PLANT PERFORMANCE ON ANTHROPOSOLS AT HUCKLEBERRY MINE, HOUSTON, BRITISH COLUMBIA by Allan Carson B.Sc., University o f Northern British Columbia, 2006 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN NATURAL RESOURCES AND ENVIRONMENTAL STUDIES (BIOLOGY) UNIVERSITY OF NORTHERN BRITISH COLUMBIA July 2012 ©Allan Carson, 2012 1+1 Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 Wellington Street Ottawa ON K1A0N4 Canada 395, rue Wellington Ottawa ON K1A 0N4 Canada Your file Votre reference ISBN: 978-0-494-94122-5 Our file Notre reference ISBN: 978-0-494-94122-5 NOTICE: AVIS: The author has granted a non­ exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distrbute and sell theses worldwide, for commercial or non­ commercial purposes, in microform, paper, electronic and/or any other formats. L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conform em ent a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. W hile these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. Canada Abstract Supplies of topsoil are often in limited supply for use in mine reclamation activities; it may be necessary to build soils (Anthroposols) using locally available substrates. Revegetation test plots were established at Huckleberry Mine, Houston, B.C., to investigate plant performance on soils supplemented with (or without) non-acid generating (NAG) sand and fertilizer. The addition o f NAG sand reduced some soil properties conducive to plant growth (e.g. cation exchange capacity), yet plant performance was not significantly lower than that observed in soil-only plots. When combined with a fertilizer application, plant performance on NAG sand-supplemented soils significantly increased. Trace element concentrations in supplemented soils were low and should not have any adverse effects on plant growth or the local environment. Plant performance o f blue wildrye (mixed genotype variety) was shown to be higher than all other species examined and is suggested as the best candidate for the revegetation at Huckleberry Mine. Table of Contents Abstract................................................................................................................................................ ii Table o f Contents...............................................................................................................................iii List o f T ables..................................................................................................................................... iv List o f Figures................................................................................................................................... vii List o f Appendices.............................................................................................................................. x G lossary.............................................................................................................................................. xi Acknowledgements......................................................................................................................... xiii 1.0 Introduction............................................................................................................................. 1 2.0 M ethods................................................................................................................................ 38 3.0 Results................................................................................................................................... 65 4.0 Discussion........................................................................................................................... 104 5.0 Conclusions and Recommendations................................................................................ 134 6.0 Literature C ited.................................................................................................................. 137 7.0 Appendices.......................................................................................................................... 147 List of Tables Table 1. A list of the nine treatment substrates for revegetation test plots established in September 2 0 0 8 ................................................................................................................................ 45 Table 2. Seed treatment types and derived bulk seed application weights for 1.0 m2 subplots based on estimated percent germination and percent purity values for each seed l o t..............49 Table 3. Chemical analysis conducted on soil samples (completed by BCMOF and ALS Canada Ltd.) collected from test plot treatment substrates in August 2009 and 2010............. 56 Table 4. Independent variables for native vegetation test plots constructed with 2 or 10 year old stockpiled so il............................................................................................................................. 63 Table 5. Dependent variables for native vegetation test plots constructed with 2- or 10-year old stockpiled soil............................................................................................................................. 64 Table 6. Mean values (± SE) of physical and chemical soil properties in 2- and 10-year old stockpiled soil supplemented with or without NAG sand............................................................ 66 Table 7. Mean values (±SE) for the concentration o f extractable elements (mg k g '1) in 2- and 10-year old stockpiled soil supplemented with or without NAG sand....................................... 68 Table 8. Mean values (±SE) for the total concentration o f rare earth and trace elements (mg k g 1) in 2- and 10-year old stockpiled soil supplemented with or without NAG sand (n=4) . 69 Table 9. Mean values (±SE) for base metal concentrations (mg k g '1) in 2- and 10-year old stockpiled soil supplemented with or without NAG sand (n=4) and NAG sand (n=4)........... 72 Table 10. Mean values (±SE) for the percent composition of major elements reported as oxides in 2- and 10-year old stockpiled soil supplemented with or without NAG s a n d ......... 73 Table 11. Mean values (± SE) of chemical soil properties of 2- and 10-year old stockpiled soil with or without the addition of fertilizer.................................................................................74 Table 12. Summary o f linear mixed-effects model results for percent emergence, seedling density, percent cover and plant height (single-species treatments) collected from 2- and 10year old stockpiled soil test plots in September 2010...................................................................76 Table 13. Summary o f linear mixed-effects model results for plant height for mixed-species treatments collected from 2- and 10-year old stockpiled soil test plots in September 2010 .. 85 Table 14. Summary of linear mixed-effects model results for percent cover per seedling (% seedling'1) calculated using cover and seedling density collected from 2- and 10-year old stockpiled soil test plots in September 2010.................................................................................. 87 Table 15. Summary o f linear mixed-effects model results for emergence, seedling density, percent cover and plant height (single-species treatments)collected from NAG sand test plots in September 2010............................................................................................................................ 89 Table 16. Summary of linear mixed-effects model results for plant height (mixed-species treatments) and percent cover per seedling (% seedling'1), calculated using cover and seedling density, collected from NAG sand testplots in September 2010...................................89 Table 17. Summary of linear mixed-effects model results for aboveground biomass, estimated belowground biomass and shoot:root ratio collected from 2- and 10-year old stockpiled soil test plots in September 2010..................................................................................92 Table 18. Summary of linear mixed-effects model results for above- and estimated belowground biomass per seedling calculated using above- and estimated belowground biomass and seedling density collected from 2- and 10-year old stockpiled soil test plots in September 2 0 1 0 ................................................................................................................................ 99 Table 19. Summary o f linear mixed-effects model results for aboveground biomass, estimated belowground biomass and shoot:root ratio collected from NAG sand testplots in September 2 0 1 0 .............................................................................................................................. 103 Table 20. Summary o f linear mixed-effects model results for above- and estimated belowground biomass per seedling calculated using aboveground biomass and seedling density collected from NAG sand testplots in September 20 1 0 ............................................... 103 List of Figures Figure 1. Overview of the Huckleberry minesite, located approximately 86 kms southwest o f Houston, BC, as photographed in the summer o f 2007................................................................28 Figure 2. Seed collection and preparation o f arctic lupine..........................................................40 Figure 3. An aerial view of Huckleberry mine (2011), located 86 kms southwest o f Houston, BC, indicating the four locations o f the revegetation test plots.................................................. 43 Figure 4. Overview diagram o f soil treatments (each square representing a subplot) for NAG sand test plots (left) and the supplemented soil test plots (right)................................................ 44 Figure 5. Photographs illustrating the construction o f the test p lo ts ......................................... 46 Figure 6. Images illustrating procedures associated with test plot seeding............................... 48 Figure 7. Overview diagram o f seeding treatments......................................................................50 Figure 8. Images from Year One vegetation sampling................................................................ 53 Figure 9. A diagram displaying subplot sampling groups for NAG sand test plots (left) and supplemented soil test plots (right)................................................................................................. 55 Figure 10. Photographs showing the harvesting o f biomass from revegetation test plots....... 59 Figure 11. Photographs illustrating preparation o f aboveground biomass and seedling samples prior to weighing................................................................................................................60 Figure 12. Overall seeding treatment effects on mean values (±SE) of seedling density (seedlings m'2) across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled so il................................................................................................................................... 77 Figure 13. Overall seeding treatment effects on mean values (±SE) of emergence (%) across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled s o il.................79 Figure 14. Overall seeding treatment effects on mean values (±SE) of plant cover (%) across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled s o il.................82 Figure 15. Overall species effects on mean values (±SE) o f plant height (cm) for single­ species seeding treatments across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled soil............................................................................................................... 83 Figure 16. Overall species effects on mean values (±SE) o f plant height (cm) for mixedspecies seeding treatments across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled soil...............................................................................................................86 Figure 17. Overall species effects on mean values (±SE) o f cover per seedling (% seedling'1) for single-species seeding treatments across all substrate treatments in 10-year old stockpiled soil.......................................................................................................................................................88 Figure 18. Overall seeding treatment effects on mean values (±SE) of emergence (%; above) and seedling density (seedlings m'2; below) in NAG sand..........................................................90 Figure 19. Overall seeding treatment effects on mean values (±SE) of aboveground biomass (g m'2) across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled soil...................................................................................................................................................... 93 Figure 20. Overall seeding treatment effects on mean values (±SE) of estimated belowground biomass (g m'2) across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled so il................................................................................................................................... 95 Figure 21. Overall seeding treatment effects on mean values (±SE) of the shoot:root ratio across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled s o il.... 97 Figure 22. Overall seeding treatment effects on mean values (±SE) of the aboveground biomass per seedling (g seedling'1) across all substrate treatments in 2-year (above) and 10year (below) old stockpiled soil..................................................................................................... 101 Figure 23. Overall seeding treatment effects on mean values (±SE) of the estimated belowground biomass per seedling (g seedling'1) across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled s o il........................................................................ 102 List of Appendices Appendix 1. Details of the dates, locations and participants for the collection o f locally native seeds o f blue wildrye, arctic lupine and M erten’s sedge within the local area o f the Huckleberry Mine (10 km radius)................................................................................................. 147 Appendix 2. Photographs of each o f the test plot locations, including a) millsite, b) lower east dam, c) upper TMF-2 and d) TMF-2 north stockpile................................................................. 148 Appendix 3. Physical and chemical analyses conducted on soil samples collected from the 12 revegetation test plots in 2009 and 2010.......................................................................................149 Appendix 4. Photographs showing the difference in soil color (i.e., darkness) for a) soils supplemented with NAG sand and b) without NAG sand......................................................... 150 Appendix 5. Photographs displaying what appears to be a) surface sealing on stockpiled soils compared to b) the surface NAG sand supplemented so ils....................................................... 151 Appendix 6. Monthly total precipitation (mm) at Huckleberry Mine for each month in 2006, 2007, 2009 and 2010.......................................................................................................................152 Appendix 7. Monthly daily minimum (above) and maximum (below) temperatures (°C) at Huckleberry Mine for each month in 2006, 2007, 2009 and 2010........................................... 153 Appendix 8. A comparison of estimated and actual percent germination (%) for 2 and 10year old stockpiled soil test plot seeding treatments................................................................... 154 Glossary Acronyms AIC AMD ANOVA ARD ATV BC BCMOF BGC BROMCIL CCME CEC CEQG EC EFL EGC ELYMGLA EPA ESSF ESSFmc FESTSAX GPS ICP-AES ICP-MS IFS LUPIARC MZE NAG NPK PAR PLS RBCM SBS SBSmc2 SE SOM TMF UNBC USA Akaike Information Criterion Acid Mine Drainage Analysis o f Variance Acid Rock Drainage All-terrain Vehicle British Columbia British Columbia Ministry o f Forests and Range Biogeoclimatic Bromus ciliatus, fringed brome Canadian Council o f Ministers of the Environment Cation Exchange Capacity Canadian Environmental Quality Guidelines Electrical Conductivity Enhanced Forestry Laboratory Environmental Growth Chamber Elymus glaucus, Blue Wildrye Environmental Protection Agency Engelmann Spruce-Subalpine Fir biogeoclimatic zone Moist Cold Subzone of the Engelmann Spruce-Subalpine Fir Zone Festuca saximontana, Rocky Mountain fescue Global Positioning System Inductively Coupled Plasma Atomic Emission Spectrometry Inductively Coupled Plasma Mass Spectrometry Industrial Forestry Service Lupinus arcticus, arctic lupine Main Zone Extension Non-acid Generating Nitrogen-Phosphorus-Potassium fertilizer Photosynthetically Active Radiation Pure Live Seed Royal British Columbia Museum Sub-boreal Spruce biogeoclimatic zone Moist Cold Subzone of the Sub-Boreal Spruce Zone Standard Error Soil Organic Matter Tailings Management Facility University o f Northern British Columbia United States o f America xi Formulas CS2 CaCC>3 Ca(OH)2 CaO FeS2 NaOH NO 3‘ NH4+ P 0 43' KOH ROCSz'lvf (where R M+ = Na+ o rK +) Carbon Disulphide Calcium Carbonate Calcium Hydroxide Calcium Oxide Iron Disulphide Sodium Hydroxide Nitrate Ammonium Orthophosphate Potassium Hydroxide alkyl and Xanthate Acknowledgements This thesis would have not have been possible without the funding provided by Huckleberry Mines Ltd., the Mathematics of Information Technology and Complex Systems (MITACS) Accelerate Internship program and Cooper Beauchesne and Associates Ltd. I would like to thank Ron Robichaud o f Huckleberry Mines Ltd. for providing assistance with coordinating project efforts and providing equipment on the mine site. Thank-you to Laurence Meadows o f MITACS Accelerate for review and comments on the funding proposal for the MITACS internship program. I am grateful for the enormous effort put forth by the students o f the 2008 School o f Exploration and Mining program (Rod Johnson, Russell Gawa, Pamela Mikolayczyk and Armin Wesley) during the construction of the test plots. I would also like to express my gratitude to Brittany Melanson for her assistance with data and sample collection during monitoring o f the test plots in 2010. Thank-you to Carla Burton o f Symbios Research and Restoration for her moral support and continuous input into various aspects o f the project. I am indebted to Andrea Eastham o f Industrial Forestry Service Ltd. for providing a supply of grass seeds for the project. I am also indebted to the facilitators o f the Enhanced Forestry Laboratory at the University o f Northern British Columbia (Steve Storch, John Orlowsky and Doug Thompson) for providing equipment and guidance with the preparation o f soil and vegetation samples and seed germination trials. I offer m y greatest appreciation to the members o f my advisory committee (Mike Rutherford, Philip Burton and Paul Sanbom) for their invaluable guidance and for their review and comments on the project proposal and thesis drafts. Lastly, and most o f all, I would like to thank my partner, Sara Sparks, for all her love and support throughout all the stages o f this project. 1.0 Introduction 1.1 Mine substrates Depending on the type o f mine operation and the mineral resources being extracted, a variety of substrates are created as a result o f ground disturbance and the production o f waste materials from excavated surface and parent material substrates as well as from the processing o f mineral concentrated materials. Major mineral resources include heavy metals (e.g., gold [Au], silver [Ag] and copper [Cu]), coal, and industrial minerals (e.g., gypsum, talc and limestone). For each of these resources, the approach to ground extraction and refinement o f the minerals differ greatly and this has a significant impact on the amount o f disturbed ground, time period of disturbance and the production o f waste materials. Overall, there are three major substrates that are commonly produced on mined lands: disturbed and degraded soil and unconsolidated overburden, waste rock materials and processed mine tailings. The conditions associated with each o f these substrates differ greatly in terms o f their physical, chemical and biological properties, which determine the concentration and mobility of contaminants as well as nutrients, and subsequently the successful establishment and growth o f vegetation. Disturbed and degraded soils are produced when the materials are removed from parent ground and transported to storage locations in the process o f site clearing for mine operations. During this process, a considerable amount o f soil can be lost as a result o f handling and transport. For soil that is successfully salvaged, soil quality, which is defined as the ability of a soil to perform the functions and provide the elements essential for plant and animal growth (Brady and Weil 2002), is often greatly reduced as a result o f compaction 1 from heavy machinery, disruption o f soil structure and long-term storage. Compacted soils impede plant growth by reducing the interconnected void space between soil particles, which allow the storage and passage of air and water throughout the soil (Schor and Gray 2007). As a result, oxygen levels within soil are reduced, hydraulic conductivity (the ease o f water movement through soil down a gradient) is decreased and the movement o f nutrients and ions is restricted (Phelps and Holland 1987). Disruption to soil structure can also reduce the ability of water, air and roots to move through the soil by eliminating pore space between soil peds. This is specifically important for surface soils; if rainfall cannot penetrate the soil surface, it becomes runoff and can result in erosion and a reduction in the availability o f water to plant roots (Troeh and Thompson 1993). Long-term storage o f soils can result in a significant reduction in microbiological activity in the soil, which can prevent or delay normal nutrient cycling (when compared to undisturbed soils) following respreading during reclamation (Harris et al. 1989). In mining, overburden is the material overlying a mineral deposit o f ore or coal, which is removed in an effort to gain access to the deposit and is either stored for future use or disposed (Aiken and Gunnett 1990). This includes materials such as rock, soil, sands, silts, clays, shale and glacial till. Overburden can be consolidated (i.e., bedrock) or unconsolidated (i.e., loose and unstratified). Overburden materials are often used as alternative media for plant growth when supplies o f soil are inadequate (e.g., glacial till at Sullivan Mine, Kimberly, British Columbia [BC]). However, due to the composition and mode of deposition, the physical and chemical properties o f overburden can vary widely; hence their ability to support plant growth will depend on how conducive these properties are to plant growth. In a greenhouse experiment conducted by Byrnes and Stockton (1980), the 2 physical properties, chemical properties and the ability to support plant growth o f 18 overburden materials from five surface coal mines in southwestern Indiana, United States o f America (USA) were evaluated. The materials examined included unconsolidated A and B soil horizons, lacustrine sediments, glacial till and consolidated, but easily weathered rock strata. From the results o f the experiment, the authors concluded that electrical conductivity (EC) and water storage capacity were the most influential properties for plant growth and ranked their suitability as a plant medium as follows: lacustrine sediment > A horizon > B horizon = glacial till > brown shale > sandstone > grey shale > black fissile shale. Waste rock, produced mostly from open pit mining, is the material excavated from the parent substrate in order to access the mineral concentrated materials and is either stored in waste treatment ponds or in large quarries. Waste rock substrates are potential sources o f heavy metal contaminant leachates because exposure to surface weathering conditions can oxidize sulphide from pyritic minerals which then enter surface and ground water, reducing pH (defined as the negative logarithm o f the hydrogen ion (H+) concentration in a water solution; Troeh and Thompson 1993) and increasing soil salinity (Borden and Black 2005). These substrates are also poor media for plant growth, with little or no organic matter content, reduced nutrient content and harsh physical and chemical conditions due to potentially high metal concentrations (Lottermoser 2010). Mine tailings are a solid waste product o f the milling and mineral concentration process which are commonly produced in base metal and coal mining operations, and are stored in large tailing impoundments (Richmond 2000). During the milling process, the tailings are mixed with water in order to safely transport and store the material in an isolated impoundment. Tailings can have a high metal and sulphate content (if they originate from a 3 sulphide based ore) and in the presence o f water and oxygen, a highly acidic, sulphateenriched slurry, can be produced. Under these conditions, the metals, including iron (Fe), Cu, molybdenum (Mo), aluminum (Al), lead (Pb), zinc (Zn) and cadmium (Cd) are released into solution, resulting in toxic conditions for plants and animals. 1.2 Acid mine drainage In the presence of oxygen and water, sulphide-based minerals such as Fe pyrite or Fe disulphide (FeS2) can produce sulphuric acid, also known as acid mine drainage (AMD). This low-pH solution increases the solubility o f metals and can result in an effluent with high metal concentrations (Jennings 2008). At a typical metal mining operation which extracts a sulphide-based ore, AMD (also known as acid rock drainage; ARD) can be generated from structures such as waste rock dumps, tailings ponds, open pits and underground workings. Once generated, AMD can have significant adverse effects on downstream aquatic ecosystems (e.g., lakes, rivers). Eliminating or reducing the risk o f AMD generation can be accomplished through a variety of techniques. Overall, by limiting the exposure of sulphide rock to either air or water (or both), the potential for the production o f AMD can be greatly reduced (Johnson and Hallberg 2005). Often, a protective cover composed o f neutral material (e.g., glacial till, soil, water or synthetic materials) is applied to the surface o f tailings impoundments and waste rock dumps to seal and prevent exposure o f the sulphide-based minerals to the air and precipitation. With open pits in which the rock walls are exposed to air and water, the most common and most practical method for reducing the potential for AMD is to flood the pit with water. 4 1.3 Desulphurization o f mine tailings In the early 1990’s, treated tailings (known as desulphurized or depyritized tailings) were introduced as a cover material for tailings impoundments (Sjoberg et al. 2001). Desulphurized tailings are essentially a more refined product o f the milling process in which the sulphidic fraction of the slurry is removed through an additional flotation stage resulting in a relatively pH-neutral material (Benzaazoua et al. 2000). During the last few years o f mine life, treated tailings are produced and then used to cap the sulphuric portion o f the tailings impoundment with a layer 1 to 2 m in depth; this layer reduces the oxidation o f remaining sulphide minerals within the reactive tailings below by limiting oxygen and water availability (Sjoberg et al. 2003). Froth flotation (separation o f hydrophobic and hydrophilic material) is a treatment stage during the processing o f ore-rich rock in which valuable ore is separated from the rock using specialized reagents. These reagents increase the hydrophobicity (the property o f being water repellent) of the ore, allowing it to float to the surface of a flotation cell and become concentrated. In addition a secondary flotation stage can be used to remove a significant portion of the remaining sulphidic fraction o f the waste material. This results in the production o f a final tailings product which has low sulphide content and a low potential to generate acidity (Sjoberg et al. 2003). The sulphidic fraction o f the waste material is stored in a separate containment area (e.g., an isolated area of the tailings impoundment) where it can be managed more easily due to the reduced volume (Benzaazoua et al. 2000). The most common reagent used for non-selective flotation of sulphides are Xanthatebased (Benzaazoua et al. 2000). Xanthate (ROCS 2 _M+ [where R = alkyl and M+ = Na+ or 5 K+]) is a salt (dithiocarbonate) that is produced by a reaction o f alcohol with sodium hydroxide (NaOH), potassium hydroxide (KOH) and carbon disulfide (CS 2 ). Other common reagents include amine acetate, thiocarbamates and mercaptobenzothiazoles (Benzaazoua et al. 2000). Regardless o f their source, the physical properties of desulphurized tailings are similar to those o f untreated tailings, consisting primarily of silt and clay to sand sized particles with a low hydraulic conductivity (Sjoberg et al. 2001). In terms o f chemical properties, the tailings maintain a relatively neutral pH (slightly alkaline at Huckleberry M ine) and low cation exchange capacity (CEC) and EC. As it is strictly a product of mined rock, the material is devoid o f organic matter and only low concentrations o f available plant nutrients are present. 1.4 The use of soil amendments and the concept of Anthroposols A soil amendment can be defined as a material that is applied to a substrate in an effort to improve its properties by restoring the essential conditions required for adequate plant growth, and/or for reducing the mobility and bioavailability o f any soil contaminants. Amendments are often used in the reclamation o f soils and unconsolidated overburden, which in almost all cases, lack adequate conditions for plant growth. These substrates can generally be described as having low levels o f plant-available nutrients (e.g., nitrogen [N] and phosphorus [P]), high concentrations o f heavy metals, low organic matter content, high salt content and lower or higher than normal pH; EPA 2006). By prescribing the appropriate type of amendment(s) and determining the optimal rate, combination and mode o f application, organic matter content and nutrient availability can be increased, pH can be neutralized, microbial soil communities can become re-established and heavy metal contaminants can be immobilized. As a result, the conditions for plant growth are improved and the mobility and incorporation o f contaminants into ground water and plant tissues can be reduced. A multitude of investigations have been completed over the last half-century that have tested many different types o f soil amendments in tandem with the operation and decommissioning o f mining operations. In some cases, the choice of amendments has been based on addressing specific soil properties (e.g., highly acidic soils, Davis et al. 1999; microbial soil communities, Kubeckova et al. 2003). Amendments tested have been both organic (e.g., sewage sludge; Alvarenga 2009) and inorganic (e.g., bentonite clay; Schuman et al. 2005) in nature. Organic amendments have included municipal wastes (Fuentes et al. 2007), farming wastes (e.g., cow, pig and chicken manure), peat (Burton 2007), soils and wood wastes (Brown and Jackson 1984) and biochar (Lehmann 2007). Inorganic amendments tested have included marble wastes (Murcia et al. 2007), commercial lime products (Stuczynski et al. 2007), commercial fertilizers and coal fly ash (Kumar and Singh 2003). Overall, the types of materials used as soil amendments have been diverse and efforts to test new residual materials with beneficial properties when added to soil from other industrial and commercial processes will likely continue. Soil amendments can be used to address two main challenges that occur within soils and overburden found on mined landscapes: reducing the mobility and bioavailability o f contaminants to plants and animals, and improving substrate conditions to create adequate conditions for plant growth. To address these issues, a full understanding o f the physical, chemical and biological conditions o f the target substrate must be developed. W ith this information, the most appropriate type of amendment, rate o f application or amendment combinations and application methods can be chosen. The type of amendment used to reclaim a minesoil substrate depends on its specific physical, biological and chemical properties. In reclamation efforts, nutrient-poor disturbed and degraded soils are often treated with commercial fertilizers in order to improve plant growth conditions and their use has been shown to increase the production o f vegetation cover for a given density o f plant seed applied (Burton and Burton 2000). Organic amendments (e.g., municipal solid wastes and farm manure) and inorganic commercial fertilizers have been applied to waste rock substrates in order to promote vegetation establishment by supplying a sufficient source o f plant available nutrients. Research by Meikle et al. (1999) tested these amendments with grass seeding treatments on waste rock substrates and found that organic amendments substantially increased vegetation cover compared to mineral fertilizers which was likely the result o f enhanced nutrient and water retention ability (i.e., CEC) o f organic matter. In the case o f acid mine tailings and metal contaminated substrates, a variety o f organic and inorganic amendments have been used to reduce acidic conditions and reduce the mobility of metal contaminants and metal uptake by plants. For example, organic residues were added to acidic, metal contaminated soils at a mine in Portugal in order to neutralize acidity, increase organic matter content and reduce the availability to plants (Alvarenga 2009). Several approaches have been used to determine the most appropriate rate or combination o f amendments for application to target substrates. Soil conditions in undisturbed areas adjacent to disturbed sites may be used as a reference for determining what target conditions should be achieved. Previous reclamation and research efforts for similar sites and soil conditions may also be used to determine target conditions (EPA 2006). However, one o f the most effective approaches is to conduct research trials which test the effects o f amendment application on plant growth and metal uptake. Amendment applications have been tested both in greenhouse experiments and field trials to determine the most successful concentrations to promote or improve plant growth. In an experiment testing various rates o f sewage sludge (wet weight h a'1) to amend minespoil at a coal mine in Utah, USA, it was determined that low application rates (14 Mg h a'1) resulted in foliar N and P concentrations in seeded grasses equivalent to those achieved under the highest application rates (83 Mg ha'1; Topper and Sabey 1986). In a greenhouse experiment conducted by Reid and Naeth (2005a), native grass species were tested on kimberlite tailings amended with combinations of peat moss, paper mill sludge, lake sediment, sewage sludge, Agri-Boost™ , inorganic fertilizer and three calcium (Ca) sources to determine the best approach for reclamation. Results revealed favourable plant growth on sewage sludge, peat moss and paper mill sludge amended tailings and that the combination o f peat moss, sewage sludge and fertilizer produced the highest levels o f vegetation cover. Applying amendments to target substrates can sometimes be difficult due to logistical limitations (such as machinery access to target substrates, limited timing windows and transport costs) and soil compaction concerns. Phelps and Holland (1987) compared the effects o f bulldozers and small skid-steer loaders on soil compaction using bulk density methods during soil replacement efforts from soil stockpiles at a strip mine in Pennsylvania, USA. It was determined that the spreading o f soil by skid steers was more successful at alleviating soil compaction during soil spreading compared to bulldozers. The timing o f amendment application can also have a significant impact on its ability to improve soil 9 conditions. It has been suggested that excessive loss o f nitrates in nutrient rich soil amendments may occur in the winter if soil amendments are applied after the growing season; as well, the workability of a land surface may be degraded if amendments are applied during the rainy season (EPA 2006). With the prescription o f the appropriate type and combination of amendment(s), and the rate and mode o f application, the substrate conditions for plant growth can be improved and the mobility and incorporation o f contaminants within ground water and plant tissues can be reduced. The addition of amendments to these nutrient-limited and contaminated substrates has been shown to increase organic matter content and nutrient availability, neutralize pH, promote increased activity o f soil microbes and immobilize heavy contaminants. However, continued research is needed in order to discover new beneficial soil amendments that will provide cost-effective and successful approaches to amending mine soils, with an emphasis on testing the use o f industrial, commercial, municipal, agricultural or mining waste products. Soils which are highly modified or constructed through human activities are defined as Anthroposols (Naeth et al. 2012). These azonal soils are described as soils where one or more o f their natural horizons are removed, replaced, added to, or are significantly altered by human activity. Many growth substrates used at mine sites for reclamation activities meet this definition as they contain layers (i.e. “horizons”) that are anthropic in origin and contain materials that are significantly altered physically and/or chemically (relative to the original pre-disturbed soil found at the site). In addition to the use o f traditional soil amendments (e.g. use of lime to improve conditions for plant growth), a variety of waste materials produced from industrial, commercial and urban development have been added to soils 10 during land reclamation activities. These materials are most often added to soils as an opportunistic approach to waste disposal (e.g. wood and construction wastes; Murcia et al. 2007), or as a supplement to soils during construction o f major infrastructure. The use o f waste rock, tailings and other amendments (often at very high application rates) are common inputs into the creation of Anthroposols during mine reclamation. Anthroposols are key to the success of many reclamation projects as they are designed to produce a suitable growth substrate where native soils are lacking, in limited supply, or, are o f poor quality. 1.5 Soil properties relevant to plant growth In order to survive, plants must be provided with a growth medium that can promote root growth, accept, hold, and supply water and mineral nutrients and allow for gas exchange (Schoensholtz et al. 2000). Within any medium, these conditions relevant for plant growth are determined by its physical and chemical properties. A variety of physical and chemical indicators of soil quality have be used in the assessment of agricultural and forest soils as well as in assessing the suitability of waste materials as a media for plant growth in the reclamation o f mined lands. 1.5.1 Physical soil properties Physical soil properties are defined as characteristics, processes or reactions o f a soil that relate to its solid particles and how they are aggregated (Brady and Weil 2002). Physical properties used in the assessment o f soil quality include soil depth, texture, porosity, color and temperature, bulk density, soil strength, water-holding capacity and hydraulic conductivity (Schoensholtz et al. 2000). 11 Soil depth influences the availability o f resources required for plant growth including nutrients, water and oxygen; hence, greater soil depths often result in a greater availability o f resources (Schoensholtz et al. 2000; Bowen et al. 2005). To determine the depth o f soil required for maximum production o f grass cover at a few mines in Wyoming, Montana and North Dakota, USA, Barth and Martin (1983) tested native and introduced grasses on soil covers over mine spoil ranging in depth and observed a significant increase in grass productivity with increasing soil depth. Similar results were also obtained by Power et al. (1981) where yields o f four grass and legume crops planted in various depths o f subsoil over sodic mine spoil in North Dakota, USA, were shown to increase with increasing soil depth. Typical target depths for construction o f a soil medium on mined areas range from 0.4 to 0.6 m (e.g., North Antellope/Rochelle Mine, Wyoming, USA [Schladweiler et al. 2005]; Huckleberry Mine, Houston, BC [Boxill 2010]) and over the long term, have shown to result in the establishment o f productive vegetation cover (Bowen et al. 2005). Soil texture is described by the relative proportions o f sand, silt and clay in a soil and determines the transport, retention and uptake o f water, nutrients, and oxygen in a medium (Brady and Weil 2002). For example, sandy soils are often easily permeable to water, air and roots, yet are limited in their ability to store water and nutrients; in comparison, soils with high clay content may have high water-holding capacities but are poorly aerated (Troeh and Thompson 1993). Soil porosity is the percentage o f total soil volume not occupied by solid particles (Coyne and Thompson 2006). Pore size distribution and pore continuity directly influence root growth by determining the amount of soil volume filled with air and water as well as the soil’s ability to transport oxygen throughout the rhizosphere. At an air-filled porosity o f less 12 than 10 %, the oxygen diffusion rate is inhibited, causing injury to roots and reducing their ability to function (Silva et al. 2004). Soil color can be attributed to its organic matter content and mineralogy, and has the potential to influence soil temperature because darker surfaces can absorb and release heat more rapidly than lighter colored surfaces (Troeh and Thompson 1993). However, the ability of soil color to influence temperature is dependent on its soil moisture content. Dark colored soils are often high in organic matter content and due to their high water-holding capacity, remain wet and cooler than lighter colored soils (Troeh and Thompson 1993). In contrast, many dark colored waste materials that are often produced on mined lands (e.g., NAG sand, Huckleberry Mine, Houston, BC), have little or no organic matter content and low water holding capacity, allowing for heat to be more readily absorbed. Soil temperature can influence plant growth by regulating the chemical and biological processes that determine the availability and absorption of water and nutrients (Brady and Weil 2002). In addition to soil color and water content, temperature within a medium is also influence by depth. Overall, soil temperature decreases with depth with the greatest variation in daily temperatures occurring in the first 5 cm below the surface and the greatest annual variation in the first 50 cm (Schaetzl and Anderson 2007). Bulk density is defined as the mass o f a unit volume o f dry soil (Brady and Weil 2002) and is a measurement used to assess soil compaction and infer root growth potential and water availability in a soil (Schoensholtz et al. 2000). Average values for bulk density for loamy soils range from approximately 1.3 to 1.5 g cm'3 (Coyne and Thompson 2006). At values between 1.4 and 1.9 g cm'3 or greater, bulk density can begin to restrict root growth as 13 a result of high soil strength (Lampurlanes and Cantero-Martinez 2003). Soil compaction often results in high bulk density, which limits the ability for plants to establish on these soils due to high penetration resistance and results in poor root growth due to reduced aeration and water availability within the medium. Soil strength is the resistance that roots meet when penetrating the soil, either at the surface or within the medium. The most common indicators used to measure soil strength are bulk density and penetration resistance. Penetration resistance measures the pressure required to penetrate a soil surface. As penetration resistance increases, root growth decreases. At values greater than 2 MPa, significant root growth reduction has been reported (Lampurlanes and Cantero-Martinez 2003), leading to decreased plant nutrient uptake and plant stress (Reintam et al. 2009). Generally, penetration resistance varies with soil moisture; as soil moisture decreases, penetration resistance increases. In very dry soils, cementing may occur at the surface, significantly reducing plant establishment from seeds deposited by natural seed rain. However, coarse-textured soils may also exhibit reduced penetration resistance when very dry (Hillel 1998). Soil compaction can also significantly increase penetration resistance (Reintam et al. 2009). The two indicators that are most often used to describe the availability and movement o f water through a soil are the available water holding capacity and hydraulic conductivity, respectively. The available water holding capacity o f a soil is defined as the portion o f water in a soil that can be readily absorbed by plant roots; hydraulic conductivity is the readiness with which water can pass through a soil (Brady and Weil 2002). Water holding capacity is dependent on various physical soil factors including soil texture, type o f clay present, organic matter content, bulk density and soil structure. Sandy soils typically have a lower water 14 holding capacity than a loam or clay soil; soils with swelling clays (e.g., montmorillonites) also will hold more water than those o f non-swelling clays (e.g., kaolinite; Hazelton and Murphy 2007). Soils with low water holding capacity may not maintain an adequate supply of water for plants through periods o f drought. Hydraulic conductivity can be a good indicator o f how quickly and how efficiently water moves through a soil column, influencing water availability and movement down through the rooting zone. 1.5.2 Chemical soil properties Organic matter is considered to be one o f the key chemical soil properties when assessing soil quality for plant growth. It has a direct influence on aggregate stability, soil porosity, gas exchange and water and nutrient storage, release and availability (Schoensholtz et al. 2000). Other important properties that describe soil quality include levels o f available macronutrients and micronutrients, pH, CEC and EC. Soil organic matter (SOM) consists o f fresh residues and humus (organic matter derived from decomposed materials) and is one o f the main components o f soil to govern both its physical and chemical properties (Troeh and Thompson 1993). W ith the addition o f organic matter, soil porosity is increased, enhancing water infiltration and improving its availability to plant roots. Organic matter also has a high water holding capacity, which helps to hold moisture within the rooting zone. As a major source o f nutrients (e.g., N, P, sulphur (S), Ca and magnesium (Mg), micronutrients and trace elements), as well as a high CEC, organic matter contributes significantly to soil quality (Coyne and Thompson 2006). The addition o f organic matter also contributes to the soil structure and tilth by promoting soil aggregation and the development o f soil structure. Primary macronutrients (required by plants in large amounts) include N, P and potassium (K). Plant-available forms o f N include nitrate (NO 3 ) and ammonium (NHV^); the available form o f P is orthophosphate (PO 4 3'), while K is released from a variety o f mineral sources into the soil solution as K+. Secondary macronutrients include S, Ca and Mg. The most important forms o f S for plant nutrition within soils are sulphates and other mineralizeable organic compounds (Vanek et al. 2008). Ca and M g are both found as divalent ions (Ca2+ and Mg24) in the soil solution and on the exchange complex, which can be readily utilized by plants. Micronutrients (elements required by plants to complete their life cycle but only in small amounts; Troeh and Thompson 1993) include a variety o f trace elements such as boron (B), chlorine (Cl), Cu, Fe, manganese (Mn), Mo and Zn and are only required by plants in trace amounts. Trace element requirements vary somewhat with plant species (Brady and Weil 2002). These elements, released from mineral or organic material in the soil, form one or more ions which then enter the soil solution and become available for plants. Generally, soils with high clay or organic matter content will retain a greater amount o f soil nutrients than soils with low clay and organic matter content (Troeh and Thompson 1993). Soil pH describes the acidity or alkalinity o f the soil solution, which strongly influences the availability of nutrients for plant growth (Troeh and Thompson 1993). A plant’s ability to survive within a specific pH range depends on its ability to utilize nutrients at the concentrations available. Overall, nutrient availability is greater in neutral to slightly acidic conditions (pH 6-7) than under alkaline conditions or strongly acid conditions (Brady and Weil 2002). pH can also influence the concentration o f metal ions in the soil solution 16 (e.g., Al and Mn) which, at low pH can reach concentrations that are toxic for plants (Troeh and Thompson 1993; Delhaize and Ryan 1995). Cation exchange capacity is the defined as the sum o f the total exchangeable ions that a soil can adsorb per unit mass; the higher the CEC, the greater the capacity o f the soil to attract, retain and exchange positively charged ions (Coyne and Thompson 2006). The CEC of a soil is determined by the relative amounts o f colloids (organic and inorganic material with a small particle size and large surface area per unit mass) and their individual CEC (Brady and Weil 2002). Generally, colloids in organic matter have the highest CEC compared to colloids o f inorganic material (i.e., sand, silt and clays), followed by clays and silts and the lowest CEC for colloids in sand. Therefore, in most soils, a m ajority o f the CEC is contributed by the organic matter content; however, in soils low in SOM, clays contribute the greatest to CEC (Brady and Weil 2002). Factors which influence the CEC in a soil include organic matter content, texture and pH (Coyne and Thompson 2006). Organic soils generally have a higher CEC compared to mineral soils due to considerably high CEC o f organic colloids. In mineral soils, soils classified with high clay content have a higher CEC compared to coarse textured and sandy soils. Clay type can also influence CEC, as the exchange capacities of clay types can differ significantly (e.g., smectites have a much higher CEC than kaolinites; Brady and W eil 2002). pH influences CEC through the constant adsorption and release o f H* and Al3+ ions between soil particles and the soil solution; as pH increases, the amount o f H+ and Al3+ ions within the soil solution decreases, which allows for a greater availability o f adsorption surfaces on soil particles, increasing the CEC (Coyne and Thompson 2006). In addition to this pH-dependent 17 CEC, many soils exhibit a non-pH dependent (permanent) CEC, based on clay mineralogy (Brady and Weil 2002). Electrical conductivity measures the ability of a soil to transmit an electric current, and is commonly employed to describe a soil’s salinity; the greater the salinity o f the soil’s solution, the greater the conductivity (Brady and Weil 2002). As a general indicator, soils with an EC value greater than 4 dS m '1 are considered saline (Coyne and Thompson 2006). Soil salinity affects a plant’s ability to take up water and nutrients by increasing the soluble ion concentration in the rooting zone, altering osmotic potential in the soil (becoming more negative) and reducing water uptake by plants (Brady and W eil 2002). Under saline conditions, plants may also take up excess amounts o f sodium through a pathway that competes with potassium uptake, resulting in nutrient deficiency (Blumwald et al. 2000). Some plant species have adapted over time to tolerate very saline soil conditions while others can only survive on non-saline soils. 1.6 Revegetation practices on mined lands Prior to and following reclamation and closure o f mined lands, much research has investigated the establishment success o f native (indigenous to the region) and non-native plant species in amended and non-amended minesoil substrates (e.g., waste rock [Sharon and Smith 2002], metaliferous tailings and soils [Kramer et al. 2000; Macyk 2002; Reid and Naeth 2005b]). The objective of these investigations has been to develop an approach to establish a cover o f self-sustaining vegetation which produces a similar level o f productivity and land capability as that which existed prior to mining. Species tested in revegetation trials have included a variety o f graminoids (e.g., grasses; Burton 2007), forbs, shrubs and trees 18 (e.g., bigleaf maple [Acer macrophyllum]; Kramer et al. 2000). Several methods for the revegetation o f minesoils have also been developed and include direct seeding (Tordoff et al. 2000), planting nursery-grown seedlings (e.g., grasses; Sharman and Smith 2002), the use o f seed-rich soils (Zhang et al. 2001) and allowing natural colonization of disturbed ground (Borden and Black 2005). 1.6.1 Plant species selection When designing seeding and planting prescriptions for the revegetation o f mined lands, care should be taken to ensure that the plant species utilized are suited to the end land use objective (e.g., agriculture, wildlife habitat, forestry). Depending on the objective, factors such as climate, cost, availability o f seed, seedling stock and site conditions (e.g., soil texture, fertility and drainage) may also play a significant role in species selection. Where the end land use objective is agriculture (e.g., Afton Gold Mine, Kamloops, BC; Schmitt et al. 2008) or the establishment o f productive grassland habitat, supplies o f appropriate native and non-native agronomic grasses and legumes are typically available from commercial seed houses and can be applied to disturbed sites with relative ease and high success (e.g., Highland Valley Copper, Logan Lake, BC). Areas targeted for these end land use objectives are mostly on plains or in lowland valleys where the growing season is longer (compared to higher elevations) and where the supply and fertility o f soils are often more favourable for revegetation. If the end land use objective is to return wildlife habitat and plant communities to conditions present prior to disturbance, the selection o f plant species for revegetation is often considerably more complex. One o f the most limiting factors for the establishment o f native 19 plant species in reclamation is the availability o f plant material. Seed for some native species can be obtained from commercial seed providers or seeds can be collected from the region o f the minesite. Both methods are costly (when compared to obtaining agronomic seed stocks) and obtaining enough stock to seed large areas is often not practical. However, seeds obtained through either method can be sent to a seed grower to increase those seed stocks, or to a local nursery for propagation, as seedlings can then be outplanted, thereby maximizing the usage o f the costly seed stock in comparison to direct seeding methods. Often, minesites that designate wildlife habitat as their end land use objective are located on terrain and at elevations which present a variety o f challenges for revegetation. These sites are usually at subalpine to alpine locations (e.g., Huckleberry Mine, Kemess South, BC) where growing seasons are short and soils are often shallow (and therefore in limited supply as a medium for reclamation), poorly developed and have low fertility due to the slow rate of nutrient cycling (due to cooler temperatures). In addition, seeds for the native species which naturally inhabit these locations are often difficult to obtain in large quantities, and successful germination for nursery propagation often requires multiple trials to determine appropriate pre-treatments to break dormancy (Kaye 1997). Where the potential for soil erosion or mass movement has been identified (e.g., steep slopes), the use o f quick-establishing agronomic seed mixes has been warranted. However, in areas where the end land use objective aims for the renewal o f natural ecosystems and creating or enhancing wildlife habitat, the use o f these agronomic species m ay represent an incompatible option for revegetation. The establishment of non-native vegetation has been shown to preclude or inhibit the natural colonization o f native plants species and therefore reduce the land’s ability to rebound along natural successional trajectories (Sharman and 20 Smith 2002; Polster and Howe 2006). Overall, these studies examining the impacts o f introduced non-native vegetation have shown that quick establishment o f agronomic species may reduce some short-term impacts o f mining on soil quality (e.g., soil erosion, plant growth inputs; Forbes and Jefferies 1999). However, the long-term impacts o f reduced species diversity and poor prospects for the regeneration of natural plant assemblages (i.e., establishment of successionally stagnant grasslands; Carson et al. 2011) raises questions of their value for the conservation of natural ecosystems (Holl 2002). In more recent reclamation efforts across BC where the end land use objective is to return wildlife habitat, attempts have been made to initiate natural successional trajectories by establishing a cover o f native plant species (e.g., Kemess South mine; Lysay et al. 2010). Once established, studies have shown high success in encouraging the establishment o f other native species and initiating successional vegetation development on a site (e.g., red alder [Alnus rubra] at Island Copper Mine, Port Hardy, BC; Polster 2001). However, establishing an initial cover of pioneering species may be difficult and costly. Depending on the species, factors such as availability of seed and site conditions may make this method impractical. As a result, incorporation o f natural successional trajectories into reclamation plans has been minimal. 1.6.2 Seeding and planting Technical reclamation and restoration involves the use o f seed mixes, seedlings and stem cuttings for revegetating disturbed environments and many studies have demonstrated that these methods provide successful results (Macyk 2002; Gretarsdottir et al. 2004; Petterson et al. 2004; Olfelt et al. 2009). However, on highly disturbed ground and mine soils, the use o f 21 direct seeding has been shown to be more successful and less costly for establishing vegetation than the use of commercially grown or locally collected seedlings and stem cuttings, which often demonstrate poor survival and growth (Macyk 2002). Yet, when applied to areas where soil disturbance is minimal, (e.g., forest cutblocks, roadsides), the use of seedlings (specifically those that are propagated commercially) and stem cuttings can be an effective means o f establishing successful tree and shrub cover. The use of direct seeding for revegetation involves creating a seeding prescription which takes into consideration factors such as species composition, seeding density and method of sowing in an attempt to produce the most successful outcome possible for the target substrate and climate conditions. In terms o f species composition, the argument over the use o f native versus non-native species has been extensively debated (Jones 2003) and a multitude o f studies have be conducted which demonstrate both negative and positive benefits of using non-native species. Forbes and Jefferies (1999) describe how non-native plant species, which are used to provide a quick and temporary establishment o f vegetation cover to reduce risks o f erosion, will often persist and spread within disturbed environments, inhibiting the establishment o f native plant assemblages. In contrast, Antonio and Meyerson (2002) argue that careful investigation o f the influences of agronomic species may help to reduce the controversy surrounding their use in restoration and suggests circumstances where their application may be practical. Seeding density is an important consideration when producing a seeding prescription in order to maximize growth and productivity o f vegetation cover for a given area o f land. Burton et al. (2006) point out that low seeding densities will not fully occupy the growing space available and high densities may result in intense competition that inhibit growth and 22 productivity of individual plants. The study by Burton et al. (2006) suggests an optimal density between 750 - 1500 PLS (pure live seed) m '2 for revegetation efforts on disturbed soils in west-central BC. When considering seed mixes that incorporate species with different life histories (e.g., grasses and shrubs), other factors must be taken into consideration as well. Hild et al. (2006) state that the complex interspecific interactions between shrubs and grasses are not well known and in their study, examined the effect o f various grass seeding densities on the growth o f big sagebrush (Artemisia tridentata Nutt, ssp. wyomingensis) at a coal mine near Gillette, Wyoming, USA. The results showed that when seeded together, increasing densities o f grass lead to a decreased growth and productivity o f big sagebrush. Various methods have been employed in the application o f seed mixes on disturbed ground; these include hydroseeding, drill seeding and broadcast seeding. Hydroseeding involves the use of a slurry which contains the seed mix, water and wood fiber (and m ay also include a tackifier and fertilizer) that is sprayed on the target area using a large tanker with a hose and pressure pump. Drill seeding requires a large piece o f machinery known as a seed drill, typically designed for agricultural use or modified from such equipment (e.g., Truax native seed drill [Truax Company Inc.]) which cuts into the soil and distributes seed at an even rate to the freshly disturbed surface. Broadcast seeding, which is often preceded by soil cultivation or harrowing, involves evenly distributing the seed throughout the area, by hand or using a hand-held mechanical device such as a seed slinger (e.g., Truax seed slinger [Truax Company Inc.]) or a mechanical seed spreader mounted on the back o f a tractor or an all-terrain vehicle (ATV). Montalvo et al. (2002) compared hydroseeding, imprinting (using machinery which produces troughs that are simultaneously seeded) and drill-seeding 23 methods for sowing seed. The study found that plant productivity differed significantly among seeding methods and that some methods proved more successful for smaller- or larger-seeded species. Tree and shrub seedlings, transplants and stem cuttings have been used extensively for reclamation and restoration efforts (Brown and Jackson 1984; Macyk 2002; Huddleston and Young 2004; Petterson 2004; Olfelt et al. 2009). Overall success of commercially grown seedlings has been shown to be high. Olfelt et al. (2009) conducted a study in which seedlings of three-toothed cinquefoil (Potentilla tridentata) and poverty grass (Danthonia spicata), propagated under greenhouse conditions, were used in the restoration o f cliff edges in Tettegouche State Park, Minnesota, USA. Following three years of observation, survival rates were between approximately 60 and 90 % for transplanted seedlings. As well, Huddleston and Young (2004) reported average survival rates o f 98 % for planted seedlings of bluebunch wheatgrass (Pseudoroegneria spicata) and Idaho fescue (Festuca idahoensis) on ground colonized by Lemon’s needlegrass (Achnatherum lemmonii). However, their research also showed that growth and productivity of planted seedlings can be significantly affected by competition from existing vegetation cover. As well, the cost associated with seedling production can be high (e.g., shrubs) as they must be grown for two seasons or more in order to ensure survival during transplanting (Sharman and Smith 2002). The use o f transplanted seedlings and stem cuttings collected from donor sites has often been less successful, compared to seeding, and survival rates have varied. Macyk (2002) described a study in which coniferous seedlings (e.g., lodgepole pine [Pinus contorta var. latifolia]) and willow (Salix spp.) cuttings were planted in reconstructed soils on coal 24 spoil dumps, but demonstrated poor growth and survival due to conditions such as frost heaving and difficulty in planting. However, Polster (1997) has shown that stem cuttings can be used successfully to establish vegetation on landslides and unstable slopes through careful stem harvesting, preparation and planting methods. 1.6.3 Natural Regeneration The process o f natural colonization and succession can be utilized either partially or entirely as a suitable and successful method for revegetation o f disturbed environments. Natural regeneration o f vegetation occurs when an opportunity for colonizing a surface (exposure o f bare soil due to surface disturbance or an increase in light availability from canopy removal) occurs in combination with the presence o f viable propagules. Seeds, spores and rhizomes can be found in soil, forest floor and peat materials, while seeds and spores can also immigrate from off-site with the aid o f dispersal vectors (e.g., wind, wildlife). Depending on the type of receiving substrate, one source o f propagules may dominate (e.g., seed dispersal for non-soil substrates), and in other cases, seed bank and off-site sources may equally contribute. For example, Skrindo and Halvorsen (2008) found a greater frequency of woody and herbaceous perennial species indigenous to adjacent forests along recently constructed forest roads in Norway where soil was applied (with an abundant seed bank) compared to where subsoil (with a sparse or no seed bank) was used. Substrate type can also have a significant effect on the rate o f colonization and the abundance and diversity of plant species. Martinez-Ruiz and Femandez-Santos (2005) compared natural colonization o f soil-amended waste rock heaps at a uranium mine in Spain to those amended with fine 25 non-toxic sediment and found greater species diversity and abundance and reduced time for vegetation establishment on waste heaps amended with soil. Soils typically have great potential for inducing natural regeneration from an existing seed bank, which was deposited over time during previous generations o f vegetation cover. In one study, Iverson and Wali (1981) described seven factors which determine the quality and quantity of seed banks within soils (prior to disturbance), including the reproductive potential of plants, dispersal mechanisms, dormancy patterns, characteristics o f soils, biotic influence, weather fluctuations and disturbances to the soil surface. Overall, the study showed decreased seed density with depth within natural grazed and ungrazed prairie grassland soils prior to surface mining and suggested that a large majority o f the seed bank occurs within approximately the first 10 cm o f soil depth. For activities where ground disturbance is minimal (e.g., forest harvesting), the soil, seed and rhizome bank plays a significant role in the re-establishment o f plant cover and greatly influences the subsequent species composition. However, where soil disturbance is high or where developed soils have been completely removed (e.g., mined lands), the seed bank may have less influence or be completely absent. When such impacts can be foreseen, efforts can be made to salvage and conserve the valuable upper soil horizons in which a majority o f the seed bank is contained. Colonization through natural seed dispersal vectors can occur wherever a dispersal vector is in proximity to disturbed ground and propagule sources are found nearby. Success of colonization resulting from dispersal is largely dependent on two main factors which include the distance o f seed source (e.g., surrounding forest cover) from the disturbed ground and the receptivity (ability to provide conditions conducive o f germination requirements) o f its surface to seed. Distance from seed source has a significant effect on the level o f seed 26 contribution by dispersal. Matsumura and Takeda (2010) examined seed dispersal in semi­ natural grasslands and found that species richness significantly decreased with increasing distance from seed source. The distance from seed source can also have a significant effect on seed rain density. In floodplain forests along the upper Mississippi River, Minnesota, USA, Adams and Sorenson (2009) observed the seed rain density under forest cover and at stratified distances from the forest edge and found a decrease in seed rain density with increasing distance from the forest edge. In terms o f seedbed receptivity, some surfaces, such as freshly disturbed soil (Skrindo and Halvorsen 2008), provide ideal conditions for dispersed seeds while others (e.g., waste rock dumps; Borden and Black 2005) provide much more inhospitable conditions for seed germination and plant establishment due to harsh chemical and physical substrate conditions. Where natural regeneration is incorporated into a revegetation plan, the costs associated with seed stocks and expensive seedlings or transplanting efforts can be reduced or eliminated and successful revegetation attained. Matesanz et al. (2006) found no difference in plant cover, species richness and aboveground biomass between hydroseeded (using a non-native erosion control seed mix) and non-hydroseeded motorways in Malaga, Spain, and concluded that in some cases, seeding was not necessary in order to successfully revegetate a site. Even seeding with site-appropriate native species can, in some cases, prove to be less successful at initiating early serai native plant communities compared to areas that are naturally regenerated. Hodacova and Prach (2003) found a much higher species diversity on areas o f coal spoil heaps in the Czech Republic that were naturally revegetated compared to areas that were technically reclaimed with native plant species; these naturally colonized spoil heaps were also found to be further along a successional trajectory. 27 1.7 Introduction to Huckleberry Mine Huckleberry mine is an open-pit copper/molybdenum mine located approximately 86 km southwest of Houston in west-central BC. Construction and operation of the mine began in 1997. The minesite occurs within the Ootsa Lake watershed at the base o f Huckleberry Mountain at an elevation of approximately 1050 m. The main features o f the site include two open pits (the East Zone Pit and the Main Zone Extension [MZE]), a tailings impoundment (TMF-2), three dams (TMF-2 dam, East Dam and East Pit Plug Dam), till borrow pit, millsite, camp and sewage treatment plant (Figure 1). C am p and M illsite Figure 1. Overview o f the Huckleberry minesite, located approximately 86 kms southwest o f Houston, BC, as photographed in the summer o f 2007. 28 At Huckleberry Mine, vegetation and ecosystems in the surrounding area belong to the Sub-Boreal Spruce (SBS) and the Engelmann Spruce-Subalpine Fir (ESSF) biogeoclimatic (BGC) zones; this includes the Babine variant o f the moist cold subzone o f the Sub-Boreal Spruce (SBSmc2), with some areas o f the moist cold subzone o f the Engelmann Spruce-Subalpine Fir (ESSFmc) at higher elevations. The growing season is short, as a deep snowpack forms earlier and stays longer than most other BGC variants in the region. Forest cover is dominated by hybrid white spruce (Picea glauca x engelmannii) and subalpine fir {Abies lasiocarpa) with lodgepole pine dominating on drier sites and black spruce {Picea mariana) on wet sites and poor soils. Within the perimeter o f the minesite (approximately 1911 ha), the landscape is composed of various intensities of disturbed and undisturbed ground with patches o f vegetation cover consisting o f early colonizing native plant species. Disturbed areas are either subject to continuous disturbance from mining activities or have not been disturbed since the construction of the minesite. These areas initially disturbed during construction, but now inactive (e.g., spur roads, soil stockpiles), have been naturally colonized by local native perennials along with a few agronomic species (e.g., timothy [Phleum pratense], alsike clover [Trifolium hybridumj). In some areas, a low to moderate cover o f native woody plants has also begun to establish (e.g., TMF-2 North Stockpile). The most common species identified on disturbed soils within the minesite include: blue wildrye {Elymus glaucus), arctic lupine {Lupinus arcticus), bluejoint {Calamogrostis canadensis), M erten’s sedge 29 {Carex mertensii), common horsetail (Equisetum arvense), fireweed (Epilobium angustifolium), Sitka alder (Alnus viridus ssp. sinuata) and thimbleberry (Rubus parviflorus). Soils in the surrounding area o f Huckleberry mine are found on a variety o f surficial materials forming a thin veneer over glacial and ablation till. The main soil subgroups found include Orthic Humo-Ferric Podzols, Orthic Ferro-Humic Podzols, Sombric Humo-Humic Podzols, Lithic Humo-Ferric Podzols, Orthic Dystric Brunisols and Terric Mesisols. Organic matter is low in these soils except for in wetlands and pond areas. Soil are typically coarse-textured and rapidly draining and the distribution o f texture classes is sandy loam (42 %), loam (32 %), loamy sand (10 %), sandy clay loam (6 %) and silt loam (3 %). N and P levels in the Huckleberry Mine soils are relatively low in all soil types and strata and the pH is mainly acidic, ranging from 4.1 to 6.1 in the B horizon and from 4.5 to 6.3 in the C horizon. In baseline soil studies prior to development, high concentrations o f Cu, Zn and arsenic (As) were found (Boxill 2010). 1.7.1 Substrates available for reclamation The substrates available for reclamation at Huckleberry Mine include stockpiled soil, desulphurized NAG sand (utilized as a cover material for TMF-2), peat and glacial till. Soils stockpiled on the minesite originated during the initial construction of the mine (1996) and during the extension of the MZE in 2006. During construction, a limited amount o f peat was salvaged where deposits existed. Glacial till was removed from selected deposits throughout the minesite with one major deposit occurring at the west end o f the property. Desulphurized NAG sand currently caps the TMF-2 impoundment and is readily available for use. 30 During construction, overlying soil was salvaged wherever possible and subsequently stored in separate stockpiles at various locations on the minesite. Soil was salvaged as a mixture of both A and B soil horizons and is therefore classified as an anthroposol. When this thesis research was initiated, stockpiled soil salvaged during initial minesite construction was approximately 10 to 12 years old and newly stockpiled soil, salvaged during the construction of the recently developed MZE pit (Figure 1) was approximately 1 to 2 years old. Peat salvage occurred during the construction o f the mine and this material is stored in a single stockpile at the base of the TMF-2 dam. Attempts were made to salvage as much peat as possible from deposits; however, due to the logistical issues with hauling the water saturated peat, only a limited amount was salvaged for reclamation. Peat which could not be salvaged was buried during the construction o f the impoundment. It is estimated that due to the logistical challenges of removing and transporting peat, only a small portion o f peat will be salvageable from the current stockpile. A major deposit o f glacial till which was utilized for the construction o f TMF-2 is located at the west end o f the minesite (till borrow pit; Figure 1) along with a recently established pit near the millsite. This material can also be utilized as a medium for plant growth. Beginning in 2006, Huckleberry Mine introduced an additional flotation step in processing of ore for the production o f desulphurized tailings. The purpose was to provide pH-neutral material that would be used to create a capping layer on TMF-2. To date, the 31 impoundment is now an abundant source o f desulphurized tailings which can be easily accessed for use as a reclamation medium as required. 1.7.2 Overview o f mine closure plan The following is a summary o f the current reclamation and mine closure plan for Huckleberry Mine, submitted to the BC Ministry o f Mines in 2010 (Boxill 2010). As designated in the closure plan, the targeted end land use has been identified as forested land for wildlife habitat. Upon closure, all open pits and TMF-2 will be flooded. Around the perimeter of the TMF-2 pond and adjacent to all other dam crests (east dam and east pit plug dam), NAG sand beaches will be constructed to act as buffer between the maximum pond elevation and the dam embankment. Existing infrastructure (all buildings, roads, power lines and pipelines) will be deconstructed and removed from the site. Resloping and placement o f reclamation media (i.e., stockpiled soil) will be completed on all areas designated to be revegetated and will be applied to site-specific depths (Boxill 2011). The most abundant media available for reclamation at the mine site will be stockpiled soil. Areas in which reclamation media will be applied include all major dam faces, upper beaches of TMF-2, all road networks, soil stockpiles (once stockpiles have been exhausted), and the site o f the sewage treatment, camp and mill. The media will be applied to a depth o f at least 30 cm (based on available quantities of soil) on dam faces and approximately 50 to 200 cm depth on NAG beaches; in all other areas, the target depth for media will be approximately 30-50 cm as specified in the current reclamation plan (Boxill 2011). Revegetation o f the minesite will involve the use of a variety of plant species and revegetation techniques. The revegetation prescription is based upon the targeted end land 32 use objective for the mine, which is to re-establish vegetation cover that provides valuable habitat features for wildlife identified in the area (e.g., pine martin, black bear, grizzly bear and moose [Boxill 2010]). Plant species will include various grass and legume species (e.g., blue wildrye, fringed brome, alsike clover [Trifolium hybridum] and arctic lupine), shrubs (e.g., soopalallie [Shepherdia canadensis], Sitka alder and black huckleberry [ Vaccinium membranaceum] and trees (e.g., lodgepole pine, subalpine fir and trembling aspen [Populus tremuloides]). Grass and legume species will be applied in seed mixes either through broadcast seeding or hydroseeding. Shrub and trees seedlings will be propagated using a local nursery and will be planted with products designed to conserve soil water (e.g., anionic polyacrylamide) and supply nutrients (e.g., Nutri-Pak®) where necessary. On all major dam faces, a grass/legume seed mix (consisting of both native and non­ native species) will be used as an initial cover crop and will be applied by hydroseeding during the fall. After several years when grass cover is presumed to be less vigorous, shrubs and tree seedlings will be planted. Trees will be planted in areas o f the dam face where the soil depth is greater than 40 cm and at a density o f 1200 to 1400 stems h a '1. The suggested density for shrubs is 400 stems ha'1. The surface o f TMF-2 consists o f upper, above water tailings beaches (coarse sand), which make up the perimeter o f the impoundment, and a lower central region (with a high concentration of fine sands) that remains flooded throughout the year. On the upper beaches, a grass/legume cover crop will first be established followed by planting both tree and shrub seedlings with species selected based on microsite moisture conditions. In the lower 33 saturated regions, shrubs adapted to wetlands (e.g., willow and scrub birch [Betula nana]) will be planted. In addition, seeding and transplanting o f a few wetland herbs (e.g., sedges [Carex spp.] and cattail [Typha latifolia]) will be experimented with in an effort to establish littoral zone habitat (Boxill 2011). Other areas designated for revegetation such as the millsite, till borrow pits and haul roads will also be initially seeded with a grass/legume cover crop, followed by planting o f tree and shrub seedlings. Lodgepole pine and aspen are suggested where sites are more south-facing, with subalpine fir and spruce limited to cooler, wetter sites. Stocking rates for trees will be between 1200 and 1600 stems ha’1 and for shrubs between 400 and 600 stems ha’1 (Boxill 2011). O f the total 1911 ha of property, approximately 404 ha has been designated as disturbed areas to be reclaimed. Areas such as the open pits (East Zone Pit and MZE) and the central portion of the TMF-2 are not included as they will be permanently flooded. Following the completion o f all major closure objectives, post-closure monitoring will include periodic assessments o f the performance o f revegetation efforts. Assessments will determine whether vegetation succession is occurring on reclaimed sites and evaluate whether further planting, seeding and/or fertilization efforts will be required. 1.7.3 Past reclamation research at Huckleberry Mine In 2006, revegetation test trials were conducted on the Huckleberry minesite which investigated the success o f 10 perennial plant species grown from seed, sown within disturbed soil during the fall and then monitored at the end o f the second growing season 34 (Burton 2007). Seed treatments compared locally collected native seeds to genetically diverse native seeds from plants grown in cultivation (Burton and Burton 2002). Overall, the growth and survival o f plants grown from locally collected seeds were superior to those from seeds collected from cultivated plants o f the same species, exhibiting significantly higher cover, shoot and root biomass and overall plant biomass (Burton 2007). 1.8 Thesis Research Objectives The availability o f NAG sand, obtained from desulphurized tailings at Huckleberry Mine, offered an opportunity to assess a possible low-cost solution to the challenge o f limited soil availability, a common problem on mined lands in BC. It was hypothesized that by supplementing stockpiled soil with NAG sand, the amount o f suitable growth medium available for reclamation could be increased. As a Masters of Science thesis project, the objective was to evaluate the use o f NAG sand and fertilizer as a supplement to stockpiled soil in an effort to increase the quantity o f suitable plant growth medium available at the site. The suitability o f the medium was assessed by comparing a) physical (e.g., texture) and chemical soil properties (e.g., trace elements and plant nutrients) and b) plant performance (establishment and productivity) between soils supplemented with or without NAG sand. To clarify measures o f plant performance, establishment here is defined as the ability o f a seed to successfully establish on a soil (e.g., % emergence, seedling density) and productivity is the defined as the size, form and mass o f a plant (e.g., plant height, % cover and above- and belowground biomass). In addition, the specific performance o f blue wildrye established from locally collected and genetically diverse seed stocks was compared to evaluate the effects of seed origin on plant establishment and productivity. 35 1.8.1 Objective 1: NAG Sand as a Supplement to Stockpiled Soil The use o f stockpiled soil for reclamation is an effective approach; however, due to poor salvaging and storage techniques, supplies o f soil are often limited. As an alternative to locating additional sources of soil, this experiment sought to determine whether the addition of NAG sand could act as a volume-enhancing supplement to stockpiled soil, increasing the amount o f suitable medium available for reclamation. In order to measure medium suitability, plant performance in soil supplemented with or without NAG sand was compared. To proclaim soil supplemented with NAG sand as a suitable medium for plant growth, it was decided that plant performance would have to be relatively equal to or greater than the performance observed on non-supplemented soil. To further test the suitability o f soil supplemented with NAG sand as a growth medium, two stockpiled soil sources, differing in age since salvage (2 and 10 years) were examined. The age and composition o f stockpiled soil at Huckleberry Mine varies considerably and therefore, it was important to ensure that the criterion o f a suitable medium was met for a variety o f soils supplemented with NAG sand. In addition, to ensure that comparison o f plant performance between soils supplemented with or without NAG sand was not limited to a single plant species, a variety of plant species with different growth forms were tested (i.e., grasses and legumes). 1.8.2 Objective 2: Comparing the Performance o f Native Grasses and Legumes When establishing vegetation cover on disturbed soils, where conditions for plant growth are often less than optimal, the most appropriate species for revegetation are those that can successfully establish in poor growing conditions, provide dense and long-term vegetation 36 cover and require little or no maintenance. To identify candidate species for revegetation at Huckleberry Mine, the performance o f five native species (three grass, one sedge and one legume) sown in supplemented and non-supplemented soils was compared. The evaluation of these species was considered important as grasses, sedges and legumes all have different regenerative strategies and a variety o f these species types are currently colonizing the minesite. The candidate species selected were native to the region of the minesite and were chosen based on the availability o f seed and ease o f seed collection. 1.8.3 Objective 3: Influence o f Genetic Origin on Performance o f Blue Wildrye Collecting native seed from areas surrounding the minesite is costly compared to obtaining similar seed stocks from local suppliers. To assess whether the higher costs o f obtaining local seed stocks are justified, the performance o f locally collected and cultivated varieties o f blue wildrye was examined in this experiment. In effect, the objective was to compare the performance o f blue wildrye established from locally collected (local genotype) and genetically diverse (mixed genotype) seed stocks to evaluate the effects o f seed origin on plant establishment and productivity. 37 2.0 M ethods 2.1 Seed Collection During the minesite reconnaissance in late June 2008, a few local native perennials were selected as candidates for incorporation into revegetation test plots. The search for these candidates was conducted in areas o f disturbed ground on the minesite property that had been colonized by native vegetation, most notably along the western boundaries. The most common species identified within these areas included blue wildrye, arctic lupine, bluejoint, reedgrass, Merten’s sedge, common horsetail, fireweed and Sitka alder. Candidate species were evaluated for their potential ease of seed collection and overall success o f propagation. Three species were chosen for incorporation into test plots; these were blue wildrye, arctic lupine and Merten’s sedge. In August and September, 2008, seeds from candidate species were collected from the local area (10 km radius) at appropriate dates to their expected ripening times. Seeds for arctic lupine were collected in the first week o f August and seed for blue wildrye and Merten’s sedge were collected in early September. Geographic coordinates (from a global positioning system [GPS] receiver) and dates were recorded for each area in which seeds were collected (Appendix 1). Once seed materials had been collected, recommended methods for each species were utilized to clean and prepare the seeds for sowing the test plots. The cleaned seeds were stored in low light conditions at room temperature between the time of collection and seeding efforts. Seed cleaning and preparation methods followed procedures recommended in a local 38 manual that included species-specific guidelines for the three species collected (Burton and Burton 2003). 2.2 Seed Preparation 2.2.1 Arctic Lupine Lupine seeds were collected on August, 2008 at four locations within the local area o f the minesite. Sites included the two locations around the TMF-2 north stockpile, a location along the Morice Forest Service Road at km 114 and an inactive airfield at km 111. At each location, lupine plants were assessed for pod ripeness (see Figure 2a for an example o f ripe pods). Where ripe pods were found, the entire inflorescence was harvested using clippers. Harvested inflorescence were collected in small brown paper bags and then transferred to larger paper bags for transport and storage. Following field collection efforts, the resulting pod stock was stored in large paper bags (sealed closed) for approximately two weeks to allow pods to fully ripen and dehisce. Seeds from dehisced pods were then collected from the bottoms o f each bag. Remaining unopened pods were crushed by hand and then crushed material was filtered through two different screens (4 mm and 1 mm mesh; Figure 2b and 2c, respectively) in order to extract seeds. A final hand screening process was completed to eliminate insect larvae from the seed stock; this insect was probably the lupine aphid, Macrosiphum alibifrons (Cohen and Mackauer 1986; Figure 2d). 39 c) d) Figure 2. Seed collection and preparation o f arctic lupine with a) example o f ripe pods (black and dry) during collection, b) screening o f crushed pod material for initial and c) final screening; d) insect larvae were separated by hand from final seed stock (close up o f larvae top left). After seed cleaning, an estimated 830 g o f pure seed were collected. An analysis o f seed purity by weight determined that the cleaned seed lot was approximately 99 % pure. The total number o f seeds available was then calculated based on an average number o f seeds per gram (determined by taking an average o f the total number o f seeds in three random one gram samples, as per Burton and Burton 2003). With an average o f 108 seeds g '1, it was estimated that approximately 90,000 lupine seeds were collected. Samples o f lupine seeds from each collection location were then sent to a commercial seed testing laboratory (20/20 Seed Labs Inc., Nisku, Alberta) for viability testing (“Between Paper” method at 20°C for 10 days; light and dark periods, 8 and 16 respectively) to determine the percent germination capacity o f each seed lot. 40 2.2.2 Blue Wildrye and M erten’s Sedge Seeds from blue wildrye and M erten’s sedge were collected from ten locations around the minesite (six within the current minesite boundaries and four outside the boundaries) between September 3rd and 16th, 2008. Pure seed from blue wildrye was collected by running hands through the inflorescence and accumulating loosened seeds, or the inflorescence was clipped from the stalk for processing later. Seed material from Merten’s sedge was collected by clipping the inflorescence from stalks. The seed material for both species was collected in brown paper bags. Seed material collected from blue wildrye was cleaned using soil screening equipment available in the metallurgical laboratory at Huckleberry Mine. Before screening, detritus material and chaff was removed by hand. Using a series o f six soil screens, decreasing in mesh size from top to bottom, the hand cleaned material was placed in the top screen, with the largest mesh size. Screens were then placed in the sieve shaker and processed (vibrated or shaken) for ten minutes. The shaking separated m ost o f the seeds from the seed heads, with seeds retained in one or two o f the middle sieves. After the sieving process, the seed stock o f blue wildrye was determined to be approximately 100 % purity by weight. For Merten’s sedge, the attempt to purify seed stock through the series o f sieves proved unsuccessful, as floral bracts were very similar to the seeds in size and shape. Therefore, seeds (with bracts) were separated from the inflorescence by hand and a seed purity analysis was conducted in order to estimate the number (proportion) o f seeds within the total seed stock material. Purity was determined to be 85.5 % by weight. 41 2.3 Test Plot Construction A total o f twelve revegetation test plots were constructed at four different locations (three test plots per location) at the Huckleberry minesite between September 22nd and October 6th, 2008 (Figure 3). Test plot locations were chosen based on criteria that included accessibility to heavy machinery for ground preparation and substrate delivery (stockpiled soil and NAG sand) and low risk o f disturbance from mine operation over the lifetime o f the project. Before test plots were constructed, the ground at each location was leveled to create similar surface conditions for each site. Following site preparation, stockpiled soil and desulphurized NAG sand were delivered to each site and piled for use in the construction o f the test plots; approximately 9 m3 o f desulphurized NAG sand and 27 m3 o f soil from stockpiles was delivered to each site. At each o f the four locations, two supplemented soil test plots were constructed using soil from either a 2-year soil stockpile (sourced from the stockpiles south o f the MZE pit; Figure 3) or 10-year old soil stockpile (sourced from the TMF-2 North Stockpile) and desulphurized NAG sand (sourced from a single location within the TMF-2 impoundment). Test plots at two locations (TMF-2 North Stockpile and Upper TMF-2 test plots) were constructed using 10-year old stockpiled soil and the remaining two locations (Lower East Dam and Millsite test plots) were constructed using 2-year old stockpiled soil. In addition, a single NAG sand test plot was constructed at each o f the four test plot locations. 42 Figure 3. An aerial view o f Huckleberry mine (2011), located 86 kms southwest ofH ouston, BC , indicating the four locations o f the revegetation test plots (TM F-2 North Stockpile, Upper TMF-2, Lower East Dam and Millsite) and the MZE stockpile. Supplemented soil test plots were composed o f 32 subplots, each measuring 10.4 m by 5.2 m, arranged in a split-strip-plot design (Little and Hills 1978) and NAG sand test plots were composed o f eight subplots (Figure 4). The supplemented soil test plots were divided into two equal sections, each consisting o f 16 subplots. In one section, the subplots were composed o f a stockpiled soil and the other section, a 50:50 mix (by volume) o f stockpiled soil and NAG sand. Within each o f these treatment substrates, each of the eight subplots was fertilized with 57.69 g (equivalent to 576.9 kg h a '1) o f 13-16-10 NPK (74.9 kg h a '1N, 92.3 kg ha'1P20 5 and 57.7 kg h a'1K20 ; Evergro™, Kelowna, BC) fertilizer and the remaining eight were left unfertilized. The rate o f fertilizer applied to subplots was based on a similar 43 rate used in revegetation research plots established at Huckleberry Mine in 2005 (Burton 2007). A list o f test plot treatment substrates is provided in Table 1. NAG Sand Test Plot NAG Sand Supplemented Soil Test Plot NAG Sand: Topsoil No Fertilizer Topsoil Fertilizer Figure 4. Overview diagram o f soil treatments (each square representing a subplot) for N A G sand test plots (left) and the supplemented soil test plots (right). Substrate composition and fertilization treatments were prepared using two 0.085m3 gas powered cement mixers and mixed material was transported between substrate source piles and subplots by shovel and wheelbarrow (Figure 5a, b). In order to create the 50:50 mix of soil and NAG sand, equal amounts o f each material were shovelled into cement mixers and allowed to mix at a medium rotation speed for approximately two minutes (Figure 5c). Fertilization followed the same procedure in which the material for the top 10 cm o f each fertilized subplot was mixed in a cement mixer with the prescribed fertilizer for two minutes (Figure 5d). 44 Table 1. A list o f the nine treatment substrates for revegetation test plots established in September 2008. Substrates Stockpiled soil Stockpiled soil Stockpiled soil Stockpiled soil Stockpiled soil Stockpiled soil Stockpiled soil Stockpiled soil NAG sand Stockpile Age 2 2 2 2 10 10 10 10 N/A 50:50 NAG Sandisoil No No Yes Yes No No Yes Yes No Fertilizer No Yes No Yes No Yes No Yes No The target surface area for each subplot was 1.0 m2. To achieve this, a 2.6 m2 ground surface area, designated for each subplot was covered with the treatment substrate, resulting in a substrate depth o f approximately 0.2-0.25 m and 0.3 m buffer around the subplot perimeters (Figure 5e). Following the completion of subplots, buffer spaces were backfilled using neutral parent soil from the area surrounding the test plots (Figure 5f). For photographs o f each o f the four test plot locations upon completion of construction, see Appendix 2. 45 Figure 5. Photographs illustrating the construction o f the test plots, showing a) mixing o f soil and N A G material, b) transporting mixed material to subplots, c) procedure for mixing substrates, d) m ixing fertilizer with the top layer o f fertilization subplots, e) standardized dimensions o f subplots and f) backfilling buffer spaces between subplots. 46 2.4 Test Plot Seeding In total, seven seeding treatments were applied to all 12 test plots initiated on the minesite (Table 2) on October 5th and 6th, 2008. Treatments consisted o f sowing either a single plant species or a mix o f species. Seeds stocks included: 1) native seeds collected from the local area o f the mine (locally specialized genotypes) and 2) seeds obtained from seed increase plots where seeds from a variety o f locations were sown to propagate genetically diverse seed crops (mixed genotype or ecovars) grown at the Industrial Forest Service Ltd. (IFS) Ness Lake Nusery, located northwest o f Prince George, BG. Treatments were applied to all subplots at a standardized sowing density o f 750 PLS m '2 which was chosen based on optimal sowing densities recommended as the result o f a previous study (Burton et al. 2006). The amount o f seed to weigh out in order to obtain the desired PLS was calculated by adjusting the bulk seed lot weights on the basis o f their germination and purity values (Table 2). For locally collected seeds, percent germination was determined from suggested values reported by Burton and Burton (2003) and percent purity and seeds g"1were determined by weight from samples of each seed lot. Percent germination and purity values and seeds g '1 for genetically diverse seeds were obtained from previous seed germination tests for seed from the same increase plots (20/20 Seed Labs Inc., Nisku, Alberta). Prior to seeding treatments, seed quantities for each subplot were weighed using a scale with a 0.01 g resolution (ProScale; LC-50; Fletcher, NC, USA) and then packaged in paper envelopes (Figure 6a). Test plots were sown during the period o f October 2008. At each location subplots for each treatment were individually hand sown (sowing was conducted by only one 47 person to prevent bias (Figure 6b). Immediately after sowing each subplot, the soil surface was lightly raked to promote good seed/soil contact and to prevent seed loss from wind (Figure 6c). An example o f seed distribution across the surface o f the subplot is shown in Figure 6d. A diagram displaying the arrangement o f seeding treatments within test plots is shown in Figure 7. c) d) Figure 6. Images illustrating procedures associated with test plot seeding, including a) w eighing and packaging o f seed quantities for subplots prior to sowing b) hand seeding procedure c) raking after sow ing to promote seed germination and d) representative seed coverage across the 1.0 m2 surface o f a subplot (before raking). 48 Table 2. Seed treatment types and derived bulk seed application weights for 1.0 m2 subplots based on estimated percent germination and percent purity values for each seed lot. Seed T reatm ent Seed Stock Composition by weight (% ) Seeds g"1 Germination (% ) Purity (% ) Bulk Seed Application (g m ) ecovar ecovar local ecovar local 100 100 100 100 100 420 219 219 1500 108 80.5 86.5 80.0 95.0 44.3 79.0 85.0 99.0 96.0 97.0 2.81 4.66 4.32 0.55 16.16 ecovar 33.3: 33.3: 33.3 Single-Species fringed brome blue wildrye blue wildrye Rocky Mountain fescue arctic lupine Mixed-Species blue wildrye: fringed brome: Rocky Mountain fescue blue wildrye: arctic lupine: Merten’s sedge , ____- ____ i » . _ i- j _____ . 1 1 .2 ( 7 5 0 P L S m "- x % c o m p o s i t io n bulk seed application = I — ;------------------------- 33.3:33.3:33.3 86.5: 80.5: 95.0 i.. 80.0: 44.0: 37.2 85.0: 79.0: 96.0 99.0: 97.0: 86.4 1.55: 0.94: 0.18 1.44: 5.39: 0.58 n ' L -..- ' “Amount o f bulk seed stock sown on each lm subplot surface. The prescribed application is based on the desired seeding density o f 750 PLS m ', where a * .. local 219: 420: 1500 219: 108: 1555 V ,™ tw \ . \ S e e d s g*‘ * % g e r m i n a ti o n * % p u r ity J 49 In addition to the seven seeding treatments mentioned above, a non-native domestic seed mix treatment was also applied to the test plots during the seeding dates. However, due to a miscalculation, the sowing rate o f the domestic seed mix treatment (estimated at >4000 PLS m'2) was considerably higher than the target seeding rate (750 PLS m '2) and thus, was not comparable with the other seeding treatments. To compensate, the domestic seed mix treatment was removed from the subplots in the following spring (top 5cm o f subplots removed by shovel) and the subplots were designated as non-vegetated subplots (Figure 7). S upplem ented Soil T est Plot NAG S and T est Plot 1-----------Blue Wildrye (Ecovar) I 1 Fringed Brome (Ecovar) | 1 Rocky Mountain F escu e (Ecovar) I 1 Native S p ecies Mix (Ecovar) i 1 Blue Wildrye (Local) 1 1 Arctic Lupine (Local) I 1 Native S p ecies Mix (Local) 1 1 Non-Vegetated NAG S an d B lue Wiidryeg (Ecovar) Blue Wildryefj (Local) F ringed B rom e® Arctic Lupine} (Ecovar) | (Local) B lue Wiidryeg (Ecovar) B lue Wildrye} (Local) F rin g ed B ro m e| Arctic Lupineg (E covar) 1 (Local) Rocky M ountain I Native S pecies] Rocky Mountain Native S p ecies} F e sc u e (E covar)i Mix (Local) F e s c u e (Ecovar) Mix (Local) Native S p e c i e s * Non.Ve^ S Native S p e c ie s i Non. Ve^ t a t e d ] Mix (Ecovar) “ —---------* Mix (Ecovar) ■ -1 NAG S and: Topsoil Topsoil Fertilizer No Fertilizer ms Figure 7. Overview diagram o f seeding treatments (each square representing a subplot) for N A G test plots (left) and supplemented soil test plots (right). Subplots were either seeded with one o f the seven seeding treatments or designated as non-vegetated. 50 2.4.1 Prior to Test Plot Construction and Seeding Due to the short time period between seed collection efforts and test plot seeding, germination testing of locally collected seeds was only conducted for arctic lupine. Four samples (100 seeds per sample) were sent to 20/20 Seed Labs Inc. where germination testing was conducted within a temperature controlled environment (daytime period o f 8 hours at 20°C) for a 20 day period. 2.4.2 Following Test Plot Constmction and Seeding Germination testing o f the seed lots used in test plot seeding treatments was initiated at the University o f Northern British Columbia (UNBC) Enhanced Forestry Laboratory (EFL) on April 8, 2009. These tests were completed in an effort to acquire actual germination rates in comparison to the estimated rates used in calculating bulk seed application rates. The actual germination rates could then be used to more accurately report sowing rate (PLS m '2) as implemented, and percent emergence. In total, four replicates o f each of the seven seeding treatments were tested within an Environmental Growth Chamber (EGC; Model GCW 30, Chagrin Falls, Ohio, USA). For each replicate, 100 seeds of a seed lot were counted and placed in a clear plastic germination test box (Tristate Plastics, Dixon, Kentucky). Additionally, a pre-treatment was applied to the locally collected arctic lupine seeds (knicking seed coat with a razor blade) to hasten germination. Seeds were applied to a thick, pre-moistened pad of cellulose wadding (Kimpack®) using distilled water in germination test boxes and then placed in the EGC for incubation. Replicates for each seeding treatment were evenly distributed throughout the chamber (4 groups o f replicates, non-randomly placed in the EGC) to account for any variability in light and temperature in the EGC. 51 The controlled conditions within the EGC were as follows: photosynthetically active radiation (PAR) at 600 W m'2 for a daytime period o f 14 hours and a day/night temperature regime o f 20/10 °C. During incubation, germination test boxes were monitored every two to three days to ensure that paper cloth substrates remained moist. As a preventative measure, a fungicide (No-Damp™) was diluted into water and used to keep cloth substrates moist (1:100 No-Damp: Water dilution). During monitoring, the number of seeds that had germinated was recorded for each germination test box. Seed germination (defined as penetration of the seed coat by the radicle; Bewley 1997) was monitored until no new seed germination was observed for a consecutive four-day period; observations were completed over a 60-day period. Germinated seeds were not removed from the germination boxes until observations were completed. 2.5 Test Plot Monitoring 2.5.1 Year One (2009) 2.5.1.1 Vegetation Sampling Sampling o f vegetation was conducted for 252 out o f the 288 1.0 m2 subplots in the twelve established revegetation test plots during late June and August, 2009 (36 subplots were designated as non-vegetated, Figure 7). Each subplot was divided into a grid consisting o f 16 sampling grid locations (625 cm2 area [25 cm x 25 cm]). Using a random number generator in Excel (Microsoft Office 2007), a single grid location for each subplot was chosen for sampling (Figure 8a). In subplots sown with a single-species seeding treatment, sampling included a count of seedlings (seedlings m '2; Figure 8b), percent cover (Figure 8c) in both June and August, and plant height (Figure 8d) in August (no seedling counts were completed 52 for Rocky Mountain fescue {Festuca saximontana) due to the difficulty in distinguishing individual seedlings). For subplots sown with mixed-species seeding treatments, cover was evaluated and measurements o f height were taken for each species. Sampled grid locations were marked using four pins and flagging tape (blue) at plot comers for future reference. ‘ ■■' ’ ■ '^ ... i t ' ’- ' ; ■ •, ■ , ■' 'V 7 'c . ^ ■■■ ■ - ' ; ' ■ fjy u ;. * v ’- ^ Jr\ A • ■ a 4-‘ • -v - "; ' f> ' t7 ~ ■’ > ■ j'-' ■ ..'- 'A - I : ■>%£•- \?'tc -J j|p*^ > V ' v ’’’ •/ k i £.' . A \ A o .& * * * ^ y - • \\ V _ \iL J ~A „ „ 'A * * * * i r A*"- A 'A A A , Y - i r * ^ ; ; ■- a *! ’£''=■ 7 = < *'• j f 1 > - n 9 A * - * w ** f . »5 i-: ~ , A A A tjv ** W b) a) ■J . fk ; A j J- ^ * v*'/r* L F* . > *-• dc£ m HW5 mem l a Figure 8. Images from Year One vegetation sampling, completed in June and August 2009 show ing a) a 25 cm x 25 cm sampling grid location (comers marked by blue flagging) with a 1 m x 1 m wood frame to delineate the 16 possible sampling grid locations, b) an example o f seedling density observed in subplots o f blue wildrye, c) measurements o f percent cover for sampling plots, and d) measuring plant height. 2.5.1.2 Substrate Treatment Sampling Soil samples representing each of the nine treatment substrates (Table 1) were collected from the 252 of the 288 1.0 m2 subplots within the 12 test plots in August 2009 (non-vegetated subplots were not sampled as a subplots were significantly disturbed during the removal of 53 domestic seed mix). Soil samples were collected from subplot sampling groups (each sampling group representing a single treatment substrate; Figure 9). The sampling design was chosen in an effort to obtain a sample o f each treatment substrate that was representative across all subplots. In each test plot, one sample was collected for each subplot sampling group (two from each o f the four NAG test plots and eight from each of the eight supplemented soil test plots) for a total o f 72 samples. Each sample was a composite, composed o f core samples collected from all subplots designated to a specific sampling group. Within each subplot, cores were collected from three random locations w ithin a 625 cm2 sampling grid using a regular 7.6 cm diameter soil auger to the depth o f approximately 15 cm (using a 16.5 cm long auger bucket). Remaining auger holes were filled with soil from the surrounding area and marked using a pin and flag to avoid soil sampling there in the future. Any damage to subplot vegetation was noted. All cores collected for each composite sample were collected into a 20 L bucket and mixed prior to packaging. Samples were packaged in ziplock bags, appropriately labelled and stored in coolers and were transported from the minesite to UNBC for sample preparation. Initial sample preparation was completed at UNBC and then sent to the Ministry of Forests and Range (BCMOF) Research Branch Laboratory in Victoria, BC, and ALS Canada Ltd., Vancouver, BC, for analysis. Sample preparation included air drying o f samples (for at least a seven day period) followed by sieving using a 2 mm sieve. Coarse fragment (material >2 mm) content (percent by mass) for each sample was recorded after sieving. The fine soil fraction material was then packaged in small cardboard sampling boxes and shipped to the BCMOF laboratory for analysis. Table 3 lists the chemical and physical properties tested; further references for analyses conducted are listed in Appendix 3. Following return o f 54 samples from the BCMOF laboratory, a subset o f the samples was sent to ALS Canada Ltd. laboratory (Vancouver, BC) for total elemental analysis (Table 3). NAG Sand Test Plot 1 2 1 2 1 2 1 □ | 1 NAG Sand n Soil Supplemented Test Plot N on-V egetated NAG Sand: Topsoil No Fertilizer Topsoil Fertilizer Figure 9. A diagram displaying subplot sampling groups from N A G sand test plots (left) and supplemented soil test plots (right). Each subplot is designated to a specific sampling group. Each sampling group (two in each N A G sand test plot and eight in each supplemented soil test plot) represents a single substrate treatment. N onvegetated subplots were not sampled. Table 3. Chemical analysis conducted on soil samples (com pleted by BCMOF and ALS Canada Ltd.) collected from test plot treatment substrates in August 2009 and 2010. Further descriptions and references are listed in Appendix 3.__________________________________________________________________________________________ L a b o ra to ry / A n a ly sis P re p a r a tio n M eth o d A n a ly tica l M ethod Y ear of A n a ly s is Particle Size A nalysis Sedim entation Rate Gravimetric 2009 Total Carbon and Nitrogen C om bustion E lem ental A nalysis E lem ental Analyzer 2 0 0 9 ,2 0 1 0 Total Inorganic Carbon C om bustion E lem ental A nalysis E lem ental Analyzer 2 0 0 9 ,2 0 1 0 Total Sulphur C om bustion E lem ental A nalysis E lem ental Analyzer 2 0 0 9 ,2 0 1 0 A vailable Phosphorus (PCL-P) O lsen ’s Extraction M ethod U V /visib le spectrophotometer 2 0 1 0 M inerizeable Nitrogen Anaerobic incubation, IN KCt extraction Colorimetric • A u to A nalyzer 2 0 1 0 Available A m m onium (N H 4 -N ) and Nitrate (N O 3 -N ) Extraction w ith 2N KC1 Colorimetric - A u to A nalyzer 2 0 1 0 Carbonate (CaC'Oi) equivalent Empirical - A cid Neutralization pH meter 2 0 1 0 pH (Calcium Chloride) 1:1 for M ineral Soil pH /Ion Meter 2 0 0 9 ,2 0 1 0 pH (water) 1:1 for M ineral S oil pH /Ion Meter 2 0 0 9 ,2 0 1 0 Electrical C onductivity Saturated Paste Conductivity M eter 2 0 0 9 ,2 0 1 0 Extractable Elem ents (AI, B , Ca, Cu, Fe, K, M g, M n, N a, P, S and Zn M ehlich III Extraction ICP-AES" 2009, 2010 Exchangeable C ations (Ca, K, M g, Na) and CEC Neutral A m m onium A cetate ICP-A E S 2 0 0 9 ,2 0 1 0 Trace Elem ents (A s, Ba, Bi, Cd, Co, Cr, Hg, M o, N i, Pb, Sb, Sn, Ti and V ) M ehlich 111 Extraction ICP-A ES 2009 Total Elemental A nalysis for Rare Earth and Trace Elem ents Lithium Borate Fusion follow ed b y A cid D issolution ICP-M Sb 2009 B ase M etals Aqua R egia D igestion IC P-A E S 2009 Major O xides Lithium M etaborate/Lithium Tetraborate Fusion follow ed b y A cid D issolu tion ICP-A ES 2009 B C M O F L a boratory A L S C anada Ltd. L a b o ra to ry aInductively Coupled Plasma Atom ic Emission Spectrometry bInductively Coupled Plasma Mass Spectrometry cThe term base metals is used by the mining industry to identify metals used in non-alloyed forms (e.g., Cu, Pb, Zn, Sn; Lottermoser 2010). 56 2.5.2 Year Two (2010) 2.5.2.1 Vegetation Sampling In late August, 2010, vegetation was sampled from 252 o f the 288 1.0 m2 subplots o f the 12 revegetation test plots (36 subplots were designated as non-vegetated, Figure 7). In each subplot, destructive vegetation sampling was completed in and outside the four central sampling grid locations and non-destructive sampling was completed in the four central sampling grid locations. The outside comers of the four central sampling grid locations were marked using four pins and pink flagging tape and a 1 m x 1 m wood frame was used to mark the subplot boundary. Non-destructive sampling was first completed in the four central sampling grid locations and included percent cover and plant height determinations. Destructive sampling from both inside and outside the four central sampling grid locations followed, and consisted of harvesting above- and belowground biomass. Inside the four central sampling grid locations, aboveground biomass (live shoot biomass above the soil surface) was harvested by clipping the stems at the soil surface; an example of a subplot where the aboveground biomass was harvested is shown in Figure 10a. The stems were then placed upright in a 20 L bucket and the number of seedlings was determined by counting the number o f stems (Figure 10b); for grasses, care was taken to distinguish primary and secondary shoots and only primary shoots were counted. No seedling counts were completed for Rocky Mountain fescue due to difficulties in distinguishing between individual seedlings. Following seedling counts, the samples were packaged in labelled brown paper bags. 57 Outside the four central sampling grid locations representative seedlings were harvested from each o f the subplots (Figure 10c). For each o f the subplots sown with single­ species seeding treatments, four seedlings were harvested and for those with mixed-species seeding treatments, between one and four seedlings o f each species present was harvested. Using a garden trowel and shovel, seedlings were harvested from the subplots by digging out the earth surrounding the root system to the base o f the subplot (-25 cm) and then carefully removing the soil from the roots by hand. To further reduce the amount o f soil around the roots, the roots o f each plant sample were soaked in water for a one minute period (Figure lOd). The seedlings were kept intact (shoot and root biomass not separated) and packaged in labelled brown paper bags. Once harvesting was complete, aboveground biomass and seedling samples were packaged in cardboard boxes and shipped to UNBC; samples arrived within 10 days o f collection with no signs o f mold or deterioration. In the EFL at UNBC, aboveground biomass and seedling samples were washed, dried and weighed over a four month period (September to December 2010). Upon arrival from the minesite, the samples were placed in a sunny, glassed-in open area and air-dried for a period of approximately two weeks in an effort to stabilize the samples (i.e., prevent any decomposition or molding). During sample preparation and drying, all samples were kept in their original labelled brown paper bags. Aboveground biomass samples were oven-dried in a laboratory oven (Isotemp® Standard Incubators, 600 series; Fisher Scientific; model: 650d; Toronto, Ontario) for 48 hours at 70 °C. After drying, biomass samples were sieved (Figure 1la) to eliminate any residual sand or soil particles and then samples were weighed using a top-loading digital balance with a 0.01 g resolution (Sartorius, model ISO 9001; Edgewood, NY, USA). 58 c) d) Figure 10. Photographs showing the harvesting o f biomass from revegetation test plots: a) a subplot after aboveground biomass has been harvested from the four central sampling grid locations; b) seedling counts and packaging o f biomass samples; c) individual seedlings with shoot and root biomass collected from outside the four central sampling grid locations; and d) seedlings bathed in water to clean residual soil on roots. Before oven-drying, seedling samples with both root and shoot biomass were soaked for a 12 hour period (to loosen soil particles still attached to the roots; Figure l i b ) and then washed with a pressurized hose over a 1 mm sieve to remove remaining soil particles (Figure 11c). Seedling samples were then oven-dried in a laboratory ovens referred to above for 48 hours at 70 °C. Once drying was completed, seedling samples were left to equilibrate to room temperature for approximately 20 minutes before further preparation and weighing. The above- and belowground biomass o f seedling samples was then separated by clipping the samples at approximately 1cm above the shoot:root transition zone (Figure lid ) , simulating the approximate location in which aboveground biomass was clipped when harvested from the test plots. Once clipped, the root and shoot mass were measured separately using a digital balance with a 0.0001 g resolution (Sartorius, model MC4105; Edgewood, NY, USA). Figure 11. Photographs illustrating preparation o f aboveground biomass and seedling sam ples prior to weighing. Images show a) aboveground biomass samples being sieved prior to weighing; b) soaking the roots o f seedling samples for a 12 hour period; c) washing seedling samples over a 1 mm sieve, d) separating the above- and belowground biomass on seedling samples. 2.5.2.2 Substrate Treatment Sampling In August 2010, soil samples were collected from subplots sown with local blue wildrye, ecovar blue wildrye and arctic lupine (total = 108 samples) single seeding treatments. A composite sample composed o f three core samples was collected from each subplot containing one of the three seeding treatments to the depth o f approximately 15 cm (using a 60 16.5 cm long auger bucket). The composite sample was mixed thoroughly in a bucket and then a smaller random sample was collected from the mix. Samples were collected into ziplock bags, labelled, packaged in coolers and then sent to the BCMOF laboratory for analysis; Table 3 lists the physical and chemical properties that were tested. Samples for bulk density measurements were collected on site using a volume excavation method (15.24 cm diameter ring, board, hammer and trowel). Four samples were collected for each o f the nine substrate treatments; sampled subplots were randomly selected using a random number generator in Excel (Microsoft Office 2007) to select four subplots for each substrate treatment. Each sample was collected by hammering the ring to a depth of approximately 4.5 cm and then extracting the core using a trowel. Bulk density samples were then processed (i.e., dried and weighed) at UNBC. Soil samples were dried within a laboratory oven (Isotemp® Standard Incubators, 600 series; Fisher Scientific; model: 650d; Toronto, Ontario) for 48 hours at 70 °C and then weighed using a scale with 0.01 g accuracy (Sartorius, model ISO 9001; Edgewood, NY, USA). In addition, the compressive strength (kg cm '1) o f the soil surface was measured during soil sampling in all subplots (except non-vegetated subplots; Figure 8) using a pocket penetrometer (Humboldt Manufacturing, model H-4200; Schiller Park, IL, USA). For each subplot, measurements were taken in four randomly selected sampling grid locations (one measurement per grid location, selected using a random number generator in Excel (Microsoft Office 2007). 61 2.6 Statistical Analyses Statistical analysis o f vegetation data was conducted using R (version 2.11.1; R Development Core Team 2010). A linear mixed effects model was used to describe the response o f dependent variables to independent treatment variables (Table 4 and 5) using the R nlme package (Pinheiro et al. 2009). Analysis o f 2- and 10-year old stockpiled soil data was completed separately. Fixed effects for models included NAG sand, fertilizer and seeding treatment; blocking (in terms o f the four separate test site locations) was incorporated into the model as a random effect. The fixed-effects structure of the maximal model fit by maximum likelihood was: [y ~ NAG sand * fertilizer * seeding treatment] where * indicates interactions. Analytical assumptions, namely homogeneity o f variances and normal distribution o f residuals, were examined by inspecting residual plots. If required, a square root transformation was used to satisfy model assumptions. Maximal models (a model that includes all fixed and random effects and interactions) were simplified by sequentially deleting non-significant terms (starting with highest-order interactions) and comparing each model using the Akaike information criterion (AIC, Burnham and Anderson 1998) and likelihood ratio tests, until minimal adequate models were retrieved. A likelihood ratio test was also used to determine whether the lowest level o f nesting should be removed. The minimal model was compared to a null m o d e l: [y ~ 1] using a likelihood ratio test to assess the validity o f the mixed effects analyses. The response of dependent variables to independent treatment variables were examined using analysis o f variance (ANOVA) on transformed or non-transformed data. 62 Where no significant interactions were detected, Tukey’s pairwise comparisons (R package multcomp; Hothom et al. 2008) were utilized to investigate differences between treatment levels. Calculated p-values < 0.05 were considered statistically significant. Where significant interactions were detected, mean values o f treatment effects were obtained (biology package; Logan 2008) and plotted on a bar graph. For soil data, mean values (± standard error [SE]) have been calculated from raw data in an effort to characterize physical and chemical properties o f substrate treatments. Soil variables were not tested for significant differences among the experimental treatments. Table 4. Independent variables for native vegetation test plots constructed with 2- or 10-year old stockpiled soil. Variable Effects Number of Treatments Description Block random 4 Test plots, two at each location NAG sand Fixed 2 50:50 mix of soil:NAG sand yes or no Fertilizer Fixed 2 Granular fertilizer, 577 kg h a'1 (13-16-10 N P K )-y e s or no Seeding Treatment Fixed 7 Seeding treatments, either single-species or mixed-species; sown at 750PLS m '2 63 Table 5. Dependent variables for native vegetation test plots constructed with 2 or 10 year old stockpiled soil. Vegetation________________________________________________________________ percent emergence (% seedling density PLS density" )a seedling density (seedlings m '2) percent cover (%) plant height (cm) cover per seedling (% seedling"1)1’ aboveground biomass (g m"2) estimated belowground biomass (g m"2)c shoot:root ratiod aboveground biomass per seedling (g seedling'1)6 estimated belowground biomass per seedling (g seedling"1)*_______________________ Soil______________________________________________________________________ coarse fragments (%) fine fragments (sand, silt and clay; %) bulk density (g cm"3) penetration resistance (Mg m"2) total carbon (organic and inorganic; %) total nitrogen (%) total sulphur (%) ammonium-N (mg kg"1) phosphorus (Mehlich III and Olsen’s; mg kg"1) nitrate-N (mg kg"1) mineralizeable nitrogen (mg kg"1) pH (water) electrical conductivity (saturated paste; mS cm"1) cation exchange capacity and exchangeable cations (cmol(+) kg"1) extractable elements (Mehlich III; mg kg"1) major elements reported as oxides (%) rare earth and trace elements (Mehlich III; mg k g '1) base metals (mg kg'1) calcium carbonate equivalent (%)_____________________________________________ “Percent emergence was calculated by dividing seedling density by actual PLS (refer to Appendix 8 for actual PLS values). bCover per seedling was calculated by dividing percent cover by seedling density. 'Estimated belowground biomass w as calculated by dividing aboveground biomass by the shootxoot ratio. dThe shootxoot ratio was calculated by dividing shoot biomass by root biomass o f seedling samples. “Aboveground biomass per seedling was calculated by dividing aboveground biomass by seedling density. Estimated belowground biomass per seedling was calculated by dividing the estimated belowground biomass by seedling density. 64 3.0 Results 3.1 Soil Characterization Physical properties of 2- and 10-year old stockpiled soil supplemented with NAG sand differed considerably from non-supplemented stockpiled soil (Table 6). Percent coarse fragments, percent silt and percent clay content were lower and percent sand was higher in ■ y supplemented compared to non-supplemented soils. Bulk density (g cm' ) and soil strength (MPa) were higher and organic matter content (SOM; %) was lower in soil supplemented with NAG sand compared to non-supplemented soils. Chemical properties differed significantly between 2- and 10-year old stockpiled soil supplemented with and without NAG sand (Table 6). Soil pH was higher in supplemented compared to non-supplemented soils. Total carbon (C; %) and N (%) were lower in supplemented 2- and 10-year old soil and total S (%) was lower in supplemented 2-year old soil, but higher in supplemented 10-year old soil compared to non-supplemented soils. The amount o f organic C (%) was lower in supplemented soils and the C/N ratio was higher in supplemented 2-year old soil and lower in supplemented 10-year old soils compared to nonsupplemented soils. Electrical conductivity (mS cm '1) was higher and calcium carbonate equivalent (CaC 0 3 ) concentration was lower in supplemented compared to nonsupplemented soils. Concentrations of exchangeable cations (Ca2+, Na+, M g2+ and K+) in supplemented soils were relatively equal to or lower than concentrations in nonsupplemented soils. 65 Table 6. Mean values (± SE) o f physical and chemical soil properties in 2- and 10-year old stockpiled soil supplemented with or without N AG sand (n=12) and NAG sand (n=12).___________________________________________________________________________________________________________________ _ 10-Year Soil 2-Year Soil NAG sand Soil + NAG sand Soil + NAG sand Soil Soil P h ysical Properties Coarse Fragments (%) 56.1 ± 6 .3 32.4 ± 5 .8 43.2 ± 4 .3 22.3 ± 3 .4 1.4 ± 0 .3 46.3 ± 0 .4 77.7 ± 0 .5 89.5 ± 0 .3 Sand (% o f fine fraction) 51.3 ± 0 .3 80.5 ± 0 .7 Silt (% o f fine fraction) 29.4 ± 0 .4 12.6 ± 0 .4 32.5 ± 0 .3 13.5 ± 0.3 6.3 ± 0 .5 C lay (% o f fine fraction 19.3 ± 0 .4 6.8± 0 .3 21.2 ± 0 .3 9.1 ± 0 .4 4.2 ± 0 .3 Bulk Density (g cm '') 0.76 ± 0 .1 2 1.17 ± 0 .0 9 0.80 ± 0 .0 4 1.18 ± 0 .0 2 1.23 ± 0 .0 2 Soil Strength (MPa) 0 .0 4 ± < 0 .0 1 0.01 ± < 0 .0 1 0.06 ± < 0 .0 1 0.03 ± < 0 .0 1 0.01 ± < 0 .0 1 Soil Organic Matter (%)b 2.75 ± 0 .1 6 1.09 ± 0 .2 8 3.13 ± 0 .1 6 0.58 ± 0.32 <0,01 C hem ical P roperties 7 .0 8 ± 0 .1 0 7.31 ± 0 .0 4 5.61 ± 0 .0 3 7.16 ± 0.03 8.25 ± 0 .0 3 Total Carbon (%) 1.25 ± 0 .0 8 0.45 ± 0.03 1.51 ± 0 .0 5 0.36 ± 0 .0 2 0.09 ± 0 .0 2 Total Nitrogen (%) 0 .0 6 ± < 0 .0 1 0.07 ± < 0 .0 1 0.09 ± < 0 .0 1 0.02± < 0 .0 1 <0.01 Total Sulphur (%) 0.42 ± 0 .0 2 0.23 ± 0 .0 1 0.04 ± < 0 .0 1 0.16 ± 0 .0 0 3 0.19 ± 0 .0 1 Total Organic Carbon (%)" 1.10± 0.06 0.43 ± 0 .1 1 1.25 ± 0 .0 7 0.23 ± 0 .1 3 <0.01 C /N Ratio" 20.96 ± 0 .1 1 22.13 ± 0 .5 9 16.52 ± 0.13 14.63 ± 0 .2 7 33.13 ± 6 .5 3 Electrical Conductivity (mS c m 1) 1.59 ± 0 .1 9 1.66± 0 .2 0 0.23 ± 0 .0 1 0.83 ± 0 .0 7 1.00 ± 0 .1 4 0.50 ± 0 .0 1 0.65 ± 0 .0 4 pH (water) Calcium Caibonate (%) 0.89 ± 0 .0 2 0.76 ± 0 .0 3 0.54 ± 0 ,0 1 CEC (cm ol(+) k g 1)11 13.98 ± 0 .9 4 5.02 ± 0 .2 8 10.47 ± 0 .1 5 4.39 ± 0 .1 4 2.11 ± 0 .0 9 Exchangeable Ca (cm ol(+) kg'1)11 19.92 ± 0 .2 9 8.75 ± 0 .6 7 6.33 ± 0 .1 3 5.63 ± 0 .2 3 4.74 ± 0 .2 8 Exchangeable Na (cm ol(+) k g 1)11 0.20± 0 .0 2 0.09 ± 0 .0 3 0.07 ± 0 .0 1 0.05 ± 0 .0 2 0,05 ± 0 .0 2 Exchangeable M g (cm ol(+) k g 1)1' 0.68 ± 0 .0 3 0.17 ± 0 .0 1 0.61 ± 0 .0 2 0.14 ± 0 .0 1 0.08 ± 0 .0 1 Exchangeable K. (cm o!(+) k g 1)1' 0.50 ± 0 .1 3 0.29 ± 0 .0 2 0.32 ± 0 .0 3 0.32 ± 0.03 0.17 ± 0 .0 1 Samples were collected from test plots during the first growing season (August 2009) with the exception o f bulk density (n=4), penetration resistance (n=32) and calcium carb o n ate eq u ivalen t (n= 16), w hich w ere sam pled in A ugust 2010. “Total Organic C = Total organic C - Total inorganic C; bSoil Organic Matter = Total Organic C x 2.5 for B horizon (Troeh and Thompson 1993); cC/N ratio = Total Carbon/Total Nitrogen; dSamples prepared using Neutral Ammonium Acetate. 66 Mean concentrations o f Mehlich III extractable elements (mg kg'1) differed between 2- and 10-year old stockpiled soil supplemented with or without NAG sand (Table 7). For 2year old stockpiled soil, concentrations o f extractable elements were lower in supplemented soil compared to non-supplemented soils with the exception o f chromium (Cr), Cu, Fe, P, titanium (Ti) and vanadium (V). For 10-year old stockpiled soil, concentrations o f elements were lower in NAG sand supplemented soils, with the exception o f cobalt (Co), Cr, Cu, Mn, S, Ti, V and Zn. Total concentrations of rare earth and trace elements (mg kg'1) differed between 2and 10-year old stockpiled soil supplemented with or without NAG sand (Table 8). In general, total concentrations o f rare earth and trace elements were similar to or lower in supplemented 2- and 10-year old soil with the exception of V, erbium (Er), rubidium (Rb) and yttrium (Y). Cesium (Cs), Gallium (Ga) and ytterbium (Yb) were also higher in 10-year old supplemented soil. In addition, total concentrations of rare earth and trace elements (mg k g '1) within all treatment substrates were found to be lower than concentrations outlined in the Canadian Council of Ministers o f the Environment (CCME) Canadian Environmental Quality Guidelines (CEQG) for agricultural land (CCME 2004), with the exception o f Cr, Mo, V and Cu (Table 8). When comparing concentrations o f these four elements between 2 and 10-year old stockpiled soils supplemented with or without NAG sand, only V was found to be higher in supplemented soils. 67 Table 7. Mean values (±SE) for the concentration o f extractable elements (mg kg'1) in 2- and 10-year old stockpiled soil supplemented with or without NAG sand (n=12) and NAG sand (n=12), prepared using Mehlich III extraction method and analysed by ICP-MS._________________________________________ 2-Year Soil 10-Year Soil Elem ent Soil Soil + N A G sand NAG sand Soil Soil + N A G sand Al B Ba Ca Cd Co Cr Cu Fe 18.41 ± 0 .3 1 1041.39 ± 16.82 0.46 ±0.01 11.72 ± 0 .4 1 1316.54 ± 2 1 .1 7 0.11 ± < 0.01 0.63 ± 0.01 0.38 ± 0 .0 1 39.20 ± 0 .8 5 320.48 ± 7.07 0.52 ± 0.03 1.61 ± 0.11 1586.08 ± 8 2 .6 7 0.09 ± < 0.01 0.36 ± < 0.01 0.92 ± 0.02 46.27 ± 1.66 162.34 ± 2 .2 2 333.39 ± 5 .9 7 474.56 ± 7.27 51.99 ± 1.00 4202.56 ± 130.36 0.17 ±< 0.01 1.07 ± 0 .0 6 0.19 ± 0 .0 2 32.89 ± 2 .1 8 836.02 ± 2 2 .9 0 0.56 ± 0 .0 2 8.46 ± 0.28 2343.55 ± 125.28 0.13 ±< 0.01 0.70 ± 0.01 0.38 ± 0.01 38.40 ± 1.02 1955.77 ± 8 .8 8 0.45 ± 0.01 36.65 ± 0 .5 1 1161.95 ± 18.69 0.13 ± < 0.01 0.41 ± 0.01 0.21 ± < 0.01 222.40 ± 2.94 421.59 ± 5 .6 6 1616.46 ± 19.96 0.69 ± 0.01 31.99 ± 2 .3 2 K 99.64 ± 2 .3 2 65.04 ± 2 .4 5 89.24 ± 2 .3 9 76.96 ± 1.60 Mg 92.73 ± 3 .7 1 50.01 ± 1.54 76.71 ± 2.24 45.62 ± 0.69 64.04 ±1.47 Mn 104.98 ± 1.34 74.35 ± 2 .1 0 24.99 ± 1.07 38.81 ±0.79 28.76 ± 1.20 Na 32.22 ± 1.42 20.22 ± 1.30 14.75 ± 0.44 15.59 ± 0 .6 4 22.87 ± 0 .7 9 Ni 0.31 ± 0.01 0.34 ± 0.01 0.38 ± 0.01 0.38 ± 0.01 0.38 ± 0.01 P 16.83 ± 1.17 17.16 ± 1.08 13.85 ± 0 .8 2 15.73 ± 0 .8 6 5.86 ± 0 .3 4 Pb 2.68 ± 0 .1 3 1.51 ± 0 .0 5 1.05 ± 0.01 1.02 ± 0 .0 3 0.67 ± 0.02 S 215.88 ± 3 0 .2 4 159.61 ± 2 1 .5 9 22.41 ± 1 .4 5 47.44 ± 4.47 59.35 ± 9 .3 7 Ti 1.87 ± 0.19 4.56 ± 0 .1 5 1.75 ± 0 .1 0 5.99 ± 0 .0 7 4.56 ±0.01 V 0.58 ± 0.01 0.67 ± 0.01 0.39 ± < 0.01 0.64 ± < 0.01 0.55 ± 0.01 Zn 2.04 ± 0.07 1.92 ± 0 .0 8 0.87 ± 0.02 1.29 ± 0 .0 4 1.84 ± 0 .0 5 Samples were collected from test plots during the first growing season (August 2009). 68 Table 8. Mean values (±SE) for the total concentration o f rare earth and trace elements (mg kg ') in 2- and 10-year old stockpiled soil supplemented with or without NAG sand (n=4) and NAG sand (n=4) and maximum recommended concentrations in the Canadian Environmental Quality Guidelines (CEQG) outlined by the Canadian Council for the Ministers o f Environment for agricultural land (CCME 2004)._______________________________________________ Elem ent 2-Year Soil 10-Year Soil Soil Soil + N A G sand Soil Soil + NAG sand CEQG NAG sand Standard <1 -1 <1 -1 <1 -1 <1 -1 <1 -1 20 Ag Ba 494.00 ± 9.80 300.00 ± 11.10 540.50 ± 7.50 370.30 ± 10.00 750 255.80 ± 14.60 Ce 36.00 ± 0.60 15.70 ± 0 .5 0 32.10 ± 1.00 19.70 ± 0 .8 0 23.20 ± 0 .2 0 Co 14.90 ± 1.10 15.20 ± 0 .8 0 12.30 ± 0 .4 0 40 27.20 ± 0 .7 0 16.90 ± 0 .2 0 Cr 64 45.00 ± 6 .5 0 67.50 ± 2 .5 0 25.00 ± 2 .9 0 65.00 ± 5 .0 0 45.00 ± 2 .9 0 Cs 8.10 ± 0.10 7.20 ± 0 .1 0 3.80 ± 0 .1 0 6.00 ± 0 .1 0 7.10 ± 0.10 Cu 270.00 ± 3 .1 2 63 529.00 ± 67.51 365.75 ± 4 4 .0 9 252.63 ± 2 .5 5 257.75 ± 16.13 Dy 4.16 ± 0 .0 4 4.43 ± 0.04 3.84 ± 0 .0 3 4.44 ± 0 .1 0 4.54 ± 0.20 Er 2.45 ± 0.03 2.77 ± 0.06 2.31 ± 0.01 2.71 ± 0.04 2.95 ± 0 .1 0 Eu 1.22 ± 0 .0 2 1.16 ± 0 .0 2 1.18 ± 0 .0 2 1.16 ± 0 .0 2 1.13 ± 0 .0 4 Ga 15.70 ± 0 .3 0 16.20 ± 0 .3 0 14.30 ± 0 .1 15.90 ± 0.10 16.40 ± 0 .3 0 Gd 4 ,1 0 ± 0.10 3.80 ± 0 .0 2 4.10 ± 0.10 3.80 ± 0 .1 0 3.80 ± 0 .1 0 Hf 2.88 ± 0.08 2.18 ± 0 .0 5 3.35 ± 0 .0 6 2.35 ± 0 .0 3 2.03 ± 0.06 Ho 0.88 ± 0.01 0.95 ± 0.02 0.81 ± 0 .0 0 2 0.95 ± 0.02 1.00 ± 0 .0 2 La 15.30 ± 0 .5 0 8.90 ± 0.40 17.00 ± 0 .3 0 10.6 ± 0 .2 0 7.00 ± 0 .3 0 Lu 0.38 ± 0.01 0.45 ± 0.01 0.35 ± <0.01 0.44 ±< 0.01 0.47 ± 0.01 Mo 28.5 ± 7 .6 0 8.50 ± 1.30 6.30 ± 0 .3 0 6.30 ± 0 .3 0 9.50 ± 3 .4 0 Nb 4.63 ± 0.03 2.63 ± 0 .1 8 5.40 ± 0 .0 4 3.15 ± 0 .0 3 1.98 ± 0 .0 5 Nd 16.90 ± 0 .6 0 11.80 ± 0 .2 0 17.9 ± 0 .4 0 13.30 ± 0 .2 0 10.80 ± 0 .3 0 5 Samples were collected from test plots during the first growing season (August 2009). Samples were prepared using lithium borate fusion followed by acid dissolution and analysed using ICP-MS. 69 Table 8 Continued. Elem ent Soil 2-Year Soil Soil + NAG sand Ni Pb Pr Rb Se Sm Sn Sr 31.30 ± 0 .3 0 24.50 ± 0 .5 0 4.10 ± 0.10 59.30 ± 1.40 <2 4.00 ± 0 .0 2 1.00 236.00 ± 2.20 Ta Soil 10-Year Soil Soil + NAG sand NAG sand 2.70 ± 0 .1 0 94.80 ± 1.20 <2 3.20 ± 0 .1 0 1.30 ± 0 .3 0 31.30 ± 0 .5 0 13.50 ± 0.30 4.40 ± 0 .1 0 44.60 ± 0 .2 0 <2 4.10 ± 0.10 1.00 23.00 ± 4 .3 0 9.00 3.10 ± 0.10 84.10 ± 0 .6 0 <2 3.50 ± 0 .0 4 1.30 ± 0 .0 3 2 .2 0 ± 0.10 108.50 ± 2 .9 0 <2 3.20 ± 0 .0 2 1.30 ± 0 .3 0 215.25 ± 2 .5 3 259.25 ± 0.62 237.13 ± 1.78 212.80 ± 6 .5 0 0.30 0.20 ± 0.03 0.40 0.20 0.10 Tb 0.68 ± 0.01 0.66 ± 0.01 0.65 ± 0.001 0.67 ±0.001 0.68 ± 0 .0 2 Th 2.88 ±0.08 1.64 ± 0 .0 6 3.25 ± 0 .0 2 1.94 ± 0 .0 4 1.24 ± 0 .0 2 TI <0.5 - 0.8 <0.5 - 0.8 <0.5 - 0.8 <0.5 - 0.8 < 0 .5 -0 .8 Tm 0.38 ± 0.01 0.42 ± 0.01 0.35 ± <0.01 0.43 ± 0.01 0.44 ± 0.02 U 1.40 ± 0 .0 5 1.00 ± 0 .0 4 1.69 ± 0 .0 2 1.13 ± 0.01 0.88 ± 0.02 23 V 148.3 ± 2 .2 0 173.0 ± 2 .3 0 124.8 ±0.9 166.0 ± 2 .3 0 190.5 ± 7 .1 0 130 W 6.50 ± 0 .5 0 15.30 ± 3 .3 0 3.00 7.50 ± 0 .6 0 28.00 ± 8 .5 0 Y 21.90 ± 0 .3 0 24.00 ± 0.20 20.00 ± 0 .2 0 23.40 ± 0 .2 0 24.50 ± 0 .8 0 2.95 ± 0.02 2.24 ± 0 .0 2 2.81 ± 0 .0 7 2.93 ± 0.11 Yb 2.47 ± 0.06 Zn Zr 112.38 ± 2 .0 7 27.50 ± 2 .9 0 11.50 ± 0 .6 0 75.00 ± 3 .6 0 72.75 ± 5.11 84.75 ± 0.67 132.50 ± 6 .0 5 88.38 ± 3 .3 5 111.50 ± 1.87 75.13 ± 1.06 Samples were collected from test plots during the first growing season (August 2009). Samples were prepared using lithium borate fusion followed by acid dissolution and analysed using ICP-MS. 35.00 ± 11.60 7.00 ± 0 .4 0 CEQG Standard 50 70 1 5 1 67.80 ± 11.30 65.50 ± 1.80 200 70 Mean concentrations o f base metals differed between 2- and 10-year old stockpiled soil supplemented with or without NAG sand (Table 9). Total concentrations o f base metals were lower in supplemented compared to in non-supplemented soils with the exception o f Cu; within 10-year old soil, concentrations o f Cu were higher in NAG sand supplemented soil. Base metal concentrations were also found to be lower than concentrations outlined in the CEQG for agricultural land (CCME 2004) with the exception o f Cu and Mo. Mean percent composition o f major elements reported as oxides for 2- and 10-year old stockpiled soil supplemented with or without NAG sand are shown in Table 10. Percent composition o f these oxides were higher in supplemented compared to in non-supplemented soils. During the second growing season (2010), differences in the level o f plant available nutrients between fertilized and unfertilized substrates for 2- and 10-year old stockpiled soil were detected (Table 11). For 10-year old soil, mineralizeable N (mg kg'1) was lower in fertilized compared to in unfertilized soils. Concentrations o f K (prepared using Mehlich III and Neutral Ammonium Acetate) and P were higher in fertilized compared to unfertilized soils. Concentrations of B and Mn were higher in fertilized 2-year old soil and B was lower in fertilized 10-year old soil compared to unfertilized soils. Concentrations o f M n in fertilized and unfertilized 10-year old soils did not differ considerably. 71 Table 9. Mean values (±SE) for base metal concentrations (mg kg'1) in 2- and 10-year old stockpiled soil supplemented with or without NAG sand (n=4) and NAG sand (n=4) and maximum recommended concentrations in the Canadian Environmental Quality Guidelines (CEQG) outlined by the Canadian Council for the Ministers o f Environment for agricultural land (CCME 2004)._______________________________________________________________________________ 2-Year Soil 10-Year Soil Soil Soil + N AG sand CEQG Soil Soil + NAG sand NAG sand Standard <0.5-0 .6 <0.5-0 .6 <0.5-0 .6 <0.5-0 .6 <0.5-0 .6 20 Ag As <5.0 <5.0 <5.0 <5.0 12 <5.0 Co 12.30 ± 0 .4 4 23.75 ± 0 .4 8 13.75 ± 0 .7 5 14.25 ± 0 .2 5 12.50 ± 0 .2 9 40 Cd 1.4 <0.5 <0.5 <0.5 <0.5 <0.5 Cu 560.50 ± 110.99 373.75 ± 78.11 256.75 ± 1.80 275.25 ± 1.03 257.75 ± 16.13 63 < 0 . 1-1 < 0 . 1-1 < 0 . 1-1 6.6 < 0 . 1-1 < 0 . 1-1 Hg Mo 6.75 ± 1.54 5.00 ± 0 .4 1 15.00 ± 2 .5 8 5.00 4.25 ± 0 .2 5 5 Ni 22.75 ± 0 .2 5 11.25 ± 0 .6 3 20.50 ± 0 .5 0 12.75 ± 0 .2 5 7.75 ± 0.25 50 Pb 21.75 ± 0 .2 5 8.00 ± 0 .4 1 5.00 ± 0 .4 1 70 10.50 ± 0 .6 5 7.25 ± 0.48 Zn 66.00 ± 3 .9 4 57.75 ± 7 .6 5 105.25 ± 4 .8 0 72.50 ± 1.55 54.25 ± 1.75 200 Samples were collected from test plots during the first growing season (August 2009). Samples were prepared using lithium borate fusion followed by acid dissolution and analysed using ICP-MS. 72 Table 10. Mean values (±SE) for the percent composition o f major elements reported as oxides in 2- and 10-year old stockpiled soil supplemented with or without NAG sand (n=4) and NAG sand (n=4). Oxide concentrations were calculated from the determined elemental concentration.___________________ 2-Year Soil 10-Year Soil Soil N AG sand Soil Soil + NAG sand Soil + NAG sand BaO S i0 2 AI2O 3 Fe20 3 CaO Cr20 3 MgO Na20 K20 < 0.01 - 0.06 57.55 ± 0 .5 7 14.58 ± 0 .1 4 8.18 ± 0 .0 4 2.58 ± 0 .0 7 < 0.1 2.16 ± 0 .0 3 < 0.01 - 0.06 59.48 ± 0.28 15.50 ± 0 .1 5 8.69 ± 0.06 3.96 ± 0 .1 0 < 0.1 2.24 ± 0.01 < 0.01 - 0.06 61.50 ± 0 .2 7 14.05 ± 0.06 6.30 ± 0.05 2.05 ± 0.03 < 0.1 1.70 ± 0.01 < 0.01 -0 .0 6 61.53 ± 0 .0 9 15.13 ± 0 .0 3 8.00 ± 0 .0 5 3.60 ± 0 .0 4 < 0.1 2.15 ±0.01 < 0.01 - 0.06 60.8 ± 0.50 15.45 ± 0 .2 0 9.00 ± 0 .2 0 4.48 ± 0.03 < 0.1 2.32 ± 0 .0 2 2.09 ± 0.05 1.42 ± 0 .0 2 2.81 ± 0 .0 5 1.57 ± 0.01 2.60 ± 0.02 1.40 ±0.01 2.93 ± 0.01 1.56 ± 0 .0 1 3.05 ± 0 .0 6 1.68 ± 0.01 T i0 2 0.71 ±< 0.01 0.85 ± 0 .0 1 0.71 ± 0.01 0.84 ± < 0.01 0.92 ± 0 .0 1 MnO 0.14 0.13 0 .1 0 ± < 0.01 0.12 ± < 0.01 0.13 ± < 0.01 P 20 5 0.18 ± 0.01 0.24 ± 0.02 0.18 ± 0.01 0.21 ± 0.01 0.28 ± 0.03 Loia 9.94 ± 0.43 4.13 ± 0.41 9.37 ± 0 .2 9 3.94 ± 0 .2 6 1.77 ± 0 .2 1 Samples were collected from test plots during the first growing season (August 2009). Samples were prepared using lithium borate fusion followed by acid dissolution and analysed using ICP-AES. "Loss on ignition at 1000°C. 73 Table 11. Mean values (± SE) o f chemical soil properties o f2 - and 10-year old stockpiled soil with or without the addition o f fertilizer. Samples were collected from test plots during the second growing season (August 2010)._____________________________________________________________________ 2-Year Soil 10-Year Soil Soil + Fertilizer Soil Soil + Fertilizer Soil pH (water) 7.08 ± 0.05 7.03 ± 0.07 6.35 ± 0 .1 6 6.31 ± 0 .1 6 1.06 ± 0 .3 1 Nitrate-N (mg kg'1) 1.37 ± 0 .2 5 1.48 ± 0 .4 0 1.64 ± 0 .4 3 Ammonium-N (mg kg'1) 0.68 ± 0.11 0.90 ± 0 .1 6 1.20 ± 0.21 1.20 ± 0.20 Mineralizeable Nitrogen (mg kg"1) 7.83 ± 0.49 8.60 ± 0 .8 5 9.02 ± 0 .5 3 7.91 ± 0 .5 1 Total Carbon (%) 0.85 ± 08 0.94 ± 0 .1 0 1.00 ± 0.11 1.00 ± 0.10 Total Nitrogen (%) 0.04 ±< 0.01 0.04 ±< 0.01 0.06 ± < 0.01 0.06 ± < 0.01 Total Sulphur (%) 0.33 ± 0.02 0.33 ± 0.02 0.09 ± 0 .0 1 0.09 ± 0.01 CEC (cmol(+)/kg)c 9.29 ± 1.04 8.73 ± 0.85 5.64 ± 0 .4 4 5.69 ± 0.44 Exchangeable Ca (cmol(+)/kg)c 14.16 ± 1.21 14.77 ± 1.34 6.55 ± 0 .1 6 6.45 ± 0 .1 4 Exchangeable Na (cmol(+)/kg)c 0.12 ± 0.01 0.10 ± 0.01 0.04 ± < 0.01 0.04 ± < 0.01 Exchangeable Mg (cmol(+)/kg)° 0.17 ± 0 .0 4 0.18 ± 0 .0 4 0.27 ± 0 .0 6 0.25 ± < 0.01 Exchangeable K (cmol(+)/kg)c 0.28 ± 0.01 0.29 + 0.01 0.23 ± <0.01 0.26 ± < 0.01 Boron (mg kg' ’)b 0.02 ± < 0.01 0.04 ± < 0.01 0.05 ± <0.01 0.03 ± <0.01 Calcium (mg kg'1)*3 4374.56 ± 149.05 4030.57 ± 208.49 1188.06 ± 2 4 .5 6 1135.85 ± 2 7 .0 9 Copper (mg kg'*)b 35.73 ± 1.24 33.19 ± 1.54 26.42 ± 1.98 25.26 ± 1.65 163.91 ± 2.71 160.77 ± 3 .5 8 Iron (mg kg'*)b 214.20 ± 3 .2 0 230.60 ± 3 .7 0 Phosphorus (mg kg'1)'1 4.58 ± 0 .3 2 7.98 ± 0 .5 6 3.21 ± 0 .1 3 5.39 ± 0 .2 7 Potassium (mg kg' 1)b 86.44 ± 2 .8 6 90.84 ± 2 .8 3 80.05 ± 1.57 84.68 ± 1.56 Magnesium (mg kg'*)b 59.11 ± 4 .3 6 55.03 ± 4.37 52.07 ± 3 .9 6 52.35 ± 3 .8 9 Manganese (mg kg‘!)b 75.29 ± 3 .3 6 66.63 ± 2.75 26.75 ± 1.45 26.19 ± 1.35 Sulphur (mg kg_l)b 215.48 ± 4 5 .5 9 216.27 ± 4 1 .7 9 22.41 ± 1 .4 5 19.80 ± 0 .8 0 1.20 ± 0.10 1.19 ± 0.13 Zinc (mg kg '1)*5 1.82 ± 0 .0 9 1.86 ± 0 .0 7 O lse n ’s ex tractio n m ethod; bM ehlich III ex traction; cN eutral A m m onium A cetate. 74 3.2 Plant Growth 3.2.1 Non-destructive Measurements The addition o f NAG sand had no significant effect on seeding density (seedlings m '2) on 2year old stockpiled soil test plots (Table 12); however, significant effects were observed with fertilizer and seeding treatment. The addition o f fertilizer significantly increased seedling density from 226±37 seedling m '2 to 333±37 seedlings mf2 (p=0.0007). Seedling density differed significantly among single-species seeding treatments (p<0.0001) with the highest value for genetically diverse blue wildrye and lowest for arctic lupine; seedling density o f genetically diverse blue wildrye was also signficantly higher than that of locally collected blue wildrye seed (Figure 12). No seedlings o f M erten’s sedge were detected on 2-year old soil. The addition o f NAG sand had no significant effect on seeding density on 10-year old stockpiled soil test plots (Table 12). Significant effects were observed with fertilizer and seeding treatment. The addition o f fertilizer significantly increased seedling density from 243±38 seedling m'2 to 322±41 seedlings m '2 (p<0.0104). Seedling density differed significantly among single-species seeding treatments (p<0.0001) with the highest value for genetically diverse blue wildrye and lowest for arctic lupine (Figure 12). No seedlings o f Merten’s sedge were detected on 10-year old soil. 75 Table 12. Summary o f linear mixed-effects model results for percent emergence, seedling density, percent cover and plant height (single-species treatments) collected from 2- and 10-year old stockpiled soil test plots in September 2010. For each o f the four response variables (columns), the details o f the minimal adequate models are listed in the rows, with explanatory variables and their interactions (first column) retained in the models, their corresponding F- and pvalues, plus the number o f observations and the minimum and maximum model AIC values. Bold font indicates significant p-values._____________________ Predictor Variable 2 -yea r Soil NAG Sand Fertilizer Seeding Treatment NAG sand X Fertilizer NAG sand X Seeding Treatment Fertilizer X Seeding Treatment NAG sand X Fertilizer X Seeding Treatment Number o f Observations AIC o f maximum model AIC o f minimum model 10- Year Soil NAG sand Fertilizer Seeding Treatment NAG sand X Fertilizer NAG sand X Seeding Treatment Fertilizer X Seeding Treatment NAG sand X Fertilizer X Seeding Treatment Number o f Observations AIC o f maximum model AIC o f minimum model Seedling Density* F, .55=0.0193 F,, 55 = 13.06 ^5.55=60.62 Excluded Excluded Excluded Excluded p=0.8901 p = 0.0007 p<0 .0 0 0 1 Excluded Excluded Excluded Excluded F,.55=0.2163 Fy.55= 13.22 F3'S5=79.16 Excluded Excluded Excluded Excluded 64 807.3 805.5 F, .55=0.6869 Ft 55=7.035 Fx55=77.23 Excluded Excluded Excluded Excluded 64 372.3317 367.7658 Percent Cover Emergence" p=0.6437 p = 0.0006 p<0 .0 0 0 1 Excluded Excluded Excluded Excluded 64 275.6 266.5 p = 0 .4 !0 8 p = 0.0104 pcO.OOOl Excluded Excluded Excluded Excluded Ft. si= 1.176 Ft 51=10.79 Fj,5i=88.70 Excluded Excluded Fj,si=3.829 Excluded 64 269.1764 262.5318 p = 0.7065 pO.OOOl p<0 .0 0 0 1 Excluded Excluded Excluded Excluded FY ioo=0.1426 Fi.ioo=34.26 F,6.i»]=9.980 Excluded Excluded Excluded Excluded Plant Height* Fy ,56=0.3860 F|,56=30.65 F ,.* = I 0 .I8 Excluded Excluded Fj,5t,=2.884 Excluded 71 280.9 263.8 112 891.2 870.7 p=0.2831 p = 0.0018 p<0 .0 0 0 1 Excluded Excluded p = 0.0149 Excluded F, 9,=0.0066 Ft .99=23.22 Ff(,m= 18.06 Excluded Excluded Excluded Excluded p = 0 .5369 p<0 .0 0 0 1 p<0 .0 0 0 1 Excluded Excluded F =0.0303 Excluded p=0.9356 p<0 .0 0 0 1 p<0 .0 0 0 1 Excluded Excluded Excluded Excluded 111 900.3047 892.685 F,.M=0.7399 F i.54= 19.86 Ft ,54=24.82 Excluded Excluded Excluded Excluded p=0.3929 p<0 .0 0 0 1 pcO . 0 0 0 1 Excluded Excluded Excluded Excluded 74 266.5264 249.2311 asquare root; AIC, Akaike information criterion; ‘excluded’ indicates terms excluded during model simplification (stepwise approach). 76 2-Year Soil 600 i 500 6CJD B 400 - o> & 300 *35 S 4) Q 200 W C> 5V =0.0303; Table 12). Mean plant height for all seeding treatments increased in response to the addition o f fertilizer, most notably for fringed brome and Rocky Mountain fescue (Figure 15). The addition o f NAG sand had no significant effect plant height for single-species seeding treatments on the 10-year old stockpiled soil test plots; however, significant effects were observed with fertilizer and seeding treatment (Table 12). The addition o f fertilizer significantly increased mean plant height from 26.6±2.4 cm to 39.2±2.6 cm (p<0.0001). Plant height significantly differed among seeding treatments (p<0.0001) with the highest values for genetically diverse blue wildrye and the lowest for Rocky Mountain fescue and arctic lupine. Genetically diverse blue wildrye had significantly higher mean plant height compared to the locally collected blue wildrye (Figure 15). In both 2- and 10-year old stockpiled soil test plots, the highest mean plant height for single-species seeding treatments were observed in soils treated with fertilizer and fertilizer + NAG sand. In 2-year old soil, the mean value o f plant height in the fertilizer + NAG sand treated soils was slightly higher than values observed in plots with just the fertilizer treatment. No considerable difference in mean plant height o f seeding treatments was visible between fertilizer and fertilizer + NAG sand substrate treatments in 10-year old soil. 81 2-Year Soil 45 1 40 - bromcil ecovar elymgla ecovar elym gla local festsax ecovar lupiarc local native mix ecovar native mix local Seeding T reatm ent Figure 14. Overall seeding treatment effects on mean values (±SE) o f plant cover (%) across all substrate treatments in 2-year (above) and 10-year (below ) old stockpiled soil. Single-species seeding treatments include seeds obtained from genetically diverse seed increase plots (ecovars) o f fringed brome (BROM CIL), blue wildrye (ELYMGLA) and Rocky Mountain fescue (FESTSAX) or seeds collected from the local area o f the minesite (local) o f blue wildrye and arctic lupine (LUPIARC). M ixed seeding treatments include an ecovar m ix (blue wildrye, fringed brome and Rocky Mountain fescue) and a local mix (blue wildrye, arctic lupine and Merten’s sedge ( C arex m erten sii). N o comparisons between single and mixed-species seed treatments were made. Bars which share a common letter are not significantly different (p<0.05). 82 2-Year Soil ■ Topsoil □ Topsoil + Fertilizer m o 0.01 U ^~h AH ihM bromcil ecovar " 1 -1 elym gla ecovar elym gla local lupiarc local Seeding Treatm ent Figure 17. Overall species effects on mean values (±SE) o f cover per seedling (% seedling'1) for single-species seeding treatments across all substrate treatments in 10-year old stockpiled soil. Single-species seeding treatments include seeds obtained from genetically diverse seed increase plots (ecovars) o f fringed brome (BROMCIL), blue wildrye (ELYMGLA) and seeds collected from the local area o f the minesite (local) o f blue wildrye and arctic lupine (LUPIARC). In 2-year old stockpiled soil test plots, mean values o f cover per seedling for single­ species seeding treatments was the highest in fertilizer + NAG sand treated soils. However, in 10-year old soil test plots, mean values o f cover per seedling for seeding treatments did not differ considerably between substrate treatments, with the exception of arctic lupine where cover per seedling considerably decreased with the addition of NAG sand. In NAG sand test plots (i.e., with no fertilizer or soil), seeding treatments had a significant effect on percent emergence (Table 15; p=0.0202) and seedling density (p=0.0344). The highest percent emergence and seedling densities on NAG sand test plots were found for genetically diverse blue wildrye and the lowest were for fringed brome (Figure 18). Seeding treatments had no significant effect on percent cover or plant height (for both single and mixed seeding treatments) on NAG sand substrates (Tables 15, 16). 88 Seeding treatment in NAG sand test plots had no significant effect on cover per seedling (% seedling'1; Table 16). Nodules o f Rhizobium were detected on the roots o f arctic lupine seedlings when they were first collected from the test plots. However, due to difficulties identifying nodulation on dried samples, the number o f inoculated seedlings and the number of nodules per seedling was not recorded. Table 15. Summary o f linear mixed-effects model results for emergence, seedling density, percent cover and plant height (single-species treatments) collected from N A G sand test plots in September 2010. D etails o f the minimal adequate models are listed in the rows, with explanatory variables and their interactions (first column) retained in the models, their corresponding F- and p-values, plus the number o f observations (bottom row). Bold font indicates significant p-values.________________________________________________________________ Predictor Variable NA G S a n d Seeding Treatment Num ber o f Observations Ft,,8=5.489 Predictor Variable N A G Sand Seeding Treatment Number o f Observations Seedling D en sity” Emergence" p = 0 .0 2 0 2 F 68= 4 .4 9 5 p = 0 .0 3 4 4 16 16 Percent Cover Plant H eight F6, ,8=1.678 p = 0 .1 7 4 4 fV r= 0.9697 26 p = 0 .3 7 5 7 6 asquare root transformed. Table 16. Summary o f linear mixed-effects model results for plant height (mixed-species treatments) and percent cover per seedling (% se ed lin g 1), calculated using cover and seedling density, collected from N A G sand testplots in September 2010. Details o f the minimal adequate model are listed in the rows, with explanatory variables and their interactions (first column) retained in the m odels, their corresponding F- and Pvalues, plus the number o f observations (bottom rows)._________________________________________________ Predictor Variable N A G Sand Seeding Treatment Number o f Observations Plant H eight F, 2=7.861 Percent Cover per Seedlin g p = 0 .2 181 5 F xt ,= 0 .6 8 0 9 p = 0 .5 8 5 6 16 89 100 1 90 80 70 - elymgla ecovar elym gla local lupiarc local Seeding Treatm ent Figure 18. Overall seeding treatment effects on mean values (±SE) o f em ergence (%; above) and seedling density (seedlings m‘2; below) in NAG sand. Single-species seeding treatments include seeds obtained from genetically diverse seed increase plots (ecovars) o f blue wildrye (ELYM GLA) and seeds collected from the local area o f the minesite (local) o f blue wildrye and arctic lupine (LUPIARC). Bars which share a common letter are not significantly different (p<0.05). 90 3.2.2 Shoot and Root Biomass The addition o f NAG sand to 2-year old stockpiled soil had no significant effect on aboveground biomass (g m '2); however, significant effects were observed with fertilizer and seeding treatment (Table 17). The addition o f fertilizer significantly increased aboveground biomass after two growing seasons from 16.59±4.72 g m’2 to 67.61±4.45 g m '2 (pO.OOOl). Aboveground biomass differed significantly among single-species seeding treatments (p=0.0001) with the highest value for genetically diverse blue wildrye and lowest for fringed brome; a comparion of mixed-species seeding treatments showed no significant difference (Figure 19). Genetically diverse blue wildrye had significantly higher aboveground biomass compared to locally collected blue wildrye (Figure 19). Within 10-year old stockpiled soil test plots, there were significant interactions between seeding treatment and NAG sand (/?=0.0072) and seeding treatment and fertilizer (p=0.0003) for aboveground biomass (Table 17). In response to the NAG sand supplement, all seeding treatments increased in aboveground biomass compared to their performance on non-supplemented soils (except for arctic lupine and the local native seed mix); Rocky Mountain fescue increased considerably more than the other seeding treatments. The interaction between seeding treatment and fertilizer showed increases in aboveground biomass for all seeding treatments in response to fertililzer with the exception o f arctic lupine (Figure 19). 91 Table 17. Summary o f linear mixed-effects model results for aboveground biomass, estimated belowground biomass and shoot:root ratio collected from 2and 10-year old stockpiled soil test plots in September 2010. For each o f the three response variables (columns), the details o f the minimal adequate models are listed in the rows, with explanatory variables and their interactions (first column) retained in the models, their corresponding F- and p-values, plus the number o f observations and the minimum and maximum model AIC values (bottom rows). Bold font indicates significant p-values.______________________ Predictor Variable 2-Y ear Soil N AG Sand Fertilizer Seeding Treatment NAG sand X Fertilizer NAG sand X Seeding Treatment Fertilizer X Seeding Treatment NAG sand X Fertilizer X Seeding Treatment Number o f Observations AIC o f maxim um model AIC o f minimum model 10-Year Soil NAG sand Fertilizer Seeding Treatment N AG sand X Fertilizer NAG sand X Seeding Treatment Fertilizer X Seeding Treatment NAG sand X Fertilizer X Seeding Treatment Number o f Observations AIC o f maximum model AIC o f minimum model Aboveground Biomass" p=0.0921 p<0.0001 p=0.0003 Excluded Excluded Excluded Excluded F,.,oo=2.893 FY,oo=52.24 FV>.ioo= 4 . 6 6 5 Excluded Excluded Excluded Excluded Shoot:Root Ratio" F u 2=4.046 F,.62= 13.84 F,.,,2=29.71 Excluded Excluded Excluded Excluded 112 616.3 588.9 112 532.5 525.4 F, .,,4=7.408 F| ,4=9.000 F4.m=7.852 Excluded F4,m=2.595 Excluded Excluded 72 110.5 88.07 p=0,2903 p=0.0001 p « ) .0 0 0 l Excluded p=0.0072 p=0.0003 Excluded F iM= 1.132 F, .*6=87.95 F6.m=22.51 Excluded FV,.*6=3.176 For,=4.703 Excluded p = 0.0486 p=0.0004 p<0.0001 Excluded Excluded Excluded Excluded Estimated Belowground Biomass" F u 5 = 11.04 F,.f,s=3.182 F4.65=48.71 Excluded Excluded Excluded Excluded 79 421.6 403.7 p = 0.0015 p = 0 .0 7 9 1 p<0.0001 Excluded Excluded Excluded Excluded 75 85.57 66.14 p=0.0083 p<0.0001 p= 0.0038 Excluded p = 0.0444 Excluded Excluded F,.64= 1.432 F,.64=19.53 F4.64=5.217 Excluded F4.64=7.282 Excluded Excluded p=0.2358 p<0.0001 p = 0 .0 0 !0 Excluded p=0.0001 Excluded Excluded 79 306.2 307.1 “square root transformed; AIC, Akaike information criterion; ‘excluded’ indicates terms excluded during model simplification (stepwise approach). 92 2-Year Soil 250 200 ce S o 150 S3 100 s•o S © ts > © > O X < 50 D ■ B ■Mi Jb b L AD BD B _jh£hl. 10-Year Soil 250 i a e* 200 ■ Topsoil □ Topsoil + NAG Sand US Topsoil + Fertilizer k s k bromcil ecovar elymgla ecovar elym gla local festsax ecovar H - lupiarc local native m ix ecovar f e native m ix local Seeding T reatm ent Figure 19. Overall seeding treatment effects on mean values (±SE ) o f aboveground biomass (g m'2) across all substrate treatments in 2-year (above) and 10-year (below ) old stockpiled soil. Single-species seeding treatments include seeds obtained from genetically diverse seed increase plots (ecovars) o f fringed brome (BROMCIL), blue wildrye (ELYMGLA) and Rocky Mountain fescue (FESTSAX) or seeds collected from the local area o f the minesite (local) o f blue wildrye and arctic lupine (LUPIARC). Mixed seeding treatments include an ecovar mix (blue wildrye, fringed brome and Rocky Mountain fescue) and a local m ix (blue wildrye, arctic lupine and Merten’s sedge ( C a rex m erten sii)). N o comparisons between single and m ixed-species seed treatments were made. For 2-year old soil (above), bars which share a com m on letter are not significantly different (p<0.05). 93 In both 2- and 10-year old stockpiled soil test plots, the highest mean aboveground biomass for seeding treatments was observed in soils treated with fertilizer and fertilizer + NAG sand. When comparing the two treatments, the mean aboveground biomass for seeding treatments in both soils was considerably higher in soils treated with fertilizer + NAG sand (with the exception of arctic lupine and the local native mix seeding treatments in 10-year old soil). Within 2-year old stockpiled soil test plots, fertilizer had a significant effect on estimated belowground biomass and there was a significant interaction between seeding treatment and NAG sand substrate treatment (Table 17). Fertilizer significantly increased estimated belowground biomass from 4.10±3.55 g m'2 to 14.89±3.22 g m '2 (p<0.0001; Figure 20). The interaction between seeding treatment and NAG sand substrate treatment showed an increase in estimated belowground biomass in response to the addition o f NAG sand for all seeding treatments with a significant increase for arctic lupine (p=0.0001; Figure 20). Fertilizer had a significant effect on estimated belowground biomass on 10-year old stockpiled soil test plots and there was a significant interaction between seeding treatment and NAG sand (Table 17). Fertilizer significantly increased estimated belowground biomass from 11.56±2.18 g m'2 to 22.77±3.07 g m'2 (p<0.0001). The interaction between the NAG sand and seeding treatments showed an increase in estimated belowground biomass for all seeding treatments on soil supplemented with NAG sand, compared to those o f non­ supplemented soil, with the exception o f arctic lupine, which showed the opposite trend (p<0.0001; Figure 20). 94 2-Year Soil 120 -| ■ Topsoil £ □ Topsoil + N A G Sand 100 V t/13 ee S o s*o fi 3 O u&£ so 10-Year Soil 120 *1 100 - ee S © 80 s*3 s 60 O u©a £ *03 40 PQ -O 3 •w 20 - s 0 bromcil ecovar elym gla ecovar elym gla local festsax ecovar lupiarc local S eed in g Treatm ent Figure 20. Overall seeding treatment effects on mean values (±SE) o f estimated belowground biom ass (g nT2) across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled soil. Single-species seeding treatments include seeds obtained from genetically diverse seed increase plots (ecovars) o f fringed brome (BROMCIL), blue wildrye (ELYMGLA) and Rocky Mountain fescue (FESTSAX) or seeds collected from the local area o f the minesite (local) o f blue wildrye and arctic lupine (LUPIARC). 95 In both 2- and 10-year old stockpiled soil test plots, the highest mean estimated belowground biomass for single-species seeding treatments was observed in soils treated with fertilizer and fertilizer + NAG sand. When comparing the two treatments, the mean estimated belowground biomass for most seeding treatments in both soils was considerably higher in soils treated with fertilizer + NAG sand (with the exception of locally collected blue wildrye and arctic lupine seeding treatments in 10-year old soil). For 2-year old stockpiled soil test plots, NAG sand, fertilizer and seeding treatments had significant effects on the shoot:root ratio (Table 17). The addition o f NAG sand significantly decreased the shoot:root ratio from 4.25±0.34 to 3.53±0,35 (p=0.0486). Fertilizer significantly increased the shoot.root ratio from 4.25±0.34 to 5.69±0.37 (p=0.0001). Significant differences in the shoot:root ratio were detected among single­ species seeding treatments (p<0.0001) with the highest values for fringed brome and locally collected blue wildrye and the lowest for arctic lupine (Figure 21). Genetically diverse blue wildrye had significantly lower shoot: root ratio compared to the locally collected blue wildrye (Figure 21). NAG sand supplement and seeding treatments had significant effects on the shootroot ratio in 10-year old stockpiled soil test plots (Table 17). The addition o f NAG sand significantly increased the shoot root ratio from 4.89±0.31 to 5.63±0.32 (p=0.0015). Significant differences for the shoot root ratio were detected among single-species seeding treatments (/K0.0001) with arctic lupine values significantly lower than all other single­ species seeding treatments (Figure 21). 96 2-Year Soil « 5 J S 4 S 3 10-Year Soil bromcil ecovar elymgla ecovar elym gla local festsax ecovar lupiarc local Seeding Treatm ent Figure 21. Overall seeding treatment effects on mean values (±SE) o f the shoot:root ratio across all substrate treatments in 2-year (above) and 10-year (below ) old stockpiled soil. Single-species seeding treatments include seeds obtained from genetically diverse seed increase plots (ecovars) o f fringed brome (BROM CIL), blue wildrye (ELYMGLA) and Rocky Mountain fescue (FESTSAX) or seeds collected from the local area o f the minesite (local) o f blue wildrye and arctic lupine (LUPIARC). N o comparisons between single and mixedspecies seed treatments were made. Bars which share a common letter are not significantly different (p<0.05). 97 In both 2- and 10-year old stockpiled soil test plots, the highest mean shoot:root ratio for single-species seeding treatments was observed in soils treated with fertilizer and fertilizer + NAG sand. In 2-year old soil, the mean shoot:root ratio was higher in fertilized soil compared to soils treated with both fertilizer + NAG sand; however, the opposite trend was observed in 10-year old soil. For 2-year old stockpiled soil test plots, the addition o f NAG sand had no significant effect on the average aboveground biomass per seedling (g seedling'1); however, a significant effect was observed for fertilizer (Table 18). Fertilizer significantly increased the aboveground biomass per seedling from 0.12±0.02 to 0.29±0.02 (p=0.0056). In the 10-year old stockpiled soil test plots, fertilizer had significant effect on the aboveground biomass per seedling and there was a significant interaction between NAG sand and seeding treatment (Table 18). Fertilizer significantly increased the aboveground biomass per seedling from 0.16±0.02 to 0.27±0.02 (p=0.0132). The interaction between seeding treatment and NAG sand substrate treatment showed a species dependent increase or decrease in aboveground biomass per seedling in response to the addition o f NAG sand, the most notable effect being an considerable decrease for arctic lupine (p=0.0039; Figure 22). 98 Table 18. Summary o f linear mixed-effects model results for above- and estimated belowground biomass per seedling calculated using above- and estimated belowground biomass and seedling density collected from 2- and 10-year old stockpiled soil test plots in September 2010. For each o f the two response variables (columns), the details o f the minimal adequate models are listed in the rows, with explanatory variables and their interactions (first column) retained in the models, their corresponding F- and p-values, plus the number o f observations and the minimum and maximum model AIC values (bottom rows). Bold font indicates significant p-values.___________________________________________________________________________________________________ Predictor Variable_________________________________________________ Aboveground Biomass per Seedling3_____________________________ Estimated Belowground B iom ass per Seedling”________ 2-Y ear Soil NAG Sand Fertilizer Seeding Treatment N AG sand X Fertilizer NAG sand X Seeding Treatment Fertilizer X Seeding Treatment NAG sand X Fertilizer X Seeding Treatment Number o f Observations AIC o f maxim um model AIC o f minimum model 10-Year Soil NAG sand Fertilizer Seeding Treatment NAG sand X Fertilizer NAG sand X Seeding Treatment Fertilizer X Seeding Treatment N AG sand X Fertilizer X Seeding Treatment Number o f Observations AIC o f maximum model AIC o f minimum model £7=0.1728 £7=0.0056 £7=0.2871 Excluded Excluded Excluded Excluded ^ 1.55= 1-908 Fi.55=8.307 ^3.55=1.290 Excluded Excluded Excluded Excluded 63 54.06 44.19 64 48.94 30.61 £7=0.1390 £7=0.0132 p=0.2869 Excluded £7=0.0039 Excluded Excluded F t.51=2.258 Ft. 51=6.582 F 3.5i= 1.292 Excluded /r3.5i=5.026 Excluded Excluded 64 -13.02 -19.44 £7=0.0592 £>=0.1234 /KO.OOOl Excluded Excluded Excluded Excluded F 1.54=3.716 F, .54=2.450 F 1. 5 4 —10.98 Excluded Excluded Excluded Excluded £7=0.0006 £7=0.1412 £7<0.0001 £1=0.0712 £7<0.0001 £7=0.8589 p=0.0078 F|,4o=13.75 FI.*,=2.245 Ft 40=26.96 F",.411=3.420 Ft.m=\3-57 Fx 40=0.2529 F 3.«=4.485 63 -29.42 -33.42 “square root transformed; AIC, Akaike information criterion; ‘excluded’ indicates terms excluded during model simplification (stepwise approach). 99 In both 2- and 10-year old stockpiled soil test plots, the highest mean aboveground biomass per seedling for seeding treatments was observed in soils treated with fertilizer and fertilizer + NAG sand. When comparing the two treatments, the mean aboveground biomass per seedling for seeding treatments in both soils was considerably higher in soils treated with fertilizer + NAG sand (with the exception o f locally blue wildrye and arctic lupine seeding treatments in 10-year old soil). For 2-year old stockpiled soil test plots, the addition o f NAG sand had no significant effect on the estimated belowground biomass per seedling (g seedling'1); however, a significant effect was observed for seeding treatment (p<0.0001; Table 18). Estimated belowground biomass per seedling for arctic lupine was significantly higher than all other seeding treatments (Figure 23). In the 10-year old stockpiled soil test plots, there was a significant interaction for estimated belowground biomass per seedling between NAG sand and seeding treatment and between NAG sand, fertilizer and seeding treatment (Table 18). Estimated belowground biomass per seedling significantly decreased in response to the addition o f NAG sand for arctic lupine (p<0.0001). In addition, fertilizer plus NAG sand further reduced estimated belowground biomass per seedling for arctic lupine (p<0.0078; Figure 23). 100 0.6 O aW «a> ■ Topsoil 0.5 □ Topsoil + N A G Sand C/5 0.4 Q. ^ S i* 93 S 0.3 - 1 1u M V3 ■8 0.2 - oL. w > u > o 0.1 C S XI < bromcil ecovar elym gla ecovar elym gla local lupiarc local Seeding T reatm ent Figure 22. Overall seeding treatment effects on mean values (±SE) o f the aboveground biom ass per seedling (g seedling'1) across all substrate treatments in 2-year (above) and 10-year (below ) old stockpiled soil. Single­ species seeding treatments include seeds obtained from genetically diverse seed increase plots (ecovars) o f fringed brome (BROMCIL), blue wildrye (ELYM GLA) or seeds collected from the local area o f the m inesite (local) o f blue wildrye and arctic lupine (LUPIARC). In both 2- and 10-year old stockpiled soil test plots, the highest mean estimated belowground biomass per seedling for seeding treatments was observed in soils treated with fertilizer and fertilizer + NAG sand (with the exception o f locally blue wildrye and arctic lupine seeding treatments in 10-year old soil). In NAG sand test plots, seeding treatments had no significant effect on aboveground and estimated belowground biomass and shoot:root ratio (Table 19). Seeding treatment also had no significant effect on aboveground and estimated belowground biomass per seedling (Table 20). 101 2-Year Soil 0.7 a. 0.6 05 a> 0.5 a5 ^*> § v L. CJ3 6# 0.4 - 5 St £4 ) —B 0.3 os 4, "O « Ol (A 0.2 td 0.1 - B B JL . JL B 10-Year Soil 0.7 ■ Topsoil 0.6 □ N A G sand ■ N A G sand + Fertilizer 73 U o> CA bromcil ecovar elym gla ecovar elym gla local lupiarc local Seeding T reatm ent Figure 23. Overall seeding treatment effects on mean values (±SE ) o f the estimated belowground biomass per seedling (g seedling'1) across all substrate treatments in 2-year (above) and 10-year (below) old stockpiled soil. Single-species seeding treatments include seeds obtained from genetically diverse seed increase plots (ecovars) o f fringed brome (BROMCIL), blue wildrye (ELYM GLA) or seeds collected from the local area o f the m inesite (local) o f blue wildrye and arctic lupine (LUPIARC). For 2 year old soils (above), bars which share a com m on letter are not significantly different (p<0.05). 102 Table 19. Summary o f linear mixed-effects model results for aboveground biomass, estimated belowground biomass and shootrroot ratio collected from N A G sand testplots in September 2010. D etails o f the minimal adequate models are listed in the rows, with explanatory variables and their corresponding F- and p-values, plus the number o f observations (bottom rows).___________________________________________________________ Predictor Variable NA G S a n d Seeding Treatment Num ber o f Observations Predictor Variable N A G Sand Seeding Treatment Num ber o f Observations A boveground B iom ass ^6.20=0.8447 P = 0.5522 Shoot:Root R atio R =0.5851 7=6,18=0.7983 28 26 Estimated B elow ground B iom ass F 4,5= 1.402 P = 0 .3757 11 Table 20. Summary o f linear mixed-effects model results for above- and estimated belowground biom ass per seedling calculated using aboveground biomass and seedling density collected from N A G sand testplots in September 2010. Details o f the minimal adequate models are listed in the rows, with explanatory variables and their corresponding F- and p-values, plus the number o f observations (bottom rows)._____________________ Predictor Variable A boveground B iom ass per Seedling Estimated Belowground B io m a ss per S eed lin g N A G Sand Seeding Treatment Num ber o f Observations R = 0.5507 F x , i=0.7473 16 F 3.h= 0 .8 7 5 3 /M 1 .4 9 3 2 16 103 4.0 Discussion The main objective o f this experiment was to determine whether NAG sand could be utilized as a supplement to stockpiled soil to effectively increase the quantity of adequate plant growth medium available for use in reclamation. The experiment set out to evaluate this option by determining how soil properties (physical and chemical) respond to the addition o f NAG sand. A number o f plant performance parameters were also used to evaluate the response o f plants to the addition o f NAG sand to soil. Plant-available nutrients (and trace elements) were estimated using a variety o f methods: exchangeable cations; Mehlich III extractable elements; Olson P, extractable P and mineral N (nitrate and ammonium). Overall concentrations of plant-available nutrients were reduced as a result of supplementing soil with NAG sand; however, physical conditions in supplemented soils (e.g., temperature, aeration) may have been more conducive for plant growth. To provide context for trace element concentrations, values were compared to CCME soil quality guidelines agricultural soil. Total concentrations of trace elements (including base metals) in soils supplemented with NAG sand were found to be below concentrations outlined in the CEQG for agricultural land, with the exception o f Cr, Mo, V and Cu (CCME 2004). However, when comparing soils supplemented with NAG sand to background levels found in non-supplemented soils, only V was found to be significantly higher in the supplemented soils. Vanadium is an element that has been shown to have low mobility and phytoavailability (Martin and Kaplan 1998). Plant performance on soils supplemented with NAG sand was shown to be equal to or higher than their performance in non-supplemented soils. With the addition o f fertilizer, 104 plant performance significantly increased in soils supplemented with or without NAG sand. When combined, the effects o f both NAG sand and fertilizer on some measures o f plant performance appeared to be additive and resulted in the highest levels of plant performance. Further to the main objective, the plant performance of three grass species (two varieties of blue wildrye, a single variety o f fringed brome and a single variety o f Rocky Mountain fescue) and a single legume (arctic lupine) were compared. Overall, the performance was highest for the two varieties o f blue wildrye compared to the other two grasses and legume. In addition, a significant difference in performance was also observed between two varieties of blue wildrye (locally specialized versus mixed genotype), with the highest performance observed for the mixed genotype variety. 4.1 Soil Characterization 4.1.1 Physical Soil Properties Differences in the physical properties between 2- and 10- year old stockpiled soils supplemented with or without NAG sand were apparent. The addition o f NAG sand to the soils resulted in an increase in the sand content and a proportional reduction in the clay, silt and organic matter. This was expected, as the 50:50 mixtures o f loam soil and NAG sand would necessarily have an intermediate soil texture. Bulk density o f the soils increased and soil porosity (which was not directly measured) decreased in response to the addition o f NAG sand. The change in bulk density and soil porosity also accompanied with a decrease in soil strength. As an incidental observation, the surfaces o f soils supplemented with NAG sand were visibly darker when compared to the non-supplemented soils, which may have influenced surface temperatures. Furthermore, visual assessments of the surface o f non­ 105 supplemented soils suggest that there may have been some surface sealing (i.e., crusting), which was not observed on soils supplemented with NAG sand. With the addition of NAG sand, the percentage o f sand in the fine fraction o f the stockpiled soils increased. Non-supplemented soils were classified as a loam and soils supplemented with NAG sand as a sandy loam (Troeh and Thompson 1993). Sandy soils are usually well aerated as most pore spaces within this soil type are large enough to allow water to enter and drain through quickly (Troeh and Thompson 1993). As a result o f the higher percentage o f sand, air and water infiltration in the soils likely increased with the addition o f NAG sand. The percentage o f clay and silt in the fine fraction o f the stockpiled soils decreased with the addition of NAG sand. Clay content contributes significantly to a soil’s ability to store water and nutrients (Brady and Weil 2002). Therefore, a reduction in clay content in soils supplemented with NAG sand likely resulted in a reduction in their water and nutrient storage capacity. The percentage o f SOM in the stockpiled soils also decreased with the addition of NAG sand. No appreciable amounts o f SOM were present in NAG sand; thus, the decrease in percent SOM of stockpiled soils in response to the addition o f NAG sand was expected. Soil organic matter is another important contributor to the storage and retention o f water and nutrients (Gregorich et al. 1994). Therefore, the decreased concentration in the SOM o f soils with the addition o f NAG sand may have also contributed to a reduction in their water and nutrient storage capacity. For soils supplemented with or without NAG sand, percent SOM (0.58 - 3.13 %) was lower than the average percent SOM in grassland soils (approximately 4 106 %) reported by Troeh and Thompson (1993). Consequently, it can be inferred that the contribution of SOM to nutrient and water storage in the stockpiled soils was substandard, regardless o f the NAG supplement. Bulk density in the stockpiled soils increased with the addition o f NAG sand. Models examining soil porosity by Stolf et al. (2011) showed that as bulk density increases, soil porosity (total porosity and macro- and microporosity) decreases. Based on these models, which estimate soil porosity using bulk density and percent sand content, the addition of NAG sand to stockpiled soils likely resulted in a decrease in total, macro- and microporosity. This seems valid as finer textured soils are associated with higher porosities and lower bulk densities (Brady and Weil 2002). However, the amount o f air-filled porosity o f stockpiled soils may have increased in response to the NAG sand supplement. In comparison to clay soils, sandy soils have higher water infiltration rates and thus, the capacity to retain water within the medium is reduced, resulting in increased air-filled porosity (Baker et al. 1999). Therefore, the addition o f NAG sand to stockpiled soils likely resulted in an increase to water infiltration and air-filled porosity. Soil strength at the surface o f soils supplemented with or without NAG sand was measured using a small handheld penetrometer. In responses to the addition o f NAG sand, soil strength decreased, and in both supplemented and non-supplemented soils, values were lower than 2 MPa, a level at which, root growth starts to be impacted (Lampurlanes and Cantero-Martinez 2003). It should be noted that soil strength varies with soil moisture. We did not measure soil moisture at the time o f penetrometer measurements but all samples were obtained over a 2 hour period. We expected soil moisture conditions to be relatively constant over the sample period. 107 A considerable difference in soil color was observed between stockpiled soils supplemented with or without NAG sand (Appendix 4). Soils supplemented with NAG sand were darker than non-supplemented soils and this difference in color may have resulted in a considerable difference in soil temperature during the growing season, especially before plant cover became well established. The darker the surface, the greater the amount o f solar radiation absorbed, which is then converted into heat. Darker soils have a low albedo, resulting in less reflection o f radiation than light-coloured soils. Thus the darker color o f soil supplemented with NAG sand may have equated to higher temperatures as compared to the non-supplemented soils. Dark soils do not always equate to warmer soils as the dark color is often due to higher organic matter content; soils with high organic matter have a very high water-holding capacity that when saturated, keeps the soil cool (Troeh and Thompson 1993). However, the dark color o f soil supplemented with NAG sand is due to the addition o f sand, a mineral material that has no organic matter content and low water-holding capacity; therefore, it is unlikely that the increased heat production from the absorbance o f solar radiation in soils supplemented with NAG sand would be impeded by wet soil conditions. In addition to soil color, soil texture may have also significantly influenced soil temperature. Soils with higher percent sand content are likely to have a lower heat capacity and greater thermal conductivity, thereby allowing for greater absorption and transfer o f heat energy throughout the soil (Shaw 1952). Therefore, soils supplemented with NAG sand may have experienced higher daily surface and sub-surface temperatures in comparison to the non-supplemented soils throughout the spring, summer and fall when snow cover did not interfere with solar irradiance. 108 Following periods of precipitation, what appeared to be the development o f a surface seal (a thin layer o f structureless material on the soil surface) was observed on soils that were not supplemented with NAG sand. In addition, other areas o f the minesite reclaimed with the same stockpiled soil exhibited water pooling on the surface too (Appendix 5). During periods o f heavy precipitation, aggregates on the soil surface can be broken down from the force o f raindrops into smaller particles (called slaking). These particles, along with dispersed clay particles, can then clog soil pores; as the soil dries, it forms a crust. Surface seals formed in this manner can significantly reduce water infiltration, limiting the availability of water for plant growth (Brady and Weil 2002). The development o f a surface seal was not observed on soils supplemented with NAG sand. 4.1.2 Chemical Soil Properties The addition o f NAG sand to 2- and 10-year old stockpiled soils resulted in considerable changes to their chemical properties. Cation exchange capacity was shown to decrease in stockpiled soils with the addition o f NAG sand. This was expected as the addition o f NAG sand would dilute the soil, lowering the organic matter and clay content, thereby reducing nutrient retention. As a result o f the lower capacity for nutrient retention, total and available macro- and micronutrient concentrations in soils supplemented with NAG sand were lower in comparison with non-supplemented soils (with a few exceptions) and concentrations o f some nutrients were considered deficient for plant growth. The pH of stockpiled soils increased with the addition of NAG sand, most notably in 10-year old soil. Electrical conductivity increased and calcium carbonate equivalence slightly decreased with the addition of NAG sand to stockpiled soils. 109 Cation exchange capacity in stockpiled soils decreased with the addition o f NAG sand, most likely in response to a reduction in SOM and clay content. The majority of negatively charged particles (micelles) that make up the exchange complex are composed o f clay and organic matter particles (Troeh and Thompson 1993). Consequently, the reduction in SOM and clay content observed in the stockpiled soils in response to the NAG sand supplement was likely responsible for the decrease in cation exchange capacity. In 2- and 10-year old stockpiled soil, several nutrient deficiencies were observed. Available concentrations (Mehlich III extraction) for P and K (10-year old soil), and total concentrations of N were found to be at levels deficient for plant growth (Marx et al. 1999). In addition, concentrations o f the micronutrients B and Zn were also found be deficient (Brady and Weil 2002). In response to the NAG sand supplement, concentrations o f macronutrients (with the exception o f P) and micronutrients (with the exception o f Cu, Fe and Ni) in the stockpiled soils decreased and further nutrient deficiencies for Mg, K (for 2year old soil) and the micronutrient Mn were observed. Concentrations o f macro- and micronutrients in soils supplemented with or without NAG sand were all below concentrations which would be considered toxic to plants (Brady and Weil 2002). W ith the addition o f NAG sand, pH in the stockpiled soils increased, most notably in 10-year old soil. It is hypothesized that the increase in pH was due to higher concentrations of calcium oxide (CaO) in NAG sand compared to stockpiled soils. With the addition of NAG sand, concentrations of CaO in the supplemented soils may have increased significantly. However, values of oxide concentrations were calculated from the determined elemental concentrations and therefore it is unknown whether Ca existed in the form o f CaO in the NAG sand or soil materials. In addition, the percent o f SOM in soils supplemented 110 with NAG sand was low and soils with a low SOM have been shown to have a lower buffering capacity as the complex structure o f SOM acts as an efficient reservoir for H+ ions, (Brady and Weil 2002). Soil pH has numerous effects on plant growth but its most important influence is on nutrient availability. Nutrient release through weathering, solubility in soil solution and storage associated with CEC is governed by soil pH. For optimum nutrient availability, a range in pH o f 6.0-7.5 has been suggested (Troeh and Thompson 1993). In response to the addition of NAG sand, both stockpiled soils where shown to have a pH at the high end o f this optimum range. Electrical conductivity increased in stockpiled soils with the addition o f NAG sand. Values o f EC in soils supplemented with or without NAG sand were not considered saline (<4 mS cm'2; Brady and Weil 2002). Soils which become saline can adversely affect plant growth by reducing their ability to absorb water and nutrients from the soil. However, from the changes in EC observed in the stockpiled soils with the addition of NAG sand, it is unlikely that soil salinity had any appreciable influence on plant growth. The percentage of calcium carbonate equivalence decreased in the stockpiled soils in response to the addition o f NAG sand. As a source o f Ca2+ ions, CaCCb (or Mg2+) can increase a soil’s pH by exchanging Ca2+ (or Mg2+) for H+ ions in the soil’s cation exchange complex. Carbonate (C O 32 ) ions may also form OH- through the hydrolysis o f water. Calcium carbonate equivalence in NAG sand was not considerably higher than the concentrations observed in stockpiled soils and overall, concentrations in both mediums are considered low. Thus, it is unlikely that CaCOs (or other carbonate) concentrations in NAG 111 sand played a significant role in the changes to pH observed in stockpiled soils with the addition o f NAG sand. With the addition of fertilizer, total and plant-available concentrations o f macro- and micronutrients in 2- and 10-year stockpiled soils were either similar to or higher than concentrations in unfertilized soils. Concentrations o f available mineral N (nitrate and ammonium) increased in 2-year old soil but in 10-year old soil remained relatively equal to concentrations observed in unfertilized soils. Available concentrations o f P (Olsen extractable) and exchangeable K (for 2-year old soil) were shown to increase in response to the addition o f fertilizer. No difference in the total concentrations of N and S, available Ca, Na, Mg, K, S or Zn was observed between fertilized and unfertilized soils. In addition, no significant change in CEC and pH was observed with the addition of fertilizer. CEC is not expected to change unless significant changes in pH are observed. Despite the increase in the availability o f N, P and K as a result o f fertilizer application, concentrations of macronutrients in fertilized and unfertilized soils may have been at levels considered to be deficient for plant growth (Marx et al. 1999). Soils with a low CEC have low capacity for nutrient retention and this is likely a factor that has contributed to deficiency levels o f macronutrients in the stockpiled soils. Concentrations of total trace elements in soils supplemented with or without NAG sand were found to be below concentrations outlined in the CEQG for agricultural land with the exception o f Cr, Mo, V and Cu (CCME 2004). However, when comparing total concentrations between soils supplemented with or without NAG sand, only V was shown to be higher in the supplemented soils. Total concentrations o f trace elements were expected to 112 be lower in soil supplemented with NAG sand compared to non-supplemented soils as it was hypothesized that NAG sand (originating from desulphurization tailings) would have relatively low total trace element content, and when combined with soil would not lead to in an increased trace element concentrations. Concentrations o f V in 2-year old soil were above CEQG concentrations and just below CEQG concentrations in 10-year old soil. With the addition of NAG sand, concentrations of V increased. This suggests that although natural levels o f V in the stockpiled soils are high, NAG sand contributed to even higher concentrations when mixed with soil. Vanadium is a transition metal that has been shown to have low mobility and phytoavailabilty in soils with low clay and organic matter content (Martin and Kaplan 1998). The term base metals refers to trace elements that are relevant to the mining industry; these are elements are usually used in pure form rather than as alloys (Lottermoser 2010). Concentrations of base metals in soils supplemented with or without NAG sand were found to be below concentrations outlined in the CEQG for agricultural land with the exception o f Cu and Mo (for 2 year old soil). Concentrations of base metals in soils supplemented with NAG sand were lower in comparison to non-supplemented soils with the exception o f Cu in 10-year old soil for which higher concentrations were found in NAG sand supplemented soils. Since NAG sand has relatively low base metal concentrations compared to nonsupplemented stockpiled soils, the lower base metal concentrations soils supplemented with NAG sand were expected. 113 4.2 Plant performance In response to the addition o f NAG sand to 2- and 10-year old stockpiled soil, plant performance was either similar to the performance observed in non-supplemented soils or showed a significant increase. When comparing measures o f plant performance between soils supplemented with or without NAG sand, no significant difference in percent emergence, seedling density, percent cover or plant height was found. In addition, no significant difference in percent cover per seedling was observed between 2-year soil supplemented with or without NAG sand; however in 10-year old soil, a significant interaction was observed between seeding treatment and NAG sand. Estimated belowground biomass for seeding treatments was significantly higher in soils supplemented with NAG sand compared to non-supplemented soils. Furthermore, the shoot:root ratio was shown to be significantly higher 2-year old soil supplemented with NAG sand compared to the nonsupplemented soil; the opposite trend was found in 10-year old soil. This evidence suggests that the addition o f NAG sand to stockpiled soils altered the physical properties o f soil in such that conditions for plant growth may have been slightly improved. It is hypothesized that surface temperature, water infiltration and soil aeration (i.e., air-filled porosity) were more favorable for plant establishment and growth in soils supplemented with NAG sand than those that were non-supplemented. Although soil temperatures were not measured in this experiment, the darker soil color and higher sand content o f soils supplemented with NAG sand suggests that they may have experienced higher soil temperatures throughout the growing season. In cold climates such as the sub-boreal/subalpine transition zone of the study area, soil temperature can significantly influence plant establishment and growth (specifically in the spring) by 114 increasing seed germination at the surface and improving nutrient and water uptake by roots (Shaw 1952). To infer germination and emergence rates, seeding densities and percent emergence were compared between soils supplemented with or without NAG sand. There was no significant increase in percent emergence or seedling density with the addition o f NAG sand. However, in 10-year old soil, slightly higher mean seedling densities were found in soils supplemented with NAG sand compared to non-supplemented soil. This finding may indicate the possibility that higher surface temperatures resulting from the darker color o f soils supplemented with NAG sand could positively influence germination rates. Soil temperature may have also influenced the production o f belowground biomass in soils supplemented with NAG sand. There was a significant interaction between the NAG sand supplement and seeding treatment in affecting the estimated belowground biomass for both 2- and 10-year old stockpiled soils. However, despite the interaction, all seeding treatments established in soils supplemented with NAG sand showed higher mean values o f estimated belowground biomass compared to seeding treatments in non-supplemented soils. The production of root mass for barley seedlings {Hordeum vulgare) grown in a growth chamber at uniform soil temperatures o f 10, 15 and 20°C as well as a vertical gradient o f 2010 °C (top to bottom) was observed by Fullner et al. (2011). The authors found that seedlings grown at the higher soil temperatures had significantly higher root mass and the highest root mass was found for seedlings grown in the vertical gradient. These findings support the hypothesis that potentially higher soil temperatures in soils supplemented with NAG sand compared to non-supplemented soils, may have contributed to a higher production of belowground biomass for seeding treatments. 115 Although it has been suggested here that soil temperatures in soils supplemented with NAG sand may have contributed to higher belowground biomass production for seedling treatments, the increase in air-filled porosity with the addition o f NAG sand may have also contributed. The production o f belowground biomass for creeping bentgrass (Agrostis stolonifera) in response to changes in soil aeration was demonstrated by Huang et al. (1999) when seedlings grown in a saturated medium were compared to seedlings grown in a well aerated medium. The authors showed that grass seedlings grown in a saturated medium had significantly lower dry root mass compared to seedlings from the well aerated medium. Although there is no indication that stockpiled soils in this experiment were frequently saturated, the addition o f sand to the soils likely increased water infiltration, reducing field capacity and increasing air-filled porosity, resulting in an increase in the productivity o f belowground biomass. When resources required by plants for growth are limited in one area, plants respond to the imbalance by allocating new biomass to organs in which the resources are most limiting. If plants are short o f C (due to shading or herbivory), they respond by allocating growth to shoots and if nutrients in the soil are limiting, plants respond by producing more root mass (Chapin et al. 1987). In this experiment, the shoot:root ratio for seedlings established in soils supplemented with or without NAG sand was compared. It was found that the shoot:root ratio was significantly lower in 2-year old soil supplemented with NAG sand compared to non-supplemented soil; the exact opposite relationship was observed for seedlings grown in 10-year old soil supplemented with or without NAG sand. In accordance with the growth allocation response explained by Chapin et al. (1987), these results suggest that 2-year old soil was more nutrient limiting in comparison to 10-year old soil. However, 116 despite this difference, seedlings compensated to the change in nutrient availability without a significant reduction in plant performance as root and shoot biomass was not adversely affected by the addition o f NAG sand. Fertilizers are often added to soil to improve plant performance by supplementing a soil’s natural fertility (Troeh and Thompson 1993). Various experiments examining plant performance in degraded soils have shown a significant increase in plant performance in response to the addition of fertilizer (Greipsson and Davy 1997; Burton and Burton 2000; Gardner et al. 2012). In this experiment, the addition o f fertilizer to 2- and 10-year old stockpiled soil significantly improved plant performance for the seeding treatments tested. Measures of plant performance, including percent emergence, seedling density, percent cover, plant height, aboveground and estimated belowground biomass, aboveground biomass per seedling and shoot:root ratio were significantly higher in fertilized soils, with the exception o f the shoot:root ratio for 10-year old soil. The increase in plant performance in fertilized soils is likely the result o f an increased availability o f primary macronutrients, including N (nitrate and ammonium), P and K, which were found to be at levels o f deficiency in the unfertilized soils. Fertilizer (13-16-10 expressed as % N, P2O5, K2O ) was added to the treatment substrates at a rate o f 576.9 kg ha'1. Slight increases in the availability o f these macronutrients were shown in fertilized soils during the second growing season, most notably for P. In this experiment, percent emergence for seeding treatments sown in 2- and 10-year old soils significantly increased in response to the addition o f fertilizer. Research testing the effects of fertilizer and sowing rate on seed emergence and seedling densities for similar grass and legume species was completed by Burton et al. (2006) at degraded sites in northern 117 BC. For all sowing densities tested, percent emergence was shown to be significantly higher on fertilized compared to non-fertilized soils. Additionally, in a glasshouse potting experiment by Agenbag and Villiers (1989), seedling emergence o f wild oat (Avena fatua) seeds sown in sandy and loamy soils treated with various concentrations o f limestone, ammonium nitrate (prills) and liquid ammonium nitrate were compared to unfertilized controls. Seedling emergence was shown to significantly increase in response to increasing concentrations o f nitrate application, all o f which had significantly higher percent seedling emergence compared to the unfertilized control treatment. The results from these two examples support the results o f our own experiment showing an increase in seedling emergence in response to the addition o f fertilizer, most notably when applied to degraded (i.e., nutrient poor) substrates. In a few experiments, the success o f seedling emergence has been shown to significantly increase in response to the addition o f fertilizers. In the experiment by Agenbag and Villiers (1989), the number o f wild oat seedlings significantly increased with increasing concentrations o f nitrate applied. Similar results were obtained by Greipsson and Davy (1997) in which the number o f dune-building grass (Leymus arenarius) seedlings, established on a sandy plain in southern Iceland, increased five-fold in response to the addition o f a slow-release fertilizer. Such findings agree with the results o f this experiment in which seedling densities in the 2- and 10 year old soil treatments significantly increased in response to the addition of fertilizer. Rather than positively influencing germination rates (thereby increasing seedling emergence), it seems more likely that the addition of fertilizer resulted in increased seedling survival during the first and second growing season, and hence higher seedling densities compared to unfertilized treatments. 118 The addition of fertilizer to 2- and 10-year old soil was shown to significantly increase the percent cover of plants established from seeding treatments. Similar results were observed by Burton and Burton (2000) during seed trials in northern British Columbia in which a specific seed mix was sown at a variety o f seeding densities on fertilized and unfertilized compacted forest soils. Regardless o f seeding density, Burton and Burton (2000) observed a significant increase in percent cover o f seedlings in response to the addition o f fertilizer. Plant height is a simple, non-destructive measurement often used in agricultural research for observing the effects o f fertilizer on plant performance. Bolton et al. (1982) measured the effects of drainage, crop rotation and fertilizer on yield, plant height and leaf nutrient composition o f com {Zea mays)', the results showed a significant increase in the height o f com plants in response to fertilizer. Similar responses o f plant height to fertilizer application have also been shown in many other agricultural experiments (Pertuit et al. 2001; Law-ogbomo and Law-ogbomo 2009). In the experiment reported here, the response o f plant height to fertilizer application was similar, with significantly greater plant heights observed for seedlings established in fertilized soils compared to seedlings growing in unfertilized soils. Biomass measurements also showed a significant increase in aboveground and estimated belowground biomass in response to fertilizer application (with the exception of estimated belowground biomass per seedling) which was consistent with the results o f our productivity proxies. Similar results were found in agricultural experiments in which fertilizer applications significantly increased crops yields for com (Bolton et al. 1982; Lawogbomo and Law-ogbomo 2009). In addition, many reclamation research trials with cover 119 crops (i.e., grasses and legumes) have also shown the same productivity response to fertilizer applications (Redente and Richards 1997; Gardner et al. 2012). In this experiment, there was a significant increase in the allocation o f growth to aboveground biomass for seedlings in fertilized compared to unfertilized soils. Increased allocation to aboveground biomass in fertilized plots suggests that soil nutrient limitations were alleviated by the fertilization treatment, as suggested by Chapin et al.’s (1987) interpretation o f limiting resources. It also suggests that available nutrients in fertilized soils were present at concentrations which may have no longer been deficient. In a greenhouse experiment by Bonifas et al. (2005), the response of root and shoot biomass o f com and velvetleaf (Abutilon theophrasti) was measured under a variety o f N applications. As the concentration o f N applied to pots was increased, the shoot:root ratio for plants increased; plants in the pots with no N added had the lowest shootrroot ratios. Therefore, the plants in this experiment allocated more growth to shoots in response to increasing N supply. Treating stockpiled soil with NAG sand or fertilizer resulted in similar or increased plant performance when compared to non-treated soils. When combined, the effect o f these treatments on some measures o f plant performance appeared to be additive and resulted in the highest levels of plant performance. The production of aboveground and estimated belowground biomass and the shoot:root ratio (10 year old soil) for most seeding treatments was shown to be highest in soils treated with both NAG sand and fertilizer, most notably for aboveground and estimated belowground biomass per seedling in 2-year old soil. No interaction between NAG sand and fertilizer was detected for any o f these measures o f plant performance. Therefore, it is likely that, when combined, the effects of NAG sand and fertilizer on soil do not significantly influence one another, but instead result in a cumulative 120 improvement to soil conditions for plant growth. An example o f the additive effects of multiple soil treatments on plant growth was likewise demonstrated by Bolten et al. (1982) who showed that the maximum yield for com crops resulted from the combined effects o f drainage spacing, crop rotation and fertilizer. 4.3 Species-specific plant performance The way in which plants respond to differences in their environment varies considerably among species. Although most plant species require a similar balance o f energy, water and mineral nutrients (Chapin et al. 1987), their strategies for responding to stress and disturbance (Grime 1977) can differ. These different strategies can confer significant advantages or disadvantages for a species under any given set o f environmental conditions (e.g., local adaptation to climate or soil conditions; Macel et al. 2007). In this experiment, significant differences were observed in the performance of four grasses and a legume sown on stockpiled soils treated with or without NAG sand and/or fertilizer. Interspecific differences in plant performance can be interpreted as resulting from different primary strategies for which they have evolved to respond to environmental stress and disturbance. Grime (1977) describes three primary plant strategies o f plant species in which species are adapted to thrive in specific permutations o f high and low levels o f stress and disturbance; these include competitors (adapted to low stress and disturbance), stress tolerators (adapted to high stress, low disturbance) and ruderals (adapted to low stress, but high disturbance). In an experiment where germination trials for seeds from 403 plant species were examined, Grime et al. (1981) ranked initial germinability o f freshly collected seed by family as follows: Poaceae >Asteraceae > Fabaceae and Cyperaceae > Apiaceae. In response to 121 storage (simulating overwintering conditions), germinability significantly increased for Poaceae and Asteraceae, but not Apiaceae and Fabaceae. The number o f seedlings which emerge each spring is dependent on the number o f seeds that germinate, and so it is logical to assume that given an equal number o f seeds, the number of grass seedlings emerging from a soil would be greater than the number o f legume seedlings due to their higher germinability. For this experiment, percent emergence and seedling density for blue wildrye, fringed brome and arctic lupine seeds sown at a similar rate (~750 PLS m'2) and under the same conditions (treated and non-treated stockpiled soils) during the second growing season were compared. Blue wildrye and fringed brome had significantly higher percent emergence compared to arctic lupine. The lower percent emergence o f arctic lupine in comparison to the blue wildrye and fringed brome agrees with the germinability ranking proposed by Grime et al. (1981) and in their findings which showed that seeds which were dark colored, had a thick seed coat and were dispersed through an explosive mechanism (dehiscence), had low germinability (often due to coat-enhanced dormancy; Bewley 1997). Such is the case for arctic lupine; where as seeds o f grass species such as blue wildrye and fringed brome, which are characterized as having features that are beneficial for germination on the soil surface (cylinder shaped with hygroscopic appendages), exhibited significantly higher germinability. In terms of its relationship to ranking o f germinability for plant families, it can only be speculated as to why a significantly lower seedling density and percent emergence was observed for fringed brome in comparison to the blue wildrye varieties as both are a member of the gramineae family. As suggested by Grime et al. (1981), species and even individuals o f the same species can differ in germination requirements (e.g., temperature and light conditions). Based on environmentally controlled germination tests for the two grass species 122 by Burton and Burton (2003), fringed brome was shown to have lower mean germination rates at cooler temperatures (15°/25 °C night/day; 57.7 %) compared to blue wildrye (79.2 %) for untreated seeds. In addition, stratification was suggested to slightly increase germination rates for blue wildrye. This evidence suggests that the lower seedling density o f fringed brome compared to blue wildrye (which were sown in our test plots located at a subalpine elevation) may be a reflection o f warmer temperature requirements for germination of fringed brome. Percent cover of vegetation is often used as a proxy for measures o f plant productivity. Grass cover is often perceived to be highly productive due to its dense and consolidated aboveground growth. In contrast, legume cover is often seen to be less productive because cover is patchy, shoots are usually highly branched and branch morphology can vary widely (e.g, sicklepod [Senna obtusifolia]; Smith and Jordan 1994). In this experiment, both varieties of blue wildrye had significantly higher percent cover compared to all other single-species seeding treatments on both 2- and 10-year old soil. On 2-year old soil, percent cover for the arctic lupine seeding treatment was significantly higher than the percent cover o f Rocky Mountain fescue and fringed brome; the trend was reversed in 10-year old soil. However, when comparing percent cover per seedling grown in 2-year old soil, arctic lupine had significantly higher percent cover compared to fringed brome and both varieties o f blue wildrye. Overall, the high aboveground productivity o f the three grass species in comparison to the single legume species suggests an ability of the grasses to more efficiently utilize resources from the environment, mostly in the first and second year. Plant height is another proxy o f plant productivity. Taller plants have a competitive advantage for light, giving them greater access to the resource, which often allows them to be 123 more productive (Falster and Westoby 2003). The mean height for the three grass species compared in this experiment was significantly greater than that o f the single legume, with the greatest heights observed for the two varieties o f blue wildrye. Therefore, when in competition, the taller stature o f the grass species would allow for greater access to the light resource in comparison to the legume. In response to fertilizer, plant height significantly increased for the grass species but did not for the legume. This further suggests that the grass species are able to utilize the increased availability of nutrients at greater rates in comparison to the legume. Competition between grass species and between grass and legume was also examined in this experiment and was shown to influence plant height. When sown together in the locally native seed mix (consisting o f blue wildrye, arctic lupine and Merten’s sedge), mean height values for blue wildrye and arctic lupine increased when compared to their mean heights in single-species seeding treatments, most notably in fertilized soil (no emergence o f Merten’s sedge was detected in the second growing season). These results suggest that when nutrient resources are not limiting (as assumed from the application of fertilizer), blue wildrye and arctic lupine respond to competition with each other by increasing in height. However, caution should be taken when interpreting these results, as the N-fixing capacity o f Rhizobium nodules (which were detected on the roots o f lupines collected from the test plots) may have contributed to the N pool in the locally native mixed seeding treatment plots, influencing the growth response of all species. For the ecovar native mix seeding treatment (consisting o f blue wildrye, Rocky Mountain fescue and fringed brome) sown in the stockpiled soils, the mean height o f fringed brome and Rocky Mountain fescue either decreased or showed no significant change in comparison to their mean heights in the single124 species seeding treatments (most notably in 10-year old soil), and the mean height o f blue wildrye increased significantly. This suggests that the competitive ability o f blue wildrye may be greater than that of fringed brome and Rocky Mountain fescue in that its ability to access light resources is enhanced in the presence of competition with other species. The most direct measure o f net productivity for a plant species is the amount of aboveground and estimated belowground biomass produced over a fixed given period o f time. In this experiment, the production of aboveground biomass in response to the NAG sand supplement or fertilizer was species dependent and overall, the most productive species were the two varieties o f blue wildrye. This result further supports the hypothesis that blue wildrye retains a greater competitive advantage in the capture and utilization o f nutrients in comparison to Rocky Mountain fescue, fringed brome and arctic lupine in the stockpiled soils. With a reduction in the availability o f nutrients (as seen with the addition o f NAG sand), the four grass species tested were able to capture and utilize enough resources to maintain productivity compared to the single legume for which productivity decreased (i.e., aboveground biomass per seedling). These findings further support the interpretation that blue wildrye has a competitive advantage in the capture and utilization o f nutrients compared to arctic lupine. In contrast, the high production o f belowground biomass for arctic lupine suggests a greater investment in the development o f biomass for the utilization of soil resources (e.g., water and mineral nutrients) in comparison to the four grass species tested. Arctic lupines are associated with nitrogen-fixing Rhizobium. Other than the production o f plant-available 125 N through atmospheric N fixation, rhizobacteria have been shown to provide a variety o f other direct and indirect benefits to plants including protection from phytopathogens, sequestration of plant-available forms o f Fe, solubilisation o f minerals including P and the production o f plant growth regulators (Noel et al. 1996). These additional benefits to arctic lupine may provide this species with a greater tolerance to stress in comparison to the grass species, which do not have these symbiotic relationships. Chapin et al. (1987) explains a large allocation o f growth to roots is usually in response to low nutrient availability in the soil. This may explain the significant increase in root biomass for arctic lupine in response to the dilution o f soil (and therefore the dilution o f mineral nutrients) through the addition o f NAG sand. Similar results were obtained by Nicholas and McGinnies (1982) when the growth of two legumes, alfalfa (Medicago sativa) and cicer milkvetch (Astragalus cicer), established in topsoil or coal mine spoil (a mixture o f shale and sandstone that was shown to have lower nutrient availability than the topsoil) was compared. Both legumes produced significantly greater root biomass when established in mine spoil compared to topsoil. All other grass species in this experiment displayed this same trend with either similar or slightly increased levels o f estimated belowground biomass in response to the NAG sand supplement. On the other hand, the decrease in estimated belowground biomass for arctic lupine in response to the addition o f NAG sand was unexpected and remains unexplained. The grass species contributed a significantly higher proportion o f their growth to aboveground biomass in comparison to the arctic lupine. Root architecture differs greatly between three grasses and the single legume species examined here. The grass species utilize a fibrous root system, whereas the legume, arctic lupine, utilizes a taproot. This difference in 126 architecture is likely what determined the difference in shoot: root ratios between the two lifeforms. Taproots have the advantage o f reaching deeper into the soil to access water, which suggests arctic lupine may have a significant advantage in dealing with drought stress compared to the fibrous rooted grass species. However, fibrous roots have the advantage o f absorbing available nutrients at greater rates than that o f taproots, suggesting a competitive advantage under nutrient-poor conditions for the grass species. Growth allocation responses to fertilizer also differed significantly between the grass and legume species. With the addition o f fertilizer, there was a significant increase in allocation to aboveground biomass for all grass species; however, allocation o f growth for arctic lupine showed no significant change. This significant shift in allocation in response to fertilizer for the grass species suggests a greater morphological plasticity in comparison to the arctic lupine which showed no significant allocation response. Species with a greater morphological plasticity are likely to have a competitive advantage (Grime et al. 1988) in that their ability to direct growth towards limited resources (as inferred by Chapin et al. 1987) is greater than species with limited morphological plasticity. The differences in plant performance among species in this experiment reflect different plant growth strategies that have evolved to respond to environmental stress and disturbance (competition, abiotic stress or frequent disturbance; Grime et al. 1988). The growth characteristics displayed by the three grass species (blue wildrye, Rocky Mountain fescue and fringed brome) suggests a primary competitive strategy. These characteristics included high seed germinability, tall stature and dense canopies (which was less apparent with fringed brome and Rocky Mountain fescue) and an ability to quickly utilize and improve access to resources in response to competition with other species (e.g., light and 127 nutrients), leading to high morphological plasticity, ha comparison to the grass species, the growth characteristics displayed by arctic lupine suggest a stress-tolerator primary strategy. Characteristics displayed by arctic lupine included low seed germinability, low stature (in comparison to the grass species), no significant response to competition with other species (i.e., blue wildrye), greater ability to store resources through the use of thick rhizomes and association with nitrogen-fixing Rhizobium and low morphological plasticity. In addition to the differences in plant performance between the grass and legume species, significant differences in plant performance were also observed between two varieties o f blue wildrye. The two varieties consisted o f seeds collected from the local area around the minesite (locally specialized genotype) and from seed increase plots (Industrial Forest Service Ltd.) in which seeds from a variety o f locations were sown to create genetically diverse seed crops (mixed genotype or ecovar). Overall, the blue wildrye mixed genotype had higher mean values for all measures o f plant performance compared to the locally specialized genotype (in most cases the differences were significant) with a few exceptions. It is therefore concluded that the benefits associated with establishment from seeds sourced from a broader geographic area, may be greater than the benefits associated with a narrower genotype supposedly adapted to local soil and climate conditions. However, let it be noted that stockpiled soils were highly disturbed (classified as Anthroposols) and therefore, not representative o f the undisturbed soil conditions for the area. Thus, the higher performance observed by the blue wildrye mixed genotype in comparison to the local genotype may also be a reflection o f its ability to establish in poor soil conditions. Although adaptations to local climate and soil conditions for a specific plant species can provide benefits to plant fitness (Macel et al. 2007), the benefits may not always result in 128 superior productivity when compared to more genetically diverse varieties o f the same species. In an experiment by Bischoff et al. (2010), performance was assessed for four species o f wildflowers, established from seeds that were collected from a variety of individuals (mother plants) from the local area o f the experimental site and from four distant provenances (distance ranged from 120 to 900 km from the experimental site). Significant differences in fitness-related traits were observed between plants established from seeds collected from four distant provenances (locations) and the local area. However, individuals established from seeds collected from the local area showed no superior fitness (significantly higher values for fitness-related traits) compared to the four provenances. Furthermore, the productivity o f genetically diverse seed plots (consisting of plants established from seeds o f 12 mother plants) was greater than that o f low diversity plots (established from seeds o f 2 mother plants). These results are similar to the results o f this experiment in which fitnessrelated traits (e.g., height, seedling density and biomass) were shown to be greater for blue wildrye plants established from a mixed genotype seeding treatment. Both experiments thus support the theory that benefits of local adaptation may not always results in superior productivity when compared to genetically diverse populations o f the same species. Theory on local adaptation, as explained by Bischoff et al. (2010), predicts that the performance o f local genotypes should be greater than genotypes o f distant provenances when grown at the local site. However, this theory assumes that conditions at the local site have not been altered from the conditions in which the local genotype has adapted to over time. In this experiment, the seeds o f the local and mixed genotypes of blue wildrye were sown in stockpiled soil in which soil conditions did not resemble those prior to disturbance. 129 In this case, a mixed genotype exhibits greater benefits when sown in such altered soil conditions to which neither genotype is specifically adapted to. In contrast to the results of this experiment, Burton (2007) observed greater plant performance (i.e., percent emergence, percent cover and total biomass) following two growing seasons (2006 and 2007) for the blue wildrye seedlings established from the local genotype in comparison to the mixed genotype (ecovar) when sown in disturbed soil at the same minesite. Weather conditions (i.e., mean monthy total precipitation and minimum and maximum temperatures; Appendix 6 and 7 respectively) during the growing seasons for this experiment and the experiment conducted by Burton (2007) were compared. Overall, no considerable differences in total precipitation and temperatures were visible between the periods of these two experiments. Therefore, it is likely that the contrasting results may be due to a considerable difference in soil conditions between the stockpiled soils and those o f the soil covered area in which the revegetation trial by Burton (2007) was conducted (although soil conditions for this experiment were not tested and therefore cannot be compared to the conditions o f the stockpiled soils). It may therefore be hypothesized that when comparing the local to mixed genotype o f blue wildrye, superior plant performance o f one genotype over the other may be strongly influenced by site specific soil conditions. This hypothesis is further supported by our findings that mean height o f blue wildrye seedlings established from local seeds was less than the mean height for the mixed genotype in 2-year old soil, but in 10-year old soil, the opposite trend was observed. 130 4.4 Limitations o f Experimental Design A few limitations became apparent in this experiment, mostly related to the design o f the revegetation test plots. During the initial construction and seeding of the text plots, the pattern in which soil and seeding treatments were applied to subplots was not different between test plots; in effect, the treatments were not randomized. It is therefore possible that the pattern in which treatments were applied may have had confounding effects in that one subplot with a specific combination o f treatments may have had unforeseen influences on its neighboring subplot and this influence may have then been expressed across all test plots. During the construction of the test plots, the test plots constructed with 2-year old soil were all located at the east end o f the minesite and those constructed of 10-year old soil were all located at the west end o f the minesite. These areas differed slightly in elevation and exposure, which may have resulted in a slight difference in the influence o f climate and weather between the two soils. In addition, a considerable difference in the amount o f original vegetation cover was noted between 2- and 10-year old soils, including the locations in which these soils were excavated for construction o f the test plots. The 2-year old stockpiled soil had almost no vegetation cover; only a few incidents of volunteer colonization were detected. However, along most o f the surface o f 10-year old stockpiled soil, a wellestablished herbaceous cover was present. This difference may have resulted in a considerable difference in soil characteristics, most notably in terms of the amount o f biological activity (which was not measured in this experiment). Therefore, in order to avoid the influence of potential differences in climate and biological soil activity, the effects o f soil and seeding treatments on plant performance in 2- and 10-year old soil were analyzed 131 separately and only minor comparisons for plant performance and soil characteristics between the two soils were made. In order to calculate the amount o f seed required for sowing seeding treatments at a standardized density, the pure live seed (PLS) count for each seed stock had to be calculated. PLS is calculated using the percent germination and percent purity of a seed stock to obtain an accurate count o f the number o f germinating seeds within a given weight o f seed stock (PLS g '1). At a targeted sowing rate o f 750 PLS m'2, the PLS g’1 was used to calculate the weight of seed stock required for seeding treatments applied to the lm2 subplots. For the seed stocks supplied by IFS Ltd. (ecovars; blue wildrye, Rocky Mountain fescue and fringed brome), values o f percent germination and percent purity from a past analysis were used to calculated PLS. For the remaining seed stocks (locally collected blue wildrye, arctic lupine and Merten’s sedge), percent purity was assessed following seed cleaning, however, due to time constraints, percent germination was estimated. Percent germination o f blue wildrye seeds was estimated at 80 %, arctic lupine at 44 % and Merten’s sedge at 32.7 % using the lowest values reported by Burton and Burton (2003). To assess the accuracy o f our estimated germination rates, a variety o f germination tests were conducted. In the fall o f 2008, germination tests o f arctic lupine seeds were conducted by 20/20 seed labs and in 2009 and 2010, seed germination trials were also conducted at UNBC for seeding treatments sown in the test plots. A comparison of germination rates used for calculating sowing rates and actual germination rates determined from germination testing o f seeding treatments is shown in Appendix 8. Due to these differences in germination rates, actual sowing rates were either higher or lower than the targeted sowing rate. The actual sowing rates are as follows: ecovar blue wildrye (416 PLS m’2), ecovar fringed brome (314 PLS m’2), ecovar Rocky Mountain 132 fescue (642 PLS m '2), local blue wildrye (703 PLS m'2) and local arctic lupine (847 PLS m ' 2), local native seed mix (1105 PLS m‘2) and ecovar native seed mix (482 PLS m '2). Assessment o f the accuracy o f estimated germination rates for Merten’s sedge is not necessary as no seedlings were detected in the test plots for this species during the second growing season. During the literature review for this experiment, an important factor in the assessment of NAG sand as a supplement to soil, which was not examined in this experiment, was recognized. One o f the major constituents in the desulphurization process for Cu tailings (and therefore in the production of NAG sand) is the compound known as xanthate. In a laboratory experiment conducted by Xu et al. (1988), the effects o f various concentrations o f xanthate on duckweed (Lemna minor) and a species o f daphnid (Daphnia magna) was examined. The results showed that concentrations o f xanthate >2 mg L '1 were toxic to both species and concluded that xanthate should be considered a micropollutant to aquatic systems. However, the experiment also revealed a rapid rate o f degradation for xanthate (2.5 to 4 day half-life) which suggests a short-lived presence in aquatic environments. In this experiment, we did not test for xanthate concentrations in NAG sand and so the potential effects of supplementing soil with NAG sand to aquatic environments downstream from which this treatment would be applied are unknown, as are its effects on terrestrial plants and microbes. 133 5.0 Conclusions and Recommendations This study confirms that the addition o f NAG sand to stockpiled soils can increase the quantity of growth medium available (i.e. Anthroposols) for reclamation while maintaining plant performance observed in non-supplemented soils. When combined w ith a fertilizer application, plant performance on soils supplemented with NAG sand can be significantly increased over non-supplemented soils. Concentrations o f total and extractable trace elements (including base metals) in soils supplemented with NAG sand are not likely to result in any adverse effects to plants or the local environment. However, it is recommended that use o f NAG sand as a soil supplement be limited to hydrologically isolated areas o f the minesite (e.g. tailings impoundments and tailings-filled pits) until further studies examine the residual concentrations o f xanthate in NAG sand, and the potential mobility o f xanthate into surrounding aquatic ecosystems. This study has shown that blue wildrye is an excellent candidate for the revegetation o f tailings and other minesoils at Huckleberry Mine. In terms o f seed source, plant performance o f blue wildrye established from a mixed genotype seed source (ecovar) performed better than that o f the local genotype. However, previous research comparing these two seed sources has shown conflicting results. Thus, in order to achieve an optimal cover of blue wildrye, the best result may come from the use o f seed stocks which are composed of both genotypes. Plant establishment and productivity on 10-year old soil appears to be greater than establishment and productivity on the 2-year old soil (however, this was not statistically tested). This result is most apparent when comparing percent cover and plant height o f 134 plants established in the two different-aged soils. Assuming the effects o f elevation and climate differences between the locations o f 2- and 10-year old stockpiled soil test plots are negligible, physical and chemical conditions in the 10-year old stockpiled soil are likely to be more conducive to plant growth. It is hypothesized that the considerably higher amount o f vegetation cover established on the 10-year old in comparison to the 2-year old stockpiled soil may have resulted in a greater amount o f biological contributions. The amount o f SOM and ammonium, mineralizeable N and total N was higher in 10-year old compared to 2-year old soil and this may be an indication a greater amount o f biological contributions resulting from the well-established vegetation cover. Huckleberry Mine may want to re-evaluate their 2010 reclamation plan, given the results o f this study. During the Huckleberry Mine Closure Meeting workshop in March 2009, the use o f a soil cover in place o f the currently proposed water cover for TMF-2 was investigated. It was determined that although the use o f a soil cover would provide an adequate cover for PAG (potentially acid generating) materials, the efforts required to obtain the necessary soil materials needed to completely cover the impoundment were deemed impractical (Boxill 2010). However, if the conservative approach to the use o f soil supplies examined in this experiment is applied to the surface o f TMF-2, successful revegetation o f the entire impoundment may be possible. In order to investigate the feasibility o f this alternative, further revegetation trials utilizing this technique are recommended. Large-scale revegetation plots (e.g. 0.25 ha), in which soil is mixed into the NAG sand surface, should be constmcted directly on the tailings impoundment and sown using a standard native seed mix. In order to assess the potential financial costs of soil application, various methods and rates o f application should be tested. 135 In addition, the use of fertilizer and peat moss as amendments to NAG sand should also be tested as an alternative method for improving plant performance on the surface o f the impoundment. Currently, analysis o f metal uptake for plant samples collected from stockpiled soils supplemented with or without NAG sand is being conducted. This information will allow us to identify whether concentrations o f metals within plant tissues possess a potential risk to browsing animals (e.g., ungulates) which feed on native grasses in the area o f the minesite. However, if further tests o f using NAG sand in reclamation are conducted, they should be accompanied by further analysis o f metal uptake. We expect metal uptake to be low, based on the low total and Mehlich III extractable metal concentrations in soils supplemented with NAG sand. 136 6.0 Literature Cited Adams, C. R., and J. W. 2009. Sorenson. Upper Mississippi River floodplain forest seed rain shows variable potential for forest regeneration. Proceedings o f the Ninety-forth Ecological Society of America Annual Meeting; 2009, August 2-7; Albuquerque, New Mexico. Ecological Society of America, http://www.esa.org/ Agenbag, G. A., and O. T. De Villiers. 1989. The effect o f nitrogen fertilizers on germination and seedling emergence o f wild oat (A .fatua L.) seed in different soils. Weed Research 29: 239-245. Aiken, G. E., and J. W. Gunett. 1990. Overburden removal, pp. 584-619. In: B. A. Kennedy (Ed.), Surface Mining. Maryland, USA: Port City Press, Inc. Antonio, C. D., and L. A. Myerson. 2002. Exotic plant species as problems and solutions in ecological restoration: a synthesis. Restoration Ecology 10: 703-713. Alvarenga, P. 2009. Reclamation o f a mine contaminated soil using biologically reactive organic matrices. Journal of Waste Management and Research 27: 101-111. Bates, D., M. Maechler, and B. Bolker. 2011. Ime4: linear mixed-effects models using s4 classes. R package version 0.999375-39. Retrieved February 5th, 2012, from http://lme4.rforge.r-proiect.org/ Baker, S. W., S. J. Mooney, and A. Cook. 1999. The effect o f sand type and rootzone amendments on golf green performance: I, soil properties. Journal of Turfgrass Science 75: 2-17. Barth, R. C., and K. M. Brooks. 1983. Soil depth requirements for revegetation o f surfacemined areas in Wyoming, Montana and North Dakota. Journal o f Environmental Quality 13: 399-404. Bendfeldt, E. S., J. A. Burger, and W. L. Daniels. 2001. Quality o f amended soils after sixteen years. Soil Science Society o f America 65:1736-1744. Benzaazoua, M., B. Bussiere, M. Kongolo, J. Mclaughlin, and P. Marion. 2000. Environmental desulphurization of four Canadian mine tailings using froth flotation. International Journal o f Mineral Processing 60: 57-74. Bewley, J. D. 1997. Seed germination and dormancy. The Plant Cell 9: 1055-1066. Bischoff, A., T. Steinger, and H. Muller-Scharer. 2010. The importance o f plant provenance and genotypic diversity o f seed material used for ecological restoration. Ecological Restoration 18: 338-348. Blumward, E., G. S. Aharon, and M. P. Apse. 2000. Sodium transport in plant cells. Biochimica et Biophysica Acta 1465: 140-151. 137 Bolton, E. F., V. A. Dirks, and M. M. McDonnel. 1982. The effect of drainage, rotation and fertilizer on com yield, plant height, leaf nutrient composition and physical properties o f Brookston Clay soil in southwestern Ontario. Canadian Journal o f Soil Science 62: 297-309. Bonifas, K. D., D. T. Walters, K. G. Cassman, and J. L. Lindquist. 2005. Nitrogen supply affects root:shoot ratio in com and velvetleaf (Abutilon theophrasti). Weed Science 53: 670675. Borden, R. K., and R. Black. 2005. Volunteer revegetation o f waste rock surfaces at the Bingham Canyon Mine, Utah. Journal o f Environmental Quality 34: 2234-2242. Bowen, C. K., G. E. Schuman, R. A. Olson, and L. J. Ingram. 2005. Influence o f topsoil depth on plant and soil attributes o f 24-year old reclaimed mined lands. Arid Land and Research Management 19: 267-284. Boxill, L. 2010. Huckleberry Mine reclamation and closure plan 2010 update. Prepared for Huckleberry Mines Ltd. by SRK Consulting Inc., Vancouver, British Columbia. Bradshaw, A. 2000. The use o f natural processes in reclamation - advantages and difficulties. Landscape and Urban Planning 51:89-100. Brady, N. C. and R. R. Weil. 2002. The nature and properties o f soils (13th ed.). New Jersey, USA: Pearson Education Inc. 960pp. Brown, L. F., and C. L. Jackson. 1984. Reclamation o f the Urad Molybdenum mine, Empire Colorado. Journal of Minerals and the Environment 6: 77-82. Brynes, W. R., and J. G. Stockton. 1980. Characteristics of coal mine overburden important to plant growth. Journal o f Environmental Quality 10: 300-308. Burton, C. M. 2007. Evaluation o f plant and tailings treatment for reclamation. Prepared for Huckleberry Mines Ltd. by Symbios Research and Restoration, Prince George, British Columbia. Burton, C. M., and P. J. Burton. 2003. A manual for growing and using seed from herbaceous plants native to the interior o f northern British Columbia. Symbios Research and Restoration, Smithers, British Columbia. 168pp. Burton, C. M., and P. J. Burton. 2001. Fertilization can reduce the amount o f seed needed to revegetate degraded soils. Ecological Restoration 19: 57-58. Burton, C. M., P. J. Burton, R. Hebda, and N. Turner. 2006. Determining the optimal sowing density for a mixture o f native plants used to revegetate degraded ecosystems. Restoration Ecology 14: 379-390. Carson, A. W., D. F. Polster, and P. J. Burton. 2011. Agronomic grasses and legumes: do they help or hinder recovery. Canadian Reclamation; Issue 3, Spring/Summer 2011. p 34-37. 138 [CCME] Canadian Council o f Ministers o f Environment. 2004. Canadian environmental quality guidelines. Update 4.0. Canadian Council o f Ministers o f Environment, Winnipeg, Manitoba. Chapin, F. S., A. J. Bloom, C. B. Field, and R. H. Waring. 1987. Plant responses to multiple environmental factors. Bioscience 37: 49-57. Cohen, M., and M. Mackauer. 1986. Lupine Aphid, Macrosiphum albifrons (Homoptera: Aphididae): distribution o f hymenopterous parasites in British Columbia (Canada). Environmental Entomology 15: 719-722. Coyne, M. S., and Thompson J. A. 2006. Fundamental soil science. New York, USA: Thomson Delmar Learning. 403pp. Crawley, M. J. 2007. The R book. W est Sussex, England: John Wiley and Sons Ltd. 942pp. Davis, A., L. E. Eary, and S. Helgen. 1999. Assessing the efficacy of lime amendment to geochemically stabilize mine tailings. Environmental Science and Technology 33: 26262632. Deshaies, A., S. Boudreau, and K. A. Harper. 2009. Assisted revegetation in subarctic environment: effects of fertilization on the performance of three indigenous plant species. Arctic, Antarctic and Alpine Research 41: 434-441. Environment Canada. 2012. Frequently asked questions: weather. Retrieved June 6, 2012, from htto://www.weatheroffice.gc.ca/mainmenu/faq e.html [EPA] Environmental Protection Agency. 2006. The use of soil amendment for remediation, revitalization and reuse (EPA Publication No: 542-R-07-013). Cincinnati, OH: U.S. Environmental Protection Agency. Fenner, M., and W. G. Lee. 1989. Growth o f seedlings o f pasture grasses and legumes deprived o f single mineral nutrients. Journal o f Applied Ecology. 26: 223-232. Forbes, B. C., and R. L. Jefferies. 1999. Revegetation o f disturbed arctic sites: constraints and applications. Biological Conservation 88: 15-24. Fuentes, D., A. Valdecantos, J. Cortina, and V. R. Vallejo. 2007. Seedling performance in sewage sludge-amended degraded meditterranean woodlands. Journal o f Ecological Engineering 31: 281-291. Fullner, K., V. M. Temperton, U. Rascher, S. Jahnke, R. Rist, U. Schurr, and A. J. Kuhn. 2011. Vertical gradient in soil temperature stimulates development and increases biomass accumulation in barley. Plant, Cell and Environment 35: 884-892. Gardner, W. C., M. A. Naeth, K. Broersma, D. S. Chanasyk, and A. M. Jobson. 2012. Influence o f Biosolids and fertilizer amendments on element concentrations and revegetation o f copper mine tailings. Canadian Journal of Soil Science 92: 89-102. 139 Gregorich, E. G., M. R. Carter, D. A. Angers, C. M. Monreal, and B. H. Ellert. 1994. Towards a minimum data set to assess soil organic matter quality in agricultural soils. Canadian Journal of Soil Science 74: 367-385. Greipsson, S., and A. J. Davy. 1997. Response o f Leymus arenarius to nutrients: improvement o f seed production and seedling establishment for land reclamation. Journal o f Applied Ecology 34: 1165-1176. Gretarsdottir, J., A. L. Aradottir, V. Vandvik, E. Heegaard, and H. J. B. Birks. 2004. Long­ term effects of reclamation treatments on plant succession in Iceland. Restoration Ecology 12: 268-278. Grime, J. P. 1977. Evidence for the existence o f three evolutionary strategies in plants and its relevance to ecological and evolutionary theory. The American Naturalist 111: 1169-1194. Grime, J. P., J. G. Hodgson, and R. Hunt. 1988. Comparative plant ecology: a functional approach to common British species. London, England: Unwin Hyman Ltd. 742pp. Grime, J. P., G. Mason, A. V. Curtis, J. Rodman, S. R. Band, M. A. G. Mowforth, A. M. Neal, and S. Shaw. 1981. A comparative study o f germination characteristics in a local flora. Journal of Ecology 69: 1017-1059. Hargis, N. E., and E. F. Redente. 1984. Soil handling for surface mine reclamation. Journal o f Soil and Water Conservation 39: 300-305. Harris, J. A., P. Birch, and K. C. Short. 1989. Changes in the microbial community and physico-chemical characteristics o f topsoils stockpiled during opencast mining. Soil Use and Management 5: 161-168. Hazelton, P., and B. Murphy. 2007. Interpreting soil test results: what do all the numbers mean. Collingwood, Australia: CSIRO Publishing. 152pp. Hild, A. L., G. E. Shuman, L .E. Vicklund, and M. I. Williams. 2006. Canopy growth and density of Wyoming Big Sagebrush with cool season perennial grasses. Arid Land Research and Management 20:183-194. Hillel, D. 1998. Environmental soil physics. California, USA: Academic Press Inc. 771pp. Hodacova, D., and K. Prach. 2003. Spoil heaps from brown coal mining: technical reclamation versus spontaneous revegetation. Restoration Ecology 11: 385-391. Holl, K. D. 2002. Long term vegetation recovery on reclaimed coal surface mines in the eastern USA. Journal o f Applied Ecology 39: 960-970. Hothom, T., F. Bretz, and P. Westfall. 2008. Simultaneous inference in general parametric models. Biometrical Journal 50: 346-363. Huddleston, R. T., and T. P. Young. 2004. Spacing and competition between planted grass plugs and pre-existing perennial grasses in a restoration site in Oregon. Restoration Ecology 12: 546-551. 140 Iverson, L. R., and M. K. Wali. 1981. Buried, viable seeds and their relation to revegetation after surface mining. Journal o f Range Management 35: 648-652. Jennings, S. R., Neuman, D. R., and Blicker, P.S. 2008. Acid mine drainage and effects on fish health and ecology: A Review. Prepared for the United States Fish and W ildlife Service by Reclamation Research Group, Bozeman, Montana. Johnson, D. B., and K. B. Hallberg. 2005. Acid mine drainage remediation options: a review. Science o f the Total Environment 338: 3-14. Jones, T. A. 2003. The restoration gene pool concept: beyond the native versus non-native concept. Restoration Ecology 3: 281-290. Kaye, T. N. 1997. Seed dormancy in high elevation plants: Implications for ecology and restoration, p. 115-120. In: T. N. Kaye, A. Liston, R. M. Love, D. L. Luoma, R. J. Meinke, and M. V. Wilson (eds.). Conservation and Management of Native Plants and Fungi. Oregon, USA: Native Plant Society o f Oregon. 296pp. Kramer, P. A, D. Zabowski, G. Scherer, and R. L. Everett. 2000. Native plant restoration o f copper mine tailings:II. Field survival, growth and nutrient uptake. Journal o f Environmental Quality 29: 1770-1777. Kubeckova, K., J. Johansen, S. D. Warren, and R. Sparks. 2003. Development o f immobilized cyanobacterial amendments for reclamation o f microbiotic soil crusts. Symposium of the International Association for Cyanophyte Research; 2003. Barcelona, Spain 48: 341-362. Kuhn, M., S. Weston, J. Wing, and J. Forester. 2010. Contrast: A collection o f contrast methods. R package version 0.13. Retrieved from http://cran.rproiect.org/web/packages/contrast/index.html Kumar, D., and B. Singh. 2003. The use o f coal fly ash in sodic soil reclamation. Land Degradation and Development 14: 285-299. Lampurlanes, J., and C. Cantero-Martinez. 2003. Soil bulk density and penetration resistance under different tillage and crop management systems and their relationship with barely root growth. Agronomy Journal 95: 526-536. Law-ogbomo, K. E., and J. E. Law-ogbomo. 2009. The performance of Zea mays as influenced byN PK fertilizer application. Notulae Scientia Biologicae 1: 59-62. Lehmann, J. 2007. Bio-energy in the black. Frontiers in Ecology and the Environment 5: 381-387. Little, T. M., and F. J. Hills. 1978. Agricultural experimentation: design and analysis. New York, USA: John Wiley and Sons. 350pp. Logan, M. 2008. Biology: extending the analysis tool kit for biologists. R package version 1.0. Retrieved from http://users.monash.edu.au/~murrav/stats/ downloads.html 141 Lottermoser, B. G. 2010. Mine wastes: characterization, treatment and environmental impacts (3rd Ed.). Heidelberg, Germany: Springer-Verlag Inc. 400pp. Lysay, G., J. McConnachie, C. Johns and J. Evans. 2010. Kemess South Mine Reclamation and Closure. Proceedings o f the Thirty Fourth Annual Mine Reclamation Symposium; 2010, September 20-23; Courtney, B.C. British Columbia Technical and Research Committee on Reclamation, http://www.trcr.bc.ca/ Macel, M., C. S. Lawson, S. R. Mortimer, M. Smilauerova, A. Bischoff, L. Cremieux, J. Dolezal, A. R. Edwards, V. Lanta, M. Bezemer, W. H. van der Putten, J. M. Igual, C. Rodriguez-Barrueco, H. Muller-Scharer, and T. Steinger. 2007. Climate vs. soil factors in local adaptation o f two common plant species. Ecology 88: 424-433. Macyk, T. M. 2002. Thirty years o f reclamation research in alpine and subalpine regions near Grande Cache, Alberta. Proceedings o f the Twenty Sixth Annual British Columbia Mine Relcamation Symposium; 2002, September 1 1 -1 4 ; Dawson Creek, B.C. British Columbia Technical and Research Committee on Reclamation, http://www.trcr.bc.ca/ Martinez-Ruiz, C., and B. Femandez-Santos. 2005. Natural revegetation on topsoiled mining-spoils according to exposure. Acta Oecologica 28: 231-238. Martin, H. W., and D. I. Kaplan. 1998. Temporal changes in cadmium, thallium and vanadium mobility in soils and phytoavailability under field conditions. Water, Air and Soil Pollution 101: 399-410. Marx, E. S., J. Hart, and R. G. Stevens. 1999. Soil test interpretation guide. Oregon State University Extension Service, Corvallis, Oregon. Matsumura, T., and T. Yoshiaki. 2010. Relationship between species richness and spatial and temporal distance from seed source in semi-natural grassland. Journal o f Applied Vegetation Science 13: 336-345. Meikle, T. W., S. Lu, and J. P. Barta. 1999. Evaluation o f nutrient and biological amendments on waste rock. Society for Mining Metallurgy and Exploration Annual Meeting; 1999, March 1-3; Denver, Colorado. Society for Mining, Metallurgy, and Exploration. http://www.smenet .ore/ Ministry o f Natural Resource Operations. 2010. Operating mines and selected major exploration projects in British Columbia, 2010. Retrieved May 31st, 2011 from http://www.empr.gov.be.ca/mining/geoscience/publicationscatalogue/ openfiles/201 l/pages/2011-l.aspx Montalvo, A.M., P. A. Mcmillan, and E.B. Allan. 2002. The relative importance o f seeding method, soil ripping and soil variables on seeding success. Restoration Ecology 10: 52-67. Murcia, F. J., M. J. Gomez, A. Palop, and G. Garcia. 2007. Assessment o f mining soil improvement after construction and demolition wastes amendment additions evaluated by microbiological analysis. Global NEST Journal 9: 224-228. 142 Naeth, A. M., H. A. Archibald, C. L. Nemirsky, L. A. Leskiw, J. A. Brierley, M. D. Bock, A. J. VanderBygaart, and D. S. Chanasyk. 2012. Proposed classification for human modified soils in Canada: Anthroposolic order. Canadian Journal o f Soil Science 92: 7-18. Nicholas, P. J., and W. J. McGinnes. 1982. An evaluation of 17 grasses and 2 legumes for the revegetation of soil and spoil on a coal strip mine. Journal of Range Management 35: 288293. Noel, T. C., C. Sheng, C. K. Yost, R. P. Pharis, and M. F. Hynes. 1996. Rhizobium leguminosarum as a plant growth-promoting rhizobacterium: direct growth promotion o f canola and lettuce. Canadian Journal o f Microbiology 42: 279-283. Olfelt, J. P., D. P. Olfelt, and J. L. Ison. 2009. Revegetation o f a trampled cliff-edge using three-toothed cinquefoil and poverty grass: a case study at Tettegouche State Park, Minnesota. Restoration Ecology 27: 200-209. Pertuit, A. J., J. B. Dudley, and J. E. Toler. 2001. Leonardite and fertilizer levels influence tomato seedling growth. HortScience 36: 913-915. Petterson, S. L., B. A. Roundy, and R. M. Bryant. 2004. Revegetation methods for high elevation roadsides at Bryce Canyon National Park, Utah. Restoration Ecology 12: 248-257. Phelps, L. B., and L. Holland. 1987. Soil compaction in topsoil replacement during mine reclamation. Environmental Geochemistry and Health 9: 8-11. Pinheiro, J., D. Bates, S. DebRoy, D. Sarkar, and the R Core Team. 2009. nlme: linear and non-linear mixed effects models. R package version 3.1-96. Retrieved from http://cran.rproiect.org/web/packages/nlme/index.html Polster, D. F. 1991. Natural vegetation succession and sustainable reclamation. Proceedings o f the Sixth Annual British Columbia Mine Reclamation Symposium; 1991, June 24-28; Kamloops, B.C. British Columbia Technical and Research Committee on Reclamation. http://www.trcr.bc.ca/ Polster, D. F. 1997. Restoration of Landslides and Unstable Slopes: Considerations for Bioengineering in Interior Locations. Proceedings o f the Twenty-First Annual British Columbia Mine Reclamation Symposium; 1997, September 22-25; Cranbrook, B.C. British Columbia Technical and Research Committee on Reclamation, http://www.trcr.bc.ca/ Polster, D. 2001. Revegetation strategies at Island Copper Mine. Proceedings o f the Twenty Fifth Annual British Columbia Mine Reclamation Symposium; 2001, September 23-27; Campbell River, B.C. British Columbia Technical and Research Committee on Reclamation. http://www.trcr.bc.ca/ Polster, D., and D. Howe. 2006. Mine reclamation in British Columbia: accolades and issues. Proceedings o f the Thirtieth Annual British Columbia Mine Reclamation Symposium; 2006, June 19-22; Smithers, B.C. British Columbia Technical and Research Committee on Reclamation, http://www.trcr.bc.ca/ 143 Power, J. F., F. M. Sandoval, R. E. Ries, and S. D. Merrill. 1981. Effects o f topsoil and subsoil thickness on soil water content and crop production on a disturbed soil. Soil Science Society o f America Journal 45: 124-129. Redente, E. F., and J. L. Richards. 1997. The effects o f lime and fertilizer amendments on plant growth in smelter impacted soils in Montana. Arid Soil Research and Rehabilitation 11: 353-366. Reid, N. B., and M. A. Naeth. 2005a. Establishment o f a vegetation cover on tundra kimberlite mine tailings: 1. a greenhouse study. Restoration Ecology 13: 594-601. Reid, N. B., and M. A. Naeth. 2005b. Establishment o f a vegetation cover on tundra kimberlite mine tailings: 2. a field study. Restoration Ecology 13: 602-608. Reintam, E., K. Trukmann, J. Kuht, E. Nugis, L. Edesi, A. Astover, M. Noormets, K. Kauer, K. Krebstein, and K. Rannik. 2009. Soil compaction effects on bulk density and penetration resistance and growth of spring barely (Hordeum vulgare L.). Acta Agriculturae Scandinavica Section B - Soil and Plant Science 59: 265-272. Richmond, T. C. 2000. The revegetation o f metalliferous tailings, pp. 801-818. In: Bamhisel, R. I., R. G. Darmody, and W. L. Daniels (eds.). Reclamation o f drastically disturbed lands. Wisconsin, USA: American Society o f Agronomy and Academic Press. 1106pp. Schaetzl, R., and S. Anderson. 2007. Soils genesis and geomorphology. Cambridge, UK: Cambridge University Press, pp. 89. Schladweiler, B. K., G. F. Vance, D. E. Legg, L. C. Munn, and R. Haroian. 2005. Topsoil depth effects on reclaimed coal mine and native area vegetation in Northeastern Wyoming. Rangeland Ecology and Management 58: 167-176. Schoenholtz, S. H., H. V. Miegroet, and J. A. Burger. 2000. A review o f chemical and physical soil properties as indicators o f forest soil quality: challenges and opportunities. Forest Ecology and Management 138: 335-356. Schor, H. J., and D. H. Gray. 2007. Landforming: an environmental approach to hillside development, mine reclamation and watershed restoration. New York, USA: John W iley and Sons. pp83. Schmitt, R., S. Ames, and D. L. Stoopnikoff. 2008. Development o f the new Afton Gold Copper Mine on a former minesite in the agricultural land reserve Kamloops, BC. Proceedings o f the Thirty-second Annual British Columbia Mine Reclamation Symposium; 2008, September 17-20; Kamloops, B.C. British Columbia Technical and Research Committee on Reclamation, http://www.trcr.bc.ca/ Sharman, K., and C. Smyth. 2002. The progression o f native species island establishment and monitoring practices on high-elevation waste rock dumps at Quintette Operation Corporation. Proceedings of the 26th Annual British Columbia Mine Reclamation Symposium; 2002, September 9-131; Dawson Creek, B.C. British Columbia Technical and Research Committee on Reclamation, http://www.trcr.bc.ca/ 144 Shaw, B. T. 1952. Soil physical conditions and plant growth. Volume 2. Agronomy: a series o f monographs. New York, USA: Academic Press Inc. 491pp. Shumanm, G. E., E. J. Depuit, J. A. Smith, and L. A King. 2005. Reclamation o f bentonite mined lands in the northern great plains. Journal of Environmental Geochemistry and Health 7: 42-47. Silva, A., S. Imhoff, and B. Kay. 2004. Plant response to mechanical resistance and air-filled porosity of soils under conventional and no tillage systems. Agricultural Science 61: 451 456. Sjoberg, D. B., G. W. Wilson, and M. Aubertin. 2003. Evaluation of a single-layer desulphurized talings cover. Proceedings o f the Sixth International Conference on Acid Rock Drainage; 2003, July 14-17; Caims, Queensland, Australia. The Australasian Institute of Mining and Metallurgy, http://www.ausimm.com.au/ Sjoberg, D. B., G. W. Wilson, and M. Aubertin. 2001. Field and laboratory characterization of a desulphurized tailings cover system. Proceedings o f the 54th Canadian Geotechnical Conference; 2001, September 15-18; Calgary, Alberta. The Canadian Geotechnical Society. http://www.cgs.ca/index.php?lang=en Skrindo, A.B., and R. Halvorsen. 2008. Natural revegetation o f forest topsoil and subsoil along roadsides in boreal forest. Journal o f Applied Vegetation Science 11:483-490. Smith, J. E., and P. W. Jordan. 1994. Stand density effects on branching in an annual legume. Armais of Botany 74: 17-25. Stirzaker, R. J., J. B. Passioura, and Y. Wilms. 1996. Soil structure and plant growth: impact of bulk density and biophores. Plant and Soil 185: 151-162. Stuczynski T., G. Siebielec, W. L. Daniels, G. McCarty, and R. L. Chaney. 2007. Biological aspects o f waste metal reclamation with biosolids. Journal o f Environmental Quality 36: 1154-1162. Topper, K. F., and B. R. Sabey. 2005. Sewage sludge as a coal minespoil amendment for revegetation in Colorado. Journal of Environmental Quality 15: 44-49. Tordoff, G. M., A. J. M. Baker, and A. J. Willis. 2000. Current approaches to the revegetation and reclamation of metalliferous mine wastes. Chemosphere 41: 219-228. Troeh, F. R., and L. M. Thompson. 1993. Soils and soil fertility (5th ed.). New York, USA: Oxford University Press Inc. 462pp. Vanek, V., J. Balik, J. Silha, and J. Cemy. 2008. Spatial variability of soil nitrogen and sulphur content at two conventionally managed field. Plant, Soil and Environment 10: 413419. Xu, Y., J. P. Lay, and F. Korte. 1988. Fate and effects o f xanthates in laboratory freshwater systems. Bulletin o f Environmental Contamination and Toxicology 41: 683-689. 145 Zhang, R., and B. Wienhold. 2002. The effect o f soil moisture on mineral nitrogen, soil electrical conductivity and pH. Nutrient Cycling in Agroecosystems 63: 251-254. 146 7.0 Appendices Appendix 1. Details o f the dates, locations and participants for the collection o f locally native seeds o f blue wildrye, arctic lupine and Merten’s sedge within the local area o f the Huckleberry Mine (10 km radius). Easting Northing UTM Date3 Collectors 625287 5957133 09U Burton, Carson Aug 1st 618402 09U 5949205 Burton, Carson, Sparks Aug 1st 618603 5949900 Burton, Carson, Sparks Aug 1st 09U 625032 09U 5952888 Burton, Carson, Sparks Aug 2nd 624729 5953458 Burton, Carson, Sparks Aug 2nd 09U 618402 5949205 Burton, Carson Aug 1st 09U 624729 5953458 Aug 2nd 09U Burton, Carson, Sparks 618728 5949241 Sept 3rd 09U Davis Sept 4th Robichaud 5348764 621505 Burton, Davis 09U Sept 15th 618728 5949241 Burton, Davis Sept 15th 09U 09U 618765 5949165 Burton, Davis Sept 15th 618485 5948939 09U Burton, Davis Sept 15th 5956214 625358 Burton, Davis Sept 16th 09U 625307 5956416 09U Burton, Davis Sept 16th ___• ' i r \ r \ n aT->_._____r _____J _ _ n _ , > „ „ , • ... , "Dates o f seed collection in 2008; b bcollectors •include Ron Robichaud, Mike Davis, Sara Sparks, Allan Carson and Carla Burton. Species arctic lupine arctic lupine arctic lupine arctic lupine arctic lupine Merten’s sedge Merten’s sedge blue wildrye blue wildrye blue wildrye blue wildrye blue wildrye blue wildrye blue wildrye blue wildrye Elevation (m) 1010 1002 1099 975 973 1002 973 1049 1067 1049 1045 1009 968 979 147 Appendix 2. Photographs o f each o f the test plot locations, including a) m illsite, b) lower east dam, c) upper TMF-2 and d) TMF-2 north stockpile. « n -III a) .A J . V 1 ' / - w /.z- l%i,: f & - La * .. .' c) <5o° ° ' .. d) 148 Appendix 3. Physical and chemical analyses conducted on soil samples collected from the 12 revegetation test plots in 2009 and 2010. L aboratory/A nalysis D escription s and R eferences o f A nalyses B CM O F L aboratory Particle Size Analysis Total Carbon, Nitrogen and Sulphur and Inorganic Carbon A vailable Phosphorus (PO 4 -P) M inerizeable Nitrogen A vailable Ammonium ( N H 4 - N ) and Nitrate ( N O 3 - N ) Carbonate (CaCOj) equivalent pH (Calcium Chloride and Water) Electrical Conductivity Extractable Elements (Ai, B, Ca, Cu, Fe, K, Mg, Mn, Na, P, S and Zn Carter, M. J. (Ed). 1993. Soil sampling and methods o f analysis. Florida, USA: Lewis publishing, pp. 503, 507-509. Carter, M. R., and E. G. Gregorich (Eds). 2008. Soil Sam pling and Methods o f Analysis (2nd edition). Florida, USA: CRC Press, pp. 226. Leco Corporation Instruction and Reference Documents for Truspec NC Analyzer Leco Corporation Instruction and Reference Documents for Truspec Add-on S Analyzer Thermo Instruments Ltd. Instruction and Reference Documents for N A -1500 Analyzer Missouri Agricultural Experiment Station SB 1001. 1998. Recommended chemical soil test procedures for the north central region. North Central R egional Research Publication NO. 221 (Revised). Bremner, J.M. 1965. Nitrogen Availability Indexes. In Methods o f soil analysis: part 2 - chemical and biological properties, C.A. Black (Ed). W isconsin, USA: American Society o f America, pp. 1324-1345. Waring, S. A., and J.M. Bremner. 1964. Ammonium production in soil under waterlogged conditions as an index o f nitrogen availability. Nature 2 0 1:951-952. Carter, M. J. (Ed). 1993. Soil sampling and methods o f analysis. Florida, USA: Lewis publishing, pp 25-37. Bremner, J.M. 1965. Inorganic forms o f nitrogen. In Methods o f soil analysis: part 2 - chem ical and biological properties, C.A. Black (Ed). W isconsin, USA: American Society o f America, pp. 1179-1237. Carter, M. J. (Ed). 1993. Soil sampling and methods o f analysis. Florida, USA: Lewis publishing, pp. 177-179. Kalra, Y .P., and Maynard, D.G. 1991. M ethods Manual for Forest Soil and Plant Analysis, Forestry Canada. Forestry Canada, Northwest Region, Northern Forestry Centre, Edmonton, Alberta. Information Report NOR-X-319E. 116pp. Atkinson, H. J., G. R. G iles, A. J. MacLean, and J. R. Wright. 1958. Chemical M ethods o f Soil Analysis - Contribution N o. 169 (Revised). Chemistry D ivision - Science Service Canada Department o f Agriculture, Ottawa. 90pp. Carter, M. R., and E. G. Gregorich (Eds). 2008. Soil Sampling and Methods o f Analysis (2n<1 ed.). Florida, USA: CRC Press. 1264pp. United States Salinity Laboratory Staff. 1954. Diagnosis and improvement o f saline and alkali soils", United States Department o f Agriculture Handbook No: 60. United States Government Printing Office, W ashington, D.C. 160pp. Kalra, Y .P., and Maynard, D.G. 1991. M ethods Manual for Forest Soil and Plant Analysis, Forestry Canada. Forestry Canada, Northwest Region, Northern Forestry Centre, Edmonton, Alberta. Information Report N O R -X -319E. 1 16pp. Carter, M. J. (Ed). 1993. Soil sampling and methods o f analysis. Florida, USA: Lew is publishing, pp. 43-49. Exchangeable Cations (Ca, K, M g, Na) and CEC Carter, M. R., and E. G. Gregorich (Eds). 2008. Soil Sampling and Methods o f Analysis (2nd edition). Florida, USA: CRC Press, pp. 203. Trace Elements (As, Ba, Bi, Cd, Co, Cr, Hg, Mo, Ni, Pb, Sb, Sn, Ti and V) Carter, M. J. (Ed). 1993. Soil sampling and methods o f analysis. Florida, USA: Lewis publishing, pp. 43-49. (M odified in-house to include additional, non-standard elements) A LS C anada Ltd. Laboratory Total Elemental A nalysis for Rare Earth and Trace Elements A prepared sample (0.200 g) is added to lithium borate flux (0.90 g), mixed w ell and fused in a furnace at 1000°C. 'rite resulting melt is then cooled and dissolved in 100 mL o f 4 % H N 0 3 / 2 % HC1 solution. This solution is then analyzed by 1CP-MS. B ase Metals A prepared sample (0.25 g) is digested with perchloric, nitric, hydrofluoric and hydrochloric acids. The residue is topped up with dilute hydrochloric acid and the resulting solution is analyzed by ICP-AES. Results are corrected for spectral inter-element interferences. Major Oxides A prepared sample (0.200 g) is added to lithium metaborate/lithium tetraborate flux (0.90 g), mixed w ell and fused 111 a furnace at 1000°C. The resulting melt is then cooled and dissolved in 100 mL o f 4 % nitric acid/2 % hydrochloric acid. This solution is then analyzed by ICP-AES and the results are corrected for spectral inter-element interferences. Oxide concentration is calculated from the determined elemental concentration and the result is reported in that format._________________________________________________________________________________________________________ 149 Appendix 4. Photographs showing the difference in soil color (i.e., darkness) for a) soils supplemented with N A G sand and b) without N A G sand using the test plot located near the m illsite as an exam ple. Soil from the area surrounding the test plot was placed in between subplots to act as a buffer and can be seen here. I- - - - - - - - - - - - - - - - - - - - r ' - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - * - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - C - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ ------------------------------------- 3 E — I 150 Appendix 5. Photographs displaying what appears to be a) surface sealing on stockpiled soils compared to b) the surface N A G sand supplemented soils where surface sealing is not visible. Where the sam e stockpiled soil was used to reclaim another area o f the minesite, surface pooling follow ing a rainfall event w as observed (c). c) 151 Appendix 6. Monthly total precipitation (mm) at Huckleberry M ine for each month in 2 006, 2 007, 2009 and 2010. Rainfall and snowfall measurements were collected from the mine weather station located on the roof o f the administration building at the camp and mill site (refer to Figure 1 for location). Total precipitation was calculated using a snow:rainfall conversion ratio o f 10:1 (Environment Canada 2012). 300 □ 2006 □ 2007 250 - E £ 02010 200 c o £ 150 - 01 ah . 75 100 o 50 - Jan Feb M ar Apr M ay Jun Jul Aug S ep Oct Nov Dec M o n th 152 Appendix 7. Monthly daily minimum (above) and maximum (below ) temperatures (°C) at Huckleberry M ine for each month in 2006, 2007, 2009 and 2010. Temperature measurements were collected from the mine weather station located on the roof o f the administration building at the camp and mill site (refer to Figure 1 for location). □ 2006 02007 < u W ES2010 3 0a> f E