18163the NW of the watershed (Figure 3.8). The selected livestock and productive-landagricultural activities represent only 11 percent of the watershed. Using the same 1989 landuse maps, Watts (1992) found that the dominant land use type was agriculture, which covers50 percent of the watershed. The land use categories of “growing forage crops” and“grazing” were not included in this study because they were considered to be less intensiveagricultural activities. This is the reason for the difference in reported percentages ofagricultural activities between the two studies.The agricultural activities selected for inclusion in this study are more commonlyfound in the Coghlan Creek sub-watershed and the Hopington Aquifer than in the watershedas a whole, indicating more intensive agricultural operations in these areas. Nineteenpercent of Coghian Creek sub-watershed and 18 percent the Hopington Aquifer, areoccupied by these intensive operations compared to 11 percent for the overall watershed.For example, berry farming is almost exclusively found on the well-drained, coarse-texturedsoils above the Hopington Aquifer. The most common livestock land use category withinthe watershed is outside animal feeding and holding areas for horses, representing 3.7percent of the land area of the watershed, and 32 percent of the selected intensiveagricultural operations. Outside animal feeding and holding areas, with no specified animal,is the next most common livestock category at 2.7 percent of the watershed. This moregeneral livestock category includes farms with mixed livestock (Sawicki and Runka 1990)and may also contain a significant number of horses. Berry production occupies 2.8 percentof the watershed. These three land-use categories represent 80 percent of the selectedintensive agricultural activities within the watershed.1645.6 Spatial relationships between land use and surface water qualityTo examine relationships between land use and water quality, the land-use activitieswithin the watershed are summarized by the amount of each land-use type within eachcontributing area (Figure 4.3). Two measures of intensity of land use are calculated: i) theproportion of land with a given land use within each delineated contributing area, hereafterreferred to as the density index; and ii) the cumulative proportion with a given land use, i.e.the proportion of a given land use in all the contributing areas upstream of the watersampling station, referred to as the cumulative density index. The density index iscalculated based on grouping the contributing areas into 13 non-overlapping areas asoutlined in Table 5.16. The contributing areas were grouped, where possible, to provideland areas of approximately equal size. For the cumulative density index, the Salmon Riverwatershed, including the Coghian and Davidson Creek sub-watersheds, is divided into 10land areas. The cumulative density index is also separately calculated for the CoghianCreek sub-watershed, divided into three land areas. The contributing areas of the smalltributaries have been grouped with the contributing area of the next downstream mainstemstation. The areas of the watershed used to calculate the cumulative density index for theland area upstream from each surface water sampling station are listed in Table 5.16. Forboth indices, contributing areas S00 and S16 have been excluded, since no land useinformation was available for S00 and no surface water quality measurements were made forS16 at the mouth of the Fraser River.The concentration of N03-N in the surface water during low flow conditionsillustrates the greatest spatial variability of the water quality parameters tested within thewatershed (Figure 5.11), therefore relationships between land use and N03-N are examined.165Table 5.16 Grouping of contributing areas for examining relationships between land useand water quality.Contributing area (ha) cumulative area percent ofarea upstream (ha) watershedFor density index:Salmon River watershedSO1-02 328 4S03 447 6S04-05 548 7S06-07 598 7S08-09 260 3Sb-il 630 8S12 585 7S13-14 619 8S15 1142 14Coghlan Creek sub-watershedC01-03 592 7C04-05 460 6C06-07 435 5Davidson Creek sub-watershedDO 1-02 403 5watershed total 7047 88For cumulative density index:Sahnon River watershedSOl 188 188 2S01-02 140 328 4S03 447 775 10S04-05 548 1324 17S06-07 598 1922 24S08-09 260 2181 27COl-SlO 1633 3814 48Sib 484 4298 54S12-S14 1204 5502 69D01-S15 1545 7047 88Coghian Creek sub-watershedC01-O3 592 592 7C04-05 460 1052 13C06-C07 435 1487 19166The relationships between surface water N03-N and the density index are examined bothgraphically and by computing the Spearman rank correlation coefficient to obtain a measureof the degree of association between the land use and N03-N variables. The relationshipsbetween surface water N03-N and the cumulative density index are examined graphicallyonly, because the use of correlation analysis was considered inappropriate due to the highdegree of spatial autocorrelation and lack of independence between each measure ofcumulative density. The land-use activities that are examined include the density of septicsystems, and the proportion of the land with all the selected agricultural activities. Themapped agricultural activities have also been grouped into the categories of berry farming,poultry operations, horse operations, and other animal operations (beef, dairy, sheep, swine,and mixed livestock farms).For the graphical presentation of the density index, the left-hand side of the Figureshows data for the Salmon River watershed, excluding the Coghlan Creek and DavidsonCreek sub-watersheds, and these two sub-watersheds are presented on the right-hand side ofthe Figure. For the cumulative density index, the left hand side of each Figure shows datafor the entire Salmon River watershed, including the Coghlan Creek sub-watershed, and theCoghlan Creek sub-watershed is presented separately on the right side.The concentration of N03-N in surface water during low flow conditions, averagedfor the three sampling periods (Figure 5.11), is used to examine relationships between landuse and N03-N in surface water. The N03-N concentration shown is the value from themost downstream station for a given delineated land area. Since the grouping of thecontributing areas is slightly different for the calculation of the density index compared tothe cumulative density index, there are small variations in the pattern of N03-N plotted in167the Figures with the density index, as compared to the cumulative density index.5.6.1 Residential Land UseThe relationship between N03-N and septic systems is examined in detail since theinformation on the number of septic systems installed in the watershed in the last twodecades is the best data set on intensity of land use activity collected during this study. Thedensity of septic systems installed since the early 1970s is used as a measure of the intensityof residential development within the watershed. In the following discussion the density ofseptic systems reported refers only to installations since the early 1970s and in some areas isa substantial underestimation of the total density. The number of septic systems in eachcontributing area is summarized in Table 5.17 and illustrated in Figure 5.32. Within somecontributing areas, there are portions that contain both a high and low density of septicsystems (e.g. S07, C06, S15), resulting in a lower average density for that contributing area.The density of septic systems per contributing area ranges from 0.03 ha1 in predominantlyagricultural areas to 1.29 ha1 near the mouth of Coghlan Creek, with an average density forthe entire watershed of 0.2 ha’ (Table 5.17). The contributing areas with the highest densityof septic systems are S06-08, C0-07, ranging from 0.4 to 1.3 septic systems ha1. These fivecontributing areas contain 44 percent of the septic systems within the watershed on 15percent of the land area. This higher density of septic systems is located on the glacialoutwash materials that make up the Hopington Aquifer. These numbers illustrate a range inseptic system densities both within the watershed and within a contributing area.168Table 5.17 Septic systems installed in each contributing area between 1970 and 1993.Data from CFVHU-MOH records.size number of septic % of total inContributing area (ha) systems number per ha watershedSalmon River headwatersS00 649.8 NASOl 187.9 5 0.03 0.3S02 140.0 10 0.07 0.6S03 447.4 13 0.03 0.8S04 209.5 28 0.13 1.7S05 339.0 63 0.19 3.7S06 95.7 25 0.26 1.5S07 502.1 268 0.53 15.8S08 90.8 72 0.79 4.3S09 168.8 131 0.78 7.7Coghian Creek sub-watershedCOl 287.6 38 0.13 2.2C02 16.5 1 0.06 0.1C03 287.9 55 0.19 3.2C04 346.0 82 0.24 4.8C05 113.8 5 0.04 0.3C06 95.4 123 1.29 7.3C07 339.7 143 0.42 8.4Salmon River middle reachesSlO 146.0 28 0.19 1.7Sil 484.2 37 0.08 2.2S12 585.2 76 0.13 4.5S13 561.4 92 0.16 5.4S14 57.4 11 0.19 0.6Davidson Creek sub-watershedDOl 123.9 18 0.15 1.1D02 278.7 18 0.06 1.1Salmon River lower reachesS15 1142.2 231 0.20 13.6S16 324.1 121 0.37 7.1Sum 8021.2 1694 100Figure5.32LocationofsepticsystemsinstalledintheSalmonRiverwatershedandHopingtonAquiferbetween1970and1993inrelationtothedelineatedcontributingareas.•SEPTICSYSTEMCONTRIBUTINGREBOUNDARYSTREAMNETWORKWATERSHEDBOUNDARY10123Km1705.6.1.1 Relationship between septic systems and surface-water N03-NDensity IndexWithin the Coghian Creek watershed, there is good agreement between the increasein density of septic systems and the increasing concentration of N03-N in surface waterfrom the headwaters to the confluence with the Salmon River (Figure 5.33). Within theSalmon mainstem, N03-N concentration increases between S03 and Si 1, and there is acorresponding increase in septic system density from contributing area S04-05 to S08-09.Contributing area S 10-il does not follow this pattern, with the density of septic systemsdropping to 0.i ha1, while the N03-N level remains high. The contribution of CoghlanCreek, with both high septic system density and N03-N concentration at the mouth, is notincluded in the calculation of the density index of 510-li. During low flow conditions, thedischarge of Coghlan Creek, at station 5, is about half of the discharge at station 4 on theSalmon River; therefore, Coghlan Creek is a large contributor to the N03-N concentration inthe Salmon River below the Coghlan confluence. The highest septic system densities occurin S04-09, prior to the N03-N peak concentration measured at Sb, which suggests a spatiallag between the higher density of septic systems and the observation of elevated N03-N inthe surface water downstream. Septic system density remains below 0.2 ha1 in contributingareas Sib through S15, with a corresponding gradual drop in N03-N concentrations in thesurface water.Thirty percent of the land area of the watershed is composed of shallow soilsoverlying glacial outwash sands and gravels. Since the overlying soils are highly permeableand shallow, bio-remediation of the effluent from septic systems is less likely to occur thanon finer textured, deeper soils. There is a higher susceptibility of ground waterx ci>-D C-‘ co C ci>c)U)900—CDoo0000-J zx ci) C >% 4-. Cl) C ci) a) > E 0C) oo04000-J 0)z ci) IFigure5.33Relationshipsbetweenresidentiallanduseandnitrate-Ninsurfacewaterduringlow-flowconditionsintheSalmonRiverwatershed.Themeasureofresidentiallanduseisthenumberofsepticsystemsinstalledbetween1970and1993.Seesection5.6forexplanationofthedensityindices.C1)U)0)—U)o0QQ‘øcth0C0oooo(I)ØU)Cl)0heactwateimouthcontributingarea.Lt)U)-0)0oo0C” 9 0 Q.suacewaterNO3contributingareaH —3H172contamination from septic systems on the glacial outwash parent materials that make up theHopington Aquifer, compared with septic systems on the finer textured glacial marine ormarine parent materials. The density of septic systems on glacial outwash parent material ineach delineated area is similar to the pattern observed when all surficial materials areconsidered, with a lag observed between peak septic system density and peak surface waterN03-N (Figure 5.33A and 5.33B). Differences are evident in contributing areas S12-14,where there is no glacial outwash surficial material, and a corresponding decrease in N03-Nconcentration is also observed. The high septic system density on glacial outwash in S15 isdue to the residential development in Fort Langley, yet there is not a corresponding increasein N03-N. Since Fort Langley is near the Fraser River, it is not known whether thegroundwater from Fort Langley flows into the Salmon or the Fraser River.Cumulative Density IndexThe land area within the watershed that is upstream of a given sampling station maypotentially contribute to the water quality at that station, and not only the adjacent land useimmediately upstream of the station. The cumulative septic system density upstream ofeach sampling station and the N03-N in the surface water are presented in Figure 5.33C.Within the Salmon mainstem, the cumulative density increases rapidly between stations S04-S05 and SlO, then gradually decreases to station S15, following closely the pattern ofN03-N concentration in the surface water. The gradual decrease in cumulative densityresults from the addition of land areas with a low density of septic systems, movingdownstream in the watershed. Similarly, in the Coghlan Creek sub-watershed the pattern ofcumulative septic density closely follows the pattern of N03-N, with a rapid increase in173N03-N between C04-05 and C06-07 occurring with a 1.8 fold increase in cumulative septicdensity. The pattern of the cumulative density index matches the pattern of N03-N insurface water more closely than the density index, and illustrates the cumulative effect ofupstream land use activities on water quality. As with the density index, a spatial lagbetween the increase in cumulative septic density and N03-N in surface water is againobserved.The cumulative density of septic systems on glacial outwash (Figure 5.33D) shows asimilar pattern to cumulative density of septic systems on all parent material types (Figure5.33C), with the spatial lag again evident. The most notable difference is that thecumulative density is higher when only septic systems on glacial outwash materials isconsidered, with the maximum at SlO of 0.45 ha1 on outwash as compared to 0.29 ha’when all parent material types are considered. The highest cumulative density of septicsystems is 0.50 observed at C06-07 at the mouth of the Coghlan Creek. These numbersillustrate the higher density of residential development on the glacial outwash materialscompared to all surficial materials within the watershed.In summary, the increase in surface water N03-N corresponds with an increase inseptic system density and cumulative septic system density, both on all surficial materialtypes and on glacial outwash material. The downstream lag between the increase in N03-Nand the increase in septic system density is likely due to the direction of ground-water flow.The pattern of surface water N03-N more closely resembles the pattern of the cumulativedensity, which indicates that the water quality at a station is not only influenced by the landarea immediately upstream from the station (the adjacent contributing area), but also reflectsthe land-use activities of the watershed upstream from the station.1745.6.2 Agricultural activitiesThe proportion of land within each contributing area with selected agriculturalactivities is used as a measure of intensity of agricultural land use (Table 5.18). Thecategory of “all agriculture” is the sum of selected agricultural activities that are listed inTable 5.15 and illustrated in Figure 5.31. The agricultural operations are also grouped intothe following categories: 1) berry production; 2) housing poultry, game, and fur bearinganimals; 3) housing or outside animal and feeding areas for horses; and 4) housing oroutside animal and feeding areas for other animals (livestock, beef, dairy, sheep, and swine).The agricultural operations selected for this study are not distributed evenlythroughout the watershed. The percentage of each contributing area with these agriculturalactivities varies from about 4 percent to 27 percent (Table 5.18). Land areas identified ashorse operations are the most common animal operation, covering 3.9 percent of the landarea of the watershed, and range from 0 to 12.6 percent of each contributing area. Landunder berry farming ranges from 0 to 14.2 percent of the contributing area, while operationswith livestock, dairy, beef, or sheep range from 0 to 11.1 percent of the contributing area.The percent of each contributing area with poultry operations is usually around 1 percent,since poultry operations are generally on a small land base. Relationships between agricultural activities and surface-water N03-NDensity IndexIn Figure 5.34A the “all agriculture” density index is compared with surface-waterN03-N, with the Salmon River shown on the left (S01-15) and Coghlan and DavidsonCreeks on the right (C01-07, D01-02). The highest density indices in the Salmon175Table 5.18 Agricultural land use in each contributing area, as the percentage eachcontributing area.Poultry, Horses Other AllBerryContributing Areaoperations game, fur animals agriculturebearersarea (ha) (%) (%) (%) (%)(%)Salmon River headwatersS00 649.8 NA NA NA NA NASOl 187.9 0.0 0.8 1.6 4.4 7.3S02 140.0 0.3 0.0 3.0 0.8 4.2S03 447.4 0.0 0.0 4.3 2.8 7.0S04 209.5 1.2 0.0 8.0 8.6 17.8S05 339.0 4.8 0.2 12.6 5.1 23.5S06 95.7 2.4 0.0 10.4 3.4 16.2S07 502.1 1.1 1.3 8.4 3.7 16.9S08 90.8 0.0 0.5 0.9 10.0 12.2S09 168.8 0.0 0.0 3.7 1.0 4.7Coghlan Creek sub-watershedCOl 287.6 0.9 0.0 2.6 11.1 15.9C02 16.5 0.0 0.0 0.0 3.9 3.9C03 287.9 5.9 0.1 12.2 5.8 26.5C04 346.0 6.8 0.7 4.3 5.7 17.8C05 113.8 7.7 0.9 0.8 1.3 10.7C06 95.4 4.9 0.0 0.8 0.2 5.8C07 339.7 9.5 1.0 10.6 2.3 23.7Salmon River middle reachesSb 146.0 0.0 1.1 5.5 8.1 15.4Sli 484.2 14.2 0.4 4.1 3.2 22.4S12 585.2 0.0 0.1 1.7 2.7 4.7S13 561.4 2.5 0.2 1.1 1.2 7.3S14 57.4 0.0 0.0 10.4 0.0 10.4Davidson Creek sub-watershedDOl 123.9 0.0 4.7 0.0 10.2 15.7D02 278.7 9.5 0.1 1.2 5.3 16.1Salmon River lower reachesS15 1142.2 0.4 0.3 1.4 2.7 5.5S16 324.1 0.0 0.0 1.3 3.7 5.2Total 8021 2.9 0.4 3.9 3.6 11.5NA = land-use infonnation not available.x ci) >•1 U) ci)-o ci) > D E C)-j 0)z IFigure5.34Relationshipsbetweenagriculturallanduseandnitrate-Ninsurfacewaterduringlow-flowconditionsintheSalmonRiverwatershed.Themeasureofagriculturallanduseisthenumberofhectaresofselectedagriculturalactivities.Seesection5.6for explanationofthedensityindices.-J 0)z dx.1V .9 >-.ci)U)csCas D U)C%JCU)r—oU)oo00•—CO4thc0C)ooooCO(0(00)U)heactwatersc)U)r—900—.CD0000009 0 Dmouthcontributingarea—C%Jc)U)0)—-U)000000000,L‘—00(00_•_..surfacewaterNO3C)U)r—00000thOOo 0contributingarea177River occur between contributing areas S04-05 and Sb-il, which corresponds with theincrease in surface water N03-N. The higher densities of 0.17 to 0.21 in contributing areasS04-S07 precede the increase in N03-N in the surface water, a spatial lag similar to thatobserved with the septic system densities. The higher densities of 0.16 to 0.20 are observedin all three land areas in the Coghlan Creek and do not correspond with the changing N03-Nin the creek’s surface water. The association between the pattern of N03-N in surface waterand the agriculture density index is weak, for both the Salmon River watershed and theCoghlan Creek sub-watershed.Greater agreement is observed between the pattern of N03-N in surface water and thedensity of the selected agricultural activities on glacial outwash parent material in theSalmon River watershed (Figure 5.34B). The increase in the density of agriculturaloperations between S04-05 and Sb-li followed by a decrease to S15 corresponds with thepattern of increase and decrease in N03-N in the surface water. This pattern is not observedin the Coghlan Creek sub-watershed, where the density of agricultural operations decreasesas the N03-N increases. The poor agreement between agriculture and surface water N03-Nin the Coghlan Creek suggests an alternate source of N03-N in this sub-watershed.When the agricultural operations are considered in the separate categories of berry,poultry, horse, and other animal operations, only weak associations are observed between thedensity index and the pattern of surface water N03-N (Appendix 22 and 23). Theagricultural activities on all surficial materials show a reasonable association between berryoperations and surface water N03-N, whereas the higher densities of poultry, horses, andother animals precedes the increase in surface water N03-N. Similar patterns are observedwhen only agricultural activities on glacial outwash surficial materials are considered178(Appendix 23).Cumulative Density IndexThe cumulative density of the selected agricultural activities upstream of eachsampling station and the N03-N in the surface water, is presented in Figure 5.34C. Withinthe Salmon mainstem, the cumulative density increases between stations S04-S05 and Sil,then gradually decreases to station Si 5. The increase in density of agricultural operationsprecedes the increase in surface water N03-N with an even greater lag between the increasein density of agricultural activities and N03-N than is observed with the septic-systemcumulative density. In the Coghian Creek sub-watershed the cumulative agriculture densityindex remains high, at about 0.20 ha1, and does not follow the pattern of N03-N. Withinthe Salmon River watershed, the agriculture cumulative density index matches the pattern ofN03-N in surface water more closely than the density index, a pattern that was alsoobserved for the septic systems. The association between the cumulative density index andN03-N is weaker when only the agricultural activities on glacial outwash materials areconsidered (Figure 5.34D). The cumulative density continues to increase despite the drop inN03-N in the Salmon River.Associations between N03-N and cumulative density are observed for some of theseparate categories of berry, poultry, horse, and other animal operations. (Appendix 24 and25). For berry operations, both on all surficial materials and on glacial outwash materialsonly, the increase in cumulative berry density lags behind the increase in N03-N, whichsuggests berry operations are not the major source of N03-N to the Salmon River. Asobserved with the density indices, the increase in cumulative density index for the poultry,179horses, and other animals, precedes the increase in N03-N. This pattern is observed whenthe agricultural activities on all surficial materials, as well as on glacial outwash materialsonly, are considered.In summary, associations between agricultural activities and surface water N03-N areobserved in the Salmon River but not in the Coghian Creek sub-watershed. The pattern ofthe density index on all surficial materials and glacial outwash alone, as well as thecumulative density index on all surficial materials follows the pattern of surface waterN03-N. These associations are not as strong as those observed with septic systems, andthere is a greater spatial lag between the increase in agricultural activities and the increase inN03-N compared with the lag observed for septic systems. This difference may be due tothe septic systems being a more concentrated and localized source of N03-N compared tothe dispersed, nonpoint source of N03-N from agriculture.5.7 Relationships between land use and ground water qualityThe median concentration of N03-N measured in well water between 1983 and 1993for each delineated land area (Figure 5.26) is used to examine relationships between landuse and N03-N in ground water. The accuracy of the median as representative of theground water concentration varies between the contributing areas, since each area in notsampled with the same frequency. When all surficial materials are considered, the numberof well-water samples varies from 1 to 39 per land area. For example, contributing areaS03 is represented by only one well measurement, therefore considered a poor representationof ground-water N03-N compared to contributing area S06-07, represented by 39measurements. For the wells located on the glacial outwash surficial materials, the number180of well samples per contributing area varies from 1 to 22 samples, with five of the landareas represented by three or fewer measurements and two land areas with no measurementsince there are no glacial outwash surficial deposits there (Appendix 26).The pattern of median well-water N03-N closely follows the pattern of density ofseptic systems in both the Salmon River watershed and Coghlan Creek sub-watershed(Figure 5.35A). The highest densities of septic systems occur in contributing areas S06-07,S08-09, and C06-07, corresponding to the highest median well-water N03-N concentrations.In the Davidson Creek sub-watershed the septic system density is low despite a high medianwell-water N03-N concentration, suggesting a different N03-N source in this small sub-watershed. A similar pattern is also observed when only wells and septic systems on glacialoutwash parent materials are considered (Figure 5.35B). The median N03-N concentrationis usually higher when only those wells on glacial outwash are considered. High nitratevalues that do not correspond to high septic system densities occur in areas SlO-il andD01-D02.An association between N03-N in well water and the agriculture density index is notas clearly distinguishable as the association between N03-N and the septic system densityindex. There is little observable association between the pattern of N03-N in well-water anddensity index of agricultural activities on all surficial material types (Figure 5.35C), althoughwhen only glacial outwash materials are considered there is closer agreement between wellwater N03-N and the agricultural density index. Similarly, no clear trends are observablebetween median well-water N03-N and the individual agricultural activities of berry, poultry,horses, and other animals (Appendix 27 and 28). Although there is not a high degree ofassociation between well-water N03-N and agriculture when the whole watershed isC’.iC’)CC)r-.O‘4)0999,’-•weIINO32øocj,10000wellNO3onoutwash-J z I‘10.0a)-D a) EFigure5.35x a)-D0.> 4- Cl) 0) VRelationshipsbetweenresidentialandagriculturallanduseandnitrate-NingroundwaterintheSalmonRiverwatershed.Thewell-waternitrate-Nisthemediannitrate-Nvaluefromwellstestedwithineachcontributingarea,frommeasurementsmadebetween1983-1993.Seesection5.6for explanationofthedensityindices.-J C)z dx ci) VCCci)>Ct5Cl)_5-C cci 0) EmouthcontributingareaC’))F.-.900,- 000C%j 9 0 0contributingarea182examined, high well-water N03-N coincides with high agricultural density index for SlO-liand DO1-02, areas with low densities of septic systems (Figure 5.35B and D). These resultssuggest different N03-N sources in different portions of the watershed.5.8 Spearman rank correlation coefficients between land use and N03-NThe degree of association between N03-N in water and the density indices for theresidential and agricultural activities are compared using Spearman rank correlationcoefficients (Table 5.19). N03-N in surface water, well water, well water from shallowwells less than 100 feet, and well water from wells located on glacial outwash surficialmaterials are considered. The null hypothesis is that there is no association or a negativeassociation between N03-N and the land use density indices, with the alternate hypothesisthat there is a positive association. Correlations significant at cx=0.05, for a one-tailed test,are indicated by an asterisk in Table 5.19.A positive association is observed between surface water N03-N and the densityindex for poultry and berry operations, and between berries and all agricultural operationswhen only those land uses on glacial outwash are considered. Positive correlations areobserved between well water N03-N and berry operations on all surficial material types, andwith all animal operations when only the shallow wells (<100 ft) are considered. WhenFort Langley is excluded from the analysis, a significant relationship is observed betweenthe septic system density and N03-N in well water (Appendix 29).Although there were general trends observed between agriculture and residential landuse, with the statistical analysis there were fewer significant correlations, as measured usingSpearman rank correlation coefficients. The lack of many simple relationships between land183Table 5.19 Spearman Rank correlation coefficients illustrating associations betweenmeasures of land use activities and nitrate-N in water in the Salmon Riverwatershed.Land use type surface-water well-waterwell-water well-waterN03-N N03-N N03-N N03-N<=100 ft on outwashn=13 n=13 n=12 n=10surface water N03-N 1.000well water N03-N 0.560* 1.000well water N03-N, <101 ft 0.147 0.517* 1.000well water N03-N, outwash only 0.821* 0.760 0.285 1.000land use on all surficial material typesdensity of septic systems 0.429 0.258 0.070berries 0.556* 0.245 0.207poultry and fur bearers 0.571* 0.582* 0.315horses -0.176 0.099 0.406other animals -0.104 0.099 0.476all animals 0.022 0.236 0.559 *other agriculture -0.088 -0.29 1 -0.37 1all agriculture 0.247 0.220 0.448land use on glacial outwash onlydensity of septic systems 0.343 0.348 0.261 -0.024berries 0.622* 0.266 -0.089 0.492poultry and fur bearers 0.353 0.298 0.261 0.098horses 0.061 0.116 0.430 -0.480other animals -0.127 -0.033 0.303 -0.578all animals 0.011 0.055 0.394 -0.559other agriculture -0.119 -0.090 0.086 -0.586all agriculture 0.644* 0.228 -0.014 0.365* indicates significant correlation at a =0.05 for a one-tailed test.184use and N03-N is likely due to a number of factors. The spatial lag between the detectionof high levels of N03-N in well water and its measurement in surface water (Figure 5.26)make it difficult to detect relationships between land-use activities and surface water N03-Nusing correlation analysis. Due to the northward ground-water flow patterns, the N03-Nobserved in surface water may not be the result of the land-use activities immediatelyupstream. Only weak correlations may be measured even though spatial trends may be seenwith graphical analysis. In the correlation analysis, each measurement is considered to beindependent, and spatial position is not considered.The grouping of the contributing areas into land units is somewhat arbitrary, andresults in land areas with mixed land uses. The combination of areas with both low andhigh density land uses (e.g. septic systems Figure 5.32) will result in weaker associationsbetween the land use and water quality variables. Errors in the mapping of land use,particularly the categories of agricultural activities, and in the delineation of contributingareas will also reduce the association observed between variables. The degree ofaggregation of the land areas used also has a large influence on the association between theland use and NO3-N. The watershed was grouped into 23, 13, and 10 land areas, and eachgrouping resulted in marked changes in the correlation coefficients. Depending on thenumber of land areas used, the land use variables that were significantly correlated withNO3-N in the water changed (Appendix 29).The graphical examination of trends between the N03-N in water and the land useindices is more likely to demonstrate possible sources of the N03-N than the statisticalcorrelation analysis. The presence of a spatial lag in the appearance of N03-N in surfacewater and the spatial autocorrelation between adjacent land areas, both in the land use and185water quality variables, limits the usefulness of the statistical approach.5.9 The effect of agriculture versus residential land use on N03-NDistinguishing the relative importance of the different land uses as N03-N sources isnot possible from the approach used in this study. The highest density of septic systemsoccurs in contributing areas S06-09 (Figure 5.33A), and the higher agricultural densityoccurs in S04-07 and S 10-11 (Figure 5.34B), and illustrates that the area within thewatershed with the greatest intensity of both land-use types occurs between S04 and Sli.This area also corresponds with the increase in N03-N in both the surface and ground water.In the Salmon River watershed, and especially on the sensitive glacial outwash parentmaterials of the Hopington Aquifer, rural residential and agricultural activities exist side byside (Figure 5.36); therefore, it is not possible to segregate sub-watersheds into categoriesbased on a dominant land-use type.The pattern of septic system density more closely matches the pattern of N03-N insurface and ground water than does the pattern of agricultural activities. The number ofseptic systems within a land area is considered a better indicator of residential nitrateloading than hectares of agricultural land use is as an indicator of agricultural nitrateloading. The more comprehensive information available on septic systems may explain thehigher association observed between septic systems and N03-N than between agriculturalland use and N03-N. The number of hectares of outside animal feeding and holding areasor animal housing does not give any measure of the number of animals at each farm. Aninventory of the number of animals per farm and the rates of fertilizer application wouldimprove the quality of information on intensity of agricultural land use. An inventory ofFigure5.36SelectedagriculturalandresidentiallanduseactivitiesontheHopingtonAquifer.Datasources:SawickiandRunka(1990)andCFVHU,BCMOH.6RICULTURcLLANOUSE1989___POULTRY,GAME,FURBEARERSHORSESOTHERANIMALSBERRIESElOTHERAGRICULTURE•SEPTICSYSTEM—STREAMNETWORKAQUIFERBOUNDARYWATERSHEDBOUNDARY10123KmIIIH187farms within the watershed is currently under way (Barbara Wernick pers. comm.).Inclusion of land used for grazing and for forage production should also be included in theassessment of agricultural activities, since these land uses can also be contributors to nitratecontamination of groundwater under certain management conditions (Ball et a!. 1979, Balland Ryden 1984, Barraclough et a!. 1983, Owens et al. 1994).Since relationships between N03-N and both septic systems and agricultural activitiesare observed, it is likely that both are probably important contributors to nitrate load.Within the watershed, it appears that the relative importance of these two sources variesbetween locations. For example, a high concentration of N03-N is observed in DavidsonCreek, a sub-watershed with little recent residential development, in contrast to CoghlanCreek that has both residential and agricultural development. Rather than attempting toseparate the impacts of septic systems from agricultural activities, it may be more useful tomodel the cumulative loading from all nitrogen sources. These results can then be used toestablish monitoring sites and for the management and remediation of nitrate contaminationof ground water.1886. SUMMARY AND CONCLUSIONSThe objectives of this study were to provide information on water quality and toidentify relationships between land use, surficial materials, and the water quality within theSalmon River watershed. The Salmon River watershed is a small watershed at the rural-urban fringe of the rapidly developing Greater Vancouver Regional District (GVRD) and hasundergone substantial land-use change in the last 20 years. This study uses trace metals andthe nutrients, nitrate and total phosphorus, as indicators of water quality. Relationshipsbetween water quality, surficial geology, and land-use activities are examined using awatershed approach combined with GIS techniques.6.1 Spatial and temporal variability in total trace-metal content of streambedsedimentsThe total trace-metal concentrations in the fine-fraction (< 63 um) of streambedsediments were used as an indicator of accumulated metal contamination over a period oftime. The metal concentration in the sediments was compared with the background metalconcentration in surficial materials from within the watershed and no evidence was found ofelevated total-metal concentrations within the sediments for the metals Zn, Cr, Cu, Co, Ni,and Mn. It is difficult to use the fine-fraction sediment as an integrator of metalcontamination in the middle reaches of the Salmon River, since accumulations of finefraction material, through several seasons, often do not occur in these gravelbed reaches ofthe river.Up to a five-fold increase in trace-metal concentration (Zn, Cu, Co, Pb, Mn) wasobserved between the sediments sampled in the 1970s and in 1991. However, since the1891991 metal concentrations were not considered to be above background levels, thesedifferences likely reflect the natural variability in trace metals due to the physico-chemicalproperties of the sediments, variable streamfiow regimes, and the sample preparation anddigestion methods. The results from this study illustrate the influence of sample preparationand analysis methodology on the concentration of trace metals in sediments and thedifficulty of making meaningful comparisons between analyses conducted with differentmethods.6.2 Spatial variability in water qualityThe variability in surface water chemistry was measured by taking grab samplesduring summer low flow conditions and winter high flow conditions. During the summerconditions the stream flow is primarily from ground water. Measures of specificconductance, chloride, and N03-N were higher during low flow conditions than at high flow,whereas the total phosphorus concentrations were lower at most stations during low flowconditions than at high flow. Specific conductance measures usually ranged from 5.0 to16 mS m1 during low flow conditions, compared to 5.0 to 8.0 mS m1 during high flowconditions. During low flow conditions chloride measurements are usually below 20 mU’,and are between 4 and 14 mg U’ during high flow. The maximum nitrate concentrationmeasured during low flow was 7.1 mg U1 compared to 5.7 mg U’ during high flow.Distinctive spatial patterns of specific conductance, chloride, and N03-N in thesurface water are observed in the watershed, with the pattern more pronounced during lowflow conditions than high flow conditions. The spatial patterns reflect both changes insurficial geology and land use within the watershed. The concentration of N03-N is low in190the headwaters of the Salmon River and Coghian Creek and gradually increase as the riverflows through the glacial outwash parent materials of the Hopington Aquifer. The highestN03-N concentrations are observed in the middle reaches of the Salmon River and inCoghlan Creek at the confluence with the Salmon River. Both specific conductance andchloride are higher in the headwaters of the Salmon and Coghlan, reflecting the glacialmarine parent materials. As observed with the N03-N measurements, the specificconductance and chloride gradually increase in downstream direction at the stations on theglacial outwash parent materials and are likely due to anthropogenic additions. Thesignificant correlation between chloride and nitrate at the stations on glacial outwash parentmaterials suggests a common source of these ions. The higher measurements of chlorideand conductivity in the lower reaches of the Salmon River likely reflect the marine andglacial marine parent materials.The concentration of total phosphorus was usually higher during high flow than lowflow conditions. During low flow the total phosphorus in the Salmon River mainstem wasusually less than 0.06 mg U1, except in the headwaters, compared with high-flow valuesusually below 0.1 mg U’. During high flow the spatial pattern of total phosphorus reflectsthe surficial geology within the watershed, with higher concentrations observed at thestations draining the fine-textured glacial marine and marine materials. There are localizedareas in the headwaters of the Salmon and at some of the tributaries where total phosphorusconcentrations are high.Of the approximately 400 wells within the Salmon River watershed and HopingtonAquifer have been tested for NO3-N during the past twenty years seven percent of thesewells had N03-N concentrations above the Canadian drinking water guideline of 10 mg U’.191Wells with N03-N levels above 1 mg U1, indicative of human influence, are located in threegeneral areas: The Hopington Aquifer, Fort Langley adjacent to the Fraser River, and asmall concentration of wells in the headwater area of Coghlan Creek. Both Fort Langleyand the Hopington Aquifer are areas of glacial outwash surficial deposits, whereas theCoghian Creek headwater area is composed of glacial marine deposits.The peak N03-N concentration during low flow conditions in the Salmon Rivermainstem is not observed at the river stations located on the glacial outwash materials of theHopington Aquifer, where the frequency of N03-N contaminated ground water is thehighest. The peak N03-N is located downstream at station 23 on the marine parentmaterial, and probably reflects the northward flow of ground water.6.3 Temporal variability in water qualityComparing historic water quality data with the current measurements is difficultbecause of differences in sampling frequency, flow conditions, location of sampling stations,methods of analysis, and detection limits. Due to these difficulties, a statistical comparisonwas not made between present samples and samples from the I 970s. For the surface waterquality variables measured, small differences were observed between the concentrationsmeasured in 1974-75 and the present. Localized increases in N03-N were observed inCoghlan Creek at the confluence with the Salmon River and in Davidson Creek. Smallincreases in specific conductivity and chloride at the stations in the glacial outwash depositsmay be due to differences in the flow conditions, analytical methods, or changes in land use.Wells on the Hopington Aquifer monitored regularly by Environment Canada andlater by BCMOELP exhibit high seasonal and temporal variability, with some showing an192increasing, and others a decreasing, trend with time. Given the dynamic and variable natureof N03-N in these wells, it is difficult to generalize from these individual well records anddetermine if a change in N03-N has occurred in the Hopington Aquifer. Using records ofapproximately 400 wells tested between 1970 and 1993, the median N03-N concentration inwells was calculated for different areas of the watershed, for the wells sampled between1972 and 1982 and between 1983 and 1993. A three-fold increase in median N03-Nconcentration is observed in well sampled post-1983 compared to pre-1983, for those areasin the middle reaches of the Salmon River, on the Hopington Aquifer. These differencesmay be due to more intensive sampling in problem areas in the second decade of samplingwhen N03-N as a ground-water contaminant was known. It is clear that N03-Ncontamination of ground water has been present in the Hopington Aquifer since the early1970s and continues to be a problem.Although there has been substantial increase in residential development, the changesin water quality within the river system have been small. The change in intensity ofagricultural activities has not yet been documented. It is possible that agricultural sources ofN03-N have been replaced by residential sources, resulting in a relatively small net changein the total amount of N03-N contamination from the different sources. If the N03-Nsources have increased, but bio-remediation of the N03-N occurs before it reaches thedischarge areas of streams, then a measurable change in surface water would not beobserved. Alternatively, the lack of a visible change in N03-N in surface water may be dueto the slow movement of N03-N contaminated ground water, and the effect of changes inland use and N03-N loadings are not yet measurable in the surface water.1936.4 Land-use activities in the Salmon River watershedSeptic systems were used as an index of residential development within thewatershed. Approximately 1700 new septic systems were installed in the watershed betweenthe early 1970s and the present. These new systems represent about half the installations inthe watershed. There is a high density of residential development in Fort Langley and in thesouthwest portion of the watershed over the Hopington Aquifer in the Salmon RiverUplands.The agricultural operations beny farming, and the animal operations of poultry,horses, and other animals (beef, dairy, swine, sheep, and mixed livestock) were chosen asindicators of intensive agricultural operations. About fifty percent of the watershed containsagricultural operations, whereas the subset of selected intensive agricultural operationsincluded in this study covers only 11 percent of the watershed.6.5 Spatial relationships between land use and water qualityHigh spatial variability is observed in N03-N in surface water within the watershed,and the N03-N is due to anthropogenic and not natural sources. This makes N03-N asuitable variable for the examination of relationships between water quality and land use.Residential and agricultural land-use activities were characterized by 1) a density index, anestimate of the land-use activity immediately upstream of the water sampling station, and ii)a cumulative density index, the density of the land-use activity in the entire watershedupstream of the water sampling station.In the middle reaches of the Salmon River the high concentration of N03-N in thesurface water is associated with high density indices of both residential and agricultural194activities. A spatial lag is observed between the higher density of land use activities and theincrease in N03-N in surface water. This lag can be explained by the northward flow ofground water from the Hopington Aquifer into the Salmon River. The pattern of thecumulative density index matches the pattern of N03-N in surface water more closely thanthe density index, and illustrates that the water quality at a point in the river is not onlyinfluenced by the land area immediately adjacent, but can also reflect the land-use activitiesof the entire watershed upstream. Higher concentrations of N03-N are also observed inwell-water in the contributing areas surrounding the middle reaches of the Salmon River andCoghlan Creek. The highest density index of both agricultural and residential activitiesoccurs on the glacial outwash parent materials that make up the Hopington Aquifer, locatedin the middle section of the watershed. The highest land-use intensity occurs on the mostsensitive soils, where ground-water contamination is most likely to occur.Examining relationships between land use and water quality in the Salmon Riverwatershed is complicated by the multiple sources of N03-N in close proximity, and changesin the relative importance of the different sources in different locations within the watershed.Positive correlations were observed between ground-water N03-N and both the “all animalactivities” and septic systems. The lack of many significant correlations between land useand N03-N could result from the difficulty of dividing the watershed into meaningful,homogeneous land units; the influence of the size and number of land units; the limitationsof the spatial extent of well-water sampling; and errors in the mapping of land use andcontributing-area boundaries.There is a substantial temporal and spatial lag between N03-N leaching from the landsurface and its detection in elevated concentrations in the ground and surface water, and this195can complicate data interpretation. The N03-N measured during baseflow conditions in theSalmon River may be the legacy of land-use activities from 10 to 30 years ago or longer.Given the rapid change in residential development observed recently, it is difficult to makecomparisons between present land use and contaminants in the water, when the age of theground water is not known. The direction of ground-water flow in the Hopington Aquifer ispoorly understood, as is the relative contributions of younger subsurface ground-water flowand older, deeper baseflow to the low flow conditions of the Salmon River. The spatial andtemporal lags between ground water N03-N and surface water N03-N, and land areas withboth agricultural and residential activities make it difficult to observe a high degree ofstatistical correlation between land use and water quality.High density indices of both agricultural and residential activities are present in theareas of high N03-N in surface and ground water, which suggests that both land-use typesare sources of N03-N. The associations between N03-N and land use observed in thisstudy, suggests that the source of N03-N to the ground water varies in different areas of thewatershed, although cause-effect relationships and relative importance of these two sourcescan not be determined with this approach. Estimation of the relative importance ofresidential and agricultural activities would require a calculation of the nitrogen loadings tothe soil from these individual land uses on the different surficial material types. Thismodelling approach would identify the contributing areas with the highest nitrogen sourcesand permit targeted management and remediation efforts.1967. RECOMMENDATIONS7.1 Recommendations for further researchRelationship between land use activities and nitrateA detectable change in surface water N03-N was observed at stations 5, 13, and 7.An examination of change in land use activities in the land area upstream of these stationsmay provide insight into the cause.Only selected agricultural activities were chosen to examine relationships betweenN03-N and agricultural land use, due to time constraints. Land used for grazing and fodderproduction is a substantial portion of the agricultural land within the watershed and shouldbe included in the examination of relationships between land use and water quality in theSalmon River watershed.Relative contribution of residential and agricultural land uses to nitrate contaminationTo determine the relative contribution of residential and agricultural activities to thenitrate load in the Hopington Aquifer a nitrogen balance should be calculated. The numberand size of septic systems, the average nitrate load from a septic system, the number ofanimals and waste production from different livestock, and the rates of fertilizer applicationare required.Septic system design for the removal of nitrogenAlternate designs for on-site sewage disposal systems are required that minimize theamount of nitrate that is leached into the ground water (USEPA 1980b, Whitmyer 1991,Winkler and Veneman 1991). A number of designs have been tested that use biological197denitrification, often with several tanks and recirculation to stimulate denitrification(Whitmyer et al. 1991). These systems often require more maintenance that the standardseptic system design.Temporal variability of nitrate in well waterThe determination of long term trends in ground-water nitrate is complicated by theseasonal variation in nitrate load and aquifer recharge. The selection of monitoring wells inareas of a single land use type, and where the land use has not changed substantially in thepast several decades, would provide information on ground water contamination fromspecific land uses. Biweekly monitoring would be required to be able to distinguishbetween seasonal and longer term trends in nitrate.Statistically designed spatial survey of nitrate contamination in ground waterA statistically designed survey, with the selection of wells based on soil type, landuse, and depth of well, would provide information on the spatial extent of ground watercontamination and illustrate relationships between land use and water quality. The lack of aconsistent sampling design for existing historic data on N03-N and the large time span overwhich the samples were collected limit the use of this database.Age of ground waterOnly two water samples were collected from the Hopington Aquifer for tritiumdating. Dating of water samples from wells of different depths in different locations on theaquifer could provide information about the age of the nitrate contamination.198Water balance for the Salmon River watershedAn important parameter in water quality is the quantity of water in the stream,particularly when there is discharge of contaminants into the water column. An adequatedilution of contaminants is required to prevent deleterious effects on the aquatic biota andlimitation of the potential water uses. Minimum flow requirements are needed to maintainthe quantity of fish habitat. A water balance must include both the ground and surfacewater supply and use. Current uses include extractions from surface and ground water forirrigation, livestock watering, and domestic uses. The drop in the water table measured atthree of the observations wells in or near the Hopington Aquifer is of concern, and it shouldbe determined if this is widespread throughout the aquifer or only a localized effect fromnearby high-capacity wells.Change in water quality during storm eventsSome contaminants are transported to the watercourse only during storm events.Better documentation of change in water quality following storm events, in particular forsuspended sediment load, bacteria, and dissolved phosphorus, is needed.Historic Changes in water qualityAlthough analysis methods change, in order to compare current results with historicdata, there is a need for well documented methods for sample locations; sample collection,preparation, and analysis methods. A standardized system for naming sampling locationsusing unique identifiers, e.g. for stream sites and wells, is required.1997.2 Management recommendationsThe highest density indices for agricultural and residential activities were observed inthe Salmon River Uplands area, in the Hopington Area. Unfortunately, this is the area withcoarse-textured permeable soils and a high sensitivity to ground water contamination. Thedecision to convert this area from agricultural activities to residential development datesfrom the 1960s when consideration was given to the agricultural potential of the area, butnot the environmental concerns of ground water contamination. In order to minimize theimpact of the land-use activities on the water quality a number of management practices aresuggested.Follow the BCMOAFF best management practices for agricultural operationsGuidelines exist for the proper storage, handling and application rates for manurefrom intensive animal operations as well as rates of fertilizer application for crop production(BCMOAFF 1992a, 1992b, Paul et a!. 1992). Encouragement of farmers to follow thecurrent best management practices will reduce the over fertilization of crops and high ratesof manure application that can result in nitrate leaching below the rooting zone andcontaminating the ground water. The participation of hobby farmers and horse operations isimportant. Although these operations are small, with often only a few animals per property,the cumulative effect of the nitrate loadings from these animals may be substantial.Registry of animal operations and animal numbers in the ESASManagement of the aquifer resource requires an understanding of the sources andloading of nitrogen. A registry of animal numbers and operations would enable the200Township, MOAFF, and MOELP staff to calculate nitrogen loading and detect localizedchanges in land-use activities that could have adverse environmental effects.Designate the aquifers within Township as ESAs with development permit restrictionsFurther intensification of land use on the sensitive glacial outwash materials of theHopington Aquifer is inappropriate, given the current level of ground-water contaminationalready recorded. Since there is often a substantial time lag, in the order of years ordecades, between the leaching of nitrate below the rooting zone and its detection in wellwater, it is expected that the level and extent of ground-water contamination in theHopington will increase in the future, even with no further land development.Regular maintenance of existing septic systems, including regular pumping toremove organic solids, will remove some of the nitrogen load that is tied up in organicmatter. Alternate methods of human-waste disposal is required. On-site sewage disposalsystems are available that include N removal and could be appropriate for the coarse-textured, glacial outwash materials (Whitmyer et al. 1991). Even if alternate waste disposalis provided, the effect of continued residential development on the recharge of the aquifermust be considered. High-density housing places additional demands on the ground waterfor garden and lawn irrigation and also reduces the permeable surface available for groundwater recharge.Education program for residents on extent of nitrate contamination of ground water and thesource of nitrogen from residential activitiesAn education program to inform residents living over the sensitive aquifers of the201household sources of nitrate to the ground water is needed. Poorly functioning or poorlymaintained septic systems and fertilization of lawns and gardens are the major householdsources. Over fertilization and over-watering of lawns could be an importance source ofleached nitrate.Comprehensive water management planWater extraction from the Salmon River and tributaries is controlled by theBCMOELP through allocation of water licences; however, extraction of ground water iscurrently not controlled. A water management plan is required that considered both thesurface-water and ground-water extractions during the critical low flow periods in themonths of June through October. Consideration should be given of the effect of furtherground water extraction for irrigated agriculture or residential use on the surface-water flow.A single well water sample is not sufficient to assess the nitrate contamination ahouseholds well-water supply. Nitrate concentrations measured in the long-term monitoringwells indicate the nitrate the concentrations exhibit high fluctuations through time.2028. LITERATuRE CITEDAckermann, F. 1980. A procedure for correction the grain size effect in heavy metalanalyses of estuarine and coastal sediments. Environmental Technology Letters 1:518-527.Ackermann, F., Bergmann, H. and U. Schleichert. 1983. Monitoring of heavy metals incostal and estuarine sediments- a question of grain-size: <20 um versus < 60 um.Environmental Technology Letters 4: 317-328.Alloway, BJ. and D.C. Ayres. 1993. Chemical principles of environmental pollution.Blackie Academic & Professional, Glasgow, UK. 291 p.Anderson, J.C. 1990. Non-point source surface and groundwater pollution in Canada:Major concerns, and their significance for sustainable development. EcologicalApplications Division, State of the Environment Reporting Branch, EnvironmentCanada. Ottawa, Ont. 66 p.Anonymous. 1991. Open files of exclusions from the BC Agriculture Land Reserve. BCAgriculture Land Commission, Burnaby, BC.APHA (American Public Health Association). 1989. Standard methods for the examinationof water and wastewater, 17th edition. American Public Health Association,Washington, D.C.Armstrong, J.E. 1980. Surficial geology, Mission, British Columbia, Geological Survey ofCanada, Map 1485A, scale 1:50,000.Armstrong, J.E. 1984. Environmental and engineering applications of the surficial geologyof the Fraser Lowland, British Columbia. Geological Survey of Canada Paper 83-23.Canadian Government Publishing Centre, Hull, Quebec. 54 p.Annstrong, J.E. and S.R. Hicock. 1980. Surficial geology, New Westminster, BritishColumbia, Geological Survey of Canada, Map 1484A, scale 1:50,000.Baker, J.L. and J.M. Laflen. 1983. Water quality consequences of conservation tillage. J.Soil Water Conserv. 38: 186-193.Ball, P.R., D.R. Keeney, P.W. Theobald, and P. Ness. 1979. Nitrogen balance in urineaffected areas of a New Zealand pasture. Agron. J. 71: 309-314.Ball, P.R. and J.C. Ryden. 1984. Nitrogen relationships in intensively managed temperategrasslands. Plant Soil 76: 23-33.Barraclough, D., M.J. Hyden, and G.P. Davies. 1983. Fate of fertilizer nitrogen applied tograssland. I. Field leaching results. J. Soil Sci. 34: 483-497.203BCMOAFF. 1992a. Environmental guidelines for beefproducers in British Columbia.BC Ministry of Agriculture, Fisheries and Food, Soils and Engineering Branch.Abbotsford, BC. 78 p.BCMOAFF. 1992b. Environmental guidelines for poultry producers in British Columbia.BC Ministry of Agriculture, Fisheries and Food, Soils and Engineering Branch.Abbotsford, BC. 70 p.BCMOEMPR. 1990. Regional geochemical survey data, NTS 92G Vancouver. BCRegional Geological Survey RGS-26, Geological Survey of Canada Open File 2184.BC Ministry of Energy, Mines and Petroleum Resources, Mineral ResourcesDivision, Geological Survey Branch, Applied Geochemistry. Victoria, BC. onediskette.BCMOE. 1991. The attainment of ambient water quality objectives in 1990. WaterQuality Branch, Water Management Branch, BC Ministry of Environment. Victoria,BC. 204 p.Beale, R.L. 1976. Analysis of the Effects ofLand Use and Soils on the Water Quality ofthe Salmon River Watershed, Langley. M.Sc. thesis, Department of Soil Science,University of British Columbia, Vancouver, BC. 243 p.Benedict, A.H., K.J. Hall, and F.A. Koch. 1973. A Preliminary Water Quality Survey ofthe Lower Fraser River System. Technical Report No. 2, Westwater ResearchCentre, University of British Columbia, Vancouver. 50 p.Bindra, K.S. and K.J. Hall. 1977. Geochemical partitioning of trace metals in sedimentsand factors affecting bioaccumulation in benthic organisms. Internal report,Westwater Research Centre, UBC., Vancouver, BC. 96 p.Burrough, P.A. 1986. Principles of Geographical Information systems for Land ResourcesAssessment. Oxford University Press, New York, New York. 194 p.Campbell, P.J.C., A. Lewis, P.M. Chapman, A.A. Crowder, W.K. Fletcher, B. Imber, S.N.Luoma, P.M. Stokes, and M. Winfry. 1988. Biological Available Metals inSediments. NRCC No. 27694. Ottawa, Canada. 298 p.Chapman, D. and V. Kimstach. 1992. The selection of water quality variables. In (Ed.) D.Chapman, Water quality assessments, a guide to the use of biota, sediments andwater in environmental monitoring. Chapman & Hall Ltd. London, UK. p. 51-119.Cherry, J.A. 1987. Groundwater occurrence and contamination in Canada. In (Eds) M.C.Healey and R.R. Wallace, Canadian aquatic resources. Can. Bull. Fish. Aquat. Sci.215: 397-426.204Chilibeck, B., G. Chislett, and G. Norris. 1992. Land Development Guidelines for theProtection ofAquatic Habitat. DFO and BCMOELP, Victoria, BC. 128 p.Clark, M.J.R. (Ed.) 1992. Status Report of the QA work group. A report of the qualityassurance work group, Canada-British Columbia water quality monitoring agreement.Clark, MJ.R., T.O. Morrison, A. Nugent, G.R. Gough, D.W. Holmes, and D.H.G. Ableson.1981. A Preliminary Study of Water Quality in the Fraser River and its Tributaries.BC. Waste Management Branch. Waste Management Branch report no. 80-12.301 p.Cogger, C.A. 1988. On-site septic systems: the risk of groundwater contamination. J.Environ. Health 51: 12-16.Colman, J.A. and R.F. Sanzolone. 1992. Geochemical characterization of streambedsediment in the upper Illinois River basin. Water Res. Bull. 28: 933-950.Combest, K.B. 1991. Trace metals in sediment: spatial trends and sorption processes.Water Res. BulL 27: 19-28.Cook, K., A. Faulkner, P. Mooney, K. Hall, M.Healey, D. Watts, S. Brown, and H. Schreier.1993. An Evaluation of Environmentally Sensitive Areas in the Township ofLangley.Volume I. ESA Analysis. Prepared for the Corporation of the Township of Langley.Westwater Research Centre, University of British Columbia, Vancouver, BC. 86 p.and appendices.Cooper, C.M. 1993. Biological effects of agriculturally derived surface water pollutants onaquatic systems - a review. J. Environ. Qual. 22: 402-408.Corporation of the Township of Langley. 1979. Official Community Plan. Bylaw No.1842. October 1, 1979. Corporation of the Township of Langley, Langley, BC.18 p. plus maps.Corporation of the Township of Langley. 1987. Township ofLangley Zoning Bylaw 1987No. 2500. Corporation of the Township of Langley. Langley, BC.Corporation of the Township of Langley. 1993. Langley Official Community Plan Bylaw1979 No. 1842 Amendment (Rural Plan) Bylaw 1993 No. 3250. The Corporation ofThe Township of Langley, Community Development Department. Langley, BC.35 p.Corporation of the Township of Langley. 1994. Population and Growth. The Corporationof The Township of Langley, Community Development Department. Langley, BC.8 p.205Crawford, P. 1993. Preserving rural character in an urban region, rural planning in theTownship of Langley. Plan Canada (March 1993): 16-23.Dakin A. 1993. Groundwater resources of the basins, lowlands, and plains, in GroundwaterResources of British Columbia. BC Environment, Water Management Division,Hydrology Branch, Groundwater Section. Victoria, BC. p. 9.1-9.78.Dayton & Knight Ltd. and Piteau Associates. 1994. The Ministry of Health SewageDisposal System Report. includes Appendix 9, Summary of communitysewered/onsite systems by township section (41 p.). Prepared for the BC Ministry ofHealth.De Groot, AJ., K.H. Zschuppe, and W. Salomons. 1982. Standardization of methods ofanalysis for heavy metals in sediments. Hydrobiologia 92: 689-695.Digital Resource Systems Limited. 1991. TerraSoft Reference Guide, The NaturalEvolution of GIS. Digital Resource Systems Limited of Nanaimo, British Columbia.Nanaimo, BC.Dorcey, A.H.J. (Ed.) 1976. The uncertain future of the Lower Fraser. Westwater ResearchCentre, University of British Columbia. Vancouver, Canada. 202 p.Elder, J.F. 1988. Metal Biogeochemistry in Surface-Water Systems, A Review of Principlesand Concepts. U.S. Geological Survey Circular 1013. 43 p.Environment Canada and BC Ministry of Environment. 1992. State of the Environment forthe Lower Fraser River Basin. SOE report no. 92-1. BC Ministry of theEnvironment and Environment Canada. Ministry of Supply and Services. Ottawa,Canada. 79 p.Environment Canada, Indicators Task Force. 1991. A Report on Canada’s ProgressTowards a National Set of Environmental Indicators. SOE Report No. 91-1. Stateof the Environment reporting, Environment Canada, Ottawa, Ontario. 98 p.Farwell, M.K., N.D. Schubert, K.H. Wilson, and C.R. Hanson. 1987. Salmon escapementsto streams entering statistical areas 28 and 29, 1951 to 1985. Can. Data Rep. Fish.and Aquat. Sci. 601. 166 p.Fergusson, J.E. 1990. The heavy metals: chemistry, environmental impact and healtheffects. Pergamon Press. Oxford, UK. 614 p.Feth, J.H. 1981. Chloride in natural continental water - a review. U.S. Geological SurveyWater-Supply Paper 2176. Washington, DC. 30 p.206Follett, R.F. and DJ. Walker. 1989. Ground water quality concerns about nitrogen. In(Ed) R.F. Follett, Nitrogen Management and Ground Water Protection. ElsevierScience Publishers B.V. Amsterdam, The Netherlands. p. 1-22.Freeze, R.A., J. Atwater, and H. Liebscher. 1993. Groundwater contamination: occurrence,sources and transport of contaminants in groundwater. In Groundwater Resources ofBritish Columbia. BC Environment, Water Management Division, HydrologyBranch, Groundwater Section. Victoria BC. p. 5.1-5.26.Freeze, R.A. and J.A. Cherry. 1979. Groundwater. Prentice-Hall, Inc. Englewood Cliffs,New Jersey. 604 p.Gartner-Lee Limited. 1992. Final Report, Fraser Valley Ground Water/Drinking WaterStudy. Prepared for the Ministry of Health, Public Protection Branch, Victoria, BC.37 p. + maps and appendices.Gartner-Lee Limited. 1993. Fraser Valley Ground Water Monitoring Program, Phase 1Report. Prepared for Ministry of Health, Environmental Health Protection, Victoria,BC. 51 p. and appendices.Gilliland, M.W. and W. Baxter-Potter. 1987. A geographic information system to predictnon-point source pollution potential. Water Resour. Bull. 23: 281-291.Goldberg, E.D., J.J. Griffin, V. Hodge, M. Koide, M., and H. Windom. 1979. Pollutionhistory of the Savannah River estuary. Envir. Sci. & Tech. 13: 588-594.GVRD Development Services. 1992. Creating our ftsture, the livable region strategic plan.Population, household, labour force, and employment projections for theMetropolitan Vancouver Region, 1991 to 2021. Development Services Department,Greater Vancouver Regional District. Burnaby, BC.Hagen, M.E. 1990a. Potential Surface Water Contamination from Miscellaneous Sources inthe Fraser River Estuary. Waste Management Activity Program Background Paper.Fraser River Estuary Management Program. FREMP, New Westminster, BC. 23 p.Hagen, M.E. 1990b. Agricultural Runoff Contamination in the Fraser River Estuary.Waste Management Activity Program Background Paper. Fraser River EstuaryManagement Program. FREMP, New Westminster, BC. 40 p.Hall, K.J. 1992. Limnology of Okanagan Lake, BC related to the proposed Kelowna STPoutfall. Prepared for Dayton and Knight Ltd., West Vancouver, BC. 53 p.Hall, K.J., F.A. Koch, and I. Yesaki. 1974. Further Investigations into Water QualityConditions in the Lower Fraser River System. Technical Report No. 4, WestwaterResearch Centre, University of British Columbia, Vancouver. 104 p.207Hall, K.J., H. Schreier, and S.J. Brown. 1991. Water quality in the Fraser River Basin. InA.H.J. Dorcey and J.R. Griggs (Eds) Water in Sustainable Development: ExploringOur Common Future in the Fraser River Basin. Westwater Research Centre,University of British Columbia. Vancouver, BC. pp. 41-75.Hall, K.J. and J.H. Weins. 1976. The quality of water in tributaries of the Lower Fraserand Sources of Pollution. In (Ed.) A.H.J. Dorcey, The Uncertain Future of theLower Fraser. Westwater Research Centre, UBC., Vancouver, BC. p. 49-83.Hall, K.J., I. Yesaki, and J. Chan. 1976. Trace Metals and Chlorinated Hydrocarbons inthe Sediments of a Metropolitan Watershed. Technical Report No. 10, WestwaterResearch Centre, University of British Columbia, Vancouver. 74 p.Hallberg, G.R. 1989. Nitrate in ground water in the United States. In (Ed.) R.F. Follett,Nitrogen Management and Ground Water Protection. Elsevier Science PublishersB.V. Amsterdam, The Netherlands. pp. 35-74.Halliday, S.L. and M.L. Wolfe. 1991. Assessing ground water pollution potential fromnitrogen fertilizer using a geographic information system. Water Resour. Bull. 27:237-245.Halstead, E.C. 1986. Groundwater Supply; Fraser Lowland, British Columbia. NationalHydrology Research Institute. Paper # 26, IWD Scientific Series # 146. NHRI,Saskatoon, 80 p.Health and Welfare Canada. 1989. Guidelines for Canadian Drinking Water Quality.Prepared by the Federal-Provincial Subcommittee on Drinking Water of the Federal-Provincial Advisory Committee on Environmental and Occupational Health.Ministry of Supply and Services. Ottawa, Canada.Hegemann, D.A. and J.D. Keenal. 1985. Measurement of watershed phosphorus: a review.Toxicol. Environ. Chem. 9: 265-289.Helsel, D.R. and G.R. Koltun. 1986. Evaluation of size-distribution effects and laboratoryprecision in the analysis of bottom materials. US Geological Survey Water-SupplyPaper 2310, Washington, DC. p. 1-11.Hem, J.D. 1985. Study and interpretation of the chemical characteristics of natural water.(3rd Edition). US Geological Survey Water Supply Paper 2254. Washington, DC.263 p.Henry, P. and W.A. Meneley. 1993. Fertilizers and groundwater nitrate, Phase I: reviewof literature, prepared for the Western Canada Fertilizer Association. 24 p.208Horowitz, A.J. 1988. A review of physical and chemical partitioning of inorganicconstituents in sediments. In: The Role of Sediments in the Chemistry ofAquaticSystems, Proceedings of the Sediment Chemistry Workshop, February 8-12, 1982.A.J. Horowitz and W.L. Bradford (Editors). US Geological Survey Circular 969, pp.7-22.Howard Paish & Associates. 1980. Cooperative Management of Watershed and SalmonidProduction. Prepared for Department of Fisheries and Oceans, SalmonidEnhancement Program. Vancouver, BC. 38 p.Inglis, S.D., M.L. Rosenau, and A.K. Loerena. 1992. Distribution and habitat of theendangered Salish Sucker (Catostomus sp.) BC Environment, Fish and WildlifeManagement. Regional Fisheries Report No. 230. 14 p.Kabata-Pendias, A. and H. Pendias. 1992. Trace Elements in Soils and Plants, 2nd Edition.CRC Press. Boca Raton, Florida.. 365 p.Kalkhoff, S.J. 1993. Using a geographic information system to determine the relationbetween stream quality and geology in the Roberts Creek watershed, Clayton County,Iowa. Water Resour. Bull. 29: 989-996.Keeney, D.R. 1986. Sources of nitrate to ground water. CRC Critical Reviews inEnvironmental Control 16: 257-304.Keeney, D.R. 1989. Sources of nitrate to ground water. In (Ed) R.F. Follett, NitrogenManagement and Ground Water Protection. Elsevier Science Publishers B.V.Amsterdam, The Netherlands. pp. 23-34.Keeney, D.R. and T.H. DeLuca. 1993. Des Moines River nitrate in relation to watershedagricultural practices: 1945 vs 1980. J. Envir. Qual. 22: 267-272.Kenk, E., M.W. Sondheim, and H. Quesnel. 1987. CAMAP Volume 1. BCMOELPThematic Mapping Geographic Information System: Data Entiy and ValidationProcedures for Soil, Agricultural Capability, Surficial Geology and all purposeentity. MOELP Manual 10. Victoria, BC. 105 p. and appendices.Kerr, A.M. 1984. A Survey of Wellwater Quality in the Hopington Area ofLangley withRegard to the Health Hazard of Nitrate Induced Methemoglobinemia. Reportprepared for Central Fraser Valley Health Unit, BC. 17 p. and appendices.Kohut, A.P., S. Sather, J. Kwong, and F. Chwojka. 1989. Nitrate contamination of theAbbotsford Aquifer, British Columbia. Proceedings of the International Associationof the Hydrogeology Symposium, Saskatoon, Sask. 22 p.Krumgalz, B. 1989. Unusual grain size effect on trace metals and organic matter incontaminated sediments. Mar. Pollut. Bull. 20: 608-611.209Kwong, J.C. 1986. Groundwater Quality Monitoring and Assessment Program- TheOccurrence ofNitrate-Nitrogen in Groundwater in the Langley-Abbotsford Area.Memorandum Report, BC. Ministry of Environment, Water Management Branch.internal report to Mr. A.P. Kohut, Groundwater Section. File 0329563-A. Volume4, 7 p. + attachments.Lachat Instruments. 1990. Methods Manual for the QuickChem Automated Ion Analyzer.Milwaukee, Wisconsin.Liebscher, H., B. Hii, and E. McNaughton. 1992. Nitrates and Pesticides in the AbbotsfordAquifer, Southwestern British Columbia. Inland Waters Directorate andEnvironmental Protection, Environment Canada, Vancouver, BC. 83 p.Likens, G.E., F.H. Bormann, R.S. Pierce, and W.A. Reiners. 1978. Recovery of adeforested ecosystem. Science 199: 492-496.Luoma, S.N. 1990. Processes affecting metal concentrations in estuarine and coastal marinesediments. In (Eds) R.W. Furness and P.S. Rainbow, Heavy Metals in the MarineEnvironment. CRC Press, Boca Raton, Florida. 256 p.Luttmerding, H.A. 1980. Soils of the Langley-Vancouver Map Area. RAB Bulletin No. 18,BC. Soil Survey Report No. 15, BCME, Kelowna.Martin, J.M. and M. Meybeck. 1979. Elemental mass-balance of material carried by majorworld rivers. Mar. Chem. 7: 173-206.Martin, J.M., P. Nirel, and A.J. Thomas. 1987. Sequential extraction techniques: promisedand problems. Mar. Chem. 22: 313-341.McKeague, J.A. and M.S. Wolynetz. 1980. Background levels of minor elements in someCanadian soils. Geoderma 24: 299-307.Meybeck, M. and R. Helmer. 1992. An introduction to water quality. In (Ed) D.Chapman, Water quality assessments, a guide to the use of biota, sediments andwater in environmental monitoring. Chapman & Hall Ltd., London, UK. p. 1-17.Millward, C.G. and P.D. Kluckner. 1989. Microwave digestion technique for the extractionof minerals from environmental marine sediments for analysis by inductively coupledplasma atomic emission spectrometry and atomic absorption spectrometry. J.Analytical Atomic Spectrometry 4: 709-713.Moody, D.W. 1990. Groundwater contamination in the United States. J. Soil WaterConserv. 45: 170-179.210Moore, J.N., E.J. Brook, and C. Johns. 1989. Grain size partitioning of metals incontaminated, coarse-grained river floodplain sediment: Clark Fork River, Montana,U.S.A. Environ. Geol. Water Sci. 14: 107-115.National Research Council of Canada. 1981. Marine Sediment Reference Materials.Division of Chemistry, Marine Analytical Chemistry Standards Program, Ottawa,Canada. 5 p.Nelson, D.W. and L.E. Sommers. 1982. Total organic carbon and organic matter. In A.L.Page (Ed.) Methods of Soil Analysis, Part 2, Chemical and MicrobiologicalProperties. Agronomy Series 9: 539-589.Nordin, R.N., J.E. Bryan, and E.V. Jensen. 1990. Nutrient controls and water quality in theOkanagan Lakes 1969-1989. In (Eds) R.Y. McNeil and J.E. Windsor, Innovations inRiver Basin Management, Proceedings of the 43 Annual Conference of the CanadianWater Resources Association.North, M.E.A. and J.M. Teversham. 1984. The vegetation of the Floodplain of the Fraser,Serpentine and Nicomekl Rivers, 1859-1890. Map.OECD. 1986. Water Pollution by Fertilizers and Pesticides. Organization for EconomicCooperation and Development. Paris, France. 144 p.Omernik, M.M. 1977. Nonpoint source-steam nutrient level relationships: a nationwidestudy. EPA-600/3-77-105, US Environmental Protection Agency, US GovernmentPrinting Office, Washington DC. 151 p. plus maps.Osborne, L.L. and M.J. Wiley. 1988. Empirical relationships between land use/cover andstream water quality in an agricultural watershed. J. Envir. Manag. 26: 9-27.Owens, L.B., W.M. Edwards, and R.W. Van Keuren. 1994. Groundwater nitrate levelsunder fertilized grass and grass-legume pastures. J. Environ. Qual. 23: 752-758.Pardue, J.H., R.D. DeLaune, and W.H. Patrick. 1992. Metal to aluminum correlation inLouisiana coastal wetlands: identification of elevated metal concentrations. J. Envir.Qual. 21: 539-545.Paul, J.W., G. Hughes-Games, and B.J. Zebarth. 1992. Manure Management Workshop,Thursday, January 30, 1992. Presented by Agriculture Canada, BCMOAFF andCanada-BC Soil Conservation Program. 42 p.Petrovic, A.M. 1990. The fate of nitrogenous fertilizers applied to turfgrass. J. Envir.Qual. 19: 1-14.Piteau Associates. 1991. Hydrogeological assessment ofAldergrove Aquifer, Aldergrove,BC. Report to the Corporation of the Township of Langley. 38 p. and maps.211Salomons, W. and U. Förstner. 1984. Metals in the Hydrocycle. Springer-Verlag. Berlin,Heidelberg. 349 p.Sather, S. 1988. Groundwater Quality Monitoring Langley/Abbotsford Area 1 987/88.Memorandum Report, BC Ministry of Environment, Water Management Branch.internal report to A.P. Kohut, Groundwater Section. File 0329563-A Volume 6. 6 p.+ attachments.Sawicki, J. and G. Runka. 1986. Land use classification in British Columbia. Prepared forthe Ministry of Agriculture and Food, Soils Branch and Ministry of Environment,Surveys and Resources Mapping Branch. Victoria, BC. MOE Manual 8. 33 p.Sawicki, J. and G. Runka. 1990. Soil Degradation and Rural Land Use Change, UplandsofLangley and Matsqui Municipalities - Fraser Valley. ARDSA Project 23010.96 p.Sawyer, C.N. and P.L. McCarty. 1978. Chemistry for Environmental Engineering.McGraw-Hill Publishing Co. New York. 532 p.Schreier, H., S.J. Brown, and K.H. Hall. 1991. The land-water interface in the Fraser RiverBasin. In A.H.J. Dorcey and J.R. Griggs (Eds) Water in Sustainable Development:Exploring Our Common Future in the Fraser River Basin. Westwater ResearchCentre, University of British Columbia. Vancouver, BC. pp. 77-116.Schropp, S.J., F.G. Lewis, H.L. Windom, J.D. Ryan, F.D. Calder, and L.C. Burney. 1990Interpretation of metal concentrations in estuarine sediments of Florida usingaluminum as a reference element. Estuaries 13: 227-235.Shacklette, H.T. and J.G. Boerngen. 1984. Element concentrations in soils and othersurficial materials of the conterminous United States. US Geological SurveyProfessional Paper 1270. US Government Printing Office, Washington, D.C. 10 p.plus maps.Sharpley, A.N., S.C. Chapra, R. Wedepohl, J.T. Sims, T.C. Daniel, and K.R. Reddy. 1994.Managing agricultural phosphorus for protection of surface waters: issues andoptions. J. Environ. Qual. 23: 437-451.Sibley, D. 1987. Spatial Application of Exploratory Data Analysis. Concepts andTechniques in Modern Geography No. 49. Geo Books. Norwich, U.K. 37 p.Siegel, S. 1956. Nonparametric statistics for the behavioural sciences. McGraw-HillBook Company. New York, New York. 312 p.Sivertun, A., L.E. Reinelt, and R. Castensson. 1988. A GIS method to aid in nonpointsource critical area analysis. mt. J. Geog. Infor. Syst. 2: 365-378.212Smith, R.A., R.B. Alexander, and M.G. Wolman. 1987. Analysis and interpretation ofwater-quality trends in major U.S. rivers, 1974-81. U.S. Geological Survey water-supply paper No. 2307. U.S. Government Printing Office. Washington D.C. 25 p.Spalding, R.F. and M.E. Exner. 1993. Occurrence of nitrate in groundwater- a review. J.Environ. Qual. 22: 392-402.Statistics Canada. 1991. Census data for the Township of Langley, provided to theCommunity Development Department, Corporation of the Township of Langley,Langley, BC.Stednick, J.D. 1991. Wildiand Water Quality Sampling and Analysis. Academic Press, Inc.San Diego, California. 217 p.Steube, M.M. and D.M. Johnstone. 1990. Runoff volume estimation using GIS techniques.Water Resour. Bull. 264: 611-620.Stevenson, F.J. 1986. Cycles of soil carbon, nitrogen, phosphorus, sulfur, micronutrients.John Wiley & Sons. New York. 380 p.Surveys and Resource Mapping Branch. 1991. NAD27 to NAD83 conversion in BritishColumbia, national transformation, version 1.1. Release 1.1 July, 1991. Surveysand Resource Mapping Branch, Ministry of Lands and Parks, Province of BritishColumbia.Surveys and Resource Mapping Branch. 1992. British Columbia Specifications andGuidelines for Geomatics, content Series Volume 3, Digital Baseline Mapping at1:20000. Release 2.0 January 1992. Surveys and Resource Mapping Branch,Ministry of Environment, Lands and Parks. Province of British Columbia.Swain, L.G. and G.B. Holms. 1985. Fraser-Delta Area. Fraser River Sub-Basin fromHope to Kanaka Creek. Water Quality Assessment and Objectives. TechnicalAppendix. Water Management Branch, Ministry of the Environment, Province ofBritish Columbia. 193 p.Systat for Windows. 1992. Statistics. Version 5 Edition. Systat Inc. Evanston, Il. 750 p.Thomas, G.W., G.R. Haszler, and J.D. Crutchfield. 1992. Nitrate-nitrogen and phosphatephosphorus in seven Kentucky streams draining small agricultural watersheds:eighteen years later. J. Environ. Qual. 21: 147-150.Thomas R. and M. Meybeck. 1992. The use of particulate material. In (Ed.) D. Chapman,Water quality assessments, a guide to the use of biota, sediments and water inenvironmental monitoring. Chapman & Hall Ltd., London, UK. p. 121-170.Tim, U.S. and R. Jolly. 1994. Evaluating agricultural nonpoint-source pollution using213integrated geographic information systems and hydrologic/water quality model. J.Environ. Qua!. 23: 25-35.Trefry, J.J. and BJ. Presley. 1976. Heavy metals in sediments from San Antonio Bay andthe northwest Gulf of Mexico. Environmental Geology 1: 283-294.Ure, A.M. and M.L. Berrow. 1982. The elemental constituents of soils. EnvironmentalChemistry 2: 94-204.USEPA. 1980a. National summary of state water quality standards, Phosphorus. Preparedby Nalesnik Associates Inc.for the US Environmental Protection Agency,Washington, DC. 11 p.USEPA. 1980b. Onsite wastewater treatment and disposal systems design manual. EPA-625/1-80-012. Cincinnati, Ohio.Ventura, S.J. and K. Kim. 1993. Modeling urban nonpoint source pollution with ageographic information system. Water Resour. Bull. 29: 189-198.Vitousek, P.M. and J.M. Melillo. 1979. Nitrate losses from disturbed forests: patterns andmechanisms. Forest Sci. 4: 605-619.Walsh, S.J. 1985. Geographic information systems for natural resource management. J.Soil Water Conserv. 40: 202-205.Watts, D. 1992. A GIS Evaluation ofLand Use Dynamics and Fish Habitat in the SalmonRiver Watershed Langley, BC. M.Sc. Thesis, Resource Management Sciences,University of British Columbia, Vancouver BC. 156 p.Whitmyer, R.W., R.A. Apfel, R.J. Otis, and R.L. Meyer. 1991. Overview of individualonsite nitrogen removal systems. In On-site wastewater treatment, volume 6.Proceedings of the sixth national symposium on individual and small communitysewage systems. ASAE Publication 10-91. American Society of AgriculturalEngineers, St. Joseph, Michigan. p. 143-154.Wilber, W.G. and J.V. Hunter. 1979. The impact of urbanization on the distribution ofheavy metals in bottom sediments of the Saddle River. Water Res. Bull. 15: 790-800.Winkler, E.S. and P.L.M. Veneman. 1991. A denitrification system for septic tank effluentusing sphagnum peat moss. In On-site wastewater treatment, volume 6. Proceedingsof the sixth national symposium on individual and small community sewage systems.ASAE Publication 10-91. American Society of Agricultural Engineers, St. Joseph,Michigan. p. 155-164.214Zimmerman, R.D. 1990. Groundwater Quality Monitoring and Assessment Program - TheOccurrence of Nitrate-Nitrogen in Groundwater in the Langley-Abbotsford Area.Memorandum Report, BC. Ministry of Environment, Water Management Branch.Internal report to A.P. Kohut, Groundwater Section. File 0329563-A. 4 p. andattachments.PERSONAL COMMUMCATIONJoe Chaylt Engineering DepartmentTownship of LangleyLangley, BC.Barry Chilibeck Department of Fisheries and OceansVancouver, BCpresentation at the Salmon River mini-conference,June 1993, Langley, BC.Shawn Heier Zenon Environmental LaboratoriesBurnaby, BC.D.W. McCallum M Sc candidateCivil Engineering, UBCVancouver, BC.Graeme Spiers Department of Land Resource Science,University of GuelphGuelph, Ontario.Michael Wei Groundwater SectionBC Ministry of Environment Lands and ParksVictoria, BC.Barbara Wernick M Sc candidateResource Management and Environmental Studies, UBCVancouver, BC.Rod Zimmerman Groundwater SectionBC Ministry of Environment Lands and ParksVictoria, BC.ciiS3DIGM3JJVAppendix1Descriptionofthesediment andwatersamplinglocationsintheSalmonR.watershed.STATIONSTATIONLOCATIONSAMPLINGLOCATIONNUMBERSTATIONSSAMPLED1974-75AND1991-3KCO1SALMONRAT96AVEKCO2SALMONRATRAWLISONCRESKCO3TRIBUTARYOFSALMONRGLOVERRDNOFRAILWAYKCO4SALMONRAT64AVEKCO5COGHLANCRATWILLIAMSPARKKCO6SALMONRAT72AVEKCO7SALMONRAT56AVEKCO8TRIBUTARYOFSALMONRUNIONCRAT56AVEKCO9SALMONRAT248STKC1OTRIBUTARYOFSALMONRATROBERTSONCRESKC11SALMONRAT272STKC12COGHLANCRAT58STKC13TRIBUTARYOFSALMONRDAVIDSONCRWOF72AVEKC14TRIBUTARYOFSALMONRDAVIDSONCRATRAWLISONCRKC15SALMONRAT256STADDITIONALSTATIONS1991-1992KC16SALMONRATTRINITYWESTERNCOLLEGEBRIDGEKC17SALMONRAT264STVANCOUVERGAMEFARMKC19COGHLANCRAT248STKC2OCOGHLANCRAT64AyE,EOF248STADDITIONALSTATIONS1993(watersamplingonly)KC21COGHLANCRAT60AVEKC22COGHLANCRTRIBUTARYAT63AVEKC23SALMONRAT232AVEKC24TRIBUTARYOFSALMONRATSADDLEHORNCRESKC25COGHLANCRAT64AVE.EASTOF240STSOF96AVE,WBANKSOFRAWLISONCRES,WBANKATCULVERTWOFGLOVERRD,SBANKATBRIDGESOF64AVEEOFPARK,ATFENCELINESOF72AVEATBRIDGE,UPSTREAM, BEFOREUNIONCREEKUPSTREAMOFSALMONIDENHANCEMENTBUILDINGATBRIDGENOFROBERTSONCRESBELOWCULVERTWOF272ST, NSIDEOFBRIDGEWOFENDOF58AyE, ABOUT5MINALONGPATHSOF72AVE,OVERFENCE, DOWNHILL,PONDEOFRAILROADTRACKS, NSIDEOFROADESIDEOFBRIDGENESIDEOFBRIDGEESIDEOFCULVERT,SMALLPONDESIDEOFROADSSIDEOFROADNSIDEOFROAD,STATIONTOREPLACEOVERGROWN12SSIDEOFROADWSIDEOFBRIDGESSIDEOFROADATENDOFROADM H217Appendix 2 Emission lines used and detection limits for ICP-AES used for thedetermination of metals in sediment and soil samples in the Salmon Riverwatershed.ELEMENT LINE READ Operational Operation detection(nm) detection limit limit for this study:for complex digested sedimentssolutions (mg kg1)(ppb)Zn 213.856 10 0.6Cr 267.716 10 0.6Cu 324. 757 8 0.5Co 238.892 10-12 0.7Ni 221.647 12 0.7Pb 220.353 20 1.2Fe 259.940 1-5Mn 259.373 1-5 3Al 309.278 10 6Na 330.237 5 ppm 3Mg 279.353 5 3Ca 393.366 1 .06Ti 334.941Zr 339.198 4 .24Sr 339.198 1 .06V 309.311 25 6P 213.618 10 15218Appendix 3 Surface water quality stations in Beale (1976) and the EQUIS and SEAMdatabases, and the equivalent station numbers from this study.this study Beale (1976) EQUIS SEAMSalmon River upstream of Coghlan CreekKC11 RB11KC1 7KC15 RB15KC1O tributary RB1OKC09 RBO9KC24 tributaryKCO7 RBO7KCO8 tributary RBO8KCO4 RBO4Coghtan CreekKC 12KC21 RB12KC2OKC1 9KC25KC22 tributaryKCO5 RBO5Salmon River downstream of Coghlan CreekKC23 0300023KCO6 RBO6 0920007KCO3 tributary RBO3KC16 0301022KCO2 RBO2 0920018KC13 tributaryKC14 tributary RB14Glover Road 1100029 0300022KCO1 RBO1 0920017 0300021219Appendix 4 Agricultural land activity codes from the 1989 land-use map, and thecategories included in this study as intensive agricultural operations.Land Activity Code Description Intensive Generalizedagriculture CategoryA000 AGRICULTURAL ACTIVI11ESA100 Productive-Land Agricultural ActivitiesAllO Growing Annual Tillage CropsAlli Growing GrainAl 12 Growing VegetablesA113 Growing Root Crops and TubersA114 Growing Seed CropsA116 Growing Flowers and BulbsAl19 OtherAl20 Growing Forage Crops and GrazingAl21 Growing Forage CropsA122 GrazingAl 30 Fruit, Berry, and Nut Production Y other agricultureAl 31 Tree Fruit Production Y other agricultureAl 33 Berry Production V berryAl 34 Tree Fruit or Grape Nursery Stock V other agricultureProductionAl 35 Nut Production V other agricultureA200 Site Agricultural Activities V other agricultureA210 Housing Livestock Other animalsA21 1 Housing Dairy Animals V Other animalsA2l2 Housing Beef Animals V other animalsA213 Housing Horses Y horsesA214 Housing Swine Y other animalsA215 Housing Sheep Y other animalsA220 Outside Animal Feeding and Holding V other animalsAreasA221 Dairy Y other animalsA222 Beef Y other animalsA223 Horse Y horsesA225 Sheep Y other animalsA231 Housing Poultry Y poultry and furbearersA232 Housing Game Birds Y poultry and furbearersA233 Housing Fur Bearing Animals Y poultry and furbearersA240 Growing Plants in Greenhouses Y other agricultureA241 Growing Vegetables Y other agricultureA270 Growing Mushrooms Y other agricultureA280 Beekeeping and Honey MakingA290 OtherAppendix 5 List of TRIM features imported into TerraSoft for the Salmon River220Watershed and Hopington Aquifer.bridgea one lane road with surface of aggregate, soil, orclaya two lane road with surface of aggregate, soil, orclaya two lane road with a surface of concrete,asphalt, or tar-gravela two lane unidirectional road with a surface ofconcrete, asphalt, or tar-gravela four lane road with a surface of concrete,asphalt, or tar gravel and where the lanes of trafficmoving in opposite directions are not separated byan obstructionan unimproved route, logging or secondary roada double track with two closely parallel rail lineson the same roadbeda single track with one set of rails on the roadbeddefinite river or stream (that is not obscured onthe aerial photography)indefinite river or stream (this is obscured on theaerial photography)intermittent river or stream is a define watercoursethat is usually dry, depending on the season andprecipitationa watercourse of width greater than 20 m, the leftshoreline of a watercourse heading downstreama watercourse of width greater than 20 m, theright shoreline of a watercourse headingdownstreamlake, definitelake, intermittent (that is normally dry at sometimeduring the year)a water-saturated area, intermittently orpermanently water covered, with grass-likevegetationa water-saturated area, intermittently orpermanently covered with water, with shrub ortree vegetationland area which is at least 6 percent covered withtrees that are more than 2 m in heightDD93250000TRIM Feature Name Assigned TerraSoft Descrlption*DA250001 10DA250001 20DA250501 80DA251 00190DA251 00210DA251 50000DE22850000DE22950000GA24850000GA248501 40GA248501 50GA900001 10GA9 0000120GB1 5300000GB1 5300140GC17100000Feature NameBRIDGEGRAVEL_RD1GRAVEL_RD2PAVED_RD2PAVED_i WAYPAVED....RD4GRAVEL_RDRAIL_DOUBLRAIL_SINGLCREEK_DEFCREEKINDEFCREEK_INTRRIVER_LEFTRIVER.RIGTLAKE_DEFLAKE_INTERMARSHGC30050000 SWAMPJA33750000 WOODS* Descriptions modified from Surveys and Resource Mapping Branch (1992).221Appendix 6 Generalization of soil types from 1:25000 sofl map (Luttmerding 1980)for the production of the surficial geology map, Figure 3.8.Symbol Soil Name Soil MaterialMarineBR Berry Moderately fine to fine-textured marine depositsCD Cloverdale Moderately fine to fine-textured marine depositsLA Langley Fine to moderately fine-textured marine depositsLV Livingstone Less than 100 cm of moderately coarse-texturedlittoral deposits over fine-textured marine depositsMR Mimer Fine to moderately fine-textured marine depositsMY Murrayville Less than 100 cm of moderately coarse to medium-textured littoral deposits over fine textured marinedepositsGlacial MarineAB Albion Moderately fine to fine-textured glaciomarine depositsHN Heron Coarse-textured littoral deposits over moderatelycoarse textured glacial till or moderately fine texturedglaciomarine depositsN Nicholson Moderately fine textured glaciomarine depositsSC Scat Moderately fine textured glaciomarine depositsW Whatcom Moderately fine textured glaciomarine depositsGlacial OutwashAD Abbotsford 20-50 cm of medium-textured eolian deposits overgravelly glacial outwashCG Coghlan Gravelly glacial outwash depositsCL Columbia Gravelly glacial outwash depositsCN Calkins More than 20 cm of medium-textured eolian depositsover glacial outwash deposits or glacial tillGP gravel pitLH Lehman Less than 30 cm medium-textured eolian depositsover gravelly glacial outwashLY Lynden Coarse-textured glacial outwash depositsMH Marble Hill More than 50 cm of medium-textured eolian depositsover gravelly glacial outwash depositsSS Sunshine Sandy littoral and glacial outwash deposits222Symbol Soil Name Soil MaterialMarineAlluvIumAN Annis 15-40 cm or organic material over moderately finetextured floodplain depositsBD Banford 40 to 160 cm of well-decomposed organic materialover medium and moderately fine textured floodplaindepositsDW Dewdney 15 to 50 cm of medium-textured, laterally accretedfloodplain deposits over sandF Fairfield Medium to moderately fine textured, laterally accretedfloodplain depositsG Grevell Coarse-textured, laterally accreted floodpiain depositsGN Gibson 40 to 160 cm of partially decomposed organic materialover floodplain depositsHD Hazelwood Fine to moderately fine textured, vertically accretedfloodplain depositsHJ Hjorth Medium-textured, laterally accreted floodplain depositsHT Hallert Medium-textured, vertically accreted floodplaindeposits containing organic strataKZ Katzie Fine to moderately fine textured, mixed floodplaindepositsPE Page Medium to moderately fine textured floodplaindepositsPR Prest Medium to moderately fine textured floodplaindepositsSB Seabird Moderately coarse to coarse-textured floodplaindepositsWL Westland Moderately fine to fine-textured mixed marine andfloodplain depositsParent Material Assigned based on neighbouring soil polygonsCV Carvolth Moderately fine textured local stream depositsRS Ross Medium to moderately fine textured local streamdepositsJN Judson 40 to 160 cm of well-decomposed organic materialunderlain by moderately fine textured glaciomarinedepositsBO Bose 30 to 160 cm of gravelly lag or glacial outwashdeposits over moderately coarse textured glacial tilland some moderately fine textured glaciomarinedepositsAppendix7Measurementoftheprecisionofthelaboratorytechniquefortracemetalanalysis,bycomparisonoftheratiooftheduplicateanalysis.Allmeasurementsreportedasmg/kg.SAMPLEDIGESTZnCrCuCoNiPbFeMnAlSD-O1Aug11.03E+021.26E+023.54E+O12.61E+013.02E+012.17E+O14.95E+048.68E+027.39E+04I1.08E+021.37E+024.07E+012.63E+011.83E+O14.19E+O14.34E+046.97E+026.81E+04ratio1.\JAppendix7continuedSAMPLEDIGESTNaMgCaTiZrSrVPSD-01Aug11.51E+048.89E÷031.71E+046.38E+031.02E+022.26E+011.16E+023.04E+0311 .46E+047.35E+031.59E÷044.94E+039.89E+012.44E÷011 .35E+027.93E÷02ratio1.÷032.02E+044.75E+035.83E+011.82E+019.O1E+012.03E+0331.78E+044.51E+036.19E+033.55E+036.58E÷012.40E+011.41E+021.34E+03ratio1.551.193.261.341.131.321.561.51SD-02Dec31.43E+046.88E÷031.05E+046.07E+035.57E+012.35E+011 .36E+021.11E+0331.13E÷045.49E+031.31E+045.21E+031.71E÷029.50E+009.17E+015.57E+02ratio1.÷045.30E+039.37E÷035.42E÷035.29E+011.94E+016.12E÷018.61E+0148.40E+034.77E+031.42E.i-044.28E+032.07E+024.70E÷011.04E+029.OIE+02ratio1.731.111.521.273.912.431.7110.46OW-0246.77E+033.78E+031.06E+045.26E+034.85E+016.19E+011.80E+023.48E+0352.72E+043.43E+031.54E+044.56E+033.15E÷014.28E+011.90E÷024.44E+03ratio4.021.101.451.151.541.441.061.28GM-036ND5.01E+031 .73E+044.91E+035.83E+019.47E÷011 .78E+027.20E+0261.08E÷024.22E+032.56E+044.42E+035.81E+018.50E+011.52E+024.35E+02ratio1.191.481.÷037.87E+017.63E+011.83E÷026.49E+026ND8.08E+039.57E+035.50E+037.88E+018.49E+011.82E+025.96E+02ratio1.091.941.’J225Summary statistics for trace metal concentration in surficial material andstreambed sediments in the Salmon River watershed. All measurements areare mg/kg except carbon in percent.Zn Cr Cu Co Ni PbSURFICIAL MATERIALSglacial marine median 8.46E+01 1.39E+02 5.79E+01 4.25E+01 3.32E+01 1.20E+00mm 8.16E+01 1.17E+02 5.27E+01 4.03E+O1 2.98E+O1 1.20E+00max 8.89E+01 1.46E+02 6.23E+O1 4.92E+ol 3.85E+O1 1.20E+00max/mm 1.09 1.25 1.18 1.22 1.29 1.00mean 8.52Ei-O1 1 .35E+02 5.78E+01 4.33Ei-01 3.35E+01 1 .20E+00s.d. 2.94E+0O 1.19E+01 4.27E+0O 3.62E+00 3.65E+O0 0.00E+0Oc.v. 3.45 8.80 7.40 8.36 10.87 0.00marine median 1.OOE+02 1.23E+02 7.02E+O1 3.17E+O1 2.99E+01 8.90E+00mm 7.41 E+O1 9.38E+01 4.82E+O1 2.86E+01 2.65E+O1 1 .20E+00max 1.14E+02 1.31 E÷02 8.32E+O1 3.94E+01 3.56E+01 1.47E+01max/mm 1.54 1.39 1.72 1.38 1.34 12.23mean 9.91 E+01 1.1 8E+02 6.54E+01 3.31 E+01 3.01 E+01 8.07E+00s.d 1.57E+01 1.52E+O1 1.42E+O1 4.64E+00 3.80E+00 6.44E+00c.v. 15.89 12.87 21.66 14.04 12.63 79.73glacial outwash median 1.1 2E+02 1 .42E+02 1 .OOE+02 2.53E+00 3.52E+00 1 .20E+00mm 5.99E+01 1.31 E-i-02 3.84E+O1 7.OOE-O1 7.OOE-01 1.20E+00max 1.31E+02 1.65E+02 2.45E+02 4.19E+00 4.73E+00 1.2OEi-00max/mm 2.18 1.26 6.38 5.99 6.75 1.00mean 1 .03E+02 1 .44E.i-02 1 .22E+02 2.40E+00 2.83E+00 1 .20E+00s.d 2.66E+01 1.30E+01 8.09E+01 1.68E+O0 2.OOE+00 0.OOE+00c.v. 25.84 9.04 66.23 69.75 70.59 0.00SALMON RIVERSTREAMBED SEDIMENTSAugust 1991 median 1.54E+02 1.21E+02 4.1OE+O1 2.63E+01 2.70E+01 3.69E+01mm 9.21E+01 1.15E+02 2.69E+O1 2.I1E+01 1.62E+01 1.086i-01max 3.44E+02 I .40E+02 8.40E+01 3.09E+O1 4.33E+01 6.74E+01max/mm 3.73 1.22 3.12 1.46 2.66 6.26mean 1.70E+02 1.24E+02 4.31E+01 2.66E+01 2.72E+01 3.91E+01s.d 6.13E+01 7.69Ei-00 1.17E+O1 2.50E+00 7.27E+00 1.66E+01c.v. 36.07 6.18 27.05 9.39 26.69 42.47December 1991 median 1.51 E+02 1 .37E+02 3.92E+O1 2.42E+01 2.24E+01 2.38E+01mm 9.28E+01 1.13E+02 1.98E+01 7.OOE-O1 4.51 E+00 1.20E+00max 2.32E+02 1 .78E+02 6.34E+01 4.48E+01 1 .83E+02 1 .36E+02max/mm 2.50 1.58 3.20 63.96 40.45 113.47mean 1.52E+02 1.36E+02 3.84E+O1 2.17E+O1 4.08E+O1 4.1OE+01s.d 3.69E+O1 1 .47E+01 9.08E+0O 1.31 E+O1 4.40E+O1 4.58E+01c.v. 24.20 10.81 23.62 60.48 107.72 111.65Appendix 8226Appendix 8 continuedFe Mn Al C(%) PSURFICIAL MATERIALSglacial marine median 5.19E+04 6.43E+02 5.67E+04 1.99E-01 4.74E+02mm 5.06E+04 6.04E+02 4.70E÷04 1.18E-01 1.72E+02max 5.21 E+04 7.80E+02 6.04E+04 8.83 E-01 8.07E+02max/mm 1.03 1.29 1.29 7.51mean 5.15E+04 6.78E+02 5.53E+04 3.58E-0l 4.83E+02s.d. 6.32E+02 7.85E+01 5.43E+03 3.28E-0l 2.35E+02c.v. 1.23 11.58 9.83 91.45marine median 6.18E+04 9.20E+02 6.08E+04 1.58E-O1 5.27E+02mm 5.39E÷04 6.83E+02 5.1 7E+04 7.89E-02 3.95E+02max 6.56E-i-04 1.O1E+03 6.73E+04 1.46E+O0 8.85E+02max/mm 1.22 1.48 1.30 18.50mean 5.98E+04 8.70E+02 5.94E+04 4.03E-01 5.87E+02s.d 5.30E+03 1.41E÷02 5.94E÷03 5.92E-01 1.85E+02c.v. 8.86 16.21 10.01 146.96glacial outwash median 7.68E+04 1 .72E+03 7.22E+04 1 .79E+00 3.96E+03mm 2.61E+04 4.12E-i-02 5.93E+04 8.83E-01 1.70E+03max 1 .30E+05 3.25E-i-03 1 .13E+05 7.83E+00 4.60E+03max/mm 4.99 7.89 1.90 8.87mean 7.31 E+04 1 .60E+03 8.08E+04 2.88E+00 3.36E+03s.d 3.80E+04 1.11E+03 2.05E÷04 2.85E+00 1.37E+03c.v. 51.99 69.67 25.40 99.07SALMON RIVERSTREAMBED SEDIMENTSAugust 1991 median 4.54E+04 1 .66E+03 6.35E+04 3.99E+00 1 .63E+03mm 3.31 E+04 5.34E+02 4.56E+04 8.40 E-01 6.38E+02max 1.14E-,-05 6.84E-i-03 8.44E+04 1.1OE+01 3.73E+03max/mm 3.44 12.82 1.85 13.08 5.84mean 4.87E+04 1 .77E-,-03 6.40E+04 4.54E+00 1 .66E+03s.d 1.72E+04 1.44E+03 9.67E+03 2.18E+00 7.51E+02c.v. 35.42 81.05 15.11 48.03December 1991 median 4.78E+04 1.08E-i-03 5.73E+04 3.37E+00 9.33E+02mm 3.72E+04 7.31 E+02 4.67E+04 1 .42E+00 1 .50E+01max 7.96E-i-04 5.80E+03 6.76E+04 6.19E+00 1.78E+03max/mmn 2.14 7.93 1.45 4.35mean 5.01 E+04 1 .40E+03 5.75E+04 3.54E+00 9.41 E+02s.d 8.73E+03 1.1OE+03 5.90E+03 1.09E+00 4.45E÷02c.v. 17.43 78.51 10.26 30.69Appendix9Total concentrationoftraceelementsinriversedimentsandparentmaterial typesintheSalmonRiverwatershedandinthethemarinereferencestandard, MESS-i.StationZnCrCuCoNiPbFeMnAlnumbermg/kg*not detected,valueisdetectionlimitGlacialMarine18.35E+O11.17E+025.43E+O14.05E÷O13.05E÷Oi1.20E+OO*5.19E÷047.43E+025.30E÷04parentmaterial28.89E÷O11.30E+025.27E+O14.92E+Oi3.85E+O11.20E+OO*5.06E+047.80E+025.92E+0438.46E+O11.43E÷026.16E+O14.25E+Oi3.58E÷O11.20E+OO*5.11E+046.04E+025.67E+0448.16E÷O11.46E+025.79E+O14.03E÷O12.98E+O11.20E+OO*5.21E÷046.20E+026.04E÷0458.74E÷O11.39E+026.23E+O14.40E+O13.32E+Oi1.20E+OO*5.19E+046.43E+024.70E+04Marine11.OOE+029.38E÷O17.02E+O12.93E+O12.65E+O11.47E+O16.56E+049.75E÷026.73E+04parentmaterial21.14E+021.30E+028.32E+O12.86E÷O13.18E+Oi1.39E+Oi6.33E÷047.63E+026.15E+0431.11E+021.31E+024.82E-i-O13.62E+O12.67E+Oi8.90E+OO5.39E+041.O1E+036.08E+0449.63E+O11.23E+027.i5E+O13.94E+O13.56E+O1i.71E+OO6.18E+049.20E÷025.57E+0457.41E+O11.i4E+025.39E+O13.17E+O12.99E÷O11.20E+OO*5.46E+046.83E+025.17E+04glacial outwash11.12E÷021.45E+023.84E+O13.89E+OO4.52E+OO1.20E+OO*5.55E+047.39E÷025.93E+04parentmaterial29.93E+O11.36E+021.OOE+024.19E÷OO4.73E+OO1.20E÷OO*7.74E+041.87E+037.22E+0431.31E+021.42E+022.45E+022.53E+OO3.52E+OO1.20E+OO*7.68E+041 .72E+038.77E-i-0445.99E-i-O11.31E+027.18E+O17.OOE-O1*7.OOE-O1*1.20E+OO*2.61E+044.12E+021.13E÷0551.14E+021.65E+021.56E+027.OOE-O1*7.OOE-Oi*1.20E+OO*1.30E+053.25E+037.18E+04Referencesediment11 .90E+027.28E÷O12.83E+O12.07E+O11 .92E+O14.89E+O13.04E+044.69E+027.04E+04MESS-i21.88E+027.22E+O13.23E+Oi2.32E÷O11.90E+O12.58E+Oi3.37E+045.17E+026.23E+0431.69E+021.03E+022.91E+O11.97E+O12.70E+Oi3.44E÷O13.46E+044.90E+026.03E+044i.87E+026.80E+Oi2.78E÷O13.40E+OO4.08E+OOi.20E+OO*3.13E+045.96E-i-024.92E+0451.85E+026.OOE+Oi2.85E+Oi7.OOE-Oi7.OOE-O1*i.20E+OO*3.37E+045.59E+026.60E+0461.65E+026.90E÷O12.64E-i-Oi2.94E÷Oi1.69E+O12.02E÷OO3.25E+045.09E+024.45E+04(‘3 (‘3Appendix9continued(pageiiof vi)StationZnCrCuCoNiPbFeMnAlnumbermQ/kgSalmonR.sediment111.09E+021.19E+023.56E+012.77E÷O12.05E+013.98E÷011.14E+056.84E÷035.37E÷04Aug21,1991171.50E-i-021.18E÷023.80E+012.50E+012.07E+012.12E+013.85E+045.34E+026.92E+04151.39E+021.26E+023.23E-,-012.43E+013.89E+O12.61E+013.31E÷047.76E+026.16E+0491.44E+021.20E+023.96E+012.68E+012.22E÷O12.53E+014.24E+041.13E+037.33E÷0471.74E+021.21E÷024.72E+012.85E+012.66E+013.69E+015.19E+042.58E+036.49E+0441 .54E+021.37E+024.21E+O13.01E÷012.84E+016.58E÷014.92E÷041.92E+037.26E+0461.52E+021.40E+024.15E+012.77E+012.40E-i-015.70E÷015.OOE+041.67E+036.35E+04161.40E+021.21E+023.91E+012.54E+012.90E+O11.08E+014.23E+046.93E+025.35E÷0421.49E÷021.18E÷023.83E÷012.70E÷011.62E+016.16E÷015.08E+041.03E+037.17E+0411.05E+021.31E+023.81E+012.62E+012.42E+013.18E+014.64E+047.82E+027.1OE+04CoghlanCr.sediment121.65E+021.16E+023.70E+012.50E+013.13E÷O13.92E+014.21E+042.37E+036.38E+04Aug21, 1991201.99E+021.18E÷024.19E+013.06E+012.73E+013.63E+014.68E+042.90E+035.89E+04191.97E+021.30E+024,59E+012.11E+012.77E+012.07E+014.21E+041.O1E+035.77E÷0451.64E÷021.29E+025.O1E+012.56E+012.70E+016.20E+014.16E÷041.72E+034.56E+04SalmonR.tributaries103.02E+021.32E+028.40E÷013.09E+013.09E+013.81E+016.02E+041.80E+037.78E÷04sediment,Aug21,199183.44E-i-021.15E÷025.32E+012.63E+O12.18E+013.62E÷014.54E+042.76E+035.40E+0431.64E÷021.17E+024.72E+012.90E÷011.78E+016.74E+015.54E÷041.66E+038.44E÷04131.85E+021.20E+024.1OE+012.59E+014.33E÷O14.18E+013.76E+047.46E+026.23E+04149.21E+011.34E+022.69E+012.33E+013.95E+012.52E+013.51E+047.85E+025.69E+04*notdetected, valueisdetectionlimitAppendix9continued(pageiii ofvi)StationZnCrCuCoNiPbFeMnAlnumbermQ/kgSalmonR.111.76E+021.19E÷024MIE+018.64E÷0O7.35E+001.20E+0O*7.96E+045.80E+035.76E÷04sediment,Dec10,1991179.28E+011.42E+021.98E+012.11E+012.21E+011.91E÷013.72E+047.31E+025.93E+04151.23E+021.21E+022.40E-i-017.OOE-01*4.51E+001.20E÷004.51E÷049.64E+025.44E+0491.28E+021.31E+023.29E+016.86E+0O1.98E÷014.25E+014.52E+041.06E÷035.20E+0471.23E+021.37E+023.44E+012.54E+019.41E+017.14E+014.78E+041.13E+035.27E+0441.51E+021.42E÷024.15E÷013.94E+019.97E+011.09E÷024.64E+041.17E+035.73E+0461.34E+021.42E+023.52E-i-013.08E+014.18E+011.09E+025.06E÷049.97E÷025.35E+04161.43E+021.23E+023.79E+012.1OE+012.24E÷011.20E÷005.41E+041.36E÷036.48E+0421.68E+021.39E+024.11E+012.74E+014.08E+016.92E+014.86E+041.07E+035.70E+0411.30E+021.47E+024.35E÷012.58E÷016.06E+011.73E÷015.12E+049.44E+026.21E+04CoghlanCr.121.83E+021.26E+023.92E+014.23E÷008.61E+001.20E+004.87E+041.08E÷034.67E+04sediment,Dec10,1991201.84E+021.13E+023.95E+012.71E+011.99E+012.38E+014.75E+041.63E+036.34E+04191.54E+021.55E+023.79E+012.42E+012.40E+012.82E+014.94E+041.33E+036.33E+0451.56E+021.78E+024.40E+013.79E+015.64E+011.36E÷024.75E÷041.13E+036.06E+04SalmonR.tributaries102.28E+021 .34E+026.34E+O18.87E+001.32E÷011.20E÷006.37E+041.89E+036.42E+04sediment,Dec10,199182.32E+021.37E+024.46E+012,12E+012.23E+012.58E+014.74E+041.45E+035.33E+0431.58E+021.27E+024.57E+014.48E÷011.83E+021.14E+024.68E+041.08E+036.76E+04131.24E+021.31E+023.58E+013.35E+012.86E+016.61E+004.56E+048.41E+025.61E+04141.07E÷021.46E+022.99E÷012.70E+006.68E+001.20E+004.BBE+049.55E+024.70E+04*not detected,valueisdetectionlimitAppendix9continued(pageivofvi)StationNaMgCaTiZrSrVPCnumbermg/kg(%)GlacialMarine13.OOE+035.86E+039.30E+035.45E+035.93E+0i7.89E+011 .74E÷024.74E÷020.5parentmaterial23.OOE+035.25E+031 .33E÷045.35E+035.52E÷016.04E÷0i1.83E+028.07E+020.933.OOE+034.61E+032.15E+044.67E÷035.82E+018.98E+011.65E+025.78E÷020.143.OOE+035.91E+031.38E+044.82E+032.76E+028.16E÷011.73E+021.72E+020.153.00E+036.03E+038.07E+034.64E+036.68E+018.08E+0i1.71E+023.87E+020.2Marine16.35E+038.43E+031.08E+045.71E+035.81E+017.05E+012.09E+025.05E+020.2parentmaterial23.69E+037.54E+031.09E+045.78E+034.97E+0i7.82E÷0i1.93E+025.27E÷020.234.53E+036.33E+039.04E+036.19E+034.97E+018.85E+01i.68E+028.85E÷021.543.OOE+037.77E+031.41E+045.65E+037.88E+018.06E+011.83E+026.23E÷020.253.OOE+036.45E+031.14E÷045.58E+035.95E+018.62E+016.OOE÷003.95E+020.1glacialoutwash17.60E÷032.99E+031.64E+044.93E+035.89E+017.OOE+0i1.45E+021.70E÷031.8parentmaterial21 .70E÷043.61E+031.30E+044.91E+034.OOE÷015.23E+0i1.85E+023.96E+030.932.74E+047.56E+034.27E÷034.07E+032.25E+02i.49E+021.80E+022.08E÷032.642.52E+043.89E+032.91E+032.53E+038.07E+0i1.88E+007.57E+014.47E+037.852.98E+042.94E+031.91E+044.08E+034.58E÷014.18E+0i2.17E+024.60E+031.2Referencesediment12.17E+041 .09E+046.41E÷034.44E+032.85E+023.99E÷016.73E+0i3.47E+02MESS-i21.97E+041.11E+045.53E÷034.52E+033.23E÷024.04E+016.84E+014.1OE+0231.41E+048.13E+034.70E+034.45E÷033.24E÷024.13E+016.39E+013.30E+0241 .32E+046.46E+035.60E+035.OOE+032.08E+023.21E÷016.93E+015.41E÷025i.51E+049.75E+036.78E+035.38E+032.07E+023.19E+019.OOE+019.94E÷0263.OOE+035.13E+033.19E+034.64E+032.21E+023.18E+017.03E+014.47E+02*not detected,valueisdetectionlimitwAppendix9continued(pagevofvi)StationNaMgCaTiZrSrVPCnumbermg/kg(%)SalmonR.sediment111.50E+046.35E÷031.25E÷043.82E+036.BOE+O12.50E+019.90E+013.73E+036.1Aug21,1991171.28E÷046.53E+031.44E+044.32E+039.42E+O17.98E÷O11.32E÷021.70E÷034.0151.74E+045.25E+031.15E+044.49E+038.59E+015.33E÷011.20E+021.20E+034.091.58E÷048.37E+031.21E+044.88E+037.63E+013.17E÷011.25E+029.61E÷023.471.16E+048.75E÷031.18E+045.04E+036.82E+O11.60E+017.53E+011.84E+035.241.56E+041.06E+041.68E÷045.68E+036.37E+012.29E÷018.73E+011.84E÷033.961.27E+047.95E+031.56E+045.24E+036.14E+011.82E+015.15E+011.37E+033.7161.96E+046.83E+036.97E+034.28E+039.41E+O12.44E+011.38E÷021.12E+032.821.64E+041.02E+049.08E÷036.17E+035.31E+012.35E÷019.78E+011.92E÷032.711.49E÷048.12E÷031.65E+045.66E÷031.O1E+022.35E+011.26E+021.92E+030.8CoghlanCr. sediment129.82E+036.15E+037.43E+034.69E+037.72E÷015.06E+011.04E+026.38E÷024.0Aug21,1991201.28E+047.30E+039.68E+034.39E+038.21E+015.03E÷011.02E÷021.16E+035.3191.70E÷047.70E+031.13E+044.08E+037.73E+012.41E+011.20E+021.60E+036.051.46E+044.93E+031.32E÷044.15E+036.20E+012.11E+011.15E+021.68E÷037.3SalmonR.tributaries109.56E÷037.59E+031.63E÷044.93E+037.76E+015.53E÷O16.68E+012.71E+035.8sediment,Aug21,199181.40E+045.79E+031.51E÷044.40E+037.82E+O12.34E÷011.05E+021.63E+0311.031.78E+041.18E-i-041.38E+046.19E+036.98E+012.83E+011.08E+022.67E+033.2131.61E+046.91E+036.83E+034.31E+038.81E÷012.20E+011.11E+021.11E+034.5141.46E+046.14E+031.07E+044.38E+037.47E+011.86E÷O11.36E÷027.45E+022.7*not detected,valueisdetectionlimitAppendix9continued(pageviofvi)StationNaMgCaTiZrSrVPCnumbermg/kg(%)SalmonR.111.33E+043.97E+031.76E+044.90E+035.1OE+015.30E+016.11E+011.36E÷034.4sediment,Dec10,1991171.1OE+044.44E÷038.06E+036.58E+033.65E÷017.83E+011.21E+028.16E÷022.3157.04E+034.49E+031.53E+045.62E+035.11E+014.72E÷011.03E+021.14E+033.591.15E+045.04E+031.18E÷044.85E+031.30E+023.32E+018.28E+014.94E+023.471.40E+045.61E+031.09E+045.59E+035.41E+017.81E+001.05E+024.84E+022.741.50E+047.54E+031.33E+045.98E+035.94E+011.OOE+011.25E+029.33E÷023.461.71E+046.03E+031.35E+045.89E+035.69E+011.17E+011.23E+027.84E+023.2168.11E+037.66E+037.23E+036.38E+035.32E÷011.35E+011.25E+029.91E+023.321.28E+046.18E+031.18E+045.64E+031.13E+021.65E+011.14E+028.35E+022.711.09E+046.89E+031.70E+045.99E+039.74E+013.85E+011.48E+021.12E+031.4CoghlanCr.121.1OE+043.55E÷031.65E+044.95E+035.1OE+019.OOE+015.17E+011.43E+023.9sediment,Dec10,1991208,58E+035.95E+036.38E+035.97E+033.56E÷011.83E+011.17E+021.12E+035.3199.71E+037.O1E+037.94E÷036.20E+033.86E+012.41E+001.31E+029.26E+024.551.55E+047.26E÷031.32E+045.52E+035.31E+011.O1E÷011.23E+021.02E+034.0SalmonR.tributaries108.92E+035.22E+032.09E+045.30E+036.13E+015.14E+017.46E÷019.30E+023.9sediment,Dec10,199181.42E+046.02E+038.59E+035.27E+037.34E÷012.02E+011.51E+021.78E+036.231.44E+047.41E+031.58E+045.84E+035.36E+012.64E+011.35E+021 .58E+032.6133.OOE+035.09E+038.89E÷036.O1E+034.71E÷017.63E÷O11.35E÷021.41E÷034.0149.44E+035.50E+031.12E+045.11E+035.19E÷018.89E+006.59E+011.50E+012.8*not detected,valueisdetectionlimitM w t\JAppendix10Samplepreparationanddigestionmethodsofsedimentsamplescollectedin1970’sand1991SampledateSampleSamplepreparationSampledigestionElementcollectionanalysissummer1974grabsample-ovendried110°Cfor 48-1.0gsubsampledigestedwith10mLof4:1AASHall(unpublisheddata),hr.concentratedHNO3:70%HCIO4methodfromHalletal.-disaggregatedinmortar-ref luxedfor1hr.1976-sievedthrougha177jim-evaporatedtodryness(80mesh)nylonscreen-dilutedwith5ml6MHCI and15mldistilledwaterMay1974grabsample-airdried-subsamplegroundtopass60mesh(0.2mm)AASBeale(1976)-groundlightlytopass2sievemmmesh-digestionwithHNO3,H2S04,HF,andHCIO4BindraandHall(1977)grabsample-wetsievedinfield-subsamplefinelygroundinanagatemortarAASthrough2mmplasticsieve-0.5gsubsampledigestedonsandbathat180°C-driedtoconstantweightwith10mLHNO3and10mLof70%HFuntildryat103°C-5mLHCIO4and5mLHNO3addedandevaporateduntildry-10mLaliquotsofHFaddedandevaporateduntilsilicatemineralsdissolved-ResiduedissolvedindiluteHCIandbrought to50mLvolumewith5%solutionof1:1HNO3:HCIthisstudy(1991)grabsample-wetsievedtopass-subsamplegroundtopass100mesh(150jim)ICP-ES63jimmeshsieve-ovendriedat60°C-ovendriedat105°C-microwavedigestionof0.5gsamplewithHNO3,HCI andHF(“3 (AJwAppendix11Summarystatisticsforthestreambedsedimentscollectedinthe1970’sandAugust1991intheSalmonRiver.Valuesareinmg/kgexceptforcarbon,reportedinpercent.ZnCrCuCoNiPbFeMnAlC1974-1977median7.1OE+O11.53E+021.60E+O11.27E+013.44E+O11.41E+012.12E+045.18E+021.21E+055.15E-O1mm1.92E+019.18E+O10.OOE+OO3.80E+001.57E+O1O.OOE+006.09E+031.70E+025.38E-i-042.1OE-01max1.37E+022.75E-i-024.79E+019.44E+O17.86E+O14.90E+O11.06E+051.38E+041.54E+052.58E÷00max/mm7.113.0024.845.0017.4181.222.8712.29mean7.43E+011.75E+021.94E÷O13.52E+014.06E+011.80E+013.14E-i-049.94E-,-021.14E+051.03E+00S.D.2.66E+015.23E+019.79E-t-0O3.62E+011.92E+O11.55E+012.33E+042.O1E+033.18E+048.46E-01c.v.%279.4334.1198.197.3211.8116.0134.849.5357.4121.5August 1991median1.54E+021.21E+024.1OE+012.63E+012.70E+O13.69E+O14.54E+041.66E+036.35E+043.99E+00mm9.21E+O11.15E+022.69E+012.11E+011.62E+011.08E+013.31E-i-045.34E+024.56E-i-048.40E-01max3.44E+021.40E+028.40E+013.09E+014.33E+O16.74E+011.14E+056.84E+038.44E+041.1OE÷013.731.223.121.462.666263.4412.821.8513.08mean1.70E+021.24E+024.31E+012.66E+012.72E+013.91E+014.87E-i.041.77E+036.40E+044.54E+00S.D.6.13E+017.69E+001.17E+012.50E+007.27E+001.66E+011.72E+041.44E+039.67E-i-032.18E+00c.v.%277.21618.3369.71065.1374.7235.5282.3123.4661.8208.2t%JwAppendix12DeterminationofpHinsurfacewaterintheSalmonRiver,August1991-September 1992.stationcontributingdatenumberdesciiptiontributary?area91.08.2191.12.1092.02.0592.08.2493.09.29SalmonR.upstreamofCoghianCr.KC11SalmonRiverheadwatersKC17KC15KC1OKC09KC24KCO7KCO8KCO4CoghianCreekKC12KC21KC2OKC1 9KC25KC22KCO5CoghianCreekheadwatersSalmonR.downstreamofCoghianCreekKC23KCO6KCO3KC16KCO2KC13KC14KCO1SalmonR.nearmouthNA=notmeasuredyesyesyesS00+S0iS02S03S04S05S06S07S08S097.1 8.1 NA 7.37.1 7.07.5NA NA NA NA NA NA NA NA NAColC02C03C04C05C06C07yesyes7.5NA 7.77.4NA NA 7.66.9NA 7.27.3NA NA 7.16.9NA 7.1 7.2NA NA 7.47.3NA 7.56.8NA NA 7.8NA NA NA NA NA NA NAyesyesyesSi0Si1S12S13S14DOlD02S15NA 7.1 7.1 6.8NA 7.1NA NA NA NA NA NA NA NAC-.-)UIAppendix13DeterminationoftemperatureinsurfacewaterintheSalmonRiver,August1991-September1993.AllvaluesreportedareindegreesCelcius.stationcontributingdatenumberdescriptiontributary?area91.08.2191.12.1092.02.0592.08.2493.09.29SalmonR.upstreamofCoghlanCr.KC11SalmonRiver headwatersS00+S0117.05.05.518.011.0KC17S0220. CoghlanCreekKC23D10NANANANA12.0KCO6Sil19.05.56.515.012.0KCO3yesS1220.,August1991-September1993.AllvaluesreportedhaveunitsmS/m.stationcontributingdatenumberdescriptiontributary?area91.08.2191.12.1092.02.0592.08.2493.09.29SalmonR.upstreamofCoghlanCr.KCIISalmonRiverheadwatersS00+S0129.$0511.,August1991-September1993.Allvaluesreportedhaveunitsmg/L.stationdatenumberdescriptiontributary?area91.08.2191.08.2191.12.1092.02.0592.08.2493.09.29SalmonR.upstreamofCoghianCr.KC11SalmonRiver headwatersS00+S01KC115.06NA3.2013.913.20KC17S02KC177.38NA3.4325.3032.00KC15S03KC152.75NA3.422.262.00KC1OyesS04KC1O7.72NA5.959.957.50KCO9S05KCO94.06NA3.454.925.40KC24yesS06KC24NANANANA9.40KC07S07KCO75.29NA3.975.435.70KCO8yesS08KCO814.44NA8.5510.1511.50KCO4S09KCO47.51NA4.376.807.00CoghlanCreekKC12CoghlanCreekheadwatersCOlKC1219.39NA8.6812.620.00KC21C02KC21NANANANA12.10KC2OC03KC2O9.90NA6.656.967.90KC19C04KC198.33NA5.936.026.60KC25C05KC25NANANANA7.40KC22yesC06KC22NANANANA4.90KCO5yesC07KCO58.53NA6.468.318.20SalmonR.downstreamofCoghlanCreekKC23Dl 0KC23NANANANA7.70KCO6SliKC068.50NA5.418.398.50KCO3yesS12KC03115.58NA14.04107.75139.00KC16S13KC1616.54NA7.2913.9317.75KCO2S14KCO217.95NA7.4116.0916.60KC13yesDOlKC136.71NA4.166.185.30KC14yesD02KC146.64NA3.726.286.40KCO1SalmonR.nearmouthS15KCO119.51NA3.9311.401.40NA=notmeasuredAppendix16Determinationoftotal phosphorusinsurfacewaterintheSalmonRiver,February1992-September 1993.Allvaluesreportedhaveunitsmg/L.stationcontributingdatenumberdescriptiontributary?area91.08.2191.12.1092.02.0592.08.2493.09.29SalmonR.upstreamofCoghlanCr.KC1 1SalmonRiver headwatersS00+S01NANA0.0810.5470.095KC17S02NANA0.1300.3710.299KC15S03NANA0.0830.0420.045KC1OyesS04NANA0.0850.0740.157KCO9S05NANA0.0670.0230.040KC24yesS06NANANANA0.077KC07S07NANA0.0530.0210.018KCO8yesS08NANA0.0260.0270.026KC04S09NANA0.0470.0690.019CoghlanCreekKC12CoghlanCreekheadwatersCOlNANA0.0360.0380.000KC21002NANANANA0.017KC20C03NANA0.0560.0260.024KC19C04NANA0.0580.0220.018KC25C05NANANANA0.017KC22yesC06NANANANA0.027KC05yesC07NANA0.0500.0180.015SalmonR.downstreamofCoghlanCreekKC23D10NANANANA0.014KCO6SilNANA0.0520.0200.013KCO3yesS12NANA0.2430.4360.185KC16S13NANA0.0670.0220.025KCO2S14NANA0.0690.0250.012KC13yesDOlNANA0.1010.1540.176KC14yesD02NANA0.0600.0260.018KCO1SalmonR.nearmouthS15NANA0.0640.0750.019NA=notmeasuredAppendix17Determinationofnitrite+nitrate-NinsurfacewaterintheSalmonR., August 1991-September 1993.Allvaluesreportedhaveunitsmg/L.stationcontributingdatenumberdescriptiontributary?area91.08.2191.12.1092.02.0592.08.2493.09.29SalmonR.upstreamofCoghianCr.KC11SalmonRiverheadwatersKC17KC1 5KC1OKCO9KC24KCO7KCO8KC04yesyesyesS00+S0iS02S03S04S05S06S07SOBS090.160.280.610.380.50 NA 1.424.623.091.881.891.654.181.77 NA 1.49 NA 2.405.722.840.120.100.790.490.81 NA 1.334.633.320.020.020.580. 6KCO2KC13KC14KCOISalmonR.nearmouthNA=notmeasuredCOlC02C03C04C05C06C07yesyes0.50 NA 0.971.80 NA NA 7.120.94 NA 2.152.32 NA NA 3.400.99 NA 2.722.74 NA NA 3.810.43 NA 0.871.65 NA NA 6.750.000.490.971.941.690.946.84yesyesyesDl 0Si1S12S13S14DOlD02Si5NA 4.612.854.544. 2.582.802.532.153.092.522.21NA 2.961.692.772.773.112.960.71NA 4.790.254.504.11 0.005.492.085.405.020.404.614.080.045.340.09I\jC241Appendix 18 Measurement of the precision of the nitrate.i-nitrite-N determinationsin surface water in the Salmon River. Measurements in mg/L.Sample Station replicate replicate replicate replicate replicate maximum!Date Number 1 2 3 4 5 minimum21/08/91 5 7.07 7.17 1.017 1.38 1.46 1.061 2.92 2.86 1.0210 0.35 0.42 1.203 2.81 2.89 1.0314 6.06 6.01 1.0110/12/91 5 3.32 3.47 1.054 2.44 2.46 1.0116 2.56 2.49 1.031 2.43 1.99 1.2210 4.24 4.11 1.0314 2.57 2.47 1.0405/02/92 5 3.79 3.79 3.83 1.017 2.38 2.42 1.024 2.81 2.87 2.85 1.026 2.87 3.05 1.061 0.83 0.59 1.4114 2.94 2.98 1.0124/08/92 1 2.08 1.96 1.062 3.99 3.97 4.26 4.2 4.13 1.074 3.32 3.32 1.006 4.88 4.71 1.047 1.28 1.5 1.26 1.28 1.1913 0 014 5.49 5.39 1.0215 0.85 0.73 1.1616 4.34 4.67 1.0819 1.55 1.76 1.83 1.1829/09/93 16 4.65 4.56 1.02242Appendix 19 Measurement of the accuracy of the nitrate÷nitrite-Ndeterminations in the surface water in the Salmon River watershednitrate-N description result mean SD expected!mgIL mgi L measured0.2 STDCHKO.2A 0.30 0.13 0.11 0.660.2 STD CHK 0.2B 0.020.2 STD CHK 0.2C 0.040.2 STDCHKO.2D 0.130.2 STDCHKO.2E 0.182.0 STDCHK2A 2.02 1.90 0.12 0.952.0 STDCHK2B 1.942.0 STDCHK2C 1.712.0 STDCHK2D 1.912.0 STD CHK 2E 1.935.0 STDCHK5A 4.94 4.80 0.18 0.965.0 STD CHK 5B 4.485.0 STD CHK 5C 4.895.0 STD CHK 5D 4.895.0 STD CHK 5E 4.77Appendix20Sitenumber identifiersusedforthe12NaquadatwellsitesandthecorrespondingSEAMidentifier codes.90129011901890199021901390149020901590169022901790100920008334092G0083340920008334092G0083340920008334092G008334092G008343092G018112092G018112092G018121092G0181210920018121092G018314IdentifierAssigned4008334.00429008334.02956008334.05669008334.06994008334.09423018112.0236008343.0062018112.00226018112.02618018121.01824018121.02441018121.0419018314.009NAQUADATSiteNo.11921632163016371024103816351626162714091631E21 3037E207083E207086E20 70804614108412147148541313104166497108810158217320NAQUADATMapNo.BCGSUniqueWellOtherSEAMWQCPTownshipSectionDepthSiteNo.WellNo.No.(ft)E207082E20708411 11 11 11 11 11 11 11 11 11 11 11 11WQ123(%J()Appendix21SummaryofsepticsysteminstallationsintheSalmonRiver watershedandHopingtonAquifer,derivedfromMinistryofHealthrecordslocatedattheCentralFraserValleyHealthUnitSalmonR.WatershedCoghlanCr.Sub-watershedHopingtonAquiferYearCountGallonsLateral Ft.CountGallonsLateral Ft.CountGallonsLateral Ft.19705233350112851267502280412525082501971639001350424009201972159300307031850730640501250197379455001496027160005240633695012010197482620502117917104003610513065010098197584518001742312740024944629300922119761671022503475726153005347102638002255419779961750246511799504030563572013662197897613002076114905035685132150114721979634000013557525501030261505047421980146934502711251328007948110705501918819811077806523349322055057458155275162581982614093012837241552046924126800730419838452600155182717100487966406501076619844628875103069535018243421500627619856439950138832214300468140249507305198670459001428632208505468473085088711987885420016558301970048205736900102021988654395013601231360041505032900875719896842848164572918950928746295481041919905234947122527460017383830467865519915233140128261912300545141263909784199231194507263627001900149400281319936422815582120044821200148Total16841083733360799446278770913601113712700220925RecordsforSalmonR.watershedincludestheCoghlanCr. sub-watershedandaportionoftheHopingtonAquifer anddoesnotincludetheportionofthewatershedwithinMatsquiDistrictf\3x ci)-o C o.U) C ci)-oAppendix22Relationshipsbetweenthedensityindexofagricultural activitiesonallsurficialmaterialsandnitrate-Ninsurfacewaterduringlow-flowconditionsintheSalmonRiver watershed.Thedensityindexistheproportionofthelandareawiththespecifiedagricultural landuse.-J 0)zx ci> C-l u) C ci)-aC’..0)‘U)C)I).C.jC’Jcu)r-U)0)U)r-CJ90099••;,-900990090••-9000U)40U)U)4U)th0CtI)—4c0000000000000oo0Cl)U(j)00000)(1)(fl(”U)0e0[eadwatersrmIhcontributingarea_______________________surfacewatercontributingareax ci V .90.>.U) C ci) VCIC’)U)r-oU)000—‘-a)4thcb0C’)oooo—coØCOCJ)a)FeadwatersrroUhC’)900—CD0 000-J 0)zci)C’)U)r—900•-‘-CD000-J 0)2.0.gz I6.Oo ct5 C,)Appendix23Relationshipsbetweenthedensityindexofagriculturalactivitiesonglacial outwashmaterialsandnitrate-Ninsurfacewaterduringlow-flowconditionsintheSalmonRiver watershed. Thedensityindexistheproportionofthelandareawiththespecifiedagriculturallanduse.x a) V C U) C ci) V0.08C” 9 0 Oc’C’)U)r-U)0o000.a),4thth0C’Ja)0000’—a)cocoa)a)_•_surfacewaterNO3CJ 9 0 0contributingareacontributingareax a) V > U) a) V a) > D E 0x ci) V C >% 4- U) C ci -o ci) > D E C)C)U)Cooo00000_surfacewaterNO3—3Appendix24Relationshipsbetweenthecumulativedensityindexofagricultural activitiesonallsurficialmaterialsandnitrate-Ninsurfacewaterduringlow-flowconditionsintheSalmonRiver watershed.Themeasureofagriculturallanduseistheproportionofthelandareawiththespecifiedlanduse.Seesection5.6for explanationofthecumulativedensityindex._J_J0)—c’jU)r—0.-o00000——[eadwatersL()—_0)0)c’j——0nuthcontributingareacontributingareaAppendix25Relationshipsbetweenthecumulativedensityindexofagricultural activitiesonglacial outwashmaterialsandnitrate-Ninsurfacewaterduringlow-flowconditionsintheSalmonRiver watershed.Themeasureofagriculturallanduseistheproportionofthelandareawiththespecifiedlanduse.Seesection5.6for explanationofthecumulativedensityindex. -J C)zd Ix ci)-o C >% 4-. U) C ci-o ci) > 2 D 0-J—0)XzI>cihC ci)C-O-wciaSDC) aS•t:0D U)cLf)r——C4c)Lf)F—0)o00C’000—0co000000CO000_•_surfacewaterNO3—cU)r0)0000000•0004cbthC9000—0c,)coIeadaters—0(j)0mh— 00(1c’J,i00C’)U)r000400000contributingareacontributingarea0)249Appendix 26 Median nitrate-N values in well water tested between 1983 and 1993 foreach contributing area.contributing wells on shallow wells, .c=100 wells on glacialarea all surficlal materials ft outwash surflcialmaterialsn= median n= median n= medianSalmon River watershed, excludingCoghlan Creek and Davidson Creeksub-watershedsS01-02 2 0.39 1 0.02 0S03 1 4.2 0 1 4.2S04-05 9 0.65 2 12.5 7 0.8S06-07 39 4.8 22 5.4 39 4.75S08-09 15 5.1 8 6.0 15 5.1Sb-li 7 3.5 3 5.7 4 7.53S12-14 7 0.02 2 2.5 0S15 13 0.26 8 0.63 2 3.6Coghian Creek sub-watershedC01-03 13 0.07 8 1.63 2 2.6C04-05 9 3.6 7 3.8 3 4.8C06-07 15 5.0 11 5.0 15 5.0Davidson Creek sub-watershedD0i-02 4 5.6 2 5.6 1 10.6x a)-D C 4- Cl)C a) Vx a) C-J 0)a) IAppendix270.12-—Relationshipsbetweenagriculturalactivitiesonallsurficialmaterialsandnitrate-Ninwellwater.Thewell-waternitrate-Nisthemediannitrate-Nvaluefromwellstestedbetween1983to1993,withineachcontributingarea.Thedensityindexistheproportionofthelandareawiththespecifiedagriculturallanduse.on0.10SafrnonR.0.08+0.06---berry ccqtncr.+z:::zoziF:0.040.02nnn.:rII0.0-J 0)z a) I0.1c_______________finC1c’U)-000004th000000teadwaters____________________0.i)otheranimals008V6.00.064.0CMC)U)F-C)LI)C’)LI)r-CM000001—9009404thth0c0—4th—0ooo’—0000Cl)0000‘4)nn*hC’)LI)N-990—.CD000000CM 9 0 0+wellNO3contributingareacontributingarea0-J c,)z d I-J z d IAppendix28Relationshipsbetweenagricultural activitiesonglacialoutwashmaterialsandnitrate-Ninwellwater.Thewell-waternitrate-Nisthemediannitrate-Nvaluefromwellstestedbetween1983to1993, withineachcontributingarea.Thedensityindexistheproportionofthelandareawiththespecifiedagriculturallanduse.xxci) 0 > .1-n Cl)CCci)a)[eadatersnuthcontributingareae.iC’)U)FU)CoC0C—oCCC—U)U)U)U)U)AwellNO3onoutwashC’)U)-.900‘-‘-CDoo00009 Ccontributingareat\J C-fl H252Appendix 29 Spearman rank correlation coefficients between land use and nitrate-N in water, whenthe Salmon River watershed is grouped into 23, 13, and 10 land areas.* significant at alpha=0.05 for a one tailed test.surface well surface well surface wellwater water water water water waternitrate-N nitrate-N nitrate-N nitrate-N nitrate-N nitrate-N23 areas 13 areas 10 areassurface water N03 1.000 1.000 1.000well water N03 0.434 * 1.000 0.560 * 1.000 0.036 1.000well water N03, <101 ft. 0.224 0.504land use on all surficial material typesdensity of septic systems 0.310 0.311 0.429 0.258 0.576 * 0.578 *berries 0.303 0.046 0.556 * 0.245 0.614 * -0.180poultry and fur bearers 0.279 0.148 0.571 * 0.582 * 0.134 -0.055horses 0.041 -0.166 -0.176 0.099 0.091 0.486other animals -0.177 -0.091 -0.104 0.099 0.103 0.079all animals -0.147 -0.156 0.022 0.236 0.273 0.395other agriculture 0.038 -0.258 -0.088 -0.291 0.406 -0.079all agriculture 0.160 -0.076 0.247 0.220 0.491 0.055land use on glacial outwash onlydensity of septic systems 0.339 0.494 * 0.343 0.348 0.423 0.609 *berries 0.435 * 0.044 0.622 * 0.266 0.800 * -0.405poultry and fur bearers 0.431 * 0.391 * 0.353 0.298 0.381 0.460horses 0.187 0.128 0.061 0.116 0.387 0.437other animals 0.262 0.359 * -0.127 -0.033 0.055 0.437all animals 0.208 0.219 0.011 0.055 0.239 0.523otheragriculture 0.071 0.038 -0.119 -0.090 0.162 0.396all agriculture 0.553 * 0.076 0.644 * 0.228 0.851 * -0.290removal of S15, Fort Langley 22 areas 12 areas 9 areassurface water N03 1.000 1.000 1.000well water N03 0.462 * 1.000 0.594 * 1.000 0.100 1.000well water N03, <101 ft. 0.223 0.465land use on all surficial material typesdensity of septic systems 0.318 0.330 0.441 0.329 0.600 * 0.603 *berries 0.320 0.058 0.557 * 0.196 0.661 * -0.164poultry and fur bearers 0.272 0.159 0.594 * 0.622 * 0.226 0.042horses 0.073 -0.192 -0.147 -0.000 0.217 0.427other animals -0.164 -0.111 -0.063 0.035 0.133 0.075all animals -0.145 -0.214 0.070 0.140 0.383 0.343other agriculture 0.050 -0.247 -0.091 -0.259 0.400 -0.059all agriculture 0.180 -0.120 0.280 0.070 0.517 0.067land use on glacial outwash onlydensity of septic systems 0.360 * 0.593 * 0.387 0.592 * 0.475 0.766 *berries 0.456 * 0.027 0.655 * 0.182 0.801 * -0.345poultry and fur bearers 0.424 * 0.430 * 0.336 0.355 0.420 0.532horses 0.195 0.151 0.056 0.092 0.407 0.409other animals 0.273 0.426 * -0.134 0.141 0.068 0.613 *all animals 0.226 0.252 0.021 0.113 0.288 0.579other agriculture 0.081 0.120 -0.109 0.094 0.201 0.587all agjclture-- 0.569 * 0.102 0.651 * 0.189 0.879 * -0.223


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items