Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Biogeochemical expressions of buried REE mineralization at the Norra Kärr Alkaline Complex, southern… Bluemel, Britt 2014

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

Item Metadata

Download

Media
24-ubc_2014_spring_bluemel_britt.pdf [ 25.41MB ]
Metadata
JSON: 24-1.0103395.json
JSON-LD: 24-1.0103395-ld.json
RDF/XML (Pretty): 24-1.0103395-rdf.xml
RDF/JSON: 24-1.0103395-rdf.json
Turtle: 24-1.0103395-turtle.txt
N-Triples: 24-1.0103395-rdf-ntriples.txt
Original Record: 24-1.0103395-source.json
Full Text
24-1.0103395-fulltext.txt
Citation
24-1.0103395.ris

Full Text

Biogeochemical Expressions of buried REE Mineralization at the Norra K?rr Alkaline Complex, southern Sweden  by Britt Bluemel B.Sc., University of Victoria, 2008   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE  REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in   The Faculty of Graduate and Postdoctoral Studies (Geological Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)    March, 2014 ? Britt Bluemel, 2014   ii  Abstract Biogeochemical exploration is an effective but underutilized method for delineating covered mineralization. Plants are capable of accumulating rare earth elements (REEs) in their tissue, and ferns (pteridophytes) are especially adept because they are one of the most primitive land plants, therefore lack the barrier mechanisms developed by more evolved plants. The Norra K?rr Alkaline Complex, located in southern Sweden approximately 300 km southwest of Stockholm, is a peralkaline nepheline syenite enriched in heavy rare earth elements (HREEs). The deposit, roughly 30 0m wide, 1300 m long, and overlain by up to 4 m of Quaternary sediments, has been well-defined by diamond drilling. The inferred REE mineral resource, over 60 million tonnes averaging 0.54% Total Rare Earth Oxide (TREO), is dominantly hosted within the pegmatitic ?grennaite? unit, a eudialyte-catapleiite-aegerine nepheline syenite. Vegetation and soil samples were collected from the surficial environment above Norra K?rr to address four key questions: which plant species is the most effective biogeochemical exploration medium; what are the annual and seasonal REE variations in that plant; how do the REEs move through the soil profile; and into which part of the plant are they concentrated.  Athyrium filix-femina (lady fern) has the highest concentration of LREEs and HREEs (up to 125.17 ppm Ce and 1.03 ppm Dy) in its dry leaves; however, there is better contrast between background and anomalous areas in Dryopteris filix-mas (wood fern), which makes it the preferred biogeochemical sampling medium.   The REE content in all fern species was shown to decrease from root > frond > stem, and chondrite normalized REE patterns within the plant displayed preferential fractionation of the LREEs in the fronds relative to the roots.   Samples collected from an area directly overlying the deposit had up to five times greater HREE content (0.74 ppm Dy) in August than the same plants did in June (0.14 ppm Dy). The elevated REE content and distinct contrast to background demonstrate that biogeochemical sampling is an effective method for REE exploration in this environment.              iii  Preface This thesis is an original intellectual product by the author, B. Bluemel.  Chapter 1  The geological units and ice flow indicators presented in Figure 1.1 and 1.4b are publically available from the Swedish Geological survey, and have been grouped according to work done by Hogdahl et al (2004). The geological data presented in Table 1.1 and Figure 1.3 were provided by Tasman Metals Ltd. and are used with permission. The highest coastline map, Figure 1.4a,  was provided by M. Johnson, University of Gothenburg, personal communication. Ages of the Norra K?rr Alkaline Complex were provided by Axel Sj?qvist, as yet unpublished data.  Chapter 3 A version of this chapter has been published as Biogeochemical expression of rare earth element and zirconium mineralization at Norra K?rr, southern Sweden (Bluemel, B. et al. 2013) in the Journal of Geochemical Exploration.                   iv  Table of Contents Abstract ............................................................................................................................................... ii Preface ................................................................................................................................................ iii Table of Contents ............................................................................................................................. iv List of Tables ................................................................................................................................... viii List of Figures ...................................................................................................................................... x List of Equations ............................................................................................................................. xvi Acknowledgments ........................................................................................................................ xvii  Chapter 1 ? Overview .....................................................................................................................................1 1.1   Rationale for Study ................................................................................................................................................. 1 1.2   Biogeochemistry of Rare Earth Elements .................................................................................................. 2 1.3   Physiographic Setting ........................................................................................................................................... 3  1.3.1 Regional Geology .......................................................................................................................................... 3  1.3.2 Detailed Geology of Norra K?rr ............................................................................................................ 5  1.3.3 Glacial History of Area ............................................................................................................................... 8  1.3.4 Anthropogenic Disturbance ................................................................................................................ 10 1.4   Classification of Heavy, Middle, and Light REE .................................................................................... 13 1.5   Eudialyte ................................................................................................................................................................... 13  1.5.1 Mineral Chemistry of Eudialyte ......................................................................................................... 13  1.5.2 Weathering of Eudialyte ........................................................................................................................ 13 1.6   Solubility and Mobility of REEs in Surface Water ............................................................................... 14 1.7   Thesis Structure .................................................................................................................................................... 16  Chapter 2 - Materials and Methods .......................................................................................................... 17 2.1   Sample Site and Media Selection .................................................................................................................. 17 2.2   Sample Collection ................................................................................................................................................. 17  2.2.1 June 2011 ...................................................................................................................................................... 17  2.2.2 August 2011 ................................................................................................................................................. 18  2.2.2 August 2012 ................................................................................................................................................. 18 2.3   Sample Prep ............................................................................................................................................................ 19 2.4   Sample Analysis .................................................................................................................................................... 20  2.4.1 Vegetation Tissue and Soil Samples by Aqua Regia (Vancouver, Canada) ................. 20 v   2.4.2 Soils by Sequential Leach (Vancouver, Canada) ....................................................................... 20  2.4.3 Vegetation Tissue by Microwave Digestion (Lulea, Sweden) ............................................ 21  2.4.4 Whole-rock Analysis of Representative Rock Types by Tasman Metals Inc            (Lule?, Sweden)  .................................................................................................................................................... 21  2.4.5 Whole-rock Analysis of Bedrock Samples Below Vertical Profile Sites      (Vancouver, Canada) .......................................................................................................................................... 21  2.4.6 Scanning Electron Microprobe (UBC, Vancouver, Canada) ................................................ 21  2.4.7 Water Chemistry in Lulea, Sweden (elements and anions) ................................................ 21  2.4.8 Regional Soils (Lulea, Sweden) .......................................................................................................... 22 2.5   Quality Control ...................................................................................................................................................... 22 2.6   Detection Limit Reliability............................................................................................................................... 23  Chapter 3 - Variation in Rare Earth Element Content in the Surficial Environment Above the Norra K?rr Alkaline Complex ...................................................................................................... 26 3.1 Lithogeochemistry ................................................................................................................................................. 27  3.1.1 The Negative Eu anomaly ..................................................................................................................... 27  3.1.2 Chondrite Normalization ...................................................................................................................... 28  3.1.3 Classification ............................................................................................................................................... 29 3.2 Soil Chemistry .......................................................................................................................................................... 33  3.2.1 General Soil Chemistry ........................................................................................................................... 33  3.2.2 Sequential Leach........................................................................................................................................ 36  3.2.3 Sequential Leach Discussion ............................................................................................................... 47 3.3 Fern Chemistry ........................................................................................................................................................ 48  3.3.1 Root .................................................................................................................................................................. 50  3.3.2 Stem ................................................................................................................................................................. 50  3.3.3 Leaf ................................................................................................................................................................... 51  3.3.4 Fern Chemistry and Sequential Leach Data ................................................................................ 51  3.3.5  Summary ...................................................................................................................................................... 52 3.4 Conclusions................................................................................................................................................................ 52  Chapter 4 - Seasonal and Species Variation .......................................................................................... 54 4.1 Species Variation .................................................................................................................................................... 54 4.1.1 REE Variation in All Species .......................................................................................................................... 54  4.1.2 Essential Nutrient Variation in All Species .................................................................................. 55 vi   4.2.1 REE Variation in Fern Species ............................................................................................................ 59 4.3 Seasonal Variation ................................................................................................................................................. 59  4.3.1 REE Variation .............................................................................................................................................. 59  4.3.2 Essential Nutrient Variation ................................................................................................................ 61 4.4 Implications for Exploration ............................................................................................................................ 62  Chapter 5 - Biogeochemical Distribution of Rare Earth Elements  in the Surficial Environment at the Norra K?rr Alkaline Complex ................................................................ 63 5.1 Regional REE Signatures .................................................................................................................................... 63  5.1.1 REE Content in the Fern Species Dryopteris filix-mas and Athyrium                                    filix-femina  ............................................................................................................................................................... 63  5.1.2 REE Content in Regional Soils ............................................................................................................ 65  5.1.3 - REE Content in Regional Surface Waters................................................................................... 68 5.2 Local REE Signature at the Norra K?rr Alkaline Complex................................................................. 71  5.2.1 Sample Site and Media Selection ...................................................................................................... 71  5.2.2 Element Correlations .............................................................................................................................. 71  5.2.3 Levelling......................................................................................................................................................... 72 5.3 Discussion .................................................................................................................................................................. 75  5.3.1 ? Fern Discussion ...................................................................................................................................... 75  5.3.2 ? Fern and Soil Discussion.................................................................................................................... 77  5.3.3 Exploration Implications....................................................................................................................... 80  5.3.4 ? Spatial Distribution of REEs in Ferns Above Norra K?rr .................................................. 82 5.4 Conclusions................................................................................................................................................................ 85  Chapter 6 - Summary, Exploration Implications, and Recommendations for                          Future Work ..................................................................................................................................... 86 6.1 Summary ..................................................................................................................................................................... 86 6.2 Exploration Implications .................................................................................................................................... 89 6.3 Handy Hints for the Burgeoning Biogeochemist.................................................................................... 90  6.3.1  Sample Station Selection ...................................................................................................................... 90  6.3.2  Species Selection ...................................................................................................................................... 90  6.3.3  Sample Collection .................................................................................................................................... 90 6.4 Recommendations for Future Work ............................................................................................................. 91 References .................................................................................................................................................... ..93 vii  Appendices ..................................................................................................................................................... 97 A1 ? Technical memo: data reliability below certified detection limits ............................................ 97 A2 ?REE content in ferns, both laboratory certified and ?below reliable  detection limit? .................................................................................................................................................. 106 A3 ? Blind Quality Assurance Controls ............................................................................................................ 114 A4 ?Ash weight to dry weight conversions ................................................................................................... 142 A5 ? Technical Memo: Element mapping at the Canadian Light Source Synchrotron ............ 160 A6 ? Field Duplicate quality control diagrams............................................................................................. 161 A7 ? Soil Reference Images .................................................................................................................................... 163 A8 ? Laboratory Certificates of geochemical data ..................................................................................... 164                          viii  List of Tables Chapter 1  Table 1.1 - Description of the main rock types in the Norra K?rr Alkaline Complex???????. 6 Chapter 2 Table 2.1 - Plant species collected in June, 2011, listed by common and Latin names, with n values and sample details ......................................................................................................................................... 18  Table 2.2 ? Precision data for blind standards ....................................................................................................... 26 Table 2.3 - Certified detection limits from Acme Analytical Laboratories, and lowest practical reliable detection limits for this thesis research. The LREEs were all above the certified detection limit, so those have been marked ?not applicable? (n/a). All values are in ppm? .. 24 Chapter 3 Table 3.1 ? Materials collected from the three vertical profile sites .......................................................... 26 Table 3.2 ? Commonly used chondrite normalization values (Nakamura, 1974; Boynton 1984; Taylor and McLennan, 1985; Wakita et al., 1971). All values are in ppm ........................................... 29 Table 3.3 - Summary of the REE content, as well as the median ?REE value, for the Ah and B soil horizons ................................................................................................................................................................................. 33 Table 3.4 - Physical descriptions of soil samples collected from each horizon at three vertical test pits ................................................................................................................................................................................... 35 Table 3.5 ? REE content in leaf, root, and stem tissue at each vertical profile site ............................. 50 Table 3.6 ? REE transfer factors from root to stem for all three vertical sample sites ..................... 52 Chapter 4 Table 4.1 - Common and Latin names of species collected from Norra K?rr, including number of each species and tissue sampled .............................................................................................................................. 54 Table 4.2 -  Average macronutrient and micronutrient content in each species, compared to standard tissue levels of essential elements required by most plants^ (Taiz and Zeiger, 2010). *All samples are leafy tissue with the exception of Norway Spruce, which is twig tissue, which has been reduced to ash and back calculated to represent dry tissue element concentration.56 Table 4.3 - REE seasonal variation in Dryopteris filix-mas ............................................................................... 59 Chapter 5 Table 5.1 - Statistical ranges of REE content in regional fern samples, grouped by species. The cells highlighted red are below reportable detection limit. Ferns analysed at ALS in Lule?, using a microwave-aided nitric acid digestion with an ICP-MS instrumental finish. The minimum and maximum values obtained from analyses of regional fern material can be used to approximate the range of biogeochemical ?background? in the surficial environment above unmineralized country rock. The background range is displayed graphically in Figure 5.6, and ix  the minimum and maximum REE values are estimates of the regional background REE content.................................................................................................................................................................................... 65 Table 5.2 - Ranges of REE content and Y (for each element) in regional soils, grouped by horizon. *Average Granite composition from Krauskopf 1967, average V?xj? granite composition (?V?xj??) from Appelquist 2010, and average Trans-Scandinavian Igneous  Belt (?TIB?) composition from Christensson 2013 ................................................................................................... 66 Table 5.3 - Ranges of REE content (for each element) in regional water, grouped by water colour and filtration........................................................................................................................................................................ 68                          x  List of Figures Chapter 1 Figure 1.1 - Geological map of southern Sweden, domains and shear zone modified from H?gdahl et al. (2004), geological polygons publically available for download from the Swedish Geological survey. The yellow star indicates the location of the NKAC ................................................... 4 Figure 1.2 - Photo of the Norra K?rr Alkaline Complex ?discovery outcrop?, with rock hammer (approximately 1m length) for scale. The pink phenocrysts are the REE-bearing zirconosilicate mineral eudialyte ................................................................................................................................. 6 Figure 1.3 - Geological Map of the Norra K?rr Alkaline Complex, after Leijd (2012) unpublished data............................................................................................................................................................................................... 7 Figure 1.4a - Map of southern Sweden with isolines representing ?highest coastline? elevation. The area around Gr?nna is above the highest coastline, so the likelihood that sediments there have been reworked by wave action is low. Norra K?rr is marked with a yellow star. (adapted from M. Johnson, associate professor of Quaternary geology, University of Gothenburg, personal communication) ............................................................................................................................................... 8 Figure 1.4b ? Geological map of Norra K?rr with ice flow direction .............................................................. 9 Figure 1.5 ? Map of surficial materials in the area around the Norra K?rr Alkaline Complex, adapted from the Swedish Geological Survey. The NKAC is marked with a yellow star on the large map, and shaded purple on the inset map................................................................................................ 11 Figure 1.6 ? Typical soil profile at Norra K?rr. Note the lack of definition between soil horizons. Image has been colour enhanced for clarity ..................................................................................................... ...12 Figure 1.7 - Airphoto of the Norra K?rr area with sample stations marked as yellow stars, and the deposit outlined in pink ......................................................................................................................................... 12 Figure 1.8 ? Depiction of eudialyte mineral group (after Johnsen et al. (2003)). The ?M? denotes cation replacement sites, and the ?X? is an anion replacement site. The chemical formula of eudialyte is (Na, Ca, REE)5(Fe2+,Mn)(Zr,Ti)[(Si3O9)2]?(OH,Cl,F) ................................................................ 14 Chapter 2 Figure 2.1 - Photo of fern frond, before being pressed and dried for transport back to Canada .. 19 Figure 2.2 ? Neodymium precision diagram in field duplicates .................................................................... 22 Figure 2.3 - Methodology example of detection limit reliability. Figure 2.3b is all Dryopteris filix-mas data, with BDL (<0.02ppm) values replaced with 0.01ppm. Figure 2.3a shows the same data, but with the BDL values left as raw ?uncertified? values. Colour scheme is based on REE (in this case, Tm) as a function of Y content. All REEs show good correlation with Y content, displayed here along the line Y = X. Data just below the certified DL (0.02ppm) is coloured green. High confidence that these samples are reliable. Samples with lower REE content, but in the same population (populations are determined based on probability plots; data plotting on a straight line without breaks, in Figure 2.3a, are deemed one population) are coloured orange, with moderate confidence that these data are reliable. Data coloured red are have low xi  REE content, or are part of a different population. Confidence that these data are reliable is very low .................................................................................................................................................................................. 25 Chapter 3 Figure 3.1 - Locations of three vertical profile sample sites over the Norra K?rr Alkaline Complex. Sites were chosen based on high REE content in ferns, which was determined  in the previous year?s study ................................................................................................................................................ 27  Figure 3.2a-c  -- Photos of rock samples collected from base of vertical profile sites. (a) NK12-072: fine grained, green-grey aegerine rich nepheline syenite. (b) NK12-039 foliated nepheline syenite with phenocrysts of catapleiite and eudialyte occurring within bands of microcline and amphibole. (c) NK12-060 strongly foliated biotite rich granite with red K-feldspar porphyroblasts ................................................................................................................................................. 30 Figure 3.3 ? REE Chondrite normalized plot (after Boynton 1984) showing a compilation of      all rock types at Norra K?rr (yellow circles, analysed by Tasman Metals Inc.) and the rocks collected for this thesis (grey squares). Note the difference in magnitude between the  granitic ?base of profile? angular fragment collected from NK12-060 compared to all other rock types .............................................................................................................................................................................. 31 Figure 3.4 ? Total alkali silica (TAS) classification of lithogeochemical samples collected for this thesis: three base of profile rock samples [NK12-072 (open circle), NK12-039 (open square), and NK12-060 (open triangle)]. Grey circles are averages of each major rock type present in the Norra K?rr Alkaline Complex (analyses provided by Tasman Metals Inc.)  ............................... 32 Figure 3.5 ? REE Chondrite normalized plot (after Boynton 1984) of soil samples collected in June 2011 (7 Ah horizon, green circles; and 7 ?B? horizon, blue circles) with representative lithogeochemical samples of all rock types in the Norra K?rr complex (red circles, analysed  by Tasman Metals Inc.) ................................................................................................................................................... 34 Figure  3.6 ? Photo of vertical test pit at site NK12-060. Note the lack of definition between   soils horizons, particularly the A and B horizons. Depth to C horizon is 70cm  ............................... 35  Figure 3.7 ? Chondrite normalized plot of all sequential leaches at each site, including the sum total of all the leaches (burgundy points) which approximate the total REE content of the surficial materials .............................................................................................................................................................. 38 Figure 3.8a - Sequential leach, four acid, and summed leach data of surficial materials and bedrock from site NK12-039, normalized to global chondrite values (Boynton, 1984). Note the positive Ce anomaly, LREE enrichment relative to HREEs, and negative Eu anomaly ......... 40 Figure 3.8b ? Sequential leach, four acid, and summed leach data of surficial materials and bedrock from site NK12-039, normalized to local bedrock. Note the same positive Ce anomaly and relative LREE enrichment, but also a weakly positive Eu anomaly in the 4acid near total ..................................................................................................................................................................................................... 40 Figure 3.8c ? Histograms of soil-forming elements, grouped by vertical profile site ......................... 41 xii  Figure 3.9a ? Sequential leach, four acid, and summed leach data of surficial materials and bedrock from site NK12-072, normalized to global chondrite values (Boynton, 1984). Note the strong negative Eu anomaly in all materials, including bedrock, and the positive Ce anomaly only in the hydroxylamine leaches (brown lines) of the B horizon. Also note the relative depletion of HREEs relative to LREEs ................................................................................................... 43 Figure 3.9b ? Sequential leach, four acid, and summed leach data of surficial materials and bedrock from site NK12-072, normalized to local bedrock. Note the ?flatness? of all leaches when normalized to local bedrock, with the exception the hydroxylamine leaches on the A horizon soil. Also note that the positive Ce anomaly in the hydroxylamine leached from  the B horizon persists ............................................................................................................................................................ 43 Figure 3.10a ? Sequential leach, four acid, and summed leach data of surficial materials and bedrock from site NK12-060, normalized to global chondrite values (Boynton, 1984). Note the appearance of a positive Eu anomaly in the samples derived from the granitic source ...... 45 Figure 3.10b ? Sequential leach, four acid, and summed leach data of surficial materials and bedrock from site NK12-060, normalized to local bedrock. Note the appearance of a positive Eu anomaly in the samples derived from the granitic source .................................................................... 45 Figure 3.10c - Histograms of La and Ce (LREEs), Eu (MREE), and Lu (HREE), grouped by vertical profile site. Note the proportion of REEs in the B horizon increases with increasing atomic number, meaning that the HREEs are more enriched (relative to the LREEs) in the B horizon than in the A horizon ....................................................................................................................................................... 46 Figure 3.11 ? Example photo of a HFSE grain on a silicate soil particle. The HFSE grain is approximately 25?m long (medium silt sized) .................................................................................................. 46 Figure 3.12 ? Differences in Fe content in the Ah and B horizon soil samples collected from the same location ....................................................................................................................................................................... 47 Figure 3.13 ? Chondrite normalized plots of leaf, root, and stem tissue from all three vertical profile sites. Note the decrease in REE content from stem < leaf < root............................................... 49 Figure 3.14 ? Chondrite normalized plots of leaf, root, and stem tissue from ferns at all three vertical profile site, with lithogeochemical data added for reference ................................................... 49 Chapter 4 Figure 4.1 ? Box and whisker plots of K (left) and Ca (right) for all sampled species. The hazelnut has much higher Ca content than any other plant, and all 3 fern species have  more K than Ca .......................................................................................................................................................................... 57 Figure 4.2 - ?REEs as a function of potassium for all species, with Norway recalculated to a dry weight basis .......................................................................................................................................................................... 58 Figure 4.3 - Chondrite normalized plot of three fern species Dryopteris filix-mas, Athyrium filix-femina, and Pteridium aquilinum. Note relative enrichment of LREEs in Athyrium filix-femina ..................................................................................................................................................................................................... 60 Figure 4.4 - REE content of Dryopteris filix-mas samples collected from the same location in June and August, coloured by season ................................................................................................................................. 60 xiii  Chapter 5 Figure 5.1 - Geological maps of the Norra K?rr area (the Alkaline Complex is marked with a yellow star), with ice flow indicators (black arrows) and regional sample station locations (green circles).  The local samples (yellow circles, Figure 1.5b, cf Section 1.3.3) cover 1.5km2, the regional samples (green circles) cover almost 750km2, and the surface trace of the deposit is 0.4km2  ................................................................................................................................................................................ 64 Figure 5.2 -  Regional soil data from C and B horizons (light and dark blue, respectively) normalized to average granite (Krauskopf, 1967). Yellow points are averaged Trans-Scandinavian Igneous Belt (TIB) samples (Appelquist, 2010) and red points are averaged V?xj? granite samples (Christensson, 2013). Dark grey points are average representative Norra K?rr mineralized nepheline syenites ........................................................................................................ 67 Figure 5.3a - Chondrite normalized REE diagram of filtered regional surface water samples. Note the strong negative Eu and Ce anomaly ..................................................................................................... 69 Figure 5.3b ? Box and whisker plots of La and Dy (representative of the LREEs and HREEs, respectively) content in filtered regional surface water samples. The highest REE content is in the brown-yellow water samples, and is attributable to the higher concentration of dissolved organic acids in these samples ................................................................................................................................... 69 Figure 5.4 - Chondrite normalized plot of surficial materials from site NK12-060, with local unfiltered lake water data. The water sample with the higher REE content was from site TAS12-43 (Fig. 5.1), which was collected from a small lake directly above the deposit, and the water sample with the lower REE content (TAS12-02) was collected from a lake 1km NE of the deposit ............................................................................................................................................................................. 70 Figure 5.5 - Two graphical representations, a histogram (above) and a ?box and whisker? plot (below) of the same data set. The data is normally distributed, with some outliers (red box) and far outliers (green box).   The box and whisker plot divides the data into four equal parts by finding the median, and then the 25th and 75th percentiles (which approximate the inter-quartile range, or IQR). The median is represented by the white line bisecting the box; the mean is represented by the white circle, in this case located just above the median. Outliers are represented by open black circles, and far outliers are represented by open   black triangles ..................................................................................................................................................................... 74 Figure 5.6 - Athyrium filix-femina (left) and Dryopteris filix-mas (right) samples collected from the surficial environment above Norra K?rr, normalized to average granite (Krauskopf, 1967). The purple shaded area represents the range of REE content in ferns sampled over unmineralized country rock. This range is assumed to represent ?background? and is quantified in Table 5.2 .................................................................................................................................................... 76 Figure 5.7 ? Yttrium content (as a surrogate for HREE) in three fern species over the deposit (left) and in the background (right). Athyrium filix-femina has higher absolute Y content over the deposit than the other two species, but the contrast between samples collected over mineralization compared to samples collected over background is much more pronounced in Dryopteris filix-mas, therefore Dryopteris filix-mas is the preferable exploration sampling medium ................................................................................................................................................................................... 77 xiv  Figure 5.8 - La, Eu, Dy, and Yb content in Athyrium filix-femina and Dryopteris filix-mas, grouped by species and coloured regionally (blues) and locally (reds) .................................................................. 78 Figure 5.9 - La (above) and Dy (below) content in surficial media, levelled by analytical method. Since the data have been levelled, the Y axis values  are standard deviations, not absolute concentrations. Sample count is displayed  over each box. In the upper diagram, regional and fern data are presented,  along with all soil data, and the variance between the four populations is similar  based on the mean and median values. In the lower diagram all regional and local soil data, from each horizon, is  presented again, but only the Dryopteris filix-mas data are displayed (bottom right) and the contrast between regional (light green) and local (dark green) fern samples is much more pronounced, particularly for Dy,  which is a proxy for the HREEs ........................................................................................................................................ 79 Figure 5.10 (a-c)     La (above) and Dy content (below), levelled by analytical method, in surficial materials collected regionally (lighter shades) and locally (darker shades) around the Norra K?rr Alkaline Complex. Dryopteris filix-mas samples both above the deposit (dark green) and in the background (light green) are  graphed against each soil horizon (A, B, and C). Compared to the A horizon soils,  Dryopteris  filix-mas  displays better contrast between the mineralized deposit and the surrounding country rock  for both the LREEs and the HREES. Dryopteris filix-mas and B horizon show comparable contrast for the HREEs,  and B horizon soils show more pronounced contrast for the LREEs. C horizon tills show a false anomaly over the regional country rock for the HREEs, and no clear contrast for the LREEs . 81 Figure 5.11 - Sm content in ferns above the Norra K?rr Alkaline Complex ............................................. 83 Figure 5.12 - Dy content in ferns above the Norra K?rr Alkaline Complex.............................................. 84 Chapter 6 Figure  6.1 ? Genetic model of the formation of the surficial environment above the Norra K?rr Alkaline Complex. The glacial material is derived from V?xj? granites in the up-ice   direction ................................................................................................................................................................................. 87 Figure 6.2 ? Schematic diagram of Norra K?rr with normalized REE plot of surficial materials. The REE plot is normalized to local bedrock from site NK12-060, not chondrite. Green squares are local bedrock, filled red circles are surficial materials from the A and B horizons, filled red triangles are glacial till, filled pink squares area glacially transported granitic fragments derived from nearby country rock. Blue circles are local surface water from lakes directly overlying the NKAC; surface water REE content has been multiplied by a  factor of 1000 for convenience of viewing. The shaded brown region represents the range of REE content of typical Swedish podzols (n = 30) collected in Scania, approximately 200km south of Norra K?rr, and analysed by microwave-aided nitric acid digestion with and ICP-MS finish (Tyler and Olsson, 2002) which is a slightly weaker strength of digestion than the sum of the sequential leach data provided by Acme Analytical (red shapes). Not the positive Eu anomaly in the granitic fragment and the glacial till, which is not visible in the local bedrock of the surface water samples .................................................................................................................................................... 88 Figure 6.3 - Diagram of a fern plant with underlying soil profile and corresponding REE profiles.  REEs have been normalized to local bedrock at site NK12-060. Note the decrease in REE xv  content in the fern plant from root > leaf > stem, and the fractionation within the plant of LREEs over HREEs. Soil samples, till samples, and water samples are shown for   comparison ........................................................................................................................................................................... 89                               xvi  List of Equations Chapter 3 Equation 3.1 ? Quantified Eu anomaly calculation after Boynton (1984) ............................................... 27 Chapter 4 Equation 4.1 - Calculation of elemental concentration in dry weight of plant material  based on elemental concentrations in ashed material and ash yield (Dunn, 2007) ....................... 55                    xvii  Acknowledgments  This research would not have been possible without encouragement and support from many different people. 	irst and foremost, I?d like to thank my excellent supervisors Colin Dunn and Craig art for all of their input and guidance, both academically and ?in life?, over these past two years. I was extremely fortunate to be surrounded by some of the best and brightest geochemical brains available; Pim, Fred, and Tans ? our endless discussions were invaluable to me, and I can?t thank you enough. This project was generously sponsored by Tasman Metals Ltd. and ioGlobal Solutions Ltd., both of whom were unfailing in their support over the years. Particular thanks to Mark Saxon, Magnus Leijd, and Henning Holmstr?m for field support in Sweden. Axel Sj?qvist: first of all, thanks for all the chess and laughs over there :) and secondly, thanks for the dates and ALL the geo-chats since then. I beat you to the publication bru, but not by much! This whole ?masters? thing wouldn?t have been half as fun without the distractions willingly, and sometimes even forcibly, provided by my friends. Bev Quist, thank you for coming to visit me in Scandinavia, you have no idea how much it meant to me that you made the trip, and that was one of our best adventures yet! A very special thanks to Leif Bailey a nd Leanne Smar (you?re the best besties anyone could hope for), and to my UBC family Brian McNulty, Erin Looby, Mike Tucker, and Irene del Real. I couldn?t have done it without you guys. Finally, thank you to my family: Patty, Ryan, Leah, Michelle, and Dom. Getting back on island time with you helped keep me centered and sane through it all, thank you and I love you.               Chapter 1 - Overview  Rare earth elements (REEs) have gained significant economic importance in the past 20 years due to their restricted supply and unique properties which make them essential for high tech and sustainable applications, such as smart phones, solar panels, wind turbines, and hybrid cars (Hatch, 2012). Sweden is rightly known as the home of REEs, as they were first discovered  in a quarry in the village of Ytterby, near Stockholm in 1794. The Norra K?rr Alkaline Complex is located in southern Sweden, and is enriched in heavy rare earth elements (HREEs).   Rationale for Study  Biogeochemical exploration is an effective but underutilized method for discovering and delineating buried mineralization. Plants are capable of accumulating REEs in their tissue, and ferns (pteridophytes) are especially adept because they are one of the most primitive land plants. Less evolved plants can accumulate higher concentrations of harmful elements, such as heavy metals or REEs, in their tissue without adverse effects on their growth and physiography than more highly evolved land plants (Dunn, 2007). This behaviour stems from the adaptation of barrier mechanisms within the plant to protect its cells from potentially toxic levels of certain elements (Kovalevsky, 1987). The Norra K?rr Alkaline Complex was discovered in 1906 by an assistant geologist at the Swedish Geological Survey (SGU) who found a small outcrop (Figure 1.2), collected a hand sample, and sent it to Professor A. E. T?rnebohm, head of the SGU. T?rnebohm made thin sections of the rock and determined it was nepheline bearing, and that it contained the REE minerals eudialyte and catapleiite (which, at the time, was spelled ?katapleite?).  Many exploration methods typically  used for discovering mineral deposits are ineffective at Norra K?rr. The nepheline syenite has similar density to the surrounding granitic country rock, and lacks sulphide minerals, hereby making it indistinguishable by gravity and induced polarity (IP) geophysical surveys, respectively.  The deposit has unusually low uranium and thorium content for a deposit of its type (S?rensen, 1997 ), and therefore does not appear in the country-wide Swedish Geological Survey radiometric surveys. Extensive glaciation of the study area presents additional challenges because conventional soil surveys are unlikely to be a reliable exploration method in areas of transported cover. Plants are an advantageous exploration medium in areas where the soil profiles are disturbed or poorly defined, because plant roots can interact with several soil horizons and yield an integrated, more representative signature. The deposit is overlain by transported Quaternary sediments and Swedish boreal forest, so biogeochemical exploration methods, a little-tested technique for exploring for REE mineralization, were evaluated  at the Norra K?rr Alkaline Complex.  This locality was selected to confirm the efficacy of biogeochemical exploration for REE mineralization because such a deposit would be challenging to detect by conventional exploration techniques.   1  As the rate of discovery of near-surface mineral deposits declines, all available tools must be utilized to successfully explore for buried mineralization. The mobility of REEs in the surficial environment, a heretofore under-appreciated phenomenon, allows biogeochemistry to be used as an effective exploration tool for buried REE deposits, such as the Norra K?rr Alkaline Complex.  1.1  Biogeochemistry of Rare Earth Elements Rare earth elements occur in measurable quantities in the tissue of various plant species (Laul et al. (1979); Markert and Li (1991); Wyttenbach et al. (1998); Tyler (2004)), but historically little attention has been paid to the biogeochemistry of REEs as a group because they were considered neither essential nor toxic.  In the past 20 years, researchers have shown that REEs in phosphate and nitrate fertilizers can be beneficial to plant growth  (Welch (1995); He et al. (1998); Pang et al. (2002)) and subsequently REEs have been added to agricultural fertilizers to increase crop yield (Guo et al., 1996). This practice has been most widely embraced in China, and as a result many researchers there have begun studying REEs in soil-plant systems to better understand the distribution and accumulation of REEs in the natural environment. Research has been carried out mainly on rice and corn (Li et al., 1998) as well as soybeans and wheat (Ding et al., 2007; Ding et al., 2006).  This pattern of decreasing concentrations conforms with that of many trace elements (Tyler, 2004).  The REE content in the fern species Matteuccia increases from stem to leaf to root, and the distribution patterns indicate that REEs in the fern were sourced from the silicate fraction of the soil and had once been in the dissolved state (Fu et al., 1998).  There can be a positive correlation in the distribution of REEs between the soil and the plant (Takada et al. (1996) Wang et al. (1997) Wyttenbach et al. (1998)). The REE transfer factor from soils to plants, i.e. the REE content ratio of plant/soil, in a natural background forest ecosystem ranges from 0.04 and 0.085, and the transfer ratio values were roughly equivalent for all the REEs, which indicates there was no significant fractionation between the light REEs (LREEs) and the heavy REEs in the studied plant species (Markert and Li (1991)).  A transfer factor of  <1 implies that there is incomplete uptake of REEs by the plant, which could be a function of REE bioavailability in the soil, and/or the minerals in the soil that host the REEs.   Biogeochemical surveys provide a viable approach to mineral exploration in most terrains because they are rapid, relatively inexpensive, and effective in areas of transported cover.  Literature reviews assisted in choosing a sample medium that could adequately accumulate REEs in its tissue and was also widespread across the Norra K?rr deposit and background areas. Findings from several authors (Dunn (1998); Fu et al. (1998); Wyttenbach et al. (1998); Zhang et al. (2002); Dunn (2007)) indicate that ferns can be a suitable sample medium for REE exploration.     2  1.2  Physiographic Setting The Norra K?rr Alkaline Complex is located approximately 300 km southwest of Stockholm and 10 km northeast of Gr?nna in southern Sweden (Figure 1.1). Norra K?rr, which is Swedish for ?northern bog?, is situated in a shallow topographic low and is slightly elevated at the centre of the intrusive complex. There is a low-lying swampy area at the western contact of the intrusion, and at the eastern edge there is a small creek draining into a lake which sits directly above the eastern contact;  the complex intrudes Paleoproterozoic V?xj?-type granites, which are ?even-grained and generally felsic? (H?gdahl et al., 2004). The Norra K?rr deposit outcrops in several places, but for the most part is overlain by 1-4m of Quaternary glacially transported sediments (see Section 1.3.3).  1.3.1 Regional Geology The Norra K?rr Alkaline Complex sits within the Paleoproterozoic Sm?land-Varmland belt (SVB). Locally, the SVB is referred to as V?xj?-type granite because the type locality is near the municipality of V?xj?, 170 km south and east of Norra K?rr. The SVB comprises a group of igneous plutons attributed to the first orogenic event responsible for the formation of the Trans-Scandinavian Igneous Belt  (TIB) (H?gdahl et al., 2004). The SVB is mainly composed of monzodioritic, monzonitic, quartz-monzonitic, granitic plutonic rocks and, particularly in Sm?land, their associated felsic volcanic rocks (Andersson et al. (2004); Johansson (1988) and references therein).  The genesis of the TIB is the subject of ongoing debate because the ages of the plutons within the region range from 1.86-1.65 Ga. As well as being a long lived orogenic event, the interpretation is complicated by multiple phases of deformation both during and after emplacement. Andersson and ikstrom (2004) ?suggest a more or less continuous magma generation and emplacement of the same type over a long time, where the earliest c. 1.85Ga rocks mark the beginning of the shift from collisional to post-collisional (extensional) regimes?.  Emplacement of the Norra K?rr Alkaline Complex is associated with a major NNW trending shear zone extending over 600 km through southern Sweden and into central Norway. This shear zone is also responsible for the formation of Lake V?ttern (Figure 1.1), which is the second largest lake in Sweden, the 6th largest lake in Europe, and sits less than 2 km west of the Norra K?rr deposit. It has been suggested that this massive structural corridor was the ?primary deep crustal / mantle tapping conduit for emplacement of the Norra K?rr nepheline syenite intrusive body? (Rankin (2011). Blaxland (1977) determined the complex was 1580?62 Ma using a whole-rock Rb/Sr isochron, but the current ?official? age of the complex, 1545??1 Ma, was recalculated using radiometric constraints by Welin (1980). Recent U/Pb isotope research using LA- ICP-MS on zircons from the complex suggests that the existing age is slightly younger, but still within the earlier-published error limits (Sj?qvist (2012), unpublished data).  3StockholmGothenburg100kmNorra K?rrAlkaline ComplexGr?nnaJ?nk?pingV?xj?Lake V?ttern50km Sveconorwegian Domain  (1.7 - 0.9Ga) Phanerozoic rocks Svecofennian Domain  (1.95 - 1.86Ga) Caledonide Mountain Range  (0.55 - 0.4Ga) Transscandinavian Igneous  Belt   (1.85 - 1.65Ga) Norra Karr Alkaline Complex Major shear zone Waterbodies TownsLegendGothenburgN 17?E17?E16?E16?E15?E15?E14?E14?E13?E13?E12?E12?E58?N58?N57?N57?NFigure 1.1 - Geological map of southern Sweden, domains and shear zone modi?ied from Hogdahl et al. 2004, geological polygons publically available for download from the Swedish Geological survey. The yellow star indicatesthe location of the Norra K?rr Alkaline Complex.4  1.3.2 Detailed Geology of Norra K?rr  The Norra K?rr deposit is a Mesoproterozoic intrusive suite wholly surrounded by Paleoproterozoic V?xj?-type granites. The elliptical complex is roughly 1300 m long, 480 m wide, elongated in the N-S direction, and dipping approximately 45? to the west.  The contact between the nepheline syenite and the surrounding granitic country rock is strongly foliated by regional metamorphism, and fenitized by hydrothermal fluids related to the emplacement of the Norra K?rr deposit. Fenitization is type of sodic hydrothermal alteration often associated with carbonatites (Adamsson, 1944). Norra K?rr may have been emplaced as a sill within the V?xj? granites (Rankin, 2011), which were concurrently foliated and fenitized during the time of emplacement. As a result of this, the contact between the V?xj? granites and the nepheline syenite is difficult to distinguish due to similar textures resulting from their shared hydrothermal and metamorphic history.  The dominant lithology of the Norra K?rr deposit is peralkaline nepheline syenite. Peralkaline rocks are deficient in Al such that Na and K are in excess of that required for feldspar, so nepheline, or similarly silica undersaturated feldspathoids, preferentially crystallize as rock forming minerals. Peralkaline nepheline syenites, such as those at Norra K?rr, have been termed ?agpaitic? (S?rensen, 1997 ) because of the presence of minerals such as eudialyte, a complex Zr-Ti-REE-(	) silicate. ?Agpaitic nepheline syenites are considered to have been formed by consolidation of melts oversaturated in alkalis, especially sodium, under conditions preventing volatiles from escaping? (S?rensen, 1997 ), and agpaitic rocks can form by extreme fractionation of alkali basaltic magma at crustal levels (Larsen and S?rensen, 1987 ). The term ?agpaitic? originated from work done on the Il?maussaq complex in Greenland, which is also where the specimen for the first detailed examination of eudialyte was collected.  The REE mineralization at Norra K?rr is mainly hosted by the zirconosilicate mineral eudialyte (Na, Ca, REE)5(Fe2+,Mn)(Zr,Ti)[(Si3O9)2?.(,Cl,	). It was named from the reek ?eu dialytos?,  meaning ?well decomposable?. The susceptibility of eudialyte to the action of hydrothermal and surface water was first remarked upon by Ussing (1912) at the Il?maussaq intrusive suite, Greenland.  Four main rock types comprise the Norra K?rr Alkaline Complex: grennaite, kaxtorpite, lakarpite, and pulaskite. All four units are type localities and named after nearby towns and local farms, with the exception of pulaskite, which was first observed in Greenland at the Il?maussaq complex. The grennaite unit contains the most eudialyte phenocrysts, and where it is coarsely pegmatitic it contains up to 70% TREO, the highest REE concentration of any of the units in the deposit. A summary of various rock types at Norra K?rr is shown in Table 1 and displayed in Figure 1.3.  1.3.2.i - Grennaite Grennaite is the dominant rock type. It is a fine-grained, grey-green nepheline syenite composed of alkali feldspar, nepheline, aegerine, eudialyte, and catapleiite. The concentration of REEs in the grennaite unit is much higher, especially where it is migmatized or pegmatitic, than any other rock types in the area.  Eudialyte occurs both as phenocrysts and as a groundmass constituent (Adamsson, 1944). The grennaite is locally schistose, and the schistosity is parallel to that of the adjoining V?xj? granites ([(T?rnebohm, 1906) in Swedish], translated by Adamsson (1944)) along the long axis of the deposit, which strikes in the same direction as the belt scale shear zone in the 5  NNW-trending structural corridor. However, at the north and south ends of the deposit (Figure 1.3) the schistosity conforms to the margins of the grennaite unit and is perpendicular to the schistosity in the granite. This implies that the schistosity of the grennaite is precrystalline, and the grennaite crystallized from a magma intruding the V?xj? granites (Adamsson, 1944).  Table 1.1 ? Description of the main rock types in the NKAC, adapted from Tasman Metals Ltd.   Figure 1.2 ? hoto of the Norra K?rr Alkaline Complex ?discovery outcrop?, with rock hammer (approximately 1m length) for scale. The pink phenocrysts are the REE-bearing zirconosilicate mineral eudialyte. 6  1.3.2.ii - Kaxtorpite Kaxtorpite is a coarse-grained, dark alkaline rock with microcline augen and an aegerine, alkali amphibole and nepheline groundmass located only in the centre of the deposit, and is effectively barren of REE mineralization. 1.3.2.iii - Lakarpite Lakarpite is ?an often medium-grained, albite-arfvedsonite-nepheline dominated rock with some microcline-rosenbuschite and minor titanite-apatite-fluorite? (Reed, 2011). Adamsson (1944) was the first to observe nepheline in this unit, as well as abundant fluorite. Lakarpite does not appear to contain eudialyte (Sj?qvist et al., 2012). 1.3.2.iv - Pulaskite Pulaskite occurs along the western side of the deposit, and is composed of albite, microcline, aegerine, alkali amphibole, minor biotite and nepheline. Rosenbuschite, apatite, titanite, and fluorite occur as accessory minerals in the pulaskite units. Typically the pulaskite and grennaite are inter-layered, and in places the grennaite appears to be brecciating the pulaskite. A zone of fenitization permeates the pulaskite and extends into the surrounding granitic country rock. This fenitized zone makes the contact between the outermost grennaite ? pulaskite and the V?xj? granites difficult to distinguish.  Figure 1.3 ? Geological Map of the Norra K?rr Alkaline Complex, after Leijd (2012) unpublished data 7  1.3.3 Glacial History of Area There were at least four glacial periods in Scandinavia: the pre-Weichselian, the first and second Weichselians, and the late Weichselian. Ice flow direction was always roughly southwards (Figure 1.4b), except the during the pre-Weichselian, when ice flowed from west to east, towards Finland, in northern Sweden (Lundqvist, 2004). The only area of permafrost in Sweden is the most northerly tip of the country, directly in contact with the Finnish border (Anders Rapp et al., 1986). The fennoscandian region (Figure 1.1) is a region of boulder depressions, and the southwestern coast, directly south of Lake Vanern, contains fossil ice wedges. Tills representing different stages of older glaciations are found below organic-bearing sediments (Lundqvist, 2004). The last glacial period was the late Weichselian, the retreat of which is marked along the southwest coast of Sweden by distinct marginal moraines. Large parts of southernmost Sweden were likely ice free by ~14,500 BP (Lundqvist and Wohlfarth, 2001). The deglaciation of southern Sweden was accompanied by isostatic rebound, such that large areas along the coast have, until relatively recently, been below sea level. This leads to the formation of ?highest coastlines? which are shown in Figure 1.4a. Depending on the position of the highest coastlines, surficial material will have been reworked to different extents according to how much time it spent below sea level. Since the Norra K?rr Alkaline Complex is approximately 200 m asl, therefore above the highest isocoastlines  (Figure 1.4a), the surficial environment may be treated equally. However, comparisons cannot be made between environments below and above the highest coastlines because they have been acted upon by very different processes.    Figure1. 4a ? Map of southern Sweden with isolines representing ?highest coastline? elevation. The area around Gr?nna is above the highest coastline, so the likelihood that sediments there have been reworked by wave action is low. Norra K?rr is marked with a yellow star. (adapted from M. Johnson, Associate Professor of Quaternary geology, University of Gothenburg) 8!!! !! !!!! !!! !!! ! ! !!!!! ! !!!!! ! !! ! !!! !!! !! !!!! ! !!! ! !! ! !! !!! ! ! !! !!! !! ! !!!!!! ! !! !! ! !!! ! !! ! !!! ! !!! !! !!! ! !!! ! !! !!!! ! !! !! ! !! ! ! !!! ! !!! !! ! !!! !!! ! !!! ! ! !!! !!! !!! !! ! !!! ! ! !!!! !!!! ! !! ! ! !! ! !! ! ! !! ! !! !! !!! !!!! !! !! !! !! ! !!!! !! ! !!! ! !! ! !!! !!!!! !!!!! ! ! !! ! !! !!!! !!! !!! ! !!! !!!! !! !!! !! ! !! !! !! !! !! ! !! !! !! !! ! !!! ! !! !!!!! !! !!!!!!! !! !!!!! ! !! !!! ! ! !! !! !! !!!! !! !! !! !!!! ! !! !!!! !!!! !! !! !!!! !! ! !! !!!! !! ! ! !! ! ! !! ! !!!! !!! !!! !! !! ! !!!! !! !! ! !!!! ! !!! !!! !!! !!!! !! !! !!! !! !! !!! !!!! ! !! !! !!! ! !!! ! ! !!!! !! !! !! !!!! !! ! !!!! !!! !! ! !!!! !! !!! ! !!! ! !!! !!! !! !! !!!! !! ! !!! ! !! !! !!!!!! !! !!! !!!!! ! !!! ! ! ! !! ! ! ! ! !! !! ! !!! ! !! !! ! !!!! !! ! ! ! !!! !!! !!!! !! !!! ! !! !! !!! !! !! !! !! !!!!!! !! ! !!!! !! !!! !! ! !!!! !! ! ! !!! !! !! !! !!! !! !!! ! ! ! !!!! !!! !!!!! !! !!! !!! ! !! !! ! !! !! !!! !! !! ! !!! !! ! !! ! !! !! !!!!! !! ! !! !!!!! !! !!! !!!! !! ! ! !!!!! !!! !! ! ! !! !!!! !!! ! !! ! ! !!!! !!! !! ! !! ! !! !! ! !! !!! !!!! !!! !!! ! ! !! ! !! !! ! !!!!!! !! ! ! !! ! !! !! !! !! !! !!! ! !!! !! ! !! !! ! ! !! !!! !!! !! !! !!! !! !! ! !!!! !! !!! !! ! ! !! !!!! !!! !! ! !! !! !!! !!! ! !!!!! ! !! !!! ! !!! !! !! !!! ! !!!! !! !! !!!!! !!!!!!! !! !!! !! ! ! !!!! ! !!!! ! !!!! ! !!! ! !! !!! !! !! !!!!!!!!!!! !!!!!! ! !!! !! !! !! ! !!! !!! !!! !! !!! !!! ! ! !! ! ! !!! !!!! ! !!!! !! !!! ! !!!! ! !! !! ! !!!!! ! !!! !!! !!! !! ! !! ! !!!! !! !!!!! !! !! ! !!! !! ! !! !! ! !!! !! !!!! ! !! !!! !!!! !! !! !! !!! ! !!! !! !! ! !!!!! !! !! !!! !!!! ! ! ! !!! !! !! !! ! !! ! !!!!! !!!! ! !! ! !!! !! !! ! !! ! ! !! !!!! ! !!!!! !! !! ! ! !! !! !! !! !! !! !!!! ! !! ! !! !! ! ! !!!!! ! ! !!! !! ! ! !! !!!! !!!!! !!!! !!! !! !! ! ! !! ! !!!!! ! ! !! !!! !! ! !!! ! ! ! !! !!! !!!!!! ! ! !!!! ! !! !! ! !! !! ! !!! ! !!!! ! !!! ! !!!! ! ! !! !!!! !!! !! !!! !! !!! !!! !! ! !!!! !!! !! ! ! !! !!! !! ! !!! ! !!!! ! !! !!!! !! !! !!! !! !!! ! ! !!! ! ! !! !!! !!!! !! !! !! !!! ! ! !!!! ! ! !!! !! !! !!! ! !!! !!!!! !! !! !! !!! ! !!!! !! ! !!!! ! !!! !! !!!! !! !!! ! !! !! !! !! !!! !!!! !!! ! !! ! !!! ! !!! ! !!!!!! !! !!! !! !! ! ! !! !! ! !!! ! !!! ! ! ! !! !!! ! !! !!! ! !!! !!! !! !! ! !!!!! !!! !!!! !! ! !!! !! !! !! !!! !!!!! !! !! !! !!!!!! !!! !! !! !! !! !! !! ! ! ! !!! !!! !! !!! ! !! !!! !! !!! !!!!!! !! !!! ! !! ! !! !! ! ! !! ! !! ! !! !! !!! !! !!!! !!! !! !! !!! ! !! ! !!! !! !!! !!! !! ! !! !!! !! !! !! ! !! !!! !! !!! ! ! !!! ! !!! !! ! ! !!!! !!! ! !!! ! !!! !!!! !!!!! !! !! !! ! ! !!! ! !! !! !! !! ! !! !! ! !!!! ! ! !! !! ! !! ! !!! !! !! !!! ! !! ! ! !! ! !! ! !!! ! !!! ! !! !! !!! ! ! !!! !! ! !!!! ! !! !! !!! !! !! !! !! !! !!! !!! !! !! !! !!!! !!! !!! !! !!! !!! ! ! !!! !! ! !! !!! !! !! !!! !! !!! ! !! ! ! !!!! ! !! !!! ! !! ! !!! ! !!! !!! !!! ! !! !! ! !!! !! !! !! !! !! !! ! !!! !!! ! ! !!!! ! ! !!! !! !!!! !! ! ! !! ! !! !! ! ! !!! !! !!!! !! !! !!! !! ! ! !! !! !! !! !!!!! !! ! !!!! !! ! !!!! ! ! !!!! ! !! !!!! !! ! !! ! !! ! !!! ! !!!! !!! ! !!! !! !! !!! !! ! !!! ! ! !! ! !! !! ! ! !!!! ! ! !! ! ! ! !!! !!! !! !! !! ! !!! ! !!!! !!! !! !!!!!!! ! ! !! !! !!! !!!! !! !! ! !! !! ! ! !! !!!!! ! !!!! !!!! !!! ! !!! ! !! !!!!!410000 440000 470000 500000 530000636000063900006420000645000064800006510000BasaltGabbro, dioriteamphibolite, doleriteGranite, syenite, monzoniteGranitoidGranitoid, tonalite,granodioriteGreywacke, quartzite,mica schist, phylliteLimestoneRhyoliteSandstoneShaleAcid and intermediatevolcanic rocksConglomerate, sandstonequartziteLimestone, sandstone,mudstoneNorra K?rr Alkaline ComplexGr? nna?Lake V?ttern50km!LegendIce flowindicatorsLakesFigure1. 4b ? Geological Map of Lake Vattern area with ice ?low direction, ice ?low data and geological polygons from the Swedish Geological Survey9  The area around Norra K?rr is mapped as ?thin or discontinuous soil cover? by the S (Figure 1.5). Bedrock outcrops in many places at topographic high points, and at lower elevations the cover is an irregular till veneer. There is a ?hummocky moraine? (	igure 1.5) approximately 4 km northeast of Norra K?rr. There are also eskers and other Quaternary landforms indicating glaciofluvial transport of surficial material between the moraine and the deposit.  Zones of lower topography are filled with till and glaciofluvial material, whereas higher elevations have residual soils developing directly on bedrock. This lack of continuity in the surface material in the region introduces complications for the explorer undertaking a soil or till geochemical survey, because there is no ubiquitous sample medium. To further complicate matters, the residual soil material, derived from local bedrock, can move down slope as colluvium and sit stratigraphically above the transported tills.   1.3.4 Anthropogenic Disturbance The surficial environment around Norra K?rr changes abruptly from grassy agricultural grazing land to typical boreal Swedish forest to dense silviculture plantations (Figure 1.7). The soil profiles (Figure 1.6) above Norra K?rr indicate that the area has been disturbed by agricultural work, which has been ongoing in the region for hundreds of years, and consequently well-developed soil horizons are rare due to anthropogenic disturbance.        10Kartan visar en generaliserad bild av jordarternas utbredning i eller n?ra jordytan. Lager med en  genomsnittlig tjocklek p? mindre ?n en halv till en meter visas vanligtvis inte men kan ibland vara markerade med en ?verbeteckning. Ett urval av ytformer visas ocks?. Kartl?ggningen bygger huvudsakligen p? flygbildstolkning. D?rf?r f?rekommer en viss os?kerhet i noggrannheten hos jordartsbest?mningar och ytavgr?nsningar. Noggrannheten ?r som b?st i omr?den som ?r t?tt genomskurna av v?gar. L?gesnoggrannheten ?r i storleksordningen en till n?gra hundra meter. F?r m?nga till?mpningar, som exempelvis fysisk planering kr?vs mer detaljerad information. Ytterligare information, om till exempel jordarternas utbredning under ytan, finns lagrad i SGUs databas och kan, liksom bland annat kartbladsbeskrivningar, best?llas fr?n SGU.? Sveriges geologiska unders?kning (SGU)Huvudkontor:Box 670751 28 UppsalaTel:  018-17 90 00E-post: kundservice@sgu.sewww.sgu.se Topografiskt underlag: Ur GSD-?versiktskartan? Lantm?teriet. MS2009/08799 Rutn?t i svart anger koordinater i SWEREF 99 TM.Gradn?tet i brunt anger latitud och longitud i referenssystemet SWEREF 99.0 1 2 3 4 5 kmSkala 1:100 000Thin or discontinuous soil coverTillPeatGlaciofluvial sediment, sandGlaciofluvial sedimentEskerPostglacial sand, gravelMoraineDrumlinSedimentary bedrockClayWaterbody5kmBedrockNorra K?rr Alkaline ComplexNSweref99 TMFigure 1.5 ? Map of surficial materials in the area around the Norra K?rr Alkaline Complex, adapted from the Swedish Geological Survey. The NKACis marked with a yellow star on the large map, and shaded purpleon the inset map.100m11   Figure 1.6? Typical soil profile at Norra K?rr. Note the lack of definition between soil horizons. Image has been colour enhanced for clarity.   Figure 1.7 ? Airphoto of the Norra K?rr area with sample stations marked as yellow stars, and the deposit outlined in pink. 12  1.3 Classification of Heavy, Middle, and Light REE The lanthanides comprise a chemically uniform group of elements that mostly have a 3+ oxidation state. They vary in crustal abundance from 0.81 ppm in the case of Ce to less than reportable detection limits for Pm, which is the least abundant REE and is invariably omitted from chondrite normalized plots because of its extremely low crustal abundance (Boynton, 1984); in fact it has been estimated that the primary source of Pm is U which accounts for only 560 g Pm in the entire Earth?s crust (Belli et al., 2007).  Despite their chemical similarity, the individual REEs vary greatly in their ionic radius due to the phenomenon known as the ?lanthanide contraction?, whereby there is a greater than expected decrease in ionic radii of the elements in the lanthanide series due to poor shielding by the 4f electrons of the attractive nuclear charge, and consequently the 6s electrons are held more closely to the nucleus than expected. Hatch (2012) wrote a concise and comprehensive synopsis on the correct distinction between light, middle, and heavy REEs, summarized hereafter. The separation between the REE weight classes is primarily based on electronic structure, so the LREEs would be those that have no 4f electrons (La, Ce, Pr, Nd, Sm, Eu, and Gd) which leaves the HREEs as (Tb, Dy, Ho, Er, Tm, Yb, and Lu). The ionic radius of Y is very similar to that of Dy, so it can be grouped with the HREEs as well. The term ?middle? REE (MREE) arose from metallurgists working to chemically separate the REEs. Promethium is technically a REE, but its crustal abundance is so extremely low that it is omitted from REE plots, and ignored compared to the other more plentiful REEs. Metallurgists have taken advantage of low m levels, because the absence of this element causes a small ?gap? between Nd and Sm in the separation process that can be exploited.  1.4  Eudialyte  1.5.1 Mineral Chemistry of Eudialyte Eudialyte is a pink to red translucent zirconosilicate mineral, commonly occurring in nepheline syenites or alkalic granites. Eudialyte is a cyclosilicate with three and nine fold rings of Si and O tetrahedra. The structure is relatively open, and has been equated to that of zeolite (Golyshev et al. (1971); Guiseppetti et al. (1971)). Johnsen et al. (2003) elucidated the structure and advanced the nomenclature (Figure 1.8).  In Figure 1.8 the M and X sites represent available cation and anion substitution sites, respectively. The M(1) site is generally occupied by Ca, the M(2) site by Fe or Mn, the M(3) by Nb, the M(4) by Si, and the X site by Cl-, F-, Oh-, or CO32-. The NA4 site readily accommodates REE, Sr, K, Y, and Ca (Johnsen, 2003) 1.5.2 Weathering of Eudialyte Eudialyte is susceptible to breakdown by the action of hydrothermal fluids, and is readily soluble in acids (Johnsen et al., 2001). Literature reviews of eudialyte decomposition revealed that its products include catalpleiite, neptunite, aegerine, britholite, monazite, muscovite, biotite, zircon, and minor zeolites, as well as zirfesite (Fe2Zr(SiO2)2(OH)), which is a byproduct of low temperature fluid interaction with eudialyte. At the Magnet Cove eudialyte-bearing pegmatite in Arkansas, the pink crystals were surrounded by a ?soft yellowish brown powder? where the eudialyte had begun to weather? Williams (1841).  At Il?maussaq, eudialyte alters mainly to catapleiite and zircon, and as 13  alteration increases the Zr content of the remaining eudialyte increases, while the alkali, REE and volatile content decreases (Coulson, 1997). The intermolecular bonding forces between eudialyte grains at Norra K?rr are more strong than the intermolecular bonding forces between eudialytes at other comparable deposits, such as Il?maussaq, Red Wine, or REE deposits of the Kola Peninsula. This was first observed by Tony Mariano (2013, personal communication), who made concentrates from each of these deposits and noticed that much more time and energy was required to crush the Norra K?rr rocks. This indicates that the eudialytes at Norra K?rr are more resistant to physical weathering, and therefore biogeochemistry is less likely to be effective here, compared to other REE deposits with similar chemistry, because fewer REEs would be released into the surficial environment at Norra K?rr.     Figure 1.8 ? Depiction of eudialyte mineral group (after Johnsen et al. (2003)). The ?M? denotes cation replacement sites, and the ?X? is an anion replacement site. The chemical formula of eudialyte is (Na, Ca, REE)5(Fe2+,Mn)(Zr,Ti)[(Si3O9)2]?(OH,Cl,F).   1.6 Solubility and Mobility of REEs in Surface Water As minerals break down due to natural weathering processes, some elements become available to the surficial environment and, if they are soluble in surface waters, these elements can 14  secondary mineral, such as an oxide, or adsorbed onto the surface of some other substance, likely an organic acid,  colloid,  or manganese or iron hydroxide (Hawkes, 1957). This hydromorphic dispersion has been well documented for a variety of elements in different mineral deposit types, and the same processes occur in the surficial environments above buried REE deposits, such that ?REE and Y may be dispersed into bedrock, soils, plants, stream and lake sediments and other media, forming an anomaly much larger than the deposit itself.? (Wood, 1990). It has long been known that uranium, an actinide, is mobile in groundwater (Langmuir, 1978) and the lanthanides, which are extremely chemically similar as a group, also share chemical traits with the actinide group. However, REEs, along with the other high field strength elements like r and Ti, are often considered ?immobile? because they are commonly used as tracers to describe the evolution of igneous rocks or metamorphic units.  In the same fashion that REEs are used as tracers in geological environments, they are also used to trace geochemical processes in natural waters. Extensive research has been undertaken to determine the solubility, speciation, and complexation involved in aqueous REE transport, including the excellent review paper by Wood (1990) who summarized all the available low temperature literature data from natural waters and examined the nature and thermodynamics of inorganic complex species of the REEs and Y. His work has been heavily cited in modelling experiments that aim to determine the complexation constants of REEs with fluoride, sulphate, phosphate, carbonate, hydroxide, organic acids, simple carboxylic acids, colloids, and their behaviour as free metal ions (Smedley (1991) Lee and Byrne (1992) Johannesson et al. (1996), Dupr? et al. (1999 ); Sholkovitz (1995) Pourret et al. (2007)). Using speciation calculations of Eu, Wood (1990) shows that ?simple ion and sulphate complexes are most important at acidic p, and that the carbonate complexes become predominant at near-neutral to basic p?. In natural waters the REEs can be divided, based on abundances and fractionation patterns, into two groups: those in solution and those existing either in suspension or as colloids. There is disagreement in more recent literature regarding which anion is most important for REE complexation in solution, but the research is concordant in that all authors agree that pH is the dominant factor influencing REE speciation complexes. Secondarily, additional factors include presence of colloids (Smedley, 1991), concentration of carbonate (CO32-) and phosphate (PO43-) anions (Lee and Byrne (1992) Johannesson et al. (1996)), availability of free ?hard? anions, such as fluoride or hydroxide, sulphate (Lewis et al., 1998), availability of humic acid (Pourret et al., 2007), oxalic acid (Schijf and Byrne, 2001), and/or dissolved organic matter ( Tang and Johannesson, 2003). Complexation constants with NO3 and Cl- are relatively low, and complexation with H2S or NH3 is negligible (Wood, 1990).  Lake water samples collected above the Norra K?rr Alkaline Complex and regionally, from an area roughly 750 km2, are all circum-neutral and all display strong correlations between the REEs and the organic content (see Chapter 5, Figure 5.3a), so while many anions may be contributing to the speciation of REEs in surface waters, only those complexed with organic acids will be considered hereafter.   15  1.7 Thesis Structure This thesis consists of 6 chapters: Chapter 1 is an outline of the scope of the thesis, and introduces the general geological units and important REE minerals, as well as broach relatively unfamiliar concepts such as the mobility of REEs  in surface waters; Chapter 2 details the sample collection, preparation, and analysis, and addresses the reliability of certified detection limits for the REEs; Chapter 3 chiefly deals with the vertical migration of REEs in the surficial environment, from the breakdown of their source mineral, eudialyte, through the overlying transported sediments, and into the overlying fern plant. Sequential leach data is used to characterize the mode of REE mobility in the surficial environment. Chapter 4 elucidates the variations in REE content in different plant species, as well as the differences in REE in fern plants attributable to seasonal variation. Chapter 5 is the product of the combined efforts of several different workers, and has been published as a stand-alone paper (Biogeochemical expression of rare earth element and zirconium mineralization at Norra K?rr, southern Sweden) in the Journal of Geochemical Exploration, in December, 2012. The detailed geological framework was supplied by Tasman Metals Ltd., and the Swedish Geological Survey contributed the regional geological context, as well as glacial history of the area. The biogeochemical expertise for media selection and sampling procedure was generously and unhesitatingly provided by Dr. Colin Dunn.  The genetic model of the surficial environment is discussed in Chapter 6, along with exploration recommendations and final comments.  The body of this thesis is followed by a series of appendices, including certified analytical data, field duplicate and blind standard quality control data, and figures illustrating the reliability of REE data below the certified laboratory detection limits.               16  Chapter 2 - Materials and Methods  2.1 Sample Site and Media Selection The Norra K?rr Alkaline Complex is situated in the southern portion of the Baltic shield, and has many features characteristic of classic ?shield? environment, including numerous glacial lakes, drumlins, and moraines that give rise to low rolling topography dominated by deciduous woodland.  Southern Sweden has been inhabited for many hundreds of years, and has been farmland for much of this time. This results in varied plant species and soil types due to differences in land use. The area around the NKAC was mainly used for grazing, interspersed with coniferous  plantations.  In total, 203 vegetation samples were collected from 157 sites on three occasions over the course of 15 months: (I) one week in mid June, 2011; (II) a brief visit in late August 2011; (III) three sampling programs in late August, 2012, including (i) extension of two original orientation lines, (ii)  regional geochemical baseline sampling from an area covering 750km2, and (iii) detailed sampling from three vertical test pits to determine the location of REEs in the vertical profile from bedrock, through overlying soils, and into the overlying fern plant (See Figures 1.5, 1.7, 3.1).  2.2 Sample Collection 2.2.1 June 2011 This was the preliminary pilot study, and little was known about the survey area with respect to distribution of vegetation. Nothing was known about the uptake of REEs by vegetation at this deposit, so the sampling program was comprehensive. The aim was to identify which plants, if any, would be able to accumulate appreciable amounts of REEs in their tissues, thereby making them an acceptable sample medium for biogeochemical exploration for REE deposits.  The majority of the samples were fern plants because they were the most widespread plant species in the survey area, and previous work has demonstrated the ability of ferns to take up REEs (Markert and Li, 1991; Ichihashi, 1992; Fu et al., 1998; Guo, 1996; Takada et al., 1996; Wang, 1997;  Wyttenbach, 1998; Ozaki and Enomoto, 2001; Tyler, 2004; Dunn, 2007). The other species sampled were: oak leaves ( Quercus robur ), alder twigs (Alnus glutinosa ), birch twigs ( Betula pendula), ash twigs ( Fraxinus excelsior ), hazelnut twigs ( Corylus avellana ), Norway Spruce twigs (Picea abies ) and feather moss (see Table 2.1).    17   Table 2.1 ? Plant species collected in June, 2011, listed by common and Latin names, with n values and sample details.  The fern fronds, less their stalks, were collected and packed into 4? x 10? Kraft paper soil sample bags. At least 50 g of material was taken for each sample, which yielded about 25 g of dry tissue. The twigs of the other species were cut from around the circumference of each tree, using the same pair of stainless steel pruning shears, and the entire branch, including the leaves, was stuffed into a 7? x 11? BC woven polypropylen  bag. It is critical in biogeochemical exploration programs to sample media consistently; both the length and width of twig should be similar for all samples collected. Previous work has shown that spruce twigs with approximately seven years growth show good reproducibility of analytical results (Dunn, 2007). The feather moss was detached from the rock, shaken vigorously to dislodge soil particulates from the rootlets, and put into the same style of polypropylene bag.  It became immediately apparent once in the field area that the soil horizons had been disturbed by anthropogenic activities, but soil samples were still collected from the Ah and B horizon in seven locations to enhance the breadth of the survey. Since the preferable sample medium is a fern plant, one of either Dryopteris filix-mas or Athyrium filix-femina was collected at each of the 82 sample sites (Markert and Li, 1991; Wyttenbach, 1998; Tyler 2004; Dunn, 2007). Multiple plant species were sampled at 28 sites, and in total 118 samples were collected.  2.2.2 August 2011 The plants sampled at the end of the summer in 2011 were collected by C. Dunn and H. Waldron while undertaking a separate field study. They collected 16 samples from seven of the same sites that samples had been collected previously. Samples were dominantly Dryopteris filix-mas and feather moss, as well as oak leaves from one site. The sample collection methodology was identical to that employed in early summer. 2.2.2 August 2012 Sampling in August, 2012 followed the same procedures as before and involved three separate collections: (I) regional fern sampling;  (II) the extension of two original orientation lines from the previous summer; (III) a vertical sampling program designed to track the movement of REEs from the bedrock, up through the soil profile, and into the overlying fern.  This involved 18  selecting a healthy, normal looking fern plant from a site previously sampled in 2011 which returned elevated REE values in the dry fern leaf tissue. Each fern plant was carefully excavated in its entirety using a large shovel. Soil around the rootlets was collected in a 7? x 11? woven polypropylene HUBCO bag, and the root bulb was separated from the complete fronds and placed into a separate HUBCO bag. The fern fronds were placed between sheets of unbleached brown paper, and press-dried as complete fronds (see Figure 2.1). A pit was dug below where the fern was growing, and soil samples were collected from each horizon until bedrock was reached, which was also sampled. In one instance, the bedrock was too deep to reach with a normal spade, so angular fragments, probably locally derived, were collected instead. Descriptions and annotated photos of soils collected from vertical sample sites are provided in Section 3.2.1ii.    Figure 2.1 ? photo of fern frond, before being pressed and dried for transport back to Canada  2.3 Sample Prep All samples were dried to a constant weight at 60?C, and then the leaves or needles were separated from the twigs. In the case of oak ( Quercus ) and fern (Athyrium filix-femina, Dryopteris filix-mas,  and  Pteridium aquilinum) samples, the leaves were prepared for analysis. The twigs of alder (Alnus glutinosa ), birch ( Betula ), ash ( Fraxinus ), hazelnut ( Corylus ), and Norway Spruce (Picea abies ) were prepared for analysis. Each sample was milled to a powder in a stainless steel, rotating blade, electric coffee grinder. The milled fern and oak leaves were placed in their original brown paper Kraft sample bags, whereas the milled twigs were shipped in new labelled envelopes. All of the samples were sent to Acme Analytical Laboratories in Vancouver, BC, for destructive chemical 19  digestion. Splits of Dryopteris filix-mas leaves and Norway Spruce twigs were sent to ALS in Australia for ionic leach testing.  Samples collected in August 2012 to assess the REE variation within each fern plant were separated in the field, dried, and transported back to Canada. The rootlets, analysed separately from the stem and leaf tissue, were thoroughly washed and dried in the new MDRU lab at UBC in Vancouver, Canada, prior to chemical analysis. Root, stem and leaf tissue were all prepared in the same manner as vegetation samples from the previous year. All soil samples were sieved to -80 mesh and that size fraction was sent for chemical analysis. A selection of samples was split and washed during the preparation stages to assess for airborne contamination, and there was no evidence in the data that airborne contamination is a problem around Norra K?rr. All sixty Norway Spruce twig samples were reduced to ash at the Pacific Geoscience Centre, in Sidney, BC. The kiln heat was ramped up to 475?C, after which time the samples remained inside for 12 hours, and then the kiln was allowed to cool for a few hours before opening. Reducing samples to ash is particularly helpful when low concentrations of elements of interest, or high background levels, interfere with obtaining  reliable results, especially near instrumental detection limits. The limitations of ashing include the increased sample prep time, and the loss of any elements, such as Hg, that will volatilize at low temperatures.   2.4 Sample Analysis 2.4.1 Vegetation Tissue and Soil Samples by Aqua Regia (Vancouver, Canada) The 1VE package provided by ACME Analytical Laboratories is a hot, modified, aqua regia digestion of 1g of dry milled vegetation tissue, preceded by a cold nitric acid pre-treatment. The modified aqua regia solution used at ACME is 1 equal part of each water, nitric acid, and HCl, compared to the ?classical? aqua regia which is ? parts HCl and 1 part nitric acid. The cold nitric acid pre-treatment is used to oxidize the organic component of the sample and prevent the reaction from occurring too vigorously and overflowing from the beaker. The 1VE digestion is the exact same strength as the 1F digestion, but is specially modified for vegetation samples due to their high organic content.  The 1F package provided by ACME Analytical Laboratories is a hot, also modified, aqua regia digestion for low or ultra-low determinations of soils. Both 1VE and 1F digestions used in this study were followed by an ICP-MS finish for 65 elements: the full suite of 53 elements and the REE add-on. The 14 soil samples collected during the first sampling campaign in June 2011, as well as the 43 fern samples collected along the extensive southern line in August 2012, were analysed by the 1F hot aqua regia package.  2.4.2 Soils by Sequential Leach (Vancouver, Canada) The Sequential Leach procedure at Acme Analytical Laboratories in Vancouver BC involves 5 leaches applied to the sample in order of increasing leach strength: de-ionized water, ammonium 20  acetate, 0.1M sodium pyrophosphate, 0.1M hydroxylamine, and 0.25M hydroxylamine. Any remaining sample material was treated with 1T, ultratrace 4 acid, for ?near total? digestion. Soil samples collected from the vertical profiles were leached sequentially to determine in which soil phases the REEs were occurring.  2.4.3 Vegetation Tissue by Microwave Digestion (Lulea, Sweden) The samples were digested in Teflon containers in a Marsexpress microwave for 30 minutes at 170? C using 5ml HNO3 (65%)+0.5ml H2O2 (30%)+20?l HF (40%). After digestion the samples were diluted up to 10 mL with MQ-water. The ICP-MS analyses were carried out according to modified US EPA method 200.8 (Creed et al., 1994). 2.4.4 Whole-rock Analysis of Representative Rock Types by Tasman Metals Inc (Lulea, Sweden) Tasman Metals Inc. had suites of representative rock types analysed at ALS Laboratories in Sweden by four-acid ?near-total? digestion (ALS package ME-MS61), as well as by lithium borate fusion (ALS package me-xrf06). The 4 acid digestion quantitatively dissolves nearly all elements for the majority of geological materials, and finishes with a combination of ICP-MS and ICP-AES instrument techniques. The XRF procedure is a lithium borate fusion with a direct analysis (XRF) determination. 2.4.5 Whole-rock Analysis of Bedrock Samples Below Vertical Profile Sites (Vancouver, Canada) The  whole rock and trace element package, provided by Acme Analytical Laboratories returns the total abundances of the major oxides (SiO2, Al2O3, Fe2O3, CaO, MgO, Na2O, K2O, TiO2, P2O5, Cr2O3) in a 0.2 g sample digested in dilute nitric acid after a lithium metaborate/tetraborate fusion, with an ICP-ES finish. The trace element portion of the analysis determines the REEs and the refractory elements by ICP-MS following a lithium metaborate/tetraborate fusion and nitric acid digest, while the precious and base metals are determined from a 0.2 g split which is digested in aqua regia and analysed by ICP-MS. 2.4.6 Scanning Electron Microprobe (UBC, Vancouver, Canada) Soil samples (n = 8) from the vertical test pits were examined under the scanning electron microscope (SEM) at UBC, and high field strength elements (HFSEs) were analysed using an energy dispersive X-ray spectrometer. The HFSEs, identifiable by their high luminosity, contained Y and Zr, but the other REEs were not detectable.  The electron microscope is  a Philips XL30, with Bruker Quanta 200 energy-dispersion X-ray microanalysis systems, equipped with an Xflash 4010 SDD detector.  2.4.7 Water Chemistry in Lulea, Sweden (elements and anions) Water samples were analysed after acidification with 1 ml of nitric acid (Suprapur) per 100 ml of water sample. Analyses were done according to EPA methods (modified) 200.7 (ICP-AES) and 200.8 (ICP-SFMS).  21  2.4.8 Regional soils (Lulea, Sweden) The sample was dried at 105 ? C, and 0.1 g of sample was digested with 0375 g of LiBO2 and dissolved in HNO3. Loss on ignition (LOI) was performed at 1000 ? C. Analysis at ALS in Lulea was done according to EPA methods (modified) 200.7. (ICP-AES) and 200.8 (ICP-QMS).   2.5 Quality control A typical mineral exploration quality control program was utilized for this survey. Field duplicates were taken every 20 stations, blind standards (fern foliage) were inserted every ten samples, and preparation duplicates and blanks were also inserted by the laboratory. Generally the worst reproducibility arises from field duplicates because the variability is influenced by natural factors, such as plant growth, whereas laboratory variability is more easily controlled. There was good reproducibility among the field duplicates, with most duplicate analyses reporting within 30% of each other (Figure 2.2), and excellent reproducibility among the laboratory duplicates. Field duplicate precision diagrams for all REEs are presented in Appendix A6. The precision for the blind standards (Table 2.2) is very good for the LREEs, and moderately good for the HREEs. The relative standard deviations are highest for Tm, Yb, and Lu because they were mostly below the detection limit, particularly in the cases of Tm and Lu.   Figure 2.2 ? Neodymium precision diagram in field duplicates   22   Table 2.2 ? Precision data for blind standards  2.6 Detection Limit Reliability Thorough investigation of the REE data suggests that it is reliable below the certified detection limits provided by Acme Analytical Laboratories, based on REE probability plots and scatter plots of the REEs as a function of yttrium; yttrium suffers the fewest detection limit interferences due to its higher crustal abundance, and therefore is one of the more reliable REE-like elements. Figure 2.3 outlines the methodology used to determine a reliable detection limit based on REE populations coupled with REE content as a function of Y. The certified detection limits, as well as the lowest reliable detection limits, for the REEs are listed in Table 2.3, and both certified and <DL replaced REE data is presented in Appendix A2. Analytical data from biogeochemical samples can be reliably used, even below the certified detection limit, because there are no matrix interferences in a biogeochemical sample. Detection limits in soil samples are complicated by the presence of silicate and resistate minerals present in the sample; however, since a biogeochemical sample is all organic material, there are no matrix interferences To assess DL reliability, probability plots of the uncertified raw as well as the certified replaced data were generated (see Figure 2.3a and 2.3b) to look for inflection points or breaks in the data that would indicate different populations. The values determined above the certified detection limit are all coloured grey (Figure 2.3a& b) and the data is assessed for overlap with the certified values. The next population, coloured green, on the Tm vs Y bivariate plot (Figure 2.3c) 23  extends from 0.019 ppm > Tm > 0.011 ppm. After that group has been selected, it is clear on the probability plot that values 0.011 ppm > Tm > 0.007 ppm belong in this population as well. Below 0.007 pm there is  a natural break, which indicates that values below this threshold belong to a different population, and are therefore unreliable.  The LREEs (La, Ce, Pr, Nd) all returned values above the certified detection limit. Of the MREEs (Sm, Eu, Gd) only Sm had no BDL values, and all of the HREEs, with the exception of Y, returned at least 11% of the values below detection limit. The least reproducible elements were Tm and Lu, with 65% and 78% respectively, returning below the certified detection limit. Since the REEs behave similarly as a group, and the LREEs didn?t return values below the detection limit, the methodology for determining reliability will be outlined here for Tm only. Yttrium was used with the HREEs because, even though it technically is not a HREE, they behave similarly due to their comparable chemical properties, so Y can be confidently grouped with the HREEs. Figures displaying reliability below certified detection limit for all other REEs are presented in Appendix A2. All Dryopteris filix-mas data from June 2011 are presented in Figure 2.3a, and Figure 2.3b is the same data set with values <0.02 ppm replaced with 0.01 ppm. Figure 2.3c is the graph of Tm as a function of Y, with no values replaced. REEs were graphed against Y to assess detection limit reliability because Y is more reproducible than most HREEs, due to its greater crustal abundance.    Table 2.3 ? Certified detection limits from Acme Analytical Laboratories, and lowest practical reliable detection limits for this thesis research. The LREEs were all above the certified detection limit, so those have been marked ?not applicable? (n/a). All values are in ppm.   24 straight line without breaks, in Figure 2.3a, are deemed one population) are coloured orange, with moderate con?idence that these data are reliable. Data coloured red are have low REE content, or are part of a different population. Con?idence that these data are reliable is very low.Tm as a function of Y in Dryopteris ?ilix-masa) b)Tm content in Dryopteris ?ilix-mas samplesTm (ppm)Tm (ppm)Y (ppm) Y (ppm)002.00.020.0100 2.0 4.0 6.0 8.00.020.040.060.08c)Lowest reliable detection limitCertified detection limitCertified detection limitCertified detection limitLowest reliable detection limitFigure 2.3 - Methodology example of detection limit reliability. Figure 2.3b is all Dryopteris ?ilix-mas data, with BDL (<0.02ppm) values replaced with 0.01ppm. Figure 2.3a shows the same data, but with the BDL values left as raw ?uncerti?ied? values. Colour scheme is based on REE (in this case, Tm) as a function of Y content. All REEs show good correlation with Y content, displayed here along the line Y = X. Data just below the certi?iable DL (0.02ppm) is coloured green. High con?idence that these samples are reliable. Samples with lower REE content, but in the same population (populations are determined based on probability plots; data plotting on a25   Chapter 3 - Variation in Rare Earth Element Content in the Surficial Environment Above the Norra K?rr Alkaline Complex   It is widely known that plants can accumulate REEs in their tissue (Laul et al., 1979; Evans, 1983; Wang et al., 1983; Fu et al., 1998; Wyttenbach et al. 1998; Ozaki and Enomoto, 2001; Zhenggui, 2001; Tyler, 2004; Dunn, 2007), and that the movement of elements into plants is principally via the interaction of roots with surrounding soils. Chemical elements may be taken up by the plant either in solution, or via ion-exchange between the root and the surface of the soil particle (Taiz and Zeiger, 2010). Many authors have shown that the concentration of REEs increases from stem < leaf < root, but these experiments have mainly been in non-natural laboratory conditions, using doped solutions (Evans, 1983; Fu et al. 1998; Ding 2005).  The concentrations of REEs in soil-plant systems have been succinctly reviewed by Tyler (2004), but the variations in REE content from bedrock to soil to plant, and hypotheses regarding the mobility of the REEs for the purpose of mineral exploration in the boreal environment remain, as yet, unstudied. An experiment was designed to assess the speciation and mobility of REEs in the environment around the Norra K?rr Alkaline Complex. In order to determine the REE content in the surficial environment around the complex, samples were collected from each component of the system:  bedrock, groundwater, glacial till, surficial and subsoil materials, and overlying biological matter.  The sites for the vertical profiles (Figure 3.1) were selected based on locations where ferns collected the previous summer yielded high REE contents (see Section 5.3.4). Specific materials collected at each site are listed in Table 3.1.   Table 3.1 ? Materials collected from the three vertical profile sites.  26   Figure 3.1 ? Locations of three vertical profile sample sites over the Norra K?rr Alkaline Complex. Sites were chosen based on high REE content in ferns, which was determined in the previous year?s study.  3.1 Lithogeochemistry 3.1.1 The Negative Eu anomaly A ?negative Eu anomaly? is the term for a lower concentration of Eu with respect to the levels of Sm and d, europium?s neighbours, in any substance, such as rocks, soils, plants, or water. The Eu anomaly can be quantified using chondrite normalizing values (see Equation 3.1), or it can be determined using a chondrite normalized plot to assess the relative value of Eu compared to Sm and Gd.   Equation 3.1 ? Quantified Eu anomaly calculation after Boynton (1984)   All of the REEs occur in the trivalent oxidation state, 3+, with the exception of Eu and Ce, which can occur in different valences (Eu2+/Eu 3+ or Ce3+/Ce 4+) based on the redox conditions of their surrounding environments. In crystallizing magma chambers, or hydrothermal and 27  metamorphic regimes, high temperatures and pressures coupled with low pH and low oxygen fugacity create a reducing environment and therefore Eu will preferentially exist as Eu2+. Conversely, in the oxidizing conditions (e.g. 25?C and 1 bar) found at the surficial environment, Ce will preferentially exist as Ce4+, the quadrivalent oxidized form. Under these surficial conditions, Eu3+ reduction to Eu2+ is ?extremely unlikely to occur, because the Eu3+/Eu 2+ redox equilibrium is located at very low oxygen fugacity (e.g. log ?O2 = -?? at p 7)? (Bau, 1991). Negative and positive Eu and Ce anomalies can be used to infer which processes have affected the formation of minerals in a particular geological environment, and which processes continue to affect the speciation of the elements in the surficial environment.  In a reducing environment, such as a crystallizing magma chamber, Eu2+ has similar properties to Ca2+, and substitutes for Ca2+ in the formation of plagioclase. This removes the Eu from the liquid phase of the magma, and produces a negative Eu anomaly. In many cases a negative Eu anomaly is caused by fractional crystallization within a reducing magma chamber.  The agpaitic nepheline syenites at Norra K?rr all have negative Eu anomalies when normalized to chondrite. This can be caused by fractional crystallization of alkali basaltic liquids at crustal levels, or extreme fractionation of nephelinitic mantle-derived melts (S?rensen, 1997 ) whereby alkali-rich melts crystallize under conditions that prevent volatiles from escaping. As this thesis is chiefly concerned with the surficial environment, attention will be focused on the oxidation state of Ce and Eu at standard conditions (25?C and 1 bar).   3.1.2 Chondrite Normalization Chondrite normalization diagrams (e.g. Figure 3.3) are generally used when discussing REE data, and the most appropriate normalizing values will depend on the nature of the sampled material. For example, evolutionary processes in mantle-derived magmas should be normalized to the primitive mantle values of McDonough and Sun (1995), while Taylor and McLennan (1985) chondrite values derived from shales can be used to approximate the upper continental crust.  The values of Wakita et al. (1971) and Nakamura (1974) are based on average chondrite, whereas Boynton (1984) normalization values are based on carbonaceous chondrite (Table 3.2). Carbonaceous, or C1, chondrite values contain carbon, water, and other volatiles, and are meant to approximate the solar nebula in composition. These normalizing values have been used in this research. Wakita?s, Nakamura?s, and Boynton?s values are nearly identical, and they fall in the middle of the range used in current literature, but Boynton?s chondrite normalizing scheme was deemed most appropriate because it uses carbonaceous chondrites, it is the most complete (Nakamura?s only has ten of the 1? REEs), and it was published more recently than the other comparable chondrite normalized values.  28   Table 3.2 ? Commonly used chondrite normalization values (Nakamura, 1974; Boynton 1984; Taylor and McLennan, 1985; Wakita et al., 1971). All values are in ppm.   3.1.3 Classification  The rock samples from the base of the vertical profiles are nepheline syenites from the two northerly sites, (NK12-072 and NK12-039), and a foliated granite from the most southerly site (NK12-060). The sample from the most northern site NK12-072 (Figure 3.2a) is a dark grey-green, fine-grained nepheline syenite with minor small catapleiite and eudialyte phenocrysts, small thin fractures filled with microcline, and minor amphibole. The sample from the central site (NK12-039,  Figure 3.2b) is a foliated nepheline syenite with phenocrysts of catapleiite (purple rectangular laths) and smaller eudialyte (rose coloured subhedral grains) occurring within bands of microcline and amphibole. The sample from the most southerly site (NK12-060, Figure 3.2c) is a strongly foliated biotite-rich granite with red K-feldspar porphyroblasts which surround quartz grains in places.   29   Figure 3.2a-c  ? Photos of rock samples collected from base of vertical profile sites. (a) NK12-072: fine grained, green-grey aegerine rich nepheline syenite. (b) NK12-039 foliated nepheline syenite with phenocrysts of catapleiite and eudialyte occurring within bands of microcline and amphibole. (c) NK12-060 strongly foliated biotite rich granite with red K-feldspar porphyroblasts.  As outlined in Section 1.3.2 (detailed geology of the NKAC), the contact between the Norra K?rr nepheline syenites and the surrounding granitic country rock is strongly foliated and fenitized as a result of regional metamorphism related to the same event responsible for the emplacement of the Norra K?rr deposit. The NKAC may have been emplaced as a sill within the V?xj? granites (Rankin, 2011), which were concurrently foliated and fenitized. As a result, the contact between the V?xj? granites and the NKAC is difficult to distinguish because of their similar textures resulting from their shared hydrothermal and metamorphic history. The best field indicator for distinguishing the V?xj? granites from the NKAC is colour, the V?xj? granites are slightly more reddish due to their high proportions of K-feldspar, and the presence of quartz grains, which are absent in the NKAC. Lithogeochemical whole rock data from the base-of-profile rock samples were obtained  using lithium metaborate/tetraborate fusion and dilute nitric acid digestion, for the major oxides and several of the minor elements, with an ICP-emission spectroscopy finish; the trace elements were determined using the same digestion but the analysis was performed using an ICP-MS. The samples were classified using a total alkali silica (TAS) diagram (Figure  3.4) devised by  Cox et al. (1979) and adapted by Wilson (1989). Based on this diagram, two of the three rocks at the base of the vertical profiles are nepheline syenites, and the other is an alkali-rich granite, which corresponds with the V?xj?-type granites surrounding the Norra K?rr deposit.  The vertical profile pit dug at the southernmost site (NK12-060) was greater than 1 m deep, and at the base of the pit a Dutch auger was used to collect till samples from the lower C horizon; bedrock was not reached.  Since a bedrock sample was not available, angular fragments were collected from the base of the vertical profile and analysed in lieu of bedrock. Site NK12-060 is almost directly above the inferred contact between the nepheline syenite and the surrounding granites; the granitic fragments collected here are angular and therefore can be assumed to be locally derived. The granitic fragments have similar textures to Norra K?rr rocks because they originated from the foliated and fenitized aureole of the V?xj? granites directly associated with the emplacement of the NKAC. In addition, a suite of representative rock types analysed by Tasman Metals Inc. was compiled and classified in the TAS diagram (Figure 3.4). The two mafic rocks in Figure 3.4 are classified by Tasman Metals Inc as MAA  ? ?mafic, probably alkaline, fine to medium grained, dark, 30  amphibole rich, intrusive rock? (Reed, 2011), and the most acidic rock, the granitic fragment from the southerly site (NK12-060), has greater than 70% SiO2, and appears to be sourced from the surrounding V?xj? granites. The analyses of the bedrock collected from the two more northerly sites (NK12-039 and NK12-072) are chemically equivalent to the major rock types of the NKAC.  The two syenite base-of-profile lithogeochemical samples have similar REE profiles when normalized to chondrite, but the third base-of-profile rock sample, the granitic fragment, has orders of magnitude lower REE content than the other lithogeochemical samples (Figure 3.3). Lithogeochemical analyses of the NKAC have a strong negative Eu anomaly, but the granitic fragment only has a minor negative Eu anomaly. This weak Eu anomaly  (see section 3.1.2) is a different magnitude and shape than the Eu anomaly present in the other lithogeochemical samples, therefore reflects a different set of geological processes.    Figure 3.3 ? REE Chondrite normalized plot (after Boynton 1984) showing a compilation of all rock types at Norra K?rr (yellow circles, analysed by Tasman Metals Inc.) and the rocks collected for this thesis (grey squares). Note the difference in magnitude between the granitic ?base of profile? angular fragment collected from NK12-060 compared to all other rock types.       31Alkali GraniteIjoliteGabbroGraniteSyeno-dioriteDioriteQuartz DioriteNepheline SyeniteSyeniteNa2O + K2O (pct)SiO 2 (pct)Ultrabasicor UltramaficBasic orMaficIntermediate Acid or FelsicTotal Alkali Silica Plutonic Classification Diagram?MAA? rock typeNK12-072NK12-039NK12-060Average of major rock types (Tasman Metals Inc.)Lithogeochemical samples collected for this thesisfragmentGraniticFigure 3.4 ? total alkali silica (TAS) classi?ication of lithogeochemical samples collected for this thesis: three base of pro?ile rock samples [NK12-072 (open circle), NK12-039 (open square), and NK12-060 (open triangle)]. Grey circles are averages (n = 210) of each major rock type present in the Norra K?rr Alkaline Complex (analyses provided by Tasman Metals Inc.)32   3.2 Soil Chemistry 3.2.1 General Soil Chemistry 3.2.1.i Soil Samples, June 2011 A suite of 14 soil samples was collected, as well as a variety of plant species, during the first sampling campaign in June 2011. Of the soil samples, seven were from the Ah horizon and seven were from the so-called ?B? horizon. As noted in Chapter 1, the developed soil horizons above the Norra K?rr complex have been extensively disturbed by agricultural practices, which have been taking place since the middle ages. Local fauna, particularly the wild boar, which is considered a pest in southern Sweden, routinely disrupt surface and subsurface soils. Since the soil horizons were disturbed and inconsistent,  soil samples were deemed an unreliable sample medium for REE exploration in this area.  The 14 soil samples all display similar geochemical patterns to the bedrock assays provided by Tasman Metals Inc., albeit at lower total rare earth element (TREE) content (Figure 3.5). The negative Eu anomaly with respect to chondrite normalization, which is apparent in all rock types at Norra K?rr, is visible in the soil samples as well, however, there is also a positive Ce anomaly in the soil profiles that does not appear in the lithogeochemistry. This could be due to several different processes, all of which are possible because Ce, like Eu, has two different oxidation states; however, Ce can exist as Ce3+ or Ce4+, depending on whether it is in a reducing or oxidizing environment. This positive Ce anomaly in the surficial environment will be treated in more detail in section 3.2.2.  There is a distinct difference in REE content between the so-called ?B? horizon soil samples and the Ah horizon soil samples. The Ah soils have consistently lower REE content than the B horizon soils taken from the same location.  The median sum of the REEs in the Ah horizon is only 23.4ppm, compared to the median sum of the REE content in the B horizon, which is 87.7ppm (Table 3.3).  This suggests that there is either some leaching process moving REEs away from the Ah horizon, or there is a REE-enrichment process acting on the B horizon soils. This will be commented on in more detail later in the chapter.    Table 3.3 - Summary of the REE content, as well as the median ?REE value, for the Ah and B soil horizons     33REE Chondrite Normalized after Boynton (1984)Ah soils?B? soilsCompiled lithogeochemicalsamples from Tasman Metals IncFigure 3.5 ? REE Chondrite normalized plot (after Boynton 1984) of soil samples collected in June 2011 (7 Ah horizon, green circles; and 7 ?B? horizon, blue circles) with representative lithogeochemical samples of all rock types in the Norra K?rr complex (red circles, analysed by Tasman Metals Inc.)34  3.2.1.ii Soil Samples, August 2012 During the second sampling campaign, in late August 2012, a suite of soil samples were collected from vertical profiles to assess the variation in REE content between each soil horizon. Eight samples were collected from three test pits, and these, as well as the 14 soil samples from 2011, were sent for sequential leach analysis. The aim of this experiment was to define which labile portion of the soil sequesters the majority of the REEs, and then infer a geochemical process responsible for the REE mobility in the surficial environment. Descriptions of the soils collected from the three vertical sample pits are listed in Table 3.4, and photos of each soil sample are provided in Appendix A7.  Table 3.4 - Physical descriptions of soil samples collected from each horizon at three vertical test pits   Figure  3.6 ? Photo of vertical test pit at site NK12-060. Note the lack of definition between soils horizons, particularly the A and B horizons. Depth to the C horizon is 70 cm. 35  3.2.2 Sequential Leach Sequential leaches are based on the same principles as selective leaches, but each selective leach is applied in order of increasing leach strength to the same sample material. Selective leaches target elements held by a specific soil phase or range of phases, thus allowing inferences of the geochemical processes acting on the sample material. The leaches target elements mobilized by groundwater, or adsorbed as ions onto clays, organic acids, or amorphous Fe and Mn oxyhydroxides.  A six step sequential leach, finishing with a ?near total? 4acid digestion,  was performed on 25 soil samples collected above Norra K?rr. Eight of the samples were collected from distinct horizons in the vertical profile test pits, and the other 17 samples were duplicates of soil samples collected the previous summer. The sequential leach process involves six leaches, which are, in order of increasing strength: de-ionized water (?20?), ammonium acetate (?N3Ac?), sodium pyrophosphate (?Na4P2O7?), 0.1M hydroxylamine (?w-NH2?), 0.25M hydroxylamine (?s-NH2?), and four acid ?near total? digestion (?4acid?). Each of these leaches is designed to liberate elements bound to a specific soil phase or range of phases. By determining which phase dominantly hosts the REEs in the surficial environment, hypotheses can be made regarding which geochemical processes are the most influential.  Each leach liberates elements bound to the sample in specific targeted phases: ? The de-ionized water liberates ions that have been hydromorphically dispersed and are weakly adhered to soil particles.  ? The ammonium acetate (?N3Ac?) targets exchangeable cations that are either adsorbed onto clays or co-precipitated with carbonates.  ? The sodium pyrophosphate (?Na4P2O7?) extracts elements adsorbed on to organic acids, such as humic or fulvic components.  ? The weaker hydroxylamine (0.1M, ?w-NH2?) attacks phases adsorbed by amorphous Mn hydroxides, which is often the reactive phase for scavenging mobile elements in arid environments.  ? The stronger hydroxylamine (0.25M, ?s-NH2?) attacks amorphous 	e hydroxides as well as crystalline Mn hydroxides.  ? 	inally, the four acid ?near total? digestion dissolves silicates and begins to break down resistate minerals, such as zircon or monazite (Cohen et al., 2010).   It is often warned that REEs can be insoluble in four acid digestions, but as outlined in Section 1.3.2, the REEs at Norra K?rr are sourced from the zirconosilicate mineral eudialyte, which is much more soluble than some of its REE mineral counterparts, such as monazite, xenotime, or zircon.  At each site, results from the sequential leaches display similar characteristics for the Ah and B soil horizons, as well as for the tills. First of all, the de-ionized water leach always returns the lowest levels of REEs, and the 4acid ?near total? digestion always returns the highest levels of REEs, approximately 100x higher than the de-ionized water leach. Secondly, the ammonium acetate and 36  both strengths of hydroxylamine return similar amounts of REEs for each profile, with the notable exception that the hydroxylamine signal, which is always slightly enriched in LREEs compared to the ammonium acetate. Lastly, and most interestingly, the sodium pyrophosphate leach returns the second highest concentrations of REEs, even though it is the third weakest leach.  Lithogeochemical analyses of the rocks from the Norra K?rr Alkaline Complex display a well documented negative Eu anomaly (see section 3.1.2, Figure 3.4, and Figure 3.5), and this pattern is apparent in all of the sequential leaches when normalized to chondrite.  Some of the sequential leaches, however, also display a positive Ce anomaly in the Ah and B horizons which is not inherent in the parent bedrock. This could be attributable to the susceptibility of secondary Ce-bearing minerals to certain leaches, and the differential mobility of Ce in an oxidizing environment compared to the other REEs.   Generally, under geological conditions, Ce exists in the reduced state, Ce3+, which is the same oxidation state as the rest of the lanthanides. When the reduced Ce3+ cation comes in contact with oxidizing conditions, such as those found in near surface weathering environments it will precipitate as CeO2 (Braun et al., 1990), and this can induce a positive Ce anomaly. If the REEs have been in solution in the near surface environment, the Ce will oxidize and immediately precipitate out of solution as CeO2 which is an extremely insoluble compound (De Baar et al., 1985). Ce (IV) can also be adsorbed on to the surfaces of oxyhydroxides, such as Fe or Mn, (de Carlo et al., 1998; Ohta and Kawabe, 2001), and in this case it will remain in the oxygenated subsoil complexed with Fe or Mn, and therefore be less available to plants than other REEs (Robinson et al., 1958). Manganese compounds present as surface coatings on soil grains can scavenge Ce4+ from the surficial environment (Hem, 1978).  Interpretation of the sequential leach data was done on a per-site basis because the bedrock material differs at each location. To this effect, special normalization schemes were devised for each site based on their parent bedrock, and these will be referred to as ?local? normalizing schemes. Normalizing to the REE content of the underlying bedrock, on a per-site basis, is helpful because subtle variations, such as Ce and Eu variations or HREE/LREE fractionations are not visible when normalizing to global average chondrite values because their ranges are large, and therefore subtle variations are masked. There is a pervasive Eu anomaly in all of the weak leach soil chemistry data (Figure 3.7) when normalized to chondrite, but when normalized to local bedrock (Figure 3.8b, Figure 3.9b) there is actually a positive  Eu anomaly at the southernmost site, in the granitic fragment as well as the till samples.  Of particular importance here is the transported material at the southernmost site (NK12-060) where an angular granitic fragment was taken as a proxy for local bedrock when in fact it originated from nearby V?xj? granites. Therefore a compilation of lithogeochemical data from the Tasman Metals Inc. drilling database was assessed to choose a suitable representative bedrock substitute for this location. The ?pulaskite? unit is a ?medium to coarse-grained alkaline rock composed of albite, microcline, aegerine, Na-amphibole, and minor biotite and nepheline? (Reed, 2011) and is representative of the soil chemistry at site NK12-060.   37REE Chondrite Normalization (after Boynton, 1984) H 2ONH 3AcNa2P4O 7w-NH 3OHs-NH 3OH4acidtotalA horizonB horizonC lowerC upperFigure 3.7 ? Chondrite normalized plot of all sequential leaches at each site, including the sum total of all the leaches (burgundy points) which approximate the total REE content of the sur?icial materials.38  3.2.2i ? Site NK12-039, the central site The sequential leach data obtained from the A and B soil samples at this site were normalized to the bedrock sample collected here. The A and B horizons both display similar REE patterns: a positive Ce anomaly of varying magnitude depending on the strength of the leach, and an overall enrichment of LREEs relative to HREEs (Figure 3.8b). The activity of Ce3+ decreases with increasing pH, such that in circum-neutral soil solutions (pH 6.98-7.31) the speciation of Ce is almost wholly controlled by the formation of CePO4 (Diatloff et al., 1996). At site NK12-039, the positive Ce anomaly could possibly be attributed to CePO4 formation in the A and B horizons, which would not be liberated by weak NH3Ac and H2O leaches. Alternatively, the positive Ce anomaly observed for both strengths of hydroxylamine leach could be due to the presence of Fe and Mn hydroxides, which can act as sinks for Ce4+ cations (Hem, 1978). Also of note is the lack of a strong positive Ce anomaly in the NH3Ac leach. The ammonium acetate leach targets exchangeable ions that are either adsorbed by clays or co-precipitated with carbonates. The disparity in Ce anomaly from this leach indicates that there is little excess Ce in the exchangeable ion fraction of the soils.  When normalized to any global chondritic values, be it Haskin or Boynton or Nakamura, the REE patterns at this site remain constant (Figure 3.8a), and all show the same positive Ce anomaly and relative LREE enrichment as seen in the specialized per-site (or ?local?) normalization scheme (Figure 3.8b), however when normalizing to the global chondrite values a deep negative Eu anomaly is apparent. This negative Eu anomaly is not a reflection of Eu mobility in the surficial environment, but rather a reflection of Eu2+ substitution for Ca2+ into crystal lattices as the Norra K?rr intrusion fractionated and crystallized. In fact, when using the special local normalization diagram, there is a slight positive  Eu anomaly in the 4acid near total digestion, which acts on the silicate fraction of the soil. This implies that there is more Eu than expected (relative to Sm and Gd content) in the silicate fraction of the soil relative to the underlying bedrock, so there must be some grains of zirconosilicate minerals, either catapleiite or eudialyte, from a more Eu-rich zone of the deposit that have been worked into the surficial environment above NK12-039.  This hypothesis was confirmed by examination of the soil samples using a scanning electron microscope (SEM) equipped with an energy dispersive X-ray spectrometer (EDS). Soil grains containing elements with large ionic radii, i.e. the high field strength elements (HFSEs), such as REEs, Zr, U or Th, were identified by their intense brightness, and then analysed with the EDS. The spectra obtained show that the HFSE grains contain both Zr and Y, but the REE content of the grains was too low to be detected. The HFSE grains are irregularly shaped, and range in size from fine silt (0.02 mm) to very fine sand (0.08 mm). Different sized grains are present in all soil horizons, so no conclusions about their provenance can be made, aside from that they are locally derived. An example of a medium silt-sized HFSE grain on a silicate soil particle is shown in Figure 3.11.    39REE  normalization to local bedrock at site NK12-039 H 2ONH 3AcNa4P2O 7w-NH 3OHs-NH 3OH4acidSUM totalA horizonB horizonRockA hor izonB hor izonR ockNK12-039: Sequential leach data normalized to global chondrite valuesREE Chondrite Normalized (Boynton, 1984)H 2ONH 3AcNa4P2O 7w-NH 3OHs-NH 3OH4acidSUM totalFigure 3.8a (above) - Sequential leach, four acid, and summed leach data of sur?icial materials and bedrock from site NK12-039, normalized to global chondrite values (Boynton, 1984). Note the positive Ce anomaly, LREE enrichment relative to HREEs, and negative Eu anomaly.Figure 3.8b (below) ? Sequential leach, four acid, and summed leach data of sur?icial materials and bedrock from site NK12-039, normalized to local bedrock. Note the same positive Ce anomaly and relative LREE enrichment, but also a weakly positive Eu anomaly in the 4acid near total.b)a)40NK12-060 NK12-039 NK12-072 NK12-060 NK12-039 NK12-072NK12-060 NK12-039 NK12-072NK12-060 NK12-039 NK12-072NK12-060 NK12-039 NK12-072NK12-060 NK12-039 NK12-072NK12-060 NK12-039 NK12-072A HorizonB HorizonC UpperC LowerAluminium CalciumIron PotassiumSodium ManganeseMagnesiumFigure 3.8c - Histograms of soil-forming elements, grouped by vertical profile site60005000200030000100020,00030,000010,00020,00030,000010,0001000000400020006000040002000040002000? Al (ppm)? Ca (ppm)? Fe (ppm)? Na (ppm)? Mg (ppm)? K (ppm)? Mn (ppm)41  The other important elements for characterizing soil profiles are Al, Ca, Fe, K, Mn, Mg, and Na. The B horizon at NK12-039 is slightly enriched in these elements, with the exception of Ca and Na, relative to the A horizon (Figure 3.8c). This could be because the A horizon has much more organic material, which could be adding Ca and Na to this horizon.    3.2.2ii ? Site NK12-072, the northern site The surficial material at NK12-072, as with the material at NK12-039, is a locally derived soil profile developed directly on mineralized bedrock. The summed total REE data in the soils are less than the REE content of the bedrock, so other components ?dilute? the REE signature, and may also dampen the Eu anomaly. These other inputs could be organic materials, or reworked and re-introduced glacial material. Two of the most conspicuous features of the bedrock-normalized REE plot at site NK12-072 are the negative Eu anomaly and positive Ce anomaly in the B horizon soils with the hydroxylamine leaches. The positive Ce anomaly is apparent in the ?total? leach as well, which is because the ?total? leach is the mathematical sum of all of the sequential leaches. The coincident positive Ce anomaly in the B horizon hydroxylamine leaches suggests that Ce may be enriched in this horizon, possibly existing as CePO4 or scavenged by Fe or Mn hydroxides.  The sums of the all the leached data for Al, Fe, K, Mn and Na are greater in the eluvial, or ?enriched?, B horizon than in the iluvial, or ?leached?, A horizon soils, while the sums of the leached data for Ca and Mg are greater in the A horizon. This indicates a greater proportion of carbonates in the A horizon (Figure 3.8c).  3.2.2iii ? Site NK12-060, the southern site The vertical profile at this site contains both A and B horizon soils, as well as glacial till originating from the nearby V?xj? granites. The different REE patterns in the till and the soils are most easily distinguished in the 4 acid digest (red lines, Figure 3.10b) because it is a near-total digestion that also attacks the silicate portion of the sample. The grey boxed line with the positive Eu anomaly is the granitic fragment (Figure 3.10a&b, labelled) in the vertical profile, and it has a markedly different REE abundance and pattern compared to the local bedrock. The upper and lower till samples (lower red lines) have a similar REE pattern to the angular granitic fragment, but with higher overall REE content, and a more subdued positive Eu anomaly. Moving up-profile, into the A and B horizons (highest red lines), the silicate fraction of the soil resembles the REE pattern of the underlying Norra K?rr bedrock, and has only a slight positive Eu anomaly that could be attributable to contamination from the V?xj? granites.     42A horizonB horizonRockREE  normalization to local bedrock at site NK12-072 H 2ONH 3AcNa4P2O 7w-NH 3OHs-NH 3OH4acidSUM totalFigure 3.9a (above) ? Sequential leach, four acid, and summed leach data of sur?icial materials and bedrock from site NK12-072, normalized to global chondrite values (Boynton, 1984). Note the strong negative Eu anomaly in all materials, including bedrock, and the positive Ce anomaly only in the hydroxylamine leaches (brown lines) of the B horizon. Also note the relative depletion of HREEs relative to LREEs.Figure 3.9b (below) ? Sequential leach, four acid, and summed leach data of sur?icial materials and bedrock from site NK12-072, normalized to local bedrock. Note the ??latness? of all leaches when normalized to local bedrock, with the exception the hydroxylamine leaches on the A horizon soil. Also note that the positive Ce anomaly in the hydroxylamine leached from the B horizon persists.a)b)A horizonB horizonRockREE Chondrite Normalized (Boynton, 1984)NK12-072: Sequential leach data normalized to global chondrite valuesH 2ONH 3AcNa4P2O 7w-NH 3OHs-NH 3OH4acidSUM total43  The process of soil formation starts at the bottom, and therefore the most evolved soils are those now at the surface. The last glacial maximum in this region was approximately 14,000 years ago (Lundqvist and Wohlfarth, 2001) and soil has been forming on top of the till veneer since then. The granitic fragment and tills represent the V?xj? granite chemistry, but since the tills were deposited on top of the Norra K?rr alkaline complex, they have been subjected to new surficial processes and are undergoing REE enrichment. This enrichment could be occurring in any one, or combination, of three ways: #1) Norra K?rr is a relative topographic high; mineralized colluvium from the deposit may have been transported such that it sits above the till, so REEs liberated via mechanical weathering (cd. Section 1.5.2) of REE-enriched silicate fragments (Figure 3.11) mix with the relatively REE-poor till.        #2) The vegetation at Norra K?rr has elevated REEs in its leafy tissue ( Bluemel et al., 2013) so decomposition of local foliage will liberate bio-available REEs, particularly the LREEs. #3) Humic and fulvic acids are abundant in the A and B horizons, and indeed any molecule with a carboxyl group will complex with trivalent lanthanides (Ln3+) and create stable REE compounds (Yamamoto et al., 2005), so through time REEs may be complexed and sequestered with organic acids, and eventually become incorporated into the till as soil horizons develop.        44NK12-060: Sequential leach data nor malized to global chondr ite values, including till and granitic cobblegraniticcobbleREE Chondrite Normalized (Boynton, 1984)A hor izonB hor izonC lowerC upperR ockH 2ONH 3AcNa4P2O 7w-NH 3OHs-NH 3OH4acidSUM totalREE  normalization to local bedrock at site NK12-060 B edrock nor malized plot of all sequential leach data from NK12-060, including till and granitic cobbleH 2ONH 3AcNa4P2O 7w-NH 3OHs-NH 3OH4acidSUM totalA hor izonB hor izonC lowerC upperR ockgraniticcobblea)b)Figure 3.10a (above) ? Sequential leach, four acid, and summed leach data of sur?icial materials and bedrock from site NK12-060, normalized to global chondrite values (Boynton, 1984). Note the appearance of a positive Eu anomaly in the samples derived from the granitic source.Figure 3.10b (below) ? Sequential leach, four acid, and summed leach data of sur?icial materials and bedrock from site NK12-060, normalized to local bedrock. Note the appearance of a positive Eu anomaly in the samples derived from the granitic source.45010203040506070809002040608010012014016000.511.522.500.511.522.533.544.5A HorizonB HorizonC UpperC LowerLanthanum CeriumEuropium LutetiumFigure 3.10c - Histograms of La and Ce (LREEs), Eu (MREE), and Lu (HREE), grouped by vertical profile site? Eu (ppm)? Lu (ppm)? La (ppm)? Ce (ppm)NK12-060 NK12-039 NK12-072 NK12-060 NK12-039 NK12-072NK12-060 NK12-039 NK12-072NK12-060 NK12-039 NK12-072Figure 3.11 - Example photo of a HFSE grain on a silicate soil particle. The HFSE grainis approximately 25um long (medium silt sized)46  3.2.3 Sequential Leach Discussion The greater than expected level of REEs in the Na4P2O7 leach, considering it is one of the weaker leaches, indicates that the REEs are being scavenged by organic acids and sequestered within the organic fraction of the soil. The affinity of REEs to form complexes with organic components increases with increasing pH (Lippold et al., 2007) because the trivalent lanthanide (Ln3+) binds to the carboxylic and phenolic functional groups of organic acids and therefore is dependent on deprotonation of these groups (Pourret et al., 2007). Yamamoto (2005) corroborated these findings and found that any molecule with a carboxyl group, even a simple molecule like acetate, could be analogous to humic and fulvic acids. Pourret et al. (2007) found that the degree of binding between humic acids and Ln3+ was also dependent on the concentration of humic acids (HAs) in the solution, and that at low concentrations of HA (5mg/L) all of the available Ln 3+ would be bound at neutral pH, while at higher HA concentrations (20mg/L) there would be complete Ln 3+ - HA complexation beginning at lower pH, as low as pH = 4.5.    The positive Ce anomaly in the hydroxylamine leaches indicates that Ce is being co-precipitated with Fe and Mn hydroxides, which are therefore acting as a Ce sink (Hem, 1978). The positive Ce anomaly is more obvious in the B horizon than in the A horizon, which corresponds with the higher Fe content in the B horizon than in the A horizon (Figure 3.12).  These findings are significant because they demonstrate two things: #1) there are definitely REEs in solution at Norra K?rr, and they are being hydromorphically dispersed, which leads to the question of the distance travelled by the REEs and their mechanism of dispersion. #2) There is a high proportion of REE content in the sodium pyrophosphate leach, compared to the other ?stronger? leaches, therefore we can conclude that a large proportion of the REEs are binding with either humic or fulvic acids and scavenged with organic complexes. This is corroborated by work done by Shan, 2002; Pourret et al. (2007); and Marsac et al. (2012).    Figure 3.12 ? Differences in Fe content in the Ah and B horizon soil samples collected from the same location 47  3.3 Fern Chemistry Complete fern plants were collected in August 2012 from locations that were selected based on high REE content in fern leaves the previous summer (see Section 5.3.4). The fern fronds and root bulbs were separated in the field (see Section 2.2) and analysed in three parts; roots, stems, and leaves, in order to establish where the fern is sequestering the REEs. This also helps to determine which part of the fern is the most effective sample media for biogeochemical exploration.   Rare earth element content in fern plants generally increases from stem < leaf < root. These findings agree with those of Fu et al. (1998), who studied spatial REE variation in the fern species Matteuccia  in order to determine the provenance of the REEs in the fern tissue. Fu et al. (1998) hypothesized that REEs had once been in solution, were originally derived from silicate particles in the soil, and were finally being stored in the cortex of the rootlet. This supports the findings at Norra K?rr, where the highest HREE content is in the root, but the highest LREE content in the leafy tissue. This will be addressed in more detail in Section 3.3.3.  There is a paucity of information regarding the limits upper threshold of REEs possible in a living plant before physiological processes are inhibited. It is likely that the presence of REEs in the natural system will have a greater effect on the soil micro-flora than on the plant physiology, therefore, which in turn will contribute to the cycling of major nutrients in the soil (Jenkinson and Ladd, 1981) particularly in the case of Gd3+, which is readily absorbed onto various microorganisms (Andres et al. 2000) The three fern plants displayed similar REE distributions within the different tissue types, even though two of the fern plants were Dryopteris filix-mas and the other was Athyrium filix-femina. Generally the Athyrium filix-femina had lower levels of HREEs, and in the stem tissue it had elevated levels of LREEs compared to the Dryopteris filix-mas samples. Of particular note is that the leaf and stem tissues are more strongly fractionated with respect to the LREEs than the root tissue (Figure 3.13). This implies that there is some fractionation mechanism within the fern plant that preferentially transports LREEs over HREEs, likely due to their different ionic radii.  There is a well documented phenomenon known as the ?lanthanide contraction?, whereby the ionic radii of the lanthanides become smaller with increasing atomic number, due to shielding effects of the 4f electrons. This is significant because the light and middle REEs have ionic radii similar to that of the Ca and P ions, which could explain in part why the LREEs appear to travel more freely than the HREEs within the biological system.  48Stem LeafR ootREE Chondrite Normalized (Boynton, 1984)REE Chondrite Normalized (Boynton, 1984)Figure 3.13 (above) ? Chondrite normalized plots of leaf, root, and stem tissue from all three vertical pro?ile sites. Note the decrease in REE content from stem < leaf < root.Figure 3.14 (below) ? Chondrite normalized plots of leaf, root, and stem tissue from ferns at all three vertical pro?ile site, with lithogeochemical data added for referenceStem LeafR ootR ock49   Table 3.5 ? REE content in leaf, root, and stem tissue at each vertical profile site  3.3.1 Root The REE content in the root samples of each fern plant is listed in Table ?.5, and the ?REE content ranges from 58.6 ppm to 183.94 ppm. The root samples have higher HREE values than the stem or leaf at all sites, but lower LREE values, particularly with respect to La and Ce. The northernmost site (NK12-072) is located directly over HREE mineralized outcrop, and the root tissue collected here actually displays lower  LREE content than the corresponding leaf tissue, but higher HREE content (Figure 3.13 and 3.14). This indicates the LREEs are more mobile within the plant than the HREEs. The root sample from the southernmost site (NK12-060) displays a negative Ce anomaly, corresponding with the soils from that site (see Section 3.2.2iii).  There is a small degree of LREE enrichment in the root samples compared to the rock samples, but generally the slope of the root REE line approximates the slope of the REE line from the bedrock (Figure 3.14).   3.3.2 Stem  The stem samples from each site, and for each fern species, contain much lower levels of all REEs than the corresponding root or leaf tissue from the same fern (Figure 3.13). This could be attributable to the mechanism by which plants protect themselves from high concentrations of harmful elements by preferentially sequestering the harmful elements in tissues that are non-essential for metabolic function. It is also possible that there is a ?threshold? REE content in fern stems. There may be a finite number of sites within the cell wall that a REE can occupy, and once all 50  of these sites are filled, the REEs must travel further through the plant system before they can be deposited.  The most notable feature of the chondrite normalized plot displaying the REE content in the stems is the enrichment of LREEs, especially compared to the roots and rocks, and the stem tissue is depleted in all REEs heavier than Sm (Figure 3.14).  3.3.3 Leaf The leaf tissue has intermediate HREE values between the stems and the roots. Like the stems, the leaves display LREE enrichment, relative to the roots, and the degree of enrichment is approximately equivalent between the stems and the leaves, meaning that the slope of the REE chondrite normalized line from La to Eu is approximately the same for the stem and leaf samples, whereas the slope of the chondrite normalized line for the root tissues is much less steep. One root tissue sample has comparable HREE levels to the leaf tissue, but generally the leaf tissues have fewer HREEs and more LREEs than their corresponding root counterparts.  The leaves also show a wide range of variability for the heaviest and lightest REEs compared to the variability observed for the MREEs. The REE content in the leaf tissue ranges from 19.7 ppm to 56.47 ppm for La, and 0.02 ppm to 0.12 ppm for Lu, but for Eu the variation is only 0.26 ppm to 0.34 ppm. The standard deviation for La is 18.46, while one standard deviation for Eu is only 0.04 ppm, but since this is based on three samples, the statistics are not robust enough for a detailed analysis. The Lu content is quite low (0.05 ppm) but this is not surprising because levels of Lu are quite low in the continental crust (Lu = 0.0381 ppm, Taylor and McLennan (1995)). Both the leaf and the stem tissue exhibit similar REE patterns; they are both highly enriched in the LREEs, enriched in the MREEs, and the depleted in HREEs with the exception, of course, of Eu.   3.3.4 Fern Chemistry and Sequential Leach Data The most striking feature of the sequential leach paired with the vegetation data is that the chondrite normalized patterns and REE magnitude in the fern leaf and root are nearly equivalent to the sodium pyrophosphate selective leach on the B and A horizons. Since sodium pyrophosphate extracts elements complexed with organic acids, such as humic or fulvic acid, the comparability of the REE content in the roots and leaves of the ferns suggests that the ?selective leach? performed by the roots on the soil is analogous to the action of sodium pyrophosphate.  This corollary is not visible in the upper or lower C samples because organic acids are not abundant in this horizon.  The transfer factors from root tissue to leaf tissue range from 0.4, in the case of the LREEs from site NK12-072, to 21.5, in the case of the HREEs from site NK12-060 (Table 3.6). In all three vertical profiles, the transfer factor for the HREEs is greater than the transfer factors for the MREEs or the LREEs, meaning the HREE concentration in the root is much greater than the concentration of HREE in the leaf. This difference of transfer factor is likely a function of increased mobility of the LREEs within the plant. 51   Table 3.6 ? REE transfer factors from root to stem for all three vertical sample sites 3.3.5  Summary #1) Norra K?rr is a relative topographic high, so any mineralized colluvium from the deposit will sit above the till and erode to liberate REEs (see Chapter 1.5.2), and REE-enriched silicate fragments will mix with the relatively REE-poor till. #2) The vegetation at Norra K?rr has elevated amounts of REEs in its leafy tissue (Bluemel, 2013) so decomposition of local foliage will liberate bio-available REEs, particularly the LREEs. #3) Humic and fulvic acids are more abundant in the A and B horizons than the underlying till. These organic compounds, and indeed any molecule with a carboxyl group, will complex with trivalent lanthanides (Ln3+) and create stable compounds (Yamamoto et al., 2005), so through time REEs may be complexed and sequestered with organic acids, and eventually become incorporated into the till as soil horizons develop.  3.4 Conclusions It appears that REEs are being mobilized by the processes of weathering and breakdown of eudialyte crystals from the bedrock at Norra K?rr and are being transported via colluvium, in solution, and recycled with organic material throughout the surficial environment. Samples collected vertically from bedrock through the developed soil horizons and into the overlying ferns show similar REE patterns to the underlying bedrock, with the exception of the leafy parts.  Elemental fractionation within the ferns, between the root and the stem, preferentially transports LREEs rather than HREEs into the leafy tissue.  This likely occurs because the LREEs have similarly sized ionic radii as the cations of essential macronutrients, such as Ca, Mg, P and K (e.g. the ionic radius of Nd3+ is 0.100 ?, whereas the ionic radius of Ca 2+ is 0.099 ?).  52  The Norra K?rr REE deposit is buried by transported Quaternary sediments, upon which soil horizons have been developing with elemental input from the REE-enriched bedrock. The glacial till and Norra K?rr bedrock have different REE signatures emphasizing that the till was derived from the regional granitoids. However, the upper soil horizons, A and B, are both enriched in REEs relative to their parent C horizon, demonstrating that REEs originating from the bedrock have been mobilized and incorporated into the upper soil horizons. The A and B horizons have higher organic matter contents, and are therefore better trap sites because REEs can be sequestered through the formation of complexes with humic and fulvic acids.                                            53  Chapter 4 - Seasonal and Species Variation   4.1 Species Variation  The ability of a plant to incorporate REEs into its living tissue is a function of its morphology and evolutionary adaptations. The more highly evolved a plant is, the more ?barrier mechanisms? (Kovalevsky, 1987) it will have which allow it to selectively uptake some elements over others through its root system. Ten different plant species were sampled in June 2011 to determine which species could accumulate the highest levels of REEs in its tissue. The Latin and common name of each species collected, as well as the median values of  all REEs, are presented in Table 4.1. Comparisons of elemental variation in the different plant species at Norra K?rr have been made using data with the below detection limit values replaced with positive half of the detection limit (see Section 2.6) because, with the exception of the fern species, the ashed Norway spruce twigs, and the feather moss, all samples either fall below the certified REE detection limit, or hover close to it, and only the fern data appear to be reliable below the detection limit.    Table 4.1 ? Common and Latin names of species collected from Norra K?rr, including number of each species and tissue sampled.  4.1.1 REE Variation in All Species The median values of the sums of the REEs, HREEs, and LREEs have been compared instead of the absolute values to highlight trends in the LREEs relative to the REEs. The ?LREEs show very similar patterns to the ?REEs, because the LREEs are present in higher abundances than the REEs, so the ?REEs is more strongly influenced by the total amounts of LREEs.  54  The highest levels of REEs are found in the fern species Athyrium filix-femina (median ?REE 56.21 ppm). There are higher levels of REEs present in the single moss mat sampled (?REE = 136.52 ppm), but since it still had its roots attached, it is extremely likely that there was contamination by soil particulates adhering to the rootlets. Lower levels of REEs, median ?REE 23.36 ppm, are found in the fern species Dryopteris filix-mas. The third highest levels of REEs are observed in the fern species Pteridium aquilinum (median ?REE 8.66 ppm), which has approximately the same REE content as the ashed twigs of the Norway Spruce (median ?REE 10.4? ppm). The levels of LREEs are slightly higher in Alnus than in Fraxinus , Quercus , and Corylus . There is no visible fractionation among the HREEs because the abundances have dropped well below the reportable detection limits. Betula  has the lowest levels of REEs in its tissue (median ?REE 0.17 ppm).  The ?LREEs and ?REEs show similar patterns at high REE concentrations; the fern species Athyrium filix-femina and Dryopteris filix-mas have the highest REE content for the ?REEs and ?LREEs, however, ashed Norway Spruce twigs have comparable ?REE content to Dryopteris filix-mas. The formula used to calculate elemental dry weight equivalencies from ashed analytical data is given in Equation 4.1.  At this point it is important to reiterate that reducing any organic material to ash, and analysing the ash will give results higher than analysing the dry tissue. In these experiments the ashed Norway spruce twigs returned REE values tens to hundreds of times greater than the calculated dry weight  values, with only Y, La, and Ce showing REE content above the certified detection limit. With this in mind, considering the goal of this research is to identify the most effective biogeochemical exploration tool for REE mineralization, the fern species Athyrium filix-femina and Dryopteris filix-mas remain the most suitable candidates.    Equation 4.1 ? Calculation of elemental concentration in dry weight of plant material based  on elemental concentrations in ashed material and ash yield  (Dunn, 2007).  4.1.2 Essential Nutrient Variation in All Species The essential macronutrients are grouped as K, Ca, Mg, P and S by Taiz and Zeiger (2010). Each macronutrient is used by the plant for a different metabolic function; K, Ca, and Mg are nutrients that remain in ionic form, K being a principle cation in establishing cell turgor and maintaining cell electroneutrality, Ca being a constituent in the middle lamellae of cell walls, and Mg being a constituent of the chlorophyll molecule. Phosphorus is a component of sugar phosphates and nucleic acids, and has a key role in reactions that involve adenosine triphosphate (ATP), and S is a component of several amino acids.  All of the macronutrients display similar behaviour in each of the species sampled, with a few interesting exceptions (Table 4.2). The Ca levels in Hazelnut, average 1.4%,  are considerably 55  higher than all of the other species, which range in average Ca content from 0.2% to 1.0% (Figure 4.1). Of particular note is that the 3 fern species have lower Ca levels than the other species sampled, but the fern species have high REE content relative to the other plants, so there appears to be an inverse relationship between Ca content and REE content.    Table 4.2 - Average macronutrient and micronutrient content in each species, compared to standard tissue levels of essential elements required by most plants^ (Epstein, 1999). *All samples are leafy tissue with the exception of Norway Spruce, which is twig tissue, which has been reduced to ash and back calculated to represent dry tissue element concentration.                56K (pct) Ca (pct)20.511Alnus glutinosa (alder)Fraxinus excelsior (ash)Betula pendula (birch)Athyrium filix-femina (lady fern)Dryopteris filix-mas (wood fern)Pteridium aquilinum (bracken)Corylus avellana (hazelnut)Picea abies (Norway spruce, dry weight)Quercus robur (oak)Figure 4.1 - Box and whisker plots of K (left) and Ca (right) for all sampled species. The hazelnut has much higher Ca content than any other plant, and all 3 fern species have more K than CaMedian Mean75th percentile25th percentileOutliers75th percentile + (1.5 x InterQuartileRange)Far outliers57  The opposite is true of potassium, because the three fern species all have high levels of K compared with the other plants (range of average K content in ferns: 1.8%-2.1%, range of average K content in other plants: 0.5%-1.2%). This is particularly interesting because there is a correlation between K and the REEs, and it is especially strong for the LREEs. This correlation is particularly visible once the Norway Spruce data has been recalculated to represent the dry weight values of the essential macronutrients. The concentration of essential macronutrients in the ashed Norway Spruce data is many times higher than the macronutrient content in the dry tissue of the other plants, so the subtle REE-K correlation is swamped by the strong K-Ca-Mg-P-S correlation (Figure 4.2). Box and whisker plots of K and Ca data are displayed in Figure 4.1.   Figure 4.2 - ?REEs as a function of potassium for all species, with Norway recalculated to a dry weight basis  There is also a correlation between the majority of the REEs and Al, as well as a correlation between Na-Al, and Na-Eu. The mineralization is hosted by a peralkaline nepheline syenite, so it?s not surprising that there is a correlation between Na and the REEs because, by definition, peralkaline rocks have excess Na and K relative to Al for the formation of feldspars (Section 1.3.2). The deposit is surrounded by granitic rocks, so the majority of the bedrock in the area is comprised of aluminosilicates, so it is not surprising to see elevated Al as well as Na in the surficial environment.    58  4.2.1 REE Variation in Fern Species The three species of ferns have significantly different chemical compositions. Pteridium aquilinum has the lowest REE values; Athyrium filix-femina has the highest REE values, and Dryopteris filix-mas has intermediate values between the other two species (Figure 4.3). Similar behaviour and mobility within the HREEs (Tb, Dy, Ho, Er, Tm, Yb, Lu) the LREEs (La, Ce, Nd) and the MREEs (Sm, Eu, and Gd), is well documented (see Section 1.6), and is observable in the HREE and LREE content in the ferns. All three fern species fractionate the LREEs preferentially, and they all exhibit a subtle negative Eu anomaly. The Eu anomaly is less apparent for Pteridium aquilinum because that species has low HREE content, and the data are more scattered on the heavy side of the chondrite normalized plot due to poor data quality nearing the detection limit (Figure 4.3).   4.3 Seasonal Variation 4.3.1 REE Variation      In August 2011 an assortment of media was collected to compare the seasonal variation at the beginning and the end of a growing season. Among the media were seven ferns, six of which were Dryopteris filix-mas. The other was Pteridium aquilinum. These six Dryopteris filix-mas samples were compared to the Dryopteris filix-mas samples collected at the beginning of the summer to establish whether there was any elemental variation in the sampled ferns that could be attributed to seasonal variation.   The median REE values for Dryopteris filix-mas samples collected at the end of the summer in 2011 are much higher than the median REE values  from the spring of that year (Table 4.3). This increase in REE content can be attributed to their longer growing season, whereby more REEs were able to accumulate in the leafy tissue through the course of the summer because they had more time and greater influx of solutes. Dryopteris filix-mas is a deciduous fern, so immature leaves will not have as much accumulation of elements as mature leaves. The Dy content in the Dryopteris filix-mas leaves is 15x higher at the end of the summer than in June (Table 4.3).   Table 4.3 - REE seasonal variation in Dryopteris filix-mas 59REE Chondrite normalized plot, after Boynton 1984Athyrium lix-feminaDryopteris lix-masPteridium aquilinumREE Chondrite normalized plot, after Boynton 1984August 2011June 2011Figure 4.3 - Chondrite normalized plot of three fern species Dryopteris ?ilix-mas, Athyrium ?ilix-femina, and Pteridium aquilinum. Note relative enrichment of LREEs in Athyrium ?ilix-femina.Figure 4.4 - REE content of Dryopteris ?ilix-mas  samples collected from the same location in June and August, coloured by season.60  Samples obtained from the same species at same location, analyzed at the same lab, but collected in different years were also compared. In the instance of NK11-054 compared with NK12-034, which are sample sites are roughly 10 m apart, the REE values doubled, for every element but Tm and Lu, in the Athyrium filix-femina sampled at the end of the summer relative to the one sampled in June. Rare Earth Element content is higher for every sample collected at the end of August, with the exception of NK11-031 (Figure 4.4). Since this site is near the end of one of the sample lines, it has much lower REE values than the other samples which were collected near the centre of the deposit. The inverse is true for site NK11-072; this sample site is directly above a catapleiite-bearing outcrop, and here samples taken in June are extremely high for all REEs. In fact, the REE content of the sample from site NK11-072 has higher REE content, with the exception of La and Ce, than all of the samples taken in August. This is obviously due to its extreme proximity to highly mineralized outcrop. The samples taken from the same site in August also have high REE content, almost identical to the REE content of the sample collected here in June, so while the majority of samples exhibit an increase in REE content throughout the growing season, any ferns directly associated with mineralization will have an ?upper threshold? of REE tolerance before the REE concentration becomes adverse to the plant?s physiology.  4.3.2 Essential Nutrient Variation  4.3.2 i ? Macronutrients ? Ca P K Mg S The levels of macronutrients in the leafy fern tissue at the beginning and the end of the summer are similar, with the exception of  Ca and P, which behave inversely. There is more Ca in the leaves at the end of the summer, but there is more P in the leaves at the beginning of the summer. This could be attributable to the increased influx of solutes in the spring, corresponding with the most rapid growth times. Phosphorus has a key role in all biochemical reactions involving ATP (Taiz and Zeiger, 2010), so when the plant is growing and photosynthesizing rapidly, there will be more ATP reactions occurring and therefore higher levels of P. The increased Ca levels in August can likely be attributable to the Ca required in the cell walls. Since the plants are larger at the end of the summer, cell walls in leaves might be thicker because they?ve been growing longer. The distribution of the essential macronutrients in fern leaves is displayed in Table 4.2. Tissue levels of macronutrients required by most plants (Epstein, 1999) are shown in italics.  4.3.2 ii ? Micronutrients -  Fe B  Na Mn Cu The levels of micronutrients within the ferns sampled at the beginning and the end of the summer are generally similar, with the exception of B and Cu. Boron was shown to be correlative with the REEs (Section 5.2) so it follows that whatever part of the leafy tissue is steadily becoming enriched in REEs throughout the growing season is also becoming enrich in B.  Copper exhibits a much larger range of values at the beginning of the summer, suggesting that some ferns sampled in June were able to accumulate more Cu in their leafy tissue than others. This could be a function of varying Cu content in soils at the specific locations, because the ferns that had high Cu in June also had relatively high Cu in August. The highest Cu values were in 61  Dryopteris filix-mas samples from NK11-031, with greater than 9 ppm Cu in June. Samples taken from the same plant in August had 6.53 ppm Cu. The second pair of samples elevated with respect to Cu were from NK11-039N, and they had a much tighter range of Cu values (6.53 ppm to 6.68 ppm). The site with high Cu values in June, NK11-031, is located at the western end of one of the sample lines. Elevated Cu here could be attributed to availability of organic matter; Howell and Gawthorne (1987) reported the importance of humic and fulvic acid for the solubility of copper. The difference in Cu content is not likely attributable to changing moisture content because none of the other samples showed the same variations, and all of them underwent the same changes in season and therefore available soil moisture. Manganese content in all of the species sampled at Norra K?rr is much higher than the standard tissue levels of essential elements required by most plants (Table 4.2; Epstein, 1999).  Oak leaves were collected from the same site, NK11-042, in June and August and for each REE, including Y and Zr, the concentration was 30 to 40x higher in the sample collected at the end of the summer. Some of the essential nutrients, such as Ca, Mg, K and B were present in lower levels at the end of the summer, while other essential nutrients, such as S and P, were higher in the samples collected in June. In general, the patterns displayed by the elemental variation in oak leaves corroborate the patterns in the fern fronds, in that the concentration of many essential nutrients, as well as the REEs, is elevated in August.   4.4 Implications for Exploration The rare earth elements, as well as many other essential micronutrients and macronutrients, are elevated in leafy plant tissue at the end of the growing season, for a variety of species. This demonstrates that the most beneficial time to sample is at the end of the growing season, because the elevated levels of REEs in the plant tissue will make chemical analysis more robust, and the highest contrast in HREEs occurred in the plants sampled in August.                  62  Chapter 5 - Biogeochemical Distribution of Rare Earth Elements  in the Surficial Environment at the Norra K?rr Alkaline Complex  Biogeochemical exploration is effective in areas where bedrock is under cover because it takes advantage of the mechanisms of natural elemental mobility within the surficial environment. Buried mineral deposits can have a larger geochemical footprint than the physical extent of the mineralized bedrock because certain labile, or ?pathfinder?, elements can be mobilized beyond the boundaries of the mineralized bedrock, thus producing a larger halo in the surficial environment for the explorer (Rose et al., 1979). The key to biogeochemical exploration, and indeed all types of geochemical exploration, is determining what elemental geochemical thresholds define ?background? in an area, so elements introduced or removed by the mineralizing process will be anomalous relative to that pre-defined threshold (Sinclair, 1974).  To establish background and anomalous REE levels in the surficial environment of southern Sweden, a suite of plant and soil samples were collected. The REE signature in the surficial environment at Norra K?rr is ?anomalous? due to proximity of the mineralized alkaline complex, compared to the ?background? REE signature in the surficial environment, which results from unmineralized country rock.   Biogeochemical samples (n = 24) were collected on a regional scale, from an area of about 750 km2, as well as on a local scale, from a 1.5 km2 area directly around the deposit, with the aim of defining biogeochemical background in this geological setting (Figure 5.1). The Norra K?rr Alkaline Complex (NKAC) has a physical footprint of 0.4 km2.  5.1 Regional REE Signatures 5.1.1 REE Content in the Fern Species Dryopteris filix-mas and Athyrium filix-femina To establish the background concentration of REEs in ferns in the Swedish boreal environment around Norra K?rr, the fern species Athyrium filix-femina (n = 17) and Dryopteris filix- mas (n = 7) were collected regionally from a large area around the deposit (Figure 5.1a).  These samples were collected using the same procedure as all other fern samples. Analyses were performed at the ALS facility in Lule?, northern Sweden (see Section 2.4.3).  Regionally, the ?REE values in the combined Athyrium filix-femina and Dryopteris filix-mas populations range from 3.7 to 134.4 ppm, with median ?REE values of 57.? and 24.  ? ppm, respectively. Athyrium filix-femina has two to three times higher median REE contents than Dryopteris filix-mas, which is consistent with the total abundances observed in the respective ferns directly associated with the NKAC (see Section 4.1.1). The REE concentrations, grouped by species, are listed in Table 5.1.     63!!! !! !!!! !!! !!! ! ! !!!!! ! !!!!! ! !! ! !!! !!! !! !!!! ! !!! ! !! ! !! !!! ! ! !! !!! !! ! !!!!!! ! !! !! ! !!! ! !! ! !!! ! !!! !! !!! ! !!! ! !! !!!! ! !! !! ! !! ! ! !!! ! !!! !! ! !!! !!! ! !!! ! ! !!! !!! !!! !! ! !!! ! ! !! !! !!!! ! !! !! !! ! !! ! ! !! ! !! !! !!! !!!! !!! !! !! ! !!!! !! ! !!! ! !! ! !!! !!!!! !!!!! ! !!! ! !! !!! !!! !!! ! !! !! !!!! !! !!! !! ! !! !! !! !! !! ! !! !! !! !! ! !!! ! !! !!!!! !! !!!!!!! !! !!!!! ! !! !!!! ! !! !! !! !!!! !! !! !! !!!! ! !! !!!! !!!! !! !! !!!! !! ! !! !!!! !! ! ! !! ! ! !! ! !!!! !!! !!! !! !! ! !!!! !! !! ! !!!! ! !!! !!! !!! !!!! !! !! !!! !! !! !!! !!!! ! !! !! !!! ! !!! ! ! !!!! !! !! !! !!!! !! ! !!!! !!! !! ! !!!! !! !!! ! !!! ! !!! !!! !! !! !!!! !! ! !!! ! !! !! !!!!!! !! !!! !!!!! !! !!! ! ! ! !! ! ! ! ! !! !! ! !!! ! !! !! ! !!!! !! ! ! ! !!! !!! !!!! !! !!! ! !! !! !!! !! !! !! !! !!!!!! !! ! !!!! !! !!! !! ! !!!! !! ! ! !!! !! !! !! !!! !! !!! ! ! ! !!!! !!! !!!!! !! !!! !!! ! !! !! ! !! !! !!! !! !! ! !!! !! ! !! ! !! !! !!!!! !! ! !! !!!!! !! !!! !!!! !! ! ! !!!!! !!! !! ! ! !! !!!! !!! ! !! ! ! !!!! !!! !! ! !! ! !! !! ! !! !!! !!!! !!! !!! ! ! !! ! !! !! ! !!! !!! !! ! ! !! ! !! !! !! !! !! !!! ! !!! !! ! !! !! ! ! !! !!! ! !! !! !! !!! !! !! ! !!!! !! !!! !! ! ! !! !!!!!!! !! ! !! !! !!! !!! !!!!!! ! !! !!!! ! !! !! !! !!! ! !!!! !! !! !!!!! !!!!!!! !! !!! !! ! ! !!!! ! !!!! ! !!!! ! !!! ! !! !!! !! !! !!!!!!!!!!! !!!!!! ! !!! !! !! !! ! !!! !!! !!! !! !!! !!! ! ! !! ! ! !!! !!!! ! !!!! !! !!! ! !!!! ! !! !! ! !!!!! ! !!! !!! !!! !! ! !! ! !!!! !! !!!!! !! !! ! !!! !! ! !! !! ! !!! !! !!!! ! !! !!! !!!! !! !! !! !!! ! !!! !! !! ! !!!!! !! !! !!! !!!! ! ! ! !!! !! !! !! ! !! ! !!!!! !!!! ! !! ! !!! !! !! ! !! ! ! !! !!!! ! !!!!! !! !!! ! ! !! !! !!!! !! !! !!!! ! !! ! !! !! ! ! !!!!! ! ! !!! !! ! ! !! !!!! !!!!! !!!! !!! !! !! ! ! !! ! !!!!! ! ! !! !!! !! ! !!! ! ! ! !! !!! !!!!!! ! ! !!!! ! !! !! ! !! !! ! !!! ! !!!! ! !!! ! !!!! ! ! !! !!!! !!! !! !!! !! !!! !!! !! ! !!!! !!! !! ! ! !! !!! !! ! !!! ! !!!! ! !! !!!! !! !! !!! !! !!! ! ! !!! ! ! !! !!! !!!! !! !! !! !!! ! ! !!!! ! ! !!! !! !! !!! ! !!! !!!!! !! !! !! !!! ! !!!! !! ! !!!! ! !!! !! !!!! !! !!! ! !! !! !! !!! !!! !!!! !!! ! !! ! !!! ! !!! ! !!!!!! !! !!! !! !! ! ! !! !! ! !!! ! !!! ! ! ! !! !!! ! !! !!! ! !!! !!! !! !! ! !!!!! !!! !!!! !! ! !!! !! !! !! !!! !!!!! !! !! !! !!!!!! !!! !! !! !! !! !! !! ! ! ! !!! !!! !! !!! ! !! !!! !! !!! !!!!!! !! !!! ! !! ! !! !! ! ! !! ! !! ! !! !! !!! !! !!!! !!! !! !! !!! ! !! ! !!! !! !!! !!! !! ! !! !!! !! !! !! ! !! !!! !! !!! ! ! !!! ! !!! !! ! ! !!!! !!! ! !!! ! !!! !!!! !!!!! !! !! !! ! ! !!! ! !! !! !! !! ! !! !! ! !!!! ! ! !! !! ! !! ! !!! !! !! !!! ! !! ! ! !! ! !! ! !!! ! !!! ! !! !! !!! ! ! !!! !! ! !!!! ! !! !! !!! !! !! !! !! !! !!! !!! !! !! !! !!!! !!! !!! !! !!! !!! ! ! !!! !! ! !! !!! !! !! !!! !! !!! ! !! ! ! !!!! ! !! !!! ! !! ! !!! ! !!! !!! !!! ! !! !! ! !!! !! !! !! !! !! !! ! !!! !!! ! ! !!!! ! ! !!! !! !!!! !! ! ! !! ! !! !! ! ! !!! !! !!!! !! !! !!! !! ! ! !! !! !! !! !!!!! !! ! !!!! !! ! !!!! ! ! !!!! ! !! !!!! !! ! !! ! !! ! !!! ! !!!! !!! ! !! ! !! !! !!! !! ! !!! ! ! !! ! !! !! ! ! !!!! ! ! !! ! ! ! !!! !!! !! !! !! ! !!! ! !!!! !!! !! !!!!!!! ! ! !!!! !! ! !!!! !! !! ! !! !! ! ! !! !!!!! ! !!!! !!!! !!! ! !!! ! !! !!!!!410000 440000 470000 500000 530000636000063900006420000645000064800006510000!!! !!! !!!!! !!!! !!!!!!!!!! !!!! !! !!!! ! !! !!!!! !!! ! !! !!! ! !!!!!!!!!!! !! !!! !! !!!!! !!!! !!! !!! !!! !!!!!!!!!! ! ! !! ! !!!! ! !!!!!!! !!! ! !! ! ! !!!! !!! !! !! ! ! !!! ! !!!! !! !!!! !!!! !!! !! !! !!! ! !!! !! ! !!!! !!! !!!! ! ! !!! !!!!!! !! ! ! !!! !! ! ! !!!! !!! ! !!!!!!!!! !! ! !!!! ! ! !!!! ! !!! ! !!! !! ! ! !! ! !!! !!!! !!!!!! ! !!! ! !!!! !!! !!! ! !!!! ! !! ! ! !!! !!!!! !!! !! !!!!464000 472000 480000 488000 496000 50400064230006432000644100064500006459000Norra K?rr Alkaline ComplexGr? nna?Lake V?tternBasaltGabbro, dioriteamphibolite, doleriteGranite, syenite, monzoniteGranitoidGranitoid, tonalite,granodioriteGreywacke, quartzite,mica schist, phylliteLimestoneRhyoliteSandstoneShaleAcid and intermediatevolcanic rocksConglomerate, sandstonequartziteLimestone, sandstone,mudstoneNorra K?rr Alkaline ComplexGr? nna?Lake V?ttern10km!LegendIce flowindicatorsRegional Sample SitesLakes50kmN22050727284011202513381426374101 0243030436 1619Figure 5.1 - Geological maps of the Norra Karr area (the Alkaline Complex is marked with a yellow star),with ice ?low indicators (black arrows) and regional sample station locations (green circles). The local samples (yellow circles, Figure 1.5b, cf Section 1.3.3) cover 1.5 km2, the regional samples (green circles) cover almost 750 km2, and the surface trace of the deposit is 0.4 km2  64    Table 5.1 ? Ranges of REE content in regional fern samples, grouped by species, analysed at ALS in Lule?, using a microwave-aided nitric-acid digestion with an ICP-MS instrumental finish. The cells highlighted red are below reportable detection limit. The minimum and maximum values obtained from analyses of regional fern material can be used to approximate the range of biogeochemical ?background? in the surficial environment above unmineralized country rock. The background range is displayed graphically in Figure 5.6, and the minimum and maximum REE values are estimates of the regional background REE content.  5.1.2 REE Content in Regional Soils The surficial environment in southern Sweden has been strongly influenced by glaciation. Bedrock outcrops in many places, particularly at topographic highs, whereas lower elevations are typically filled with till and glaciofluvial materials. Residual soils have developed on top of bedrock and till alike in the past 14,000 years, since the last glacial maximum in southern Sweden (see Section 1.?.?), and the surficial materials have been mapped as ?thin or discontinuous soil cover? by the SGU (Figure 1.5). Regional soil samples were collected from 21 sites in southern Sweden to determine the natural REE contents in the Swedish boreal forest (Figure 5.1). Soil samples were collected from the B (n = 17) and C (n = 4) horizons. The median ?REE value for the B horizon is 128.14 ppm, while the ?REE value for the C horizon ranges from 104.14 ppm to 4?5.71 ppm.  The REE patterns in the regional, or ?background?, soil samples from both the C and B horizons reflect the underlying geology (Figure 5.2). The NKAC is wholly surrounded by V?xj? granites, which are part of the Trans-Scandinavian Igneous Belt (TIB, see Chapter 1.3.1, H?gdahl et al., 2004). The TIB has slightly lower LREE content and slightly higher HREE content than average granites (Krauskopf and Bird, 1967), and the B and C horizon samples have the same chondrite-65  normalized REE patterns. Mineralized nepheline syenites from the NKAC (dark grey points) have many times more REE content than either the  TIB (yellow points) or V?xj? granites (red points), but the REE pattern, normalized to average granite, is consistent (Figure 5.2). This indicates that the NKAC, TIB, and V?xj? granites share a magmatic source and fractionation history, but since Norra K?rr is slightly younger (the alkaline complex is 1.545?61Ga, the TIB is 1.86-1.65Ga), it is more fractionated, and therefore more enriched in REEs.    Table 5.2 ? Ranges of REE content and Y in regional soils, grouped by horizon. *Average Granite composition from Krauskopf 1??7, average V?xj? granite composition (?V?xj??) from Appelquist 2010, and average Trans-Scandinavian Igneous Belt (?TIB?) composition from Christensson 201?.           66Regional C horizonRegional B horizonAverage TIB (n = 11)Average V?xj? granite (n = 4)Representative Norra K?rrmineralized nepheline syenite110La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu YFigure 5.2 - Regional soil data from C and B horizons (light and dark blue,respectively) normalized to average granite (Krauskopf, 1967). Yellow points areaveraged Transscandinavian Igneous Belt (TIB) samples (Appelquist, 2010) and red points are averaged V?xj? granite samples (Christensson, 2013). Dark grey points are average representative Norra K?rr mineralized nepheline syenite.Normalized to Average Granite(after Krauskopf 1967)10067  5.1.3 - REE content in Regional Surface Waters Water samples, both filtered and unfiltered, were collected from 42 lakes in the vicinity of the NKAC, and analysed within 48 hours of collection as outlined in Chapter 2.4.8. The unfiltered water samples had higher REE contents than the filtered water samples, median ?REE ? 7.47 ppm compared to median ?REE ? 5.1? ppm, respectively (Table 5.3, Figure 5.3a&b), and the REE content is higher in lakes with more dissolved organic content, i.e. those with brownish water. Interestingly, the darkest brown water, which could be assumed to be the richest in dissolved organic matter, did not have the highest REE concentrations. The brown-yellow lakes had the highest REE content (median ?REE ? ?.41 ppm) compared to the brown (median ?REE ? 2.?4 ppm) or the colourless (median ?REE ? 0.  ?3ppm) lakes. Individual REE concentrations are listed, grouped by lake-water colour, in Table 5.3. An interesting pattern in the REE content of the regional surface waters is that they all display negative Ce and Eu anomalies, based on the shape of the chondrite normalized diagram (Figure 5.3a). This dual negative Ce and Eu anomaly is also visible in the weak leach data for the soil samples collected at site NK12-060, situated at the eastern contact of the Norra K?rr intrusion with the surrounding V?xj? granites (Figure 5.4). This indicates that C horizon tills from site NK12-060 have a ?regional? REE signature, even though they were collected proximal to the deposit.    Table 5.3 ? Median REE content of regional water lake water samples, grouped by water colour and filtration.  68REE Chondr ite nor malized plot, af ter Boynton 1984BrownBrown-yellowLight yellowColourless4195116411051313Dy (ppb)La (ppb)10.10.10.010.01Figure 5.3a - Chondrite normalized diagram of ?iltered regional surface water samples.  Note the strong negative Eu and Ce anomalies.Figure 5.3b - Box and whisker plots of La and Dy (representative of the LREEs and HREEs, respectively) content in ?iltered regional surface water samples.The highest REE content is in the brown-yellow water samples, and isattributable to the higher concentration of dissolved organic acids in these samples.69A horizonB horizonUpper C horizonLower C horizonWater sampleSummed Sequential Leach Data1T near total ?4acid?De-ionized water (?H2O?)Ammonium acetate (?NH3Ac?)REE Chondr ite nor malized plot, af ter Boynton 1984Sodium pyrophosphate  (?Na4P2O 7?)0.1M hydroxylamine (?w-NH2OH?)0.25M hydroxylamine (?s-NH2OH?)Surface water sampleFigure 5.4 ? Chondrite-normalized REE diagram of sur?icial materials from site NK12-060, with local un?iltered lake water data. The water sample with the higher REE content was from site TAS12-43 (Fig. 5.1), which was collected from a small lake directly above the deposit, and the water sample with the lower REE content (TAS12-02) was collected from a lake 1 km NE of the deposit.70  5.2 Local REE Signature at the Norra K?rr Alkaline Complex 5.2.1 Sample Site and Media Selection Vegetation samples (n = 142) were collected from 126 stations above and proximal to the Norra K?rr Alkaline Complex. Soil samples (n = 14) were also collected from the surficial environment around Norra K?rr, but they were deemed unsuitable as a sampling medium because of the extensive disturbance of the soil horizons by anthropogenic activity in the area (see Chapter 1.3.4). Undisturbed soils were present at the base of trees, but were generally inconsistent throughout the survey area. Conversely, both the fern species Dryopteris filix-mas and Athyrium filix-femina were ubiquitous in the survey area, making ferns the preferred sample media.  The first set of fern samples (n = 92) was collected in June 2011, the second set of samples (n = 7) was collected from the same locations in August 2011 to assess seasonal variability, and the third set of samples (n = 43) was collected in August 2012, along an extended southerly line, to ensure samples were taken from the ?background? surficial environment. All samples were collected and analysed according to the procedure outlined in Chapter 2.  The stations were located along east-west lines transecting the deposit, designed to intersect each different rock type and continue 1.5 km into the surrounding country rock. Extensive drilling has already taken place over the deposit, so its lateral extent is well constrained. The drillhole sections were oriented E-W, and particular care was taken in choosing the location of the biogeochemistry sample lines to avoid possible contamination. The drills were lubricated with clean water only, to avoid contamination from drilling muds. In the rare instance that a biogeochemical sample site was nearby a drill collar, the vegetation was collected from at least ten metres away to minimize the likelihood of contamination. Also, many of the samples in the eastern part of the survey were taken from within densely forested areas, lowering the likelihood of wind-borne contaminants. Samples were collected along five survey lines across the deposit. These lines varied in length depending on the width of the intrusion at that latitude. The northernmost survey line was truncated in the west because there is a major highway less than 300 m away from this part of the complex, and the southernmost line continues 1.7 km past the eastern edge of the NKAC.  5.2.2 Element Correlations The REEs, major elements, macronutrients, and micronutrients were tabulated in a Spearman Ranked Correlation (rs) matrix, using data from all 92 samples from the first sampling campaign, to determine which elements are correlated. There is a correlation (rs = 0.6 and 0.53, respectively) between the LREEs, B, and Ca. There is a good correlation (rs = 71  0.75) between the LREEs and Y, with confidence greater than the 75th percentile, as well as a strong correlation, confidence greater than the 90th percentile, between Y and the HREEs (rs = 0.97), which is not surprising due to their chemical similarities. Calcium is an essential macronutrient, and is widely available to plant roots in neutral and alkaline environments (Lucas and Davis, 1961). Though Ca2+ has many metabolic functions, it is particularly useful to the plant as a structural element in the cell wall. Matoh and Kobayashi (1998) found that, when coupled with Ca, B forms a B-RG-II complex that is ?ubiquitous in higher plants?; these two elements bind to pectic polysaccharides and maintain the integrity of cell walls. The correlation between B, Ca, and the LREEs leads to the supposition that REEs taken up into the plant structure are being sequestered in the cell wall, but more detailed analyses are required for confirmation. This is in agreement with findings by Ding et al. (2007) who determined that REE fractionation in soybean roots was controlled by REE-phosphate precipitation and cell-wall absorption.  5.2.3 Levelling 5.2.3i ? Levelling by fern species Different species of ferns can naturally accumulate different amounts of REEs in their plant structure (see Chapter 4.2, Figure 4.2), so to make a coherent biogeochemical response map over the deposit and surrounding area, REE values must be levelled to account for variation in sample species. The levelling method undertaken here was Z-Score 10log-transformation. Employing a 10log-transformation is appropriate because the data is log-normally distributed. Z-Score transformation was used instead of median calculation because levelling by the median does not account for the spread of the data, so variables with a wide distribution and outliers will dominate in a median-levelled distribution. The other advantage to using a Z-Score transformation is that the outliers, ranging from 1.5 to 3 standard deviations, are preserved for each variable. Outliers are classified as outside of 1.5x (Q3-Q1), and are represented as open circles; far outliers are classified as outside of 3x (Q3-Q1), and are represented as open triangles. Box and whisker plots  (e.g. Fig. 5.5) are an effective way to display levelled data from related populations with similar distributions but different outliers. These plots divide the data into four parts by calculating the median, 75th, and 25th percentiles, which are represented as the black line in the centre, and the upper and lower limits of the coloured box, respectively. Any values that fall outside of 1.5x this range are considered ?outliers? and will be displayed as points either above or below the vertical coloured line, which is referred to as the ?whisker?. The black circle near the centre of the coloured box represents the mean of the population.  72  5.2.3ii ? Levelling by analytical method The regional and local samples, for ferns as well as soils, were analysed at different labs by different methods, so in order to compare the local REE signature at the NKAC with the regional background signature, the data must be levelled. The two suitable levelling parameters are ?laboratory? or ?analytical method?. The former is appropriate because the regional soils and ferns were analysed at ALS Analytical Laboratories in Lule?, by lithium metaborate fusion and nitric acid digestion, and the local ferns and soils were analysed at Acme Analytical Laboratories in Vancouver, by aqua regia digestion (see Section 2.4). However, levelling by ?analytical method? is preferable to levelling by ?laboratory? because the soils and ferns were analysed by different techniques at each lab.  When comparing different analytical techniques, it is important to remember that differences in analytical procedure, particularly with respect to the strength of digestion, can result in quantifiable differences in elemental concentrations. For example, aqua regia digestion, such as that performed by Acme Analytical Laboratories on the locally collected fern plants, is a ?near total? digestion, meaning it will target carbonates, soluble or weakly adsorbed phases, oxides, and some silicates. Contrarily, lithium metaborate/tetraborate is a ?fusion? digestion, which targets all the same phases as aqua regia, plus any resistate minerals, such as garnet or xenotime. It is unreasonable to expect to be able to compare aqua regia data with fusion data without first taking their different strengths into consideration; levelling both datasets by ?analytical method? allows comparison of outliers and median values, rather than absolute concentrations, so in an exploration program, where contrast with background is more critical than total element concentration, comparing levelled data is acceptable.   730255075100Figure 5.5 - Two graphical representations, a histogram (above) and a ?box and whisker? plot (below) of the same data set.  The data is normally distributed, with some outliers (red box) and far outliers (green box). The box and whisker plot divides the data into four equal parts, by ?inding the median, and then the 25th and 75th percentiles (which approximatethe interquartile range, or IQR). The median is represented by the white linebisecting the box; the mean is represented by the white circle, in this case located just above the median. Outliers are represented as open black circles,and far outliers are represented as open black triangles. 74  5.3 Discussion  5.3.1 ? Fern Discussion The biogeochemical samples collected above mineralization are elevated in both the heavy and light REEs, particularly the HREEs, compared to samples collected from the surficial environment above unmineralized country rock (Figure  5.6). The range of REE contents from the fern samples collected over the ?background? area is listed in Table 5.2 and displayed in Figure 5.6.  Of the three fern species sampled, Athyrium filix-femina has the highest levels of LREEs, and the highest average value for the HREEs. The REE and Y content is higher in both species of ferns over the deposit than at background sites, but there is clearer contrast in Dryopteris filix-mas than in Athyrium filix-femina (Figure 5.6 and  5.7). This is because, in general, Athyrium filix-femina can have higher REE content in its leafy tissue than Dryopteris filix-mas (Figure 5.8). However, when the surficial environment is enriched in REEs, such as above a REE deposit, Dryopteris filix-mas will accumulate more REEs in its tissue, relatively, than Athyrium filix-femina. This is critical because the physiology of Dryopteris filix-mas prevents it from accumulating REEs in its leafy tissue unless there is an abundance of REEs in the growing conditions. This heightened contrast, relative to the contrast observable in Athyrium filix-femina, makes Dryopteris filix-mas the preferable sample medium (Figure  5.7).         75La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu0.010.1110100Local Athyrium filix-feminaLocal Dryopteris filix-masAverage TIB (n = 11)Average V?xj? granite (n = 4)Representative Norra Karrmineralized nepheline syeniteRange of regional fern dataa) b)Normalized to Average Granite(after Krauskopf 1967)Figure 5.6 - Athyrium ?ilix-femina (left) and Dryopteris ?ilix-mas (right) samples collected from the sur?icial environment above Norra K?rr, normalized to average granite (Krauskopf, 1967). The purple shaded arearepresents the range of REE content in ferns sampled over unmineralizedcountry rock. This range is assumed to represent ?background? and is quanti?ied in Table 5.2.76   Figure 5.7 ? Yttrium content (as a surrogate for HREE) in three fern species over the deposit (left) and in the background (right). Athyrium filix-femina has higher absolute Y content over the deposit than the other two species, but the contrast between samples collected over mineralization compared to samples collected over background is much more pronounced in Dryopteris filix-mas, therefore Dryopteris filix-mas is the preferable exploration sampling medium.  The fern samples collected over unmineralized country rock display the same relative REE variation by species (Figure 5.8). Regionally, Dryopteris filix-mas has a median ?REE value of 24.3 ppm, while Athyrium filix-femina has a higher ?REE value of 57.? ppm (Table 5.1). Athyrium filix-femina has higher ?background? REE content, which supports the earlier conclusion that Dryopteris filix-mas is the preferable sample medium, as it displays better contrast between mineralized and unmineralized areas.    5.3.2 ? Fern and Soil Discussion The fern and soil samples collected both regionally and locally were combined and levelled by analytical methods (see Section 5.2.3ii and Section 2.4),  to determine which sample medium is the most effective for identifying covered REE mineralization. In all cases, the soil has higher total REE abundance than the overlying fern plant, though in some cases, for instance the samples collected directly over mineralized bedrock, the leafy tissue of the fern plant is particularly enriched in LREEs and is comparable to the LREE content of the soils. The La and Dy content in all regional (lighter shades) and local (darker shades) fern (green boxes) and soil (brown boxes) data is displayed in Figure 5.9. The data have been levelled by analytical method, so the Y-axis values are standard deviations, not absolute concentrations. The variance in REE content over mineralization compared to unmineralized country rock is similar for both the fern data and the soil data, based on mean and median values (Figure 5.9).   77Dryopteris filix-mas Athyrium filix-feminaMedianMean75th percentile25th percentileOutliers75th percentile + (1.5 x InterQuartileRange)Far outliersLocal Samples   Dryopteris filix-mas   Athyrium filix feminaRegional Samples   Dryopteris filix-mas   Athyrium filix femina   Figure 5.8 ? La, Eu, Dy, and Yb content in Athyrium ?ilix-femina and Dryopteris ?ilix-mas, grouped by species and coloured regionally (blue) and locally (red).781034426 17 603026 1710344266030 26 17172-21-10All fern and soil dataDryopteris filix-mas only,with all soil dataDy content by levelled Z-score La content by levelled Z-score 2-21-10Regional fern samplesLocal fern samplesRegional soil samplesLocal soil samplesMedianMean75th percentile25th percentileOutliers75th percentile + (1.5 x InterQuartileRange)Far outliersFigure 5.9 -  La (above) and Dy (below) content in sur?icial media, levelledby analytical method. Since the data have been levelled, the Y axis values are standard deviations, not absolute concentrations. Sample count is displayed over each box. In the upper diagram, regional and fern data is presented, along with all soil data, and the variance between the four populations is similar based on the mean and median values.In the lower diagram all regional and local soil data, from each horizon, is presented again, but only the Dryopteris ?ilix-mas data is displayed (bottomright) and the contrast between regional (light green) and local (dark green)fern samples is much more pronounced, particularly for Dy, which is a proxy for all the HREEs. 79  By comparing the regional data to the local data using only Dryopteris filix-mas, the contrast between background and deposit becomes more pronounced (Figure 5.9 and  5.10). The fern samples collected over the deposit (dark green) have higher Dy contents than those collected regionally (light green). The contrast becomes stronger still if the soil samples are grouped into populations based on soil horizons. Using Dy as a proxy for all the HREEs and La as a proxy for the LREEs, background contrast between Dryopteris filix-mas  samples and each soil horizon was compared (Figure 5.10). The contrast between regional and local samples for the C horizon soils is poor; in fact, the regional C horizon soils have higher  median REE content than those collected directly over the deposit, so a false anomaly is generated over unmineralized country rock. Since the C horizon is composed of transported glacial till, it is reasonable to expect the REE signature to be different from the underlying bedrock (Figure 5.10c). The contrast between the regional and local B horizon soils is comparable to the Dryopteris filix-mas samples for HREEs, and slightly better for LREEs (Figure 5.10b). The contrast between the A horizon soils collected regionally and those collected locally is poor; A horizon soil results have large variability of REE distribution based on a small (n = 2) sample group, preventing further interpretation of the A horizon soils.  5.3.3 Exploration Implications  Both Dryopteris filix-mas and B horizon soil samples demonstrate good contrast over mineralized bedrock compared to regional background samples; however, considering the paucity of well developed soil horizons due to anthropogenic disturbances and the glacial history of the area, Dryopteris filix-mas is the most reliable sample medium.  Athyrium filix-femina, as well as A and C horizon soils, are deemed less useful sample media in this environment when exploring for buried REE mineralization.   80603021160302010603048603021130 2010La content by levelled Z-score Dy content by levelled Z-score Regional Dryopteris filix-masLocal Dryopteris filix-masRegional soil samplesLocal soil samplesMedianMean75th percentile25th percentileOutliers75th percentile + (1.5 x InterQuartileRange)Far outliers12-2-3-1012-2-1012-2-1012-2-3-10A horizon C horizonB horizonFigure 5.10 (a-c)  -  La (above) and Dy content (below), levelled by analytical method, in sur?icial materials collected regionally (lighter shades) and locally (darker shades) around the Norra K?rr Alkaline Complex.Dryopteris ?ilix-mas samples both above the deposit (dark green) and in the background (light green) are graphed against each soil horizon (A, B, and C). Compared to the A horizon soils,  Dryopteris ?ilix-mas  displays better contrast between the mineralized deposit and the surrounding country rock for both the LREEs and the HREES. Dryopteris ?ilix-mas and B horizon show comparable contrast for the HREEs, and B horizon soils show more pronounced contrast for the LREEs. C horizon tills show a false anomaly over the regional country rock for the HREEs, and no clear contrast for the LREEs.  6030486081  5.3.4 ? Spatial Distribution of REEs in Ferns Above Norra K?rr The spatial variation of REEs around the NKAC was assessed by collecting 142 fern samples from 126 sample stations above and proximal to the mineralized zone. Profile lines showing variations in Sm, representing the LREEs, and Dy, representing the HREEs, are displayed in Figure 5.11 and Figure 5.12. The LREE signature associated with mineralization is most clearly visible in the Line 4, on the eastern edge of the complex. This is because Norra K?rr forms a relative topographic high, and both east and west  margins are low-lying areas. On the western margin there is a swampy zone, and on the eastern margin there is a small stream running southwards. The LREE peak in the southeast corner of the deposit is likely attributable to increased surface-water flow, thereby allowing increased REE mobility in this area and a stronger LREE response (Figure 5.11 ? Sm profile) The response of Dy in the profile lines above the deposit is clearer, likely because Norra K?rr is enriched in HREEs. There is a strong Dy peak in the third line, directly over the pegmatitic grennaite unit, which has the highest HREE content of all the rock types present at Norra K?rr. There is also a positive response in the second line. In Line 4 there is a strong positive response in the same location as the Sm peak. The peaks in these two elements are likely both caused by increased moisture content in the soil, which facilitates Ln3+ mobility. The other Dy peaks are also recognized in Sm, but not as strongly as the response in the south-eastern part of the deposit (Figure 5.12 ? Dy profile). Both Dy and Sm show weak positive responses in the northwest corner of the deposit, coincident with the pegmatitic catapleiite zone. The depletion of LREEs and HREEs in the southernmost line could be attributable to the presence of kaxtorpite, which has the lowest REE content of any rock type in the NKAC with similar patterns displayed in the LREE profile (Figure 5.11) to the HREE profile (Figure 5.12) but the central line, which shows a strong response for the HREEs, has slightly more minor irregular variation (?noise?) in the LREE data. The peaks in Sm and Dy off the eastern side of the deposit, in Line 3, could also be caused by increased soil moisture, even though it is technically off of the deposit. Further studies are required determine whether the REE signature here is due to Ln3+ ions mobilized in the surficial environment.   82Line 4Line 3Line 2Line 1Figure 5.11 ? Sm content in ferns above the Norra K?rr Alkaline Complex0 100 200 300 40050MetersSWEREF 99TM?Geological Map of Norra K?rr with Sm contentProfile LinesSm (ppm)02Sm (ppm)Sm (ppm)0025Sm (ppm)0583Geological Map of Norra K?rr with Dy contentProfile Lines0 100 200 300 40050MetersSWEREF 99TM?Line 4Line 3Line 2Line 1Dy (ppm)Dy (ppm)Dy (ppm)024602 020510Figure 5.12 ? Dy content in ferns above the Norra K?rr Alkaline Complex84  5.4 Conclusions This biogeochemical orientation survey shows the following: a) All sampled fern species, particularly Athyrium filix-femina and Dryopteris filix-mas, are able to incorporate appreciable amounts of REEs into their leafy tissue.  b) Regionally, C and A-horizon soil samples have lower, or comparable, REE content than C and A-horizon soil samples collected locally, but the B horizon soils at Norra K?rr have a positive REE response when compared to the background B-horizon soils. The range of ?background? REE content in regional soils is quantified in Table 5.2.  c) Compared to regional samples, local Athyrium filix-femina samples do not show good contrast over the deposit; conversely, Dryopteris filix-mas shows strong contrast when comparing regional and local samples. The upper limit of REE content in samples collected over unmineralized bedrock is assumed to represent the ?background? threshold in this environment (Table 5.1).  d) There is a correlation between the REEs, Ca, and B, which may be associated with the physiological roles of Ca and B within the plant. The correlation between Ca and the REEs suggests that the REEs may be incorporated into the cell walls.  e) There are positive responses in the ferns collected over areas of known mineralization, particularly the pegmatitic grennaite units and zones with phenocrysts of catapleiite.  f) Athyrium filix-femina fractionates the LREEs more strongly than Dryopteris filix-mas or Pteridium aquilinum.  g) Athyrium filix-femina has higher absolute contents of LREEs and HREEs in its leaves; however, there is better contrast between background and anomalous areas when sampling Dryopteris filix-mas, which makes it the preferable biogeochemical sampling medium. 85   Chapter 6 - Summary, Exploration Implications, and Recommendations for Future Work 6.1 Summary The geochemical environment around the Norra K?rr Alkaline Complex has developed over several different time scales, from the evolution of the bedrock geology over millions of years, to the development of the surficial environment over tens of thousands of years since the last glacial maximum, to the perennial growth of a fern plant over several years, to the changes in metal content in that fern plant over the course of one summer.  The study of rare earth elements (REEs) provides a tool to help unravel these intertwined geochemical systems; REEs are a unique group of elements due to their similar chemical properties and hence they have been studied extensively as tracer elements in igneous geochemistry, hydrogeology, aqueous geochemistry, and natural water chemistry. Rare earth element research has been topical in the past decade due to the high demand and limited supply of REEs for the burgeoning technology market, and their increased use has enabled advances in both green technology and nanotechnology.  Biogeochemical exploration methods presented herein, were successfully used to delineate the Norra K?rr Alkaline Complex, a HREE-enriched, eudialyte bearing, Mesoproterozoic nepheline syenite. The NKAC was emplaced as a sill in a N-S trending shear zone within the Paleoproterozoic Trans-Scandinavian Igneous Belt (TIB) during a stage of crustal scale deformation, followed by post-mineralization thrust faulting in the Ediacaran. Subsequent glaciation of southern Sweden, terminating approximately 14,000 years ago, buried the deposit with transported Quaternary materials derived from the V?xj? granites, a subset of the TIB. Since the last glacial maximum, soil horizons have begun developing on the glacial till, and they differ chemically from the till because of input from other sources. The NKAC is, locally, a topographic high point (Figure 6.1); in places where the REE-enriched bedrock is exposed, erosion and subsequent deposition of colluvial material derived from the NKAC endows the till in the surrounding region with HREE-bearing zirconosilicate fragments (Figure 3.11). 86StockholmGothenburg100kmFigure 6.1 ? Genetic model of the formation of the surficial environment above the Norra K?rr Alkaline Complex. The glacial material is derived from V?xj? granites inthe up-ice direction.    Sveconorwegian Domain  (1.7 - 0.9Ga) Phanerozoic rocks Svecofennian Domain  (1.95 - 1.86Ga) Caledonide Mountain Range  (0.55 - 0.4Ga) Transscandinavian Igneous  Belt   (1.85 - 1.65Ga) Norra K?rr Alkaline Complex (1.54?0.6Ga) Major shear zone Waterbodies TownsLegend The REE signature of the buried bedrock mineralization was mobilized into the sur?icial environment following the mechanical weathering of the zirconosilicate mineral eudialyte. As eudialyte breaks down, REEs are dissolved and transported by surface waters until they are complexed and re-deposited. Of the REEs, Ce is the least mobile in the sur?icial environment because it has two oxidation states, 3+ and 4+. Under typical sur?icial conditions, Ce will exist in the oxidized form, Ce4+, which is extremely insoluble, and will precipitate immediately as CeO2 or CePO4, which is why there is often a negative Ce anomaly observed in natural surface waters (Figure 6.2). The other REEs can adsorb onto colloids or Mn hydroxides, form stable compounds with carbonate, phosphate, ?luoride or hydroxide anions, or, most likely, form complexes via bonding with the carboxyl group of any dissolved organic acid (see Section 1.6 and references therein). Different REE patterns in a sur?icial environment can be used to infer the provenance of the sur?icial materials, be it soils or plants or glacial till, therefore subtle variations in these REE pattern, discernible from normalizing plots, can be used to distinguish materials derived from different sources. In Figure 6.2, there are positive Eu anomalies in the three different materials: the glacial till sample collected above the NKAC (red triangles); granitic fragments (pink squares) taken as a proxy for V?xj?-type granites; and a suite of soil samples from a distal, typical, boreal, podzol in southern Sweden (brown shaded region). All of these media have the same REE pattern, indicating that they originated from the same source. 3)Nemplacement*glaciationfaultingcolluvial depositionColluvium from the NKACAngular granitic fragmentLocally derived nepheline syeniteErosional levelGlacial till with foreign clastsFaultShear zoneV?xj? granites (subset of the TIB) 1)2)5)4) erosion87Figure 6.2 ? Schematic Diagram of Norra K?rr withnormalized REE plot of surficial materials. The REE plot is normalized to local bedrock from site NK12-060, not chondrite. Green squaresare local bedrock, filled red circles are surficial material from the A and B horizons, filled redriangles are glacial till, filled pink squares are glacially transported granitic fragments derived from nearby country rock. Blue circles are local surface waters from lakes directly overlying the NKAC; surface water REE content has been multiplied by a factor of 1000 for convenience of viewing. The shaded brown region represents the range of REE content of typical Swedish podzols (n = 30) collected in Scania, approximately 200km south of Norra K?rr, and analysed by microwave-aided nitric acid digestion, with an ICP-MS finish (Tyler and Olsson, 2002) which is a slightly weaker strength of digestion than the sum of the sequential leach data provided by Acme Analytical (red shapes). Note the positive Eu anomaly in the granitic fragment and theglacial till, which is not visible in the local bedrock or surface water samples.La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Typical Swedish podzolic soil samples collected and analysed by Tyler and Olsson (2002), feature a strong positive Nd anomaly (brown dashed lines represent the actual data range), which is substantially different than any of the rocks in the Trans-Scandinavian Igneous Belt (see Section 5.1.2); speculation on the cause of this anomaly is outside the scope of this thesis, so the Nd content of the podzolic soil data has been estimated based on the Pr and Sm content. {{{{Angular granitic fragmentLocally derived nepheline syenite fragmentsGlacially transported foreign clasts Surface waterLocal bedrockA and B horizon soils C horizon glacial tills Angular granitic fragmentLocal surface water samplesRange of typical southern Swedish podzol0.10.010.0011 The highest concentrations of REEs in the soil pro?ile above the NKAC are found in the B horizon soils, and are due to a combination of the presence of colluvial zirconosilicate mineral grains, and downward translocation of REEs via complexation with soluble organic acids from the eluvial ?A horizon? into the illuvial ?B horizon?. The precipitation of these metal-organic complexes at depth could be attributable to metal saturation, organic polymerization, or changes in pH (Drever and Vance, 1994). Organic acids, which are measurable as the dissolved organic carbon content in soil water, surface water and groundwater, provide an ef?icient trap site to sequester REEs.  The B horizon soils at Norra K?rr developed from the underlying glacially transported till, and show a weak positive Eu anomaly (Figure 6.3) therefore are not representative of the underly-ing bedrock.8810.10.010.001La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb LuNor ra K ar rbedrockGlacial till with angular granitic cobbles?B ? hor izonAh hor izonIIIIIIIIIIIILocal bedrockA and B horizon soil samplesC horizon glacial till samplesAngular granitic fragmentFern tissue samples(I = root, II = leaf, III = stem)Range of typical southern Swedish podzol{{{{Figure 6.3 - Diagram of a fern plant with underlying soil pro?ile and corresponding REE pro?iles.  REEs  have been normalized to local bedrock at site NK12-060. Note the decrease in REE content in the fern plant from root > leaf > stem, and the fractionation within the plant of LREEs over HREEs. Soil samples, till samples, and V?xj? granitic samples are shown for comparison. As the rate of discovery of near-surface mineral deposits declines, all available tools must be utilized to successfully explore for buried mineralization. The mobility of REEs in the sur?icial envi-ronment, a heretofore under-appreciated phenomenon, allows biogeochemistry to be used as an effective exploration tool for buried REE deposits, such as the Norra K?rr Alkaline Complex.6.2 Exploration Implications Fern plants are an ideal biogeochemical sampling medium when exploring for HREE miner-alization because they can have measurable amounts of REEs in their leaves, and they provide an integrated chemical signature based on all the components of the sur?icial environment. Plants utilize mycorrhizal fungi and transport vast quantities of surface water to acquire nutrients, and therefore their tissue is a composite sample of the geochemical environment, involving surface water, soil substrate, and bedrock, in which they grow (Figure 6.3). The fern plants above the Norra K?rr Alkaline Complex contain appreciable amounts of REEs in varying quantities (root > leaf > stem) in their tissue. The root tissue has comparable REE content with the till samples collected from the same site, and the leaf tissue has similar LREE levels, but considerably less HREE content than the soils or the root tissue. This indicates there is some preferential physiological mechanism within the plant that transports LREEs more readily than HREEs, likely due to differences in their ionic radii or charge density.   Other plant species were collected and analyzed at Norra K?rr as well, but ferns, particularly Dryopteris ?ilix-mas, displayed the best contrast between mineralization and background, so was deemed the effective biogeochemical sampling medium for evaluating an area of covered boreal forest for buried HREE mineral potential.89  6.3 Handy Hints for the Burgeoning Biogeochemist 6.3.1  Sample Station Selection As with any geochemical exploration program, the target size will dictate the optimal sample spacing. It is critical to undertake an orientation survey first, and extend the orientation survey lines as far as is practical for cost and time efficiency. If there are some limitations in the surrounding environment preventing long sample lines, go to an area nearby that you?re sure has no mineralization and collect a few samples there, to establish ?background?.  6.3.2  Species Selection It?s all well and good to know that Douglas fir trees concentrate As in unusually high levels (Warren et al., 1964), but if you only have maple trees on your property... you?re probably going to sample maple trees. I hear the syrup is delicious... But this brings us to the first and most important rule of geochemical sampling: be consistent!!  In the same way that you cannot properly compare soils collected from different horizons, you also cannot compare different plant species, it?s like the age old adage chiefly concerned with the differences between apples and oranges. Depending on the species you choose, you can usually sample bark, leaves, twigs, or cones. If you?re not sure which media shows the best response in your area, the orientation survey is a great way to find out.  6.3.3  Sample Collection 1. Collect at least 100 g of sample material, because at least half is water weight. 2. Place the samples in woven polypropylene bags (not plastic, not zip loc ) so the samples don?t sweat, mold, and rot before they get to the lab 3. dry the samples out thoroughly as soon as possible. Not only will this save you valuable exploration dollars on you shipping costs, but it will prevent sample rot. 4. If you aren?t totally sure what species you?re sampling, and even if you?re pretty sure, collect a specimen, press it nicely in your notebook, and refer to it later in the survey, or in the sample prep stages to confirm you?ve collected the right species consistently (see Rule ?1). If you have a botanist friend, try and convince them to come sampling with you! 5. Use the same tools throughout the entire survey ? you wouldn?t switch shovels part way through a soil survey, so don?t switch scissors/shears mid-survey when you?re collecting biogeochemical samples.  6. When sampling, for instance,  a tree, there are several different media available for collection: bark, twigs, leaves/needles, or cones. Be sure to create a composite sample by collecting the selected tissue from all around the perimeter of the tree. In some cases, particular elements may be concentrated on the north (or south) aspect of the tree, depending on the individual species physiology. 90  7. Be sure to sample plants that have similar age and health. An older tree will have a more developed root system that interacts with a greater volume of soil, so could give you a false positive anomaly, if you otherwise only sampled young trees. 8. Take detailed notes of the sample site (topography, moisture content, slope aspect, species present) as well as detailed notes and photos of the individual plant being sampled (age, size, health), and while you?re on the ground, do some geology? In many instances you?re the only person who will go to that site, so be sure to capture as much information as possible.  Now get out there and go do some biogeochemistry!  6.4 Recommendations for Future Work As with any research, this thesis posed questions that could be addressed with further study. Some recommended areas for future research are: 1) Laboratory based experiments  to assess the detailed vertical migration of REEs in a controlled environment: a. Modelling the breakdown of REE-enriched substrate under controlled ?natural? weathering conditions. This could include agitating REE-bearing mineral grains in solutions of varying acidity for different lengths of time and monitoring the REE concentration of the resulting solution. This will assist in quantifying the effects of mechanical and chemical weathering on the REE-bearing mineral grain, therefore allow the explorer to predict the rate of weathering under natural conditions. b. Tracing the flow of REEs through different media using doped isotopic REE solutions. Many metal ions are translocated through the surficial environment via complexation with organic acids, and research has been done testing the rate of complexation of REEs with varying concentrations of REEs and humic acids under conditions of varying pH. However, this research was all performed in solution. Experiments modelling  the transport of REEs under more natural conditions, mimicking those of soil pore water, have not been attempted. These conditions should encompass a range of pH values, as well as a variety of substrates, such as clean quartz sand, to organic rich clays and silts of varying permeability due to compaction. This will assist in hypothesizing the rate of transport under natural conditions, and allow estimations of expected REE distribution in the natural environment based on the age of the deposit and the nature of the surficial environment.  2) Biogeochemical orientation surveys over other styles of REE deposit (monazite-bearing placers, carbonatites, pegmatites associated with metaluminous granites, Fe oxide phosphate rocks, laterites) to assess the relative size of the biogeochemical footprint relative to the style of mineralization. The main factor inducing REE solubility is the nature of the host REE minerals; for instance, a churchite- bearing (hydrous REE 91  phosphate) laterite deposit will have a much larger REE footprint than a xenotime-bearing (REE, Zr phosphate) pegmatite because its REEs are more easily weathered.  3) Physiological studies of plants to determine what internal mechanism is responsible for the preferential mobility of LREEs over HREEs. This preferential mobility is observable by the increased proportion of LREEs in the leaf tissue compared to the root tissue. In this thesis, experiments were performed on the CLS synchrotron in Saskatoon to determine which part of the plant was hosting the REEs; unfortunately, the level of REEs was too low to be measured. Since REEs are toxic to plants at the high concentrations necessary for detection, plants should be grown in a REE-free or REE-poor solution, and then immersed in a REE-rich solution at increased air temperature to promote solution uptake into the plant. This may ultimately kill the plant, but not before the REEs infiltrate the plant tissue.  4) Lateral migration of REEs sourced from mineralization through groundwater or along aquifers. Since REEs are mobilized mainly via complexation with the carboxyl groups of organic acids, there may not be significant REE concentrations in groundwater reservoirs, however, surface water flow, especially in swampy or marshy areas, could be a significant consideration in the contributions to the REE budget.                      92  References  Adamsson, O. J., 1944, The petrology of the Norra Karr district: Geologiska F?ereningan i Stockholm, v. 66, no. 2, p. 142. Anders Rapp, Rolf Nyberg, and Lindh, L., 1986, Nivation and local glaciation in N. and S. Sweden. A progress report.: Geografiska Annaler, v. 68, p. 8. Andersson, U. B., Sjostrom, H., Hogdahl, K., and Eklund, O., 2004, The Transscandinavian igneous belt, evolutionary models: SPECIAL PAPER-GEOLOGICAL SURVEY OF FINLAND, v. 37, p. 104. Andres, Y., Thouand, G., Boualam, M., and Mergeay, M. 2000. Factors influencing the biosorption of gadolinium by micro-organisms and its mobilisation from sand. Applied Microbiology and Biotechnology, 54:262 ? 267. Bau, M., 1991, Rare-earth element mobility during hydrothermal and metamorphic fluid-rock interaction and the significance of the oxidation state of europium: Chemical Geology, v. 93, no. 3, p. 219-230. Belli, P.; Bernabei, R.; Cappella, F. et al. (2007). ?Search for ? decay of natural Europium?. Nuclear Physics A 789 (1?4): 15?29. Blaxland, A., 1977, Agpaitic magmatism at Norra Karr? Rb-Sr isotopic evidence: Lithos, v. 10, p. 8. Bluemel, B., Leijd, M., Dunn, C., Hart, C., Saxon, M., and Sadeghi, M., 2013, Biogeochemical expression of Rare Earth Element and Zirconium mineralization at Norra K?rr, Southern Sweden: Journal of Geochemical Exploration. Boynton, W. V., 1984, Cosmochemistry of the rare earth element: meteorite studies., Amsterdam, Elsevier, Rare Earth Element Geochemistry. Braun, J.-J., Pagel, M., Muller, J.-P., Bilong, P., Michard, A., and Guillet, B., 1990, Cerium anomalies in lateritic profiles: Geochimica et Cosmochimica Acta, v. 54, no. 3, p. 781-795. Cohen, D. R., Kelley, D. L., Anand, R., and Coker, W. B., 2010, Major advances in exploration geochemistry, 1998-2007: Geochemistry: Exploration, Environment, Analysis, v. 10, p. 13. Coulson, I. M., 1997, Post-magmatic alteration in eudialyte from the North Qoroq centre, South Greenland: Mineralogical Magazine, v. 61, p. 10. Cox, K. G., Bell, J. D., and Pankhurst, R. J., 1979, The interpretation of igneous rocks, G. Allen & Unwin. Creed, J. T., Brockhoff, C. A., and Martin, T. D., 1994, Method 200.8 Determination of trace elements in waters and wastes by inductively coupled plasma - mass spectrometry, in development., E. m. s. l. o. o. r. a., ed.: Cincinnati Ohio. De Baar, H. J. W., Bacon, M. P., Brewer, P. G., and Bruland, K. W., 1985, Rare earth elements in the Pacific and Atlantic Oceans: Geochimica et Cosmochimica Acta, v. 49, no. 9, p. 1943-1959. Diatloff, E., Asher, C., and Smith, F., 1996, Concentrations of rare earth elements in some Australian soils: Soil Research, v. 34, no. 5, p. 735-747. Ding, S., Liang, T., Yan, J., Zhang, Z., Huang, Z., and Xie, Y., 2007, Fractionations of rare earth elements in plants and their conceptive model: Science in China Press, v. 50, p. 9. Ding, S., Liang, T., Zhang, C., Huang, Z., Xie, Y., and Chen, T., 2006, Fractionation mechanisms of Rare Earth Elements (REEs) in hydroponic wheat: An application for metal accumulation by plants: Environmental Science and Technology, v. 40, p. 7. Drever, J. I., and Vance, G. F., 1994, Role of soil organic acids in mineral weathering processes, Organic acids in geological processes, Springer, p. 138-161. Dunn, C. E., 1998, Regional and detailed biogeochemical surveys in the Nechako NATMAP area and in the Babine Porphyry Belt, in: New Geological Constraints on Mesozoic to Tertiary Metallogenesis and on Mineral Exploration in Central British Columbia: Nechako NATMAP Project., in L.C. Struik, D. G. M., ed., Short Course Extended Abstracts, Cordilleran Section of the Geological Association of Canada, p. 17. 93  Dunn, C. E., 2007, Biogeochemistry in Mineral Exploration, Amsterdam, Elsevier, Handbook of Exploration and Environmental Geochemistry. Dupr?, B., Viers, J., Dandurand, J.-L., Polve, M., B?n?zeth, P., Vervier, P., and Braun, J.-J., 1999, Major and trace elements associated with colloids in organic-rich river waters: ultrafiltration of natural and spiked solutions: Chemical Geology, v. 160, p. 17. Epstein, E., 1999, Silicon: Annual Review of Plant Physiology & Plant Molecular Biology, v. 50, p. 4. Fu, F., Akagi, T., and Shinotsuka, K., 1998, Distribution Pattern of Rare Earth Elements in Fern - Implications for intake of fresh silicate particles by plants: Biological Trace Element Research, v. 64, p. 14. Golyshev, V. M., Simonov, V. I., and Belov, N. V., 1971, Kristallografiya: Sov. Phys. Crystallogr., v. 16, p. 56. Guiseppetti, G., Mazzi, F., and Tadini, C., 1971, TMPM Tschermaks Mineral: Petrogr. Mitt., v. 16, p. 105. Guo, F., Wang, Y., Sun, J., and Chen, H., 1996, REE Bound Proteins in Natural Plant Fern Dicranopteris dichitoma  by MAA.: Journal of Radioactivity and Nuclear Chemistry, v. 209, p. 9. Hatch, G., 2012, Dynamics in the Global Market for Rare Earths: Elements, An International Magazine of Mineralogy, Geochemistry, and Petrology, v. 8, p. 6. Hawkes, H. E., 1957, Principles of geochemical prospecting, US Govt. Print. Off., v. 1000. He, Y., Wang, J., Fang, N., Gan, W., and Zhao, G., 1998, Effects of rare earth micro-fertilizer on plant physiological indexes and yield of hot pepper: Chinese Rare Earth, v. 19, no. 2, p. 36-40. Hem, J. D., 1978, Redox processes at surfaces of manganese oxide and their effects on aqueous metal ions: Chemical Geology, v. 21, no. 3?4, p. 199-218. H?gdahl, K., Andersson, U. B., and Eklund, O., 2004, The Trans-Scandinavian Igneous Belt (TIB) in Sweden: a review of its character and evolution, in Finland, G. S. o., ed., Volume Special Paper 37: Finland, Gological Survey of Finland, p. 125. Howell, J., and Gawthorne, J., 1987, Copper in animals and man. Jenkinson, D. S and Ladd, J. N. 1981.  Microbial mass in soil: measurement and turnover, in: Paul, E. A., Jia, W. Z., Di, G. T. (Eds.), Rare Earths and Agriculture. Agriculture Press, Beijing, 1-3.  Johannesson, K. H., Stetzenbach, K. J., Hodge, V. F., and Lyons, W. B., 1996, Rare earth element complexation behavious in circumneutral pH groundwaters: assessing the role of carbonate and phophate ions: Earth and planetary science letters, v. 139, p. 14. Johansson, ?., 1988, The age and tectonic setting of the Sm?land-V?rmland granite-porphyry belt: Geologiska F?ereningan i Stockholm, v. 110, no. 2, p. 5. Johnsen, O., Ferraris, G., Gault, R. A., Grice, J. D., Kampf, A. R., and Pekov, I. V., 2003, The nomenclature of eudialyte group minerals: The Canadian Mineralogist, v. 41, p. 9. Johnsen, O., Grice, J. D., and Gault, R. A., 2001, The eudialyte group: a review: Geology of Greenland Survey Bulletin, v. 190, p. 65-72. Kovalevsky, A. L., 1987, Biogeochemical Exploration for Mineral Deposits, The Netherlands, VNU Science Press BV. Krauskopf, K. B., and Bird, D. K., 1967, Introduction to geochemistry, McGraw-Hill New York. Langmuir, D., 1978, Urnaium solution-mineral equilibria at low temperatures with applications to sedimentary ore deposits: Geochimica et Cosmochimica Acta, v. 42, p. 22. Larsen, L. M., and S?rensen, H., 1987, The Il?maussaq intrusion - progressive crystallization and formation of layering in an agpaitic magma: Alkaline Igneous Rocks, v. 30, p. 15. Laul, J. C., Weimer, W. C., and Rancitelli, L. A., 1979, Biogeochemical distribution of rare earths and other trace elements in plants and soils: Physics and chemistry of the earth, v. 11, p. 8. Lee, J. H., and Byrne, R. H., 1992, Complexation of trivalent rare earth elements (Ce, Eu, Gd, Tb, Yb) by carbonate ions: Geochimica et Cosmochimica Acta, v. 57, p. 7. 94  Lewis, A. J., Komninou, A., Yardley, B. W. D., and Palmer, M. R., 1998, Rare earth element speciation in the geothermal fluids from yellowstone national park, Wyoming, USA: Geochimica et Cosmochimica Acta, v. 62, no. 4, p. 6. Li, F., Shan, X., Zhang, T., and Zhang, S., 1998, Evaluation of plant availability of rare earth elements in soils by chemical fractionation and multiple regression analysis: Environmental Pollution, v. 102, p. 9. Lippold, H., Evans, N., Warwick, P., and Kupsch, H., 2007, Competitive effect of iron (III) on metal complexation by humic substances: Characterisation of ageing processes: Chemosphere, v. 67, no. 5, p. 1050-1056. Lucas, R. E., and Davis, J. F., 1961, Relationships between pH values of organic soils and availabilities of 12 plant nutrients: Soil Science, v. 92, no. 3, p. 177. Lundqvist, J., 2004, Glacial History of Sweden: Quaternary Glaciations - Extent and Chronology, v. 3, p. 11. Lundqvist, J., and Wohlfarth, B., 2001, Timing and east-west correlation of south Swedish ice marginal lines during the late Weichselian: Quaternary Science Reviews, v. 20, p. 22. Markert, B., and Li, Z. D., 1991, Natural background concentrations of rare-earth elements in a forest ecosystem: The science of the total environment, v. 103, p. 8. Marsac, R., Davranche, M., Gruau, G., Dia, A., and Bouh-Le Coz, M., 2012, Aluminium competitive effect on rare earth elements binding to humic acid: Geochimica et Cosmochimica Acta, v. 89, p. 9. Matoh, T., and Kobayashi, M., 1998, Boron and Calcium, Essential Inorganic Constituents of Pectic Polysaccharids in Higher Plant Cell Walls: Journal of Plant Research, v. 111, p. 12. McDonough, W. F., and Sun, S.-S., 1995, The composition of the Earth: Chemical Geology, v. 120, no. 3, p. 223-253. Nakamura, N., 1974, Determination of REE, Ba, Fe, Mg, Na and K in carbonaceous and ordinary chondrites: geochimica et cosmochimica acta, v. 38, no. 5, p. 757-775. Ozaki, T., and Enomoto, S., 2001, Uptake of rare earth elements by Dryopteris erythrosora (autumn fern): RIKEN REVIEW, p. 84-87. Pang, X., Li, D., and Peng, A., 2002, Application of rare-earth elements in the agriculture of China and its environmental behavior in soil: Environmental Science and Pollution Research, v. 9, no. 2, p. 143-148. Pourret, O., Davrache, M., Gruau, G., and Dia, A., 2007, Rare earth elements complexation with humic acid: Chemical Geology, v. 243, p. 13. Rankin, L., 2011, Structural setting of the Norra K?rr intrusive complex, central Sweden: Australia, p. 40. Reed, G. C., 2011, NI 43-101 Technical Report, Norra Karr REE-Zirconium Deposit, Granna, Sweden. Robinson, W. O., Bastron, H., and Murata, K. J., 1958, Biogeochemistry of rare earth elements with particular references to hickory trees: Geochimica et Cosmochimica Acta, v. 14, p. 12. Rose, A. W., Hawkes, H. E., and Webb, J. S., 1979, Geochemistry in mineral exploration, Academic Press London. Schijf, J., and Byrne, R. H., 2001, stability constants for mono- and dioxalato-complexes of Y and the REE, potentially important species in groundwaters and surface freshwaters: Geochimica et Cosmochimica Acta, v. 65, no. 7, p. 9. Sholkovitz, E. R., 1995, The aquatic chemistry of rare earth elements in rivers and estuaries: Aquatic Geochemistry, v. 1, p. 34. Sinclair, A., 1974, Selection of threshold values in geochemical data using probability graphs: Journal of Geochemical Exploration, v. 3, no. 2, p. 129-149. Sj?qvist, A., 2012, U/Pb isotope research, in Bluemel, B., ed.: University of G?teborg, Sweden. 95  Sj?qvist, A., Cornell, D., Andersen, T., Erambert, M., Ek, M., and Leijd, M., 2012, Three varieties of eudialyte-group REE-ore mineral from the Norra K?rr Alkaline Complex, southern Sweden: Minerals, v. 2, p. 28. Smedley, P. L., 1991, The geochemistry of rare earth elements in groundwater from the Carnmenellis area, southwest E: Geochimica et Cosmochimica Acta, v. 55, p. 12. S?rensen, H., 1997, The agpaitic rocks - an overview: Mineralogical Magazine, v. 61, p. 14. Stanley, C. R., 2003, Statistical evaluation of anomaly recognition performance: Geochemistry: Exploration, Environment, Analysis, v. 3, no. 1, p. 3-12. Taiz, L., and Zeiger, E., 2010, Plant Physiology, USA, Sunauer Associates, Inc., 782 p.: Takada, J., Sumino, T., Nishimura, K., Tanaka, Y., and Akaboshi, M., 1996, Determination of rare earth elements in fern leaves using instrumental neutron activation analysis: Journal of Radioactivity and Nuclear Chemistry, v. 214, no. 2, p. 13. Tang, J., and Johannesson, K. H., 2003, Speciation of rare earth elements in natural terrestrial waters: assessing the role of dissolved organic matter from the modelling approach: Geochimica et Cosmochimica Acta, v. 67, no. 13, p. 8. Taylor, S. R., and McLennan, S. M., 1985, The continental crust: its composition and evolution. Taylor, S. R., and McLennan, S. M., 1995, The geochemical evolution of the continental crust: Reviews of Geophysics, v. 33, no. 2, p. 241-265. T?rnebohm, A. E., 1906, Katapleiit-syenit, en nyupptackt varietet af nefelinsyenit i Sverige, in SGU, ed., Volume C: Stockholm. Tyler, G., 2004, Rare earth elements in soil and plant systems - A review: Plant and Soil, v. 267, p. 16. Tyler, G., and Olsson, T., 2002, Conditions related to solubility of rare and minor elements in forest soils: Journal of Plant Nutrition and Soil Science, v. 165, no. 5, p. 594-601. Ussing, N. V., 1912, Geology of the country around Julianehaab, Greenland. Wakita, H., Rey, P., and Schmitt, R., Abundances of the 14 rare-earth elements and 12 other trace elements in Apollo 12 samples: five igneous and one breccia rocks and four soils, in Proceedings Lunar and Planetary Science Conference Proceedings1971, Volume 2, p. 1319. Wang, Y. Q., Sun, J. X., Chen, H. M., and Guo, F. Q., 1997, Determination of the contents and distribution chacteristics of REE in natural plants by NAA: Journal of Radioactivity and Nuclear Chemistry, v. 219, p. 4. Warren, H. V., Delavault, R. E., and Barakso, J., 1964, The role of arsenic as a pathfinder in biogeochemical prospecting: Economic Geology, v. 59, no. 7, p. 1381-1385. Welch, R. M., 1995, Micronutrient nutrition of plants: Critical Review of Plant Science, v. 14, p. 33. Welin, E., 1980, Tabulation of recalculated radiometric ages published 1960-1979 for rocks and minerals in Sweden: Geologiska F?ereningan i Stockholm, no. 104, p. 12. Williams, J. F., 1841, Eudialyte and Eucolite from Magnet Cove, Arkansas: American Journal of Science, v. 40, p. 5. Wilson, B. M., 1989, Igneous petrogenesis a global tectonic approach, Chapman & Hall. Wood, S. A., 1990, The aqueous geochemistry of the rare-earth elements and yttrium: Chemical Geology, v. 82, p. 27. Wyttenbach, A., Furrer, V., Schleppi, P., and Tobler, L., 1998, Rare earth elements in soil and soil-grown plants: Plant and Soil, v. 199, p. 7. Yamamoto, Y., Takahashi, Y., and Shimizu, H., 2005, Systematics of stability constants of fulvate complexes with rare earth ions: Chemistry Letters, v. 34, no. 6, p. 880-881. Zhang, Z., Wang, Y., Li, F., Xiao, H., and Chai, Z., 2002, Distribution characteristics of rare earth elements in plants from a rare earth ore area.: Journal of Radioactivity and Nuclear Chemistry v. 252, p. 5.    96Appendix A1 Technical Memo: Data Reliability Below Certified Detection Limits  The detection limit (D